This week in Neurology® Highlights of the January 13 issue
A multidisciplinary study of patients with early-onset PD with and without parkin mutations This study indicates that carriers of parkin mutations are clinically indistinguishable from other early-onset patients. Nevertheless, parkin patients had significantly lower daily doses of dopaminergic treatment and greater delay in the development of levodopa-related fluctuations. See p. 110; Editorial, p. 106
Comorbidity delays diagnosis and increases disability at diagnosis in MS The authors studied over 8,000 patients with MS and observed longer delays in diagnosis and more disability at diagnosis in patients with pre-existing chronic conditions. This indicates that physicians seeing patients with chronic
MRI correlates of cognitive decline in CADASIL: A 7-year follow up study This paper investigates which of the MRI hallmarks in CADASIL are associated with cognitive decline. The authors found that increase in lacunar infarcts, microbleeds and ventricular volume, but not white matter lesions or atrophy, are associated with cognitive decline in CADASIL. See p. 143
Surrogate consent for dementia research: A national survey of older Americans Alzheimer disease clinical research often involves subjects who have lost their ability to consent; however, surrogate consent for dementia research remains controversial. This paper discusses a national survey of older Americans, which reveals broad public support for surrogate-based dementia research. See p. 149
conditions and new neurologic symptoms must be circumspect about attributing those symptoms to the existing conditions. See p. 117; Editorial, p. 108
Postmenopausal hormone therapy and subclinical cerebrovascular disease: The WHIMS-MRI Study The authors examine the association of conjugated equine
Cerebral microbleeds are a risk factor for warfarin-related intracerebral hemorrhage Underlying microbleeds on brain MRI are an independent risk factor for incident warfarin-related intracerebral hemorrhage. This study provides new radiological information on prognosis after long-term warfarin medication. See p. 171
estrogen-based hormone therapy (HT) versus placebo and silent cerebrovascular disease as a mechanism for negative effects of HT on cognition in older women. HT is not associated with increased lesion volume on brain MRI conducted 8 years after randomization to HT. See p. 135
Postmenopausal hormone therapy and regional brain volumes: The WHIMS-MRI Study The authors found that older women randomized to conjugated equine estrogen-based hormone therapy (HT) versus placebo have smaller hippocampal and frontal volumes on MRI assessed a mean of 8 years after
SPECIAL ARTICLES
Practice Parameter: Evaluation of distal symmetric polyneuropathy: Role of autonomic testing, nerve biopsy, and skin biopsy (an evidence-based review) Practice Parameter: Evaluation of distal symmetric polyneuropathy: Role of laboratory and genetic testing (an evidence-based review) These papers outline the accuracy and usefulness of laboratory, genetic, and autonomic testing, as well as nerve and skin biopsy in the evaluation and management of patients with distal symmetric polyneuropathy. See p. 177 & p. 185
randomization. HT effects on regional volumes are most pronounced in women with lower cognitive function at baseline or higher ischemic lesion burden. See p. 135
Podcasts can be accessed at www.neurology.org
Copyright © 2009 by AAN Enterprises, Inc.
105
EDITORIAL
Parkinson disease(s) Is “Parkin disease” a distinct clinical entity?
Christine Klein, MD Katja Lohmann, PhD
Address correspondence and reprint requests to Dr. Christine Klein, Department of Neurology, University of Lu¨beck, Ratzeburger Allee 160, 23538 Lu¨beck, Germany christine.klein@neuro. uni-luebeck.de
Neurology® 2009;72:106–107
The discovery of several monogenic forms has clearly established Parkinson disease (PD) as an etiologically heterogeneous condition.1 However, initial high expectations of well-defined genotype–phenotype correlations have remained largely unmet, both at the clinical and at the pathologic level.2 To date, very few studies have systematically addressed the natural history of genetic PD, resulting in an almost complete lack of longitudinal data on genetic vs non-genetic forms. Likewise, it is currently unknown whether the frequency and type of nonmotor signs, a well-recognized feature of idiopathic PD, might help differentiate genetic from idiopathic PD. These considerations are of major importance as they will not only improve our general understanding of the various forms of PD but will likely also impact on a better prediction of the individual patient’s prognosis, on recommendations for genetic testing, and on the choice of therapeutic options including type and doses of antiparkinsonian medication and potential treatment of nonmotor signs. Parkin mutations are the most common known cause of early-onset parkinsonism3,4 and appear to be associated with a slower disease progression and an overall better response to dopaminergic treatment.5 In their article on a multidisciplinary study of patients with early-onset PD with and without Parkin mutations, Lohmann and colleagues confirm their previous observation of a more favorable disease course in the mutation carriers who maintained an excellent response to lower doses of L-dopa than patients without mutations.6 The novelty of this study lies in the detailed assessment of cognitive function and psychiatric features of affected and unaffected Parkin mutation carriers vs patients with idiopathic early-onset PD.6 Patients with Parkin mutations did not differ from patients without mutations with respect to occurrence of cognitive impairment, nor regarding specific behavioral or psychiatric symptoms that had previously been suggested as a key feature of “Parkin disease.”7
However, the lack of observed clinical differences between mutation carriers and noncarriers might, at least in part, be attributed to the relatively young age of many of the carriers of two or one mutated Parkin alleles, raising the possibility that the full-blown picture of the disease has not yet developed. As discussed by Lohmann et al.,6 another limitation of their study represents the relatively small sample size (21 patients with and 23 without Parkin mutations). This may also explain why they did not detect the well-established observation of an earlier age at onset in mutation carriers compared to patients without a Parkin mutation and why many of their comparisons yielded nonsignificant differences. A previous investigation focusing on special clinical features in a series of 24 Parkin mutation carriers emphasized early instability, autonomic failure, early or atypical dyskinesias, and recurrent psychosis as part of the phenotypic spectrum of “Parkin disease.” 7 While all of these signs may occur in individual Parkin mutation carriers, none of them seems to hold up as a red flag that may be generally attributed to patients with Parkin mutations.6 Indeed, a systematic literature review on cognitive and psychiatric features in Parkin mutation carriers found the rates for depression (31%), anxiety (26%), hallucinations (7%), and dementia (5%; unpublished data) to be comparable to or even lower than those reported for patients with idiopathic PD.8 Adding a further level of complexity, it has been suggested that the type and localization of Parkin mutations may play a role in shaping different Parkin-associated phenotypes.9 This potential source of bias needs to be considered when assessing phenotype– genotype correlations, as was accounted for by Lohmann and colleagues in their previous study.5 In spite of the aforementioned caveats, an increasing body of evidence suggests that Parkin disease is indeed characterized by a particularly benign disease
See page 110 e-Pub ahead of print on November 5, 2008, at www.neurology.org. From the Department of Neurology, University of Lu¨beck, Germany. C. Klein is supported by a Lichtenberg grant from the Volkswagen Foundation and recipient of a career development award from the Hermann and Lilly Schilling Foundation. Disclosure: The authors report no disclosures. 106
Copyright © 2009 by AAN Enterprises, Inc.
course with slow progression, excellent treatment response, and few, if any, additional motor and nonmotor signs. What we do not know at this point is whether Parkin disease is clinically different from PINK1 or DJ-1 disease and whether Parkin mutation carriers employ certain, particularly effective compensatory mechanisms counteracting the effects of their neurodegeneration. Finding answers to these intriguing questions will require a combined approach of large-scale, longitudinal clinical studies, biomarker development, and an improved understanding of the pathways and mechanisms involved in “Parkin disease” and other forms of PD.
3.
4.
5.
6.
7.
8. REFERENCES 1. Klein C, Schlossmacher MG. Parkinson disease, 10 years after its genetic revolution: multiple clues to a complex disorder. Neurology 2007;69:2093–2104. 2. Marras C, Lang AE. Changing concepts in Parkinson disease: moving beyond the decade of the brain. Neurology 2008 (in press).
9.
Lu¨cking CB, Du¨rr A, Bonifati V, et al. Association between early-onset Parkinson’s disease and mutations in the parkin gene. N Engl J Med 2000;342:1560–1567. Hedrich K, Eskelson C, Wilmot B, et al. Distribution, type, and origin of Parkin mutations: review and case studies. Mov Disord 2004;19:1146–1157. Lohmann E, Periquet M, Bonifati V, et al. How much phenotypic variation can be attributed to parkin genotype? Ann Neurol 2003;54:176–185. Lohmann E, Thobois S, Lesage S, et al., and the French Parkinson’s Disease Genetics Study Group. A multidisciplinary study of patients with early-onset PD with and without parkin mutations. Neurology 2009;72:110–116. Khan NL, Graham E, Critchley P, et al. Parkin disease: a phenotypic study of a large case series. Brain 2003;126: 1279–1292. Schrag A, Schott JM. Epidemiological, clinical, and genetic characteristics of early-onset parkinsonism. Lancet Neurol 2006;5:355–363. Hampe C, Ardila-Osorio H, Fournier M, Brice A, Corti O. Biochemical analysis of Parkinson’s disease-causing variants of Parkin, an E3 ubiquitin-protein ligase with monoubiquitylation capacity. Hum Mol Genet 2006; 15:2059–2075.
Neurology 72
January 13, 2009
107
EDITORIAL
Selection bias in observational studies Out of control?
Christopher A. Beck, PhD
Address correspondence and reprint requests to Dr. Christopher A. Beck, Department of Biostatistics and Computational Biology, University of Rochester Medical Center, 601 Elmwood Ave, Box 630, Rochester, NY 14642
[email protected]
Neurology® 2009;72:108–109
Bias comes in many flavors. Observational studies are especially prone to its many forms, mainly due to the investigator’s lack of control over the study. Selection bias may result when a study fails to select a representative sample from the population of interest, limiting the applicability of the study’s results.1,2 In this issue of Neurology®, Marrie et al.3 report an analysis of questionnaires completed by participants of the North American Research Committee on Multiple Sclerosis (NARCOMS) Registry,4 a self-report database for patients with multiple sclerosis (MS). They obtained demographic and clinical information from a questionnaire participants completed at the time of enrollment in the registry, and comorbidity information from a follow-up questionnaire in 2006. The study suggested that comorbidity lengthens the delay between onset and diagnosis of MS, and that comorbidity increases the severity of disability at diagnosis. However, after accounting for the delay in diagnosis, the association between comorbidity and disability at diagnosis diminished and in most cases became no longer significant. In other words, the delay in diagnosis partly mediated the effect of comorbidity on disability at diagnosis. For this study, 16,141 NARCOMS participants were eligible to receive the comorbidity questionnaire, and 8,983 (55.7%) responded. The authors restricted their primary analyses to the 2,375 (26.4%) responders who had enrolled in the registry within 2 years of diagnosis. Given the low response rate from registry participants, what is the population to which these results can be generalized? Ideally, this would be the general MS population, but bias in the selection of the sample may be an issue in the external validity of the study results. There are at least three sources of bias in the current study; the authors noted and addressed all three. These include the use of a disease registry database,5 nonresponse to the comorbidity questionnaire,6 and the restriction to subjects who enrolled in the registry within 2 years of diagnosis. The authors addressed the first two of these by comparing the characteristics of their sample to all NARCOMS participants and to MS subjects nationally, and found minor differences in race, gender, socioeconomic
status, type of insurance coverage, disease course, and therapies used. The most concerning difference between their sample and other patients with MS relates to the restriction to patients enrolled in the registry within 2 years of diagnosis. This restriction was necessary in order to assume that demographic and disability data from the enrollment questionnaire would not have changed between diagnosis of MS and enrollment in the registry. This resulted in a sample of patients with milder disability and shorter duration of disease than the general MS population, potentially limiting the generalizability of the association detected between comorbidity and disability at diagnosis. However, the finding that comorbidity lengthens diagnostic delay is supported by additional analyses based on all responders to the comorbidity questionnaire. Thus, the generalizability of this association does not appear to be limited by the restriction to subjects who enrolled in the registry within 2 years of diagnosis. Effective prevention of selection bias can be achieved using random sampling. This technique selects subjects independently and randomly from the population, with each subject having an equal chance of being selected. In research involving humans, a truly random sample is nearly impossible since subjects have to agree to participate in a study. The extent of selection bias in these studies should be assessed by comparing volunteers and non-volunteers whenever possible.7,8 Even randomized clinical trials are not immune to selection bias. Such studies use randomization in the assignment of treatments to enrolled subjects, but not in the selection of subjects to be enrolled. This results in treatment groups that tend to be comparable to each other in all respects, but potentially different from the general population. These treatment groups would be affected equally by any bias in the sample selection. In this case, if the purpose of the trial is to compare the efficacy of the treatments, selection bias is only a concern if the treatment effect interacts with any characteristics that differ between the sample and the population.9 The effect of selection bias should be carefully considered in the interpretation of results based on nonrandom
See page 117 From the Department of Biostatistics and Computational Biology, University of Rochester Medical Center, NY. Disclosure: The author reports no disclosures. 108
Copyright © 2009 by AAN Enterprises, Inc.
samples, and any extrapolations should be made with extreme caution. Although selection bias can never be completely controlled, its effect can be minimized and better understood through proper study design and execution. Marrie et al. recognize these issues, and although the generalizability of their results do not appear to be compromised, they are correct in concluding that their findings require replication in population-based cohorts.
4.
5. 6.
7. REFERENCES 1. Ellenberg JH. Selection bias in observational and experimental studies. Stat Med 1994;13:557–567. 2. Williams WH. How bad can “good” data really be? Am Stat 1978;32:61–65. 3. Marrie RA, Horwitz R, Cutter G, Tyry T, Campagnolo D, Vollmer T. Comorbidity delays diagnosis and increases disability at diagnosis in MS. Neurology 2009;72:117–124.
8.
9.
The Consortium of Multiple Sclerosis Centers. North American Research Committee on Multiple Sclerosis Patients Registry. Available at: www.mscare.org/cmsc/CMSC-NARCOMSInformation.html. Accessed August 7, 2008. Dambrosia JM, Ellenberg JH. Statistical considerations for a medical data base. Biometrics 1980;36:323–332. Rupp I, Triemstra M, Boshuizen HC, Jacobi CE, Dinant HJ, van den Bos GAM. Selection bias due to non-response in a health survey among patients with rheumatoid arthritis. Eur J Public Health 2002;12:131–135. Kim SYH, Holloway RG, Frank S, et al. Volunteering for early phase gene transfer research in Parkinson disease. Neurology 2006;66:1010–1015. Coronary artery surgery study (CASS): a randomized trial of coronary artery bypass surgery. Comparability of entry characteristics and survival in randomized patients and nonrandomized patients meeting randomization criteria. J Am Coll Cardiol 1984;3:114–128. Piantadosi S. Clinical Trials: A Methodologic Perspective, 2nd ed. New York: Wiley; 2005.
Neurology 72
January 13, 2009
109
ARTICLES
A multidisciplinary study of patients with early-onset PD with and without parkin mutations E. Lohmann, MD S. Thobois, MD, PhD S. Lesage, PhD E. Broussolle, MD, PhD S. Tezenas du Montcel, MD, PhD M.-J. Ribeiro, PhD P. Remy, MD, PhD A. Pelissolo, MD B. Dubois, MD L. Mallet, MD P. Pollak, MD, PhD Y. Agid, MD, PhD A. Brice, MD The French Parkinson’s Disease Genetics Study Group*
ABSTRACT
Objective: To establish phenotype– genotype correlations in early-onset Parkinson disease (EOPD), we performed neurologic, neuropsychological, and psychiatric evaluations in a series of patients with and without parkin mutations.
Background: Parkin (PARK2) gene mutations are the major cause of autosomal recessive parkinsonism. The usual clinical features are early-onset typical PD with a slow clinical course, an excellent response to low doses of levodopa, frequent treatment-induced dyskinesias, and the absence of dementia.
Methods: A total of 44 patients with EOPD (21 with and 23 without parkin mutations) and 9 unaffected single heterozygous carriers of parkin mutations underwent extensive clinical, neuropsychological, and psychiatric examinations. Results: The neurologic, neuropsychological, and psychiatric features were similar in all patients, except for significantly lower daily doses of dopaminergic treatment and greater delay in the development of levodopa-related fluctuations (p ⬍ 0.05) in parkin mutation carriers compared to noncarriers. There was no major difference between the two groups in terms of general cognitive efficiency. Psychiatric manifestations (depression) were more frequent in patients than in healthy single heterozygous parkin carriers but did not differ between the two groups of patients.
Conclusion: Carriers of parkin mutations are clinically indistinguishable from other patients with Address correspondence and reprint requests to Prof. Alexis Brice, INSERM UMR S_679, Hoˆpital Pitie´-Salpeˆtrie`re, 47 boulevard de l’Hoˆpital, F-75013 Paris, France
[email protected]
young-onset Parkinson disease (PD) on an individual basis. Severe generalized loss of dopaminergic neurons in the substantia nigra pars compacta in these patients is associated with an excellent response to low doses of dopa-equivalent and delayed fluctuations, but cognitive impairment and special behavioral or psychiatric symptoms were not more severe than in other patients with early-onset PD. Neurology® 2009;72:110–116 GLOSSARY CPRS ⫽ Comprehensive Psychopathological Rating Scale; EOPD ⫽ early-onset Parkinson disease; FAB ⫽ Frontal Assessment Battery; MADRS ⫽ Montgomery-Asberg Depression Rating Scale; MDRS ⫽ Mattis Dementia Rating Scale; MINI ⫽ Mini International Neuropsychiatric Interview; MMSE ⫽ Mini-Mental State Examination; TMT ⫽ Trail Making Test; UPDRS ⫽ Unified Parkinson’s Disease Rating Scale; WCST ⫽ Wisconsin Card Sorting Test.
Mutations in the parkin gene are considered to be the predominant cause of early-onset Parkinson disease (EOPD) particularly when the family history is compatible with autosomal recessive inheritance.1,2 Parkin-linked PD has a broad range of clinical phenotypes, some atypical, but is generally early-onset parkinsonism, with a slow clinical course, excellent response to low doses of levodopa, frequent treatment-induced dyskinesias, and no dementia.3,4 Cognitive function remains normal in the majority of patients,3-6 but behavioral disorders have been reported, including anxiety, psychosis, panic attacks, depression, and disturbed sexual, behavioral, and obsessive-compulsive disorders.4,5,7-9 Editorial, page 106 e-Pub ahead of print on November 5, 2008, at www.neurology.org. *The French Parkinson’s Disease Genetics Study Group members are listed in the appendix. Authors’ affiliations are listed at the end of the article. Supported by INSERM/AP-HP (PCR02006-P011104), the NIH grant NS41723-01A1, and the European commission (EU Contract No.LSHMCT-2003-503330/APOPIS). Disclosure: The authors report no disclosures. 110
Copyright © 2009 by AAN Enterprises, Inc.
Mutations have been found in the homozygous or compound heterozygous state, compatible with recessive transmission, but occasionally as single heterozygous mutations.2,5,10 In these patients, it is still not clear whether a second mutation outside the open reading frame of the gene may be overlooked or whether some heterozygous mutations are sufficient to cause the disease.11 To establish pheno– genotype correlations in EO parkinsonism, we performed a clinical study of 21 patients with and 23 patients without parkin mutations. We expected behavioral or psychiatric problems in parkin mutation carriers because neuropathologic studies of parkin patients show severe generalized loss of dopaminergic neurons in the substantia nigra pars compacta that would greatly decrease dopaminergic efferents to the limbic and sensory-motor systems. In addition, although it is generally assumed that parkin gene mutations carriers have no dementia, systematic assessment of cognitive functions in this population was not performed. We aim to fill this gap. Additionally, we evaluated 9 healthy sibs of our patients with single heterozygous parkin mutations. METHODS Forty-four patients with isolated or familial earlyonset parkinsonism (⬍45 years) recruited in Paris (n ⫽ 25), Grenoble and Lyon (n ⫽ 19) underwent neurologic, neuropsychological, and psychiatric evaluation. The inclusion criteria for PD were at least two of the parkinsonian triad of signs (bradykinesia, rigidity, rest tremor) and at least 30% improvement under levodopa therapy, in familial or isolated cases. Exclusion criteria were the existence of extensor plantar reflexes, ophthalmoplegia, early dementia, or early autonomic failure. There were 13 women and 31 men; mean age at onset was 33.1 years ⫾ 8.3 (12– 44) and mean disease duration was 17.4 years ⫾ 7.8. (5–34). Twenty-five patients had known family histories of PD. Among the 44 patients originating from France (n ⫽ 41), Asia (n ⫽ 2), and North Africa (n ⫽ 1), 21 had homozygous or compound heterozygous parkin mutations and 23 patients no parkin mutations. Five female and four male heterozygous parkin carriers originating from France (n ⫽ 7) and North Africa (n ⫽ 2), healthy sibs of the parkin patients, also participated in the study. Their age at examination was 47.9 years ⫾ 12.6 (28 – 68). A standardized form was used to assess the history of the disease in the patient and the family, the clinical signs, additional diseases, and the response to present treatment. All patients were evaluated with the Unified Parkinson’s Disease Rating Scale (UPDRS) I to VI, with and without treatment (“on” and “off” state), except UPDRS I and IV, which were evaluated only in the “off” state. The “off” state was usually reached after interruption of antiparkinsonian medication for at least 12 hours, and the best “on” state was obtained after administration of a single su-
Table 1
Parkin mutations in affected and unaffected carriers
Mutations in unaffected heterozygous carriers (n ⴝ 9)
Mutations in affected carriers (n ⴝ 21)
ex8-9del/N (2 patients)
ex8-9del/ex8-9del
ex3-4del/N
ex6dupl/c.255delA (2 patients)
ex3dupl/N
ex5del/Cys411Arg (4 patients)
ex2-3del/N
ex5del/c255delA
ex2tripl/N
ex4del/Met1stop (2 patients)
promotor-ex1del/N (2 patients)
ex3-6del/Arg275Trp
Met1stop/N
ex3-4del/Arg275Trp (3 patients) ex3del/Arg275Trp ex3del /c.202-203delAG ex3del/promotor-ex1del (2 patients) ex3dupl/ex3dupl ex3del/ex3del (2 patients)
prathreshold dose of levodopa (50 mg higher than the usual effective morning dose). The response to treatment was calculated from the “off” and “on” values of UPDRS III. We used the conversion factors proposed by Thobois12 to calculate the daily levodopa dose-equivalent taken by the patients. Neuropsychological and psychiatric tests were performed while the patients received their usual treatment. The MiniMental State Examination (MMSE) and the Mattis Dementia Rating Scale (MDRS) were used to assess global intellectual efficiency. Verbal episodic memory was investigated with the Grober and Buschke test to control for effective encoding and to facilitate retrieval with the same semantic cueing, providing a comparison between free and cued recall. Two parallel forms of the Grober and Buschke test were used to control test-retest effects. Executive functions were assessed with the simplified version of the Wisconsin Card Sorting Test (WCST), the category and literal fluency test, the Trail Making Test (TMT), the Frontal score, and the Frontal Assessment Battery (FAB). The psychiatric evaluation was based on the Montgomery-Asberg Depression Rating Scale (MADRS), the Comprehensive Psychopathological Rating Scale (CPRS), and a standardized Mini International Neuropsychiatric Interview (MINI). Molecular analysis of the parkin gene was performed by denaturing high performance liquid chromatography and sequencing and semiquantitative multiplex PCR as described before.13 The G2019S mutation of the LRRK2 gene and mutations in the DJ-1 and Pink1 genes were excluded in all patients.14 Data are expressed as the mean ⫾ SD (range) or percentage (n). Group comparisons were made using the Kruskal-Wallis test for quantitative variables and the Fisher exact test for qualitative variables. When significant differences were detected, post hoc comparisons were performed with a Bonferroni correction for multiple comparisons. For the delay before dyskinesia, dystonia, and fluctuations, Kaplan-Meier estimations were obtained. The Sidak correction for multiple comparisons was applied by discipline (clinical, psychiatric, neuropsychological). The SAS 8.1 Neurology 72
January 13, 2009
111
Table 2
Clinical characteristics of patients with Parkinson disease (PD) with parkin and without parkin mutations PD patients with parkin mutations (n ⴝ 21)
PD patients without parkin mutations (n ⴝ 23)
Age at onset, y
31.1 ⫾ 8.3 (12–44)
34.9 ⫾ 6.2 (16–44)
PD duration, y
17.4 ⫾ 7.8 (5–34)
12.9 ⫾ 7.7 (4–34)
UPDRS I (/16) UPDRS II off (/52) UPDRS II on (/52)
2.1 ⫾ 1.9 (0–6)
2.3 ⫾ 1.8 (0–6)
12.1 ⫾ 6.4 (2–23)
17.3 ⫾ 8.2 (6–34)
4.3 ⫾ 4 (0–17)
5 ⫾ 4.1 (0–16)
UPDRS III off (/108)
32.4 ⫾ 16.2 (7–63)
UPDRS III on (/108)
10.2 ⫾ 6.5 (1–24)
12.4 ⫾ 9.7 (1–37)
5.2 ⫾ 3.1 (0–12)
7.3 ⫾ 3.5 (2–17)
UPDRS IV (/23)
40.2 ⫾ 16.6 (19–70)
Hoehn and Yahr off (/5)
2.6 ⫾ 0.8 (1–4)
2.8 ⫾ 0.9 (1–5)
Hoehn and Yahr on (/5)
1.5 ⫾ 0.9 (0–3)
1.4 ⫾ 0.9 (0–3)
Schwab and England off (0–100%) Schwab and England on (0–100%)
78.1 ⫾ 12 (50–90) 91.9 ⫾ 8 (80–100)
Daily doses of levodopa equivalent, mg/d*
636 ⫾ 462 (120–2,150)
Daily doses of levodopa, mg/d
528 ⫾ 439 (50–2,000)
Duration of treatment, y
13.2 ⫾ 6.5 (4–31) (n ⫽ 20)
Occurrence of dyskinesia Delay before dyskinesia after treatment, y† Occurrence of dystonia
89.6 ⫾ 11.9 (60–100) 1139 ⫾ 451 (300–1,900) 778 ⫾ 343 (300–1,700) 9.9 ⫾ 6.3 (0.3–20) (n ⫽ 19)
71.4% (n ⫽ 15) 12 (8–16) 52.4% (n ⫽ 11)
Delay before dystonia after treatment, y†
17 (10–23)
Occurrence of fluctuations
57.1% (n ⫽ 12)
Delay before fluctuations after treatment, y*†
63.9 ⫾ 19.9 (20–90)
14 (9–28)
60.9% (n ⫽ 14) 10 (6–20) 52.2% (n ⫽ 12) 13 (7–20) 95.7% (n ⫽ 22) 5 (4–8)
*p ⬍ 0.05. †Data are expressed as the median (95% CI), otherwise they are mean ⫾ SD (range). UPDRS ⫽ Unified Parkinson’s Disease Rating Scale.
statistical package (SAS Institute, Cary, NC) was used for the analyses. The study was approved by the Ethical Committee of the Salpeˆtrie`re University Hospital Paris, France, and all patients gave their written informed consent.
Twenty-one out of 44 patients with EOPD had homozygous or compound heterozygous parkin mutations (table 1). Nine healthy sibs of the patients with parkin mutations had a single heterozygous mutation (table 1). Age at examination did not differ significantly among the three groups (47.5 years ⫾ 9.1 for affected parkin mutation carriers, 48.0 years ⫾ 8.0 for affected nonmutation carriers, and 47.9 years ⫾ 12.6 for unaffected single heterozygous carriers). Age at onset (31.1 years ⫾ 8.3 vs 34.9 years ⫾ 6.2) and disease duration (17.4 years ⫾ 7.8 vs 12.9 years ⫾ 7.7) were similar in patients with and without parkin mutations (table 2). Clinical characteristics RESULTS
112
Neurology 72
January 13, 2009
such as mentation, behavior, and mood (UPDRS I), activities of daily living (UPDRS II, “off” and “on” drug), parkinsonian motor disability (UPDRS III, “off” and “on” drug), Hoehn and Yahr Scale (UPDRS V), and the Schwab and England Scale (UPDRS VI) were also similar in patients with and without parkin mutations (table 2). However, the daily levodopa dose equivalents were significantly lower in patients with parkin mutations than in noncarriers (636 mg ⫾ 462 vs 1,139 mg ⫾ 451), although the duration of treatment was similar (13.2 years ⫾ 6.5 vs 9.9 years ⫾ 6.3). The UPDRS IV scores, which evaluate levodoparelated complications, did not differ between the two groups, but the delay before the appearance of levodopa-related fluctuations after treatment was significantly shorter in noncarriers than in patients with mutations (median: 5 years 95% CI [4 –18] vs 14 years [9 –28]). One affected non-mutation carrier but none of the patients with mutations had orthostatic hypotension, and none of the patients in either group had urinary incontinence. Patients from both groups had problems with sleep, such as sleep disruption, daytime sleepiness, or sleep behavior disorder (16/23 non-mutation carriers vs 10/21 parkin mutation carriers, data not shown), but the difference was not significant. Careful neurologic examination revealed no parkinsonism in the sibs with single heterozygous parkin mutations. Detailed neuropsychological examinations, including the MMSE, MDRS, Grober and Buschke test, WCST, TMT, FAB, and frontal score, did not detect any significant differences among the three groups (table 3). The results of the MDRS and the FAB were significantly different among the three groups, but after the Sidak correction for multiple comparisons, the difference was no longer significant. Four patients carrying two parkin mutations, one healthy heterozygous mutation carrier and one non-mutation carrier, had MDRS scores below the threshold of 136/144. The patient without parkin mutations (SPD-150) was a 55-year-old man, who started PD at age 33 and developed epilepsy during the 24-year evolution of his disease. The low cognitive efficiency of the patient (MMSE: 24/30 and MDRS: 110/144) was attributed to either his epileptic status or his carbamazepine treatment. The psychiatric profiles obtained with the MADRS, CPRS, and MINI tests were similar in patients with and without parkin mutations. Interestingly, patients tended to experience more depressive episodes and had higher MADRS scores (p ⫽ 0.2) than unaffected heterozygous parkin carriers (table 4). Additionally, six patients without and three pa-
Table 3
Neuropsychological evaluations of patients with Parkinson disease (PD) with and without parkin mutations and unaffected heterozygous parkin carriers (ⴙⴚ) PD patients with parkin mutations, n ⴝ 21 28.4 ⫾ 1.5 (25–30)
MMSE (/30) Mattis Dementia Rating Scale (/144)
138.5 ⫾ 4.1 (n ⫽ 19) (126–143)
PD patients without parkin mutations, n ⴝ 23
Healthy carriers of heterozygous parkin mutations, n ⴝ 9
28.4 ⫾ 2.1 (23–30)
29.5 ⫾ 0.9 (28–30)
139.0 ⫾ 6.7 (110–144)
142.4 ⫾ 1.4 (n ⫽ 8) (132–144)
Grober and Buschke 28.8 ⫾ 4 (n ⫽ 19) (20–34)
Free recall
30.2 ⫾ 5.0 (18–38)
35.1 ⫾ 6.2 (n ⫽ 8) (28–48)
Total recall
46.7 ⫾ 1.6 (n ⫽ 19) (43–48)
46.3 ⫾ 2.7 (37–48)
45.1 ⫾ 5.8 (n ⫽ 8) (31–48)
Delayed recall
10.9 ⫾ 2.5 (n ⫽ 19) (6–15)
11.8 ⫾ 2.4 (7–16)
13.3 ⫾ 1.3 (n ⫽ 8) (12–16)
Delayed total recall
15.7 ⫾ 0.6 (n ⫽ 19) (14–16)
Frontal score (/60)
15.8 ⫾ 0.5 (14–16)
15.6 ⫾ 1.1 (n ⫽ 8) (13–16)
54 ⫾ 5.6 (n ⫽ 18) (43–60)
54.7 ⫾ 5.3 (n ⫽ 21) (41–60)
54.4 ⫾ 10.3 (n ⫽ 8) (30–60)
17 ⫾ 1 (n ⫽ 18) (13–18)
16.5 ⫾ 1.5 (n ⫽ 22) (12–18)
17.9 ⫾ 0.4 (n ⫽ 8) (17–18)
Frontal Assessment Battery (/18) MCST
5 ⫾ 1.5 (n ⫽ 19) (1–6)
Categories Perseverations
1.7 ⫾ 2.8 (n ⫽ 18) (0–11)
5.3 ⫾ 1.1 (n ⫽ 21) (3–6)
6.0 (n ⫽ 8) (6–6)
1.3 ⫾ 2.3 (n ⫽ 21) (0–10)
0.6 ⫾ 1.4 (n ⫽ 8) (0–4)
Lexical fluency Category
20.7 ⫾ 5.1 (n ⫽ 18) (12–32)
Literal
12.8 ⫾ 4.1 (n ⫽ 18) (2–18)
21 ⫾ 7 (2–31)
23.3 ⫾ 6.0 (n ⫽ 8) (15–31)
13.2 ⫾ 4.9 (n ⫽ 22) (3–20)
13.1 ⫾ 3.6 (n ⫽ 8) (9–20)
MMSE ⫽ Mini-Mental State Examination; MCST ⫽ Modified Card Sorting Test.
tients with parkin mutations had MADRS scores ⬎15; five affected non-mutation carriers and three affected mutation carriers had scores ⬎18. However, the mean scores for both groups were not significantly different.
Table 4
This is one of the most detailed clinical studies to combine neurologic, neuropsychological, and psychiatric investigations to establish phenotype– genotype correlation in early-onset parkinsonism. We expected to find particular behavioral
DISCUSSION
Psychiatric evaluation of patients with Parkinson disease (PD) with and without parkin mutations and unaffected heterozygous parkin carriers PD patients with parkin mutations, n ⴝ 21
PD patients without parkin mutations, n ⴝ 23
MDE
43% (9)
48% (11)
MDE with melancholic features
10% (2)
9% (2)
Healthy carriers of heterozygous parkin mutations, n ⴝ 9
MINI 33% (3) 0
Dysthymia
6% (1)
4% (1)
0
Suicidality
14% (3)
4% (1)
0
Manic episode
14% (3)
4% (1)
0
Panic disorder
14% (3)
22% (5)
0
Agoraphobia
19% (4)
4% (1)
0
Social anxiety disorder
14% (3)
17% (4)
0
Obsessive-compulsive disorder
5% (1)
17% (4)
0
Posttraumatic stress disorder
10% (2)
4% (1)
0
Generalized anxiety disorder
33% (7)
26% (6)
11% (1)
CPRS (390)
34.1 ⫾ 29.2 (0–95)
42.7 ⫾ 27.7 (5–110)
17.1 ⫾ 15.3, n ⫽ 7 (4–44)
MADRS (30)
8.8 ⫾ 6.5 (0–21)
12.8 ⫾ 7.2 (2–28)
4 ⫾ 3.2, n ⫽ 7 (2–11)
None of the subjects were alcohol/substance abusers or had psychotic disorders, anorexia/bulimia nervosa, or antisocial personality disorders. MINI ⫽ Mini International Neuropsychiatric Interview; MDE ⫽ major depressive episode; CPRS ⫽ Comprehensive Psychopathological Rating Scale; MADRS ⫽ Montgomery-Asberg Depression Rating Scale. Neurology 72
January 13, 2009
113
or psychiatric pattern and maybe cognitive disorders in parkin mutation carriers because neuropathologic studies of these patients have shown severe generalized loss of dopaminergic neurons in the substantia nigra pars compacta13 that would greatly decrease dopaminergic efferents to the limbic and subcorticalfrontal and sensory-motor systems. The 21 patients with and 23 patients without parkin mutations were appropriately matched for age at onset (31.1 years ⫾ 8.3 vs 34.9 years ⫾ 6.2) and disease duration (17.4 years ⫾ 7.8 vs 12.9 years ⫾ 7.7). Detailed clinical evaluations including UPDRS I–VI, with and without treatment, did not detect any significant differences between the two groups. The daily doses of levodopa were also similar in parkin mutation and non-mutation carriers (528 mg ⫾ 439 vs 778 mg ⫾ 343), but after calculation of the daily levodopa dose equivalents,12 a significant difference between the groups was observed: parkin mutation carriers improved to the same extent as noncarriers (68.4% ⫾ 13.1 vs 70.2% ⫾ 18.6, data not shown) with significantly lower daily levodopa dose equivalents (636 mg ⫾ 462 vs 1,139 mg ⫾ 451). Previous studies have already shown in large cohorts of patients that parkin carriers had more levodopa induced dyskinesias, brisk reflexes, onset with foot dystonia,3,15-17 and a better response to low doses of medication, even after a long evolution of the disease,4,5 although none of these studies compared patients matched for age and disease duration. Our study confirms that parkin patients have very good responses to low doses of antiparkinsonian treatment. Furthermore, our parkin mutation carriers did not have more levodopa-induced complications than noncarriers, but the delay before the appearance of fluctuations after initiation of treatment was significantly longer in parkin mutation carriers than in patients without parkin mutations (14 years ⫾ 5.1 vs 5 years ⫾ 1.4). This delay in the development of levodopa-related dyskinesia was probably a result of the significantly lower doses of medication, but, curiously, the delay between the development of levodopa-related dyskinesias and dystonia did not differ between the two groups (10 vs 12 years and 13 vs 17 years, table 2). Interestingly, we show that dysautonomia, orthostatic hypotension, and urinary incontinence, which are absent in patients with parkin mutations, are also rare in the other early-onset PD cases. All unaffected single heterozygous mutation carriers, with ages ranging from 28 to 68, had normal detailed neurologic examinations. Statistically significant reductions in [18F]fluorodopa uptake have been observed in carriers of a single heterozygous parkin mutation15,18 compared to controls, and some were reported to have 114
Neurology 72
January 13, 2009
subtle extrapyramidal signs, such as resting tremor, reduced arm swing, or a mask-like face.3,4,15,18 It cannot be excluded that these healthy single heterozygous mutation carriers will develop parkinsonian symptoms in the future, but six of them are already more than 15 years older than the age at onset of their affected sibs. This observation supports the hypothesis that a single, even truncating, mutation may not be sufficient to trigger PD. These unaffected sibs of affected parkin mutation carriers are useful controls, however, for neuropsychological and psychiatric evaluations, because they share genetic and environmental factors but not the disease with their affected sib. Neuropsychological examinations did not reveal major differences in general cognitive efficiency (MMSE, Mattis DRS, episodic memory, Grober and Buschke test) or executive functions (frontal score and FAB) in patients with and without parkin mutations and unaffected heterozygous parkin carriers. This is consistent with and confirms, with more detailed neuropsychological evaluations than in previous studies, that cognitive function remains normal in the majority of patients with parkin mutations,4,5,19 even after 45 years of evolution.4 Additionally, we showed that patients with early-onset PD without parkin mutations do not have cognitive decline even after more than 30 years of disease evolution. However, it is interesting to note that only one patient without a parkin mutation, who also has epilepsy, four patients with parkin mutations, and one healthy single heterozygous parkin carrier had abnormal results on the MDRS (⬍136), although the groups did not differ significantly after the Sidak correction for multiple testing. Functional or structural abnormalities in the caudate nucleus have been postulated to play a role in frontal-subcortical cognitive impairment or dementia in patients with basal ganglia disease. Interestingly, several studies3,15,17,18,20-22 reported that the decrease in [18F]fluorodopa uptake in nigrostriatal terminals in the caudate nucleus is generally greater in patients with parkin mutations than in patients without this mutation. This pattern of nigrostriatal dysfunction might result in a different neuropsychological profile. However, this discussion remains speculative and the results must be confirmed by further studies with larger patient groups. Behavioral disorders, including anxiety and psychosis, panic attacks, depression, disturbed sexual behavior, and obsessive-compulsive disorders, have been reported with variable frequency in patients with parkin mutations,4,5,7-9 and were suggested to be a distinctive feature of parkin disease.4 Dopaminergic dysfunction in cortical areas which might contribute to the psychiatric disorders, as postulated for patients with idiopathic PD,23 has been demonstrated in 13 homozygous or compound heterozygous parkin mutation carriers using
PET with 11C-raclopride.22 However, our detailed psychiatric examinations including MINI, MADRS, and CPRS did not detect any significant qualitative or quantitative differences between parkin mutation carriers and noncarriers. Psychiatric manifestations were present in both groups of patients, but at a similar rate. They were less frequent, however, in the unaffected heterozygous parkin carriers, supporting the hypothesis that dopaminergic dysfunction or antiparkinsonian drugs might account for their greater frequency in patients. Nevertheless, they do not appear to be more frequent in parkin-related parkinsonism than in other early-onset patients. The large spectrum of parkin gene defects, which differ in their predicted consequences on the function of the protein, also raises the question of their role in the variability of the phenotype. Despite the large number of parkin mutation carriers included, the number of cases in each of the genotype specific groups was too small for a specific neurologic, neuropsychological, or psychiatric pattern to emerge. The results of this detailed clinical study indicate that patients with PD with parkin mutations are clinically indistinguishable from other early-onset patients. Severe generalized loss of dopaminergic neurons in the substantia nigra pars compacta in these patients is associated with an excellent response to low doses of levodopa-equivalent and delayed fluctuations. Their neuropsychological performance is not distinctive. Interestingly, behavioral problems and psychiatric symptoms, which have been considered to be markers of parkin disease, are observed at similar rates in both groups of early-onset patients and are more frequent in these patients than in unaffected heterozygous parkin mutation carriers. AUTHORS’ AFFILIATIONS From INSERM (E.L., S.L., Y.A., A.B.), UMR_S679 Neurologie & The´rapeutique Expe´rimentale, Paris; AP-HP (E.L., B.D., Y.A., A.B.), Pitie´-Salpeˆtrie`re Hospital, Department of the Nervous System Disorders, Paris; UPMC Univ Paris 06 (E.L., S.L., Y.A., A.B.), UMR_S679, Paris; University of Lyon I and INSERM UMR 864 and The Pierre Wertheimer Neurological Hospital (S.T., E.B.), Department of Neurology, Lyon; AP-HP (S.T.d.M.), Pitie´-Salpeˆtrie`re Hospital, Department of Public Health, Unit of de Biostatistics and Medical Information and Unit of Medical Research, Paris; UPMC Univ Paris 06 (S.T.d.M.), EA3974 Modelisation in Clinical Research, Paris; CEA (M.-J.R.), I2BM, Service Hospitalier Fre´de´ric Joliot, Orsay; CEA (P.R.), I2BM, URA-CEA-CNRS 2210, Orsay; CHU Henri Mondor (P.R.), AP-HP et Faculte´ de Me´decine Paris 12, Cre´teil; AP-HP (A.P.), Pitie´-Salpeˆtrie`re Hospital, Department of Psychiatry, Paris; INSERM UMR 610 Neuroanatomie Fonctionnelle du Comportement et de ses Troubles (B.D.), Paris; AP-HP (B.D.), Pitie´Salpeˆtrie`re Hospital, Centre de Re´fe´rence sur la Maladie de Pick, Paris; Inserm Avenir Group IFR 70 Behaviour (L.M.), Emotion and Basal Ganglia, Center of Clinical Investigation, Paris; Department of Clinical and Biological Neurosciences (P.P.), University Hospital of Grenoble; AP-HP (Y.A.), Pitie´-Salpeˆtrie`re Hospital, Clinical Investigation Center, Paris; and AP-HP (A.B.), Pitie´-Salpeˆtrie`re Hospital, Department of Genetics and Cytogenetics, Paris, France.
ACKNOWLEDGMENT The authors thank the patients and their families; Ce´line Chamayou and Aure´lie Funkiewiez for the neuropsychological data, Ce´cile Behar and Mirce´a Polosan for psychiatric advice, Merle Ruberg for critical reading of the manuscript, and the DNA and Cell Bank of the IFR 070 for sample preparation; and the nurses of the Centre d’Investigation Clinique who provided care for the patients.
APPENDIX The French Parkinson’s Disease Genetics Study Group: Y. Agid, A.-M. Bonnet, M. Borg, A. Brice, E. Broussolle, Ph. Damier, A. Deste´e, A. Du¨rr, F. Durif, E. Lohmann, M. Martinez, C. Penet, P. Pollak, O. Rascol, F. Tison, C. Tranchant, M. Ve´rin, F. Viallet, M. Vidailhet, and J.-M. Warter (deceased).
Received March 10, 2008. Accepted in final form June 27, 2008. REFERENCES 1. Kitada T, Asakawa S, Hattori N, et al. Mutations in the parkin gene cause autosomal recessive juvenile parkinsonism. Nature 1998;392:605–608. 2. Periquet M, Latouche M, Lohmann E, et al. Parkin mutations are frequent in patients with isolated early-onset parkinsonism. Brain 2003;126:1271–1278. 3. Khan NL, Brooks DJ, Pavese N, et al. Progression of nigrostriatal dysfunction in a parkin kindred: an [18F]dopa PET and clinical study. Brain 2002;125:2248–2256. 4. Khan NL, Graham E, Critchley P, et al. Parkin disease: a phenotypic study of a large case series. Brain 2003;126: 1279–1292. 5. Lohmann E, Periquet M, Bonifati V, et al. How much phenotypic variation can be attributed to parkin genotype? Ann Neurol 2003;54:176–185. 6. Lucking CB, Brice A. Alpha-synuclein and Parkinson’s disease. Cell Mol Life Sci 2000;57:1894–1908. 7. Tassin J, Durr A, de Broucker T, et al. Chromosome 6-linked autosomal recessive early-onset Parkinsonism: linkage in European and Algerian families, extension of the clinical spectrum, and evidence of a small homozygous deletion in one family. The French Parkinson’s Disease Genetics Study Group, and the European Consortium on Genetic Susceptibility in Parkinson’s Disease. Am J Hum Genet 1998;63:88–94. 8. Wu RM, Shan DE, Sun CM, et al. Clinical, 18F-dopa PET, and genetic analysis of an ethnic Chinese kindred with early-onset parkinsonism and parkin gene mutations. Mov Disord 2002;17:670–675. 9. Yamamura Y, Hattori N, Matsumine H, Kuzuhara S, Mizuno Y. Autosomal recessive early-onset parkinsonism with diurnal fluctuation: clinicopathologic characteristics and molecular genetic identification. Brain Dev 2000;22 suppl 1:S87–91. 10. West A, Periquet M, Lincoln S, et al. Complex relationship between Parkin mutations and Parkinson disease. Am J Med Genet 2002;114:584–591. 11. Klein C, Lohmann-Hedrich K, Rogaeva E, Schlossmacher MG, Lang AE. Deciphering the role of heterozygous mutations in genes associated with parkinsonism. Lancet Neurol 2007;6:652–662. 12. Thobois S. Proposed dose equivalence for rapid switch between dopamine receptor agonists in Parkinson’s disease: a review of the literature. Clin Ther 2006;28:1–12. 13. Lesage S, Magali P, Lohmann E, et al. Deletion of the parkin and PACRG gene promoter in early-onset parkinsonism. Hum Mutat 2007;28:27–32. Neurology 72
January 13, 2009
115
14.
Lesage S, Lohmann E, Tison F, Durif F, Durr A, Brice A. Rare heterozygous parkin variants in French early-onset Parkinson disease patients and controls. J Med Genet 2008;45:43–46. 15. Hilker R, Klein C, Ghaemi M, et al. Positron emission tomographic analysis of the nigrostriatal dopaminergic system in familial parkinsonism associated with mutations in the parkin gene. Ann Neurol 2001;49:367–376. 16. Ishikawa A, Tsuji S. Clinical analysis of 17 patients in 12 Japanese families with autosomal-recessive type juvenile parkinsonism. Neurology 1996;47:160–166. 17. Lucking CB, Chesneau V, Lohmann E, et al. Coding polymorphisms in the parkin gene and susceptibility to Parkinson disease. Arch Neurol 2003;60:1253–1256. 18. Khan NL, Horta W, Eunson L, et al. Parkin disease in a Brazilian kindred: manifesting heterozygotes and clinical follow-up over 10 years. Mov Disord 2005;20:479–484.
19.
20.
21.
22.
23.
Lucking CB, Durr A, Bonifati V, et al. Association between early-onset Parkinson’s disease and mutations in the parkin gene. N Engl J Med 2000;342:1560–1567. Portman AT, Giladi N, Leenders KL, et al. The nigrostriatal dopaminergic system in familial early onset parkinsonism with parkin mutations. Neurology 2001;56:1759–1762. Sawle GV, Leenders KL, Brooks DJ, et al. Doparesponsive dystonia: [18F]dopa positron emission tomography. Ann Neurol 1991;30:24–30. Scherfler C, Khan NL, Pavese N, et al. Striatal and cortical pre- and postsynaptic dopaminergic dysfunction in sporadic parkin-linked parkinsonism. Brain 2004;127:1332– 1342. Williams-Gray CH, Foltynie T, Brayne CE, Robbins TW, Barker RA. Evolution of cognitive dysfunction in an incident Parkinson’s disease cohort. Brain 2007;130: 1787–1798.
Learn. Earn. Network. 2009 AAN Annual Meeting: An Excellent Value • Learn about the latest scientific advances in neurology • Earn valuable CME credit and fulfill Maintenance of Certification requirements • Network with your peers at exciting social events all week long • Enjoy the convenience and value of all this and more—in just one meeting Early registration and hotel deadline is March 20, 2009. Register today at www.am.com/AM2009.
Calling All New and International Members! Don’t miss these FREE AAN Annual Meeting events designed just for you: • New Member Information Session Sunday, April 26 / 5:00 p.m. to 6:00 p.m. Learn about the AAN, its resources and benefits, and network with Academy leaders. • International Attendee Summit Monday, April 27 / 7:00 a.m. to 9:00 a.m. Meet Academy leaders and make your voice heard on matters most important to you. Learn more at www.aan.com/specialevents.
116
Neurology 72
January 13, 2009
Comorbidity delays diagnosis and increases disability at diagnosis in MS
R.A. Marrie, MD, PhD R. Horwitz, MD G. Cutter, PhD T. Tyry, PhD D. Campagnolo, MD T. Vollmer, MD
Address correspondence and reprint requests to Dr. Ruth Ann Marrie, Health Sciences Center, GF 543, 820 Sherbrook Street, Winnipeg, MB, R3A 1R9, Canada
[email protected]
ABSTRACT
Background: Comorbidity is common in the general population and is associated with adverse health outcomes. In multiple sclerosis (MS), it is unknown whether preexisting comorbidity affects the delay between initial symptom onset and diagnosis (“diagnostic delay”) or the severity of disability at MS diagnosis.
Objectives: Using the North American Research Committee on Multiple Sclerosis Registry, we assessed the association between comorbidity and both the diagnostic delay and severity of disability at diagnosis. In 2006, we queried participants regarding physical and mental comorbidities, including date of diagnosis, smoking status, current height, and past and present weight. Using multivariate Cox regression, we compared the diagnostic delay between participants with and without comorbidity at diagnosis. We classified participants enrolled within 2 years of diagnosis (n ⫽ 2,375) as having mild, moderate, or severe disability using Patient Determined Disease Steps, and assessed the association of disability with comorbidity using polytomous logistic regression.
Results: The study included 8,983 participants. After multivariable adjustment for demographic and clinical characteristics, the diagnostic delay increased if obesity, smoking, or physical or mental comorbidities were present. Among participants enrolled within 2 years of diagnosis, the adjusted odds of moderate as compared to mild disability at diagnosis increased in participants with vascular comorbidity (odds ratio [OR] 1.51, 95% CI 1.12–2.05) or obesity (OR 1.38, 95% CI 1.02–1.87). The odds of severe as compared with mild disability increased with musculoskeletal (OR 1.81, 95% CI 1.25–2.63) or mental (OR 1.62, 95% CI 1.23–2.14) comorbidity. Conclusions: Both diagnostic delay and disability at diagnosis are influenced by comorbidity. The mechanisms underlying these associations deserve further investigation. Neurology® 2009;72: 117–124 GLOSSARY BMI ⫽ body mass index; EDSS ⫽ Expanded Disability Status Scale; MS ⫽ multiple sclerosis; NARCOMS ⫽ North American Research Committee on Multiple Sclerosis; NINDS ⫽ National Institute of Neurological Disorders and Stroke; NMSS ⫽ National Multiple Sclerosis Society; OR ⫽ odds ratio; PDDS ⫽ Patient Determined Disease Steps.
Comorbidity is common in the general population1; it influences a broad range of health outcomes, including diagnostic delays and disease severity.2–5 Potentially, individuals with preexisting chronic illnesses are diagnosed earlier because of more frequent medical contacts.5 Conversely, preexisting disease may mask the symptoms of a new disease, negatively affect access to care, or prevent the consideration of etiologies other than the preexisting disease for new signs and symptoms.6 Comorbidity is common in multiple sclerosis (MS) at diagnosis.7 It is unknown, however, whether preexisting comorbidity affects the delay between symptom onset and diagnosis or the Supplemental data at www.neurology.org Editorial, page 108 e-Pub ahead of print on October 29, 2008, at www.neurology.org. From the Department of Medicine (R.A.M.), University of Manitoba, Winnipeg, Canada; Department of Medicine (R.H.), Stanford University, CA; Department of Biostatistics (G.C.), University of Alabama at Birmingham, AL; and Division of Neurology (T.T., D.C., T.V.), Barrow Neurological Institute, Phoenix, AZ. Supported partly by NIH, National Institute of Child Health and Human Development, Multidisciplinary Clinical Research Career Development Program Grant K12 HD04909. The NARCOMS Registry is supported by the Consortium of Multiple Sclerosis Centers. Disclosure: Author disclosures are provided at the end of the article. Copyright © 2009 by AAN Enterprises, Inc.
117
severity of disability at diagnosis. Previous work suggested that greater disability early in the disease course is a negative prognostic factor for long-term disability,8,9 and diseasemodifying therapies are most effective early in the disease course, when disability is mild.10 Thus, comorbidity-associated differences in time to diagnosis or disability at diagnosis would be important prognostically and therapeutically. Using the North American Research Committee on Multiple Sclerosis (NARCOMS) Registry, we aimed to determine the association between preexisting comorbidities or health behaviors on the degree of disability at MS diagnosis, as measured using Patient Determined Disease Steps (PDDS), and on the delay between symptom onset and diagnosis. We hypothesized that NARCOMS participants with preexisting comorbid illness or health behaviors would have more disability at diagnosis, after accounting for potential confounders. METHODS Study design and population. The NARCOMS Registry is a self-report registry for patients with MS,11 approved by the institutional review board at St. Joseph’s Hospital and Medical Center. At enrollment, participants provide demographic and clinical information, including date of birth, age at initial symptom onset, and age at and year of diagnosis. In October 2006, 18,000 active participants were eligible to receive the Fall Update Questionnaire. Per participant preference, we mailed a paper questionnaire (6,757) or e-mailed an invitation to complete the questionnaire online (11,243). To maximize response rates, we used a first-class postcard, two e-mail reminders, and a reminder in a lay publication provided quarterly to NARCOMS participants. We queried NARCOMS participants regarding physical and mental comorbidities. Questionnaire development is outlined on the Neurology® Web site at www.neurology.org. Participants indicated the presence or absence of a comorbidity and, if present, the year of diagnosis. Participants also reported past and present smoking status, height, and past and present body weight using questions from the Behavioral Risk Factor Surveillance Survey.12 Based on a literature review, we classified physical comorbidities as very likely to be accurately self-reported, moderately likely to be accurately self-reported, and least likely to be accurately reported.13,14 For the primary study, eligible participants were those living in the United States with complete data regarding date of birth, age at symptom onset, age at diagnosis, and age at symptom onset ⱖ16 years and ⬍60 years (n ⫽ 16,141). These criteria were intended to reduce heterogeneity in diagnostic testing and access to care, permit determination of the onset of the comorbidity in relation to MS onset and diagnosis, and limit heterogeneity due to differences in prognosis among persons with earlyor late-onset MS.15,16 118
Neurology 72
January 13, 2009
For this analysis, we grouped comorbidities into categories: physical, mental, vascular, autoimmune, visual, musculoskeletal, and gastrointestinal (see Neurology® Web site). Too few participants reported comorbid cancer for individual analysis. The validity of self-reported diagnoses is variable, being reasonably accurate for well-defined, chronic conditions requiring ongoing care but less accurate for diseases with less explicit diagnostic criteria.13,14 To account for this, we also created a category for any physical comorbidity, which included only those conditions very likely to be accurately self-reported based on literature review,13,14 including diabetes, hypertension, heart disease, breast cancer, colon cancer, rectal cancer, and lung cancer. Body mass index (BMI) was calculated from self-reported height and weight. Overweight was defined as a BMI ⱖ25 kg/m2 and ⬍30 kg/m2, and obesity was defined as a BMI ⱖ30 kg/m2.17 Participants were categorized according to the presence or absence of comorbidity at MS symptom onset and diagnosis.
Delay between symptom onset and diagnosis. For each participant, we calculated the delay between initial symptom onset and diagnosis (diagnostic delay) in years. Associations between the diagnostic delay and the presence or absence of comorbidity, and other covariates were analyzed with Wilcoxon or Kruskal–Wallis tests, or by using large-sample Z tests and CIs. We constructed a series of multivariate Cox proportional hazards models with diagnostic delay as the dependent variable and comorbidity categories as the independent variables.18 Covariates included sex, age at symptom onset, and year of symptom onset as independent variables.19 Year of symptom onset was categorized as ⱕ1980, 1981–1984, 1985–1989, 1990 –1994, 1995– 1999, or ⱖ2000. Age at symptom onset was categorized as ⱕ25, 25–39, or ⱖ40 years based on previous work regarding the diagnostic delay distribution in Denmark.20 We did not use income, education, or region of residence data for this model because there was sizeable potential for change in these variables between symptom onset, MS diagnosis, and registry enrollment. The proportional hazards assumption was tested using time-dependent covariates and graphical methods.21
Disability at diagnosis. To determine whether preexisting comorbidity influenced the degree of disability at diagnosis, we restricted the analysis to participants enrolled in the NARCOMS Registry within 2 years of diagnosis (n ⫽ 2,375); this allowed the use of demographic and disability data from the enrollment questionnaire. We assumed that changes in these variables were small or none in such a short time interval, based on previous examination of registry participants (data not shown).19 We report the characteristics of this subgroup and those of the whole sample, because this restriction truncates the distribution of diagnostic delay in patients with more recent symptom onset. PDDS is a self-reported surrogate measure of the Expanded Disability Status Scale (EDSS).22 Using PDDS, participants were classified as having mild (EDSS ⱕ3), moderate (EDSS 4 –5.5), or severe (EDSS ⱖ6) disability.23 Using polytomous logistic regression, we assessed the association between comorbidity and severity of disability at diagnosis after adjustment for potential confounders. Polytomous logistic regression is a technique used when the dependent variable is a categorical variable with greater than two classes but not necessarily monotonically ordered.24 The natural ordering of the data is lost, but all available data are used when calculating parameter estimates. Thus, we compared the odds of having moderate disability as compared with mild disability, and the odds of having severe disability as compared with mild disability. We report adjusted odds ratios (ORs) and 95% CIs as measures of association between comorbidity
Table 1
Demographic and clinical characteristics of the entire study population and of the disability at diagnosis subcohort
Entire population, n ⴝ 8,983
Disability at diagnosis subcohort, n ⴝ 2,375
Female
6,811 (75.8)
1,946 (81.9)
Male
2,172 (24.2)
429 (18.1)
White
8,442 (94.3)
2,237 (94.6)
Black
218 (2.4)
50 (2.1)
Other
293 (3.3)
78 (3.3)
Characteristic Sex, no. (%)
Race, no. (%)
Education, no. (%) <12 y
165 (1.9)
29 (1.2)
High school diploma
3,128 (35.0)
732 (30.9)
Associate’s or technical degree
1,448 (16.2)
422 (17.8)
Bachelor’s degree
2,395 (26.8)
681 (28.9)
Postgraduate degree
1,790 (20.1)
502 (21.2)
Married/cohabiting
6,049 (67.6)
1,636 (69.2)
Never married/divorced/widowed
2,904 (32.4)
730 (30.9)
Marital status, no. (%)
Annual income, no. (%) 888 (11.9)
<$15,000
165 (8.2)
$15,000–30,000
1,292 (17.3)
272 (13.4)
$30,000–50,000
1,605 (21.5)
391 (19.3)
$50,000–100,000
2,418 (32.3)
739 (36.5)
>$100,000
1,276 (17.1)
458 (22.6)
Private
6,601 (75.2)
1,989 (85.6)
Public
2,057 (23.4)
298 (12.8)
None
126 (1.4)
37 (1.6)
West
2,158 (24.0)
571 (24.0)
Midwest
2,333 (26.0)
623 (26.2)
South
2,335 (26.0)
622 (26.2)
East
2,157 (24.0)
Health insurance, no. (%)
Region, no. (%)
559 (23.5)
Age, mean (SD), y
52.7 (10.4)
47.1 (9.3)
Age at symptom onset, mean (SD), y
31.2 (9.0)
34.2 (9.1)
Age at diagnosis, mean (SD), y
38.2 (9.5)
41.2 (8.9)
Diagnostic delay, mean (SD), y
7.0 (7.4)
Disease duration, mean (SD), y
21.5 (11.2)
7.1 (7.5) 12.9 (8.0)
Patient Determined Disease Steps (categorized), no. (%) Mild
3,178 (35.4)
1,318 (55.5)
Moderate
1,067 (11.9)
350 (14.7)
Severe
4,738 (52.7)
707 (29.8)
and degree of disability at diagnosis. Potential confounders of the association between comorbid illness and disability progression considered were demographic characteristics, including age; sex; race; socioeconomic status, including highest education level reached, annual household income, and health insurance status;
and region of residence in the United States. Clinical characteristics considered as potential confounders were age at symptom onset and clinical course at onset; these were captured from the enrollment questionnaire. Education was included as indicator variables for ⬍12 years (reference group), high school diploma, associate’s degree or technical degree, bachelor’s degree, and postgraduate degree. Annual household income was included as indicator variables for ⬍$15,000 (reference group), $15,000 –30,000, $30,000 – 50,000, $50,000 –100,000, and ⬎$100,000. Insurance status was included as indicator variables for private, public (reference group), and none. Region of residence was included as indicator variables for West (reference group), Midwest, South, and East as defined by the US Census Bureau. Race was included as indicator variables for white (reference group), African-American, and other. Clinical course was defined as relapsing or progressive at onset. Delay from symptom onset to diagnosis was continuous. Age at symptom onset was categorized as ⱕ25, 25–39, and ⱖ40 years as indicated above, and included as indicator variables with ⱕ25 years as the reference group.
Of 16,141 participants meeting the inclusion criteria for the primary comorbidity study, 8,983 (55.7%) responded.7 Most respondents were white (94.3%) women (75.8%), with characteristics similar to those reported for the general MS population.25 Nonresponders were less likely to be white and tended to have lower socioeconomic status.7 Of the 8,983 participants in the primary study, 2,375 (26.4%) enrolled within 2 years of diagnosis (diagnostic delay subcohort). As compared with the entire cohort, this subcohort included a higher proportion of women, and participants with higher incomes and more private health insurance (table 1). As expected, the subcohort reported less disability and shorter disease duration than the entire cohort. The subcohort also included a higher proportion of participants with a relapsing course at onset (96.5% vs 88.5%) and a higher proportion currently receiving diseasemodifying therapy (84.2% vs 77.4%), but a lower proportion currently receiving immunosuppressive therapies (9.7% vs 14.3%). Members of the subcohort reported several comorbidities at diagnosis. This included 520 participants (22.6%) with vascular, 56 (2.5%) with visual, 277 (12.0%) with autoimmune, 345 (14.9%) with gastrointestinal, 265 (11.6%) with musculoskeletal, and 668 (29.4%) with mental comorbidities. More than 50% of participants were overweight or obese, 640 (27.8%) smoked, and 529 (23.0%) were exsmokers. RESULTS
Diagnostic delay. In the entire study sample, the mean (SD) diagnostic delay was 7.03 (7.4) years. In the disability at diagnosis subcohort, the mean diagnostic delay was 7.08 (7.5) years. The diagnostic delay decreased steadily with later year of symptom onset, from 10.6 (9.1) years in participants with onNeurology 72
January 13, 2009
119
Vascular ⫽ diabetes, hypertension, heart disease, peripheral vascular disease, hypercholesterolemia; Autoimmune ⫽ autoimmune thyroid disease, uveitis, systemic lupus erythematosus, rheumatoid arthritis, inflammatory bowel disease, Sjo ¨gren disease; Gastrointestinal ⫽ peptic ulcer disease, irritable bowel syndrome, inflammatory bowel disease, liver disease; Musculoskeletal ⫽ fibromyalgia, arthritis, rheumatoid arthritis, systemic lupus erythematosus, hip replacement, knee replacement; Visual ⫽ uveitis, cataracts, glaucoma; Mental ⫽ depression, anxiety, bipolar disorder, schizophrenia. p values and CIs based on large-sample Z tests. *p ⬍ 0.0001. †p ⫽ 0.06. ‡p ⫽ 0.01. § p ⫽ 0.3. NARCOMS ⫽ North American Research Committee on Multiple Sclerosis.
0.3 (⫺0.2 to 0.8)§ 4.2 (4.3) 3.9 (4.2) 2.0 (1.6–2.5)* 6.0 (6.3) 9.9 (8.9) Mental
14.6 (10.0)
6.3 (4.9–7.6)*
8.0 (6.8)
0.6 (⫺0.06 to 1.3)†
1.9 (0.7–3.2)‡ 5.6 (5.0)
4.5 (4.6) 3.9 (4.2)
3.7 (4.0) 4.1 (2.7–5.5)*
2.0 (1.4–2.7)* 8.2 (7.0)
10.1 (7.8)
6.2 (6.4)
6.0 (6.2)
4.5 (3.2–5.7)*
10.3 (7.5–13.0)*
14.6 (9.9)
20.2 (9.2) 9.9 (8.7)
10.1 (9.0) Gastrointestinal
Neurology 72
Visual
2.3 (1.6–3.0)*
1.4 (0.8–2.1)* 5.2 (4.7)
6.0 (5.2) 3.7 (4.0)
3.7 (4.1) 2.7 (2.0–3.4)* 8.9 (7.1)
10.7 (7.4)
6.2 (6.4)
6.0 (6.3)
6.3 (4.9–7.6)*
10.1 (8.8) Musculoskeletal
19.0 (10.6)
10.1 (9.0) Autoimmune
16.4 (9.8)
8.9 (7.4–10.5)*
4.7 (3.9–5.4)*
1.3 (0.9–1.8)* 4.9 (4.7) 3.6 (3.9) 3.4 (2.9–4.0)* 9.3 (7.6) 5.9 (6.1) 8.8 (7.5–10.1)* 18.7 (11.0) 9.9 (8.7) Vascular
Difference (95% CI) Affected, mean (SD) Unaffected, mean (SD) Comorbidity category
Unaffected, mean (SD)
Affected, mean (SD)
Difference (95% CI)
Unaffected, mean (SD)
Affected, mean (SD)
Difference (95% CI)
>40 y >25 y and <40 y <25 y
Age at symptom onset
Mean (SD) diagnostic delay in years among NARCOMS participants by age at symptom onset, and presence or absence of comorbidity at diagnosis of multiple sclerosis (n ⴝ 8,983) Table 2 120
January 13, 2009
set in 1980 or earlier to 1.12 (1.9) years in participants with onset in 2000 or later (p ⬍ 0.0001, Kruskal–Wallis test). The diagnostic delay was shorter in men (p ⫽ 0.009, Wilcoxon test) and persons with a later age at symptom onset (p ⬍ 0.0001, Kruskal– Wallis test). After stratification by age at symptom onset, the mean diagnostic delay was consistently longer in the presence of vascular, autoimmune, musculoskeletal, gastrointestinal, visual, and mental comorbidities (table 2). The mean diagnostic delay was 6.49 (7.0) years in nonsmokers, 6.58 (6.9) years in active smokers, and 9.13 (8.4) years in ex-smokers (p ⬍ 0.0001, Kruskal–Wallis test). These differences persisted after stratification by age at symptom onset (data not shown). The mean diagnostic delay was slightly longer in participants who were overweight [7.29 (7.6) years] and obese [7.29 (7.6) years] at diagnosis than in those who were not [6.65 (7.0) years] (p ⬍ 0.0001, Kruskal–Wallis test). For comorbidities and health behaviors, the difference in the diagnostic delay decreased substantially with increasing age at symptom onset. This evidence of effect modification was so great that it is not appropriate to provide a summary estimate of the mean diagnostic delay across the different ages of symptom onset; further, age at symptom onset was included as a stratification variable in multivariable models. Because of small numbers of participants reporting other races, the multivariable analysis was restricted to whites. In multivariable Cox proportional hazards models, all comorbidity categories, smoking, and obesity remained associated with a longer delay between symptom onset and diagnosis as demonstrated by hazard ratios less than 1 (table e-1). Comorbidity and disability at diagnosis. Because of small numbers of participants reporting other races, this analysis also was restricted to whites. After multivariable adjustment, participants with any physical comorbidity had increased odds of reporting moderate as compared with mild disability at diagnosis (OR 1.66, 1.18 –2.35). To assess dose–response, we included the count of comorbidities in the model as a continuous variable. For every additional physical comorbidity, the odds of moderate as compared with mild disability were 1.13 (1.03–1.23), and the odds of severe as compared with mild disability were 1.18 (1.08 –1.28). This is illustrated in the figure. Similarly, vascular, musculoskeletal, and mental comorbidities and obesity were associated with increased severity of disability at diagnosis (table 3). Other comorbidities and smoking were not associated with degree of disability at diagnosis (data not shown). To determine whether the association of comorbidity and disability was mediated by the diagnostic
Figure
Proportion of NARCOMS participants enrolled within 2 years of diagnosis who reported severe disability at diagnosis by number of physical comorbidities present
to the method of sample selection. Specifically, we 1) restricted the analysis to persons enrolled within a year of diagnosis, 2) expanded the analysis to include persons enrolled within 3 years of diagnosis, and 3) restricted the analysis to persons with a relapsing course at onset and enrolled within 2 years of diagnosis. Our results did not change apart from slight changes in the size of the CIs. The literature suggests that the delay in diagnosis of MS is affected by sex, age at symptom onset, and year of symptom onset.19,20 We also found that comorbidity is associated with longer delays in the time between symptom onset and diagnosis. Factors influencing the patient’s time from symptom onset to presentation for evaluation include perceived seriousness of the symptoms, socioeconomic status, and comorbidity.5,6 Factors influencing the time to specialist referral once the patient presents to a general practitioner include patient sex, socioeconomic status, comorbidity, and others. The findings regarding comorbidity and diagnostic delays for other conditions are variable, with both shorter and longer delays reported.6,27 The cancer literature suggests that for some cancers, comorbidity increases the odds of practitioner delay, possibly because of misattribution of new symptoms to the preexisting condition. Comorbidity, however, may make patients present sooner for evaluation. Our findings suggest that practitioners treating persons with chronic diseases should not attribute new neurologic signs or symptoms to existing conditions without careful consideration, but this must be balanced against DISCUSSION
NARCOMS ⫽ North American Research Committee on Multiple Sclerosis.
delay, we included that variable in the models. For any physical comorbidity, the odds of reporting moderate as compared with mild disability at diagnosis remained elevated (OR 1.49, 1.05–2.11). After including diagnostic delay in the model, the associations between disability and comorbidity or obesity were slightly attenuated (by 13%–37%),26 in some cases becoming marginally nonsignificant, suggesting that the associations could be partially mediated by the diagnostic delay. Sensitivity analyses. For the analyses using the disability of diagnosis subcohort, we performed additional analyses to assess the sensitivity of our results
Table 3
Odds ratios and 95% CIs for the association of comorbidity category at diagnosis and degree of disability at diagnosis in white NARCOMS participants enrolled within 2 years of diagnosis (n ⴝ 2,237) Adjusted*
Adjusted for diagnostic delay*
Moderate vs mild
Severe vs mild
Moderate vs mild
Severe vs mild
Comorbidity category
OR
95% CI
OR
95% CI
OR
95% CI
OR
95% CI
Vascular
1.51
1.12–2.05
1.06
0.77–1.44
1.32
0.97–1.80
0.87
0.63–1.20
Musculoskeletal
1.54
1.04–2.28
1.81
1.25–2.63
1.35
0.91–2.01
1.55
1.06–2.27
Mental
1.29
0.97–1.71
1.62
1.23–2.14
1.23
0.92–1.63
1.53
1.16–2.02
Overweight†
1.08
0.78–1.50
1.03
0.74–1.42
1.02
0.74–1.43
0.93
0.68–1.29
Obesity†
1.38
1.02–1.87
1.24
0.91–1.67
1.33
0.98–1.80
1.16
0.86–1.57
Vascular ⫽ diabetes, hypertension, heart disease, peripheral vascular disease, hypercholesterolemia; Musculoskeletal ⫽ fibromyalgia, arthritis, rheumatoid arthritis, systemic lupus erythematosus, hip replacement, knee replacement; Mental ⫽ depression, anxiety, bipolar disorder, schizophrenia; Overweight ⫽ body mass index (BMI) ⱖ25 kg/m2 and ⬍30 kg/m2; Obesity ⫽ BMI ⱖ30 kg/m2. Autoimmune, gastrointestinal, and visual comorbidities were not associated with disability at diagnosis. *All models adjusted for sex, age at symptom onset, year of symptom onset, and income. †Included in the same Cox model as indicator variables with normal weight as the reference group. NARCOMS ⫽ North American Research Committee on Multiple Sclerosis; OR ⫽ odds ratio. Neurology 72
January 13, 2009
121
overinvestigation. Further research is needed to better understand these issues. We also found that an increasing number of comorbidities, and obesity, vascular, musculoskeletal, and mental comorbidities were associated with a greater degree of disability at diagnosis. To our knowledge, other studies have not addressed the association of comorbidity and severity of disability at diagnosis, but some have focused on comorbidity and disability progression. One populationbased study reported that autoimmune disease was not associated with increased disability progression.28 Studies show conflicting results regarding the association of smoking and disability progression in MS.29,30 We found no association between smoking or autoimmune disease and more severe disability at diagnosis. Several possible explanations exist for the association of some comorbidities with more disability at diagnosis. First, a clinician could mistakenly attribute MS symptoms to a preexisting condition,27 increasing the time from symptom onset to diagnosis, and consequently disability at diagnosis. The attenuation of some of the observed associations when the statistical models included the diagnostic delay supports this idea, but this must be evaluated further. Second, comorbidities could act pathophysiologically to increase disease progression. Vascular conditions, for example, are associated with increased peripheral inflammation, and elevated cytokines are associated with increased brain atrophy.31 Third, having two or more comorbidities that independently cause similar impairments could additively or synergistically increase disability; in older adults, certain combinations of chronic diseases are associated with reduced mobility and functional status.32,33 Our finding of more severe disability at diagnosis in persons with comorbidity could be important for several reasons. Patients with comorbidity and increased disability at diagnosis might need or use more health care resources, might adhere differently to medication, or might respond differently to medication.34,35 They may also be at risk of undertreatment of their MS; a recent study found that persons with uncontrolled hypertension and unrelated comorbidities were less likely to have their hypertension addressed.36 We do not know whether disability progression is affected by comorbidity or whether these individuals should be treated differently. In diabetes, for example, aggressive management of comorbidities is increasingly emphasized to reduce diabetes-related complications.37,38 Similarly, in MS, comorbidities potentially could be treated more aggressively. Because of the burden of comorbidity in MS,7 these issues warrant further investigation. 122
Neurology 72
January 13, 2009
Our study has limitations. The NARCOMS Registry is a volunteer registry, but the characteristics of the registry population are similar to those reported for MS patients from the National Health Interview Survey.25 Nonresponders tended to be nonwhite, and much of our analysis was restricted to whites; thus, we do not know whether our findings generalize to other racial groups. Nonresponders had lower socioeconomic status. Although we did not identify any interactions between socioeconomic status and our findings, persons of lower socioeconomic status are at increased risk of comorbidity; so our findings may underestimate the impact of comorbidity in this portion of the population. Disability status was selfreported, but a substantial literature supports the ability of MS patients to report health status accurately,22,39 and the instruments used are validated.40 Persons with comorbidity could overreport their degree of disability at diagnosis, but the finding that only some of the comorbidities studied were associated with greater disability argues against this possibility. Study strengths include the large size of the cohort studied and the robustness of our results to sensitivity analyses. We analyzed our data with respect to a priori defined comorbidity categories to avoid excessive statistical comparisons, and we carefully considered confounding factors, including socioeconomic status. Comorbidity is associated with greater diagnostic delays, and increased disability at diagnosis in MS. Diagnostic delays may partially account for the association between comorbidity and disability at diagnosis. These findings need to be replicated in population-based cohorts. Future studies should evaluate the underlying mechanisms of these associations and determine how treatment of persons with MS and comorbidity can be optimized. AUTHOR CONTRIBUTIONS R.A.M. performed the statistical analysis.
DISCLOSURES R.A.M. has received research support from NIH, the Consortium of Multiple Sclerosis Centers, Serono, Berlex, Sanofi Aventis, and BioMS Technology Corporation. R.H. has received research support from NIH. Gary Cutter has served on Data and Safety Monitoring Committees for AntiSense Pharmaceuticals, Sanofi-Aventis, Bayhill Pharmaceuticals Inc., BioMS Pharmaceuticals, Enzo Pharmaceuticals Esai Pharmaceuticals, GlaxoSmithKline Pharmaceuticals, Genentech Pharmaceuticals, Glycomids Pharmaceuticals, Incyte Pharmaceuticals, Millennium Pharmaceuticals, Novartis Pharmaceuticals, Protein Design Labs, Roche Pharmaceuticals, National Heart, Lung, and Blood Institute, National Institute of Neurological Disorders and Stroke (NINDS), and the National Multiple Sclerosis Society (NMSS). He has served as a consultant to Amgen Pharmaceuticals, CibaVision, Millennium Pharmaceuticals, Consortium of MS Centers, MS-CORE and NMSS funded research group, Practice Based Research Network NYU, and Klein-Buendel Incorporated. T.T. has served as a consultant to Serono. T.V. has received support from NIH/NINDS U01NS45719-01A1, NIH ITN020AI, Abbott, Acorda,
Bayhill Therapeutics Inc., Biogen Idec, Genentech, Protein Design Laboratory, Serono, Pfizer, Teva, Novartis, and Berlex. D.C. has received research support from Accorda Therapeutics, Eli Lilly & Company, Avanir Pharmaceuticals, Bayer HealthCare Pharmaceuticals Inc., Bayhill Therapeutics Inc., Biogen Idec, BioMS Technology Corporation, Daiichi Sankyo Pharma Development, Genentech Inc., Genzyme Corporation, Merck Serono International SA, MSDx, LLC, National Institute of Allergy and Infectious Diseases/Immune Tolerance Network, Novartis, PDL, BioPharma Inc., Serono International SA, Pfizer, and Teva Neurosciences. She has served as a consultant or on speaker’s bureaus for ALZA, Kalos Therapeutics, Teva Neurosciences, Bayer HealthCare Pharmaceuticals, Serono-Pfizer, Xenoport, and Biogen Idec.
15. 16.
17.
Received March 16, 2008. Accepted in final form July 7, 2008. 18. REFERENCES 1. Hoffman C, Rice D, Sung HY. Persons with chronic conditions: their prevalence and costs. JAMA 1996;276:1473– 1479. 2. Braunstein JB, Anderson GF, Gerstenblith G, et al. Noncardiac comorbidity increases preventable hospitalizations and mortality among Medicare beneficiaries with chronic heart failure. J Am Coll Cardiol 2003;42:1226–1233. 3. Dunlop DD, Manheim LM, Sohn MW, Liu X, Chang RW. Incidence of functional limitation in older adults: the impact of gender, race, and chronic conditions. Arch Phys Med Rehabil 2002;83:964–971. 4. Gijsen R, Hoeymans N, Schellevis FG, Ruwaard D, Satariano WA, van den Bos GA. Causes and consequences of comorbidity: a review. J Clin Epidemiol 2001;54:661–674. 5. Fleming ST, Pursley HG, Newman B, Pavlov D, Chen K. Comorbidity as a predictor of stage of illness for patients with breast cancer. Med Care 2005;43:132–140. 6. Macdonald S, Macleod U, Campbell N, Weller D, Mitchell E. Systematic review of factors influencing patient and practitioner delay in diagnosis of upper gastrointestinal cancer. Br J Cancer 2006;94:1272–1280. 7. Marrie RA, Horwitz RI, Cutter G, Tyry T, Campagnolo D, Vollmer T. Comorbidity, socioeconomic status, and multiple sclerosis. Mult Scler 2008;14:1091–1098. 8. Bergamaschi R, Berzuini C, Romani A, Cosi V. Predicting secondary progression in relapsing-remitting multiple sclerosis: a Bayesian analysis. J Neurol Sci 2001;189:13–21. 9. Weinshenker BG, Rice GP, Noseworthy JH, Carriere W, Baskerville J, Ebers GC. The natural history of multiple sclerosis: a geographically based study, 3: multivariate analysis of predictive factors and models of outcome. Brain 1991;114:1045–1056. 10. Marrie RA, Cohen JA. Interferons in secondary progressive multiple sclerosis. In: Cohen JA, Rudick RA, eds. Multiple Sclerosis Therapeutics, 2nd ed. London: Martin Dunitz, 2003:347–362. 11. Consortium of Multiple Sclerosis Centers. NARCOMS Multiple Sclerosis Registry. Available at: www.mscare.org/ cmsc/CMSC-NARCOMS-Information.html. Accessed January 5, 2008. 12. Centers for Disease Control and Prevention. Behavioral Risk Factor Surveillance System Survey Questionnaire. Atlanta: US Department of Health and Human Services, 1995. 13. Katz JN, Chang LC, Sangha O, Fossel AH, Bates DW. Can comorbidity be measured by questionnaire rather than medical record review? Med Care 1996;34:73–84. 14. Okura Y, Urban LH, Mahoney DW, Jacobsen SJ, Rodeheffer RJ. Agreement between self-report questionnaires
19.
20.
21. 22.
23.
24. 25.
26.
27.
28.
29.
30.
31.
32.
33.
and medical record data was substantial for diabetes, hypertension, myocardial infarction and stroke but not for heart failure. J Clin Epidemiol 2004;57:1096–1103. Polliack ML, Barak Y, Achiron A. Late-onset multiple sclerosis. J Am Geriatr Soc 2001;49:168–171. Gusev E, Boiko A, Bikova O, et al. The natural history of early onset multiple sclerosis: comparison of data from Moscow and Vancouver. Clin Neurol Neurosurg 2002; 104:203–207. Centers for Disease Control and Prevention. Overweight and obesity: defining overweight and obesity. Available at: http://www.cdc.gov/nccdphp/dnpa/obesity/defining.htm. Accessed April 15, 2006. Collett D. Survival analysis. In: Chatfield C, Tanner M, Zidek J, eds. Modelling Survival Data in Medical Research, 2nd ed. Boca Raton: Chapman & Hall/CRC, 2003:1–2. Marrie RA, Cutter G, Tyry T, Hadjimichael O, Campagnolo D, Vollmer T. Changes in the ascertainment of multiple sclerosis. Neurology 2005;65:1066–1070. Esbjerg S, Keiding N, Koch-Henriksen N. Reporting delay and corrected incidence of multiple sclerosis. Stat Med 1999;18:1691–1706. Kleinbaum DG, Klein M. Survival Analysis: A SelfLearning Text. New York: Springer-Verlag, 1997. Hohol MJ, Orav EJ, Weiner HL. Disease steps in multiple sclerosis: a longitudinal study comparing disease steps and EDSS to evaluate disease progression. Mult Scler 1999;5:349–354. Marrie RA, Cutter G, Tyry T, Vollmer T, Campagnolo D. Does multiple sclerosis-associated disability differ between races? Neurology 2006;66:1235–1240. Agresti A. An Introduction to Categorical Data Analysis. New York: John Wiley & Sons, 1996. Collins JG. Prevalence of selected chronic conditions, United States, 1986-88. National Center for Health Statistics. Vital Health Statistics 1993;10. Shishehbor MH, Litaker D, Pothier CE, Lauer MS. Association of socioeconomic status with functional capacity, heart rate recovery, and all-cause mortality. JAMA 2006; 295:784–792. Bjerager M, Palshof T, Dahl R, Vedsted P, Olesen F. Delay in diagnosis of lung cancer in clinical practice. Br J Gen Pract 2006;56:863–868. Kirby S, Brown MG, Murray TJ, et al. Progression of multiple sclerosis in patients with other autoimmune diseases. Mult Scler 2005;11:S28–S29. Hernan MA, Jick SS, Logroscino G, Olek MJ, Ascherio A, Jick H. Cigarette smoking and the progression of multiple sclerosis. Brain 2005;128:1461–1465. Koch M, van Harten A, Uyttenboogaart M, De Keyser J. Cigarette smoking and progression in multiple sclerosis. Neurology 2007;69:1515–1520. Jefferson AL, Massaro JM, Wolf PA, et al. Inflammatory biomarkers are associated with total brain volume: the Framingham Heart Study. Neurology 2007;68:1032–1038. Fried LP, Bandeen-Roche K, Kasper JD, Guralnik JM. Association of comorbidity with disability in older women: the Women’s Health and Aging Study. J Clin Epidemiol 1999;52:27–37. Rijken M, van Kerkhof M, Dekker J, Schellevis FG. Comorbidity of chronic diseases: effects of disease pairs on physical and mental functioning. Qual Life Res 2005;14: 45–55.
Neurology 72
January 13, 2009
123
34.
Kalsekar ID, Madhavan SS, Amonkar MM, et al. Impact of depression on utilization patterns of oral hypoglycemic agents in patients newly diagnosed with type 2 diabetes mellitus: a retrospective cohort analysis. Clin Ther 2006;28:306–318. 35. Briesacher BA, Andrade SE, Fouayzi H, Chan KA. Comparison of drug adherence rates among patients with seven different medical conditions. Pharmacotherapy 2008;28:437–443. 36. Turner BJ, Hollenbeak CS, Weiner M, Ten Have T, Tang SSK. Effect of unrelated comorbid conditions on hypertension management. Ann Intern Med 2008;148:578–586. 37. Berlowitz DR, Ash AS, Hickey EC, Glickman M, Friedman R, Kader B. Hypertension management in patients-
with diabetes: the need for more aggressive therapy. Diabetes Care 2003;26:355–359. 38. Bakris GL, Weir MR, Shanifar S, et al. Effects of blood pressure level on progression of diabetic nephropathy: results from the RENAAL study. Arch Intern Med 2003; 163:1555–1565. 39. Ratzker PK, Feldman JM, Scheinberg LC, et al. Selfassessment of neurologic impairment in multiple sclerosis. J Neurol Rehabil 1997;11:207–211. 40. Marrie RA, Goldman MD. Validity of performance scales for disability assessment in multiple sclerosis. Mult Scler 2007;13:1176–1182.
AAN CME: Quick, Convenient, Smart. The AAN makes earning your continuing medical education (CME) credits more convenient than ever with a host of unparalleled CME opportunities. Whether you choose online, print, or classroom, rest assured that AAN CME is designed by top experts in neurology specifically to help you fulfill your Maintenance of Certification requirements and stay current in the field. Quick and convenient has never been so smart. • Neurology Online CME • Continuum: Lifelong Learning in Neurology® • Quintessentials® • Online CME Tracker • 2009 Winter Conference, Lake Buena Vista, FL • 2009 AAN Annual Meeting, Seattle, WA • More Learn more at www.aan.com/cme.
124
Neurology 72
January 13, 2009
Postmenopausal hormone therapy and subclinical cerebrovascular disease The WHIMS-MRI Study
L.H. Coker, PhD P.E. Hogan, MS N.R. Bryan, MD L.H. Kuller, MD K.L. Margolis, MD K. Bettermann, MD R.B. Wallace, MD Z. Lao, MD R. Freeman, MD M.L. Stefanick, PhD S.A. Shumaker, PhD For the Women’s Health Initiative Memory Study*
Address correspondence and reprint requests to Dr. Laura H. Coker, Division of Public Health Sciences, Wake Forest University Health Sciences, Medical Center Blvd., Winston-Salem, NC 27157
[email protected]
ABSTRACT
Objective: The Women’s Health Initiative Memory Study (WHIMS) hormone therapy (HT) trials reported that conjugated equine estrogen (CEE) with or without medroxyprogesterone acetate (MPA) increases risk for all-cause dementia and global cognitive decline. WHIMS MRI measured subclinical cerebrovascular disease as a possible mechanism to explain cognitive decline reported in WHIMS.
Methods: We contacted 2,345 women at 14 WHIMS sites; scans were completed on 1,424 (61%) and 1,403 were accepted for analysis. The primary outcome measure was total ischemic lesion volume on brain MRI. Mean duration of on-trial HT or placebo was 4 (CEE⫹MPA) or 5.6 years (CEE-Alone) and scans were conducted an average of 3 (CEE⫹MPA) or 1.4 years (CEEAlone) post-trial termination. Cross-sectional analysis of MRI lesions was conducted; general linear models were fitted to assess treatment group differences using analysis of covariance. A (two-tailed) critical value of ␣ ⫽ 0.05 was used. Results: In women evenly matched within trials at baseline, increased lesion volumes were significantly related to age, smoking, history of cardiovascular disease, hypertension, lower post-trial global cognition scores, and increased incident cases of on- or post-trial mild cognitive impairment or probable dementia. Mean ischemic lesion volumes were slightly larger for the CEE⫹MPA group vs placebo, except for the basal ganglia, but the differences were not significant. Women assigned to CEE-Alone had similar mean ischemic lesion volumes compared to placebo.
Conclusions: Conjugated equine estrogen– based hormone therapy was not associated with a significant increase in ischemic brain lesion volume relative to placebo. This finding was consistent within each trial and in pooled analyses across trials. Neurology® 2009;72:125–134 GLOSSARY 3MSE ⫽ modified Mini-Mental State Examination; BMI ⫽ body mass index; CEE ⫽ conjugated equine estrogen; CVD ⫽ cerebrovascular disease; HT ⫽ hormone therapy; MCI ⫽ mild cognitive impairment; MPA ⫽ medroxyprogesterone acetate; MRIQCC ⫽ MRI Quality Control Center; ROI ⫽ region of interest; WHIMS ⫽ Women’s Health Initiative Memory Study.
Dementia is increasing due to an aging population.1 Clinical stroke is often considered the first sign of cerebrovascular disease (CVD), but silent stroke and white matter lesions are more prevalent and commonly predate clinical CVD.2 Large cohort studies of older adults indicate that clinical and subclinical CVD increase the risk of cognitive decline3,4 and dementia.5 The Women’s Health Initiative Memory Study (WHIMS) reported that hormone therapy (HT)— conjugated equine estrogen 0.625 mg/d (CEE) with or without medroxyprogesterone 2.5 mg/d (MPA)—increased the risk for dementia6,7 and global cognitive decline in women age 65 and older.8,9 The WHI reported that both CEE ⫹ MPA and CEE-Alone were associated with an increased risk of clinical stroke.10-12 These results question whether HT increased the prevalence of subclinical CVD as a possible mechanism for cognitive decline in WHIMS. The main objective of the WHIMS Magnetic Resonance Imaging Study (WHIMS-MRI) was to See page 135 *See the appendix for details about the WHIMS centers. Authors’ affiliations are listed at the end of the article. The Women’s Health Initiative is funded by the National Heart, Lung, and Blood Institute of the National Institutes of Health, US Department of Health and Human Services. The Women’s Health Initiative Memory Study was funded in part by Wyeth Pharmaceuticals, Inc., St. Davids, PA. Disclosure: The authors report no disclosures. Copyright © 2009 by AAN Enterprises, Inc.
125
determine whether HT compared to placebo, randomly assigned at enrollment to the WHI hormone trials, was associated with volumetric subclinical CVD on post-trial brain MRI. The primary outcome measure of WHIMS MRI was total ischemic lesion volume on brain MRI. Lesion volumes in the basal ganglia and in the white and gray matter outside the basal ganglia were secondary outcomes. We hypothesized that women randomized to CEE-based HT would have significantly increased ischemic lesion volumes on brain MRI. A companion article by Resnick at al.13 reports whether regional and total brain volumes differed by WHI treatment assignment. METHODS The ClinicalTrials.gov Identifier for this study is NCT00000611.
Design of WHI and WHIMS. The WHI randomized HT trials were designed to evaluate postmenopausal HT and prevention of disease, with coronary heart disease the primary outcome. Hip fracture, other fractures, other cardiovascular disease, and endometrial, colorectal, and other cancers were secondary outcomes.14 A geographically diverse group of postmenopausal women aged 50 –79 were enrolled in the parallel CEE⫹MPA (n ⫽ 16,608) or CEE-Alone (n ⫽ 10,739 women with prior hysterectomy) randomized placebocontrolled trials.15 Ancillary to the WHI, WHIMS followed women from these two trials to examine the effects of postmenopausal CEE-based HT on the risk of all-cause dementia and global cognitive functioning in healthy women who were 65–79 years old at WHIMS enrollment. Of age-eligible participants who were approached, 4,532 (92.6%) consented to participate in the WHIMS CEE⫹MPA trial and 2,947 (92.1%) consented to participate in the WHIMS CEE-Alone Trial. The WHIMS study designs, eligibility criteria, and recruitment procedures were described previously.16 WHIMS participants underwent cognitive screening with the modified Mini-Mental State Examination (3MSE)17 at enrollment and annually. Women who scored below an education-adjusted cutpoint on the 3MSE underwent a more comprehensive cognitive evaluation,18 assessment of mood, and a neuropsychiatric examination. If probable dementia was suspected, the participant underwent laboratory tests and non-contrast x-ray computerized tomography of the brain. Final classifications (normal, mild cognitive impairment [MCI], or probable dementia) were adjudicated at the WHIMS Clinical Coordinating Center at the Wake Forest University Health Sciences. The WHI hormone trials were terminated earlier than planned due to significantly more non-cognitive adverse events associated with HT compared to placebo.10,11 As a consequence, the ancillary WHIMS CEE⫹MPA and CEE-Alone trials were also prematurely terminated.6-9 Post-trial follow-up of participants continued annually for cognitive assessment and adjudication of MCI and probable dementia in the WHIMS Extension Study. Design of WHIMS-MRI. WHIMS-MRI was designed to compare neuroradiologic outcomes among women with an average on trial HT exposure of 4.0 years (CEE⫹MPA) or 5.6 years (CEE-Alone). Brain scanning was conducted, on average, 8.02 126
Neurology 72
January 13, 2009
years (CEE⫹MPA) or 7.97 (CEE-Alone) years following randomization and 3.0 years (CEE⫹MPA) or 1.4 years (CEEAlone) after termination of the WHIMS HT trials. WHIMS-MRI was conducted in 14 of the 39 WHIMS clinical sites. Following Institutional Review Board approval, WHIMS-MRI recruitment began in January 2005 and was completed in April 2006. Women who provided consent underwent screening to determine if they were acceptable candidates for MRI. Exclusion criteria included the presence of items that would make the MRI procedure hazardous (pacemakers, prohibited medical implants, and foreign bodies); shortness of breath or inability to lie flat; and conditions that can be exacerbated by stress (anxiety panic disorders, claustrophobia) severe enough to preclude MRI. WHIMS participants who enrolled in WHIMS-MRI tended to be younger and more highly educated than those who did not; they also had been relatively healthier at the time they enrolled in the WHI and tended to have experienced less on-trial decline in global cognition.19
MRI protocol. The MRI scanning protocol was developed by investigators at the MRI Quality Control Center (MRIQCC) in the Department of Radiology, University of Pennsylvania. Scanning and reading were done by individuals masked to treatment assignment. The scans were conducted by a standardized protocol; scanning pulse sequences were performed in the following order: • Series one: three-plane gradient echo localizer for positioning. • Series two: sagittal T1-weighted spin echo midslice image to demonstrate anatomic location of the AC/PC for slice angle and slice position. • Series three: oblique axial spin density/T2-weighted spin echo images from the vertex to skull base parallel to the AC/PC plane. • Series four: oblique axial FLAIR T2-weighted spin echo images matching slice positions in series three. • Series five: oblique axial three-dimensional T1-weighted gradient echo images from the vertex to the skull base parallel to the AC/PC plane. The field of view was 22 cm and the acquisition matrix was 256 ⫻ 256 for series three, four, and five. Scans obtained in series three, four, and five were used for analyses of ischemic lesions. The WHIMS-MRI technologist at each site immediately reviewed all scans for protocol compliance and technical problems. Imaging data were transmitted by an encrypted DICOM image transfer mechanism to the permanent study archive via the Web.
WHIMS-MRI primary outcome measure. The primary outcome measure for WHIMS-MRI is total ischemic lesion volume, measured in cubic centimeters, detected from a standardized imaging and reading protocol. Secondarily, ischemic lesion volumes in the basal ganglia and the cerebral white and gray matter outside the basal ganglia were measured.20 Ischemic lesion volume defined and identified by this methodology generally corresponds to what has been called small vessel ischemic disease (ischemic white matter disease and lacunar infarctions). Basically unknown before the advent of MRI, this process is now accepted as a non-necrotic, ischemic effect on myelin that is secondary to the effects of aging, hypertension, and other small vessel pathologic processes of the brain.21,22 The earliest reports of this vasculopathy were by anecdotal observations which were quickly superseded by semiquantitative, human observer scoring sys-
Figure
Enrollment and flow of participants through Women’s Health Initiative Memory Study–MRI
*Multiple reasons were given, so totals are not additive.
tems such as those used in the Cardiovascular Health Study23 and the Rotterdam Study.24 While these systems are strongly correlated with each other in terms of rank order, their scores are not directly comparable, and these manual systems have limited reproducibility and restricted dynamic ranges.23,25 The methodology for detecting and quantifying ischemic tissue used in this report reflects the evolution in image processing from manual human observer to automatic, quantitative computerized digital image analytical techniques that are not only correlated with human observers and the semiquantitative scoring systems, but are very reproducible and offer a greater dynamic range.26,27 Our methodology classifies all brain tissue into either normal or ischemic gray or white matter and assigns the tissue type to each of 92 anatomic regions of interest (ROIs) of the cerebrum. These are organized in an anatomically hierarchal system that is collapsed into three ROIs for this analysis: the separate volumes of cerebral gray and white matter and basal ganglia (gray and white matter). Abnormal matter from these three ROIs form the basis for our secondary outcomes; the sum of the three is the primary outcome.
Statistical analysis. Patricia E. Hogan, MS, Department of Biostatistical Sciences, Division of Public Health Sciences, Wake Forest University School of Medicine, was the statistician. Characteristics at the time of randomization to the WHI HT trials and cognitive measures during follow-up were compared between women assigned to HT vs placebo using 2 tests. Because the distributions of the ischemic lesion volumes were highly skewed, a logarithmic transformation was employed and back-transformed (geometric) means are presented.
Analysis of covariance was used to assess the associations of outcomes with cognitive risk factors and other participant characteristics after adjustment for clinical site, age at randomization, time from randomization to MRI, intracranial volume, trial, and treatment assignment (active vs placebo). Analysis of covariance was also used to contrast means of our outcome measures between treatment groups, with adjustments for clinical site, age at randomization, time from randomization to MRI scan, intracranial volume, and trial. Additional models also involved adjustment for other baseline factors associated with ischemic brain lesion volumes and factors affecting participation in WHIMSMRI (education, smoking, body mass index [BMI], prior cardiovascular disease, hypertension, and diabetes). We also examined subgroups of women defined by baseline 3MSE scores and compliance to study medications for evidence of differential treatment effects. Women were analyzed according to WHI randomization assignment. Treatment arm differences are presented within trial and pooled across trials. Because WHIMS-MRI was designed to examine potential mechanisms underlying the adverse findings of the WHIMS trials, Type I error was not controlled for multiple comparisons: individual inferences were conducted according to two-sided significance levels of 0.05. Differential effects of treatment were examined for subgroups defined by baseline factors including age, BMI, smoking, prior cardiovascular disease, diabetes, and hypertension; however, none reached significance. RESULTS Of the 2,859 active WHIMS participants in the 14 WHIMS-MRI sites at the beginning of the study, 2,345 (82.0%) were contacted about enrollment in WHIMS-MRI, and 1,527 (65.1%) consented. MRI scans were completed on 1,424 participants (61%) and 1,403 met study criteria for central reading. Of 1,403 participants, 883 had previously been enrolled in the CEE⫹MPA trial (436 active and 447 placebo) and 520 in the CEE-Alone trial (257 active and 263 placebo) (figure). There were no within or across trial treatment assignment differences in WHIMS-MRI participants at WHI baseline by age, years since menopause, education, ethnicity, smoking status, or alcohol consumption (table 1). At the time of MR scanning, the age range across both trials was 71 to 88 years and the mean age was 78.5 years. There were no differences between the active and placebo groups within or across the two trials by BMI, hypertension, prior cardiovascular disease (stroke, myocardial infarction, and/or history of angina, angioplasty, or coronary artery bypass graft surgery), diabetes, prior use of hormones, or baseline 3MS scores (table 2). In participants who showed MCI or probable dementia during or after WHIMS, there were increased but not significantly more cases of probable dementia in the treatment arm vs placebo of the CEE⫹MPA trial, and these findings parallel results previously reported for the overall WHIMS trial.6 Age and hypertension were the strongest predictors of increasing geometric mean ischemic lesion Neurology 72
January 13, 2009
127
Table 1
WHI baseline demographic, socioeconomic status, and lifestyle characteristics of WHIMS MRI women by treatment assignment: Frequency (%) WHIMS-MRI (n ⴝ 883) CEEⴙMPA
WHIMS-MRI (n ⴝ 520)
CEEⴙMPA (n ⴝ 436)
Placebo (n ⴝ 447)
CEE-Alone (n ⴝ 257)
Placebo (n ⴝ 263)
65–69
229 (53)
231 (52)
124 (48)
131 (50)
70–74
153 (35)
153 (34)
94 (37)
92 (35)
54 (12)
63 (14)
39 (15)
40 (15)
Variable
CEE-Alone
Age, y, n (%)
75ⴙ Years since menopause, n (%) <15 15–24 25ⴙ
92 (22)
105 (24)
17 (8)
29 (12)
241 (57)
253 (58)
79 (36)
88 (38)
90 (21)
78 (18)
123 (56)
117 (50)
Education, n (%)
16 (4)
22 (5)
16 (6)
High school/GED
88 (20)
98 (22)
70 (27)
69 (26)
Some college
179 (41)
165 (37)
95 (37)
118 (45)
College graduate
152 (35)
160 (36)
76 (30)
67 (26)
18 (4)
16 (4)
13 (5)
17 (6)
405 (93)
409 (92)
224 (88)
238 (91)
13 (3)
21 (5)
18 (7)
Never
257 (59)
252 (57)
149 (59)
Former
159 (37)
172 (39)
96 (38)
Current
18 (4)
17 (4)
9 (3)
None
163 (37)
190 (42)
131 (51)
135 (52)
<1/d
208 (48)
205 (46)
107 (42)
104 (40)
1ⴙ/d
65 (15)
52 (12)
19 (7)
22 (8)
White, non-Hispanic Other
CEEⴙMPA vs placebo
CEE-Alone vs placebo
HT vs no HT
0.75
0.92
0.81
0.40
0.19
0.18
0.54
0.18
0.97
0.39
0.06
0.88
0.77
0.49
0.67
0.18
0.86
0.35
9 (3)
Ethnicity, n (%) Black/African American
p Values
7 (3)
Smoking status, n (%) 148 (56) 99 (38) 15 (6)
Alcohol intake
WHIMS ⫽ Women’s Health Initiative Memory Study; CEE ⫽ conjugated equine estrogen; MPA ⫽ medroxyprogesterone acetate; HT ⫽ hormone therapy.
volumes across all three measures (p ⬍ 0.0001) (table 3). Smoking was significantly associated while prior cardiovascular disease was of borderline significance across the three measures. Neither history of high cholesterol nor diabetes had a significant association with brain lesions, possibly due to small numbers of cases. Increased BMI was significantly associated with smaller ischemic lesion volumes except for those in the basal ganglia. There was no significant association between global cognition (3MSE scores) at WHI baseline and cerebral ischemic lesion volumes on MRI. However, women who scored highest on the 3MSE obtained nearest to the MRI scan date had the smallest mean ischemic lesion volumes across all measures (p ⬍ 0.05). We also examined changes in 3MSE scores over time, from WHI baseline to the WHIMS cogni128
Neurology 72
January 13, 2009
tive testing session nearest the MRI scan. Women whose test scores improved by at least 2 points had the smallest lesions while women with worsening scores were more likely to have the largest lesions. This pattern was significant within the basal ganglia (p ⬍ 0.01). Women who developed on or post-trial MCI or probable dementia had larger ischemic lesion volumes, which were significant across all measures. Table 4 shows there were no significant differences between geometric mean lesion volumes by treatment arm within or across trials, after initial adjustment for clinical site, age at randomization, time from randomization to MRI scan, and intracranial volume. After controlling for education, smoking, BMI, prior cardiovascular disease, hypertension, and diabetes (plus the previously mentioned variables), the results were very similar. Mean ischemic volumes
Table 2
WHI baseline clinical characteristics of WHIMS-MRI women by treatment assignment: Frequency (%)
p Values
Variable
WHIMS-MRI CEEⴙMPA
WHIMS-MRI CEE-Alone
CEEⴙMPA
CEE-Alone
Placebo
Placebo
Body mass index, kg/m2, n (%) <25
138 (32)
156 (35)
60 (24)
64 (24)
25–29
165 (38)
155 (35)
98 (38)
108 (41)
30–34
90 (21)
90 (20)
62 (24)
58 (22)
35ⴙ
42 (10)
45 (10)
35 (14)
32 (12)
Hypertension status, n (%) None Current/controlled* Current/uncontrolled
234 (54)
240 (54)
122 (47)
139 (53)
55 (13)
49 (11)
51 (20)
54 (21)
147 (34)
158 (35)
84 (33)
70 (27)
Prior cardiovascular disease, n (%) No History of stroke History of other cardiovascular disease†
413 (95)
425 (95)
236 (92)
5 (1)
3 (1)
4 (2)
21 (5)
17 (4)
18 (7)
20 (8)
No
420 (96)
425 (95)
241 (94)
242 (92)
Yes
16 (4)
22 (5)
16 (6)
21 (8)
Prior use of hormone therapy, n (%)
Yes
340 (78)
346 (77)
131 (51)
127 (48)
96 (22)
101 (23)
126 (49)
136 (52)
19 (4)
20 (8)
19 (7)
Baseline 3MSE, n (%) <90 90–94 95–100
22 (5)
CFE-Alone vs placebo
HT vs no HT
0.71
0.85
0.81
0.71
0.30
0.75
0.43‡
0.93‡
0.56
0.36
0.44
0.23
0.84
0.54
0.59
0.45
0.76
0.36
0.13
0.70
0.36
239 (91)
2 (0)
Diabetes, n (%)
No
CFEⴙMPA vs placebo
66 (15)
80 (18)
50 (20)
58 (22)
347 (80)
341 (78)
185 (73)
184 (71)
On- or post-trial MCI/probable dementia, n (%) No
416 (95)
435 (97)
247 (96)
251 (95)
Yes
20 (5)
12 (3)
10 (4)
12 (5)
*Measured to be less than 140/90 mm Hg. †Other cardiovascular disease defined as myocardial infarction, angina, percutaneous transluminal angioplasty, or coronary artery bypass grafting. ‡Based on Fisher exact test. WHIMS ⫽ Women’s Health Initiative Memory Study; CEE ⫽ conjugated equine estrogen; MPA ⫽ medroxyprogesterone acetate; HT ⫽ hormone therapy; 3MSE ⫽ modified Mini-Mental State Examination; MCI ⫽ mild cognitive impairment.
were slightly but not significantly larger in the CEE⫹MPA group compared to placebo, except in the basal ganglia. We also examined subgroups of women defined separately by baseline 3MSE scores and compliance to study medication; no differential treatment effects were detected. In this analysis of older women enrolled in the WHIMS HT trials, who were evenly matched within trials on demographic and clinical characteristics, we found no marked differences
DISCUSSION
within or across trials by treatment assignment (CEE⫹MPA or CEE-Alone vs placebo) on total ischemic lesion volume—the primary outcome of the WHIMS-MRI study. Further, we found no differences in ischemic lesion volumes in the basal ganglia or in the white and gray matter outside the basal ganglia. We hypothesized that women assigned to HT would have significantly larger ischemic lesion volumes, providing a mechanism to explain the earlier reports from WHIMS that HT increased the Neurology 72
January 13, 2009
129
Table 3
Geometric mean (SE) ischemic brain lesion volumes (cubic centimeters) by demographics, socioeconomic status, lifestyle, and clinical characteristics*
Variable
No. (%)
White and gray matter (outside of basal ganglia), mean (SE)
Basal ganglia, mean (SE)
Total lesion volume, mean (SE)
3.78 (0.12)
0.54 (0.01)
4.21 (0.14)
Age, y* 65–69
715 (51)
70–74
492 (35)
5.00 (0.20)
0.74 (0.02)
5.61 (0.22)
75ⴙ
196 (14)
6.47 (0.40)
0.98 (0.03)
7.36 (0.47)
⬍0.0001
⬍0.0001
⬍0.0001
p Value Education
63 (5)
3.80 (0.42)
0.56 (0.03)
4.25 (0.48)
325 (23)
4.47 (0.22)
0.66 (0.02)
5.01 (0.25)
Some college
557 (40)
4.54 (0.17)
0.66 (0.01)
5.07 (0.19)
College grad
455 (33)
4.62 (0.19)
0.69 (0.01)
5.19 (0.22)
p Value
0.61
0.62
0.60
Ethnicity Black/African American white, non-Hispanic Other
64 (5) 1276 (91) 59 (4)
p Value
5.35 (0.61)
0.65 (0.04)
5.86 (0.68)
4.49 (0.11)
0.69 (0.01)
5.03 (0.13)
4.21 (0.49)
0.65 (0.04)
4.73 (0.56)
0.40
0.98
0.50
Smoking status Never
806 (58)
4.15 (0.13)
0.62 (0.01)
4.65 (0.14)
Former
526 (38)
4.95 (0.19)
0.72 (0.01)
5.53 (0.21)
Current
59 (4)
6.07 (0.69)
0.88 (0.05)
6.88 (0.79)
p Value
0.001
0.004
0.001
Alcohol intake None
619 (44)
4.49 (0.16)
0.64 (0.01)
5.01 (0.18)
<1/d
624 (45)
4.56 (0.16)
0.69 (0.01)
5.12 (0.18)
1ⴙ/d
158 (11)
4.39 (0.31)
0.68 (0.02)
4.93 (0.35)
p Value
0.92
0.57
0.90
Prior cardiovascular disease No
1313 (94)
4.43 (0.11)
0.66 (0.01)
4.96 (0.12)
History of stroke
14 (1)
5.17 (1.21)
0.62 (0.08)
5.67 (1.35)
History of other cardiovascular disease†
76 (5)
5.99 (0.60)
0.85 (0.04)
6.67 (0.68)
p Value
0.05
0.12
0.05
Hypertension at WHI enrollment No
735 (52)
3.97 (0.13)
0.59 (0.01)
4.44 (0.14)
Yes, controlled‡
209 (15)
4.32 (0.26)
0.62 (0.02)
4.81 (0.30)
Yes, uncontrolled
459 (33)
5.61 (0.23)
0.83 (0.02)
6.31 (0.26)
⬍0.0001
⬍0.0001
⬍0.0001
1153 (84)
4.48 (0.11)
0.65 (0.01)
5.01 (0.13)
223 (16)
4.91 (0.29)
0.76 (0.02)
5.53 (0.33)
p Value History of high cholesterol No Yes p Value
0.23
0.07
0.19
Diabetes No
1328 (95)
Yes
75 (5)
p Value
4.46 (0.11)
0.67 (0.01)
5.00 (0.12)
5.43 (0.55)
0.62 (0.03)
6.01 (0.62)
0.12
0.92
0.14 —Continued
130
Neurology 72
January 13, 2009
Table 3
Continued
Variable
No. (%)
White and gray matter (outside of basal ganglia), mean (SE)
Basal ganglia, mean (SE)
Total lesion volume, mean (SE)
2
Body mass index, kg/m <25
418 (30)
4.90 (0.21)
0.72 (0.02)
5.49 (0.24)
25–29
526 (37)
4.63 (0.18)
0.68 (0.01)
5.18 (0.20)
30–34
300 (21)
4.25 (0.21)
0.61 (0.02)
4.76 (0.24)
35ⴙ
154 (11)
3.75 (0.27)
0.59 (0.02)
4.20 (0.30)
p Value
0.04
0.12
0.04
Baseline 3MS <90 90–94 95–100
80 (6)
4.14 (0.41)
0.65 (0.03)
4.64 (0.47)
254 (18)
4.64 (0.26)
0.61 (0.02)
5.14 (0.29)
1057 (76)
4.50 (0.12)
0.68 (0.01)
5.05 (0.14)
p Value
0.71
0.39
0.76
3MS closest to MRI <90 90–94 95–100
67 (5)
5.60 (0.60)
0.80 (0.04)
6.25 (0.68)
174 (12)
5.20 (0.35)
0.81 (0.03)
5.86 (0.40)
1162 (83)
4.36 (0.11)
0.64 (0.01)
4.88 (0.13)
p Value
0.04
0.01
0.03
Change in 3MSE (MRI – baseline) <ⴚ1
450 (32)
4.90 (0.20)
0.77 (0.02)
5.52 (0.23)
>ⴚ1 and <1
432 (31)
4.50 (0.19)
0.63 (0.01)
5.01 (0.21)
>2
509 (37)
4.18 (0.16)
0.60 (0.01)
4.68 (0.19)
p Value
0.08
0.002
0.06
On- or post-trial MCI/probable dementia No Yes
1349 (96) 54 (4)
p Value
4.44 (0.10)
0.65 (0.01)
4.97 (0.12)
6.61 (0.79)
0.97 (0.06)
7.34 (0.89)
0.006
0.006
0.007
*After adjustment for trial, clinical site, age, time from randomization to MR scan, intracranial volume, education, smoking, body mass index, prior cardiovascular disease, hypertension, and diabetes. *Not adjusted for age. †Other cardiovascular disease defined as myocardial infarction, angina, percutaneous transluminal angioplasty, or coronary artery bypass grafting. ‡Measured to be less than 140/90 mm Hg. WHI ⫽ Women’s Health Initiative; 3MSE ⫽ modified Mini-Mental State Examination; MCI ⫽ mild cognitive impairment.
risk of all-cause dementia6,7 and global cognitive decline in women age 65 years and older,8,9 and the association of HT and clinical stroke reported by the WHI.10-12 Instead, we found no evidence that HT exposure increased ischemic lesion volumes. Increased lesion volumes on MRI demonstrated internal validity; they were associated with several vascular risk factors including increased age, smoking, history of cardiovascular disease, and hypertension at WHI baseline. Increased lesion volumes were also associated with development of on- or post-trial MCI and probable dementia in WHIMS and lower 3MSE scores at time of MRI scanning. History of stroke and diabetes were not associated with ischemic brain lesions in WHIMS-MRI due to few participants with these conditions. We found a protective
association between increasing BMI and ischemic lesion volumes. Although recent MRI findings demonstrate the association of obesity with generalized and regional brain atrophy,28 and increased white matter volumes,29 the underlying mechanisms (diabetes with associated elevated insulin levels and metabolic syndrome) are not prevalent in the WHIMS-MRI cohort. Although the literature records a link between midlife obesity and risk for future dementia,30 as age increases, overweight and moderate obesity may have a neutral31 or protective effect on cognition in women,32,33 but not in men.34 The association of BMI (low and high) with brain MRI findings and the underlying mechanisms require further study. Compared to the general age-matched US population, WHIMS-MRI participants have lower prevaNeurology 72
January 13, 2009
131
Table 4
Geometric mean (SE) ischemic brain lesion volumes (cubic centimeters) by treatment assignment* WHIMS-MRI CEEⴙMPA trial
WHIMS-MRI CEE-Alone trial
p Values
CEEⴙMPA, (n ⴝ 436)
Placebo, (n ⴝ 447)
CEE-Alone, (n ⴝ 257)
Placebo, (n ⴝ 263)
CEEⴙMPA vs placebo
CEE-Alone vs placebo
HT vs No HT
White and gray matter (outside of basal ganglia)
4.57 (0.19)
4.18 (0.17)
4.84 (0.26)
4.77 (0.26)
0.21
0.87
0.27
Basal ganglia
0.65 (0.01)
0.65 (0.01)
0.68 (0.02)
0.71 (0.02)
0.89
0.66
0.88
Total brain lesion volume
5.10 (0.21)
4.70 (0.20)
5.41 (0.30)
5.35 (0.29)
0.25
0.91
0.32
Variable
Consistency of treatment effects of CEE⫹MPA vs CEE-Alone. White and gray matter (outside of basal ganglia): p ⫽ 0.53; basal ganglia: p⫽ 0.66; total lesion volume: p ⫽ 0.5. *After adjustment for trial, clinical sites, age, time from randomization to MR scan, intracranial volume, education, smoking, body mass index, prior cardiovascular disease, hypertension, and diabetes. WHIMS ⫽ Women’s Health Initiative Memory Study; CEE ⫽ conjugated equine estrogen; MPA ⫽ medroxyprogesterone acetate; HT ⫽ hormone therapy.
lence rates of risk factors for cardiovascular disease and stroke, including diabetes, hypercholesterolemia, or uncontrolled hypertension35,36 and history of cigarette smoking.37 Thus, HT may not have the same adverse effects in individuals with few or no vascular risk factors compared to those at high risk.38 Post hoc calculations indicate that WHIMS-MRI provided 80% statistical power (based on a two-sided test with ␣ ⫽ 0.05) to detect a mean difference of 19% in the primary outcome, total ischemic lesion volume, between the active and placebo arms. The mean we observed was a nonsignificant difference of only 5.7%. To put this in perspective, we examined the cross-sectional age increases in total lesion volume observed in the placebo group. The geometric means were 3.93 cm3 and 6.03 cm3 for the 70 –74 and 80⫹ age groups, respectively, after adjusting for trial differences. This corresponds to a 53% change over 10 years or approximately 5.3% per year. Therefore, the mean treatment difference detected (5.7%) was equivalent to about a 1-year age increase, whereas we were powered to detect an age increase of approximately 3.5 years. Additional analyses of sequences of pre-scan cognitive data provided reassurance that our null findings could not be attributed to differential enrollment between women who had been assigned to HT vs placebo. We repeated our principal analyses using propensity scores adjustment to account for potential differential enrollment, 39 using all factors identified as having independent associations with enrollment21 which resulted in essentially identical results. Our sample may have under-represented the participants most at risk for hormone-related ischemic brain lesions and cognitive decline. In addition, our sample size may be too small to measure the small differences in blood pressure by treatment assignment in the WHIMS trial. 132
Neurology 72
January 13, 2009
The cross-sectional analysis in WHIMS-MRI limits the ability to measure the frequency of new ischemic brain lesions associated with prior assignment to HT compared to placebo. The difference in the frequency of new lesions may be masked by the much higher underlying prevalence of lesions at trial entry. This suggests the need for repeat scanning with longitudinal analyses of ischemic lesions. Furthermore, the detrimental cognitive effects of HT reported in WHIMS may be primarily related not to vascular disease but to neurodegeneration, which would not have been detected by MRI data analysis limited to ischemic lesion volume measurements. HT may promote glial or neuronal damage, for instance, by triggering the accumulation of amyloid or other toxic compounds causing primarily cell death and brain atrophy. Perhaps older women have fewer estrogen receptors in the brain, and thus the vasodilatory effect of estrogen seen in younger women may be lost. This could cause a decrease in cerebral blood flow, which, according to the neurovascular hypothesis of Alzheimer disease, could cause long-term accumulation of amyloid and glial/neuronal loss.40 A companion article by Resnick et al.13 reports the association of HT and brain volumes by MRI. AUTHORS’ AFFILIATIONS From the Division of Public Health Sciences (L.H.C., P.E.H., S.A.S.), Wake Forest University School of Medicine, Winston-Salem, NC; Department of Radiology (N.R.B.), University of Pennsylvania, Philadelphia; Department of Epidemiology (L.H.K.), University of Pittsburgh, PA; Health Partners Research Foundation (K.L.M.), Minneapolis, MN; Department of Neurology (K.B.), Penn State College of Medicine, Hershey, PA; Department of Epidemiology (R.B.W.), University of Iowa College of Public Health, Iowa City; Eastman Kodak (Z.L.), Rochester, NY; Department of Obstetrics and Gynecology and Women’s Health (R.F.), Albert Einstein College of Medicine, Montefiore Medical Center, Bronx, NY; and Stanford Prevention Research Center (M.L.S.), Department of Medicine, Stanford University, CA.
APPENDIX WHIMS-MRI Clinical Centers. Albert Einstein College of Medicine, Bronx, NY: Sylvia Wassertheil-Smoller, Mimi Goodwin, Richard DeNise, Michael Lipton, James Hannigan; Medical College of Wisconsin, Milwaukee: Jane Morley Kotchen, Diana Kerwin, John Ulmer, Steve Censky; Stanford Center for Research in Disease Prevention, Stanford University, CA: Marcia L. Stefanick, Sue Swope, Anne Marie SawyerGlover; The Ohio State University, Columbus: Rebecca Jackson, Rose Hallarn, Bonnie Kennedy; University of California at Davis, Sacramento: John Robbins, Sophia Zaragoza, Cameron Carter, John Ryan; University of California at Los Angeles: Lauren Nathan, Barbara Voigt, Pablo Villablanca, Glen Nyborg; University of Florida, Gainesville/Jacksonville: Marian Limacher, Sheila Anderson, Mary Ellen Toombs, Jeffrey Bennett, Kevin Jones, Sandy Brum, Shane Chatfield; University of Iowa, Davenport: Jennifer Robinson, Candy Wilson, Kevin Koch, Suzette Hart; University of Massachusetts, Worcester: Judith Ockene, Linda Churchill, Douglas Fellows, Anthony Serio; University of Minnesota, Minneapolis: Karen Margolis, Cindy Bjerk, Chip Truwitt, Margaret Peitso; University of Nevada, Reno: Robert Brunner, Ross Golding, Leslie Pansky; University of North Carolina, Chapel Hill: Carol Murphy, Maggie Morgan, Mauricio Castillo, Thomas Beckman; University of Pittsburgh, PA: Lewis Kuller, Pat McHugh, Carolyn Meltzer, Denise Davis. WHIMS-MRI Clinical Coordinating Center. Wake Forest University Health Sciences, Winston-Salem, NC: Sally Shumaker, Mark Espeland, Laura Coker, Jeff Williamson, Debbie Felton, LeeAnn Andrews, Steve Rapp, Claudine Legault, Maggie Dailey, Julia Robertson, Patricia Hogan, Sarah Jaramillo, Pam Nance, Cheryl Summerville, Josh Tan.
6.
7.
8.
9.
10.
11.
WHIMS-MRI Quality Control Center. University of Pennsylvania, Philadelphia: Nick Bryan, Christos Davatzikos, Lisa Desiderio. WHIMS-MRI Working Group. Wake Forest University Health Sciences, Winston-Salem, NC: LeeAnn Andrews; University of Pennsylvania, Philadelphia: Nick Bryan; Wake Forest University Health Sciences, Winston-Salem, NC: Laura Coker; Wake Forest University Health Sciences, Winston-Salem, NC: Mark Espeland; Wake Forest University Health Sciences, WinstonSalem, NC: Debbie Felton; University of Pittsburgh, PA: Lew Kuller; University of Minnesota, Minneapolis: Karen Margolis; University of Minnesota, Minneapolis: Anne Murray; National Institute on Aging, Baltimore, MD: Susan Resnick; Wake Forest University Health Sciences, Winston-Salem, NC: Sally Shumaker; Wake Forest University Health Sciences, WinstonSalem, NC: Jeff Williamson. US NIH. National Institute on Aging, Bethesda, MD: Neil Buckholtz, Susan Molchan, Susan Resnick; National Heart, Lung, and Blood Institute, Bethesda, MD, Jacques Rossouw, Linda Pottern.
12.
13.
14.
15.
16.
Received June 13, 2008. Accepted in final form August 28, 2008.
REFERENCES 1. Brookmeyer R, Gray S, Kawas C. Projections of Alzheimer’s disease in the United States and the public health impact of delaying disease onset. Am J Public Health 1998;88:1337–1342. 2. Bryan RN, Cai J, Burke G, et al. Prevalence and anatomic characteristics of infarct-like lesions on MR images of middle-aged adults: The Atherosclerosis Risk in Communities Study. Am J Neuroradiol 1999;20:1273–1280. 3. Mosley TH Jr, Knopman DS, Catellier DJ, et al. Cerebral MRI findings and cognitive functioning: The Atherosclerosis Risk in Communities Study. Neurology 2005;64: 2056–2062. 4. Prins ND, van Dijk EJ, den Heijer T, et al. Cerebral smallvessel disease and decline in information processing speed, executive function and memory. Brain 2005;128:2034– 2041. 5. Kuller LH, Lopez OL, Newman A, et al. Risk factors for dementia in the Cardiovascular Health Cognition Study. Neuroepidemiology 2003;22:13–22.
17. 18.
19.
20.
21. 22.
Shumaker SA, Legault C, Rapp SR, et al. The effects of estrogen plus progestin on the incidence of dementia and mild cognitive impairment in postmenopausal women: The Women’s Health Initiative Memory Study: a randomized controlled trial. JAMA 2003;289:2651–2662. Shumaker SA, Legault C, Kuller L, et al. Conjugated equine estrogens and incidence of probable dementia and mild cognitive impairment in postmenopausal women: Women’s Health Initiative Memory Study. JAMA 2004; 291:2947–2958. Rapp SR, Espeland MA, Shumaker SA, et al. Effect of estrogen plus progestin on global cognitive function in postmenopausal women: The Women’s Health Initiative Memory Study: a randomized controlled trial. JAMA 2003;289:2663–2672. Espeland MA, Rapp SR, Shumaker SA, et al. Conjugated equine estrogens and global cognitive function in postmenopausal women: Women’s Health Initiative Memory Study. JAMA 2004;291:2959–2968. Rossouw JE, Anderson GL, Prentice RL, et al. Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results from the Women’s Health Initiative randomized controlled trial. JAMA 2002;288:321–333. Effects of conjugated equine estrogen in postmenopausal women with hysterectomy: The Women’s Health Initiative randomized controlled trial. JAMA. 2004;291:1701–1712. Wassertheil-Smoller S, Hendrix SL, Limacher M, et al. Effect of estrogen plus progestin on stroke in postmenopausal women. JAMA 2003;289:2673–2684. Resnick SM, Espeland MA, Jaramillo SA, et al., for the Women’s Health Initiative Memory Study. Postmenopausal hormone therapy and regional brain volumes: the WHIMS-MRI Study. Neurology 2009;72:135–142. Anderson GL, Manson J, Wallace R, et al. Implementation of the Women’s Health Initiative study design. Ann Epidemiol 2003;13:S5–17. Design of the Women’s Health Initiative clinical trial and observational study. The Women’s Health Initiative Study Group Controlled Clin Trials. 1998;19:61-109. Shumaker SA, Reboussin BA, Espeland MA. The Women’s Health Initiative Memory Study (WHIMS): a trial of the effect of estrogen therapy in preventing and slowing the progression of dementia. Controlled Clin Trials 1998;19:604–621. Teng EL, Chui HC. The Modified Mini-Mental State (3MS) examination. J Clin Psychiatry 1987;48:314–318. Morris JC, Heyman A, Mohs RC, et al. The Consortium to Establish a Registry for Alzheimer’s Disease (CERAD). Part I. Clinical and neuropsychological assessment of Alzheimer’s disease. Neurology 1989;39:1159–1165. Jaramillo SA, Felton D, Andrews L, et al. Enrollment in a brain magnetic resonance study: Results from the Women’s Health Initiative Memory Study Magnetic Resonance Imaging Study (WHIMS-MRI). Acad Radiol 2007;14:603–612. Lao Z, Shen D, Liu D, Jawad AF, Melhem ER, Launer LJ, Bryan RN, Davatzikos C. Computer-assisted segmentation of white matter lesions in 3D MR images using support vector machine. Acad Radiol 2008;15:300–313. Pantoni L, Garcia JH. Pathogenesis of leukoaraiosis: A review. Stroke 1997;28:652–659. Moody DM, Bell MA, Challa VR. Features of the cerebral vascular pattern that predict vulnerability to perfusion or oxygenation deficiency: an anatomic study. Am J Neuroradiol 1990;11:431–439.
Neurology 72
January 13, 2009
133
23.
24.
25.
26.
27.
28. 29.
30.
31.
134
Bryan RN, Manolio TA, Schertz LD, et al. A method for using MR to evaluate the effects of cardiovascular disease on the brain: The Cardiovascular Health Study. Am J Neuroradiol 1994;15:1625–1633. de Groot JC, de Leeuw FE, Oudkerk M, et al. Periventricular cerebral white matter lesions predict rate of cognitive decline. Ann Neurol 2002;52:335–341. Mantyla R, Erkinjuntti T, Salonen O, et al. Variable agreement between visual rating scales for white matter hyperintensities on MRI. Comparison of 13 rating scales in a poststroke cohort. Stroke 1997;28:1614–1623. Anbeek P, Vincken KL, van Osch MJ, Bisschops RH, van der Grond J. Automatic segmentation of different-sized white matter lesions by voxel probability estimation. Med Image Anal 2004;8:205–215. Yoshita M, Fletcher E, Harvey D, et al. Extent and distribution of white matter hyperintensities in normal aging, MCI, and AD. Neurology 2006;67:2192–2198. Jagust W. What can imaging reveal about obesity and the brain? Curr Alzheimer Res 2007;4:135–139. Haltia LT, Viljanen A, Parkkola R, et al. Brain white matter expansion in human obesity and the recovering effect of dieting. J Clin Endocrinol Metab 2007;92:3278–3284. Kivipelto M, Ngandu T, Fratiglioni L, et al. Obesity and vascular risk factors at midlife and the risk of dementia and Alzheimer disease. Arch Neurol 2005;62:1556–1560. Bagger YZ, Tanko LB, Alexandersen P, Qin G, Christiansen C. The implications of body fat mass and fat distribution for cognitive function in elderly women. Obes Res 2004;12: 1519–1526.
Neurology 72
January 13, 2009
32.
33.
34.
35.
36.
37.
38. 39.
40.
Brownbill RA, Ilich JZ. Cognitive function in relation with bone mass and nutrition: cross-sectional association in postmenopausal women. BMC Womens Health 2004;4:2. Kuo HK, Jones RN, Milberg WP, et al. Cognitive function in normal-weight, overweight, and obese older adults: An analysis of the Advanced Cognitive Training for Independent and Vital Elderly cohort. J Am Geriatr Soc 2006; 54:97–103. Elias MF, Elias PK, Sullivan LM, Wolf PA, D’Agostino RB. Obesity, diabetes and cognitive deficit: The Framingham Heart Study. Neurobiol Aging. 2005;26 Suppl 1:11–16. Arnold AM, Psaty BM, Kuller LH, et al. Incidence of cardiovascular disease in older Americans: the Cardiovascular Health Study. J Am Geriatr Soc 2005;53:211–218. Goff DC Jr, Bertoni AG, Kramer H, et al. Dyslipidemia prevalence, treatment, and control in the Multi-Ethnic Study of Atherosclerosis (MESA): gender, ethnicity, and coronary artery calcium. Circulation 2006;113: 647–656. Centers for Disease Control and Prevention (CDC). Cigarette smoking among adults–United States, 2004. Morb Mortal Weekly Rep 2005;54:1121-1124. Petitti DB. Hormone replacement therapy for prevention: more evidence, more pessimism. JAMA 2002;288:99–101. Rosenbaum PR, Rubin DB. The central role of the propensity score in observational studies for causal effects. Biometrika 1983;70:41–55. Kalaria RN. The role of cerebral ischemia in Alzheimer’s disease. Neurobiol Aging 2000;21:321–330.
Postmenopausal hormone therapy and regional brain volumes The WHIMS-MRI Study
S.M. Resnick, PhD M.A. Espeland, PhD S.A. Jaramillo, MS C. Hirsch, MD M.L. Stefanick, PhD A.M. Murray, MD, MSc J. Ockene, PhD Med C. Davatzikos, PhD For the Women’s Health Initiative Memory Study*
Address correspondence and reprint requests to Dr. Susan M. Resnick, Laboratory of Personality and Cognition, Biomedical Research Center/04B317, 251 Bayview Blvd., Baltimore, MD 21224
[email protected]
ABSTRACT
Objectives: To determine whether menopausal hormone therapy (HT) affects regional brain volumes, including hippocampal and frontal regions.
Methods: Brain MRI scans were obtained in a subset of 1,403 women aged 71– 89 years who participated in the Women’s Health Initiative Memory Study (WHIMS). WHIMS was an ancillary study to the Women’s Health Initiative, which consisted of two randomized, placebo-controlled trials: 0.625 mg conjugated equine estrogens (CEE) with or without 2.5 mg medroxyprogesterone acetate (MPA) in one daily tablet. Scans were performed, on average, 3.0 years post-trial for the CEE ⫹ MPA trial and 1.4 years post-trial for the CEE-Alone trial; average on-trial follow-up intervals were 4.0 years for CEE ⫹ MPA and 5.6 years for CEE-Alone. Total brain, ventricular, hippocampal, and frontal lobe volumes, adjusted for age, clinic site, estimated intracranial volume, and dementia risk factors, were the main outcome variables.
Results: Compared with placebo, covariate-adjusted mean frontal lobe volume was 2.37 cm3 lower among women assigned to HT (p ⫽ 0.004), mean hippocampal volume was slightly (0.10 cm3) lower (p ⫽ 0.05), and differences in total brain volume approached significance (p ⫽ 0.07). Results were similar for CEE ⫹ MPA and CEE-Alone. HT-associated reductions in hippocampal volumes were greatest in women with the lowest baseline Modified Mini-Mental State Examination scores (scores ⬍90).
Conclusions: Conjugated equine estrogens with or without MPA are associated with greater brain atrophy among women aged 65 years and older; however, the adverse effects are most evident in women experiencing cognitive deficits before initiating hormone therapy. Neurology® 2009;72: 135–142 GLOSSARY 3MS ⫽ modified Mini-Mental State Examination; AC-PC ⫽ anterior commissure–posterior commissure; BMI ⫽ body mass index; CEE ⫽ conjugated equine estrogens; GM ⫽ gray matter; MPA ⫽ medroxyprogesterone acetate; WHI ⫽ Women’s Health Initiative; WHIMS ⫽ Women’s Health Initiative Memory Study; WM ⫽ white matter.
Supplemental data at www.nehurology.org
The Women’s Health Initiative Memory Study (WHIMS) trials1-4 showed that conjugated equine estrogens (CEE) alone or combined with medroxyprogesterone acetate (MPA) increase dementia risk and adversely affect global cognition in women aged 65 years or older. In view of these results and findings that hormone therapy (HT) increases the risk of clinical stroke in older women,5,6 we examined potential mechanisms through MRI scans of former WHIMS participants. HT may influence clinical outcomes through vascular changes or effects on regional brain volumes, including neuronal architecture and synaptic density. Increases in gray matter7,8 and hippocampal volumes,7,9,10 hippocampal blood flow,11 and temporal glucose metabolism12,13 have been reported in observational studies of estrogen users. Effects of HT on frontal function
See page 125 *See the appendix for details about the WHIMS-MRI Clinical Centers. Authors’ affiliations are listed at the end of the article. The Women’s Health Initiative and WHIMS-MRI Study are funded by the National Heart, Lung, and Blood Institute of the NIH, US Department of Health and Human Services. WHIMS was funded in part by Wyeth Pharmaceuticals, Inc., St. Davids, PA. S.M.R. is supported by the Intramural Research Program, National Institute on Aging, NIH. Disclosure: Author disclosures are provided at the end of the article. Copyright © 2009 by AAN Enterprises, Inc.
135
also were reported.14,15 These generally small studies were conducted in cohorts with average ages less than 70 years. However, the increased risk of stroke and thromboembolic disease associated with HT in older women6,16 may offset potential neurocognitive benefits, resulting in a net increase in dementia risk. We investigated whether global and regional brain volumes differ post-trial between older women who had been randomly assigned to HT or placebo during the Women’s Health Initiative (WHI) HT trials. We focused on whether total brain, hippocampal, and frontal lobe volumes, measured by MRI, differed by treatment assignment. A companion article17 reports findings on lesion volume, the primary outcome of the WHIMS-MRI study. Analysis of global cognitive function in the WHIMS trials uncovered only one factor moderating the adverse HT effects: baseline cognitive function at WHI enrollment. Women with lower baseline scores on the modified Mini-Mental State Examination (3MS)18 had significantly greater on-trial HTassociated declines in cognitive function than women with higher scores.4 Thus, a second goal is to determine whether a low 3MS score at baseline is associated with a greater HT effect on global and regional brain volumes. Finally, we tested whether HT benefits women with the lowest vascular lesion burden, as suggested by animal models,19 by comparing HT effects on brain volumes in women with the lowest ischemic lesion volume to the remaining women. METHODS This trial is registered at ClinicalTrials.gov with identifier NCT00000611. WHIMS was an ancillary study to WHI, which consisted of parallel placebo-controlled randomized clinical trials of 0.625 mg/d CEE therapy alone in women after hysterectomy and in combination with 2.5 mg/d MPA in women with a uterus. WHIMS design, eligibility criteria, and recruitment procedures have been described.20 Participants were recruited from 39 of the 40 clinical centers participating in the WHI CEE-Alone or CEE ⫹ MPA clinical trials. To be eligible for WHIMS, women were aged 65–79 years at enrollment and were free of dementia.20 Written informed consent was obtained; institutional review boards for participating institutions and the NIH approved the protocols and consent forms. The WHIMS CEE ⫹ MPA trial terminated earlier than planned (July 2002)1,3 because of an adverse risk-to-benefit profile in the main WHI trial. Subsequently, the WHI, and ancil136
Neurology 72
January 13, 2009
lary WHIMS, CEE-Alone trial also terminated early (February 2004).2,4 WHIMS-MRI was designed to contrast MRI outcomes post-trial among WHIMS participants who had been assigned to active treatment vs placebo. It was conducted in 14 of the 39 WHIMS clinical centers, selected on the basis of interest, experience with multicenter MRI studies, participation in the WHI Study of Cognitive Aging, and availability of necessary equipment. Participants in these centers were eligible for recruitment to WHIMS-MRI, regardless of prior adherence to the WHI study protocol, on-trial use of study medications, on-study measures of cognitive function, or willingness to continue post-trial follow-up.21 Scans were performed, on average, 3.0 years posttrial for the CEE ⫹ MPA trial and 1.4 years post-trial for the CEE-Alone trial; average on-trial follow-up intervals were 4.0 years for CEE ⫹ MPA and 5.6 years for CEE-Alone. Exclusion criteria included the presence of pacemakers and other implants or foreign bodies contraindicated for MRI. Baseline demographic, lifestyle, and clinical factors were collected via self-report and standardized assessments. We included body mass index (BMI), because lower values may signal underlying brain pathologies in older individuals,22 and education, because higher education may identify individuals whose cognitive function is less responsive to atrophy.23 The 3MS18 was administered by a centrally trained and certified technician. It measures temporal and spatial orientation, immediate and delayed recall, executive function, naming, verbal fluency, abstract reasoning, praxis, writing, and visuoconstructional abilities. Scores range from 0 to 100 (higher score reflecting better cognitive functioning).
MRI protocol. MRI scans were conducted using a standardized protocol, developed by investigators at the MRI Quality Control Center in the Department of Radiology, University of Pennsylvania, Philadelphia. Additional detail and quality control procedures are provided in Coker et al.17 MRI series were acquired with field of view ⫽ 22 and matrix ⫽ 256 ⫻ 256. They included oblique axial spin density/T2-weighted spin echo (3,200/0/30,120/3), fluid-attenuated inversion recovery T2-weighted spin echo (8,000/2,000/100/3), and oblique axial three-dimensional T1-weighted gradient echo (flip angle 30; 21/0/8/1.5) images from the vertex to skull base parallel to the anterior commissure–posterior commissure (AC-PC) plane. To quantify regional brain volumes, the T1-weighted volumetric MRI scans were first preprocessed according to a standardized protocol24: 1) alignment to the AC-PC orientation; 2) removal of extracranial material; and 3) segmentation of brain parenchyma into gray matter (GM), white matter (WM), and CSF. Regional volumetric measurements of GM, WM, and CSF were subsequently obtained via a validated, automated computer-based template warping method.25 This technique is based on a digital atlas labeled for brain lobes and individual structures, including the hippocampus. Atlas definitions were transferred to MRI scans via an image-warping algorithm performing pattern matching of anatomically corresponding brain regions. The volumes of GM, WM, and CSF of each labeled brain region were obtained by summing the number of respective voxels within each region. Volumes of brain lesions and periventricular abnormal WM were also measured separately via the same procedure, using the three sets of images; total lesion volume was measured, as described in the accompanying article.17 Volumes of GM and WM reported in this article refer to normal brain tissue only. Intracranial volume (ICV) was estimated as the total cerebral
Figure
Enrollment and follow-up of WHIMS-MRI participants
*Multiple reasons were given, so totals are not additive. CEE ⫽ conjugated equine estrogens; MPA ⫽ medroxyprogesterone acetate; WHIMS ⫽ Women’s Health Initiative Memory Study.
hemispheric volumes, including ventricular CSF and the CSF within the sulcal spaces.
Statistical methods. Characteristics of participants at the time of WHI enrollment were described, and differences among treatment groups were compared using 2 tests. Differences in volumes of the total brain, ventricles, hippocampus, and frontal lobe (prespecified as secondary outcomes) were contrasted among women grouped by WHI treatment assignment, both separately within each trial and pooled across trials, using analyses of covariance that adjusted for age at WHI enrollment, time between enrollment and scanning, intracranial volume, clinical center site (the WHIMS stratification factor), and other baseline dementia risk factors (education level, ethnicity, smoking status, BMI, hypertension status, prior cardiovascular disease, diabetes, prior HT, and baseline 3MS score). Dementia risk factors were included to account for the possibility that balance among the groups originally developed by randomization had been diminished by attrition, nonconsent, and MRI-related eligibility. Each volume measure was analyzed separately. Because WHIMS-MRI was primarily designed to provide mechanistic support for the findings of the WHIMS trials, no adjustment for comparisons of its multiple endpoints was specified in its protocol. Associations between MRI outcomes and dementia risk factors were assessed with analyses of covariance. To test the hypothesis that the effect of HT on MRI volumes varied by baseline 3MS, we fitted an interaction term between treatment effect and baseline 3MS
scores as a continuous variable and presented fitted means for women grouped by baseline scores. We also grouped women according to total ischemic lesion volume, which includes infarcts and WM signal abnormalities,17 using the cutpoint of ⬍2 cm3 (lowest quartile) vs ⱖ2 cm3 (upper three quartiles). Analyses of covariance for total brain, ventricular, hippocampal, and frontal volumes were repeated using this grouping as a stratification factor to test the hypothesis that women with the lowest ischemic volume and the healthiest brains might show a benefit of HT on regional volumes. RESULTS WHIMS-MRI contacted 2,345 WHIMS participants, of which 1,527 (65.1%) provided consent. Of these, 1,424 (93.3%) received brain MRI scans, of which 1,403 (98.5%) met central reading criteria for analysis: 883 women in the CEE ⫹ MPA trial and 520 women in the CEE-Alone trial. The study flow diagram is shown in the figure. Compared with the 1,610 WHIMS participants at the 14 WHIMS-MRI sites who did not join the MRI study, WHIMS-MRI women were younger (mean age 77.5 vs 78.3 years; p ⬍ 0.001), had higher baseline 3MS scores (mean 96.1 vs 95.1; p ⬍ 0.01), and were fewer years postmenopausal (mean 28.7 vs 30.5 years; p ⬍ 0.001). However, participation rates did not differ among treatment assignments (p ⫽ 0.10), race (p ⫽ 0.36), education (p ⫽ 0.10), or BMI (p ⫽ 0.15). Table 1 presents dementia risk factors within the WHIMS-MRI cohort by WHI treatment assignment at the time of WHI enrollment. Although there were differences with respect to many risk factors between women enrolled in the CEE ⫹ MPA vs CEEAlone trials, there were no marked differences between women who had been randomly assigned to HT vs placebo. The mean (SD) age at the time of the MRI was 78.5 (3.7) years, which occurred an average of 8.0 years after WHI enrollment. The overall deficit in 3MS performance in association with HT observed on trial was apparent in WHIMS-MRI women at their annual evaluation preceding the MRI scan [treatment effect of 0.43 (0.21) units]. Mean (SE) ICV, an estimate of cranial size, was similar between HT and placebo groups: 1,095.9 (5.06) vs 1,087.1 cm3 for the CEE ⫹ MPA trial (p ⫽ 0.19) and 1,088.0 (5.96) vs 1,086.4 (6.66) cm3 for the CEE-Alone trial (p ⫽ 0.86). Table 2 presents mean volumes for total brain (GM plus WM), ventricles, hippocampus, and frontal lobe after adjustment for age at WHI enrollment, time between enrollment and scan, ICV, clinic site, and dementia risk factors listed in table 1. Mean hippocampal (p ⫽ 0.05) and frontal lobe (p ⫽ 0.004) volumes were lower in HT-treated women, and mean overall brain volumes were slightly lower among women who had been assigned to HT compared with placebo (p ⫽ 0.07). These differences were consistent between the Neurology 72
January 13, 2009
137
Table 1
Demographic, socioeconomic status, and lifestyle characteristics at the time of WHI enrollment by treatment assignment WHIMS-MRI CEE ⴙ MPA
WHIMS-MRI CEE-Alone
CEE ⴙ MPA, n ⴝ 436
Placebo, n ⴝ 447
CEE-Alone, n ⴝ 257
Placebo, n ⴝ 263
65–69 y
229 (52.5)
231 (51.7)
124 (48.2)
131 (49.8)
70–74 y
153 (35.1)
153 (34.2)
94 (36.6)
92 (35.0)
54 (12.4)
63 (14.1)
39 (15.2)
40 (15.2)
Variable Age, no. (%)
0.81
75ⴙ y Education, no. (%)
0.97
16 (3.7)
22 (4.9)
16 (6.2)
High school/GED
88 (20.0)
98 (22.0)
70 (27.2)
69 (26.2)
>High school <4 y college
179 (41.2)
165 (37.1)
95 (37.0)
118 (44.9)
9 (3.4)
>4 y college
152 (35.0)
160 (36.0)
76 (29.6)
67 (25.5)
3 (1.2)
0 (0.0)
Ethnicity, no. (%) American Indian/Alaskan
0.88‡ Native
Asian/Pacific Islander Black/African American Hispanic/Latino White, non-Hispanic Other
1 (0.2)
0 (0.0)
4 (0.9)
14 (3.1)
3 (1.2)
2 (0.8)
18 (4.1)
16 (3.6)
13 (5.1)
17 (6.5)
7 (1.6)
5 (1.1)
7 (2.8)
2 (0.8)
405 (92.9) 1 (0.2)
409 (91.7) 2 (0.4)
224 (87.8) 5 (2.0)
238 (90.7) 3 (1.2)
Smoking status, no. (%)
0.67
Never
257 (59.2)
252 (57.1)
149 (58.7)
Former
159 (36.6)
172 (39.0)
96 (37.8)
Current
18 (4.2)
17 (3.8)
9 (3.5)
148 (56.5) 99 (37.8) 15 (5.7)
Body mass index, no. (%) <25 kg/m
HT vs no HT, p Value
2
0.81 138 (31.7)
156 (35.0)
60 (23.5)
64 (24.4)
25–29 kg/m2
165 (37.9)
155 (34.8)
98 (38.4)
108 (41.2)
30–34 kg/m2
90 (20.7)
90 (20.2)
62 (24.3)
58 (22.1)
>35 kg/m2
42 (9.7)
45 (10.1)
35 (10.1)
32 (12.2)
240 (53.7)
122 (47.5)
139 (52.8)
Hypertension status, no. (%) None
0.75 234 (53.7)
Current/controlled* Current/uncontrolled
55 (12.6)
49 (11.0)
51 (19.8)
54 (20.5)
147 (33.7)
158 (35.4)
84 (32.7)
70 (26.6)
413 (94.7)
435 (95.1)
236 (91.8)
239 (90.9)
Prior cardiovascular disease, no. (%) No History of stroke
0.56
2 (0.5)
5 (1.1)
3 (1.2)
4 (1.5)
21 (4.8)
17 (3.8)
18 (7.0)
20 (7.6)
No
420 (96.3)
425 (95.1)
241 (93.8)
242 (92.0)
Yes
16 (3.7)
22 (4.9)
16 (6.2)
21 (8.0)
340 (78.0)
346 (77.4)
131 (51.0)
127 (48.3)
96 (22.0)
101 (22.6)
126 (49.0)
136 (51.7)
<90
22 (5.10)
19 (4.3)
20 (7.8)
19 (7.3)
90–94
66 (15.2)
80 (18.1)
50 (19.6)
58 (22.2)
347 (79.8)
342 (77.6)
185 (72.6)
184 (70.5)
History of other cardiovascular disease† Diabetes, no. (%)
0.23
Prior hormone therapy, no. (%) No Yes
0.59
3MS score, no. (%)
95–100
0.36
Data are presented as frequency (percent). *Measured to be less than 140/90 mm Hg. †Other cardiovascular disease defined as myocardial infarction, angina, percutaneous transluminal angioplasty, or coronary artery bypass grafting. ‡Based on collapsing to three categories (African American, white, and other). WHI ⫽ Women’s Health Initiative; WHIMS ⫽ Women’s Health Initiative Memory Study; CEE ⫽ conjugated equine estrogens; MPA ⫽ medroxyprogesterone acetate; HT ⫽ hormone therapy; GED ⫽ general equivalency diploma; 3MS ⫽ modified MiniMental State Examination.
138
Neurology 72
January 13, 2009
Table 2
Volumes (cubic centimeters) by treatment assignment after adjustment for age, time since enrollment, intracranial volume, clinic site, and other potential confounders listed in table 1 Total brain volume
Ventricle volume
HT
798.37 (1.30)
37.62 (0.55)
Placebo
801.69 (1.29) ⫺3.32 (1.84)
Hippocampal volume
Frontal lobe volume
Pooled trials
Difference p Value
0.07
5.69 (0.04)
282.72 (0.57)
37.15 (0.55)
5.79 (0.04)
285.09 (0.57)
0.47 (0.78)
⫺0.10 (0.05)
⫺2.37 (0.81)
0.55
0.05
0.004
CEE ⴙ MPA trial CEE ⴙ MPA
800.92 (1.63)
37.84 (0.68)
5.72 (0.04)
283.61 (0.72)
Placebo
803.11 (1.63)
36.53 (0.68)
5.83 (0.04)
285.46 (0.72)
⫺2.19 (2.32)
1.31 (0.97)
⫺0.11 (0.06)
⫺1.85 (1.03)
Difference p Value
0.35
0.18
0.09
0.07
CEE-Alone trial CEE-Alone
794.53 (2.21)
Placebo
799.03 (2.16) ⫺4.50 (3.13)
Difference p Value Consistency of treatment effects of CEE ⴙ MPA vs CEE-Alone, p value
37.53 (0.95)
5.63 (0.06)
281.47 (0.95)
37.85 (0.94)
5.75 (0.06)
284.25 (0.94)
⫺0.33 (1.36)
⫺0.12 (0.09)
⫺2.78 (1.36)
0.15
0.81
0.18
0.04
0.36
0.20
0.99
0.45
Data are presented as mean (SE). Tissue volumes include gray and white matter but not CSF. HT ⫽ hormonal therapy; CEE ⫽ conjugated equine estrogens; MPA ⫽ medroxyprogesterone acetate.
CEE ⫹ MPA and CEE-Alone trials. Mean ventricular volumes were unaffected by prior HT assignment. Associations that volumes had with dementia risk factors are described in table e-1 on the Neurology® Web site at www.neurology.org. Consistent with expectation, mean adjusted brain volumes were lower among women with higher age, lower BMI, uncontrolled hypertension, prior cardiovascular disease, or diabetes (all p ⱕ 0.05). Higher educational level also was associated with lower brain volumes. Older women had larger mean ventricular volumes and Table 3
Fitted mean difference in volumes (cubic centimeters) for women assigned to hormone therapy vs placebo, after adjustment for age, time since enrollment, intracranial volume, clinic site, and other potential confounders listed in table 1 Baseline 3MS score
Region
<90
90 –94
95–100
p Value*
Total brain
⫺16.93 (7.71)
⫺7.40 (4.34)
⫺1.41 (2.10)
0.07
3.19 (3.29)
⫺0.69 (1.85)
0.52 (0.90)
0.77
Hippocampus
⫺0.53 (0.21)
⫺0.21 (0.12)
⫺0.04 (0.06)
0.02
Frontal
⫺7.62 (3.40)
⫺2.59 (1.92)
⫺1.96 (0.93)
0.43
Ventricles
Data are presented as mean (SE). *p Values are based on interaction terms between treatment effect and baseline 3MS score as a continuous variable. 3MS ⫽ modified Mini-Mental State Examination.
smaller mean hippocampal and frontal lobe volumes. Lower BMI was associated with smaller hippocampal and frontal lobe volumes. Table 3 presents mean differences in volumes between women assigned to HT vs placebo who are grouped according to 3MS score at WHI enrollment, with adjustment for age, ICV, and clinic site, and additionally for all other dementia risk factors in table 1. Decrements in hippocampal volumes associated with HT therapy were greatest in women with the lowest pretreatment 3MS scores. Parallel analyses found that the association of HT assignment with the brain volume measures did not seem to depend on age. Women whose total ischemic lesion volume was below the approximate 25% percentile (2 cm3) were selected to represent those with relatively little evidence of vascular disease: 359 women, 26.1% of HT group and 25.1% of placebo group (p ⫽ 0.65). Table 4 contrasts mean HT-related differences in total brain, ventricular, hippocampal, and frontal volumes among women with lesion volumes ⬍2 cm3 with those with lesion volumes ⱖ2 cm3, with adjustment for all covariates. The small differences between treatment groups were not significant among women with the lowest ischemic lesion volumes. However, for women with ischemic lesion volumes ⱖ2 cm3, mean total brain (p ⬍ 0.05), hippocampal (p ⬍0.01), and frontal (p ⬍ 0.01) volumes were lower among women who had been assigned to HT. Through post-trial MRI scans of WHIMS participants, we found that randomization to CEE, with or without MPA, was associated with small but significant mean decrements in frontal (2.37 ⫾ 0.81 cm3) and hippocampal (0.10 ⫾ 0.05 cm3) volumes. Women randomly assigned to HT continued to express a persistent treatment-related deficit in 3MS scores through the time of the MRI assessment. Analysis of brain volume measures as a function of 3MS scores at WHIMS baseline showed that HT-associated reductions in hippocampal volume were greatest in women with the lowest cognitive function at WHI enrollment. These associations were similar for CEE ⫹ MPA and CEE-Alone trials. In addition, HT-associated reductions in total brain, hippocampal, and frontal volumes were apparent in women with vascular lesion burden volumes of 2 cm3 or larger, but not lower than 2 cm3. In contrast to several earlier reports of increased volumes of the hippocampus and other brain regions in HT users,7-10 we found no evidence of increased frontal, hippocampal, or total brain volumes in women randomly assigned to CEE ⫹ MPA or CEEAlone compared with placebo. Our findings are DISCUSSION
Neurology 72
January 13, 2009
139
Table 4
Volumes (cubic centimeters) by treatment assignment for women grouped according to total abnormal tissue volumes: <2 cm3 or >2 cm3, after adjustment for age, time since enrollment, intracranial volume, clinic site, and other potential confounders listed in table 1 Total brain volume
Ventricle volume
Hippocampal volume*
Frontal lobe volume
HT
788.12 (2.41)
30.90 (0.93)
6.08 (0.07)
278.15 (1.05)
Placebo
785.51 (2.43)
31.71 (0.94)
5.93 (0.07)
278.66 (1.06)
Lesion volume <2 cm3
Difference p Value
2.62 (3.52)
⫺0.81 (1.36)
0.41
0.55
HT
802.27 (1.55)
39.78 (0.67)
Placebo
806.93 (1.54) ⫺4.67 (2.20)
0.15 (0.10) 0.13
⫺0.51 (1.54) 0.74
Lesion volume > 2 cm3
Difference p Value
0.03
5.57 (0.04)
284.29 (0.69)
39.13 (0.66)
5.73 (0.04)
287.31 (0.68)
0.66 (0.95)
⫺0.16 (0.06)
⫺3.01 (0.98)
0.49
0.005
0.002
Data are presented as mean (SE). *Significant hormone therapy (HT) ⫻ lesion volume interaction, p ⫽ 0.01.
based on the largest sample of postmenopausal women studied to date. However, our sample differs from most prior reports in that we studied older women, with a mean age of 77.5 years at the time of MRI assessment, who initiated HT at age 65 years and older within the framework of the WHI clinical trial, and who had discontinued study medications an average of 3.0 years (CEE ⫹ MPA trial) and 1.4 years (CEE-Alone trial) before the MRI. In contrast, studies reporting increased volumes of the hippocampus and other gray matter regions in HT users7-10 were based on younger women who were long-term users of HT, generally initiated close to menopause, but not all studies have reported increased brain volumes in association with HT in younger women and long-term HT users.26,27 Moreover, hormone use before WHI enrollment was not associated with differences in regional brain volumes in WHIMS-MRI. The relationships between HT and hippocampal volumes varied significantly with baseline cognitive function, with a trend to similar effects for total brain volume. HT-associated reductions in hippocampal volume were greater in women with lower cognitive function (3MS score ⬍90) at WHIMS baseline before WHI HT randomization. Reductions in total brain, hippocampal, and frontal volumes in women randomly assigned to HT also were observed in the 75% of women with vascular lesion burdens greater than or equal to 2 cm3, but not in women with lesion volumes less than 2 cm3. These findings parallel the earlier WHIMS report that the degree to which HT adversely affected cognitive function was greatest in 140
Neurology 72
January 13, 2009
women with the lowest baseline 3MS scores (p ⬍ 0.001).4 It also is consistent with the short time frame in which HT increased risk of dementia (4 –5 years on average),1,2 which seems to be too rapid to be linked to the primary initiation of a protracted disease process. Greater vulnerability of postmenopausal women with low baseline cognitive function and higher lesion volumes to reduced brain volumes in association with HT is consistent with other evidence of the greater vulnerability of an already compromised brain28 and the potential that estrogen may adversely affect cognition among women with existing pathology.19 These findings also point to the growing body of evidence that vascular lesions and Alzheimer-type pathology act additively to influence the risk for clinical dementia.29 Because hippocampal volume loss is a well-documented risk factor for dementia30 and may be a biomarker for Alzheimer-type neuropathology,31 our findings suggest a possible contributory mechanism to HT-associated increase in dementia risk in women with low baseline cognitive function or existing neuropathology. Further research is required to elucidate whether the contribution of HT to lower total and regional brain volumes results from acceleration of Alzheimer-type pathology, from vascular disease, or some other mechanism. The mechanism underlying this possible neurotoxicity is unclear. Results from the companion article17 suggest that the effect is not conveyed primarily through an increase in ischemic lesions. It may be that there is an optimal level of estrogen exposure beyond which HT is neurotoxic.32 The optimum level may vary as a function of age or time since menopause as estrogen receptors may lose sensitivity in the absence of hormone exposure.33 CEE contains many equine estrogens that are not normally found in human blood and that have varying affinities to estradiol binding sites and a range of biologic activities.34,35 Many constituents seem to have neuroprotective properties,36 whereas the role of others remains unclear. Although ours is the largest study conducted to date of possible HT effects on brain structure, a number of issues limit the generality of our findings. We investigated the effects of particular CEE-based hormone regimens in older postmenopausal women, aged 65 years and older at initiation of treatment, and did not address possible effects in younger postmenopausal women. However, adverse effects of CEE ⫹ MPA on verbal memory (word list recall) were similar in older WHI participants37 and younger menopausal women with cognitive symptoms.38 Another limitation is that MRI scans were conducted post-trial, on average 3.0 and 1.4 years
post-trial for CEE ⫹ MPA and CEE-Alone. Because pretreatment MRI scans were not obtained, we have no information on brain volumes at baseline. However, the HT and placebo groups were well balanced with respect to many dementia risk factors. We repeated analyses in table 2 using propensity scores adjustment to account for potential differential enrollment,39 which resulted in essentially identical results. The automated approach to image processing may be prone to image registration errors, especially in some small regions. However, previous validation studies of this methodology40 have confirmed its accuracy in measuring hippocampal and lobar volumes. Moreover, total and regional brain volumes showed the predicted relationships with age and medical comorbidities such as uncontrolled hypertension and diabetes, providing an internal validation of our approach. More refined analyses of smaller regions, including voxel-based analysis, may identify other regions of vulnerability to HT that potentially cannot be resolved via the current methodology. Finally, our study is cross-sectional, and longitudinal volumetric studies may yield greater sensitivity to HT effects on the brain. Our findings emphasize the need for continued investigation of the joint effects of brain volume changes and vascular changes to further understanding of HT effects on cognitive and brain aging. AUTHORS’ AFFILIATIONS From the Laboratory of Personality and Cognition (S.M.R.), Intramural Research Program, National Institute on Aging, NIH, Biomedical Research Center, Baltimore, MD; Division of Public Health Sciences (M.A.E., S.A.J.), Wake Forest University School of Medicine, WinstonSalem, NC; Departments of Internal Medicine and Public Health Sciences (C.H.), University of California, Davis, CA; Stanford Prevention Research Center (M.L.S.), Department of Medicine, Stanford University, CA; Chronic Disease Research Group (A.M.M.), Hennepin County Medical Center, Minneapolis, MN; University of Massachusetts (J.O.), Worcester, MA; and Department of Radiology (C.D.), University of Pennsylvania, Philadelphia, PA.
DISCLOSURE M.A.E. received salary support from Wyeth Pharmaceuticals from 1995 to 2003 as a Women’s Health Initiative Memory Study (WHIMS) investigator. He also was compensated by Wyeth Pharmaceuticals from 2002 to 2005 for serving on a monitoring board for an unrelated clinical trial. The remaining authors have nothing to disclose.
APPENDIX WHIMS-MRI Clinical Centers. Albert Einstein College of Medicine, Bronx, NY: Sylvia Wassertheil-Smoller, Mimi Goodwin, Richard DeNise, Michael Lipton, James Hannigan; Medical College of Wisconsin, Milwaukee, WI: Jane Morley Kotchen, Diana Kerwin, John Ulmer, Steve Censky; Stanford Center for Research in Disease Prevention, Stanford University, CA: Marcia L. Stefanick, Sue Swope, Anne Marie Sawyer-Glover; The Ohio State University, Columbus, OH: Rebecca Jackson, Rose Hallarn, Bonnie Kennedy; University of California at Davis, Sacramento, CA: John Robbins, Sophia Zaragoza, Cameron Carter, John Ryan; University of California at Los Angeles, CA: Lauren Nathan, Barbara Voigt, Pablo Villablanca, Glen Nyborg; University of
Florida, Gainesville/Jacksonville, FL: Marian Limacher, Sheila Anderson, Mary Ellen Toombs, Jeffrey Bennett, Kevin Jones, Sandy Brum, Shane Chatfield; University of Iowa, Davenport, IA: Jennifer Robinson, Candy Wilson, Kevin Koch, Suzette Hart; University of Massachusetts, Worcester, MA: Judith Ockene, Linda Churchill, Douglas Fellows, Anthony Serio; University of Minnesota, Minneapolis, MN: Karen Margolis, Cindy Bjerk, Chip Truwitt, Margaret Peitso; University of Nevada, Reno, NV: Robert Brunner, Ross Golding, Leslie Pansky; University of North Carolina, Chapel Hill, NC: Carol Murphy, Maggie Morgan, Mauricio Castillo, Thomas Beckman; University of Pittsburgh, PA: Lewis Kuller, Pat McHugh, Carolyn Meltzer, Denise Davis. WHIMS-MRI Clinical Coordinating Center. Wake Forest University Health Sciences, Winston-Salem, NC: Sally Shumaker, Mark Espeland, Laura Coker, Jeff Williamson, Debbie Felton, LeeAnn Andrews, Steve Rapp, Claudine Legault, Maggie Dailey, Julia Robertson, Patricia Hogan, Sarah Jaramillo, Pam Nance, Cheryl Summerville, Josh Tan. WHIMS-MRI Quality Control Center. University of Pennsylvania, Philadelphia, PA: Nick Bryan, Christos Davatzikos, Lisa Desiderio. WHIMS-MRI Working Group. Wake Forest University Health Sciences, Winston-Salem, NC: LeeAnn Andrews; University of Pennsylvania, Philadelphia, PA: Nick Bryan; Wake Forest University Health Sciences, Winston-Salem, NC: Laura Coker; Wake Forest University Health Sciences, Winston-Salem, NC: Mark Espeland; Wake Forest University Health Sciences, Winston-Salem, NC: Debbie Felton; University of Pittsburgh, PA: Lew Kuller; University of Minnesota, MN: Karen Margolis; University of Minnesota, Minneapolis, MN: Anne Murray; National Institute on Aging, Baltimore, MD: Susan Resnick; Wake Forest University Health Sciences, Winston-Salem, NC: Sally Shumaker; Wake Forest University Health Sciences, Winston-Salem, NC: Jeff Williamson. US NIH. National Institute on Aging, Bethesda, MD: Neil Buckholtz, Susan Molchan, Susan Resnick; National Heart, Lung, and Blood Institute, Bethesda, MD, Jacques Rossouw, Linda Pottern.
Received June 27, 2008. Accepted in final form September 26, 2008.
REFERENCES 1. Shumaker SA, Legault C, Thal L, et al. Estrogen plus progestin and the incidence of dementia and mild cognitive impairment in postmenopausal women: the Women’s Health Initiative Memory Study—a randomized controlled trial. JAMA 2003;289:2651–2662. 2. Shumaker SA, Legault C, Kuller L, et al. Conjugated equine estrogens and incidence of probable dementia and mild cognitive impairment in postmenopausal women: Women’s Health Initiative Memory Study. JAMA 2004; 291:2947–2958. 3. Rapp SR, Espeland MA, Shumaker SA, et al. Effect of estrogen plus progestin on global cognitive function in postmenopausal women: the Women’s Health Initiative Memory Study—a randomized controlled trial. JAMA 2003;289:2663–2672. 4. Espeland MA, Rapp SR, Shumaker SA, et al. Conjugated equine estrogens and global cognitive function in postmenopausal women: Women’s Health Initiative Memory Study. JAMA 2004;291:2959–2968. 5. Wassertheil-Smoller S, Hendrix SL, Limacher M, et al. Effect of estrogen plus progestin on stroke in postmenopausal women: the Women’s Health Initiative—a randomized trial. JAMA 2003;289:2673–2684. 6. Anderson GL, Limacher M, Assaf AR, et al. Effects of conjugated equine estrogen in postmenopausal women with hysterectomy: the Women’s Health Initiative randomized controlled trial. JAMA 2004;291:1701–1712. 7. Boccardi M, Ghidoni R, Govoni S, et al. Effects of hormone therapy on brain morphology of healthy postmenoNeurology 72
January 13, 2009
141
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18. 19.
20.
21.
22. 23.
142
pausal women: a voxel-based morphometry study. Menopause 2006;13:584–591. Erickson KI, Colcombe SJ, Raz N, et al. Selective sparing of brain tissue in postmenopausal women receiving hormone replacement therapy. Neurobiol Aging 2005;26: 1205–1213. Eberling JL, Wu C, Haan MN, Mungas D, Buonocore M, Jagust WJ. Preliminary evidence that estrogen protects against age-related hippocampal atrophy. Neurobiol Aging 2003;24:725–732. Lord C, Buss C, Lupien SJ, Pruessner JC. Hippocampal volumes are larger in postmenopausal women using estrogen therapy compared to past users, never users and men: a possible window of opportunity effect. Neurobiol Aging 2008;29:95–101. Resnick SM, Maki PM, Golski S, Kraut MA, Zonderman AB. Estrogen effects on PET cerebral blood flow and neuropsychological performance. Horm Behav 1998;34:171– 182. Rasgon NL, Small GW, Siddarth P, et al. Estrogen use and brain metabolic change in older adults: a preliminary report. Psychiatry Res 2001;107:11–18. Eberling JL, Reed BR, Coleman JE, Jagust WJ. Effect of estrogen on cerebral glucose metabolism in postmenopausal women. Neurology 2000;55:875–877. Eberling JL, Wu C, Tong-Turnbeaugh R, Jagust WJ. Estrogen- and tamoxifen-associated effects on brain structure and function. Neuroimage 2004;21:364–371. Shaywitz SE, Shaywitz BA, Pugh KR, et al. Effect of estrogen on brain activation patterns in postmenopausal women during working memory tasks. JAMA 1999;281: 1197–1202. Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results From the Women’s Health Initiative randomized controlled trial. JAMA 2002;288:321–333. Coker LH, Hogan PE, Bryan NR, et al., for the Women’s Health Initiative Memory Study. Postmenopausal hormone therapy and subclinical cerebrovascular disease: the WHIMS-MRI Study. Neurology 2009;72:125–134. Teng EL, Chui H. The Modified Mini-Mental State (3MS) Exam. J Clin Psychiatry 1987;48:314–318. Wise PM, Dubal DB, Rau SW, Brown CM, Suzuki S. Are estrogens protective or risk factors in brain injury and neurodegeneration? Reevaluation after the Women’s Health Initiative. Endocr Rev 2005;26:308–312. Shumaker SA, Reboussin BA, Espeland MA, et al. The Women’s Health Initiative Memory Study (WHIMS): a trial of the effect of estrogen therapy in preventing and slowing the progression of dementia. Control Clin Trials 1998;19:604–621. Jaramillo SA, Felton D, Andrews L, et al. Enrollment in a brain magnetic resonance study: results from the Women’s Health Initiative Memory Study Magnetic Resonance Imaging Study (WHIMS-MRI). Acad Radiol 2007;14:603– 612. Gustafson D. A life course of adiposity and dementia. Eur J Pharmacol 2008;585:163–175. Coffey CE, Saxton JA, Ratcliff G, Bryan R, Lucke JF. Relation of education to brain size in normal aging: implications for the reserve hypothesis. Neurology 1999;53:189–196.
Neurology 72
January 13, 2009
24.
25.
26.
27.
28.
29.
30.
31.
32.
33. 34.
35.
36.
37.
38.
39.
40.
Goldszal AF, Davatzikos C, Pham DL, Yan MX, Bryan RN, Resnick SM. An image-processing system for qualitative and quantitative volumetric analysis of brain images. J Comput Assist Tomogr 1998;22:827–837. Shen D, Davatzikos C. HAMMER: hierarchical attribute matching mechanism for elastic registration. IEEE Trans Med Imaging 2002;21:1421–1439. Low LF, Anstey KJ, Maller J, et al. Hormone replacement therapy, brain volumes and white matter in postmenopausal women aged 60-64 years. Neuroreport 2006;17: 101–104. Greenberg DL, Payne ME, MacFall JR, Provenzale JM, Steffens DC, Krishnan RR. Differences in brain volumes among males and female hormone-therapy users and nonusers. Psychiatry Res 2006;147:127–134. Gamaldo A, Moghekar A, Kilada S, Resnick SM, Zonderman AB, O’Brien R. Effect of a clinical stroke on the risk of dementia in a prospective cohort. Neurology 2006;67: 1363–1369. Schneider JA, Wilson RS, Cochran EJ, et al. Relation of cerebral infarctions to dementia and cognitive function in older persons. Neurology 2003;60:1082–1088. Jack CR Jr, Shiung MM, Weigand SD, et al. Brain atrophy rates predict subsequent clinical conversion in normal elderly and amnestic MCI. Neurology 2005;65:1227–1231. Bobinski M, de Leon MJ, Wegiel J, et al. The histological validation of post mortem magnetic resonance imagingdetermined hippocampal volume in Alzheimer’s disease. Neuroscience 2000;95:721–725. Chen S, Nilsen J, Brinton RD. Dose and temporal pattern of estrogen exposure determines neuroprotective outcome in hippocampal neurons: therapeutic implications. Endocrinology 2006;147:5303–5313. Toran-Allerand CD. Estrogen as a treatment for Alzheimer disease [letter; comment]. JAMA 2000;284:307–308. de Lignieres B, Silberstein S. Pharmacodynamics of oestrogens and progestogens. Cephalalgia 2000;20:200– 207. Turgeon JL, Carr MC, Maki PM, Mendelsohn ME, Wise PM. Complex actions of sex steroids in adipose tissue, the cardiovascular system, and brain: insights from basic science and clinical studies. Endocr Rev 2006;27:575–605. Zhao L, Brinton RD. Select estrogens within the complex formulation of conjugated equine estrogens (Premarin) are protective against neurodegenerative insults: implications for a composition of estrogen therapy to promote neuronal function and prevent Alzheimer’s disease. BMC Neurosci 2006;7:24. Resnick SM, Maki PM, Rapp SR, et al. Effects of combination estrogen plus progestin hormone treatment on cognition and affect. J Clin Endocrinol Metab 2006;91: 1802–1810. Maki PM, Gast MJ, Vieweg AJ, Burriss SW, Yaffe K. Hormone therapy in menopausal women with cognitive complaints: a randomized, double-blind trial. Neurology 2007; 69:1322–1330. Rubin DB. Using multivariate matched sampling and regression adjustment to control bias in observational studies. J Am Stat Assoc 1979;74:318–324. Shen D, Moffat S, Resnick SM, Davatzikos C. Measuring size and shape of the hippocampus in MR images using a deformable shape model. Neuroimage 2002;15:422–434.
MRI correlates of cognitive decline in CADASIL A 7-year follow-up study
M.K. Liem, MD S.A.J. Lesnik Oberstein, MD, PhD J. Haan, MD, PhD I.L. van der Neut, MSc M.D. Ferrari, MD, PhD M.A. van Buchem, MD, PhD H.A.M. Middelkoop, PhD J. van der Grond, PhD
Address correspondence and reprint requests to Dr. M.K. Liem, Department of Radiology, Leiden University Medical Center, C2S, Albinusdreef 2, 2333 ZA Leiden, the Netherlands
[email protected]
ABSTRACT
Background: Cognitive decline is one of the clinical hallmarks of cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL), a cerebrovascular disease caused by NOTCH3 mutations. In this 7-year follow-up study, we aimed to determine whether there are associations between the different radiologic hallmarks in CADASIL and decline in specific cognitive domains.
Methods: Twenty-five NOTCH3 mutation carriers and 13 controls had standardized neuropsychological testing and MRI examinations at baseline and after a follow-up of 7 years. To identify longitudinal associations between MRI abnormalities and cognitive decline, correlation analysis was used. Results: At follow-up, mutation carriers showed a decline in global cognitive function (CAMCOG, p ⬍ 0.01) and in the cognitive domains language, memory, and executive function, compared to controls. Cognitive decline, especially executive dysfunction, was associated with increase in lacunar infarcts, microbleeds, and ventricular volume. In contrast, WMHs and brain atrophy were not associated with cognitive decline. Conclusion: Increase in lacunar infarcts, microbleeds, and ventricular volume, but not white matter lesions or atrophy, are associated with cognitive decline in the process of CADASIL in younger-aged, mildly affected patients with CADASIL. Neurology® 2009;72:143–148 GLOSSARY ANOVA ⫽ analysis of variance; CADASIL ⫽ cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy; CAMCOG ⫽ Cambridge Cognitive Examination; DSM-IV ⫽ Diagnostic and Statistical Manual of Mental Disorders, 4th edition; MC ⫽ mutation carrier; MMSE ⫽ Mini-Mental State Examination; nonMC ⫽ non-mutation carrier; WMH ⫽ white matter hyperintensity; WMS ⫽ Wechsler Memory Scale.
Cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL) is a hereditary microangiopathy caused by mutations in the NOTCH3 gene.1 One of the clinical hallmarks of CADASIL is cognitive dysfunction. Most patients develop a marked and progressive cognitive decline before the age of 60.2 Cognitive dysfunction in CADASIL usually starts with impairment of executive functions which slowly progresses to global impairment of all cognitive domains.3,4 Although several studies have demonstrated that general cognitive decline in CADASIL is associated with a number of CADASIL-associated MRI abnormalities, the exact mechanism that leads to this cognitive decline remains unclear. Two recent studies showed that lacunar infarct load is independently associated with cognitive dysfunction in CADASIL.5,6 In other studies it was found that whole brain volume and diffusion tensor imaging parameters are associated with global cognitive dysfunction.7,8 The main limitations of these studies are their cross-sectional or short follow-up design,5,6 as well as that only general cognitive decline was measured, and not specific cognitive domains.7,8 The possible relationship between decline in Supplemental data at www.neurology.org From the Departments of Radiology (M.K.L., I.L.v.d.N., M.A.v.B., J.v.d.G.), Clinical Genetics (S.A.J.L.O.), Neurology (J.H., M.D.F.), and Neuropsychology (H.A.M.M.), Leiden University Medical Center; the Department of Neurology (J.H.), Rijnland Hospital, Leiderdorp; and the Department of Psychology (H.A.M.M.), Leiden University, The Netherlands. Disclosure: The authors report no disclosures. Copyright © 2009 by AAN Enterprises, Inc.
143
diverse cognitive domains and radiologic hallmarks of CADASIL, i.e., lacunar infarcts, white matter hyperintensities (WMHs), and microbleeds, has not been investigated yet. The aim of this study is to investigate longitudinal associations between radiologic changes and cognitive decline in CADASIL over a 7-year interval. METHODS Patients. Participants were drawn from a Dutch CADASIL cohort evaluated at baseline in 1999/2000,9 which included 40 symptomatic and asymptomatic NOTCH3 mutation carriers (MCs) and 22 non-mutation carriers (nonMCs) from 15 unrelated families. All living participants were invited for a follow-up visit. Informed consent was obtained from the participant or from a family member if the participant was unable to provide informed consent. A total of 38 participants from 12 unrelated families consented to participate in the follow-up study. Twenty-five were MCs and 13 were nonMCs, the latter serving as controls. The medical ethics committee of the Leiden University Medical Center approved the study. We took a full medical history of all participants and obtained their medical records from their physicians and general practitioners. Clinical, neuropsychological, and radiologic examinations were performed blinded to NOTCH3 mutation status.
Neuropsychological assessment. All individuals followed a standardized neuropsychological test battery, lasting 3 hours, at baseline and at follow-up. Details regarding administration, scoring, and clinical value of the administered neuropsychological tests have been extensively described by Spreen and Strauss.10 Global cognitive functioning was assessed using the Cambridge Cognitive Examination (CAMCOG),11 which incorporates the Mini-Mental State Examination (MMSE).12 The CAMCOG provides a total score for global cognitive functioning as well as subscores for specific cognitive functions (orientation, attention, memory, language, praxis, gnosis, calculation, abstract thinking). Memory was additionally evaluated using the Wechsler Memory Scale (WMS).13 For testing of executive function we used the Trail Making Test A, a test of processing speed with relatively low executive burden, Trail Making Test B, a test of processing speed with relatively high executive burden,14 and the colorinterference section of the Stroop Color and Word test.15 For data analysis we used MMSE and CAMCOG as overall scores and we used the subtests that correspond to the five domains of cognition according to the DSM classification of dementia: memory, language, gnosis, praxis, and executive function.16 Raw scores of the tests were used, except for the WMS “memory quotient” which was conventionally transformed into a scaled score.10 Trail Making Test A and B scores were also analyzed after transformation to a logarithmic scale, because of a skewed distribution to the right, caused by long task completion times in patients with severe executive dysfunction. Since results were similar with and without using this transformation, only the non transformed scores are shown. MRI. Image acquisition. A uniform MRI protocol was performed on the same 1.5 T MR system (Philips Medical Systems, Best, The Netherlands) at the baseline and the follow-up examination. No significant hardware updates have been made to the scanner in the 7-year follow-up interval. The protocol included axial T1, T2, FLAIR, and T2* GE sequences. See e-Methods on 144
Neurology 72
January 13, 2009
the Neurology® Web site at www.neurology.org for additional details. Image postprocessing. For volumetric MRI measurements of brain parenchyma, CSF, and WMHs we used locally developed semiautomated segmentation software (SNIPER, Software for Neuro-Image Processing in Experimental Research) that combines knowledge-based fuzzy clustering and region-growing techniques.17 Measurements were performed on dual spin-echo and FLAIR images. The FLAIR and proton density MRI were registered affine to the T2-weighted MRI using 12 degrees of freedom. The T1 sequences were not used for segmentation. As a first step the intracranial volume was determined after skull stripping. In the second step brain parenchyma and CSF were segmented separately. Then CSF was separated into ventricular CSF and CSF surrounding the brain (peripheral CSF). Manual correction steps were used between these steps to correct for random segmentation errors. Brain atrophy was defined as follows: brain atrophy ⫽ (intracranial volume ⫺ brain parenchymal volume)/intracranial volume ⫻ 100%. Ventricular CSF volume and peripheral CSF volume were also expressed as percentage of total intracranial volume. The final segmentation step was segmentation of WMHs. proton density, T2, and FLAIR images were used for WMH segmentation. WMHs were defined as white matter areas with increased signal intensity on all three sequences. The software computed an additional T2/proton density image to distinguish the lesions from CSF. Volume of WMH was corrected for total brain volume by dividing the individual volume of WMH by total brain volume and expressed in percent. The number of lacunar infarcts and microbleeds were counted on a digital workstation by one observer (M.L.), according to scoring criteria previously described (for more details, see e-Methods).5 A second observer (M.v.B.) reviewed the scores and in case of conflicting scores agreement was reached with a third observer (J.v.d.G.). All observers were blinded to patient data. After manual identification of lacunar infarcts, they were also marked on the MRI segmentations.
Statistics. Statistical analysis was performed by M.K. Liem, I.L. van der Neut, and J. van der Grond, all from the Department of Radiology of the Leiden University Medical Center. Statistical analysis was performed using the SPSS-14 statistical software package (SPSS Inc., Chicago, IL). Differences between MCs and nonMCs in demographic variables, MRI parameters, and neuropsychological testing results at baseline were analyzed using Student t tests for the normally distributed continuous variable Wechsler Memory Scale memory quotient, Mann–Whitney U test for the other non-normally distributed continuous variables, and 2 tests for categorical variables. In order to detect possible effects of selection bias on the study population, we also compared age and neuropsychological testing scores at baseline between MCs who did and who did not provide follow-up data, using Student t test for age and Wechsler Memory Scale memory quotient and Mann–Whitney U test for the other variables. Changes in neuropsychological testing results between baseline and follow-up were calculated for nonMCs. The mean change in nonMCs was considered to be a result from aging and learning effect, and was used as a correction factor for the neuropsychological testing scores of the MCs. Each individual MC testing score at follow-up was corrected by adding the mean change of the nonMC group to that score. The resulting differences in neuropsychological testing results in MCs were tested with repeated measurements analysis of variance (ANOVA) for Wechsler Memory Scale memory quotient and paired samples
Table 1
Neuropsychological testing results at baseline and follow-up (n ⴝ 38) NonMC (n ⴝ 13), mean (SD)
Cognitive domain Measure Global
Baseline
MC (n ⴝ 25), mean (SD)
Follow-up Baseline
Follow-up
Corrected difference* ⫺0.9 (4.6)
MMSE
28 (1.2)
28 (1.4)
28 (1.4)
27 (4.7)
CAMCOG total
95 (4.5)
96 (5.1)
94 (5.4)
89 (13.4)
⫺5.8 (10)†
27 (2.2)
28 (1.9)
27 (2.5)
26 (3.8)
⫺1.7 (2.3)†
9.9 (0.3)
9.6 (0.9)
9.6 (0.9)
Language CAMCOG language Gnosis
CAMCOG gnosis
9.1 (1.3)
⫺0.2 (1.1)
Praxis
CAMCOG praxis
11.5 (0.7) 11.4 (0.7) 11.3 (0.8) 10.6 (2.0)
⫺0.6 (2.0)
Memory
WMS-MQ
110 (12)
117 (16)
106 (14)
108 (18)
⫺4.7 (12)‡
29 (10)
30 (10)
38 (17)
50 (67)
⫹11 (54)
0.4
⫹0.2 (2.0)
Executive Trails A, speed (sec) function Trails A (errors)
0
0
0.2
⫹19 (49)
Trails B, speed (sec)
73 (24)
70 (15)
92 (53)
107 (73)
Trails B (errors)
0.2 (0.1)
0.1 (0.1)
0.4 (0.1)
0.7 (1.4)
⫹0.3 (1.5)
Stroop interference
39 (15)
32 (10)
40 (17)
48 (39)
⫹14.4 (35)‡
Baseline–follow-up comparisons: paired samples t test for WMS-MQ. Paired samples Wilcoxon test for the other variables. *Difference corrected for change in nonMCs. †Difference (p ⬍ 0.01) between baseline and follow-up. ‡Difference (p ⬍ 0.05) between baseline and follow-up. NonMC ⫽ non-mutation carriers; MC ⫽ mutation carriers; MMSE ⫽ Mini-Mental State Examination; CAMCOG ⫽ Cambridge Cognitive Examination; WMS-MQ ⫽ Wechsler Memory Scale memory quotient.
Wilcoxon tests for the other variables. Differences in MRI parameters between baseline and follow-up were also tested with repeated measurements ANOVA for normally distributed variables global brain atrophy, central atrophy, and peripheral atrophy and with paired samples Wilcoxon tests for the other nonnormally distributed variables. To correct for possible age differences between the MC and nonMC group, covariance analysis with age as a covariate was also used when needed. The 7-year differences in MRI parameters were correlated with differences in neuropsychological testing results using Pearson correlation coefficient for correlations between Wechsler Memory Scale memory quotient and brain atrophy, central atrophy, or peripheral atrophy, and using Spearman correlation coefficient for all other correlations. In order to determine the relative contribution of each MRI variable to cognitive decline, an additional multivariate analysis was performed, using multiple linear regression analysis of variables that showed an association (p ⬍ 0.05) on univariate analysis. Significance thresholds for associations between MRI parameters and cognitive decline were set at p ⬍ 0.01. Associations with a p value between 0.01 and 0.05 were considered as trends. RESULTS Patients. Of the 62 original participants, 7 died during the follow-up period. Of the 55 remaining participants, 38 consented to participate in the follow-up study. Reasons for not participating were severe disease (3), disease-related immobility (2), hospital anxiety (2), and lack of interest or motivation (8). Two individuals did not provide a reason for nonparticipation. Of the 24 people who did not participate, 15 were MCs. These MCs were significantly older (nonparticipants: mean age 51.4, SD 9.5; participants: mean age 42.2; SD 9.5) and had
worse cognitive testing scores (CAMCOG total, p ⫽ 0.02; CAMCOG language, p ⫽ 0.04; CAMCOG praxis, p ⫽ 0.04; WMS memory quotient, p ⫽ 0.01; TMT-A, p ⫽ 0.03) than MCs who participated in the follow-up study. Of the 38 participants in the follow-up study, 25 were NOTCH3 mutation carriers. The average time between the two examinations was 7.1 years (range 6.4 –7.6). There were no significant differences in male/female ratio, years of secondary education, and age between MCs and nonMCs, although the average age in the MC group was slightly higher than in the nonMC group (MCs: mean 42.2, SD 9.5; nonMCs: mean 36.7, SD: 8.4). Neuropsychological test results of the MC and nonMC group at baseline and follow-up are shown in table 1. No significant differences in test results were found at baseline. At follow-up, MCs had a decrease in CAMCOG total score (p ⬍ 0.01), CAMCOG language subscore (p ⬍ 0.01), WMS-MQ (p ⬍ 0.05), and Stroop interference score (p ⬍ 0.05). The MRI characteristics of the MC and nonMC group at baseline and follow-up are shown in table 2. MCs demonstrated an increase in lacunar infarcts (p ⬍ 0.01), WMHs (p ⬍ 0.01), and microbleeds (p ⬍ 0.05) at follow-up. NonMCs had no lacunar infarcts or microbleeds and no significant WMHs at baseline or follow-up. At baseline, level of brain atrophy, ventricular volume, and peripheral CSF volume was similar between MCs and nonMCs. At followup, ventricular volume had increased significantly only in MCs, whereas the level of brain atrophy increased by the same amount in both MCs and nonMCs. The yearly rate of brain atrophy was 1.7 mL per year for both MCs and nonMCs. The associations between the changes in MRI parameters and neuropsychological testing results in MCs after the 7-year follow-up are shown in table 3. A decrease in global cognitive functioning was associated with an increase in microbleeds (CAMCOG: p ⫽ 0.005) and showed a trend toward an association with increase in lacunar infarcts (CAMCOG: p ⫽ 0.026) and increase in ventricular volume (MMSE: p ⫽ 0.025; CAMCOG: p ⫽ 0.022). Global cognitive decline was not associated with changes in WMH volume, brain atrophy, or peripheral CSF volume. Cognitive decline in the memory domain was associated with increase in microbleeds (p ⫽ 0.005), showed a trend toward an association with increase in ventricle volume (p ⫽ 0.02), but was not associated with increase in WMH volume, brain atrophy, or peripheral CSF volume. Decrease of executive function testing scores was associated with increase in lacunar infarcts (Stroop interference: p ⫽ 0.005), microbleeds (TMT-B: p ⫽ 0.007; Stroop inNeurology 72
January 13, 2009
145
Table 2
decline, especially in the executive function domain (TMT-B and Stroop: p ⬍ 0.01).
MRI characteristics at baseline and follow-up (n ⴝ 38) NonMC (n ⴝ 13)
MC (n ⴝ 25)
This study shows that in addition to lacunar infarcts, cognitive decline in mild CADASIL is also associated with microbleeds and ventricular volume, especially executive dysfunction. In contrast, WMHs and brain atrophy are not associated with cognitive decline. This association between microbleeds and cognitive decline in CADASIL has not been described before. In this study we found that, next to global cognitive decline, the decline in the specific cognitive domains of executive function and memory are most strongly associated with an increase in microbleeds. Possibly, this association can be explained by focal tissue damage caused by microbleeds that leads to cognitive dysfunction, as has been suggested by studies in the general population.18,19 However, it is also possible that they are a proxy measure of vascular pathology and subcortical white matter injury, rather than a major cause of focal tissue damage. Another finding in this study is the association between ventricular volume and cognition in CADASIL, as shown on both univariate and multivariate analysis. This is in line with a general population study showing that ventricular volume is an important determinant of cognitive decline.20 Associations between ventricular volume and cognition have also been described in other diseases such as Alzheimer disease and normal pressure hydrocephalus.21-23 However, the pathophysiologic mechanism behind this association is unknown. The increasing ventricular volume in NOTCH3 MCs may DISCUSSION
MRI parameter
Baseline
Follow-up
WMH volume, % (SD)
0.0
0.0
Infarcts, n (range)
0
0
6.2 (0–27)*
9.9 (0–38)†
Microbleeds, n (range)
0
0
1.6 (0–35)
3.5 (0–40)‡
Brain atrophy, % (SD)
19.2 (1.7)
Ventricular volume, % (SD) Peripheral CSF volume, % (SD)
20.0 (1.3)‡
Baseline
Follow-up
5.0 (3.9)*
7.4 (5.3)†
17.4 (2.9)
18.2 (3.2)‡
1.9 (0.7)
2.1 (1.0)
2.5 (1.0)
2.9 (1.4)†
17.3 (1.9)
17.9 (1.7)
14.9 (3.0)
15.2 (2.9)
MC–NonMC comparisons: Mann-Whitney U test for WMH volume, infarcts, and microbleeds. Covariance analysis of means with age as a covariate for brain atrophy, ventricular volume, and peripheral CSF volume. Baseline–follow-up comparisons: paired samples Wilcoxon test for WMH volume, infarcts, and microbleeds. Repeated measures analysis of variance for the other variables. *Difference (p ⬍ 0.01) between MCs and nonMCs at baseline. †Difference (p ⬍ 0.01) between baseline and follow-up. ‡Difference (p ⬍ 0.05) between baseline and follow-up. NonMC ⫽ non-mutation carriers; MC ⫽ mutation carriers; WMH ⫽ white matter hyperintensity.
terference: p ⫽ 0.002), and ventricle volume (TMT-B: p ⫽ 0.008; Stroop interference: p ⫽ 0.02). WMH volume, brain atrophy, and peripheral atrophy had no effect on changes in executive function. Cognitive decline in the language, gnosis, and praxis domains was, overall, not associated with any of the MRI parameters. There were only some trends toward associations between brain atrophy and peripheral CSF volume with CAMCOG praxis (p ⫽ 0.02 and p ⫽ 0.04) and between lacunar infarcts and CAMCOG language (p ⫽ 0.04). Multivariate analysis showed that increase in ventricular volume was the most important relative contributor to cognitive Table 3
Correlations between changes in MRI parameters and changes in neuropsychological test results in mutation carriers (n ⴝ 25) ⌬ Infarcts
Cognitive domain
Measure
r
Global
⌬ MMSE
⫺0.39
⌬ WMH volume
p
⌬ Microbleeds
p
r
⫺0.19
⫺0.30
p
⌬ Brain atrophy r
p
⌬ Ventricle volume
⌬ Peripheral CSF volume
r
p
r
⫺0.17
⫺0.45
0.025
⫺0.08
p
0.022*
⫺0.06
⌬ CAMCOG total
⫺0.45
0.026
⫺0.26
⫺0.54
⫺0.09
⫺0.46
Language
⌬ CAMCOG language
⫺0.41
0.041
⫺0.13
⫺0.25
⫺0.20
⫺0.36
0.00
Gnosis
⌬ CAMCOG gnosis
⫺0.22
⫺0.02
⫺0.25
⫺0.17
⫺0.29
⫺0.09
Praxis
⌬ CAMCOG praxis
⫺0.05
0.12
⫺0.38
⫺0.47
⫺0.32
⫺0.41
Memory
⌬ WMS-MQ
⫺0.21
0.02
⫺0.55
⫺0.22
⫺0.46
Executive function
⌬ Trails A (speed)
⫺0.02
⫺0.16
0.01
⫺0.14
0.32
0.01
⌬ Trails A (errors)
⫺0.06
⫺0.33
0.15
⫺0.42
0.31
0.26
⌬ Trails B (speed)
0.34
0.19
0.56
⌬ Trails B (errors)
0.38
0.16
0.14
⌬ Stroop interference
0.57
0.25
0.62
0.005
0.005
0.005
0.007†
0.002
0.02
0.33
0.55
⫺0.08
0.32
0.16
0.48
0.023*
0.008†
0.04
⫺0.07
0.10 ⫺0.18
0.024†
0.06
WMS completed by 24 mutation carriers, Stroop test by 22 mutation carriers, Trail-Making Test by 22 mutation carriers. Pearson correlation coefficient for WMS-MQ with NBV, VV, PCSF. Spearman rho for all other correlations. Only p values lower than 0.05 are shown. *p ⬍ 0.05, †p ⬍ 0.01 in multivariate analysis. WMS-MQ ⫽ Wechsler Memory Scale memory quotient. 146
Neurology 72
January 13, 2009
be secondary to other MRI abnormalities in CADASIL. WMHs, lacunar infarcts, and microbleeds are mainly located in deep white matter tracts or in gray matter structures that surround the ventricles. Damage to these structures may cause central atrophy, leading to enlarged ventricles with relatively little change in peripheral CSF volumes. The results of the multivariate analysis indicated that of all neuroradiologic findings ventricular enlargement is the main determinant of cognitive decline. Our data also demonstrate that an increase in lacunar infarcts is associated with cognitive decline, whereas increase in WMHs is not. This confirms the findings from recent cross-sectional studies showing lacunar infarcts, and not WMHs, to be the main predictor of cognitive decline.5,6 We found no significant role for brain atrophy on cognition. This is in contrast to the findings of one recent study demonstrating an association between brain atrophy and cognitive decline in CADASIL.7 It is possible that this difference is caused by the smaller sample size in our study. However, it should be noted that our results do show an effect of ventricular enlargement on cognition, which can be considered a measure of central atrophy. In our study the amount of ventricular CSF volume reflects only 15% of the total amount of CSF volume. It is possible that the changes in ventricular volume are too small too cause a measurable change in total CSF volume and total brain volume. The previous research looked at global atrophy, without distinguishing between central atrophy and peripheral atrophy. It is unclear what the results would have been if ventricular volume was included separately in that study. Global brain atrophy is likely to correlate with cognitive decline in an older, more impaired CADASIL sample. It should be noted that even with 7 years of follow-up the overall rate of cognitive decline in our research population was relatively small, even in sensitive measures of executive function such as the Trail Making Test B. Possible explanations for this are that our study population also included younger MCs who do not have any noticeable cognitive dysfunction and that there may be a selection bias, because of loss to follow-up of MCs with rapid disease progression. Another possible explanation for the low rate of cognitive decline may be a learning effect, as the cognitive testing scores in the control group show a slight improvement at follow-up for most cognitive tests. After correcting for this improvement, a significant decline in MCs was found in three out of five cognitive domains. The correlations we found mainly involved executive dysfunction and memory, which is in agreement with the profile of cognitive dysfunction early in the CADASIL disease process.24
Our finding that the rate of cognitive decline was highest in the language domain may seem to disagree with this early cognitive profile in CADASIL. However, it should be noted that the CAMCOG language subtest also includes a test of word fluency, which is a test that measures both language and executive function. Additional analysis of the different subtests of CAMCOG language revealed that word fluency was the main contributor to this decline in CAMCOG language score. Word fluency also contributed most to the significant association between CAMCOG language and lacunar infarcts. A limitation of this study is the relatively small study population, which may have limited the statistical power to prove some correlations between MRI parameters and cognitive decline. Another limitation is the small age difference between MCs and nonMCs, which may have affected analysis of cognitive testing scores. However, a strength of our study is the longitudinal design, which allowed us to measure the effect of relatively small changes in MRI parameters on cognitive performance within the same individual. This advantage is especially important in our study population, because CADASIL is known for the large interindividual variability in MRI parameters and clinical phenotype.2,25 No long-term follow-up studies describing the correlations between changes in MRI parameters and cognition have been performed before. Another potential problem is that multiple statistical tests were used to associate MRI parameters and neuropsychological testing results. This may have resulted in false positive findings. We applied an a priori control of the alpha level to 0.01 to decrease this risk. However, it should be noted that the total number of significant associations and trends toward associations for specific MRI parameters was much higher than the rate of 5% false positives that would be expected on the basis of chance. Lacunar infarcts, microbleeds, and ventricular volume showed 27%, 36%, and 45% positive associations with neuropsychological testing results. In contrast, the three other MRI variables only showed 2 out of 33 associations (6%) which did not reach the significance threshold of p ⬍ 0.01. Therefore, we believe that the overall results of this study are realistic and cannot be explained by chance alone. ACKNOWLEDGMENT The authors thank Sanneke van Rooden and Mirjam van Zuiden, both from the Department of Neuropsychology, Leiden University Medical Center, for performing the neuropsychological testing and for their assistance with data entry.
Received May 13, 2008. Accepted in final form October 2, 2008. Neurology 72
January 13, 2009
147
REFERENCES 1. Joutel A, Vahedi K, Corpechot C, et al. Strong clustering and stereotyped nature of Notch3 mutations in CADASIL patients. Lancet 1997;350:1511–1515. 2. Dichgans M, Mayer M, Uttner I, et al. The phenotypic spectrum of CADASIL: clinical findings in 102 cases. Ann Neurol 1998;44:731–739. 3. Buffon F, Porcher R, Hernandez K, et al. Cognitive profile in CADASIL. J Neurol Neurosurg Psychiatry 2006;77: 175–180. 4. Charlton RA, Morris RG, Nitkunan A, Markus HS. The cognitive profiles of CADASIL and sporadic small vessel disease. Neurology 2006;66:1523–1526. 5. Liem MK, van der Grond J, Haan J, et al. Lacunar infarcts are the main correlate with cognitive dysfunction in CADASIL. Stroke 2007;38:923–928. 6. Viswanathan A, Gschwendtner A, Guichard JP, et al. Lacunar lesions are independently associated with disability and cognitive impairment in CADASIL. Neurology 2007;69:172–179. 7. Peters N, Holtmannspotter M, Opherk C, et al. Brain volume changes in CADASIL: a serial MRI study in pure subcortical ischemic vascular disease. Neurology 2006;66: 1517–1522. 8. Holtmannspotter M, Peters N, Opherk C, et al. Diffusion magnetic resonance histograms as a surrogate marker and predictor of disease progression in CADASIL: a two-year follow-up study. Stroke 2005;36:2559–2565. 9. van den Boom R, Lesnik Oberstein SA, Ferrari MD, Haan J, Van Buchem MA. Cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy: MR imaging findings at different ages–3rd-6th decades. Radiology 2003;229:683–690. 10. Spreen O, Strauss E. A Compendium of Neuropsychological Tests: Administration, Norms, and Commentary. 2nd ed. New York: Oxford University Press; 1998. 11. Roth M, Tym E, Mountjoy CQ, et al. CAMDEX. A standardised instrument for the diagnosis of mental disorder in the elderly with special reference to the early detection of dementia. Br J Psychiatry 1986;149:698–709. 12. Folstein MF, Folstein SE, McHugh PR. “Mini-mental state”: a practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 1975;12:189–198.
148
Neurology 72
January 13, 2009
13. 14.
15. 16.
17.
18.
19. 20.
21.
22.
23.
24. 25.
Wechsler DA. Standardized memory scale for clinical use. J Psychol 1945;19:87–95. Reitan R. Validity of the Trail Making Test as an indication of organic brain damage. Percept Mot Skills 1958;8: 271–276. Stroop JR. Studies of interference in serial verbal reactions. J Exp Psychol 1992;121:15–23. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders. DSM-IV. 4th ed. Washington, DC: 1994. Admiraal-Behloul F, van den Heuvel DMJ, Olofsen H, et al. Fully automatic segmentation of white matter hyperintensities in MR images of the elderly. Neuroimage 2005; 28:607–617. Werring DJ, Frazer DW, Coward LJ, et al. Cognitive dysfunction in patients with cerebral microbleeds on T2*weighted gradient-echo MRI. Brain 2004;127:2265–2275. Schneider JA. Brain microbleeds and cognitive function. Stroke 2007;38:1730–1731. Carmichael OT, Kuller LH, Lopez OL, et al. Ventricular volume and dementia progression in the Cardiovascular Health Study. Neurobiol Aging 2007;28:389–397. Anderson RC, Grant JJ, de la PR, Frucht S, Goodman RR. Volumetric measurements in the detection of reduced ventricular volume in patients with normalpressure hydrocephalus whose clinical condition improved after ventriculoperitoneal shunt placement. J Neurosurg 2002;97:73–79. Slansky I, Herholz K, Pietrzyk U, et al. Cognitive impairment in Alzheimer’s disease correlates with ventricular width and atrophy-corrected cortical glucose metabolism. Neuroradiol 1995;37:270–277. Thompson PM, Hayashi KM, de Zubicaray GI, et al. Mapping hippocampal and ventricular change in Alzheimer disease. Neuroimage 2004;22:1754–1766. Amberla K, Waljas M, Tuominen S, et al. Insidious cognitive decline in CADASIL. Stroke 2004;35:1598–1602. Opherk C, Peters N, Holtmannspotter M, Gschwendtner A, Muller-Myhsok B, Dichgans M. Heritability of MRI lesion volume in CADASIL: evidence for genetic modifiers. Stroke 2006;37:2684–2689.
Surrogate consent for dementia research A national survey of older Americans
S.Y.H. Kim, MD, PhD H.M. Kim, ScD K.M. Langa, MD, PhD J.H.T. Karlawish, MD D.S. Knopman, MD P.S. Appelbaum, MD
Address correspondence and reprint requests to Dr. Scott Kim, Bioethics Program, 300 North Ingalls St, 7C27, Ann Arbor, MI 48109
[email protected]
ABSTRACT
Background: Research in novel therapies for Alzheimer disease (AD) relies on persons with AD as research subjects. Because AD impairs decisional capacity, informed consent often must come from surrogates, usually close family members. But policies for surrogate consent for research remain unsettled after decades of debate.
Methods: We designed a survey module for a random subsample (n ⫽ 1,515) of the 2006 wave of the Health and Retirement Study, a biennial survey of a nationally representative sample of Americans aged 51 and older. The participants answered questions regarding one of four randomly assigned surrogate-based research (SBR) scenarios: lumbar puncture study, drug randomized control study, vaccine study, and gene transfer study. Each participant answered three questions: whether our society should allow family surrogate consent, whether one would want to participate in the research, and whether one would allow one’s surrogate some or complete leeway to override stated personal preferences.
Results: Most respondents stated that our society should allow family surrogate consent for SBR (67.5% to 82.5%, depending on the scenario) and would themselves want to participate in SBR (57.4% to 79.7%). Most would also grant some or complete leeway to their surrogates (54.8% to 66.8%), but this was true mainly of those willing to participate. There was a trend toward lower willingness to participate in SBR among those from ethnic or racial minority groups. Conclusions: Family surrogate consent– based dementia research is broadly supported by older Americans. Willingness to allow leeway to future surrogates needs to be studied further for its ethical significance for surrogate-based research policy. Neurology® 2009;72:149–155 GLOSSARY AD ⫽ Alzheimer disease; CI ⫽ confidence interval; HRS ⫽ Health and Retirement Study; LAR ⫽ legally authorized representatives; LP ⫽ lumbar puncture; OR ⫽ odds ratio; RCT ⫽ randomized controlled trial; SBR ⫽ surrogate-based research.
Alzheimer disease (AD) is common, incurable, and devastating. In 2000, there were 4.5 million Americans with AD and by 2050, this number is projected to be 12.5 million if no effective interventions are found to alter the trend.1 Current treatments are of only modest benefit. Much needed innovative and promising research on AD can involve invasive procedures with unpredictable risks.2,3 Although some patients with mild AD are capable of providing their own research consent, the disease usually leads to early decisional incapacity4,5 and surrogate consent is necessary for research.6,7 Federal regulations allow surrogate consent by legally authorized representatives (LAR) (45CFR46.116), but policy uncertainties have continued for three decades, for at least two reasons. First, the regulations defer to states for defining the LAR but few states have done so.8 Thus, in most jurisdictions, the common practice of obtaining family consent with subject Supplemental data at www.neurology.org Authors’ affiliations are listed at the end of the article. Dr. Kim was supported by a grant from the NIA (R01 AG029550) and Greenwall Foundation Faculty Scholars in Bioethics award. Dr. Langa was supported by a grant from the NIA (R01 AG027010) and a Paul Beeson Physician Faculty Scholars in Aging Research award. Dr. Karlawish was supported by a Greenwall Foundation Faculty Scholars in Bioethics award and the Marian S. Ware Alzheimer Program. The National Institute on Aging (NIA) provided funding for the Health and Retirement Study (U01 AG09740) which is performed at the Survey Research Center, Institute for Social Research, University of Michigan. Disclosure: The authors report no disclosures. Copyright © 2009 by AAN Enterprises, Inc.
149
assent7 occurs under a cloud of policy uncertainty. This uncertainty has led to some institutions not allowing any surrogate consent for adults in research, the Federal government temporarily shutting down such research in one institution, a self-imposed temporary moratorium at another institution, and lawsuits.6 Second, there is no consensus on how much special protection is needed when subjects are enrolled based on surrogates’ consent.9-11 Perhaps the most critical issue is how to balance the risks and the potential benefits of such research.6 Three recent state laws have diverged on this issue; internationally, there are discrepancies as well.12,13 For example, the California law14 does not set a limit on the level of allowable risk for nontherapeutic research, whereas both New Jersey15 and Virginia16 laws limit such risk to minor increase over minimal. Because policy discussions regarding surrogatebased research (SBR) have continued for three decades without a clear resolution, it may be especially important to assess the public’s views about SBR in working toward a solution. However, there have been very few attempts to understand the attitudes of the lay public or of stakeholder groups regarding SBR.17-19 We therefore designed a survey to assess the views of a nationally representative, policy-relevant sample of the general public (namely, older Americans) regarding surrogate consent for four research scenarios of varying degrees of risk and potential benefit. Because racial and ethnic minorities are underrepresented in AD research,20 we also examined their attitudes toward SBR. Finally, we examined how much latitude or leeway21,22 people are willing to confer on their surrogates. Many studies examining clinical treatment decisions have found that what people say they would want and what their surrogates think those people would want do not always match.23 These studies did not ask about leeway. But the significance of the discrepancy may depend in part on how strictly people want their preferences followed. METHODS Health and Retirement Study. The HRS is a nationally representative study of persons 51 and older, designed to investigate the health, social, and economic implications of the aging of the American population with a sample of 150
Neurology 72
January 13, 2009
more than 30,000 individuals.24 Interviews are conducted with all HRS respondents either by telephone or in person, with the latter used preferentially for those aged 80 or older. The baseline participation rate was 80%, and the reinterview participation rates for subsequent waves have been 92 to 95%. Additional HRS information is available at http://hrsonline.isr.umich.edu. Informed consent is obtained orally from all respondents prior to each HRS interview. The HRS was approved by the Institutional Review Board of the University of Michigan.
Module on SBR. Our module survey was a modified version of a paper-based survey that we had used in a previous study,17 adapted for the brief interview format of the HRS modules (appendix e-1 on the Neurology® Web site at www.neurology.org). After a brief introductory background on AD and the rationale for the survey, the subjects were given one of four SBR scenarios: a lumbar puncture (LP) study, a randomized controlled trial (RCT) of a new drug, a vaccine study, and a first-in-human gene transfer neurosurgical study. These scenarios approximated real studies in AD.2,3 Then the subjects were asked three questions (see table 2, column 1). The module was administered to a random subsample of the 2006 HRS participants. Of the 1,517 subjects approached for the module, two declined at the outset. The remaining 1,515 subjects were randomized to one of the four SBR scenarios: 374 to the LP scenario, 398 to the new drug RCT scenario, 375 to the vaccine scenario, and 368 to the gene transfer scenario. Depending on the survey question, 52 to 71 persons (out of 1,515) answered “don’t know” or refused to answer; they were excluded from the analyses below. The randomization was successful in that the groups were not significantly different on age, gender, race/ethnicity, education, net worth, marital status, or religion.
Questions from Health and Retirement Study database. In analyzing our module survey responses, we included additional variables from the HRS parent survey listed in table e-1. We also examined four additional HRS parent survey questions for the leeway question analysis: 1) “How much do they [spouse or partner] really understand the way you feel about things?” 2) “How much can you rely on them if you have a serious problem?” The same questions were also asked regarding their children. The response choices were a lot, some, a little, or not at all. These questions were chosen because surrogates of patients with AD tend to be spouses and children. We wished to examine if the nature of those relationships predict the respondent’s attitudes regarding leeway, since identifying strong predictors of leeway could have policy implications. Although responses to these extra questions are available only in subsets of participants who have spouses or children (ranging from n ⫽ 494 to n ⫽ 633 of our module sample of 1,515, depending on the question), neither the availability nor distribution of these variables was associated with the four scenarios.
Analysis. Descriptive. For each SBR scenario, the proportions of respondents who find the SBR research acceptable from a societal perspective and from a self perspective were calculated. The proportions of respondents willing to allow no, some, or complete leeway were calculated for each scenario. Scenario and perspective effects. The effect of the scenario on willingness to allow family consent for SBR, on willingness to participate in SBR, and on willingness to allow leeway were each assessed using 2 tests. We assessed the effect of perspective for each scenario, using the McNemar test. Race/ethnicity effect. Race/ethnicity categories were defined as non-Hispanic white, non-Hispanic black, Hispanic, or other.
Table 1
Characteristics of 1,515 respondents to the 2006 Health and Retirement Study (HRS) module on surrogate consent for dementia research*
Subject characteristics
Total (n ⴝ 1,515), n (%)
Age, y <60
388 (32.5)
60–69
508 (35.0)
70–79
401 (20.2)
>80
218 (12.3)
Gender Men
581 (42.3)
Women
934 (57.7)
Race/ethnicity White
1,126 (81.7)
Black
223 (10.4)
Hispanic
135 (6.2)
Other
31 (1.6)
Education, y <12
862 (51.4)
13–15
338 (24.3)
>16
310 (24.3)
Net worth <$45,000
342 (20.2)
$45,000–$163,000
385 (24.3)
$163,001–$435,000
349 (25.8)
>$435,000
389 (29.8)
Marital status Unmarried
546 (35.0)
Married
967 (65.0)
*Cell values are N (weighted percentages) with weighted percentages derived using the HRS respondent population weights.
A logistic regression model was fit for each scenario using responses to each of the three main survey questions as the dependent variable. Since our primary interest was in assessing how the responses to the three main survey questions vary by race/ ethnicity, adjusting for potential confounders and other variables potentially related to attitudes toward SBR, the logistic regression models were adjusted for all of the variables in table 1 and table e-1. Leeway question analysis. We examined two types of models. First, we evaluated the predictors of willingness to allow leeway (dichotomized as some/complete leeway vs not) using logistic regression models, where the potential predictors examined were the HRS variables in table 1 and table e-1, the scenario dummy variables, as well as a variable for willingness to participate (because it turned out to be a strong predictor in bivariate analysis). The covariates were modeled carefully to assess their appropriate functional relationships to the response variable by using polynomials (e.g., for variables such as age) and categorical or dichotomized dummy variables. Second, similar but separate logistic regression models were used to test if allowing leeway was
associated with relying on spouse or children and feeling understood by spouse or children. Statistical analyses were performed using Stata 9.2 (Stata Corp, College Station, TX). All analyses except for McNemar tests were adjusted for the complex sampling design of the HRS survey using the HRS respondent level sampling weights (http:// hrsonline.isr.umich.edu/meta/tracker/desc/wghtdoc.pdf). Dr. H.M. Kim conducted the statistical analysis.
Table 2 summarizes the responses for the three main survey questions. Most of the respondents are supportive of allowing families to make surrogate consent decisions for dementia research (67.5% to 82.5%). Even for a first-inhuman gene transfer scenario, nearly 68% state that our society should allow families to make such surrogate decisions. The response patterns were significantly different across the four scenarios for the societal question and for the self perspective question, and the pattern trended toward a difference for the leeway question. The proportion of those willing to participate in the various SBR scenarios is generally similar to the level of support for family surrogate consent, but a significantly greater proportion would allow family consent in the vaccine study than would themselves want to participate (odds ratio [OR] ⫽ 3.09; 95% confidence interval [CI] ⫽ 1.89 –5.25; p ⬍ 0.001). For other scenarios, the societal to self perspective comparisons showed an OR of 1.24 (95% CI ⫽ 0.77–2.00; p ⫽ 0.42) for the lumbar puncture study, 1.59 (0.96 –2.68; p ⫽ 0.07) for the drug RCT scenario, and 1.29 (0.81–2.06; p ⫽ 0.31) for the gene transfer scenario. RESULTS
Race/ethnicity. The relationship between race/eth-
nicity and attitudes toward the various SBR scenarios are summarized in table 3. The direction of the race/ethnicity effect for the societal perspective question is not consistent across scenarios and none of the effects are significant. In absolute terms, the majority in all race/ethnicity groups responded that society should allow family consent for all scenarios. For the self perspective question, the overall pattern trends toward black and Hispanic respondents being less willing to participate in SBR, although the only significant results are that black respondents are less willing than white respondents to participate in the drug RCT and Hispanic respondents are less willing to participate in the gene transfer study. For the leeway issue (not shown in the table), the only significant difference is that Hispanic respondents are less likely than white respondents to allow some/complete leeway for the gene transfer study (adjusted OR ⫽ 0.26; 95% CI ⫽ 0.09 – 0.79; p ⫽ 0.02). Willingness to allow leeway. Depending on the scenario, between 55 and 67% would allow their surroNeurology 72
January 13, 2009
151
Table 2
Proportion* of respondents providing positive responses for each of the four surrogate-based research scenarios for the three main survey questions
Questions
Lumbar puncture
Drug RCT
Vaccine
Gene transfer
p Value
72.0
82.5
70.5
67.5
⬍0.001
70.8
79.7
57.4
68.7
⬍0.001
41.4
33.2
45.2
39.6
39.3
40.6
37.9
39.2
19.3
26.2
16.9
21.1
Societal perspective: If patients cannot make their own decisions about being in studies like this one, should our society allow their families to make the decision in their place? (n ⴝ 1,463) [allow, not allow] Self perspective: Suppose you wanted to give a close family member instructions for the future, in case you ever became unable to make decisions for yourself. Would you say you would want to participate in the study? (n ⴝ 1,444) [yes, no] Leeway question: How much freedom or leeway would you give the close family member to go against your preference and instead [opposite of answer to 2: enroll/not enroll] you in the study? (n ⴝ 1,456) [In descending order in the column: no leeway, some leeway, complete leeway]†
0.09
*Weighted proportions. The respondents answered the three questions for one of the four scenarios randomly assigned. Those responding “don’t know” or “refuse to answer” are not included. †The responses are listed, from top to bottom: no leeway; some leeway; complete leeway. RCT ⫽ randomized controlled trial.
gates some or complete leeway to decide against their stated wishes regarding SBR participation. Among those who do not desire to participate in SBR, 26% would allow some or complete leeway; among those who desire to participate, 76% would allow some or complete leeway. Table 4 summarizes the adjusted ORs of the variables retained in the final logistic regression model to predict leeway question responses. The strongest predictor of whether one is willing to allow at least some leeway remains whether one is willing to participate in research, with an adjusted OR ⫽ 8.31. There were
Table 3
other significant, although weaker, predictors of willingness to grant leeway. Married people, women, those in excellent health, and respondents to the gene transfer scenario (although not among those who considered religion very important) were more likely to give at least some leeway than not. In separate models, the more the respondent felt understood by a child, the more likely he or she would allow leeway (adjusted OR ⫽ 1.37, 95% CI ⫽ 1.07–1.75; p ⫽ 0.01), corresponding to a 37% increase in the likelihood of willingness to allow leeway with every one level increase in feeling under-
Race/ethnicity and attitudes toward four surrogate-based dementia research scenarios of varying risks and benefits Lumbar puncture
Drug RCT
Vaccine
Gene transfer
%*
OR† (95% CI)
%*
OR† (95% CI)
%*
OR† (95% CI)
%*
OR† (95% CI)
1.0
70.5
1.0
66.5
1.0
Societal perspective White
71.3
1.0
85.1
Black
75.8
1.36 (0.43–4.37)
76.7
0.40 (0.15–1.08)
69.5
0.82 (0.29–2.31)
73.7
2.25 (0.93–5.43)
Hispanic
66.4
0.58 (0.15–2.27)
70.3
0.35 (0.10–1.19)
69.2
1.10 (0.38–3.15)
77.8
1.94 (0.52–7.19)
Self perspective White
71.1
1.0
81.8
1.0
59.0
1.0
71.5
1.0
Black
69.4
0.82 (0.32–2.07)
60.1
0.27‡ (0.08–0.86)
53.5
0.54 (0.17–1.69)
63.8
1.00 (0.41–2.46)
Hispanic
65.3
0.68 (0.15–3.12)
78.6
0.80 (0.25–2.60)
47.5
0.97 (0.40–2.38)
49.2
0.26‡ (0.08–0.88)
*All percents are weighted. †Ratio of odds for allowing surrogate-based research for a race category relative to white Americans. See table 1 and table e-1 for covariates included in the models to generate these adjusted odds ratios. ‡p ⬍ 0.05. RCT ⫽ randomized controlled trial; OR ⫽ odds ratio; CI ⫽ confidence interval. 152
Neurology 72
January 13, 2009
Table 4
Factors associated with subjects’ willingness to allow some/complete leeway to their surrogates regarding future participation in dementia research
Variables
Adjusted* OR (95% CI)
Want to participate (vs not want to)
8.31 (6.09–11.34)
Female (vs male)
1.46 (1.02–2.08)
Married (vs not married)
1.50 (1.05–2.15)
Excellent self-rated health (vs less than excellent)
1.68 (1.03–2.75)
Gene transfer scenario (vs other scenarios)
2.07 (1.02–4.20)
Religion is very important (vs somewhat or not too important)
1.23 (0.87–1.75)
Interaction of gene transfer scenario x religion very important
0.47 (0.24–0.91)
*This final multiple logistic regression model is also adjusted for race/ethnicity and religion in addition to variables included in the table. OR ⫽ odds ratio; CI ⫽ confidence interval.
stood (out of four levels, ranging from “not at all” to “a lot”). Being understood by one’s spouse or partner had a similar effect (adjusted OR ⫽ 1.44, 95% CI ⫽ 1.07–1.94; p ⫽ 0.02). The degree of reliance on one’s spouse or child, however, was not related to one’s willingness to allow leeway. DISCUSSION We found broad majority support— ranging from 68% to 83%, depending on the scenario—for a societal policy of family surrogate consent for AD research. A previous survey of adults at increased risk for AD also showed majority support for family consent for the same four SBR scenarios.17 Another previous study of research participants at risk for AD19 found very high willingness to participate in SBR of relatively low risk. A survey in Quebec18 showed similar results but its emphasis on examining a variety of surrogate types (legal guardian, family) makes it difficult to compare with our results. The responses from the societal and self perspectives are closely related, but they should not be treated as equivalent. Personal willingness to participate is not a proxy for views about the ethical appropriateness of SBR because many people may choose not to participate for a variety of reasons. If the point is to characterize public opinion on the ethical permissibility of SBR, then respondents’ views from a societal perspective should be primary. Of course, it may not be ideal if many more people are willing to allow SBR as a societal practice than those who are willing to participate. However, even for the vaccine
study scenario, in which the gap is greatest, the majority (57%) of respondents still desire to participate. The significant underrepresentation of racial and ethnic minorities in AD research is a perennial concern.20 Recent studies have shown that racial and ethnic minorities, despite the legacy of abuses such as the Tuskegee study, are similar to whites in their attitudes toward and participation in research.25,26 We found that regarding the social policy of family surrogate consent for SBR, the differences among racial and ethnic groups are small and the overall level of support is high. In terms of a desire to participate in SBR, the overall pattern among minority respondents is a somewhat lower willingness to participate. However, responses vary a great deal by scenario. For instance, although black Americans are significantly less likely to volunteer for a drug RCT, they appear more likely than whites to endorse the gene transfer neurosurgical study for AD with family consent (and are no less willing to participate). Leeway is an important concept because it may signify a subject’s recognition that the future cannot be pinned down in detail and it allows subjects to convey their trust in their surrogates who will have more complete information at the time of actual decision-making. The majority of our respondents would allow some or complete leeway for all the scenarios. However, it should be noted that a significant minority (up to 45% for the gene transfer scenario) would not allow leeway. Those who are willing to participate in research are highly likely to allow leeway. Relational factors (such as feeling understood by their potential surrogates) clearly play a role as well—and this may have important consequences on how we view current practices. Because there is ample cross-sectional evidence that even impaired dementia patients generally express “reasonable” preferences (i.e., similar to healthy controls) for treatment27 and for research participation,28 if such preferences are predictive of a desire for conferring leeway then the current practice of surrogate consent with subject assent seems well supported. However, only a longitudinal study would provide direct evidence regarding whether a potential subject’s current preference (as reflected in the subject’s assent) is reflective of a longstanding desire for granting leeway. Even among those who state that they do not desire to participate in SBR, a significant minority (over a quarter) are willing to confer some or complete leeway on their future surrogates. If these subjects are added to those who are willing to participate in the four SBR scenarios, the proportions of potential participants would rise to approximately 80% for the LP study, 88% for the drug RCT, 79% for the Neurology 72
January 13, 2009
153
vaccine study, and 74% for the gene transfer study. A recent study of elderly subjects from senior centers and primary care clinics regarding an LP SBR scenario found similar results.22 However, this point needs to be balanced by the fact that the surrogates of those who are willing to participate and are willing to grant leeway will have the latitude to override that willingness to participate; leeway does not necessarily imply an increase in enrollment. This study’s main limitation is that the scientific and policy issues surrounding surrogate consent for research are quite complicated. This brief survey captures only a snapshot impression regarding SBR. It may be that as people learn and reflect more, their attitudes may change. We are therefore currently conducting a series of all-day deliberative democracy sessions with laypersons and caregivers regarding SBR.29 Another limitation of the study is that it focuses on dementia research and may not generalize to other areas of SBR.30,31 An important strength of this study is its large and representative sample with an excellent response rate. The Federal government is revisiting the issue of surrogate consent and recently solicited public input.32 As a nationally representative study on the topic, our study provides new data for an important policy discussion. The primary ethical goals in overseeing SBR are to minimize the potential harms of SBR (already mandated by the current regulations) and to minimize the potential for enrolling persons whose values may not be compatible with participation in SBR. In this regard, the current evidence, including the results of our study, provides some guidance. First, it appears that even for invasive studies the prior probability of an older American’s willingness to participate in SBR is high. Second, even among those who are not willing to participate, there is a sizable minority who are willing to confer some leeway on their surrogates. More research on attitudes toward leeway may provide further guidance to policy makers and to researchers at the time subjects are approached for research. Third, even when patients suffer from dementia with impaired decisional capacity, they may still retain some ethically relevant abilities. Persons with AD express personal choices regarding research participation that are similar to those expressed by competent adults, suggesting that risk-related preferences are relatively preserved.28 Thus, the practice of requiring the assent of even incapable subjects may provide another level of assurance. Patients with AD may also have relatively preserved abilities to appoint a trusted relative to help make research decisions for them.33 Finally, it is quite likely that persons acting as family surrogates will be protective and provide a 154
Neurology 72
January 13, 2009
gatekeeper function.17 Future research should aim to provide further guidance on how to minimize the enrollment of those whose values are not compatible with SBR. AUTHORS’ AFFILIATIONS From the Center for Behavioral and Decision Sciences in Medicine, Bioethics Program, and Department of Psychiatry (S.Y.H.K.), Center for Statistical Consultation and Research (H.M.K.), and Division of General Medicine, Department of Medicine, and Institute for Social Research (K.M.L.), University of Michigan, Ann Arbor; Veterans Affairs Center for Practice Management and Outcomes Research (K.M.L.), Ann Arbor, MI; Department of Medicine (J.H.T.K.), Division of Geriatrics, Alzheimer’s Disease Center, Center for Bioethics, and the Leonard Davis Institute for Health Economics, University of Pennsylvania; Department of Neurology (D.S.K.), Mayo Clinic, Rochester, MN; and Division of Law, Ethics, and Psychiatry (P.S.A.), Department of Psychiatry, Columbia University and New York State Psychiatric Institute, New York, NY.
Received August 12, 2008. Accepted in final form October 3, 2008.
REFERENCES 1. Hebert LE, Scherr PA, Bienias JL, Bennett DA, Evans DA. Alzheimer disease in the US population: prevalence estimates using the 2000 census. Arch Neurol 2003;60:1119– 1122. 2. Orgogozo JM, Gilman S, Dartigues JF, et al. Subacute meningoencephalitis in a subset of patients with AD after A42 immunization. Neurology 2003;61:46–54. 3. Tuszynski MH, Thal L, Pay M, et al. A phase 1 clinical trial of nerve growth factor gene therapy for Alzheimer disease. Nat Med 2005;11:551–555. 4. Kim SYH, Caine ED, Currier GW, Leibovici A, Ryan JM. Assessing the competence of persons with Alzheimer’s disease in providing informed consent for participation in research. Am J Psychiatry 2001;158:712–717. 5. Okonkwo O, Griffith HR, Belue K, et al. Medical decision-making capacity in patients with mild cognitive impairment. Neurology 2007;69:1528–1535. 6. Kim SYH, Appelbaum PS, Jeste DV, Olin JT. Proxy and surrogate consent in geriatric neuropsychiatric research: update and recommendations. Am J Psychiatry 2004;161: 797–806. 7. Karlawish JHT, Knopman D, Clark CM, et al. Informed consent for Alzheimer’s disease clinical trials: a survey of clinical investigators. IRB Ethics Hum Res 2002;24:1–5. 8. Saks E, Dunn L, Wimer J, Gonzales M, Kim S. Proxy consent to research: legal landscape. Yale J Health Law Policy Ethics 2008;8:37–78. 9. National Bioethics Advisory Commission. Research Involving Persons with Mental Disorders That May Affect Decisionmaking Capacity. Volume 1 ed. Rockville, MD: NBAC; 1998. 10. Maryland Attorney General’s Research Working Group. Final Report of the Attorney General’s Research Working Group. 1998. 11. New York Department of Health Advisory Work Group on Human Subject Research Involving the Protected Classes. Recommendations on the oversight of human subject research involving the protected classes. 1999. 12. Geiselmann B, Helmchen H. Demented subjects’ competence to consent to participate in field studies: the Berlin Ageing Study. Medicine Law 1994;13:177–184.
13.
UNESCO. Universal Declaration on Bioethics and Human Rights. 10-19-2005. Paris: UNESCO. 14. Amendment to Section 24178 of the California Health and Safety Code. 2002. 15. New Jersey. Access to Medical Research Act. 26, 14.114.5. 2008. 16. Code of Virginia. Title 32.1, Section 162.16-162.19. 2002. 17. Kim SYH, Kim H, McCallum C, Tariot P. What do people at risk for Alzheimer’s disease think about surrogate consent for research? Neurology 2005;65:1395–1401. 18. Bravo G, Paquet M, Dubois MF. Opinions regarding who should consent to research on behalf of an older adult suffering from dementia. Int J Soc Res Pract 2003;2:49–65. 19. Wendler D, Martinez RA, Fairclough D, Sunderland T, Emanuel E. Views of potential subjects toward proposed regulations for clinical research with adults unable to consent. Am J Psychiatry 2002;159:585–591. 20. Olin JT, Dagerman KS, Fox LS, Bowers B, Schneider LS. Increasing ethnic minority participation in Alzheimer disease research. Alzheimer Dis Assoc Disord 2002;16 suppl 2:S82–S85. 21. Sehgal A, Galbraith A, Chesney M, Schoenfeld P, Charles G, Lo B. How strictly do dialysis patients want their advance directive followed? JAMA 1992;267:59–63. 22. Karlawish J, Rubright J, Casarett D, Cary M, Tenhave T, Sankar P. Do older adults support enrolling noncompetent subjects in research that does not benefit the subjects? Am J Psychiatry 2008 (in press). 23. Shalowitz DI, Garrett-Mayer E, Wendler D. The accuracy of surrogate decision makers: a systematic review. Arch Intern Med 2006;166:493–497. 24. Juster F, Suzman R. An overview of the Health and Retirement Study. J Hum Resour 1995;30 (suppl):S7–S56.
25.
Wendler D, Kington R, Madans J, et al. Are racial and ethnic minorities less willing to participate in health research? PLoS Med 2006;3. 26. Katz R, Green L, Kressin N, Claudio C, Wang M, Russell S. Willingness of minorities to participate in biomedical studies: confirmatory findings from a follow-up study using the Tuskegee Legacy Project Questionnaire. J Nat Med Assoc 2007;99:1052–1060. 27. Marson DC, Ingram KK, Cody HA, Harrell LE. Assessing the competency of patients with Alzheimer’s disease under different legal standards. A prototype instrument. Arch Neurol 1995;52:949–954. 28. Kim SYH, Cox C, Caine ED. Impaired decision-making ability and willingness to participate in research in persons with Alzheimer’s disease. Am J Psychiatry 2002;159:797– 802. 29. Gutmann A, Thompson D. Deliberating about bioethics. Hastings Center Report 1997;27:38–41. 30. Silverman HJ, Luce JM, Schwartz J. Protecting subjects with decisional impairment in research: the need for a multifaceted approach. Am J Respir Crit Care Med 2004;169: 10–14. 31. Ciroldi M, Cariou A, Adrie C, et al. Ability of family members to predict patient’s consent to critical care research. Intens Care Med 2007;33:807–813. 32. Request for Information and Comments on Research That Involves Adult Individuals With Impaired Decisionmaking Capacity. Fed Reg 72[171], 50966 –50970. 9-52007. 33. Kim SYH, Appelbaum PS. The capacity to appoint a proxy and the possibility of concurrent proxy directives. Behav Sci Law 2006;24:469–478.
AAN Store: Your One-Stop Shop for Practice Tools Looking for an easy, convenient way to order your practice essentials? Visit the AAN Store online for an array of quality products backed with our high standards. Save up to 20 percent with your AAN member discount. The AAN Store is your source for diagnostic tools and references, including: ● Exclusive AAN Products (pocket guidelines, coding guides, syllabi on CD, patient education books, and brochures) ● Top-quality Diagnostic Tools (reflex hammers, visual and sensory testing tools, pocket guidelines) ● Educational Resources for You, Your Practice, and Your Patients (CDs, DVDs, books, 3D models, posters, and brochures) Shop quickly and easily at www.aan.com/aanstore today!
Neurology 72
January 13, 2009
155
␣-Internexin expression identifies 1p19q codeleted gliomas
F. Ducray, MD E. Crinie`re, MSc A. Idbaih, MD, PhD K. Mokhtari, MD Y. Marie, MSc S. Paris, MSc S. Navarro, MD F. Laigle-Donadey, MD C. Dehais, MD J. Thillet, PhD K. Hoang-Xuan, MD, PhD J.-Y. Delattre, MD M. Sanson, MD, PhD
Address correspondence and reprint requests to Dr. M. Sanson, Service de Neurologie Mazarin, Hoˆpital de la Salpeˆtrie`re, 47 Bd de l’Hoˆpital, 75013 Paris, France
[email protected]
ABSTRACT
Background: ␣-Internexin (INA) is a proneural gene encoding a neurofilament interacting protein that is upregulated in some gliomas, particularly oligodendrogliomas.
Methods: INA expression was evaluated by immunohistochemistry in a series of 122 gliomas, and correlated to the 1p19q codeletion, a favorable prognostic marker of oligodendroglial tumors.
Results: INA expression was strong (⬎10% positive cells) in 22 cases (22 oligodendroglial tumors and 0 astrocytic tumors), weak (⬍10% cells) in 14 cases (12 oligodendroglial tumors, 2 glioblastoma with an oligodendroglial component, and 0 astrocytic tumors), and negative in 86 cases (49 oligodendroglial tumors, 9 glioblastoma with an oligodendroglial component, and 28 astrocytic tumors). Among the 27 tumors exhibiting the 1p19q codeletion (all with an oligodendroglial phenotype), INA was detected in 96% (26/27, 18 strong, 8 weak) as compared to 11% (10/95, 4 strong, 6 weak) in the tumors without 1p19q codeletion (with an oligodendroglial or an astrocytic phenotype) (p ⬍ 0.001). In oligodendroglial tumors, INA expression specificity for 1p19q codeletion was 86%, sensitivity 96%, positive predictive value 76%, and negative predictive value was 98%. The prognostic impact of INA expression could be evaluated in grade III oligodendroglial tumors. Similar to 1p19q deletion, positive INA expression was correlated with better progression-free survival (52.6 vs 8.7 months [p ⫽ 0.001]) and overall survival (121.1 vs 31.4 months [p ⫽ 0.0001]). Conclusion: ␣-Internexin (INA) expression appears to be a simple, reliable prognostic marker and a surrogate marker of 1p19q codeletion. Neurology® 2009;72:156–161 GLOSSARY CGH ⫽ comparative genomic hybridization array; FISH ⫽ fluorescent in situ hybridization; GBMO ⫽ glioblastoma with oligodendroglial component; HR ⫽ hazard ratio; INA ⫽ ␣-internexin; LOH ⫽ loss of heterozygosity; OS ⫽ overall survival; PFS ⫽ progression-free survival.
The chromosomes 1p19q codeletion is a specific molecular signature, strongly associated with an oligodendroglial phenotype and favorable prognosis.1-3 It has recently been shown to be mediated by a specific t(1;19)(q10;p10) translocation,4 but reliable detection of 1p19q codeletion requires an appropriate and expensive technique, such as comparative genomic hybridization array (CGH-array). Indeed, most widely used techniques, such as loss of heterozygosity (LOH) analysis, may not distinguish this signature from partial distal 1p and 19q deletion or gain, which have radically different prognostic implications.2 There is, therefore, a need for a simple, widely available, and reliable technique to identify oligodendrogliomas with 1p19q codeletion. To date, no oligodendrocyte marker has been shown to be of clinical relevance.5 Recently, a proneural expression profile has been related to better prognosis.6 Using gene expression profiling, we found that 1p19q codeleted gliomas display a proneural gene expres-
From Unite´ INSERM U711 (F.D., E.C., A.I., Y.M., S.P., J.T., K.H.-X., J.Y.-D., M.S.), Service de Neurologie Mazarin (F.D., A.I., F.L.-D., C.D., K.H.-X., J.Y.-D., M.S.), Laboratoire de Neuropathologie (K.M.), and Service de Neurochirurgie (S.N.), CHU Pitie´-Salpeˆtrie`re, Universite´ Paris VI, France. Supported by a grant of the Institut National du Cancer (INCA, PL046), the De´le´gation a` la Recherche Clinique (MUL032), and by the Ligue Nationale contre le Cancer. Disclosure: The authors report no disclosures. 156
Copyright © 2009 by AAN Enterprises, Inc.
Table
␣-Internexin (INA) expression according to histology and 1p19q codeletion INA expression
Histology
Negative
Weak
Strong
Total
No 1p19q codeletion Pilocytic astrocytoma
2
0
0
2
Grade 2 astrocytoma
6
0
0
6
Grade 3 astrocytoma
6
0
0
6
14
0
0
14
9
2
0
11
Glioblastoma Glioblastoma with oligodendroglial component Grade 2 oligodendroglioma
5
1
3
9
Grade 3 oligodendroglioma
11
0
1
12
Grade 2 oligoastrocytoma
13
0
0
13
Grade 3 oligoastrocytoma
19
3
0
22
85
6
4
95
Grade 2 oligodendroglioma
1
4
8
13
Grade 3 oligodendroglioma
0
4
8
12
Grade 3 oligoastrocytoma
0
0
2
2
1
8
18
27
86
14
22
122
Total nondeleted 1p19q codeletion
Total deleted Total
Assay [ref QT00232148; Qiagen]), Alas forward primer 5=TGCAGTCCTCAGCGCAGT3=, and Alas reverse primer 5=TGGCCCCAACTTCCATCAT3=. 2 tests were used to compare INA protein expression between the two groups (1p19q codeleted and non-codeleted gliomas). Wilcoxon test was used to compare the amount of INA transcript (quantified by RT-PCR) between INA⫹ and INA⫺ samples and between 1p19q deleted and nondeleted tumors. Overall survival (OS) and progression-free survival (PFS) were studied in grade III oligodendroglial tumors (oligodendrogliomas and oligoastrocytomas) according to INA protein expression. OS was defined as the time from diagnosis to death or last follow-up. PFS was defined as the time from diagnosis to recurrence or last follow-up. Patients lost to follow-up were censored on the last known day of life. Kaplan-Meier methods were used to obtain survival curves, and a two-sided Log-rank test was used to compare strata. A multivariate Cox regression model was used to adjust the effect of INA expression on survival or PFS for age, type of surgery, Karnofsky index, and postoperative treatment. Independent covariates were selected by stepwise methods.
The histology, 1p19q status, and INA expression of the 122 gliomas are reported in the table: 27 patients had a 1p19q codeletion (13 grade II oligodendrogliomas [OII], 12 grade III oligodendrogliomas [OIII], 2 grade III oligoastrocytomas [OAIII]), 14 had an EGFR amplification (1 grade II oligoastrocytoma [OAII], 2 OIII, 5 OAIII, 1 glioblastoma with oligodendroglial component [GBMO], 5 glioblastomas), and no tumor had both abnormalities (figure 1). According to the criteria expressed above, independent lecture (by K.M. and E.C.) was consistent in 118/122 cases (97%). The four discrepant cases were resolved after reexamination (K.M., E.C., and M.S.). INA expression was completely negative in 86 tumors (70.5%), strongly positive in 22 (18%), and weak in 14 (11.5%). INA expression was tightly related to the oligodendroglial phenotype (particularly “pure” oligodendroglioma) being present in 63% of pure oligodendrogliomas (16/22 OII, 13/24 OIII), 15% of mixed tumors (0/13 OAII, 5/24 OAIII, 2/11 GBMO), and none of the 28 pure astrocytic tumors (p ⬍ 10⫺4). INA expression was tightly related to 1p19q codeletion (table and figure 2). Indeed, 26/27 tumors (18 strong, 8 weak) with 1p19q deletion expressed INA compared to only 10/95 (4 strong, 6 weak) nondeleted tumors (p ⬍ 10⫺4). Estimated specificity of INA testing for 1p19q codeletion was 90%, and sensitivity was 92%. Negative INA expression was strongly predictive of nondeleted tumor with an estimated probability of 98%. A positive expression, however, was predictive of 1p19q deletion with a probability of 74%, but, in this setting, it is important to distinguish a strong INA expression (involving more than 10% of tumor cells), predictive of 1p19q deleted tumor with an estimated probability of 82%, from a low expression RESULTS
sion profile and that INA is one of the most overexpressed neuronal genes in gliomas with 1p19q codeletion.7 We tested INA immunohistochemistry in a large series of gliomas as a prognostic marker and a surrogate marker for the 1p19q codeletion signature which was assessed by CGHarray analysis. METHODS Glioma samples were snap frozen in liquid nitrogen and analyzed by CGH-array as previously described.8 The corresponding clinical annotations were collected on the neurooncology department database. Immunohistochemistry was carried out on 4 m paraffin sections of formalin-fixed tumor samples using antibodies directed against internexin neuronal intermediate filament protein ␣ (INA) (Novocastra, http://vision-bio.com). Slides were read independently by E.C. and K.M. Discrepant cases were reviewed together with M.S. to reach a consensus. Labeling was defined as strong (⬎10% of positive cells), weak (⬍10% of positive cells), or negative (no positive tumor cells detected). Total RNA was isolated according to the manufacturer’s instructions (Qiagen, Rneasy Lipid Tissue Mini Kit). cDNA was prepared from 1 g RNA using a combination of random primers (Promega, France) and MMLV reverse transcriptase (Invitrogen). Real-time PCR (RT-PCR) was carried out in a 25 L volume containing 5 L of 20-fold diluted cDNA, 600 nM of each primer, and 12.5 L of 2X Syber Green buffer (Abgene, Courtaboeuf, France). Values were normalized to the expression levels of a housekeeping gene, ALAS1 (delta-aminolevulinate synthase1). Primers were as follows: 1X INA (Quantitect primer
Neurology 72
January 13, 2009
157
Figure 1
Genomic data (CGH array) and INA expression (immunohistochemistry)
(A) Genomic profile of the 122 gliomas. Each column corresponds to a sample. Lines correspond to BACs located on chromosomes 1p, 7, 9p, 10, and 19. These chromosomes contain the main genomic regions altered in gliomas. Yellow indicates normal genomic copy number, green indicates a loss (see the loss of 1p and 19q, the loss of 9p and 10), and red indicates a gain (see the gain of chromosome 7). A dark red indicates a high copy number gain (i.e., an amplification, as shown for EGFR). On the top of the figure, blue indicates positive (strong or weak) and purple indicates negative ␣-internexin (INA) expression. The order of the samples, from left to right, is the following: gliomas with 1p19q codeletion, gliomas with EGFR amplification characterized by a red dark signal on 7p12 (white arrow), gliomas with other genomic characteristics. (B) Top: representative strong INA expression in an oligodendroglioma with 1p19q codeletion. Middle: weak INA expression (i.e., involving less than 10% of the cells) in an oligoastrocytoma without 1p19q codeletion. Bottom: negative INA expression in a glioblastoma.
(involving less than 10% of tumor cells) found in 12% of the tumors with weak predictive value (8/ 14). Interestingly, no tumor with epidermal growth factor receptor gene (EGFR) amplification expressed INA (figure 1). RNA was available for 48 tumors. RT-PCR results are indicated on figure 2. INA mRNA expression was strongly associated with immunohistochemistry data (p ⬍ 10⫺4) and with 1p19q codeletion (p ⬍ 10⫺4). 158
Neurology 72
January 13, 2009
When considering only oligodendroglial tumors grade II and III (table), estimated specificity for 1p19q codeletion was 86% (48/56), sensitivity was 96% (26/27), positive predictive value was 76% (26/34), and negative predictive value was 98% (48/49). We then analyzed the impact of INA expression on survival of grade 3 gliomas (follow-up was insufficient in grade 2 gliomas, and INA was expressed in only two grade 4 gliomas). As expected, INA expres-
Figure 2
INA expression correlates with 1p19q deletion and predicts outcome in grade III gliomas
(A) Percentage of gliomas with and without 1p19q codeletion with strong, weak, and negative ␣-internexin (INA) immunostaining. (B) Relative INA gene expression assessed by real time PCR; from left to right: strong, weak, negative INA protein expression, gliomas with and without 1p19q codeletion. (C-I/C-II) Overall and progression-free survival according to INA protein expression in grade III oligodendroglial tumors (OIII and OAIII).
sion reflected the strong favorable prognostic impact of 1p19q. In anaplastic oligodendroglial tumors, PFS was 52.6 vs 8.7 months (p ⫽ 0.001) and OS was 121.1 vs 31.4 months (p ⫽ 0.0001) for positive tumors vs negative tumors (figure 2). The impact of INA expression on survival was comparable to 1p19q codeletion (PFS was 52.6 vs 10.4 months [p ⫽ 0.0007] and OS was 121.1 vs 32.9 months [p ⫽ 0.0012]), for deleted tumors vs nondeleted tumors. In addition to INA expression and 1p19q codeletion, age, grade, and type of surgery (biopsy vs tumor removal) were all related to survival. Cox regression multivariate analysis indicated that the hazard ratio (HR) associated with INA expression, adjusted for grade, age, and type of surgery, remained significant (HR ⫽ 0.29; 95% confidence interval, 0.15 to 0.54; p ⫽ 0.0001); however, when adjusted for 1p19q codeletion, HR associated with INA expression was
no longer significant, confirming that both variables are tightly related. Oligodendroglial tumors have better prognosis and are more chemosensitive than astrocytic tumors,9 but histologic diagnosis is subjective, giving rise to discrepancies. The 1p19q codeletion status (related to an unbalanced t(1;19)(q10;p10)) represents by far the strongest prognostic factor in oligodendrogliomas. In this context, a critical issue for clinicians is not only differentiating oligodendroglioma from astrocytoma but above all identifying the 1p19q status of the tumor. Direct test for translocation is rarely performed in clinical practice: 1p19q codeleted oligodendroglioma is difficult to grow and it is therefore unrealistic to perform a karyotype. The translocation can be suspected using a fluorescent in situ hybridization (FISH) fusion test
DISCUSSION
Neurology 72
January 13, 2009
159
with two probes flanking the breakpoint on the 1q19p derived chromosome.4 However, in practice, the most widely available techniques detect the 1p19q codeletion and not the translocation. They include FISH, LOH, multiplex ligation-dependent probe amplification (MLPA),10 and CGH array; the last is the most complete, reliable, and sophisticated but also the most expensive technique. In fact, all of them have their limitations: contamination with normal cells may impair the genetic analysis; LOH and FISH are inadequate to distinguish whole 1p and whole 19q loss from partial 1p loss frequently associated with 19p and 19q loss, and these prognosis are radically different2; and the amount of available tissue, in case of biopsy, may be insufficient for CGH array. This study suggests that INA expression could be a surrogate marker for 1p19q codeleted tumors. This test can be done quickly from a simple biopsy, is reliable (with 97% consistency) and inexpensive, and does not need any special equipment. Our data indicate that the absence of INA expression in an oligodendroglial tumor makes the 1p19q deletion very unlikely (1/48 case ⫽ 2%). In contrast, when more than 10% of the cells express INA, there is a more than 80% chance to find a 1p19q codeletion in the tumor (table). It is unclear from our results whether INA is simply a surrogate marker of the 1p19q status or is a prognostic marker per se. Indeed, the expression of proneural genes characterizes a group of malignant gliomas with better prognosis compared to the “mesenchymal” subtype that also includes anaplastic astrocytomas and glioblastomas6; therefore, the prognostic value of INA expression could encompass the subset of 1p19q codeleted oligodendrogliomas. Although the rare tumors with strong INA expression and no 1p19q codeletion on CGH array were all alive at the time of analysis, the delay was too short to be relevant in three of them (low grade gliomas with 8, 9, and 23 months follow-up), but the only grade III tumor (a 1p19q intact anaplastic oligodendroglioma) was still alive at an 11.6-year follow-up. Our results are consistent with three published gene expression studies in which INA was significantly overexpressed in oligodendrogliomas compared to glioblastomas11 and in 1p19q codeleted oligodendrogliomas compared to non-codeleted gliomas.12,13 In a tissue microarray study, INA protein expression was associated with oligodendroglial phenotype, but the correlation with 1p19q codeletion was not studied.14 The gene encoding INA maps to 10q24.33: large deletions of 10q, including the INA locus, are frequent in glioblastomas, and are often associated with EGFR amplification. EGFR amplification and 10q loss are mutually exclusive with 1p19q codeletion 160
Neurology 72
January 13, 2009
(see figure 1A). One could therefore speculate that 10q deletion explains the downregulation of INA. However, INA is also absent in many gliomas without 10q deletion (figure 1A) and in normal glial cells, suggesting that its expression in glial tumor should rather be considered an abnormal feature, characteristic of 1p19q codeleted gliomas. INA is a class IV neuronal intermediate filament involved in the morphogenesis of neurons.15 It has been implicated in neurodegeneration,16 and its overexpression has been linked to neuronal cell death.17 Recently, INA has been identified as one of the many proteins phosphorylated by ATM and ATR in response to DNA damage,18 but it has no demonstrated role in oncogenesis, and, thus, its expression rather suggests that the cell of origin of oligodendrogliomas could be a progenitor cell giving rise to both neurons and oligodendrocytes.19,20 Indeed, INA expression has been demonstrated in tumors of neuronal origin such as neuroblastomas and medulloblastomas.21,22 Its expression in oligodendrogliomas is in agreement with a recent study showing that oligodendrogliomas display both ultrastructural and immunohistochemical neuronal features.23 Although preliminary, these data are strong enough to suggest that searching for 1p19q codeletion is unnecessary in tumors with negative INA expression, even with oligodendroglial phenotype. Further prospective studies are needed to determine whether INA expression is only a surrogate marker of 1p19q signature or an independent prognostic factor that encompasses this particular subset of gliomas. INA is, after YKL40, IQGAP1, and IGFBP2, a new immunohistochemical marker, identified through gene expression studies and shown to be of potential clinical value in gliomas.18,24 This demonstrates that microarray technology is a powerful technology useful for identifying immunohistochemical markers for everyday practice. Received June 19, 2008. Accepted in final form September 30, 2008.
REFERENCES 1. Reifenberger J, Reifenberger G, Liu L, et al. Molecular genetic analysis of oligodendroglial tumors shows preferential allelic deletions on 19q and 1p. Am J Pathol 1994;145: 1175–1190. 2. Idbaih A, Marie Y, Pierron G, et al. Two types of chromosome 1p losses with opposite significance in gliomas. Ann Neurol 2005;58:483–487. 3. Cairncross JG, Ueki K, Zlatescu MC, et al. Specific genetic predictors of chemotherapeutic response and survival in patients with anaplastic oligodendrogliomas. J Natl Cancer Inst 1998;90:1473–1479. 4. Jenkins RB, Blair H, Ballman KV, et al. A t(1;19)(q10; p10) mediates the combined deletions of 1p and 19q and
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
predicts a better prognosis of patients with oligodendroglioma. Cancer Res 2006;66:9852–9861. Mokhtari K, Paris S, Aguirre-Cruz L, et al. Olig2 expression, GFAP, p53 and 1p loss analysis contribute to glioma subclassification. Neuropathol Appl Neurobiol 2005;31:62–69. Phillips HS, Kharbanda S, Chen R, et al. Molecular subclasses of high-grade glioma predict prognosis, delineate a pattern of disease progression, and resemble stages in neurogenesis. Cancer Cell 2006;9:157–173. Ducray F, Idbaih A, de Reynies A, et al. Anaplastic oligodendrogliomas with 1p19q codeletion have a proneural gene expression profile. Mol Cancer 2008;7:41. Idbaih A, Marie Y, Lucchesi C, et al. BAC array CGH distinguishes mutually exclusive alterations that define clinicogenetic subtypes of gliomas. Int J Cancer 2008;122: 1778–1786. Cairncross JG, Macdonald DR. Successful chemotherapy for recurrent malignant oligodendroglioma. Ann Neurol 1988;23:360–364. Gilhuis HJ, Anderl KL, Boerman RH, et al. Comparative genomic hybridization of medulloblastomas and clinical relevance: eleven new cases and a review of the literature. Clin Neurol Neurosurg 2000;102:203–209. Shirahata M, Iwao-Koizumi K, Saito S, et al. Gene expression-based molecular diagnostic system for malignant gliomas is superior to histological diagnosis. Clin Cancer Res 2007;13:7341–7356. Mukasa A, Ueki K, Ge X, et al. Selective expression of a subset of neuronal genes in oligodendroglioma with chromosome 1p loss. Brain Pathol 2004;14:34–42. Mukasa A, Ueki K, Matsumoto S, et al. Distinction in gene expression profiles of oligodendrogliomas with and without allelic loss of 1p. Oncogene 2002;21:3961–3968. Ikota H, Kinjo S, Yokoo H, Nakazato Y. Systematic immunohistochemical profiling of 378 brain tumors with 37 antibodies using tissue microarray technology. Acta Neuropathol 2006;111:475–482.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
Chan SO, Chiu FC. Cloning and developmental expression of human 66 kd neurofilament protein. Brain Res Mol Brain Res 1995;29:177–184. Cairns NJ, Uryu K, Bigio EH, et al. alpha-Internexin aggregates are abundant in neuronal intermediate filament inclusion disease (NIFID) but rare in other neurodegenerative diseases. Acta Neuropathol 2004;108: 213–223. Chien CL, Liu TC, Ho CL, Lu KS. Overexpression of neuronal intermediate filament protein alpha-internexin in PC12 cells. J Neurosci Res 2005;80:693–706. Matsuoka S, Ballif BA, Smogorzewska A, et al. ATM and ATR substrate analysis reveals extensive protein networks responsive to DNA damage. Science 2007; 316:1160–1166. Menn B, Garcia-Verdugo JM, Yaschine C, et al. Origin of oligodendrocytes in the subventricular zone of the adult brain. J Neurosci 2006;26:7907–7918. Nunes MC, Roy NS, Keyoung HM, et al. Identification and isolation of multipotential neural progenitor cells from the subcortical white matter of the adult human brain. Nat Med 2003;9:439–447. Foley J, Witte D, Chiu FC, Parysek LM. Expression of the neural intermediate filament proteins peripherin and neurofilament-66/alpha-internexin in neuroblastoma. Lab Invest 1994;71:193–199. Kaya B, Mena H, Miettinen M, Rushing EJ. Alphainternexin expression in medulloblastomas and atypical teratoid-rhabdoid tumors. Clin Neuropathol 2003;22: 215–221. Vyberg M, Ulhøi BP, Teglbjaerg PS. Neuronal features of oligodendrogliomas-an ultrastructural and immunohistochemical study. Histopathology 2007;50:887–896. Nutt CL, Betensky RA, Brower MA, et al. YKL-40 is a differential diagnostic marker for histologic subtypes of high-grade gliomas. Clin Cancer Res 2005;11:2258–2264.
Neurology 72
January 13, 2009
161
Visual evoked potentials with CRT and LCD monitors When newer is not better
Aatif M. Husain, MD Susan Hayes, REEG/EP T Margaret Young, REEG T Dharmen Shah, MD
Address correspondence and reprint requests to Dr. Aatif M. Husain, 202 Bell Building, Box 3678, Duke University Medical Center, Durham, NC 27710
[email protected]
ABSTRACT
Background: The stimulus for pattern reversal visual evoked potentials (PRVEP) has traditionally been delivered by a cathode ray tube (CRT) monitor. Liquid crystal display (LCD) monitors have become more affordable and are being used instead of CRT monitors for many applications. We tested the hypothesis that LCD monitors were equivalent to CRT monitors when used for PRVEP.
Methods: Monocular, full field PRVEP with a 32= check size were obtained in six normal subjects with a CRT monitor and LCD monitors having 2 msec, 8 msec, and 30 msec response times. The average P100 latency with the CRT screen was compared to the latencies with the LCD screens. Results: The mean P100 latency of the CRT monitor was 107.7 (⫾6.6) ms, for the LCD 2 msec monitor was 115.7 (⫾6.9; p ⬍ 0.0001) ms, for the LCD 8 msec monitor was 118.5 (⫾6.5; p ⬍ 0.0001) ms, and the LCD 30 msec monitor was 156.8 (⫾6.8; p ⬍ 0.0001) ms.
Conclusions: Currently available liquid crystal display (LCD) monitors do not provide data comparable to cathode ray tube (CRT) monitors. LCD monitors cannot replace CRT monitors for pattern reversal visual evoked potentials unless new normative data are obtained. Neurology® 2009;72:162–164 GLOSSARY CRT ⫽ cathode ray tube; LCD ⫽ liquid crystal display; PRVEP ⫽ pattern reversal visual evoked potentials; VEP ⫽ visual evoked potentials.
Visual evoked potentials (VEP) are scalp-recorded responses to visual stimulation. Visual stimulation can be either patterned or unpatterned. VEP to patterned stimuli allow detection of minor abnormalities of the visual pathways as there is less interindividual variability. The commonest type of pattern used is a black and white checkerboard pattern. Alteration of the checkerboard color results in a pattern reversal VEP (PRVEP), which is used clinically.1 The checkerboard pattern used for PRVEP is generated by a pattern generator and displayed on a monitor. A cathode ray tube (CRT) monitor is used most often as it is readily available and cheap. With liquid crystal display (LCD) monitors becoming cheaper and replacing CRT monitors, we wanted to determine whether CRT monitors could be replaced by LCD monitors for PRVEP.2 METHODS Monocular, full field PRVEP were obtained in six subjects with a CRT monitor (Viewsonic E70F, 17 inch) and three LCD monitors with response times of 2 msec (KDS K-22mdwb2, 22 inch), 8 msec (Planar 2010m, 20 inch), and 30 msec (NEC Multisync 1860NX, 18 inch). A Nicolet Biomedical 2015 Visual Stimulator (Madison, WI) was used to generate the PRVEP stimulus. American Clinical Neurophysiology Society guidelines on obtaining PRVEP were followed.3 This study was approved by the local institutional review board. Visual acuity was determined for all subjects, and corrective lenses were used if they were worn habitually. A 32= check size was used with all monitors. To keep the visual angle constant, the distance between the subject and the monitor was increased when larger monitors were used.3 The contrast between the black and white checks was 98.4% or greater for all monitors, and the ambient light was kept constant. Each eye was stimulated independently at a rate of 4.1 per second. The monitors were presented in random order. Electrodes were applied to the scalp according to the Queen Square System.3 Positivity at MO was displayed downward. For each trial, 300 responses were averaged; each trial was repeated to confirm reproducibility. A 25 msec time base was used (250 msec full screen).
From the Department of Medicine (Neurology) (A.M.H., D.S.), Duke University Medical Center; and Neurodiagnostic Center (A.M.H., S.H., M.Y., D.S.), Veterans Affairs Medical Center, Durham, NC. Disclosure: The authors report no disclosures. 162
Copyright © 2009 by AAN Enterprises, Inc.
Table
Mean latencies of the P100 with different monitors
Monitor
P100 latency (SD) (ms)
CRT
110.3 (3.0)
p Value*
LCD 2 ms
118.1 (4.5)
⬍0.0001
LCD 8 ms
120.9 (3.7)
⬍0.0001
LCD 30 ms
159.4 (3.7)
⬍0.0001
*All p values are compared to CRT. CRT ⫽ cathode ray tube; LCD ⫽ liquid crystal display.
Figure
Examples of the P100 obtained with different monitors
The P100 latency of the left and right eyes of each subject with each monitor was noted. The mean P100 latencies with the LCD 2 msec, 8 msec, and 30 msec monitors were compared to the CRT monitor with Student t tests. A p value ⬍0.01 was considered significant to account for the multiple (three) comparisons.
The mean age of the six subjects was 43.0 (⫾10.6) years; five were women. The mean preauricular-to-preauricular distance was 35.5 (⫾2.7) cm. Corrected visual acuity was 20/20 for both eyes in all subjects. Reproducible PRVEP were obtained in all subjects with each monitor. The mean latencies of the P100 waveforms with the different monitors are presented in the table. The P100 latency was shortest for the CRT monitor, 107.7 (⫾6.6) ms. The LCD 30 msec monitor had the longest P100 latency at 156.8 (⫾6.8) ms (p ⬍ 0.0001). Even the fastest LCD monitor (2 msec) had a significantly longer P100 latency at 115.7 (⫾6.9) ms (p ⬍ 0.0001). An example of the PRVEP obtained with the different monitors is presented in the figure.
RESULTS
The trend of replacing CRT monitors with LCD monitors should be tempered in the case of monitors used to present the PRVEP stimulus. This study shows that with all other parameters kept constant, substituting a CRT monitor with a LCD monitor results in a significant prolongation of the P100 latency. Several features of LCD monitors may cause the delay seen in PRVEP latency. LCD monitors consist of two pieces of polarized glass with a liquid crystal between them.4 Light passes through the first polarized glass and liquid crystal to be displayed as an image on the second polarized glass. Depending on the electric current, liquid crystals either allow the light to pass or not (“on” or “off” state). The time it takes for the liquid crystal to go from an “off” state to an “on” state back to an “off” state is called the response time. Lower response times indicate better, faster monitors. Monitors with a higher response time can produce blurring of fast moving images.4 LCD monitors also have a short lag between when the image is presented and when it is displayed.4 This “input lag” occurs because the signal is processed before it is displayed. One common source of input lag is the scaling of an image to the native resolution of the monitor. The input lag may be as high as 68 msec. Whereas the response time is clearly stated by the manufacturer, the input lag time is not. CRT monitors use a different technology to display images on a screen.4,5 These monitors contain millions of blue, green, and red phosphor dots that glow when struck by an electron beam. The glowing DISCUSSION
(A) Cathode ray tube, (B) liquid crystal display (LCD) 2 msec, (C) LCD 8 msec, and (D) LCD 30 msec. Time scale: 25 msec/division. Amplitude scale 5 V/division.
Neurology 72
January 13, 2009
163
dots produce the image seen on the monitor. The electron beam sweeps from top to bottom of the screen in a non-interlaced (each line at a time) or interlaced (even or odd number lines) manner. Two complete sweeps are needed when the interlaced method is used. Though each phosphor dot is illuminated very briefly and consecutively, persistence of vision makes it seem as if the entire screen changes instantaneously. Persistence of vision is a property of the eye which allows an image presented for a nanosecond to be perceived for milliseconds.6 CRT monitors do not have a “response time”; rather, they have a refresh rate which measures how quickly the electron beam passes over the screen. Most current monitors have a refresh rate of at least 60 – 80 per second, which is equivalent to a response time of about 12.5 msec. In this study, LCD monitors with a variety of response times were used. However, even the one with the lowest response time (2 msec) produced a P100 latency longer than the CRT monitor. Slower monitors produced P100 latencies that were much greater, clearly in the abnormal range in most laboratories. Additionally, the input lag also contributed to the longer P100 latency in LCD monitors; however, the degree of this contribution cannot be determined as the exact amount of input lag is not available from the monitor manufacturers. The P100 latency change could not be attributed to other technical issues. The luminance of the pattern was checked by a photometer, and the contrast between the black and white checks was maintained greater than 98%. The alertness of the subjects was also monitored by the technologist, and the normal morphology of the waveforms documented subject
164
Neurology 72
January 13, 2009
attention to the stimulus. The ambient room light was also kept constant between trials. A shortcoming of this study is the number of subjects enrolled. However, despite using only six subjects and correcting the significance level for multiple comparisons, the differences in the P100 latencies were significant. All LCD monitors produced longer P100 latencies than the CRT monitor. Newer is not always better. This study demonstrates that even the best LCD monitors currently available cannot produce PRVEP similar to those seen with CRT monitors. Laboratories should be very cautious about “upgrading” to LCD monitors for VEP; if they do, normative data must be obtained again. Whenever LCD monitors are changed, the normative data will need to be updated as well. Received May 22, 2008. Accepted in final form October 1, 2008. REFERENCES 1. Chiappa KH, Hill RA. Pattern-shift visual evoked potentials: interpretation. In: Chiappa KH, ed. Evoked Potentials in Clinical Medicine, 3rd ed. Philadelphia: Lippincott-Raven; 1997:95–130. 2. Saunders RS, Samei E, Baker J, et al. Comparison of LCD and CRT displays based on efficacy for digital mammography. Acad Radiol 2006;13:1317–1326. 3. ACNS. Guideline 9B: Guidelines on visual evoked potentials. J Clin Neurophysiol 2006;23:138–156. 4. Badano A. Principles of cathode-ray tube and liquid crystal display devices. In: Samei E, Flynn MJ, eds. Advances in Digital Radiography: Categorical Course in Diagnostic Radiology Physics. 2003 Syllabus. Oak Brook, IL: RSNA; 2003:91–102. 5. Cook LT, Cox GG, Insana MF, et al. Comparison of a cathode-ray-tube and film for display of computed radiographic images. Med Phys 1998;25:1132–1138. 6. Coltheart M. The persistences of vision. Philos Trans R Soc Lond B Biol Sci 1980;290:57–69.
Benefits and risks of stavudine therapy for HIV-associated neurologic complications in Uganda N. Sacktor, MD N. Nakasujja, MB, ChB R.L. Skolasky, ScD K. Robertson, PhD S. Musisi, FRCP A. Ronald, MD E. Katabira, FRCP D.B. Clifford, MD, FAAN
Address correspondence and reprint requests to Dr. Ned Sacktor, Department of Neurology, Johns Hopkins Bayview Medical Center, 4940 Eastern Avenue, B Bldg. Rm. 123, Baltimore, MD 21224
[email protected]
ABSTRACT
Background: The frequency of HIV dementia in a recent study of HIV⫹ individuals at the Infectious Disease Institute in Kampala, Uganda, was 31%. Coformulated generic drugs, which include stavudine, are the most common regimens to treat HIV infection in Uganda and many other parts of Africa.
Objective: To evaluate the benefits and risks of stavudine-based highly active antiretroviral therapy (HAART) for HIV-associated cognitive impairment and distal sensory neuropathy. The study compared neuropsychological performance changes in HIV⫹ individuals initiating HAART for 6 months and HIV⫺ individuals receiving no treatment for 6 months. The risk of antiretroviral toxic neuropathy as a result of the initiation of stavudine-based HAART was also examined. Methods: At baseline, 102 HIV⫹ individuals in Uganda received neurologic, neuropsychological, and functional assessments; began HAART; and were followed up for 6 months. Twenty-five HIV⫺ individuals received identical clinical assessments and were followed up for 6 months. Results: In HIV⫹ individuals, there was improvement in verbal memory, motor and psychomotor speed, executive thinking, and verbal fluency. After adjusting for differences in sex, HIV⫹ individuals demonstrated significant improvement in the Color Trails 2 test (p ⫽ 0.025) compared with HIV⫺ individuals. Symptoms of neuropathy developed in 38% of previously asymptomatic HIV⫹ patients after initiation of the stavudine-based HAART.
Conclusions: After the initiation of highly active antiretroviral therapy (HAART) including stavudine, HIV⫹ individuals with cognitive impairment improve significantly as demonstrated by improved performance on a test of executive function. However, peripheral neurotoxicity occurred in 30 patients, presumably because of stavudine-based HAART, suggesting the need for less toxic therapy. Neurology® 2009;72:165–170 GLOSSARY AVLT ⫽ Auditory Verbal Learning Test; CES-D ⫽ Center for Epidemiologic Studies–Depression Scale; D-drug ⫽ dideoxynucleoside antiretroviral drug; GEE ⫽ generalized estimating equation; GP ⫽ Grooved Pegboard test; HAART ⫽ highly active antiretroviral therapy; HIV-SN ⫽ HIV-associated sensory neuropathy; IHDS ⫽ International HIV Dementia Scale; MSK ⫽ Memorial Sloan-Kettering; NA ⫽ not applicable; NS ⫽ not significant; SDMT ⫽ Symbol Digit Modalities Test; UCLA ⫽ University of California-Los Angeles; WHO ⴝ World Health Organization.
The majority of HIV cases globally, an estimated 25 million people, are in sub-Saharan Africa.1 HIV-1–associated dementia complex (HIV dementia) is characterized by disabling cognitive, behavioral, and motor dysfunction.2 HIV-associated sensory neuropathies (HIV-SNs) are another common neurologic manifestation of advanced HIV infection seen in approximately 35% of HIV-infected patients.3-5 HIV-SNs can be caused by HIV infection itself as a distal symmetric polyneuropathy or by exposure to dideoxynucleoside antiretroviral drugs (D-drugs), which include d4T (stavudine) as an antiretroviral toxic neuropathy. The frequency of HIV
From the Departments of Neurology (N.S.) and Orthopedic Surgery (R.L.S.), Johns Hopkins University School of Medicine, Baltimore, MD; Departments of Psychiatry (N.N., S.M.), and Medicine (E.K.), Makerere University, Kampala, Uganda; Department of Neurology (K.R.), University of North Carolina, Chapel Hill, NC; Department of Internal Medicine (A.R.), University of Manitoba, Winnipeg, Canada; and Department of Neurology (D.B.C.), Washington University, St. Louis, MO. Supported by the Neurologic AIDS Research Consortium, which receives support from the National Institute of Neurological Diseases and Stroke (NS 32228), and MH71150. Disclosure: The authors report no disclosures. Copyright © 2009 by AAN Enterprises, Inc.
165
dementia, the most severe form of HIVassociated neurocognitive disorders, in a recent study of ambulatory HIV⫹ patients in an Infectious Disease Clinic in Kampala, Uganda, was 31%.6 The frequency of HIV-SN in this same study was 32%. Highly active antiretroviral therapy (HAART) can improve cognitive performance in some patients with HIV dementia in the United States.7-9 In a pilot study of 23 HIV⫹ individuals in Uganda, HAART also was associated with improvement in neuropsychological test performance.10 This pilot study was without a control group so that the study was limited by the possibility of “practice effects.” Furthermore, it is possible that HAART may improve peripheral neuropathy in some individuals, whereas it is a recognized toxicity in others.11 Generic drugs are the most commonly used antiretroviral drugs in sub-Saharan Africa. Triommune is a coformulation of stavudine, lamivudine, and nevirapine. Because of cost and availability, it is the most frequent generic regimen prescribed. The objective of this study was to evaluate the benefits and risks of stavudine-based HAART for HIV-associated neurologic complications, namely HIV-associated cognitive impairment and HIV-SN in Uganda. Specifically, the study compared neuropsychological test performance changes in HIV⫹ individuals initiating Triommune for 6 months and HIV⫺ individuals receiving no treatment for 6 months. The risk of antiretroviral toxic neuropathy as a result of the initiation of Triommune was also examined. METHODS Participants. The study was conducted, from September 2005 to January 2007, with 102 HIV⫹ individuals at the Infectious Disease clinic in Mulago Hospital and with 25 HIV⫺ individuals from the AIDS Information Center in Kampala, Uganda.12 The study was approved by the institutional review boards and ethical standards committees at Johns Hopkins University in Baltimore, Maryland, and Makerere University in Kampala, Uganda. Written patient consent was obtained from all individuals. All HIV⫺ subjects had a documented negative ELISA HIV-1 test result. HIV⫹ individuals in the clinic were chosen to receive HAART using the following inclusion criteria: advanced HIV infection with a CD4 lymphocyte count ⬍200 cells/L, attendance of at least two clinic visits in the past 6 months, residence within a 20-km radius of Kampala, antiretroviral drug naive, performance on a screening test for HIV dementia (International HIV Dementia Scale [IHDS] score ⱕ10) 166
Neurology 72
January 13, 2009
suggestive of HIV-associated cognitive impairment,13 and ability to provide written informed consent. Exclusion criteria included age less than 18 years, an active or known past CNS opportunistic infection, fever ⬎37.5°C, a history of a chronic neurologic disorder, active psychiatric disorder, alcoholism, physical deficit (e.g., amputation), severe functional impairment (Karnofsky Performance Scale score ⬍50),14 or severe medical illness that would interfere with the ability to perform the study evaluations. The evaluations were translated into the local language, Luganda.
Clinical assessments. HIV⫹ individuals received clinical assessments using standardized questionnaires assessing demographic information, including primary language and reading abilities; medical, psychiatric, and neurologic history, including an assessment of peripheral neuropathy symptoms; and a neurologic examination, including assessments of vibration and pin sensation, limb strength and coordination, and deep tendon reflexes. The neurocognitive assessment included the IHDS13 and a full neuropsychological test battery. The neuropsychological testing battery included the World Health Organization (WHO)–University of California-Los Angeles (UCLA) Auditory Verbal Learning Test (AVLT) for verbal memory15; the Timed Gait, Finger Tapping, and Grooved Pegboard tests to assess motor performance; the Symbol Digit Modalities test16 and the Color Trails test15 to assess psychomotor speed and executive functioning performance; Digit Span Forward and Backward to assess attention; and the category naming test to assess verbal fluency. The functional assessment included the Karnofsky Performance Scale.14 These assessments were used to assign a Memorial Sloan-Kettering (MSK) dementia stage of 0, 0.5, or ⱖ1 by a consensus conference including a neurologist, a psychiatrist, and a neuropsychologist.17 All HIV⫹ subjects received a baseline CD4 lymphocyte count. Follow-up CD4 counts were obtained at 3 and 6 months after baseline. Plasma and CSF HIV RNA as well as neuroimaging were not available in this study. HIV⫺ individuals received identical clinical assessments except for the absence of CD4 lymphocyte counts. HIV⫺ individuals were also followed up for 6 months. Antiretroviral treatment. After the baseline visit, all of the HIV⫹ subjects began Triommune (stavudine, lamivudine, and nevirapine). HIV⫺ individuals received no treatment.
Data analysis. For each neuropsychological test, a Z score was calculated using age- and education-adjusted normative data obtained from 100 HIV⫺ individuals in Uganda.6 Distributional tests have confirmed that the resultant Z scores follow a normal distribution, and scores are summarized as mean (SE). To account for loss-to-follow-up, longitudinal changes in the mean Z score for each neuropsychological test were evaluated at 3 and 6 months using generalized linear models for handling nonignorable dropouts for continuous outcomes.18 Statistical inference for the model parameters was based on generalized estimating equations (GEEs).19 The GEE model included HIV serostatus, study visit, and the interaction between HIV serostatus and study visit as important parameters. Significance of the interaction term was taken as inferential evidence of a difference in longitudinal performance between the HIV⫺ and HIV⫹ groups. Type I error rates were adjusted for multiple comparisons using a Bonferroni correction. Distributional tests have confirmed that CD4 count, IHDS screening test, and Karnofsky Performance Scale scores did not follow a normal distribution, and scores were summarized as median and interquartile range (25th and 75th per-
Table 1
Demographics of HIVⴙ and HIVⴚ individuals HIVⴙ (n ⴝ 102)
Age, mean (SD), y Education, mean (SD), y Male, n (%)
34.2 (6.4)
30.3 (4.0)
9.1 (4.3)
10.3 (4.2)
29 (28)
CD4 count, mean (SD)
129 (79.0)
Karnofsky score, mean (SD)
HIVⴚ (n ⴝ 25)
84 (8.5)
15 (60)
p Value 0.004 NS 0.003
NA 98 (4.1)
⬍0.001
MSK HIV dementia stage, baseline, n (%) 0 ⴝ no impairment
12 (12)
NA
0.5 ⴝ equivocal/subclinical
48 (48)
NA
1 ⴝ mild dementia
33 (33)
NA
7 (7)
NA
2 ⴝ moderate dementia
NS ⫽ not significant; NA ⫽ not applicable; MSK ⫽ Memorial Sloan-Kettering.
centiles). Longitudinal changes in these measures were evaluated using a nonparametric one-way analysis of variance.
The demographics for the 102 HIV⫹ subjects and 25 HIV⫺ subjects are summarized in table 1. The HIV⫹ subjects [mean (SD) age ⫽ 34.2 (6.4) years] were slightly older than the HIV⫺ subjects [mean (SD) age ⫽ 30.3 (4.0) years] (p ⫽ 0.004). There were no differences in level of education between the two groups. The HIV⫹ group (28% male) had a greater proportion of women than the HIV⫺ group (60% male) (p ⫽ 0.003). The HIV⫹ subjects in our study are representative of the demographics of HIV infection in Uganda. Using the inclusion criteria of an IHDS score ⱕ10 suggesting HIV-associated neurocognitive disorders, the frequency of HIV dementia at baseline was high, with 33% presenting to the ambulatory clinic with mild HIV dementia (MSK dementia stage 1) and 7% presenting with moderate HIV dementia (MSK dementia stage 2). None of the HIV⫺ individuals had impairment on their neuropsychological testing to consider a diagnosis of dementia. The HIV⫹ individuals also had
RESULTS
Figure 1
Memorial Sloan-Kettering HIV dementia stage changes after 3 and 6 months of highly active antiretroviral therapy
0 ⫽ Memorial Sloan-Kettering (MSK) 0 (normal cognitive function); 0.5 ⫽ MSK 0.5 (equivocal symptoms or signs without impairment in capacity to perform activities of daily living); 1 ⫽ MSK 1 (mild HIV dementia); 2 ⫽ MSK 2 (moderate HIV dementia).
more functional impairment at baseline as measured by the Karnofsky score [HIV⫹ group Karnofsky mean (SD) ⫽ 84 (8.5) vs HIV⫺ group Karnofsky mean (SD) ⫽ 98 (4.1); p ⬍ 0.001]. The follow-up rates for the HIV⫹ and HIV⫺ subjects were excellent. Among the HIV⫹ patients, 92% returned for their 6-month follow-up visit. Among the HIV⫺ subjects, 84% returned for their 6-month follow-up visit. There was improvement in the mean CD4 count among the HIV⫹ subjects at both 3 and 6 months after the initiation of HAART. The mean CD4 count improved from 129 at baseline to 268 at 3 months (p ⬍ 0.001) and to 272 at 6 months (p ⬍ 0.01). Neurocognitive improvement with HAART. As shown in the figure 1, improvements in the MSK HIV dementia stage were noted among the HIV⫹ subjects after 6 months of HAART. After 3 months, 23% of the HIV individuals had mild HIV dementia (MSK dementia stage 1) and 3% had moderate HIV dementia (MSK dementia stage 2). After 6 months, 13% of the HIV⫹ individuals had mild HIV dementia (MSK dementia stage 1) and 3% had moderate HIV dementia (MSK dementia stage 2) (p ⬍ 0.001). Table 2 shows the changes in the mean neuropsychological test performance for each of the neuropsychological tests after 3 and 6 months of HAART among the HIV⫹ subjects. Compared with baseline, improvement was seen in tests of verbal memory (WHO–UCLA AVLT total score and delayed recall at 3 and 6 months), psychomotor speed and executive functioning (Color Trails 1 and 2 at 3 and 6 months, Symbol Digit Modalities test at 6 months only), motor performance (Timed Gait and Grooved Pegboard dominant hand at 6 months only, Grooved Pegboard nondominant hand and Finger Tapping at 3 and 6 months), and verbal fluency (category naming at 3 and 6 months). Table 3 shows the changes in the mean neuropsychological test performance for each of the neuropsychological tests after 6 months among the untreated HIV⫺ subjects. Compared with baseline, among the HIV⫺ subjects, improvement was seen in a test of motor speed (Grooved Pegboard nondominant hand at 3 months, Grooved Pegboard dominant hand at 6 months) and psychomotor speed (Color Trails 1 at 6 months only). Also, there was borderline improvement in tests of verbal memory [AVLT total score (p ⫽ 0.09) and delayed recall (p ⫽ 0.08) at 6 months]. No other neuropsychological test showed improvement compared with baseline among the HIV⫺ subjects. Compared with the HIV⫺ subjects, HIV⫹ subjects showed improvement on the Color Trails 2 test (p ⫽ 0.02) after adjusting for differences in sex (figNeurology 72
January 13, 2009
167
Table 2
score of 90 indicates an individual who is able to do normal activity without effort but shows minor disease symptoms or signs.
Neuropsychological test performance at baseline and 3 and 6 months after HAART among HIVⴙ patients
Neuropsychological test
Baseline Z score* (n ⴝ 102)
3-Month Z score* (n ⴝ 94)
p Value†
6-Month Z score* (n ⴝ 95)
p Value†
AVLT total score
⫺1.2 (0.1)
⫺0.4 (0.1)
⬍0.001
⫺0.1 (0.2)
⬍0.001
AVLT delayed recall
⫺1.2 (0.1)
⫺0.4 (0.2)
0.003
⫺0.1 (0.1)
⬍0.001
Color Trails 1
⫺1.7 (0.3)
⫺0.7 (0.2)
⬍0.001
⫺0.4 (0.3)
⬍0.001
⫺1.3 (0.2)
⬍0.001
⫺0.3 (0.1)
0.002
Color Trails 2
⫺2.8 (0.3)
⫺1.9 (0.4)
SDMT
⫺0.8 (0.1)
⫺0.5 (0.1)
0.001 NS
0 (0.2)
⫺0.2 (0.1)
NS
0.4 (0.1)
0.014
GP, nondominant hand
⫺0.7 (0.2)
0.1 (0.1)
⬍0.001
0.3 (0.1)
⬍0.001
Finger tapping
⫺1.0 (0.1)
⫺0.5 (0.1)
0.001
⫺0.5 (0.1)
0.002
GP, dominant hand
Timed gait
⫺2.7 (0.3)
⫺2.3 (0.3)
Verbal fluency
⫺0.4 (0.4)
⫺0.1 (0.1)
NS 0.007
⫺1.8 (0.3)
0.028
0 (0.1)
⬍0.001
*Mean (SE). †Compared with performance at baseline. HAART ⫽ highly active antiretroviral therapy; AVLT ⫽ Auditory Verbal Learning Test; SDMT ⫽ Symbol Digit Modalities Test; NS ⫽ not significant; GP ⫽ Grooved Pegboard test.
ure 2). Other tests, e.g., WHO–UCLA AVLT, showed trends for greater improvement among the HIV⫹ subjects compared with the HIV⫺ subjects, but this difference was not significant. Functional improvement with HAART. There was im-
provement in the functional performance (Karnofsky scale) at both 3 and 6 months after the initiation of HAART among HIV⫹ subjects. The mean Karnofsky score improved from 84 at baseline to 87 at 3 months (p ⫽ 0.002) and to 89 at 6 months (p ⬍ 0.001). A Karnofsky score of 80 indicates an individual who is able to do normal activity with effort but shows some disease symptoms or signs. A Karnofsky
Table 3
Neuropsychological test performance at baseline and 3 and 6 months after HAART among HIVⴚ patients
Neuropsychological test
Baseline Z score* (n ⴝ 25)
3-Month Z score* (n ⴝ 25)
p Value†
AVLT total score
⫺0.4 (0.3)
⫺0.4 (0.4)
NS
0.2 (0.2)
NS
AVLT delayed recall
⫺0.6 (0.2)
⫺1.1 (0.6)
NS
0.1 (0.2)
NS
Color Trails 1
⫺0.3 (0.3)
0.2 (0.3)
NS
0.8 (0.2)
Color Trails 2
⫺0.8 (0.3)
⫺0.4 (0.4)
NS
⫺0.4 (0.3)
NS
SDMT
⫺0.4 (0.3)
⫺0.2 (0.3)
NS
0.4 (0.3)
NS
0.3 (0.3)
0.9 (0.2)
NS
0.8 (0.2)
0.032
GP, nondominant hand
⫺0.1 (0.2)
0.7 (0.2)
0.011
0.6 (0.2)
NS
Finger tapping
⫺0.0 (0.2)
⫺0.1 (0.3)
NS
0.1 (0.2)
NS
Timed gait
⫺1.5 (0.5)
⫺0.4 (0.5)
NS
⫺1.7 (0.4)
NS
Verbal fluency
⫺0.0 (0.2)
⫺0.2 (0.3)
NS
0.4 (0.3)
NS
GP, dominant hand
6-Month Z score* (n ⴝ 25)
p Value†
Neurology 72
January 13, 2009
subjects, 37% had foot numbness at baseline.20 Signs of neuropathy (loss of vibration or pinprick sensation, decreased or absent ankle reflexes) were present in 43% of the HIV⫹ subjects at baseline. In contrast, only 1 (4%) of the HIV⫺ controls had foot numbness at baseline, and 1 (4%) of the HIV⫺ controls had signs of neuropathy at baseline. After 6 months, 54% of the HIV⫹ subjects had numbness in the feet. Symptoms of neuropathy were described in 38% (n ⫽ 30) of previously asymptomatic HIV⫹ patients after initiation of the stavudine-based HAART. Signs of neuropathy were detected in 31% (p ⫽ 0.03) of the HIV⫹ individuals who did not have signs at baseline after initiation of HAART. There were no differences in the age, education, sex, or CD4 count among asymptomatic HIV⫹ patients who developed symptoms of neuropathy and asymptomatic HIV⫹ patients who did not develop symptoms of neuropathy after initiation of the stavudine-based HAART. Similarly, there were no differences in the age, education, sex, or CD4 count among HIV⫹ patients without signs of neuropathy at baseline who developed signs of neuropathy and HIV⫹ patients who never developed signs of neuropathy after initiation of the stavudine-based therapy. In contrast, only 1 (4%) of the HIV⫺ controls had an increase in neuropathy symptoms from baseline, and only 1 (4%) had an increase in neuropathy signs from baseline. After the initiation of the stavudine-based HAART, some HIV⫹ subjects had improvement in neuropathy symptoms and signs present at baseline. Among HIV⫹ subjects with foot numbness at baseline, 22% no longer had numbness after 6 months of the stavudine-based HAART. Among HIV⫹ subjects with signs of neuropathy at baseline, 23% no longer had neuropathy signs after 6 months of the stavudine-based HAART.
0.006
*Mean (SE). †Compared with performance at baseline. HAART ⫽ highly active antiretroviral therapy; NS ⫽ not significant; AVLT ⫽ Auditory Verbal Learning Test; SDMT ⫽ Symbol Digit Modalities Test; GP ⫽ Grooved Pegboard test. 168
Neuropathy changes over time. Among the HIV⫹
The results from this study support the clinical benefit of HAART for stavudine-based therapy in HIV⫹ individuals in Uganda. After the initiation of HAART, HIV⫹ individuals had improvement in their level of immune suppression with increased CD4 counts. HIV⫹ individuals had improvement in their overall functional performance as measured by the Karnofsky scale. HIV⫹ individuals also had improvements in their neurocognitive performance in tests of verbal memory, psychomotor and motor speed performance, executive functioning, and verbal fluency similar to the results of a prior
DISCUSSION
Figure 2
Color Trails 2 changes over 6 months
Case ⫽ HIV⫹ individual; control ⫽ HIV⫺ individual. Y-axis corresponds to the Z score on the Color Trails 2 test.
pilot study in Uganda evaluating the effect of HAART on neurocognitive performance.10 The neuropsychological test improvement among HIV⫹ individuals is likely due to a combination of both HAART-associated improvements in immune suppression and neurocognitive function as well as practice effects. Compared with HIV⫺ individuals, HIV⫹ individuals showed greater improvement in Color Trails 2, a test of executive functioning, suggesting that practice effects alone cannot account for the improvements seen in executive functioning. However, a ceiling effect for performance on this test, where the HIV⫺ subjects could not improve significantly from baseline normal function, could have limited the learning effects relevant to the control group. Other cognitive domains, e.g., verbal memory, showed trends for greater improvement among HIV⫹ individuals compared with HIV⫺ individuals. However, the sample size of the study may not have been large enough to detect a difference between the two groups. In addition, it is possible that some of the neurocognitive impairment among the HIV⫹ individuals at baseline may have been due to confounding factors other than HIV infection itself, and thus unlikely to improve with HAART. The prevalence of neuropathy symptoms and signs at baseline in our study is similar to the prevalence of neuropathy symptoms and signs in Western countries among HIV⫹ patients with advanced immunosuppression,3-5 because the mean CD4 count of our HIV⫹ cohort was 129 cells. Stavudinebased HAART is also associated with some measurable neurologic morbidity. In Western countries, D-drug based HAART is rarely used now because of the risk of peripheral neuropathy. In our study in Uganda among HIV⫹ patients without peripheral neuropathy at baseline, 31% to 38% of these patients developed either symptoms or signs of neuropathy during the study. The onset of the symptomatic neuropathy experienced after starting therapy was prompt within the first 3 months of the trial, sug-
gesting that stavudine may have been the cause of the neuropathy or stavudine decreased the threshold for developing an HIV-related neuropathy. The severity of these symptoms did not lead to HAART discontinuation in any of these patients. However, the study duration was only 6 months. At present, it is unclear whether a cumulative toxicity of stavudine over longer periods of treatment might result in a greater proportion affected or more severe neuropathy in those taking this therapy. Future studies of neuropathy should include a more extensive period of follow-up. There are limitations to the study. The study was conducted among HIV⫹ patients from an urban setting in Kampala, Uganda, which is likely to be representative of the urban HIV⫹ community in Uganda. However, it may not necessarily reflect the HIV⫹ population in rural Uganda. All study participants received a screening test for depression symptomatology, the Center for Epidemiologic Studies–Depression Scale (CES-D).21 At baseline, the mean CES-D score for the HIV⫹ group was 18.1 (SD 11.4). Thus, some patients may have had major depression. However, a psychiatrist on site evaluated any patient in which an active psychiatric condition was a concern. HIV subtypes D and A are the predominant subtypes in Uganda, whereas subtype C is the predominant subtype in southern Africa. If HIV subtype has an effect on HIV neuropathogenesis, our results in Uganda may not necessarily reflect the HIV⫹ population throughout Africa. HIV⫹ patients did not receive either neuroimaging or CSF examinations to rule out another CNS disease as the cause for an individual’s cognitive impairment. Other infectious diseases, such as tuberculosis, syphilis, or malaria, or malnutrition could have contributed to the cognitive impairment in our study participants. However, all HIV⫹ patients did receive a detailed neurologic history and examination (including an evaluation of fever, headache, neck stiffness, and focal abnormalities). Any HIV⫹ patient with a suspected CNS opportunistic infection or neoplasm was excluded from the study. The neuropsychological tests that showed the largest improvement among the HIV⫹ patients were the tests that had the largest baseline Z score deficits. Thus, regression to the mean may have been a contributing factor for some of the neurocognitive improvement seen among the HIV⫹ patients. HIV⫹ individuals demonstrated greater improvement in functional performance compared with HIV⫺ individuals. However, the HIV⫺ individuals had normal functional performance in genNeurology 72
January 13, 2009
169
eral, so this group had little capacity for further improvement. The cause of the neuropathy could not be fully determined due to the design of the study. A screening question for diabetes was included, and two HIV⫹ individuals reported a history of diabetes. Screening tests for other causes of neuropathy and electrophysiological studies were not available in Uganda. Ideally, a group of HIV⫹ individuals with the same inclusion criteria at study entry who were not on a stavudine-based regimen could have been evaluated, but this group of patients did not exist at the time of the study. HAART is a lifesaving therapy for HIV⫹ patients with advanced immunosuppression in sub-Saharan Africa, and if a patient meets clinical criteria, HAART should be initiated without delay. If no other antiretroviral drug options are available, stavudine-based HAART can provide benefit with respect to immune system restoration and improvement in executive functioning in HIV⫹ individuals with neurocognitive impairment. However, this study provides data demonstrating that peripheral neurotoxicity is a significant problem from stavudinebased HAART, and if possible, alternative antiretroviral drug combinations should be provided. It is anticipated that future trends will have decreased use of stavudine-based therapies in resource limited countries. Additional resources to provide HAART regimens without stavudine would provide a great benefit by preventing a potential increase in the prevalence of painful peripheral neuropathy among millions of HIV⫹ individuals who will need HAART within the next several years in sub-Saharan Africa.
4.
5.
6.
7.
8.
9.
10.
11.
12. 13.
14.
15.
16. AUTHOR CONTRIBUTIONS
17.
Statistical analysis was conducted by R.L.S.
18. Received June 5, 2008. Accepted in final form October 1, 2008. 19. REFERENCES 1. Joint United Nations Program on HIV/AIDS. UNAIDS and the World Health Organization. Report on AIDS Epidemic Update, 2006. 10th Anniversary ed. 2006. 2. Sacktor N. The epidemiology of human immunodeficiency virus-associated neurological disease in the era of highly active antiretroviral therapy. J Neurovirol 2002; 8 (suppl 2):115–121. 3. Schifitto G, McDermott MP, McArthur JC, et al. Incidence of and risk factors for HIV-associated distal sensory polyneuropathy. Neurology 2002;58:1764–1768.
170
Neurology 72
January 13, 2009
20.
21.
de la Monte SM, Gabuzda DH, Ho DD, et al. Peripheral neuropathy in the acquired immunodeficiency syndrome. Ann Neurol 1988;23:485–492. So YT, Holtzman DM, Abrams DI, et al. Peripheral neuropathy associated with acquired immunodeficiency syndrome: prevalence and clinical features from a population-based survey. Arch Neurol 1988;45:945– 948. Wong M, Robertson K, Nakasujja N, et al. Frequency of and risk factors for HIV dementia in an HIV clinic in sub-Saharan Africa. Neurology 2007;68:350–355. Sacktor NC, Skolasky RL, Lyles RH, et al. Improvement in HIV-associated motor slowing after antiretroviral therapy including protease inhibitors. J Neurovirol 2000;6: 84–88. Sacktor NC, Lyles RH, Skolasky RL, et al. Combination antiretroviral therapy including protease inhibitors improves HIV-associated cognitive impairment. Neurology 1999;52:1640–1647. Robertson KR, Robertson WT, Ford S, et al. Highly active antiretroviral therapy improves neurocognitive functioning. J Acquir Immune Defic Syndr 2004;36:562–566. Sacktor N, Nakasujja N, Skolasky R, et al. Antiretroviral therapy improves cognitive impairment in HIV⫹ individuals in sub-Saharan Africa. Neurology 2006;67:311–314. Martin C, Solders G, Sonnerborg A, et al. Antiretroviral therapy may improve sensory function in HIV-infected patients: a pilot study. Neurology 2000;54:2120–2127. Quinn TC. The Academic Alliance for AIDS Care and Prevention in Africa. Hopkins HIV Rep 2001;13:1–4. Sacktor NC, Wong M, Nakasujja N, et al. The International HIV Dementia Scale: a new rapid screening test for HIV dementia. AIDS 2005;19:1367–1374. Karnofsky DA, Abelman WH, Craver LF, et al. The use of nitrogen mustards in the palliative treatment of carcinoma. Cancer 1948;1:634–656. Maj M, Satz P, Janssen R, et al. WHO Neuropsychiatric AIDS Study, cross-sectional phase II: neuropsychological and neurological findings. Arch Gen Psychiatry 1994;51: 51–61. Smith A. Symbol Digit Modalities Test. Los Angeles, CA: Western Psychological Service; 1982. Price RW, Brew BJ. The AIDS dementia complex. J Infect Dis 1988;158:1079–1083. Fitzmaurice GM, Laird NM. Generalized linear mixture models for handling nonignorable dropouts in longitudinal studies. Biostatistics 2000;1:141–156. Zeger SL, Liang KY, Albert PS. Models for longitudinal data: a generalized estimating equation approach. Biometrics 1988;44:1049–1060. Peters K, Nakasujja N, Robertson KR, et al. Development of peripheral neuropathy after stavudine-based antiretroviral therapy in HIV positive (HIV⫹) patients in subSaharan Africa. Presented at: 132nd Annual Meeting of the American Neurological Association; October 7, 2007; Washington, DC; 2007. Radloff LS. The CES-D scale: a self-report depression scale for research in the general population. Appl Psychol Meas 1977;1:385–401.
Cerebral microbleeds are a risk factor for warfarin-related intracerebral hemorrhage
Seung-Hoon Lee, MD, PhD* Wi-Sun Ryu, MD* Jae-Kyu Roh, MD, PhD
Address correspondence and reprint requests to Dr. Jae-Kyu Roh, Department of Neurology, Seoul National University Hospital, 28 Yongon-dong, Jongno-gu, Seoul, 110-744, Republic of Korea
[email protected]
ABSTRACT
Background: Cerebral microbleeds are known to be indicative of bleeding-prone microangiopathy and may predict incident intracerebral hemorrhage (ICH). In this study, we investigated whether microbleeds are associated with the incidence of warfarin-related ICH.
Methods: Twenty-four patients with ICH while on outpatient treatment with warfarin were selected from a consecutive cohort. Control, warfarin-using subjects with no history of ICH were randomly selected during the same time period (n ⫽ 48). We compared demographic factors, vascular risk factors, laboratory findings, and radiologic findings including microbleeds between the groups. Result: There were more cases of patients with microbleeds in the ICH than control group (79.2% vs 22.9%: p ⬍ 0.001), and the number of microbleeds was much higher for the ICH group (9.0 ⫾ 26.8 vs 0.5 ⫾ 1.03: p ⬍ 0.001). Moreover, the number of microbleeds was significantly correlated with the presence of warfarin-related ICH (r ⫽ 0.299; p ⬍ 0.001). Conditional logistic regression analysis showed that increased prothrombin time and the presence of microbleeds were independently related to the incidence of warfarin-related ICH (microbleeds: adjusted OR, 83.12).
Conclusion: This study suggests that underlying microbleeds are independently associated with an incidence of warfarin-related intracerebral hemorrhage. Future research should focus on elucidating the risks and benefits of warfarin medication in patients with microbleeds. Neurology® 2009;72:171–176 GLOSSARY CI ⫽ confidence interval; GRE ⫽ gradient-echo; ICH ⫽ intracerebral hemorrhage; INR ⫽ international normalized ratio; NS ⫽ not significant; OR ⫽ odds ratio; PT ⫽ prothrombin time; WMH ⫽ white matter hyperintensity.
Spontaneous intracerebral hemorrhage (ICH) results from a rupture of the small penetrating cerebral arteriole. ICH accounts for 10 to 15% of acute first-ever strokes, and it reaches an incidence rate of 30% in Asian countries like Japan and Korea.1 Moreover, ICH is associated with the highest mortality of all cerebrovascular events, and most survivors never regain functional independence.2 The incidence of ICH is increasing in elderly people in some developed countries, in part because of the increased use of warfarin medication.3,4 Warfarin remains a highly effective therapy for prevention of thromboembolic strokes in common clinical situations like atrial fibrillation, but it increases both the risk of developing ICH and mortality. More severe outcomes in warfarin-related ICH are associated with an increased risk of hematoma expansion.5 Thus, the prevention of ICH in patients with warfarin medication is critical, and identification of risk factors for bleeding remains an important issue. Cerebral microbleeds are seen as small round hypointense lesions on T2*-weighted gradientecho (GRE) MRI, and they are pathologically tiny extravasations of blood from lipohyalinized Supplemental data at www.neurology.org *These authors contributed equally. From the Department of Neurology (S.-H.L., W.-S.R., J.-K.R.) and Clinical Research Center for Stroke, Clinical Research Institute (S.-H.L.), Seoul National University Hospital, Seoul, Republic of Korea. Supported by grants of the Korea Health 21 R&D Project, Ministry of Health and Welfare, Republic of Korea (A060143, A060171, and A060263). Disclosure: The authors report no disclosures. Copyright © 2009 by AAN Enterprises, Inc.
171
cerebral arterioles. 6 Hypertension, 7 old age,8 low serum cholesterol,9 cerebral amyloid angiopathy,10 and glycated hemoglobin11 have been demonstrated as risk factors for these lesions. It is of potential importance that the lesions are closely associated with spontaneous ICH,12 and they may predict future occurrence or recurrence of ICH.13,14 These lesions are also associated with the severity of ICH. 15 Microbleeds may be indicative of a bleeding-prone state in the brain, and they may be regarded as risk lesions of spontaneous ICH. However, it has been understood that the lesions are not associated with hemorrhagic transformation after acute ischemic stroke in patients treated with thrombolysis.16,17 This difference on predictability of subsequent hemorrhagic events may be related to differences of bleeding mechanism-reperfusion to the ischemic tissue (hemorrhagic transformation) vs rupture of arteriosclerotic arterioles (ICH). To our knowledge, an association between microbleeds and incident warfarin-related ICH has not been studied. In this case-control study, we investigated whether microbleeds are associated with the incidence of ICH in patients taking warfarin medication. METHODS Study population. All patients with ICH while on outpatient treatment with warfarin were selected from consecutive patients aged ⱖ45 years admitted to the Seoul National University Hospital with ICH. The electric medical record system in our hospital was used to identify patients between January 2002 and July 2007. We reviewed medical records to confirm whether patients were eligible for our study. Patients were excluded if their ICH was due to head trauma or acute ischemic stroke with hemorrhagic transformation. Patients with brain tumor, vascular malformation, vasculitis of the CNS, or invalid medical records were also excluded. We excluded patients who were not taking warfarin at the time of hemorrhage. From the database, 54 patients were matches for the diagnosis of symptomatic warfarin-related ICH. The initial provisional diagnosis of ICH was not confirmed by imaging studies in eight subjects. Twenty-two patients did not undergo brain MRI before or at the time of ICH incidence (economic problem, n ⫽ 13; metallic materials in the body [e.g., pacemaker], n ⫽ 3; poor cooperation, n ⫽ 6). Thus, 24 patients were finally included as cases for analysis (table e-1 on the Neurology® Web site at www. neurology.org). Characteristics of selected patients were not different from those of unselected patients (table e-2). We selected control subjects from warfarin users with no history of ICH during the same time period. From the database, we found 1,970 patients who meet these conditions. We reviewed medical records to identify eligible controls. We excluded 172
Neurology 72
January 13, 2009
patients who had a previous ICH or who did not receive brain MRI. Two controls per a case were randomly selected, and subjects were matched for sex and age. Based on the patients’ medical records, most of the brain MR imaging studies were performed to evaluate suspicious neurologic symptoms (headache, dizziness, or subjective weakness). Finally, 48 patients were included as controls. All study procedures were approved by our institutional review board.
Clinical information. We reviewed medical records for cases and controls to obtain demographic data, concurrent antiplatelet medication, previous medical history, and an indication for warfarin. Hypertension and coronary heart disease were defined as present if they had been diagnosed and treated by a physician. Diabetes was defined as present if the patient’s fasting glucose level was at least 7.0 mmol/L or if the patient had taken a hypoglycemic agent. A diagnosis of hypercholesterolemia was based on a history of hypercholesterolemia with medication or a fasting serum cholesterol level ⬎6.2 mmol/L. A history of smoking was defined as present if the subject was a current smoker or an ex-smoker who had quit smoking within 5 years of admission. Indications of warfarin included atrial fibrillation, valvular heart disease, undetermined causes, and other-determined causes. The prothrombin time (PT)–international normalized ratio (INR) values for patients with ICH were recorded during presentation to the emergency room. For the matched control subjects, the PT-INR values were determined by selecting the date from our database that was closest to the date of emergency room presentation.
Image analysis. MRI was performed on a 1.5 T superconducting magnet system (GE Medical System, Milwaukee, WI). GRE MRI was performed as a part of the routine protocol in our hospital, and the images were obtained in the axial plane with the following parameters: repetition time/echo time, 500/15 msec; flip angle, 26°; matrix size, 256 ⫻ 192; slice thickness, 6 mm; and gap width, 2 mm. Standard T2-weighted and fluidattenuated inversion recovery (FLAIR) sequences were also obtained. Cerebral microbleeds were defined as well-defined focal areas of low signal on the GRE MRI of less than 5 mm in diameter, and they were counted throughout the whole brain by study neurologists (S.-H.L., W.-S.R.) blinded to the clinical characteristics. We did not include the microbleeds around the symptomatic ICH lesion because the lesions might be caused by the ICH itself. Interobserver agreement on the presence of microbleeds was found to be excellent ( ⫽ 0.88). The presence and numbers of microbleeds were finally determined by consensus between the readers. Because some cases exhibited microbleeds and symptomatic ICH in the same GRE images, total blinding of the readers of the MRI for the presence of ICH was difficult. The GRE images were re-reviewed by other investigators (Drs. E.K. Bae and J.I. Cha) who did not know the hypothesis of this study and did not participate in this study as authors. Because interobserver agreement between the original data and the re-reviewed data were excellent ( ⫽ 0.83), we do not believe that our reading was biased by the lack of blinding to the presence of ICH. White matter hyperintensities (WMHs) were diagnosed according to the criteria of Fazekas et al.18 based on whether the subject had one of the following: 1) multiple periventricular hyperintense punctate lesions and early confluence or 2) multiple areas reaching confluence seen on T2-weighted or FLAIR images.
Statistical analysis. For baseline comparisons, dichotomous variables were compared between groups using the 2 test or Fisher exact analysis, and continuous variables (age, PT-INR, and number of microbleeds) were compared by the Wilcoxon
Table 1
ICH cases (n ⴝ 24)
Variables Gender (men)
9 (37.5)
Controls (n ⴝ 48)
p Value
18 (37.5)
NS
Age, y
65.0 ⫾ 8.4
65.0 ⫾ 8.5
NS*
Hypertension
13 (54.2)
23 (47.9)
NS
Diabetes
8 (33.3)
11 (22.9)
NS
Hyperlipidemia
6 (25.0)
9 (18.8)
NS
Smoking
1 (4.2)
9 (18.8)
NS†
Previous stroke
12 (50)
Previous use of antiplatelet agents
24 (50)
3 (12.5)
4 (8.3)
7 (29.2)
20 (41.7)
10 (41.7)
12 (25.0)
7 (29.2)
16 (33.3)
Indication for warfarin medication Atrial fibrillation Valvular replacement Others
NS NS† NS
41.2 ⫾ 15.3
Duration of warfarin treatment
35.4 ⫾ 18.8
Successful duration of warfarin treatment
NS* NS†
0–25%
1 (4.2)
5 (10.4)
26–50%
10 (41.7)
14 (29.2)
51–75%
11 (45.8)
23 (47.9)
76–100%
2 (8.3) 3.22 ⫾ 1.88
PT-INR
6 (12.5) 2.21 ⫾ 0.62
0.001*
Values are n (%) or mean ⫾ standard deviation. Pearson 2 test, *Student t test, and †Fisher exact test were used. ICH ⫽ intracerebral hemorrhage; NS ⫽ not significant; PT-INR ⫽ prothrombin timeinternational normalized ratio.
rank sum test. To find a correlation between the number of microbleeds and the presence of warfarin-related ICH, we used Spearman correlation analysis. Finally, we used conditional logistic regression models with the matched sets to determine which variable was associated with an occurrence of ICH. In this analysis, WMHs and microbleeds were analyzed as dichotomous variables. A two-tailed p value of ⬍0.05 was considered to be significant. Data analysis was performed with SPSS (version 13.0, Chicago, IL).
Table 2
Radiologic findings
Variables
ICH cases (n ⴝ 24)
Controls (n ⴝ 48)
Presence of microbleeds
19 (79.2%)
11 (22.9%)
No. of microbleeds
9.0 ⫾ 26.8
p Value ⬍0.001
0.5 ⫾ 1.03
⬍0.001*
Location of microbleeds Lobar area
14 (58.3%; 73.7%‡)
5 (10.4%; 45.5%‡)
⬍0.001§
Basal ganglia
10 (41.7%; 52.6%‡)
2 (4.2%; 18.2%‡)
⬍0.001†§ ⬍0.001§
Thalamus
7 (29.2%; 36.8%‡)
6 (12.5%; 54.5%‡)
Brainstem
5 (20.8%; 26.3%‡)
0 (0%; 0%‡)
0.005†§
Cerebellum
7 (29.2%; 36.8%‡)
2 (4.2%; 18.2%‡)
0.008†§
WMH
12 (50.0%)
Characteristics of the patients included in this study are described in table 1. There were higher numbers of women (62.5% in both groups) with no significant difference. Hypertension was the most frequent vascular risk factor in both ICH case and control subjects, and smoking was more frequent in control subjects. However, none of the demographic variables were significantly different between groups. Warfarin was largely used for the prevention of stroke in the patients with atrial fibrillation or valvular replacement treatment. The durations of warfarin treatment were not different between the ICH cases and controls and, when we set the target range of PT-INR to 2.0 –3.0 according to the general recommendation, the successful time of anticoagulation was not also different between the groups. PT-INR was significantly higher in the ICH cases than in the control subjects (3.22 ⫾ 1.88 vs 2.21 ⫾ 0.62; p ⫽ 0.001). In the patients with ICH, the location of ICH was quite typical: 8 patients (33.3%) had ICH in the lobar area, 7 (29.2%) in the basal ganglia, 2 (8.3%) in the thalamus, 1 (4.2%) in the brainstem, and 6 (25.0%) in the cerebellum. In addition, ratio of patients under adequate blood pressure control in the hypertensive subjects (⬍140/90 mm Hg) was 84.6% in the ICH cases and 82.6% in the controls. See table e-2 for comparisons with unselected cases (n ⫽ 22) and controls (n ⫽ 837). As illustrated in table 2, there were more patients with microbleeds in the ICH than control group (79.2% vs 22.9%: p ⬍ 0.001), and the number of microbleeds was much higher in the ICH than control group (9.0 ⫾ 26.8 vs 0.5 ⫾ 1.03: p ⬍ 0.001). Moreover, Spearman correlation analysis revealed that a greater number of microbleeds was associated with a greater risk of ICH (r ⫽ 0.299; p ⬍ 0.001). Distribution of microbleeds appeared to be different between the groups. The most frequently involved area was the lobar area (58.3%) for ICH cases but thalamus (12.5%) for control subjects. In all the brain areas, the number of patients with microbleeds was larger for ICH cases than control subjects. Finally, we examined whether the effects of microbleeds on warfarin-related ICH were independent of other important clinical variables. Some variables were associated with warfarin-related ICH in the univariate analyses: PT-INR (odds ratio [OR], 2.51; 95% confidence interval [CI], 1.32– 4.79), WMHs (OR, 3.00; 95% CI, 1.07– 8.43), and the presence of microbleeds (OR, 12.78; 95% CI, 3.88 – 42.15). With regard to the location, microbleeds in the lobar area, basal ganglia, and cerebellum were significantly associated with warfarin-related ICH. Microbleeds in the brainstem and thalamus were not significantly associated (data not shown). In the conditional logisRESULTS
Baseline characteristics
12 (25.0%)
0.063
Pearson 2 test, *Student t test, and †Fisher exact test were used. ‡Percentages are proportions among the patients with microbleeds (n ⫽ 19 vs n ⫽ 11). § The percentages of total cases between both groups were compared. ICH ⫽ intracerebral hemorrhage; WMH ⫽ white matter hyperintensity.
Neurology 72
January 13, 2009
173
Table 3 Variables PT-INR Presence of microbleeds WMH
Conditional logistic regression analysis OR 4.13 83.12 3.60
95% CI
p Value
1.40–12.77
0.011
5.96–1,159.10
0.001
0.70–113.64
0.093
OR ⫽ odds ratio; CI ⫽ confidence interval; PT-INR ⫽ prothrombin time-international normalized ratio; WMH ⫽ white matter hyperintensity.
tic regression analysis, the presence of microbleeds was independently related to warfarin-related ICH. Furthermore, the effect of microbleeds was the most powerful among the variables (table 3). In contrast, WMH did not remain significant after conditional logistic regression analysis (p ⫽ 0.093). DISCUSSION In this case-control study, we found that PT-INR and the presence of microbleeds on GRE MRI were independently associated with incident ICH in the patients undergoing warfarin treatment. In addition, the number of microbleeds was correlated with the incidence of warfarin-related ICH, and microbleeds in the lobar area and basal ganglia were more significantly associated. WMHs were associated with the incidence of warfarinrelated ICH in a univariate analysis, but the significance did not remain after adjustment for PT-INR and the presence of microbleeds. Cerebral microbleeds indicate previous extravasation of blood and signify bleeding-prone cerebral microangiopathy.6,15,19,20 It was first suggested that these lesions would predict all types of hemorrhages in the brain, including primary ICH and hemorrhagic transformation after acute ischemic stroke. With regard to hemorrhagic transformation after acute ischemic stroke, two observational studies19,21 indicated that presence of baseline microbleeds predicted incident hemorrhagic transformation in patients with or without thrombolytic treatment. However, recent prospective multicenter studies have failed to confirm these results in the patients who underwent IV thrombolysis.16,17,22 Furthermore, we recently found that the presence of baseline microbleeds is not associated with subsequent hemorrhagic transformation after acute atherothrombotic stroke regardless of thrombolytic therapy.23 Considering these results, microbleeds are not likely to be associated with the risk of hemorrhagic transformation in acute ischemic stroke. In contrast, microbleeds are closely associated with primary ICH. Several cross-sectional and prospective studies indicated that the presence or severity of microbleeds is strongly related to the incidence or recurrence of primary ICH.12-14 We believe that 174
Neurology 72
January 13, 2009
the differential effects of microbleeds on hemorrhagic transformation and ICH may be related to differences in the pathogenesis of hemorrhage. The underlying histopathologies of ICH include the rupture of penetrating arterioles damaged by hyaline degeneration and microaneurysm formation caused by longstanding hypertension or aging.24-26 This is quite the same in microbleeds, but the hemorrhagic transformation is caused by ischemic injury to the microvasculature in an extensive brain infarction.27 Warfarin-related ICH may share the mechanism of hemorrhage with primary ICH rather than hemorrhagic transformation. In a pooled analysis of five primary stroke prevention trials, adjusted-dose warfarin reduced all strokes by approximately 70% in patients with atrial fibrillation.28 This preventive effect of warfarin is diminished by a significant 0.2% increase in the annual risk of ICH.29 Several risk factors for warfarin-related ICH have been reported. First, advancing age is one of the most important risk factors consistently reported by most studies.30,31 The relationship might be associated with an increased frequency of vascular rupture-related microangiopathic findings in cerebral amyloid angiopathy or hypertension in aged patients. In addition, racial background was also suggested as a new risk factor. A recent retrospective multiethnic study with a stroke-free atrial fibrillation cohort32 indicated that the risk of warfarin-related ICH was significantly higher in Hispanic (hazard ratio, 2.04) and Asian (hazard ratio, 4.06) than Caucasian white patients. Furthermore, an increased PTINR value reaching 4.0 to 5.0 is another important risk factor for ICH.31,33,34 In many cases, however, warfarin-related ICHs occurred in the patients with the optimal range of PT-INR.35 In the present study, the PT-INR value in ICH cases was slightly increased (3.22 ⫾ 1.88) compared to that of controls. Finally, underlying microangiopathy, such as cerebral amyloid angiopathy, may be associated with warfarin-related ICH; however, this causality has not yet been established. Aging is a global phenomenon, and hemorrhagic stroke caused by cerebral amyloid angiopathy would be more important than expected. In this context, the differential effects of cerebral amyloid angiopathy and hypertensive microangiopathy on warfarin-related ICH may be important in clinical practice. Thus, further studies on this issue should be conducted. The Stroke Prevention In Reversible Ischemia Trial first suggested that white matter hypodensity lesions in CT scans might be associated with an increased risk of warfarin-related ICH, and this risk remained significant after adjustment for age, blood pressure, and PT-INR value.30 This suggestion was
confirmed by a case-control study conducted in a single hospital.35 In the present study, we investigated an association between the presence of WMHs seen on T2-weighted or FLAIR images and the incidence of warfarin-related ICH. The effect of WMHs was significant in univariate analysis, but it did not remain significant after adjustment for possible confounders like the presence of microbleeds. These results signify that microbleeds are more specifically effective on incident warfarin-related ICH than WMHs. In fact, the presence and severity of microbleeds are strongly correlated with those of WMHs.36 However, microbleeds have a greater association with the incidence or recurrence of primary ICH than WMHs.37 This may arise because WMHs are closely associated with microangiopathic findings of ischemic nature whereas microbleeds represent a bleedingprone microangiopathy. There are important caveats in this study. First, in some patients, microbleeds and ICH were found on the same GRE images. The total blinding of presence of ICH was not possible in this context. To achieve sufficient validity, we invited two independent neurologists as reviewers who did not know the hypothesis of this study as indicated in Methods. As a result, we found that the re-reviewed data were nearly identical to the original review data. We do not believe that the lack of total blinding in this study seriously affected the study results. Second, because this study was conducted by a retrospective collection of data, selective brain MRI scanning possibly occurred in this situation. Moreover, the clinical practice guidelines have not recommended brain MRI scanning as a primary tool for diagnosis of acute hemorrhagic stroke. However, our hospital has performed brain MRI on all stroke patients regardless of stroke subtype to detect concomitant cerebrovascular lesions. Finally, because the cases and controls were only matched for age and sex, the groups were not fully matched for warfarin indications. Statistical methods were used to adjust for differences in warfarin indications, but we acknowledge that the adjustment may not be complete. Microbleeds are a radiologic finding that is more often associated with ICH than other stroke subtypes. General application of brain MRI is not recommended for detecting the risk of ICH in the asymptomatic elderly population with vascular risk factors due to its high cost. If it is limited to the patients taking warfarin medication, however, brain MRI might be useful because warfarin-related ICH is a very critical complication in this population. Future research should focus on the risks and benefits of warfarin medication in patients with microbleeds.
ACKNOWLEDGMENT The authors thank Drs. Eun-Gi Bae and Jung-In Cha for technical assistance.
Received April 21, 2008. Accepted in final form October 3, 2008.
REFERENCES 1. Broderick J, Connolly S, Feldmann E, et al. Guidelines for the management of spontaneous intracerebral hemorrhage in adults: 2007 update: a guideline from the American Heart Association/American Stroke Association Stroke Council, High Blood Pressure Research Council, and the Quality of Care and Outcomes in Research Interdisciplinary Working Group. Stroke 2007;38:2001–2023. 2. McGuire AJ, Raikou M, Whittle I, Christensen MC. Long-term mortality, morbidity and hospital care following intracerebral hemorrhage: an 11-year cohort study. Cerebrovasc Dis 2007;23:221–228. 3. Lovelock CE, Molyneux AJ, Rothwell PM. Change in incidence and aetiology of intracerebral haemorrhage in Oxfordshire, UK, between 1981 and 2006: a population-based study. Lancet Neurol 2007;6:487–493. 4. Flaherty ML, Kissela B, Woo D, et al. The increasing incidence of anticoagulant-associated intracerebral hemorrhage. Neurology 2007;68:116–121. 5. Flibotte JJ, Hagan N, O’Donnell J, Greenberg SM, Rosand J. Warfarin, hematoma expansion, and outcome of intracerebral hemorrhage. Neurology 2004;63:1059–1064. 6. Fazekas F, Kleinert R, Roob G, et al. Histopathologic analysis of foci of signal loss on gradient-echo T2*-weighted MR images in patients with spontaneous intracerebral hemorrhage: evidence of microangiopathy-related microbleeds. AJNR Am J Neuroradiol 1999;20:637–642. 7. Tanaka A, Ueno Y, Nakayama Y, Takano K, Takebayashi S. Small chronic hemorrhages and ischemic lesions in association with spontaneous intracerebral hematomas. Stroke 1999;30:1637–1642. 8. Roob G, Schmidt R, Kapeller P, Lechner A, Hartung HP, Fazekas F. MRI evidence of past cerebral microbleeds in a healthy elderly population. Neurology 1999;52:991–994. 9. Lee SH, Bae HJ, Yoon BW, Kim H, Kim DE, Roh JK. Low concentration of serum total cholesterol is associated with multifocal signal loss lesions on gradient-echo magnetic resonance imaging: analysis of risk factors for multifocal signal loss lesions. Stroke 2002;33:2845–2849. 10. Greenberg SM, Finklestein SP, Schaefer PW. Petechial hemorrhages accompanying lobar hemorrhage: detection by gradient-echo MRI. Neurology 1996;46:1751–1754. 11. Viswanathan A, Guichard JP, Gschwendtner A, et al. Blood pressure and haemoglobin A1c are associated with microhaemorrhage in CADASIL: a two-centre cohort study. Brain 2006;129:2375–2383. 12. Lee SH, Bae HJ, Kwon SJ, et al. Cerebral microbleeds are regionally associated with intracerebral hemorrhage. Neurology 2004;62:72–76. 13. Fan YH, Zhang L, Lam WW, Mok VC, Wong KS. Cerebral microbleeds as a risk factor for subsequent intracerebral hemorrhages among patients with acute ischemic stroke. Stroke 2003;34:2459–2462. 14. Greenberg SM, Eng JA, Ning M, Smith EE, Rosand J. Hemorrhage burden predicts recurrent intracerebral hemorrhage after lobar hemorrhage. Stroke 2004;35:1415– 1420. Neurology 72
January 13, 2009
175
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25. 26.
27.
176
Lee SH, Kim BJ, Roh JK. Silent microbleeds are associated with volume of primary intracerebral hemorrhage. Neurology 2006;66:430–432. Derex L, Nighoghossian N, Hermier M, et al. Thrombolysis for ischemic stroke in patients with old microbleeds on pretreatment MRI. Cerebrovasc Dis 2004;17:238–241. Fiehler J, Albers GW, Boulanger JM, et al. Bleeding risk analysis in stroke imaging before thromboLysis (BRASIL): pooled analysis of T2*-weighted magnetic resonance imaging data from 570 patients. Stroke 2007;38:2738–2744. Fazekas F, Chawluk JB, Alavi A, Hurtig HI, Zimmerman RA. MR signal abnormalities at 1.5 T in Alzheimer’s dementia and normal aging. AJR Am J Roentgenol 1987; 149:351–356. Kidwell CS, Saver JL, Villablanca JP, et al. Magnetic resonance imaging detection of microbleeds before thrombolysis: an emerging application. Stroke 2002;33:95–98. Jeerakathil T, Wolf PA, Beiser A, et al. Cerebral microbleeds: prevalence and associations with cardiovascular risk factors in the Framingham Study. Stroke 2004;35:1831– 1835. Nighoghossian N, Hermier M, Adeleine P, et al. Old microbleeds are a potential risk factor for cerebral bleeding after ischemic stroke: a gradient-echo T2*-weighted brain MRI study. Stroke 2002;33:735–742. Kakuda W, Thijs VN, Lansberg MG, et al. Clinical importance of microbleeds in patients receiving IV thrombolysis. Neurology 2005;65:1175–1178. Lee SH, Kang BS, Kim N, Roh JK. Does microbleed predict haemorrhagic transformation after acute atherothrombotic or cardioembolic stroke? J Neurol Neurosurg Psychiatry 2008;79:913–916. Fisher CM. Pathological observations in hypertensive cerebral hemorrhage. J Neuropathol Exp Neurol 1971;30: 536–550. Fisher CM. Cerebral miliary aneurysms in hypertension. Am J Pathol 1972;66:313–330. Cole FM, Yates PO. Comparative incidence of cerebrovascular lesions in normotensive and hypertensive patients. Neurology 1968;18:255–259. del Zoppo GJ, von Kummer R, Hamann GF. Ischaemic damage of brain microvessels: inherent risks for thrombo-
Neurology 72
January 13, 2009
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
lytic treatment in stroke. J Neurol Neurosurg Psychiatry 1998;65:1–9. Risk factors for stroke and efficacy of antithrombotic therapy in atrial fibrillation. Analysis of pooled data from five randomized controlled trials. Arch Intern Med 1994;154: 1449–1457. van Walraven C, Hart RG, Singer DE, et al. Oral anticoagulants vs aspirin in nonvalvular atrial fibrillation: an individual patient meta-analysis. JAMA 2002;288:2441– 2448. Gorter JW. Major bleeding during anticoagulation after cerebral ischemia: patterns and risk factors. Stroke Prevention In Reversible Ischemia Trial (SPIRIT). European Atrial Fibrillation Trial (EAFT) study groups. Neurology 1999;53:1319–1327. Palareti G, Leali N, Coccheri S, et al. Bleeding complications of oral anticoagulant treatment: an inception-cohort, prospective collaborative study (ISCOAT). Italian Study on Complications of Oral Anticoagulant Therapy. Lancet 1996;348:423–428. Shen AY, Yao JF, Brar SS, Jorgensen MB, Chen W. Racial/ ethnic differences in the risk of intracranial hemorrhage among patients with atrial fibrillation. J Am Coll Cardiol 2007;50:309–315. Optimal oral anticoagulant therapy in patients with nonrheumatic atrial fibrillation and recent cerebral ischemia. The European Atrial Fibrillation Trial Study Group. N Engl J Med 1995;333:5–10. A randomized trial of anticoagulants versus aspirin after cerebral ischemia of presumed arterial origin. The Stroke Prevention in Reversible Ischemia Trial (SPIRIT) Study Group. Ann Neurol 1997;42:857–865. Smith EE, Rosand J, Knudsen KA, Hylek EM, Greenberg SM. Leukoaraiosis is associated with warfarin-related hemorrhage following ischemic stroke. Neurology 2002;59: 193–197. Koennecke HC. Cerebral microbleeds on MRI: prevalence, associations, and potential clinical implications. Neurology 2006;66:165–171. Lee SH, Heo JH, Yoon BW. Effects of microbleeds on hemorrhage development in leukoaraiosis patients. Hypertens Res 2005;28:895–899.
SPECIAL ARTICLE
Practice Parameter: Evaluation of distal symmetric polyneuropathy: Role of autonomic testing, nerve biopsy, and skin biopsy (an evidence-based review) Report of the American Academy of Neurology, American Association of Neuromuscular and Electrodiagnostic Medicine, and American Academy of Physical Medicine and Rehabilitation
J.D. England, MD G.S. Gronseth, MD, FAAN G. Franklin, MD G.T. Carter, MD L.J. Kinsella, MD J.A. Cohen, MD A.K. Asbury, MD K. Szigeti, MD, PhD J.R. Lupski, MD, PhD N. Latov, MD R.A. Lewis, MD P.A. Low, MD M.A. Fisher, MD D.N. Herrmann, MD J.F. Howard, Jr., MD G. Lauria, MD R.G. Miller, MD M. Polydefkis, MD, MHS A.J. Sumner, MD
ABSTRACT
Address correspondence and reprint requests to the American Academy of Neurology, 1080 Montreal Avenue, St. Paul, MN 55116
[email protected]
GLOSSARY
Supplemental data at www.neurology.org
Background: Distal symmetric polyneuropathy (DSP) is the most common variety of neuropathy. Since the evaluation of this disorder is not standardized, the available literature was reviewed to provide evidence-based guidelines regarding the role of autonomic testing, nerve biopsy, and skin biopsy for the assessment of polyneuropathy.
Methods: A literature review using MEDLINE, EMBASE, and Current Contents was performed to identify the best evidence regarding the evaluation of polyneuropathy published between 1980 and March 2007. Articles were classified according to a four-tiered level of evidence scheme and recommendations were based upon the level of evidence.
Results and Recommendations: 1) Autonomic testing should be considered in the evaluation of patients with polyneuropathy to document autonomic nervous system dysfunction (Level B). Such testing should be considered especially for the evaluation of suspected autonomic neuropathy (Level B) and distal small fiber sensory polyneuropathy (SFSN) (Level C). A battery of validated tests is recommended to achieve the highest diagnostic accuracy (Level B). 2) Nerve biopsy is generally accepted as useful in the evaluation of certain neuropathies as in patients with suspected amyloid neuropathy, mononeuropathy multiplex due to vasculitis, or with atypical forms of chronic inflammatory demyelinating polyneuropathy (CIDP). However, the literature is insufficient to provide a recommendation regarding when a nerve biopsy may be useful in the evaluation of DSP (Level U). 3) Skin biopsy is a validated technique for determining intraepidermal nerve fiber density and may be considered for the diagnosis of DSP, particularly SFSN (Level C). There is a need for additional prospective studies to define more exact guidelines for the evaluation of polyneuropathy. Neurology® 2009;72:177–184
AAN ⫽ American Academy of Neurology; AANEM ⫽ American Academy of Neuromuscular and Electrodiagnostic Medicine; AAPM&R ⫽ American Academy of Physical Medicine and Rehabilitation; ART ⫽ autonomic reflex testing; BRSI ⫽ baroreflex sensitivity index; CASS ⫽ composite autonomic scoring scale; CIDP ⫽ chronic inflammatory demyelinating polyneuropathy; DSFN ⫽ distal small fiber neuropathy; DSP ⫽ distal symmetric polyneuropathy; EDx ⫽ electrodiagnosis; EFNS ⫽ European Federation of Neurological Societies; HRV ⫽ heart rate variability; IAN ⫽ idiopathic autonomic neuropathy; IENF ⫽ intraepidermal nerve fibers; MSNA ⫽ muscle sympathetic nerve activity; NCSs ⫽ nerve conduction studies; PGP 9.5 ⫽ protein-gene-product 9.5; PN ⫽ peripheral neuropathy; PRT ⫽ blood pressure recovery time; QAE ⫽ quantitative autonomic examination; QSART ⫽ quantitative sudomotor axon reflex test; QSS ⫽ Quality Standards Subcommittee; QST ⫽ quantitative sensory testing; SFSN ⫽ small fiber sensory polyneuropathy; TST ⫽ thermoregulatory sweat testing.
Polyneuropathy is a relatively common neurologic disorder.1 The overall prevalence is approximately 2,400 (2.4%) per 100,000 population, but in individuals older than 55 years, the prevalence rises to approximately 8,000 (8%) per 100,000.2,3 Since
there are many etiologies of polyneuropathy, a logical clinical approach is needed for evaluation and management. This practice parameter provides recommendations for the evaluation of distal symmetric polyneu-
See page 185 e-Pub ahead of print on December 3, 2008, at www.neurology.org. Published simultaneously in PM&R and Muscle & Nerve. Authors’ affiliations are listed at the end of the article. The AAN Mission Statement, Classification of evidence, Classification of recommendations, and Conflict of Interest Statement (appendices e-1 through e-4), as well as references e1– e16, are available as supplemental data on the Neurology威 Web site at www.neurology.org. Approved by the Quality Standards Subcommittee on November 10, 2007; by the AAN Practice Committee on January 30, 2008; by the Neuromuscular Guidelines Steering Committee on April 22, 2008; by the AAN Board of Directors on August 20, 2008; by the AANEM Board of Directors on May 1, 2008; and by the AAPM&R Board of Governors on April 7, 2008. Disclosure: Author disclosures are provided at the end of the article. Copyright © 2009 by AAN Enterprises, Inc.
177
ropathy (DSP) based upon a prescribed review and analysis of the peer-reviewed literature. The parameter was developed to provide physicians with evidence-based guidelines regarding the role of autonomic testing, nerve biopsy, and skin biopsy for the assessment of polyneuropathy. The diagnosis of DSP should be based upon a combination of clinical symptoms, signs, and electrodiagnostic criteria as outlined in the previous case definition.1 See Mission statement (appendix e-1 on the Neurology® Web site at www.neurology.org) for details. The Polyneuropathy Task Force included 19 physicians with representatives from the American Academy of Neurology (AAN), the American Academy of Neuromuscular and Electrodiagnostic Medicine (AANEM), and the American Academy of Physical Medicine and Rehabilitation (AAPM&R). All of the task force members had extensive experience and expertise in the area of polyneuropathy. Additionally, four members had expertise in evidence-based methodology and practice parameter development. Two are current members (J.D.E., G.F.), and two are former members (G.S.G, R.G.M.) of the Quality Standards Subcommittee (QSS) of the AAN. The task force developed a set of clinical questions relevant to the evaluation of DSP, and subcommittees were formed to address each of these questions.
FORMATION OF EXPERT PANEL
DESCRIPTION OF THE ANALYTIC PROCESS
The literature search included OVID MEDLINE (1966 to March 2007), OVID Excerpta Medica (EMBASE; 1980 to March 2007), and OVID Current Contents (2000 to March 2007). The search included articles on humans only and in all languages. The search terms selected were peripheral neuropathy, polyneuropathy, and distal symmetric polyneuropathy. These terms were cross-referenced with the terms diagnosis, electrophysiology, autonomic testing, nerve biopsy, and skin biopsy. Panel experts were asked to identify additional articles missed by the initial search strategy. Further, the bibliographies of the selected articles were reviewed for potentially relevant articles. Subgroups of committee members reviewed the titles and abstracts of citations identified from the original searches and selected those that were potentially relevant to the evaluation of polyneuropathy. Articles deemed potentially relevant by any panel member were also obtained. Each potentially relevant article was subsequently reviewed in entirety by at least three panel members. Each reviewer graded the risk of bias in each article by using the diagnostic test classification-of-evidence scheme (appendix e-2). In this scheme, articles at178
Neurology 72
January 13, 2009
taining a grade of Class I are judged to have the lowest risk of bias, and articles attaining a grade of Class IV are judged to have the highest risk of bias. Disagreements among reviewers regarding an article’s grade were resolved through discussion. Final approval was determined by the entire panel. The AAN’s method for determining the strength of recommendation was used (appendix e-3). The QSS (AAN; appendix 1), the Practice Issues Review Panel (AANEM; appendix 2), and the Practice Guidelines Committee (AAPM&R; appendix 3) reviewed and approved a draft of the article. The draft was next sent to members of the AAN, AANEM, and AAPM&R for further review and then to Neurology® for peer review. Boards of the AAN, AANEM, and AAPM&R reviewed and approved the final version of the article. At each step of the review process, external reviewers’ suggestions were explicitly considered. When appropriate, the expert panel made changes to the document. ANALYSIS OF EVIDENCE The search yielded 1,045 references with abstracts. After reviewing titles and abstracts, 106 articles were reviewed and classified.
Role of clinical autonomic testing in the evaluation of polyneuropathy. Autonomic nervous system dysfunc-
tion occurs in several phenotypes. It may occur as one component of a generalized polyneuropathy such as DSP of diabetes. Such polyneuropathies are usually diagnosed by a combination of neuropathic symptoms, decreased or absent ankle reflexes, decreased distal sensation, distal muscle weakness or atrophy, and abnormal nerve conduction studies (NCSs).1 The majority of these features constitute evidence of “large fiber” sensory and motor involvement. However, signs of autonomic nervous system involvement may also constitute findings indicative of DSP. In DSP with autonomic involvement, the most common clinical findings are abnormalities of sweating and circulatory instability in the feet.1,3 A second phenotype is that of an autonomic neuropathy such as in amyloidosis and autoimmune autonomic neuropathy, where autonomic nerves are affected disproportionately relative to somatic nerves.4 In these neuropathies, autonomic fibers can be affected in isolation and their involvement may precede somatic fiber involvement.5 A third relatively common phenotype is distal small fiber sensory polyneuropathy (SFSN), which can manifest as burning pain affecting the feet, often with allodynia and sometimes with erythromelalgia (red hot and painful skin). Involvement of autonomic and somatic C fibers usually occurs concurrently in small fiber polyneuropathy.5
QSART ⫽ quantitative sudomotor axon reflex testing; QST ⫽ quantitative sensory testing; HRV ⫽ heart rate variability; EDx ⫽ electrodiagnosis; PN ⫽ peripheral neuropathy; QAE ⫽ quantitative autonomic examination; ART ⫽ autonomic reflex testing; CASS ⫽ composite autonomic severity score; BRSI ⫽ baroreflex sensitivity index; MSNA ⫽ muscle sympathetic nerve activity; PRT ⫽ blood pressure recovery time; DSFN ⫽ distal small fiber neuropathy; DN ⫽ diabetic neuropathy; IAN ⫽ idiopathic autonomic neuropathy.
90% 95% III Unmasked/ independent N Concurrent comparative 38 11 Neurologic exam CASS 2004 5
DSFN, PN, DN, IAN
⬎90% ⬎90% 2005 12
Multisystem atrophy, PN
PRT, CASS
Clinical exam
162
32
Concurrent comparative
B
Unmasked/ independent
II
⬎90% 86% II Unmasked/ independent B Concurrent comparative 29 84 MSNA BRSI 2007 13
Adrenergic autonomic failure
⬎90% ⬎90% 1993 9
Diabetic PN
CASS
EDx and standard clinical exam
78
350
Concurrent comparative
N
Unmasked/ independent
II
94% ART: 93%; QSART: 73% III N N Noncomparative 357* (per Dr. Low) 126 Clinical evaluation QSART, ART, CASS 2001 19
Painful neuropathy
93% QSART: 80%; QST: 67% III Unmasked/ independent B Concurrent comparative 357* (per Dr. Low) 138 EDx QSART, QST, clinical symptoms 1999 18
Peripheral (small fiber) neuropathy
⬎90% ⬎90% 1997 7
PN, Parkinson, multisystem atrophy
QSART
Older scale
18
557
Concurrent comparison
B
Unmasked/ independent
II
⬎90% QAE: 97% II Unmasked/ independent B Concurrent comparative 357* 380 EDx QAE 1992 6
Diabetic PN
Spec
72% 80%
Sens Class
III N
Masked Spectrum
N Retrospective review
Design Controls
129 40
Cases Reference standard
Neurologic exam and EDx QSART, QST, HRV
Predictor Target disorder
Distal small fiber neuropathy
Year
1992
Reference
17
Evidence table for autonomic testing Table
What is the usefulness of clinical autonomic testing in the evaluation of polyneuropathy, and which tests have the highest sensitivity and specificity? Currently available autonomic tests can provide indices of cardiovagal, adrenergic, and postganglionic sudomotor function. As such, they provide indices for both parasympathetic and sympathetic autonomic function. Heart rate variability testing is a simple and reliable test of cardiovagal function. It detects the presence of diabetic polyneuropathy with nearly the same sensitivity as NCSs (Class II).6 Specificity is high (97.5%) for identifying parasympathetic deficits if the recommended age-controlled values are used (Class II).7 Intrinsic cardiac disease can affect the results of this test, and this possibility must be considered in the interpretation. Cardiovagal function can be evaluated using different indices in the time and frequency domains.8 There is no compelling evidence that one method is better than another or that the use of multiple indices confers any advantage. Heart rate variability to deep breathing is the most widely used test of cardiovagal function and has a specificity of approximately 80% (Class II).9 The vagal component of the baroreflex can be evaluated by quantitating the heart period response to induced changes in BP. A well-studied test is the modified Oxford method.10 The test consists of an evaluation of heart period responses to induced increases and decreases in arterial BP. The increase is evoked by IV phenylephrine and decrease by nitroprusside in incremental doses. Baroreflex sensitivity is defined by the slope of the heart period to BP relationship. Linearity is required (R ⬎ 0.85). The advantage of this test is that it evaluates vagal baroreflex sensitivity; however, the disadvantage is that the test is invasive and not widely performed. Approximation of this method is possible by relating heart period alterations to changes in BP induced by the Valsalva maneuver.11 The sensitivity and specificity of invasive and noninvasive tests of baroreflex function are high, but these tests are not generally used in the study of neuropathy since their value is considered only additive to current tests of cardiovagal function (Class II).4,9,12,13 Thermoregulatory sweat testing (TST) is a sensitive test of sudomotor function that utilizes an indicator substance whose color changes upon exposure to sweat.14,15 The test results can be semiquantitated by estimating the percentage of skin surface that is anhidrotic. Since the test is tedious, messy, and timeconsuming, it is not routinely done. Additionally, TST is not able to distinguish between postganglionic, preganglionic, and central lesions.14,15 The most quantitative test of sudomotor function is the Neurology 72
January 13, 2009
179
quantitative sudomotor axon reflex test (QSART).16 QSART is mediated by impulses traveling antidromically then orthodromically along the postganglionic sympathetic sudomotor axon. QSART can detect distal sudomotor loss with a sensitivity of 75–90% (Class III).7,17-19 Several studies have demonstrated that QSART can determine sudomotor abnormalities with relatively high sensitivity and specificity in many types of polyneuropathies (Class II and III). 4-7,16,17,19-21 In three Class III studies, QSART was shown capable of detecting distal small fiber polyneuropathy with a sensitivity of ⬎75%.17-19 Skin vasomotor reflexes assessed by monitoring skin blood flow using laser Doppler flowmeter have not been well studied. Limited data from one Class III study using this technique demonstrated an unacceptably large coefficient of variation.22 Analysis of the available Class II and III studies on autonomic testing indicates that a combination of autonomic reflex screening tests provides distinct advantages over single modality methods (table). The composite autonomic scoring scale (CASS), which includes QSART, orthostatic blood pressure, heart rate response to tilt, heart rate response to deep breathing, the Valsalva ratio, and beat-to-beat blood pressure measurements during phases II and IV of the Valsalva maneuver, tilt, and deep breathing, provides a useful 10-point scale of autonomic function (Class II).4,9 In a study of 78 patients with graded autonomic failure obtained by selecting approximately equal numbers of patients with multiple system atrophy, Parkinson disease, autonomic neuropathies, and idiopathic peripheral neuropathies, this combination of tests provided a noninvasive, sensitive, specific, and reproducible methodology for grading the degree of autonomic dysfunction (Class II).9 Conclusions. Autonomic testing is probably useful in documenting autonomic nervous system involvement in polyneuropathy (Class II and III). The sensitivity and specificity vary with the particular test. The utilization of the combination of autonomic reflex screening tests in the CASS probably provides the highest sensitivity and specificity for documenting autonomic dysfunction (Class II). Recommendations. Autonomic testing should be considered in the evaluation of patients with polyneuropathy to document autonomic nervous system involvement (Level B). Autonomic testing should be considered in the evaluation of patients with suspected autonomic neuropathies (Level B) and may be considered in the evaluation of patients with suspected distal SFSN (Level C). The combination of autonomic screening tests in the CASS should be 180
Neurology 72
January 13, 2009
considered to achieve the highest diagnostic accuracy (Level B). Role of nerve biopsy in the evaluation of polyneuropathy. Nerve biopsy is generally accepted as useful in
the diagnosis of inflammatory diseases of nerve such as vasculitis, sarcoidosis, CIDP, infectious diseases such as leprosy, or infiltrative disorders such as tumor or amyloidosis.3 Nerve biopsy is most valuable in mononeuropathy multiplex or suspected vasculitic neuropathy. There are no studies regarding the role of nerve biopsy in the evaluation of DSP although on occasion the above noted diseases may present in that fashion. What is the usefulness of nerve biopsy in determining the
Out of 50 articles judged to be relevant, no article attained a grade greater than Class IV. Most of the articles discussed the nerve biopsy findings in specific diseases, the clinical suspicion of which had prompted the biopsy.23-34 No article provided guidance regarding when to perform a nerve biopsy in the evaluation of DSP. Conclusions. There is no evidence to support or refute a conclusion regarding the role of nerve biopsy in the evaluation of DSP (Class IV). Recommendations. No recommendations can be made regarding the role of nerve biopsy in determining the etiology of DSP (Level U). etiology of distal symmetric polyneuropathy?
Role of skin biopsy in the evaluation of polyneuropathy. Skin biopsy is being increasingly used to evalu-
ate patients with polyneuropathy. The most common technique involves a 3 mm punch biopsy of skin from the leg. After sectioning by microtome, the tissue is immunostained with anti-protein-geneproduct 9.5 (PGP 9.5) antibodies and examined with immunohistochemical or immunofluorescent methods. This staining allows for the identification and counting of intraepidermal nerve fibers (IENF). PGP 9.5 immunohistochemistry has been validated as a reliable method for IENF density determination with good intra- and interobserver reliability in normal controls and patients with DSP. 35-38 In March 2005, the European Federation of Neurological Societies (EFNS) published a guideline on the use of skin biopsy in peripheral neuropathy.35 This comprehensive review focused on the technical aspects of skin biopsy as well as normative data and correlations with other clinical, physiologic, and pathologic tools. The EFNS concluded that skin biopsy is a safe, validated, and reliable technique for the determination of IENF density. The major conclusion was that skin biopsy (IENF density) was diagnostically efficient at distinguishing polyneuropathy patients (including small fiber neuropathy) from normal controls. The EFNS guideline also reviewed
the literature on IENF morphologic changes such as axonal swellings as a measure of distal symmetric polyneuropathy.35,39,40 The EFNS concluded that axonal swellings may be predictive of progression of polyneuropathy but further studies were needed to determine their diagnostic accuracy.35 What is the usefulness and diagnostic accuracy of skin
Beyond distinguishing asymptomatic normal controls from polyneuropathy patients, one clinical question not addressed by the EFNS guideline was the diagnostic accuracy of skin biopsy in distinguishing symptomatic patients with polyneuropathy from symptomatic patients without polyneuropathy. For example, in patients with painful feet, would skin biopsy accurately distinguish patients with polyneuropathy from patients with other conditions causing painful feet? To address this separate question, a subgroup of the Polyneuropathy Task Force (J.D.E., R.A.L., D.H., G.L., M.P., and G.S.G.) independently reviewed the literature regarding the diagnostic accuracy of skin biopsy in DSP and in the SFSN form of DSP. To be considered for review, studies needed to determine IENF density in patients with and without polyneuropathy. Furthermore, the data from studies had to be presented in such a way as to allow calculation of the sensitivity and specificity of skin biopsy for polyneuropathy. Nine studies met inclusion criteria.36,39-40,e1-e6 One was a prospective cohort survey of patients presenting with bilateral painful feet and normal strength, but skin biopsy was done only in those with normal NCS.e1 Patients with reduced IENF density and normal NCS were assumed to have painful small fiber neuropathies. However, the study did not compare the results of the IENF density to an independent reference standard to confirm the presence of small fiber neuropathy. Thus, for the purposes of determining the diagnostic accuracy of skin biopsy for polyneuropathy, this study was graded Class IV. The remaining studies employed a case-control design.36,39,40,e2-e6 In these studies, the investigators performed skin biopsies on patients with established polyneuropathy and normal controls. No study included patients with conditions causing lower extremity pain or sensory complaints that might be confused with polyneuropathy. Thus, all studies had potential spectrum bias. Following the evidence classification scheme for studies of diagnostic accuracy, all of these studies were graded Class III. All of the case control studies showed a significant reduction in IENF density in polyneuropathy patients as compared to controls.36,39,40,e2-e6 The sensitivity of decreased IENF density for the diagnosis of polyneuropathy was moderate to good (range 45 to biopsy in the evaluation of polyneuropathy?
90%). The specificity of normal IENF density for the absence of polyneuropathy was very good (range 95 to 97%). Thus, the absence of reduced IENF density (using the clinical impression as the diagnostic reference standard) would not “rule out” polyneuropathy, but the presence of reduced IENF density would importantly raise the likelihood of polyneuropathy. The form of DSP for which IENF assessment is particularly diagnostically attractive is SFSN for the following reasons: 1) IENF are the nerve terminals of somatic unmyelinated C fibers, which are hypothesized to be predominantly affected in SFSN. 2) There has been a lack of a direct objective measure of small fiber sensory nerves since objective measures of large fiber function (e.g., NCS) are by most definitions normal in SFSN.e7 3) Patients in whom SFSN is clinically suspected manifest with symptoms of small fiber sensory dysfunction (e.g., tingling, numbness, and neuropathic pain) but few objective signs, making it difficult to diagnose and to distinguish SFSN from non-neurologic causes of sensory complaints.e7 Since no validated objective gold standard exists for the diagnosis of SFSN, the authors considered whether demonstration of a pathologic lesion (small sensory fiber pathology on skin biopsy) should be the de facto diagnostic standard or whether a clinical impression of SFSN should be the independent reference standard. For the purposes of this parameter, a clinical impression of SFSN was adopted as the independent reference standard for calculation of sensitivity and specificity of IENF density in the detection of SFSN. In order to assess the diagnostic accuracy of IENF density assessment for SFSN, the literature was surveyed for studies assessing IENF density in subjects with clinically suspected SFSN (symptoms or symptoms and signs of DSP but with normal NCS) and controls where the diagnostic accuracy of IENF density for clinically defined SFSN could be determined. Four Class III studies met these criteria.e6,e8-e10 The sensitivity of IENF density assessment at the ankle for DSP with normal NCS was 58% (20% for subjects with symptoms but no signs of SFSN; 100% for subjects with symptoms and signs of SFSN),e8 90%,e6 and 24%.e9 In these studies, the specificity of the test ranged from 95% to 97.5%.e6,e8,e9 The other case control study found that among patients with symptoms of SFSN and an abnormal pinprick examination in the feet, but normal ankle reflexes, normal vibration sensibility, and normal NCS, an IENF density of ⬍8 fibers/mm at the dorsal foot provided a sensitivity of 88%, a specificity of 91%, a positive predictive value of 0.9, and a negative predictive value of 0.83 for the diagnosis of SFSN.e10 Conclusions. IENF density assessment using PGP 9.5 immunohistochemistry is a validated, reproducNeurology 72
January 13, 2009
181
ible marker of small fiber sensory pathology. Skin biopsy with IENF density assessment is possibly useful to identify DSP which includes SFSN in symptomatic patients with suspected polyneuropathy (Class III). Recommendations. For symptomatic patients with suspected polyneuropathy, skin biopsy may be considered to diagnose the presence of a polyneuropathy, particularly SFSN. (Level C) RECOMMENDATIONS FOR FUTURE RESEARCH
This comprehensive review reveals several weaknesses in the current approach to the evaluation of polyneuropathy and highlights opportunities for research. • Autonomic testing. Autonomic testing can with a high degree of accuracy document autonomic system dysfunction in polyneuropathy. This is particularly relevant to small fiber polyneuropathy and the autonomic neuropathies. Research is necessary to determine whether the documentation of autonomic abnormalities is important in modifying the evaluation and treatment of polyneuropathy. Specific tests such as QSART can document small fiber (i.e., sudomotor axon) loss with a high degree of sensitivity, making the test useful to confirm the diagnosis of small fiber polyneuropathy. Since skin biopsy with determination of IENF density can also document small fiber loss, there is a need for studies that compare and correlate the two techniques. • Nerve biopsy. There are no studies of nerve biopsy in the evaluation of DSP. Although it would be useful to know the outcome of welldesigned prospective studies in this area, it is unlikely that such studies will be done. • Skin biopsy. Skin biopsy with determination of IENF density is a technique that has come of age for the objective documentation of small fiber loss. This technique provides a unique opportunity for research in different varieties of neuropathy. Further studies are needed to characterize the diagnostic accuracy of skin biopsy in distinguishing patients with suspected polyneuropathy, particularly SFSN, from patients with sensory complaints or pain unrelated to peripheral neuropathy. Prospective studies with appropriate “other disease” controls should be done to assess the sensitivity, specificity, and predictive values of IENF density measurement to identify SFSN in patients with lower extremity pain or sensory complaints. A predetermined independent reference standard for the diagnosis of SFSN should be specifically stated in such studies. 182
Neurology 72
January 13, 2009
• A case definition of SFSN should be developed. Investigators need to determine whether this case definition should be based upon clinical criteria, pathologic criteria (e.g., skin biopsy), or a combination of clinical, paraclinical, and pathologic criteria. • The diagnostic accuracy of morphologic changes (e.g., axonal swellings) in the diagnosis of SFSN vs healthy controls and disease controls needs to be better defined. • Studies exploring other uses for skin biopsy beyond identification and quantification of DSP and SFSN have been reported and should be further explored. Biopsies of glabrous skin and dermal skin include myelinated nerve fibers, and have been shown to have potential utility in the diagnosis of immune-mediated neuropathies, Charcot-Marie-Tooth, and related diseases.e2 Other studies have employed skin biopsy for detection or monitoring of leprosy, hereditary amyloidosis, vasculitic neuropathy, and Fabry disease.e11-e14 Additional studies are required to determine the usefulness of skin biopsy in the diagnosis and monitoring of these and other varieties of neuropathy. • Serial IENF density measurements and IENF regenerative capacity are being studied and used as outcome measures in therapeutic trials.e15,e16 Further studies are needed to validate and determine the value of skin biopsy for this purpose. AUTHORS’ AFFILIATIONS From the Louisiana State University Health Sciences Center (J.D.E., A.J.S.), New Orleans; University of Kansas (G.S.G.), Kansas City; University of Washington (G.F.), Seattle; Providence Health System (G.T.C.), Southwest Washington; St. Louis University School of Medicine (L.J.K.), St. Louis, MO; Dartmouth Hitchcock Medical Center (J.A.C.), Lebanon, NH; University of Pennsylvania School of Medicine (A.K.A.), Philadelphia; Baylor College of Medicine (K.S., J.R.L.), Houston, TX; Weill Medical College of Cornell (N.L.), New York, NY; Wayne State University School of Medicine (R.A.L.), Detroit, MI; Mayo Clinic (P.A.L.), Rochester, MN; Loyola University Chicago Stritch School of Medicine and the Hines VAH (M.A.F.), IL; University of Rochester Medical Center (D.H.), NY; University of North Carolina (J.F.H.), Chapel Hill; Fondazione IRCCS National Neurological Institute “Carlo Besta” (G.L.), Milan, Italy; California Pacific Medical Center (R.G.M.), San Francisco; and Johns Hopkins Medical Institutions (M.P.), Baltimore, MD.
DISCLOSURE J.D.E. holds financial interests in Pfizer and has received research support from Wyeth and Pfizer. G.S.G. has received speaker honoraria from Pfizer, GlaxoSmithKline, and Boehringer Ingelheim and served on the IDMC Committee of Ortho-McNeil. He estimates that ⬍2% of his clinical effort is spent on EMG and EEG. G.F., A.K.A., and K.S. have nothing to disclose. G.T.C estimates that 30% of his clinical effort is spent on EMG. J.A.C. has received speaker honoraria from Athena Diagnostics and estimates that 40% of his clinical effort is spent on EMG/NCS, 10% on autonomic testing, and 10% on botulinum toxin injections. L.J.K. has received speaker honoraria from American Medical Seminars, Cross Country Education, Therapath Laboratories, and CME, LLC, and holds equity in Passnet Air Ambulance. He estimates 25% of his clinical effort is spent on NCS/EMG, 4% on skin biopsy for nerve fiber counting, and 8%
on autonomic studies, and has received payment for expert testimony in legal proceedings. J.R.L. holds financial interests in Athena Diagnostics and has received research funding from NIH/NEI, NIH/NIDCR, Charcot-Marie-Tooth Association, and the March of Dimes. N.L. serves as a consultant for Talecris Biopharmaceuticals and Quest Diagnostics, receives royalties from Athena Diagnostics, and holds equity and is a partner in Therapath LLC. He is the Medical and Scientific Director for the Neuropathy Association, estimates that ⬍1% of his clinical effort is spent on skin biopsy, and has received research support from Talecris Biotherapeutics. R.A.L. has consulted for Talecris and has received research funding from MDA, Baxter Pharmaceuticals, and CMTA. He estimates that 33% of his clinical effort is spent on electromyography. He has received payment for expert testimony regarding the use of IVIg in CIDP and neuropathic pain after breast reduction. P.A.L. estimates 25% of his clinical effort is spent on autonomic reflex screening. D.H. has received research funding from NIH, Astellas Pharmaceutical Company, and MDA/ CMT Association. He estimates that 25% of his clinical effort is spent on EMG and 20% on skin biopsies. J.F.H. holds financial interests in FEMI, Johnson & Johnson, Pfizer, and General Electric. He estimates that 40% of his clinical effort is spent on EMG/NCS. G.L. holds financial interests in GlaxoSmithKline and Formenti-Grunenthal. In addition, he has received research funding from Pfizer, Formenti-Grunenthal, Italian Ministry of Health, and Regione Lombardia. He estimates that 25% of his clinical effort is spent in an outpatient pain center, 25% on out- and inpatient clinical examination, 25% on skin biopsy examination, and 25% on research. R.G.M. holds financial interests in Celgene, Knopp Neurosciences, Medivation, Teva Neuro, Taiji Biomedicals, and Translational Genomics. M.P. serves on the scientific advisory board of GSK, the editorial board of Journal of the Peripheral Nervous System, the speakers’ bureau of Pfizer and participated in the Joslin diabetes CME programs. He has received research funding from Astellas Pharma and Mitsubishi Pharma and reads clinical skin biopsies, runs an EMG lab, and cares for patients with peripheral nerve diseases. A.J.S. has received payment for expert testimony in the possible neurotoxic injury of the peripheral nerve.
DISCLAIMER The diagnosis and evaluation of polyneuropathy is complex. The practice parameter is not intended to replace the clinical judgment of experienced physicians in the evaluation of polyneuropathy. The particular kinds of tests utilized by a physician in the evaluation of polyneuropathy depend upon the specific clinical situation and the informed medical judgment of the treating physician. This statement is provided as an educational service of the AAN, AANEM, and AAPM&R. It is based upon an assessment of current scientific and clinical information. It is not intended to include all possible proper methods of care for a particular neurologic problem or all legitimate criteria for choosing to use a specific test or procedure. Neither is it intended to exclude any reasonable alternative methodologies. The AAN, AANEM, and AAPM&R recognize that specific care decisions are the prerogative of the patient and physician caring for the patient, based on all of the circumstances involved.
APPENDIX 1 Quality Standards Subcommittee (AAN): Jacqueline French, MD, FAAN (chair); Charles E. Argoff, MD; Eric Ashman, MD; Stephen Ashwal, MD, FAAN (ex-officio); Christopher Bever, Jr., MD, MBA, FAAN; John D. England, MD, FAAN (QSS facilitator); Gary M. Franklin, MD, MPH, FAAN (ex-officio); Deborah Hirtz, MD (ex-officio); Robert G. Holloway, MD, MPH, FAAN; Donald J. Iverson, MD, FAAN; Steven R. Messe´, MD; Leslie A. Morrison, MD; Pushpa Narayanaswami, MD, MBBS; James C. Stevens, MD, FAAN (ex-officio); David J. Thurman, MD, MPH (ex-officio); Dean M. Wingerchuk, MD, MSc, FRCP(C); and Theresa A. Zesiewicz, MD, FAAN.
APPENDIX 2 Practice Issues Review Panel (AANEM): Yuen T. So, MD, PhD (chair); Michael T. Andary, MD; Atul Patel, MD; Carmel Armon, MD; David del Toro, MD; Earl J. Craig, MD; James F. Howard, Jr, MD; Joseph V. Campellone Jr., MD; Kenneth James Gaines, MD; Robert Werner, MD; and Richard Dubinsky, MD.
APPENDIX 3 Practice Guidelines Committee (AAPM&R): Dexanne B. Clohan, MD (chair); William L. Bockenek, MD; Lynn Gerber, MD; Edwin Hanada, MD; Ariz R. Mehta, MD; Frank J. Salvi, MD, MS; and Richard D. Zorowitz, MD.
Received April 24, 2008. Accepted in final form August 29, 2008.
REFERENCES 1. England JD, Gronseth GS, Franklin G, et al. Distal symmetric polyneuropathy: a definition for clinical research: report of the American Academy of Neurology, the American Association of Electrodiagnostic Medicine, and the American Academy of Physical Medicine and Rehabilitation. Neurology 2005;64:199–207. 2. Martyn CN, Hughes RAC. Epidemiology of peripheral neuropathy. J Neurol Neurosurg Psychiatry 1997;62:310–318. 3. England JD, Asbury AK. Peripheral neuropathy. Lancet 2004;363:2151–2161. 4. Low PA, Vernino S, Suarez G. Autonomic dysfunction in peripheral nerve disease. Muscle Nerve 2003;27:646–661. 5. Singer W, Spies JM, McArthur J, et al. Prospective evaluation of somatic and autonomic small fibers in selected neuropathies. Neurology 2004;62:612–618. (Class III) 6. Dyck PJ, Karnes JL, O’Brien PC, Litchy WJ, Low PA, Melton LJ. The Rochester Diabetic Neuropathy Study: reassessment of tests and criteria for diagnosis and staged severity. Neurology 1992;42:1164–1170. (Class II) 7. Low PA, Denq JC, Opfer-Gehrking TL, Dyck PJ, O’Brien PC, Slezak JM. Effect of age and gender on sudomotor and cardiovagal function and blood pressure response to tilt in normal subjects. Muscle Nerve 1997;20:1561–1568. (Class II) 8. Ziegler D, Dannehl K, Muhlen, Spuler M, Gries FA. Presence of cardiovascular autonomic dysfunction assessed by spectral analysis, and standard tests of heart rate variation and blood pressure response at various stages of diabetic neuropathy. Diabet Med 1992;9:806–814. 9. Low PA. Composite Autonomic Scoring Scale for laboratory quantification of generalized autonomic failure. Mayo Clin Proc 1993;68:748–752. (Class II) 10. Ebert TJ, Morgan BJ, Barney JA, Denahan T, Smith JJ. Effects of aging on baroreflex regulation of sympathetic activity in humans. Am J Physiol 1992;263:798–803. 11. Trimarco B, Volpe M, Ricciardelli B, et al. Valsalva maneuver in the assessment of baroreflex responsiveness in borderline hypertensives. Cardiology 1983;70:6–14. 12. Vogel ER, Sandroni P, Low PA. Blood pressure recovery from Valsalva maneuver in patients with autonomic failure. Neurology 2005;65:1533–1537. (Class II) 13. Schrezenmaier C, Singer W, Muenter-Swift N, Sletten D, Tanabe J, Low PA. Adrenergic and vagal baroreflex sensitivity in autonomic failure. Arch Neurol 2007;64:381– 386. (Class II) 14. Low PA, Walsh JC, Huang C, McLeod JG. The sympathetic nervous system in diabetic neuropathy: a clinical and pathological study. Brain 1975;98:341–356. 15. Fealey RD, Low PA, Thomas JE. Thermoregulatory sweating abnormalities in diabetes mellitus. Mayo Clin Proc 1989;64:617–628. 16. Low PA, Caskey PE, Tuck RR, Fealey RD, Dyck PJ. Quantitative sudomotor axon reflex test in normal and neuropathic subjects. Ann Neurol 1983;14:573–580. Neurology 72
January 13, 2009
183
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
184
Stewart AG, Low PA, Fealey RD. Distal small fiber neuropathy: results of tests of sweating and autonomic cardiovascular reflexes. Muscle Nerve 1992;15:661–665. (Class III) Tobin K, Guliani MJ, LaComis D. Comparison of different modalities for detection of small fiber neuropathy. Clin Neurophysiol 1999;110:1909–1912. (Class III) Novak V, Freimer ML, Kissel JT, et al. Autonomic impairment in painful neuropathy. Neurology 2001;56:861– 868. (Class III) Low PA, Zimmerman BR, Dyck PJ. Comparison of distal sympathetic with vagal function in diabetic neuropathy. Muscle Nerve 1986;9:592–596. Low PA, Opfer-Gehrking TL, Proper CJ, Zimmerman I. The effect of aging on cardiac autonomic and postganglionic sudomotor function. Muscle Nerve 1990;13:152–157. Low PA, Neumann C, Dyck PJ, Fealey RD, Tuck RR. Evaluation of skin vasomotor reflexes by using laser Doppler velocimetry. Mayo Clin Proc 1983;58:583–592. (Class III) Argov Z, Steiner I, Soffer D. The yield of sural nerve biopsy in the evaluation of peripheral neuropathies. Acta Neurol Scand 1989;79:243–245. Bosboom WMJ, VanderBerg LH, Franssen H, et al. Diagnostic value of sural nerve demyelination in chronic inflammatory demyelinating polyneuropathy. Brain 2001; 124:2427–2438. Chia L, Fernandez A, Lacroix C, Adams D, Plante´ V, Said G. Contribution of nerve biopsy findings to the diagnosis of disabling neuropathy in the elderly: a retrospective review of 100 consecutive patients. Brain 1996;119:1091–1098. Deprez M, Ceuterick-DeGroote C. Clinical and neuropathological parameters affecting the diagnostic yield of nerve biopsy. Neuromuscul Disord 2000;10:92–98. Deprez M, Ceuterick-DeGroote C, Schoenen J, Reznik M, Martin JJ. Nerve biopsy: indications and contribution of the diagnosis of peripheral neuropathy. Acta Neurol Belg 2000;100:162–166. Flachenecker P, Janka M, Goldbrunner R, Toyka KV. Clinical outcome of sural nerve biopsy: a retrospective study. J Neurol 1999;246:93–96.
Neurology 72
January 13, 2009
29.
30.
31.
32. 33. 34. 35.
36.
37.
38.
39.
40.
Gabriel CM, Howard R, Kinsella N, et al. Prospective study of the usefulness of sural nerve biopsy. J Neurol Neurosurg Psychiatry 2000;69:442–446. Molenaar DSM, Vermeulen M, de Haan R. Diagnostic value of sural nerve biopsy in chronic inflammatory demyelinating polyneuropathy. J Neurol Neurosurg Psychiatry 1998;64:84–89. Rappaport WD, Valente J, Hunter GC, et al. Clinical utilization and complications of sural nerve biopsy. Am J Surg 1993;166:252–256. Said G. Indications and value of nerve biopsy. Muscle Nerve 1999;22:1617–1619. Said G. Value of nerve biopsy. Lancet 2001;357:1220– 1221. Said G. Indications and usefulness of nerve biopsy. Arch Neurol 2002;59:1532–1535. Lauria G, Cornblath DR, Johansson O, et al. EFNS Guidelines on the use of skin biopsy in the diagnosis of peripheral neuropathy. Eur J Neurol 2005;12:1–12. McArthur JC, Stocks EA, Hauer P, Cornblath DR, Griffin JW. Epidermal nerve fiber density: normative reference range and diagnostic efficiency. Arch Neurol 1998;55: 1513–1520. (Class III) Goranson LG, Mellgren SI, Lindal S, Omdal R. The effect of age and gender on epidermal nerve fiber density. Neurology 2004;62:774–777. Smith AG, Howard JR, Kroll R, et al. The reliability of skin biopsy with measurement of intraepidermal nerve fiber density. J Neurol Sci 2005;228:65–69. Lauria G, Morbin M, Lombardi R, et al. Axonal swellings predict the degeneration of epidermal nerve fibers in painful neuropathies. Neurology 2003;61:631–636. (Class III) Herrmann DN, McDermott MP, Henderson D, Chen L, Akowuah K, Schifitto G, and The North East AIDS Dementia (NEAD) Consortium. Epidermal nerve fiber density, axonal swellings and QST as predictors of HIV distal sensory neuropathy. Muscle Nerve 2004;29:420–427. (Class III)
SPECIAL ARTICLE
Practice Parameter: Evaluation of distal symmetric polyneuropathy: Role of laboratory and genetic testing (an evidence-based review) Report of the American Academy of Neurology, American Association of Neuromuscular and Electrodiagnostic Medicine, and American Academy of Physical Medicine and Rehabilitation
J.D. England, MD G.S. Gronseth, MD, FAAN G. Franklin, MD G.T. Carter, MD L.J. Kinsella, MD J.A. Cohen, MD A.K. Asbury, MD K. Szigeti, MD, PhD J.R. Lupski, MD, PhD N. Latov, MD R.A. Lewis, MD P.A. Low, MD M.A. Fisher, MD D.N. Herrmann, MD J.F. Howard, Jr., MD G. Lauria, MD R.G. Miller, MD M. Polydefkis, MD, MHS A.J. Sumner, MD
Address correspondence and reprint requests to the American Academy of Neurology, 1080 Montreal Avenue, St. Paul, MN 55116
[email protected]
ABSTRACT
Background: Distal symmetric polyneuropathy (DSP) is the most common variety of neuropathy. Since the evaluation of this disorder is not standardized, the available literature was reviewed to provide evidence-based guidelines regarding the role of laboratory and genetic tests for the assessment of DSP.
Methods: A literature review using MEDLINE, EMBASE, and Current Contents was performed to identify the best evidence regarding the evaluation of polyneuropathy published between 1980 and March 2007. Articles were classified according to a four-tiered level of evidence scheme and recommendations were based upon the level of evidence.
Results and Recommendations: 1) Screening laboratory tests may be considered for all patients with polyneuropathy (Level C). Those tests that provide the highest yield of abnormality are blood glucose, serum B12 with metabolites (methylmalonic acid with or without homocysteine), and serum protein immunofixation electrophoresis (Level C). If there is no definite evidence of diabetes mellitus by routine testing of blood glucose, testing for impaired glucose tolerance may be considered in distal symmetric sensory polyneuropathy (Level C). 2) Genetic testing should be conducted for the accurate diagnosis and classification of hereditary neuropathies (Level A). Genetic testing may be considered in patients with cryptogenic polyneuropathy who exhibit a hereditary neuropathy phenotype (Level C). Initial genetic testing should be guided by the clinical phenotype, inheritance pattern, and electrodiagnostic features and should focus on the most common abnormalities which are CMT1A duplication/HNPP deletion, Cx32 (GJB1), and MFN2 mutation screening. There is insufficient evidence to determine the usefulness of routine genetic testing in patients with cryptogenic polyneuropathy who do not exhibit a hereditary neuropathy phenotype (Level U). Neurology® 2009;72:185–192 GLOSSARY AAN ⫽ American Academy of Neurology; AANEM ⫽ American Academy of Neuromuscular and Electrodiagnostic Medicine; AAPM&R ⫽ American Academy of Physical Medicine and Rehabilitation; CMT ⫽ Charcot-Marie-Tooth; DSP ⫽ distal symmetric polyneuropathy; EDX ⫽ electrodiagnostic; GTT ⫽ glucose tolerance testing; IFE ⫽ immunofixation electrophoresis; QSS ⫽ Quality Standards Subcommittee; SPEP ⫽ serum protein electrophoresis.
Polyneuropathy is a relatively common neurologic disorder.1 The overall prevalence is approximately 2,400 (2.4%) per 100,000 population, but in individuals older than 55 years, the prevalence rises to approximately 8,000 (8%) per 100,000.2,3 Since there are many etiologies of polyneuropathy, a logical
clinical approach is needed for evaluation and management. This practice parameter provides recommendations for the role of laboratory and genetic tests in the evaluation of distal symmetric polyneuropathy (DSP) based upon a prescribed review and analysis of
Supplemental data at www.neurology.org See page 177 e-Pub ahead of print on December 3, 2008, at www.neurology.org. Published simultaneously in PM&R and Muscle & Nerve. Authors’ affiliations are listed at the end of the article. The AAN Mission Statement, Classification of evidence, Classification of recommendations, and Conflict of Interest Statement (appendices e-1 through e-4), as well as tables e-1, e-2, and e-3, are available as supplemental data on the Neurology威 Web site at www.neurology.org. Approved by the AAN Quality Standards Subcommittee on November 10, 2007; by the AAN Practice Committee on January 20, 2008; by the Neuromuscular Guidelines Steering Committee on April 22, 2008; by the AAN Board of Directors on August 20, 2008; by the AANEM Board of Directors on May 1, 2008; and by the AAPM&R Board of Governors on April 7, 2008. Disclosure: Author disclosures are provided at the end of the article. Copyright © 2009 by AAN Enterprises, Inc.
185
the peer-reviewed literature. The parameter was developed to provide physicians with evidence-based guidelines regarding the role of laboratory and genetic tests for the assessment of polyneuropathy. The diagnosis of DSP should be based upon a combination of clinical symptoms, signs, and electrodiagnostic criteria as outlined in the previous case definition.1 See Mission statement (appendix e-1 on the Neurology® Web site at www.neurology.org) for details. The Polyneuropathy Task Force included 19 physicians with representatives from the American Academy of Neurology (AAN), the American Academy of Neuromuscular and Electrodiagnostic Medicine (AANEM), and the American Academy of Physical Medicine and Rehabilitation (AAPM&R). All of the task force members had extensive experience and expertise in the area of polyneuropathy. Additionally, four members had expertise in evidence-based methodology and practice parameter development. Two are current members (J.D.E., G.F.), and two are former members (G.S.G., R.G.M.) of the Quality Standards Subcommittee (QSS) of the AAN. The task force developed a set of clinical questions relevant to the evaluation of DSP, and subcommittees were formed to address each of these questions.
FORMATION OF EXPERT PANEL
DESCRIPTION OF THE ANALYTIC PROCESS
The literature search included OVID MEDLINE (1966 to March 2007), OVID Excerpta Medica (EMBASE; 1980 to March 2007), and OVID Current Contents (2000 to March 2007). The search included articles on humans only and in all languages. The search terms selected were peripheral neuropathy, polyneuropathy, and distal symmetric polyneuropathy. These terms were cross-referenced with the terms laboratory test, diagnosis, electrophysiology, and genetic testing. Panel experts were asked to identify additional articles missed by the initial search strategy. Further, the bibliographies of the selected articles were reviewed for potentially relevant articles. Subgroups of committee members reviewed the titles and abstracts of citations identified from the original searches and selected those that were potentially relevant to the evaluation of polyneuropathy. Articles deemed potentially relevant by any panel member were also obtained. Each potentially relevant article was subsequently reviewed in entirety by at least three panel members. Each reviewer graded the risk of bias in each article by using the diagnostic test classification-of-evidence scheme (appendix e-2). In this scheme, articles attaining a grade of Class I are judged to have the lowest risk of bias, and articles attaining a grade of Class IV are judged to have the highest risk of bias. Dis186
Neurology 72
January 13, 2009
agreements among reviewers regarding an article’s grade were resolved through discussion. Final approval was determined by the entire panel. The AAN’s method for determining the strength of recommendation was used (appendix e-3). The QSS (AAN; appendix 1), the Practice Issues Review Panel (AANEM; appendix 2), and the Practice Guidelines Committee (AAPM&R; appendix 3) reviewed and approved a draft of the article. The draft was next sent to members of the AAN, AANEM, and AAPM&R for further review and then to Neurology® for peer review. Boards of the AAN, AANEM, and AAPM&R reviewed and approved the final version of the article. At each step of the review process, external reviewers’ suggestions were explicitly considered. When appropriate, the expert panel made changes to the document. ANALYSIS OF EVIDENCE The search yielded 4,500 references with abstracts. After reviewing titles and abstracts, 450 articles were reviewed and classified.
Role of laboratory testing in the evaluation of polyneuropathy. With the exception of electrodiagnostic (EDX)
studies, laboratory tests are not utilized to diagnose polyneuropathy; however, laboratory tests are routinely utilized in patients with a diagnosis of polyneuropathy as a screening test for specific etiologies. Several questions regarding the use of laboratory testing as a screening tool in the evaluation of polyneuropathy were assessed. What is the yield of screening laboratory tests in the evaluation of DSP, and which tests should be performed?
The cause of most polyneuropathies is evident when the information obtained from the medical history, neurologic examination, and EDX studies are combined with simple screening laboratory tests. Such a comprehensive investigation yields an etiologic diagnosis in 74 to 82% of patients with polyneuropathy.4-13 Laboratory test results must be interpreted in the context of other clinical information since the etiologic yield of laboratory testing alone is limited by the low specificity of many of the tests. For example, one study of idiopathic polyneuropathy found that laboratory tests alone had only a 37% diagnostic yield (Class III).6 In another study, laboratory abnormalities were documented in 58% of 91 patients with chronic cryptogenic polyneuropathy, but only 9% were etiologically diagnostic (Class III).10 The majority of studies indicated that screening laboratory tests comprised of a complete blood count, erythrocyte sedimentation rate, comprehensive metabolic panel (blood glucose, renal function, liver function), thyroid function tests, serum B12, and serum protein immunofixation electrophoresis are indicated for most patients with polyneuropathy.4-13 Five Class III studies indicated that the highest yield of abnormality was seen with screening for blood glucose,
serum B12, and serum protein immunofixation electrophoresis.4,6,10,13,14 The test with the highest yield was the blood glucose, consistent with the well-known fact that diabetes mellitus is the most common cause of DSP. In patients with DSP, blood glucose was elevated in approximately 11%, serum protein electrophoresis or immunofixation was abnormal in 9%, and serum B12 was low in approximately 3.6%. Two Class III studies showed that routine CSF analysis had a low diagnostic yield except in demyelinating polyneuropathies, which usually showed an increased CSF protein level.5,8 Vitamin B12 deficiency was relatively frequent in patients with polyneuropathy, and the yield was greater when the metabolites of cobalamin (methylmalonic acid and homocysteine) were tested (Class II and III).4,14-17 Serum methylmalonic acid and homocysteine were elevated in 5–10% of patients whose serum B12 levels were in the low normal range of 200 –500 pg/dL.16,17 In large series of patients with polyneuropathy, between 2.2 and 8% of patients had evidence of B12 deficiency as indicated by elevations of these metabolites. 4,14 In one Class III study involving 27 patients with polyneuropathy and B12 deficiency, 12 (44%) had B12 deficiency based upon the finding of abnormal metabolites alone.14 Thus, serum B12 assays with metabolites (methylmalonic acid and homocysteine) are useful in documenting B12 deficiency. Although both methylmalonic acid and homocysteine are sensitive for B12 deficiency, methylmalonic acid is more specific. In a large Class III study involving 434 patients with vitamin B12 deficiency, serum methylmalonic acid was elevated in 98.4% and serum homocysteine was elevated in 95.9%.17 In the same study, serum methylmalonic acid was elevated in 12.2%, but serum homocysteine was elevated in 91% of 123 patients with isolated folate deficiency.17 Homocysteine may also be elevated in pyridoxine deficiency and heterozygous homocystinemia. Both homocysteine and methylmalonic acid may be elevated in hypothyroidism, renal insufficiency, and hypovolemia. Several studies highlight the relatively high prevalence of prediabetes (impaired glucose tolerance) in patients with DSP who do not fulfill the criteria for definite diabetes mellitus (Class III).18-20 In these studies, glucose tolerance testing (GTT) was performed in patients with idiopathic DSP. Impaired glucose tolerance was documented in 25–36% of patients compared to approximately 15% of controls. Additionally, patients with painful sensory polyneuropathies were more likely to have impaired glucose tolerance than those with painless sensory polyneuropathies. Only one major study has not found an increased prevalence of impaired glucose tolerance in chronic idiopathic axonal polyneuropathy (Class III).21
Monoclonal gammopathies are more common in patients with polyneuropathy than in the normal population. IgM monoclonal gammopathies may be associated with autoantibody activity, type I or II cryoglobulinemia, macroglobulinemia, or chronic lymphocytic leukemia. IgG or IgA monoclonal gammopathies may be associated with myeloma, POEMS syndrome, primary amyloidosis, or chronic inflammatory conditions. In one Class III study of 279 consecutive patients with polyneuropathy of otherwise unknown etiology seen at a referral center, 10% had monoclonal gammopathy, a significant increase over that reported in community studies.22 Serum protein immunofixation electrophoresis (IFE) is more sensitive than serum protein electrophoresis (SPEP), especially for detecting small or nonmalignant monoclonal gammopathies. Ten of 58 (17%) monoclonal gammopathies, including 10 of 36 (30%) with IgM ⬍5 g/L, were identified by IFE but not by SPEP.23 Conclusions. Screening laboratory tests are possibly useful in determining the cause of DSP, but the yield varies depending upon the particular test (Class III). The tests with the highest yield of abnormality are blood glucose, serum B12 with metabolites (methylmalonic acid with or without homocysteine), and serum protein immunofixation electrophoresis (Class III). Patients with distal symmetric sensory polyneuropathy have a relatively high prevalence of diabetes or prediabetes (impaired glucose tolerance), which can be documented by blood glucose or GTT (Class III). Recommendations. Screening laboratory tests may be considered for all patients with DSP (Level C). Although routine screening with a panel of basic tests is often performed (table e-1), those tests with the highest yield of abnormality are blood glucose, serum B12 with metabolites (methylmalonic acid with or without homocysteine), and serum protein immunofixation electrophoresis (Level C). When routine blood glucose testing is not clearly abnormal, other tests for prediabetes (impaired glucose tolerance) such as a GTT may be considered in patients with distal symmetric sensory polyneuropathy, especially if it is accompanied by pain (Level C). Although there are no control studies (Level U) regarding when to recommend the use of other specific laboratory tests, clinical judgment correlated with the clinical picture will determine which additional laboratory investigations (table e-2) are necessary. Role of genetic testing in the evaluation of polyneuropathy. Hereditary neuropathies are an important
subtype of polyneuropathy with a prevalence of approximately 1:2,500 people. DSP is the predominant phenotype, but phenotypic heterogeneity may be present even within the same family; therefore, when Neurology 72
January 13, 2009
187
Table 1
Evidence table for genetic testing
Reference
Data collection
Setting*
Sampling
Completeness gene dependent
Masking
Class
26
Prospective
Referral center
NA
PMP22 dup
Waived
II
27
Prospective
Referral center
Consecutive
PMP22 dup
Waived
II
28
Prospective
Referral center
Consecutive
PMP22 mut, Cx32, MPZ
Waived
I
29
Prospective
Referral center
Consecutive
PMP22 dup, del, mut, Cx32, MPZ
Waived
I
30
Prospective
Referral center
Consecutive
PMP22 dup, del, Cx32
Waived
II
31
Prospective
NA
NA
PMP22 dup
Waived
III
32
Prospective
Referral center
Consecutive
PMP22 dup, del
Waived
I
33
Prospective
Referral center
Consecutive
PMP22 dup, mut, Cx32, MPZ
Waived
II
34
Prospective
Referral center
Consecutive
PMP22 dup, del, mut, Cx32, MPZ
Waived
II
35
Prospective
Referral center
Consecutive
PMP22 dup, mut, Cx32, MPZ
Waived
I
24
Prospective
Referral center
Consecutive
PMP22 dup, mut, Cx32, MPZ
Waived
I
36
Prospective
Referral center
Consecutive
PMP22 dup, mut, Cx32, MPZ
Waived
I
37
Prospective
Referral center
Selected
MFN2
Waived
II
*Referral center for test, not for patient; patients come from general neurology clinics.
genetic testing is contemplated all neuropathy phenotypes need to be considered. In the evaluation of polyneuropathy, a comprehensive family history should always be elicited. A high index of suspicion for a hereditary neuropathy phenotype is essential. Since molecular diagnostic techniques are available, guidelines for their usefulness in the evaluation of polyneuropathy are needed. The majority of genetically determined polyneuropathies are variants of Charcot-Marie-Tooth (CMT) disease, and genetic testing is available for an increasing number of these neuropathies. The clinical phenotype of CMT is extremely variable, ranging from a severe polyneuropathy with respiratory failure through the classic picture with pes cavus and “stork legs” to minimal neurologic findings.24,25 Since a substantial proportion of CMT patients have de novo mutations, a family history of neuropathy may be lacking.24-26 Additionally, different genetic mutations can cause a similar phenotype (genetic heterogeneity) and different phenotypes can result from the same genotype (phenotypic heterogeneity). How accurate is genetic testing for identifying patients with genetically determined neuropathies? The CMT phe-
notype has been linked to 36 loci and mutations have been identified in 28 different genes, several of which can be identified by commercially available genetic testing. Previous segregation studies followed by several prospective cohort studies have documented that the results of currently available genetic testing are unequivocal for diagnosis of established pathogenic mutations, providing a specificity of 100% (i.e., no false positives) and high sensitivity (Class I and II).27-36 The interpretation of novel mutations may require further characterization available in specialized centers. Data from six Class I, six Class II, and 188
Neurology 72
January 13, 2009
one Class III study indicate that genetic testing is useful for the accurate classification of hereditary polyneuropathies.24,26-37 See table 1 for details. Which patients with polyneuropathy should be screened
Genetic studies of hereditary neuropathies have tested the prevalence of various mutations in selected patients with the classic CMT phenotype with and without a family history of polyneuropathy27,31-36 (Class III evidence for screening). For these patients, the yield of genetic tests has been relatively high. Data from seven studies indicate that the demyelinating form of Charcot-Marie-Tooth (CMT1) is the most prevalent, and about 70% of these patients have a duplication of PMP22 gene (CMT1A).27,31-36 CMT1A is also the most common variety of sporadic CMT1, accounting for 76 –90% of cases.26,32 Six studies showed that when the test for CMT1A duplication is restricted to patients with clinically probable CMT1 (i.e., autosomal dominant, primary demyelinating polyneuropathy), the yield is 54 – 80% as compared to testing a cohort of patients suspected of having any variety of hereditary peripheral neuropathy where the yield is only 25–59% (average of 43%).27,29,30,32,34,36 Axonal forms of Charcot-Marie-Tooth (CMT2) are most commonly caused by MFN2 mutations, which account for approximately 33% of the cases.37 MFN2 mutations have not occurred in the CMT1 group. Data from eight studies indicate that Cx32(GJB1) mutations cause an X-linked neuropathy (CMTX), which may present with either a predominantly demyelinating or axonal phenotype and account for approximately 12% of all cases of CMT.27-30,33-36 If the pedigree is uninformative as to whether the inherifor hereditary neuropathies?
tance is autosomal dominant or X-linked (lack of father to son transmission), Cx32(GJB1) mutation is in the differential diagnosis for both predominantly demyelinating and axonal neuropathies. Data from seven studies have established average mutation frequencies of 2.5% for PMP22 point mutations, and 5% for MPZ mutations in the CMT population.27-29,34-36 CMT caused by other genes is much less frequent (figure). Given the relationships between pattern of inheritance, EDX results, and specific mutations, the efficiency of genetic testing can be improved by following a stepwise evaluation of patients with possible hereditary neuropathy. First, a clinical classification that includes EDX studies should be performed to determine whether the neuropathy is primarily demyelinating or primarily axonal in type. Since EDX studies are sometimes problematic in children, some physicians may opt to proceed directly to genetic testing of symptomatic children suspected of having CMT. Secondly, the inheritance pattern (autosomal dominant, autosomal recessive, or X-linked) should be ascertained. Based upon this information, the most appropriate genetic profile testing can then be performed. The figure indicates an evidence-based, tiered approach for the evaluation of suspected hereditary
Figure
neuropathies, and table 2 shows the relative frequency of the most common genetic abnormalities accounting for the CMT phenotype from population studies. The previous discussion applies to patients with polyneuropathy and a classic hereditary neuropathy phenotype with or without a family history. The authors were not able to find studies of the yield of genetic screening in polyneuropathy patients without a classic hereditary neuropathy phenotype. Some patients with CMT genetic mutations have minimal neurologic findings and do not have the classic CMT phenotype.24,25 Thus, some patients with cryptogenic polyneuropathies without the classic CMT phenotype may also have hereditary neuropathies. The prevalence of mutations in this population is unknown. Conclusions. Genetic testing is established as useful for the accurate diagnosis and classification of hereditary polyneuropathies (Class I). For patients with a cryptogenic polyneuropathy who exhibit a classic hereditary neuropathy phenotype, routine genetic screening may be useful for CMT1A duplication/ deletion and Cx32 mutations in the appropriate phenotype (Class III). Further genetic testing may be considered guided by the clinical question. There is
Evaluation of suspected hereditary neuropathies
Decision algorithm for use in the diagnosis of suspected hereditary polyneuropathies using family history and NCSs. *PMP22 denotes peripheral myelin protein 22; MPZ myelin protein zero; PRX periaxin; GDAP1 ganglioside-induced differentiation-associated protein 1; GJB1 gap-junction beta-1 protein (connexin 32); MFN2 mitofusin 2; EGR2 early growth response 2; LITAF lipopolysaccharide-induced tumor necrosis factor ␣; RAB7 small guanosine triphosphatase late endosomal protein; GARS glycyl-transfer RNA synthetase; NEFL neurofilament light chain; HSPB1 heat shock protein beta-1. Neurology 72
January 13, 2009
189
Table 2
Mutation frequencies for Charcot-Marie-Tooth (CMT) and related neuropathies in various populations
Population
Cohort (no. of patients), total/CMT1/HNPP
CMT1A duplication (%), total/CMT1
HNPP deletion (%), total/HNPP
American27
75/63
56/68
ND
3.9
7.2
3.3
Spanish28
52
Excluded
Excluded
3.8
19.2
9.6
0.8*
7.7*
3.8*
Belgian29
443
24.6
10.6
2.7
5.4
0.7
30
157
40.7
26.1
ND
7.6
ND
71
81
ND
ND
ND
ND
Finnish
Slovene31 European
32
PMP22 mutation (%), total/CMT1
Cx32 mutation (%), total/CMT1
MPZ mutation (%), total/CMT1
975/819/156
59.4/70.7
13.4/84
ND
ND
ND
Australian33
224
61
ND
1.3
12
3.1
Russian34
174 /108/3
33.9 /53.7
100
1.1/1.9
6.8/7.4
3.4, 5.6
Italian35
172
57.6
ND
1.2
6.9
2.3
Korean
36
57
Average
26/54
ND
1.7
5.3
5.3
43/70
11/92
2.50
12
5
The mutation frequencies are given in the total CMT cohort and in the clinical phenotypes (CMT1 and HNPP) when available. Bold ⫽ CMT1 subpopulation; italics ⫽ HNPP subpopulation. *Extrapolated total number and mutation frequencies recalculated for the total number. For the estimation of the total number, we calculated with the average frequencies for CMT1A duplication and HNPP deletion derived from the other studies.
insufficient evidence to determine the usefulness of routine genetic screening in cryptogenic polyneuropathy patients without a classic hereditary neuropathy phenotype. Recommendations. Genetic testing should be conducted for the accurate diagnosis and classification of hereditary neuropathies (Level A). Genetic testing may be considered in patients with a cryptogenic polyneuropathy and classic hereditary neuropathy phenotype (Level C). There is insufficient evidence to support or refute the usefulness of routine genetic testing in cryptogenic polyneuropathy patients without a classic hereditary phenotype (Level U). Clinical context. To achieve the highest yield, the genetic testing profile should be guided by the clinical phenotype, inheritance pattern (if available), and EDX features (demyelinating vs axonal). See the figure for guidance. RECOMMENDATIONS FOR FUTURE RESEARCH
This comprehensive review reveals several weaknesses in the current approach to the evaluation of polyneuropathy and highlights opportunities for research. • Laboratory testing. The finding of a laboratory abnormality does not necessarily mean that the abnormality is etiologically significant. For instance, there is a relatively high prevalence of impaired glucose tolerance in patients with DSP; however, whether this is etiologically diagnostic is not known. This and other such examples point to the need for more research into the basic pathobiology of the peripheral ner190
Neurology 72
January 13, 2009
vous system. As an extension of this area of research, there is a need to determine whether aggressive treatment or reversal of specific laboratory abnormalities improves or alters the course of polyneuropathy. • Genetic testing. The genetic revolution has provided great insights into the mechanisms of hereditary neuropathies. Genetically determined neuropathies are more common and clinically diverse than previously appreciated. Further research to identify genotype–phenotype correlation is needed to improve the evaluation process for patients with suspected hereditary neuropathies. The issue of cost/benefit ratio of genetic testing is important since an ever-increasing number of genetic tests are commercially available. More clearly defined guidelines for genetic testing are needed to maximize yield and to curtail the costs of such evaluations. Continued exploration into the genetic basis of neuropathies has tremendous potential for the understanding of basic pathophysiology and treatment of neuropathies. AUTHORS’ AFFILIATIONS From the Louisiana State University Health Sciences Center (J.D.E., A.J.S.), New Orleans; University of Kansas (G.S.G.), Kansas City; University of Washington (G.F.), Seattle; Providence Health System (G.T.C.), Southwest Washington; St. Louis University School of Medicine (L.J.K.), St. Louis, MO; Dartmouth Hitchcock Medical Center (J.A.C.), Lebanon, NH; University of Pennsylvania School of Medicine (A.K.A.), Philadelphia; Baylor College of Medicine (K.S., J.R.L.), Houston, TX; Weill Medical College of Cornell (N.L.), New York, NY; Wayne
State University School of Medicine (R.A.L.), Detroit, MI; Mayo Clinic (P.A.L.), Rochester, MN; Loyola University Chicago Stritch School of Medicine and the Hines VAH (M.A.F.), IL; University of Rochester Medical Center (D.H.), NY; University of North Carolina (J.F.H.), Chapel Hill; Fondazione IRCCS National Neurological Institute “Carlo Besta” (G.L.), Milan, Italy; California Pacific Medical Center (R.G.M.), San Francisco; and Johns Hopkins Medical Institutions (M.P.), Baltimore, MD.
DISCLOSURE J.D.E. holds financial interests in Pfizer and has received research support from Wyeth and Pfizer. G.S.G. has received speaker honoraria from Pfizer, GlaxoSmithKline, and Boehringer Ingelheim and served on the IDMC Committee of Ortho-McNeil. He estimates that ⬍2% of his clinical effort is spent on EMG and EEG. G.F., A.K.A., and K.S. have nothing to disclose. G.T.C estimates that 30% of his clinical effort is spent on EMG. J.A.C. has received speaker honoraria from Athena Diagnostics and estimates that 40% of his clinical effort is spent on EMG/NCS, 10% on autonomic testing, and 10% on botulinum toxin injections. L.J.K. has received speaker honoraria from American Medical Seminars, Cross Country Education, Therapath Laboratories and CME, LLC, and holds equity in Passnet Air Ambulance. He estimates 25% of his clinical effort is spent on NCS/EMG, 4% on skin biopsy for nerve fiber counting, and 8% on autonomic studies, and has received payment for expert testimony in legal proceedings. J.R.L. holds financial interests in Athena Diagnostics and has received research funding from NIH/NEI, NIH/NIDCR, Charcot-Marie-Tooth Association, and the March of Dimes. N.L. serves as a consultant for Talecris Biopharmaceuticals and Quest Diagnostics, receives royalties from Athena Diagnostics, and holds equity and is a partner in Therapath LLC. He is the Medical and Scientific Director for the Neuropathy Association, estimates that ⬍1% of his clinical effort is spent on skin biopsy, and has received research support from Talecris Biotherapeutics. R.A.L. has consulted for Talecris and has received research funding from MDA, Baxter Pharmaceuticals, and CMTA. He estimates that 33% of his clinical effort is spent on electromyography. He has received payment for expert testimony regarding the use of IVIg in CIDP and neuropathic pain after breast reduction. P.A.L. estimates 25% of his clinical effort is spent on autonomic reflex screening. D.H. has received research funding from NIH, Astellas Pharmaceutical Company, and MDA/ CMT Association. He estimates that 25% of his clinical effort is spent on EMG and 20% on skin biopsies. J.F.H. holds financial interests in FEMI, Johnson & Johnson, Pfizer, and General Electric. He estimates that 40% of his clinical effort is spent on EMG/NCS. G.L. holds financial interests in GlaxoSmithKline and Formenti-Grunenthal. In addition, he has received research funding from Pfizer, Formenti-Grunenthal, Italian Ministry of Health, and Regione Lombardia. He estimates that 25% of his clinical effort is spent in an outpatient pain center, 25% on out- and inpatient clinical examination, 25% on skin biopsy examination, and 25% on research. R.G.M. holds financial interests in Celgene, Knopp Neurosciences, Medivation, Teva Neuro, Taiji Biomedicals, and Translational Genomics. M.P. serves on the scientific advisory board of GSK, the editorial board of Journal of the Peripheral Nervous System, the speakers’ bureau of Pfizer and participated in the Joslin diabetes CME programs. He has received research funding from Astellas Pharma and Mitsubishi Pharma and reads clinical skin biopsies, runs an EMG lab, and cares for patients with peripheral nerve diseases. A.J.S. has received payment for expert testimony in the possible neurotoxic injury of the peripheral nerve.
DISCLAIMER The diagnosis and evaluation of polyneuropathy is complex. The practice parameter is not intended to replace the clinical judgment of experienced physicians in the evaluation of polyneuropathy. The particular kinds of tests utilized by a physician in the evaluation of polyneuropathy depend upon the specific clinical situation and the informed medical judgment of the treating physician. This statement is provided as an educational service of the AAN, AANEM, and AAPM&R. It is based upon an assessment of current scientific and clinical information. It is not intended to include all possible proper methods of care for a particular neurologic problem or all legitimate criteria for choosing to use a specific test or procedure. Neither is it intended to exclude any reasonable alternative methodologies. The AAN,
AANEM, and AAPM&R recognize that specific care decisions are the prerogative of the patient and physician caring for the patient, based on all of the circumstances involved. The clinical context section is made available in order to place the evidence-based guideline into perspective with current practice habits and challenges. No formal practice recommendations should be inferred.
APPENDIX 1 Quality Standards Subcommittee (AAN): Jacqueline French, MD, FAAN (chair); Charles E. Argoff, MD; Eric Ashman, MD; Stephen Ashwal, MD, FAAN (ex-officio); Christopher Bever, Jr., MD, MBA, FAAN; John D. England, MD, FAAN (QSS facilitator); Gary M. Franklin, MD, MPH, FAAN (ex-officio); Deborah Hirtz, MD (ex-officio); Robert G. Holloway, MD, MPH, FAAN; Donald J. Iverson, MD, FAAN; Steven R. Messe´, MD; Leslie A. Morrison, MD; Pushpa Narayanaswami, MD, MBBS; James C. Stevens, MD, FAAN (ex-officio); David J. Thurman, MD, MPH (ex-officio); Dean M. Wingerchuk, MD, MSc, FRCP(C); and Theresa A. Zesiewicz, MD, FAAN.
APPENDIX 2 Practice Issues Review Panel (AANEM): Yuen T. So, MD, PhD (chair); Michael T. Andary, MD; Atul Patel, MD; Carmel Armon, MD; David del Toro, MD; Earl J. Craig, MD; James F. Howard, Jr, MD; Joseph V. Campellone Jr., MD; Kenneth James Gaines, MD; Robert Werner, MD; and Richard Dubinsky, MD.
APPENDIX 3 Clinical Quality Improvement Committee (AAPM&R): Dexanne B. Clohan, MD (chair); William L. Bockenek, MD; Lynn Gerber, MD; Edwin Hanada, MD; Ariz R. Mehta, MD; Frank J. Salvi, MD, MS; and Richard D. Zorowitz, MD.
Received April 24, 2008. Accepted in final form August 29, 2008. REFERENCES 1. England JD, Gronseth GS, Franklin G, et al. Distal symmetric polyneuropathy: a definition for clinical research: Report of the American Academy of Neurology, the American Association of Electrodiagnostic Medicine, and the American Academy of Physical Medicine and Rehabilitation. Neurology 2005;64:199–207. 2. Martyn CN, Hughes RAC. Epidemiology of peripheral neuropathy. J Neurol Neurosurg Psychiatry 1997;62:310– 318. 3. England JD, Asbury AK. Peripheral neuropathy. Lancet 2004;363:2151–2161. 4. Barohn RJ. Approach to peripheral neuropathy and myopathy. Semin Neurol 1998;18:7–18. (Class III) 5. Jann S, Beretta S, Bramerio M, Defanti CA. Prospective follow-up study of chronic polyneuropathy of undetermined cause. Muscle Nerve 2001;24:1197–1201. (Class III) 6. Lubec D, Muellbacher W, Finsterer J, Mamoli B. Diagnostic work-up in peripheral neuropathy: an analysis of 171 cases. Postgrad Med J 1999;75:723–727. (Class III) 7. Wolfe GI, Baker NS, Amato AA, et al. Chronic cryptogenic sensory polyneuropathy: clinical and laboratory characteristics. Arch Neurol 1999;56:540–547. (Class III) 8. Notermans NC, Wokke JH, Franssen H, et al. Chronic idiopathic polyneuropathy presenting in middle or old age: a clinical and electrophysiological study of 75 patients. J Neurol Neurosurg Psychiatry 1993;10:1066–1071. (Class III) 9. Notermans NC, Wokke JH, van der Graaf Y, Franssen H, van Dijk GW, Jennekens FG. Chronic idiopathic axonal Neurology 72
January 13, 2009
191
10. 11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
192
polyneuropathy: a five year follow up. J Neurol Neurosurg Psychiatry 1994;57:1525–1527. (Class III) Fagius J. Chronic cryptogenic polyneuropathy. Acta Neurol Scand 1983;67:173–180. (Class III) Dyck PJ, Oviatt KF, Lambert EH. Intensive evaluation of referred unclassified neuropathies yields improved diagnosis. Ann Neurol 1981;10:222–226. (Class IV) McLeod JG, Tuck RR, Pollard JD, Cameron J, Walsh JC. Chronic polyneuropathy of undetermined cause. J Neurol Neurosurg Psychiatry 1984;47:530–535. (Class III) Johannsen L, Smith T, Havsager A-M, et al. Evaluation of patients with symptoms suggestive of chronic polyneuropathy. J Clin Neuromuscul Disord 2001;3:47–52. (Class III) Saperstein DS, Wolfe GI, Gronseth GS, et al. Challenges in the identification of cobalamin-deficient polyneuropathy. Arch Neurol 2003;60:1296–1301. (Class III) Lindenbaum J, Rosenberg IH, Wilson PWF, Stabler SP, Allen RH. Prevalence of cobalamin deficiency in the Framingham elderly population. Am J Clin Nutr 1994;60:2– 11. (Class II) Lindenbaum J, Savage DG, Stabler SP, et al. Diagnosis of cobalamin deficiency: II: relative sensitivities of serum cobalamin, methylmalonic acid and total homocysteine concentrations. Am J Hematol 1990;34:99–107. (Class II) Savage DG, Lindenbaum J, Stabler SP, Allen RH. Sensitivity of serum methylmalonic acid and total homocysteine determinations for diagnosing cobalamin and folate deficiencies. Am J Med 1994;96:239–246. (Class III) Novella SP, Inzucchi SE, Goldstein JM. The frequency of undiagnosed diabetes and impaired glucose tolerance in patients with idiopathic sensory neuropathy. Muscle Nerve 2001;24:1229–1231. (Class III) Singleton JR, Smith AG, Bromberg MB. Painful sensory polyneuropathy associated with impaired glucose tolerance. Muscle Nerve 2001;24:1225–1228. (Class IV) Sumner CJ, Sheth S, Griffin JW, Cornblath DR, Polydefkis M. The spectrum of neuropathy in diabetes and impaired glucose tolerance. Neurology 2003;60:108–111. (Class III) Hughes RA, Umapathi T, Gray IA, et al. A controlled investigation of the cause of chronic idiopathic axonal polyneuropathy. Brain 2004;127:1723–1730. (Class III) Kelly JJ, Kyle RA, O’Brien PC, Dyck PJ. Prevalence of monoclonal proteins in peripheral neuropathy. Neurology 1981;31:1480–1483. (Class III) Kahn SN, Bina M. Sensitivity of immunofixation electrophoresis for detecting IgM paraproteins in serum. Clin Chem 1988;34:1633–1635. Boerkoel CF, Takashima H, Garcia CA, et al. CharcotMarie-Tooth disease and related neuropathies: mutation distribution and genotype-phenotype correlation. Ann Neurol 2002;51:190–201. (Class I)
Neurology 72
January 13, 2009
25.
26.
27.
28.
29.
30.
31.
32.
33. 34.
35.
36.
37.
Boerkoel CF, Takashima H, Lupski JR. The genetic convergence of Charcot-Marie-Tooth disease types 1 and 2 and the role of genetics in sporadic neuropathy. Curr Neurol Neurosci Rep 2002;2:70–77. Hoogendijk JE, Hensels GW, Gabreels-Festen AA, et al. De-novo mutation in hereditary motor and sensory neuropathy type I. Lancet 1992;339:1081–1082. (Class II) Wise CA, Garcia CA, Davis SN, et al. Molecular analyses of unrelated Charcot-Marie-Tooth (CMT) disease patients suggest a high frequency of the CMTIA duplication. Am J Hum Genet 1993;53:853–863. (Class II) Bort S, Nelis E, Timmerman V, et al. Mutational analysis of the MPZ, PMP22 and Cx32 genes in patients of Spanish ancestry with Charcot-Marie-Tooth disease and hereditary neuropathy with liability to pressure palsies. Hum Genet 1997;99:746–754. (Class I) Janssen EA, Kemp S, Hensels GW, et al. Connexin32 gene mutations in X-linked dominant Charcot-MarieTooth disease (CMTX1). Hum Genet 1997;99:501– 505. (Class I) Silander K, Meretoja P, Juvonen V, et al. Spectrum of mutations in Finnish patients with Charcot-Marie-Tooth disease and related neuropathies. Hum Mutat 1998;12: 59–68. (Class II) Leonardis L, Zidar J, Ekici A, Peterlin B, Rautenstrauss B. Autosomal dominant Charcot-Marie-Tooth disease type 1A and hereditary neuropathy with liability to pressure palsies: detection of the recombination in Slovene patients and exclusion of the potentially recessive Thr118MetPMP22 point mutation. Int J Mol Med 1998;1:495–501. (Class III) Nelis E, Van Broeckhoven C, De Jonghe P, et al. Estimation of the mutation frequencies in Charcot-Marie-Tooth disease type 1 and hereditary neuropathy with liability to pressure palsies: a European collaborative study. Eur J Hum Genet 1996;4:25–33. (Class I) Nicholson GA. Mutation testing in Charcot-Marie-Tooth neuropathy. Ann NY Acad Sci 1999;883:383–388. (Class II) Mersiyanova IV, Ismailov SM, Polyakov AV, et al. Screening for mutations in the peripheral myelin genes PMP22, MPZ and Cx32 (GJB1) in Russian Charcot-Marie-Tooth neuropathy patients. Hum Mutat 2000;15:340–347. (Class II) Mostacciuolo ML, Righetti E, Zortea M, et al. CharcotMarie-Tooth disease type 1 and related demyelinating neuropathies: mutation analysis in a large cohort of Italian families. Hum Mutat 2001;18:32–41. (Class I) Choi BO, Lee MS, Shin SH, et al. Mutational analysis of PMP22, MPZ, GJB1, EGR2 and NEFL in Korean Charcot-Marie-Tooth neuropathy patients. Hum Mutat 2004;24:185–186. (Class I) Verhoeven K, Claeys KG, Zuchner S, et al. MFN2 mutation distribution and genotype/phenotype correlation in CharcotMarie-Tooth type 2. Brain 2006;129:2093–2102. (Class II)
Clinical/Scientific Notes
N. Weiss, MD D. Hasboun, MD, PhD S. Demeret, MD B. Fontaine, MD, PhD F. Bolgert, MD O. Lyon-Caen, MD D. Chabas, MD, PhD
Supplemental data at www.neurology.org
PAROXYSMAL HYPOTHERMIA AS A CLINICAL FEATURE OF MULTIPLE SCLEROSIS
Hypothermia is rare in multiple sclerosis (MS). Only 14 patients with MS with hypothermia have been reported1,2 (table e-1 on the Neurology® Web site at www.neurology.org). Here we present a patient with MS with recurrent episodes of profound hypothermia, likely due to demyelination. Case report. In March 2000, a 41-year-old man with a 7-year MS history was admitted to the intensive care unit for severe hypothermia. The patient initially developed recurrent episodes of paraparesis and ataxia in 1993. In 1995, the brain MRI showed multiple white matter lesions and the patient began treatment with interferon beta-1b. In 1996, he developed rapidly worsening paraparesis, ataxia, and internuclear ophthalmoplegia. The repeat brain MRI scan showed new T2-weighted hyperintensities (not shown). The patient received monthly infusions of mitoxantrone (20 mg) for 6 months and the disease subsequently stabilized with significant residual disability (Expanded Disability Status Score of 7). In February 2000, the patient developed rapidly progressive slurred speech, hypothermia, paranoid delusions, and vivid auditory, visual, and tactile sensations suggestive of hypnagogia, over a period of 3 weeks. Upon admission, the body temperature was 30.0°C, heart rate was 100 beats/minute, and blood pressure was 120/70 mm Hg. Paraplegia, bilateral upper extremity weakness, bilateral facial droop, and miosis were noted. After 6 hours of external rewarming, the temperature rose to 33°C. The patient subsequently became comatose and required respiratory assistance. He was treated presumptively for sepsis; cultures were negative. Blood and urine toxicology screens, TSH, TRH, and cosyntropin stimulation test were normal. Platelets were 113,000/mm3, prothrombin time (PT) was 1.1, and serum sodium was normal. Brain MRI showed an increased overall lesion burden and a new T2-weighted hypothalamic hyperintensity (figure). Complete clinical recovery ensued over 6 weeks. At discharge, the patient was normothermic (36 –37°C). He subsequently received five monthly IV methylprednisolone infusions (1 g/month). In the following 3 years, he
developed four similar hypothermic episodes (30.4 –33°C). Each lasted 2 to 4 weeks; three required mechanical ventilation. Psychiatric manifestations recurred each time the temperature was below 31°C. Platelets were low (64,000 –90,000/ mm3), PT was elevated (1.1–1.4), and serum sodium remained within normal range. Thyroid, adrenal insufficiency, and infectious workups were negative each time except for one urinary tract infection. The patient recovered systematically within a week of receiving high dose IV methylprednisolone (1 g/day for 3 days). Between hypothermic episodes, the patient noticed that he felt best when his body temperature was between 35.5°C and 36°C, while his neurologic status would worsen below 35°C or above 37°C. The patient eventually died in April 2003 during the sixth hypothermic exacerbation (31°C) prior to receiving steroids. No autopsy was performed. Discussion. Most patients with MS with symptomatic hypothermia typically have a long history of MS and are significantly disabled (Expanded Disability Status Scale 6 –9) (table e-1). In most reported cases, the temperature was above 31°C, although one patient presented with a temperature of 29°C. Deep hypothermia was usually accompanied by nonspecific symptoms such as dysarthria, delusions, miosis, bradycardia, thrombopenia, and altered PT, as with our patient or with patients without MS who have experienced acute hypothermia of accidental origin.3 Hypothermia may be monophasic (5/14) or have a recurrent (6/14) course, as with our patient. In the case of our patient, relapses responded well to steroids, suggesting an inflammatory origin of the hypothermia. Paroxysmal hypothermia is almost invariably associated with a lesion in the preoptic area and may be associated with various causes.4 It is likely that the new hypothalamic MRI lesion, including the preoptic area, concomitant to the first hypothermic attack was causal in the case of our patient (figure). The hypothalamic involvement is further supported by the presence of concomitant dream-like hallucinations. The absence of any further documented hyperthermic episodes in response to infection also Neurology 72
January 13, 2009
193
Figure
MRI lesions in the hypothalamic area
The brain MRI scan included axial (A) and coronal (B, C) FLAIR images. The planes of intersection in A, B, and C are represented in D (sagittal view of the right medial hypothalamus nuclei in the lateral wall and the floor of the third ventricle). An elongated FLAIR-weighted hyperintense signal was seen on the axial cut between the paraterminal gyrus and the posterior hypothalamus bilaterally (A, stars), which included the preoptic area (arrows). This lesion was also seen on coronal images including the middle (B, stars) and posterior (C, stars) hypothalamus. A ⫽ anterior nucleus (n); Ar ⫽ arcuate n; DM ⫽ dorsomedial n; LT ⫽ lamina terminalis; MB ⫽ mammillary body; Po ⫽ preoptic n; Pv ⫽ paraventricular n; P ⫽ posterior n; SC ⫽ suprachiasmatic n; So ⫽ supraoptic n; VM ⫽ ventromedial n.
suggests a dysregulation of hyperthermic response in the preoptic area.5 Finally, both hypothermia and hyperthermia (Uhthoff phenomenon) have been reported to cause nerve slowing. Remarkably, our patient reported neurologic worsening alternatively in both conditions. 194
Neurology 72
January 13, 2009
One postmortem pathologic analysis and one MRI study found lesions around the hypothalamic area in hypothermic patients with MS,1,2 although none specified the involvement of the preoptic area. The good response to steroids in our patient suggests that the lesion was recent. This is in line with
the fact that 60% of reported hypothalamic MS plaques are active.6 Our case indicates that paroxysmal hypothermia can be profound and still relatively well tolerated in MS, possibly because the MS flare at the origin of hypothermia has a rapidly progressive onset, and not acute, allowing some time for adaptation. Still, hypothermia can be fatal in patients with MS. From the Service de Re´animation Neurologique (N.W., S.D., F.B.), Service de Neuroradiologie (D.H.), Fe´de´ration des Maladies du Syste`me Nerveux Central (B.F., O.L.-C., D.C.), Assistance Publique-Hoˆpitaux de Paris, Groupement Hospitalier Pitie´Salpeˆtrie`re, Paris, France; and Universite´ Pierre et Marie Curie (D.H., B.F., O.L.-C., D.C.), INSERM U 546, Paris. D.C. is currently with the Multiple Sclerosis Center, University of California, San Francisco. Disclosure: The authors report no disclosures. Received February 20, 2008. Accepted in final form July 31, 2008. Address correspondence and reprint requests to Dr. Dorothe´e Chabas, UCSF Multiple Sclerosis Center, 350, Parnassus Avenue, Suite 908, San Francisco, CA 94117;
[email protected] Copyright © 2009 by AAN Enterprises, Inc.
E. Bartoccioni, PhD F. Scuderi, PhD A. Augugliaro, PhD S. Chiatamone Ranieri, MD D. Sauchelli, MD P. Alboino M. Marino, MD, PhD A. Evoli, MD
HLA CLASS II ALLELE ANALYSIS IN MuSKPOSITIVE MYASTHENIA GRAVIS SUGGESTS A ROLE FOR DQ5
Myasthenia gravis (MG) is caused by autoantibodies targeting, in most cases, the acetylcholine receptor (AChR-MG). Different disease subtypes are distinguished on the basis of clinical characteristics and thymus pathology. In 40% of patients with anti-AChR negative generalized MG, the disease appears to be mediated by antibodies against the muscle specific kinase (MuSK-MG).1 We evaluated HLA-DRB1*, DQA1*, and DQB1* allele profile in MuSK-MG in comparison with a control population and non-thymoma early onset AChR-MG (AChR-EOMG). We chose to compare these two clinical entities as they share a high prevalence in women and a proportion of MuSKMG patients have early onset disease. Patients. Our study includes consecutive unrelated patients, all with generalized MG. Patients gave informed consent to inclusion in the study, which was approved by the local Ethics Committee. The MuSK-MG group included 37 patients (8 men/29 women, onset age: 6 – 62 years), the thymus was normal for age in 10 patients who underwent thymectomy, thyroid autoimmunity was associated in 4/37 cases (10.5%). The AChR-EOMG group comprised 28 patients (4 men/24 women, onset age: 9 –39 years), thymic hyperplasia was found in 24/26 thymectomized cases, and different autoimmune disorders were associated in 8/28 (28.6%).
ACKNOWLEDGMENT The authors thank Ellen Mowry, MD, and Elizabeth Crabtree-Hartman, MD, for editing the manuscript.
1.
2.
3.
4.
5.
6.
Linker RA, Mohr A, Cepek L, Gold R, Prange H. Core hypothermia in multiple sclerosis: case report with magnetic resonance imaging localization of a thalamic lesion. Mult Scler 2006;12:112–115. White KD, Scoones DJ, Newman PK. Hypothermia in multiple sclerosis. J Neurol Neurosurg Psychiatry 1996; 61:369–375. Fischbeck KH, Simon RP. Neurological manifestations of accidental hypothermia. Ann Neurol 1981;10:384– 387. Plum F, Van Uitert R. Nonendocrine diseases and disorders of the hypothalamus. Res Publ Assoc Res Nerv Ment Dis 1978;56:415–473. Lazarus M, Yoshida K, Coppari R, et al. EP3 prostaglandin receptors in the median preoptic nucleus are critical for fever responses. Nat Neurosci 2007;10:1131–1133. Huitinga I, Erkut ZA, van Beurden D, Swaab DF. The hypothalamo-pituitary-adrenal axis in multiple sclerosis. Ann NY Acad Sci 2003;992:118–128.
All patients were from Central Italy, with Italian ancestors. For the allele profile, the control group consisted of 380 unrelated individuals representative of Central Italian population.2 For haplotype frequency, patients were compared with a control Italian population of 53 unrelated individuals.3 Methods. Anti-AChR and anti-MuSK antibodies were assayed by radioimmunoprecipitation (RSR Limited, Cardiff, UK). PBL genomic DNA was extracted by High Pure PCR Template Preparation Kit (Roche, Switzerland). HLA-DQA1*, -DQB1*, and -DRB1* loci were typed by PCR-SSO. Highresolution HLA-DQA1* and DQB1* typing and low-resolution DRB1* typing were performed (Innogenetics, Italy). Statistical analysis. Statistical analysis was performed by Flavia Scuderi, PhD, and was revised by Alberto Mari, PhD, in Statistical Sciences. Odds ratios (ORs) with 95% confidence intervals (CIs) according to the Woolf test and two-sided Fisher exact test were used; p values were corrected for multiple comparisons according to the Bonferroni method (pc) and were considered significant when p ⬍ 0.05. Results. No association was found for DQA1* alleles in patients with MG (data not shown). Table 1 shows the DQB1* and DRB1* allele frequency in patients and controls. We found a highly significant association of AChR-EOMG with the Neurology 72
January 13, 2009
195
the fact that 60% of reported hypothalamic MS plaques are active.6 Our case indicates that paroxysmal hypothermia can be profound and still relatively well tolerated in MS, possibly because the MS flare at the origin of hypothermia has a rapidly progressive onset, and not acute, allowing some time for adaptation. Still, hypothermia can be fatal in patients with MS. From the Service de Re´animation Neurologique (N.W., S.D., F.B.), Service de Neuroradiologie (D.H.), Fe´de´ration des Maladies du Syste`me Nerveux Central (B.F., O.L.-C., D.C.), Assistance Publique-Hoˆpitaux de Paris, Groupement Hospitalier Pitie´Salpeˆtrie`re, Paris, France; and Universite´ Pierre et Marie Curie (D.H., B.F., O.L.-C., D.C.), INSERM U 546, Paris. D.C. is currently with the Multiple Sclerosis Center, University of California, San Francisco. Disclosure: The authors report no disclosures. Received February 20, 2008. Accepted in final form July 31, 2008. Address correspondence and reprint requests to Dr. Dorothe´e Chabas, UCSF Multiple Sclerosis Center, 350, Parnassus Avenue, Suite 908, San Francisco, CA 94117;
[email protected] Copyright © 2009 by AAN Enterprises, Inc.
E. Bartoccioni, PhD F. Scuderi, PhD A. Augugliaro, PhD S. Chiatamone Ranieri, MD D. Sauchelli, MD P. Alboino M. Marino, MD, PhD A. Evoli, MD
HLA CLASS II ALLELE ANALYSIS IN MuSKPOSITIVE MYASTHENIA GRAVIS SUGGESTS A ROLE FOR DQ5
Myasthenia gravis (MG) is caused by autoantibodies targeting, in most cases, the acetylcholine receptor (AChR-MG). Different disease subtypes are distinguished on the basis of clinical characteristics and thymus pathology. In 40% of patients with anti-AChR negative generalized MG, the disease appears to be mediated by antibodies against the muscle specific kinase (MuSK-MG).1 We evaluated HLA-DRB1*, DQA1*, and DQB1* allele profile in MuSK-MG in comparison with a control population and non-thymoma early onset AChR-MG (AChR-EOMG). We chose to compare these two clinical entities as they share a high prevalence in women and a proportion of MuSKMG patients have early onset disease. Patients. Our study includes consecutive unrelated patients, all with generalized MG. Patients gave informed consent to inclusion in the study, which was approved by the local Ethics Committee. The MuSK-MG group included 37 patients (8 men/29 women, onset age: 6 – 62 years), the thymus was normal for age in 10 patients who underwent thymectomy, thyroid autoimmunity was associated in 4/37 cases (10.5%). The AChR-EOMG group comprised 28 patients (4 men/24 women, onset age: 9 –39 years), thymic hyperplasia was found in 24/26 thymectomized cases, and different autoimmune disorders were associated in 8/28 (28.6%).
ACKNOWLEDGMENT The authors thank Ellen Mowry, MD, and Elizabeth Crabtree-Hartman, MD, for editing the manuscript.
1.
2.
3.
4.
5.
6.
Linker RA, Mohr A, Cepek L, Gold R, Prange H. Core hypothermia in multiple sclerosis: case report with magnetic resonance imaging localization of a thalamic lesion. Mult Scler 2006;12:112–115. White KD, Scoones DJ, Newman PK. Hypothermia in multiple sclerosis. J Neurol Neurosurg Psychiatry 1996; 61:369–375. Fischbeck KH, Simon RP. Neurological manifestations of accidental hypothermia. Ann Neurol 1981;10:384– 387. Plum F, Van Uitert R. Nonendocrine diseases and disorders of the hypothalamus. Res Publ Assoc Res Nerv Ment Dis 1978;56:415–473. Lazarus M, Yoshida K, Coppari R, et al. EP3 prostaglandin receptors in the median preoptic nucleus are critical for fever responses. Nat Neurosci 2007;10:1131–1133. Huitinga I, Erkut ZA, van Beurden D, Swaab DF. The hypothalamo-pituitary-adrenal axis in multiple sclerosis. Ann NY Acad Sci 2003;992:118–128.
All patients were from Central Italy, with Italian ancestors. For the allele profile, the control group consisted of 380 unrelated individuals representative of Central Italian population.2 For haplotype frequency, patients were compared with a control Italian population of 53 unrelated individuals.3 Methods. Anti-AChR and anti-MuSK antibodies were assayed by radioimmunoprecipitation (RSR Limited, Cardiff, UK). PBL genomic DNA was extracted by High Pure PCR Template Preparation Kit (Roche, Switzerland). HLA-DQA1*, -DQB1*, and -DRB1* loci were typed by PCR-SSO. Highresolution HLA-DQA1* and DQB1* typing and low-resolution DRB1* typing were performed (Innogenetics, Italy). Statistical analysis. Statistical analysis was performed by Flavia Scuderi, PhD, and was revised by Alberto Mari, PhD, in Statistical Sciences. Odds ratios (ORs) with 95% confidence intervals (CIs) according to the Woolf test and two-sided Fisher exact test were used; p values were corrected for multiple comparisons according to the Bonferroni method (pc) and were considered significant when p ⬍ 0.05. Results. No association was found for DQA1* alleles in patients with MG (data not shown). Table 1 shows the DQB1* and DRB1* allele frequency in patients and controls. We found a highly significant association of AChR-EOMG with the Neurology 72
January 13, 2009
195
Table 1
Frequencies of most representative HLA alleles in MuSK- and AChR-EOMG patients and controls
Controls, n (%)
AChRⴙ, n (%)
MuSKⴙ, n (%)
AChRⴙ/controls, OR (95% CI)
MuSKⴙ/controls, OR (95% CI)
HLA DQB1* 0201 0301/0313 0502 0503 Total alleles
64 (17.3)
13 (23.2)
5 (6.8)
1.381 (0.704–2.708)
117 (31.6)
17 (30.4)
16 (21.6)
0.973 (0.535–1.769)
0.596 (0.329–1.082)
17 (4.6)
2 (3.6)
18 (31.6)a
0.742 (0.167–3.297)
6.674 (3.248–13.717)
0.343 (0.045–2.620)
3.785 (1.400–10.229)
5.579 (2.848–10.930)
0.871 (0.303–2.508)
18 (4.8)
1 (1.8)
6 (8.1)
370 (100)
56 (100)
74 (100)
0.346 (0.134–0.893)
HLA DRB1* 03
40 (6.1)
15 (26.8)b
4 (5.4)
11
177 (27.2)
14 (25.0)
15 (20.3)
0.891 (0.475–1.671)
0.679 (0.356–1.229)
13
68 (10.5)
9 (16.1)
3 (4.1)
1.639 (0.769–3.491)
0.362 (0.111–1.179)
14
46 (7.0)
1 (1.8)
7 (9.5)
0.239 (0.032–1.764)
1.372 (0.596–3.159)
16
14 (2.2)
2 (3.6)
17 (23.0)c
1.682 (0.373–7.597)
13.549 (6.352–28.899)
650 (100)
56 (100)
Total alleles
74 (100)
Only significant differences are indicated with aMuSK⫹/controls, pc ⫽ 7.3 ⫻ 10⫺6; bAChR⫹/controls, pc ⫽ 5.8 ⫻ 10⫺5; c MuSK⫹/controls, pc⫽ 2.3 ⫻ 10⫺9.
DRB1*03 allele (pc 5.8 ⫻ 10⫺5) and of MuSKMG with DQB1*0502 (pc 7.3 ⫻ 10⫺6) and DRB1*16 (pc 2.3 ⫻ 10⫺9) alleles. When we considered the serologic DQ5 corresponding to molecular DQB1*0501/02/03, its allele frequency was 44.6% in MuSK-MG and 21.1% in controls (OR 3.013, CI 1.788 –5.078, p 5.4 ⫻ 10⫺5, pc 2.2 ⫻ 10⫺4). Moreover, 27/37 MuSK-MG patients (73%) showed one or two DQ5 alleles. Haplotype distribution is shown in table 2. The DRB1*16, DQA1*0102, DQB1*0502 haplotype was associated with MuSK-MG (17/74 haplotypes) (MuSK-MG vs controls, OR 6.025, CI 2.111–17.193), while the DRB1*03, DQA1*0501, DQB1*0201 haplotype was associated with AChRTable 2
EOMG (13/56 haplotypes) (AChR-EOMG vs controls, OR 5.039, CI 1.797–14.132). An association trend was found for DRB1*14, DQA1*0104, DQB1*0503 haplotype with MuSK-MG and for DRB1*13, DQA1*0102, DQB1*0601 with AChRMG patients. In the MuSK-MG group, 16 patients (43.24%) carried the DR16-DQ5 haplotype and 7 (18.92%) carried the DR14-DQ5 haplotype; only one patient had both haplotypes. Two of four MuSK-MG patients with thyroid autoimmunity carried the DRB1*03, DQA1*0501, DQB1*0201 haplotype, which was also present in four of eight AChR-MG subjects with other autoimmune disorders.
Frequencies of most representative HLA haplotypes in MuSK- and AChR-EOMG patients and controls
-
Controls, n (%)
Haplotype DRB1*03; DQA1*0501; DQB1*0201
DRB1*13; DQA1*0102†;DQB1*0601‡
DRB1*14; DQA1*0104; DQB1*0503
DRB1*16; DQA1*0102; DQB1*0502
Total haplotypes
6 (5.7)
10 (9.5)
4 (3.8)
5 (4.7)
106 (100)
AChRⴙ, n (%) 13 (22.4)*
10 (17.2)
1 (1.7)
2 (3.6)
56 (100)
MuSKⴙ, n (%) 4 (5.4)
3 (4.0)
6 (8.1)
17 (23.0)§
AChRⴙ/controls, OR (95% CI)
MuSKⴙ/controls, OR (95% CI)
5.039 (1.797–14.132)
0.952 (0.259–3.500)
0.058–0.617
1.620–17.274
2.087 (0.812–5.366)
0.406 (0.108–1.528)
0.057–0.858
1.166–17.616
0.464 (0.051–4.250)
2.250 (0.612–8.271)
0.567–41.523
0.024–1.673
0.748 (0.140–3.985)
6.025 (2.111–17.193)
1.776–36.515
0.027–0.563
74 (100)
Only significant differences are indicated with *AChR⫹/controls (13/56 vs 6/106) pc ⫽ 0.0037; §MuSK⫹/controls (17/74 vs 5/106) pc ⫽ 0.001. †DQA1*0102 and 0103. ‡DQB1*0601– 04. 196
Neurology 72
January 13, 2009
Discussion. We confirm the strong positive association of AChR-EOMG with DRB1*03 DQA1* 0501 and DQB1*0201 alleles as previously reported. MuSK-MG is a rare and relatively newly described clinical entity4; in our patients, it appears to be strongly associated with DRB1*16 and DQB1*0502 alleles. This association was recently reported in a family including a HLA-DR1-DQ5, DR3-DQ2 AChR-MG mother with a HLADR16-DQ5, DR1-DQ5 MuSK-MG daughter.5 A previous study of 23 Dutch patients reported a significant association with the DR14-DQ5 haplotype,6 where the serologic analysis could not discriminate differences between allele molecular profiles. In our series, DR14 allele showed only a trend of association with MuSK-MG, but the fact that DR16 and DR14 are both in linkage with DQ5 points to the relative importance of this last allele. When we compared the whole DQ5 allele frequency (DQB1*0501/02/03) corresponding to serologic DQ5 in our MuSK-MG population with that of the Dutch study, we found very similar results (73% vs 78.3%). In our population, the association of MuSK-MG with HLA-DQ5 appears to be related to DQB1*0502 allele. We show that MuSK-MG and AChR-EOMG, which have distinct clinical characteristics and thymus changes, significantly differ in HLA class II allele frequency. Similarly to AChR-EOMG, which is linked with DR3 and DQ2 alleles in patient series from different countries, MuSK-MG appears to be
associated with DQ5 alleles in populations with diverse ancestries both from Southern and Northern Europe. From the General Pathology Institute (E.B., F.S., A.A., S.C.R., M.M.) and the Department of Neuroscience (D.S., P.A., A.E.), Catholic University, Rome, Italy. Supported by financial grants of Catholic University to E.B. and A.E. Disclosure: The authors report no disclosures. Received May 22, 2008. Accepted in final form August 5, 2008. Address correspondence and reprint requests to Dr. Emanuela Bartoccioni, General Pathology Institute, Catholic University, Largo F. Vito, 1, 00168 Rome, Italy;
[email protected] Copyright © 2009 by AAN Enterprises, Inc. ACKNOWLEDGMENT The authors thank Dr. Alberto Mari for the revision of statistical analysis.
1. 2. 3.
4.
5.
6.
Vincent A, Palace J, Hilton-Jones D. Myasthenia gravis. Lancet 2001;357:2122–2128. Adorno D, Piancatelli D, Canossi A, et al. HLA 1998. Lenexa, KS: ASHI; 1998. Lulli P, Grammatico P, Brioli G, et al. HLA-DR and -DQ alleles in Italian patients with melanoma. Tissue Antigens 1998;51:276–280. Evoli A, Tonali PA, Padua L, et al. Clinical correlates with anti-MuSK antibodies in generalized seronegative myasthenia gravis. Brain 2003;126:2304–2011. Lavrnic D, Nikolic A, De Baets M, et al. Familial occurrence of autoimmune myasthenia gravis with different antibody specificity. Neurology 2008;70:2011–2013. Niks EH, Kuks JB, Roep BO, et al. Strong association of MuSK-positive myasthenia gravis and HLA-DR14-DQ5. Neurology 2006;66:1772–1774.
Neurology 72
January 13, 2009
197
NEUROIMAGES
A radiologic “alcohol breathalyzer” test
Figure
Magnetic resonance spectroscopy showing the ethanol triplet resonance of 1.16 ppm (highlighted in white box) 1
We did not find changes of hepatic encephalopathy, which are a reduction in the myoinositol and choline resonances and an increase in the glutamate/ glutamine composite resonance.
A 52-year-old man with 3 years of progressive unsteadiness admitted to previous alcohol excess but denied ongoing consumption. Magnetic resonance spectroscopy (MRS) (figure) was performed looking for changes of subclinical hepatic encephalopathy, which was not found. Instead, it showed the characteristic triplet resonance peak for ethanol—a confirmation of alcohol consumption that day, which was admitted by the patient. This image is a helpful reminder that MRS (in the appropriate clinical context along with EEG and hyperammonemia) can help diagnose hepatic encephalopathy1 (clinically excluded in our patient), and in an interesting twist, may also be a radiologic alcohol breathalyzer test. Sui H. Wong, MRCP, Trevor Smith, FRCR, Richard P. White, MD, FRCP, Liverpool, UK Disclosure: The authors report no disclosures. Address correspondence and reprint requests to Dr. Sui Wong, Neurology Specialist Registrar, Department of Neurology, Walton Centre for Neurology and Neurosurgery NHS Trust, Lower Lane, Fazakerley, Liverpool L9 7LJ, UK;
[email protected] 1.
198
Grover VP, Dresner MA, Forton DM, et al. Current and future applications of magnetic resonance imaging and spectroscopy of the brain in hepatic encephalopathy. World J Gastroenterol 2006;12:2969–2978.
Copyright © 2009 by AAN Enterprises, Inc.
RESIDENT & FELLOW SECTION Section Editor Mitchell S.V. Elkind, MD, MS
Chafic Karam, MD Azita Khorsandi, MD Daniel J. MacGowan, MD
Address correspondence and reprint requests to Dr. Chafic Karam, 353 East 17th Street, #22D, New York, NY 10003
[email protected]
Clinical Reasoning: A 23-year-old woman with paresthesias and weakness SECTION 1
A 23-year-old woman presented with a 6-month history of progressive left hand weakness associated with left ulnar distribution numbness and paresthesias. At the onset of these symptoms, she recalled shooting pain up and down the medial left forearm. She denied any neck pain. The patient had been diagnosed with acute myeloid leukemia (AML) type 5a and appendicitis 9 months previously. At that time lumbar puncture and brain and spine MRI were negative for CNS involvement. A right Hickman catheter was placed. She was treated with cytarabine and idarubicin and was thought to be in complete remission after re-
peated bone marrow biopsies showed no blasts. No intrathecal chemotherapy was given. The appendicitis was treated for 1 month with moxifloxacin prior to an elective appendectomy. Bone marrow examination and peripheral blood smear were normal at the time of neurologic presentation. Questions for consideration: 1. Where can the lesion be localized and what could be its nature in the context of this patient’s history? 2. How could the differential diagnosis be narrowed further?
GO TO SECTION 2
From the Departments of Neurology (C.K., D.J.M.) and Neuro-radiology (A.K.), Beth Israel Medical Center, Albert Einstein College of Medicine, New York, NY. Disclosure: The authors report no disclosures.
Copyright © 2009 by AAN Enterprises, Inc.
e5
SECTION 2
The evaluation of a patient with hand weakness, numbness, and paraesthesias would start with localizing the lesion. A lesion in the CNS, involving the contralateral precentral gyrus and anterior aspect of the postcentral gyrus,1 is highly unlikely, especially with the shooting pain up and down the medial left forearm. In the peripheral nervous system, radiculopathy, plexopathy, or a nerve lesion could be responsible for combined motor and sensory impairment in the hand. Minor repetitive trauma can cause nerve damage, resulting in carpal tunnel syndrome or ulnar neuropathy at the elbow. The radial, ulnar, or median nerve could have been damaged during the patient’s recent surgery. Other causes of isolated neuropathy in this patient include a mononeuritis multiplex caused by blast cell infiltration of the left ulnar nerve. The brachial plexus may also be compressed or infiltrated by an extramedullary myeloid tumor (EMT) in the setting of leukemia. Neurotoxicity secondary to the chemotherapy is more likely to be bilateral, symmetric, and ascending.
In this patient, the neurologic examination showed 4/5 strength of the left interossei, flexor carpi ulnaris, ulnar flexor digitorum profundus, and adductor pollicis. There was no other weakness. There was reduced pin sensation in the medial aspect of the 4th digit, entire 5th digit, hypothenar eminence, and medial one third of the dorsal hand. Tinel sign was positive in the left ulnar cubital tunnel. Sensation of the left medial forearm was intact. Tone, reflexes, vibration, coordination, and gait were normal. There was no Horner syndrome. The neurologic examination indicates that the lesion is localized to the peripheral nervous system. The symptoms can be explained by left ulnar neuropathy, left C8 radiculopathy, or a lower trunk/ medial cord plexopathy. The absence of neck pain argues against a nerve root lesion. Questions for consideration: 1. What are the clinical findings in an ulnar neuropathy and a lower trunk/medial cord plexopathy? 2. What is the role of an EMG/NCS study?
GO TO SECTION 3
e6
Neurology 72
January 13, 2009
SECTION 3
Patients with ulnar neuropathy usually present with numbness and paresthesias involving the 5th finger and the ulnar half of the 4th digit. Some patients may notice a dull ache down the ulnar border of the forearm. The symptoms can be elicited by having the patient flex the elbow or by tapping over the ulnar nerve at the cubital tunnel (Tinel sign). C8 radiculopathy causes pain and numbness of the 4th and 5th digits. T1 root pain causes pain in the shoulder joint radiating down the medial side of the upper arm and forearm with numbness of the medial arm and forearm. Atrophy and weakness reflect motor axon or anterior horn cell loss. A proximal ulnar neuropathy at the elbow (UNE) results in weakness in the interosseous muscles, adductor pollicis, and long flexors of the 4th and 5th digits. Usually it spares the flexor carpi ulnaris (FCU).2 A T1 root lesion will preferentially involve the abductor pollicis brevis (APB) and opponens. A C8 root lesion will cause weakness in the flexor carpi ulnaris, deep finger flexors, flexor pollicis longus, interossei, adductor pollicis, and extensor indicis proprius muscles. Medial cord brachial plexopathy would result in weakness of the muscles innervated by the ulnar nerve in addition to the median-innervated intrinsic hand muscles, i.e., opponens and APB. Thus in order to differentiate UNE from C8/T1 radiculopathy and lower trunk/ medial cord brachial plexopathy, one would search for sensory loss extending into the medial forearm
Table
Motor and nerve conduction studies of the ulnar nerve
Latency, ms
Amplitude, mV
Conduction velocity, m/s
Motor nerve conduction studies L ulnar/ADM Wrist
4.55
3.8
Below the elbow
9.3
3.2
42.1
Above the elbow
11.7
1
41.7
L ulnar/FDI Wrist
4.4
4.2
Below the elbow
8.85
2.7
44.9
Above the elbow
11.95
2.1
32.3
Sensory nerve conduction studies R ulnar-V-antidromic
2.7
70.3
L ulnar-V-antidromic
4.45
4.9
R ulnar-dorsal-antidromic
1.45
14.4
L ulnar-dorsal-antidromic
2.3
3.7
50 27 55.2 37
The moderately prolonged left distal ulnar motor response latency and conduction velocity slowing in the forearm and across the elbow reflects a loss of left ulnar large fiber axons. There is no evidence of selective slowing in the left ulnar nerve across the elbow. These findings were felt to be consistent with a proximal, axonal left ulnar neuropathy rather than a lower trunk plexopathy since the medial antebrachial cutaneous sensory, median motor conduction studies, and needle EMG of the abductor pollicis brevis were normal.
and assess for weakness of C8/T1 non-ulnar innervated muscles such as flexor pollicis longus, extensor indicis proprius, opponens and abductor pollicis brevis. This patient had sensory symptoms of an ulnar neuropathy at the elbow, with a positive Tinel sign, which is suggestive of UNE. However, the shooting pain up and down her medial forearm is atypical. This distribution of pain fits a C8 and T1 radiculopathy or a lower trunk/medial cord plexopathy. The motor findings suggest an ulnar pattern of weakness, although FCU weakness is atypical. EMG/NCS studies could help determine the affected muscles and sensory nerve deficits, thus indicating if her symptoms are secondary to UNE secondary to minor elbow trauma or positioning at the time of her surgery vs a lower trunk/medial cord brachial plexopathy. In 1999, the American Association of Electrodiagnostic Medicine (AAEM) issued a Practice Parameter for Electrodiagnostic Studies in Ulnar Neuropathy at the Elbow. The strongest evidence of ulnar neuropathy at the elbow includes an absolute motor nerve conduction velocity (NCV) from above elbow (AE) to below elbow (BE) of less than 50 m/s and an AE-to-BE segment greater than 10 m/s slower than BE-to-wrist (W) segment. Other findings include a decrease in compound muscle action potential (CMAP) negative peak amplitude from BE to AE greater than 20% suggesting a conduction block or temporal dispersion indicative of focal demyelination and a significant change in CMAP configuration at the AE site compared to the BE site. Ulnar sensory responses should be recorded from the fifth fingers and dorsal ulnar palm. In this patient, the left ulnar distal motor response latency was prolonged. There was diffuse left ulnar forearm and across elbow slowing with reduced left ulnar motor response amplitudes. The ulnar F-waves done with supramaximal stimulation at the wrist were prolonged. There was no conduction block, focal area of slowing, or temporal dispersion. The left ulnar and dorsal digital ulnar cutaneous sensory response amplitudes were very reduced. The bilateral medial cutaneous nerve of forearm sensory responses were normal. Median motor and sensory conduction studies and F-waves were normal. Needle EMG of muscles in the left upper extremity showed spontaneous fibrillations and positive sharp waves with reduced recruitment of prolonged and polyphasic motor units in all left ulnar innervated muscles including the FCU. All other muscles were normal, including APB, flexor pollicis longus, and extensor indicis proprius (table). The findings are consistent with subacute, proximal, and axonal left ulnar neuropathy. The proNeurology 72
January 13, 2009
e7
longed ulnar F-waves were thought to be due to large fiber axon loss in the ulnar nerve. The normal left medial antebrachial cutaneous sensory response and needle EMG of the abductor pollicis brevis favors isolated ulnar neuropathy. However, the involvement of the flexor carpi ulnaris (FCU) fibers is unusual in UNE. The patient had a left ulnar nerve transposition procedure but postoperatively developed worsening of the left hand weakness with new right toe extension weakness and right dorsal foot numbness. Her neurologic examination showed weakness in the following muscles on the left: APB and opponens 4/5, wrist and finger extensors 4⫹/5, FCU 4/5, FDP
IV/V 4/5, interossei and adductor pollicis 2/5; and on the right: extensor hallucis longus (EHL) 2/5 and extensor digitorum brevis (EDB) 4/5. All other muscles were normal. Pin sensation was reduced in the left hypothenar eminence, 5th and medial 4th digits, and medial forearm. Vibration and proprioception were normal. Reflexes were intact apart from trace right and absent left ankle jerks. Plantar responses were flexor. Questions for consideration: 1. What is the next step in assessing the plexopathy and why is there progression of the patient’s symptoms with multifocal weakness? 2. What is the best approach and management?
GO TO SECTION 4
e8
Neurology 72
January 13, 2009
Figure 1
Coronal spin echo T1 and postcontrast FMSPGRs demonstrate a bulky, enhancing lesion involving the divisions of the left brachial plexus surrounding the distal left subclavian and axillary arteries (arrows)
SECTION 4
Subsequent EMG/NCS showed evolution consistent with lower trunk brachial plexopathy evidenced by new absence of the left medial cutaneous nerve of forearm sensory response denervation in the extensor indicis proprius, flexor pollicis longus, and abductor pollicis brevis. The right lower extremity showed absent right and reduced left superficial peroneal sensory responses, absent right and prolonged left H-reflexes, and denervation activity in the right more than left EHL and EDB. MRI is the best neuroimaging technique for the assessment of plexopathies. MRI demonstrated bilateral brachial and lumbosacral plexus thickening and enhancement consistent with chloromas, worst in the left brachial plexus (figures 1 and 2). MRI with and without gadolinium of the brain and entire spine were negative. The peripheral blood smear now showed AML relapse. A second lumbar puncture was negative, with no leukemic cells in the CSF after cytocentrifugation, flow cytometry, and immunocytochemistry. The patient was treated with reinduction and consolidation therapy. Total body irradiation with a boost to the left brachial plexus provided marked improvement in her symptoms, with complete resolution of the chloromas on imaging. She underwent allogeneic bone marrow transplantation that was complicated by graft vs host disease. One year later the patient was found to have imaging recurrence of the left brachial plexus chloroma without clinical worsening. This resolved completely following repeated local irradiation. She is now believed to be in complete remission.
Figure 2
Postgadolinium fat suppressed T1-weighted image enhancement of the right lumbosacral trunk (arrows)
DISCUSSION In this patient the initial EMG and NCS study showed normal left medial antebrachial cutaneous sensory response and needle EMG of the left abductor pollicis brevis. This led to the diagnosis of UNE. Plexopathies may initially present with selective fascicular involvement or sparing that may lead to a misdiagnosis of a peripheral nerve lesion, such as ulnar neuropathy as in this case. The FCU weakness with radiating pain up and down the medial arm and forearm was a clue to plexopathy in this case since the FCU is usually spared in UNE. The false reassurance provided by evidence of hematologic remission at neurologic presentation also led to missed consideration of the diagnosis of plexopathy. The explanation for the apparent ulnar neuropathy in this case of lower trunk brachial plexopathy resides in the nature of the plexus lesion. Extramedullary myeloid tumors (EMT) are depositions of blasts outside the blood vessels. They can compress or infiltrate surrounding tissue as in the lower brachial plexus in this patient’s case, resulting in progressive symptoms as different fascicles are compressed and infiltrated earlier than others. Peripheral nervous system involvement in acute myeloid leukemia has different manifestations. Most commonly, leptomeningeal metastasis with nerve root involvement results in radiculopathy.3 Plexus compression by EMT is rare4 as is infiltration of peripheral nerves resulting in mononeuritis multiplex.5 Gadolinium enhanced MRI of the spine is useful for showing radicular infiltrations, while mononeuritis can be diagnosed by biopsy of the involved nerve. Chloromas, more correctly called EMT, were first Neurology 72
January 13, 2009
e9
described by Burns in 1811. They occur most frequently in AML type M2, M4, and M5, the latter being the type in our case.6 Although these tumors commonly have an indolent course, they can present as true neurologic emergencies. Controversies exist regarding whether EMT affect the prognosis of AML.7 The diagnosis can be suspected clinically if the lesion is superficial as in a cutaneous chloroma or orbital myeloblastoma. EMT can be seen on CT and MRI.8 These tumors usually respond rapidly to irradiation. However, they often recur. Other treatment options include surgical decompression, IV chemotherapy, or any combination of these treatments. There are no studies showing superiority of any treatment modality. The blood–nerve barrier shares some similarities with the blood– brain barrier.9 It has a protective role toward the endoneurium, isolating it form the extracellular fluid, making metastases to peripheral nerves a rare incident.10 At the same time, malignant cells that have succeeded in infiltrating this barrier would be protected from systemic chemotherapy. Malignant cells can then nest in the peripheral nervous system and relapse after treatment of the primary tumor, often preceding hematologic relapse.5 In summary, chloromatous infiltration of plexi and nerves should be considered in patients with AML who have neurologic symptoms, even in the presence of hematologic remission.
e10
Neurology 72
January 13, 2009
REFERENCES 1. Phan TG, Evans BA, Huston J. Pseudoulnar palsy from a small infarct of the precentral knob. Neurology 2000;54: 2185. 2. Campbell WW, Pridgeon RM, Riaz G, Astruc J, Leahy M, Crostic EG. Sparing of the flexor carpi ulnaris in ulnar neuropathy at the elbow. Muscle Nerve 1989;Dec 12: 965–967. 3. Anuradha S, Singh NP, Anand KS, Prasad A. A rare case of radiculopathy. Postgrad Med J 1999;75:53–55. 4. Stork JT, Cigtay OS, Schellinger D, Jacobson RJ. Recurrent chloromas in acute myelogenous leukemia. AJR Am J Roentgenol 1984;142:777–778. 5. Lekos A, Katirji MB, Cohen ML, Weisman R Jr, Harik SI. Mononeuritis multiplex: a harbinger of acute leukemia in relapse. Arch Neurol 1994;51:618–622. 6. Byrd JC, Edenfield WJ, Shields DJ, Dawson NA. Extramedullary myeloid cell tumors in acute nonlymphocytic leukemia: a clinical review. J Clin Oncol 1995;13:1800– 1816. 7. Bisschop MM, Revesz T, Bierings M, et al. Extramedullary infiltrates at diagnosis have no prognostic significance in children with acute myeloid leukemia. Leukemia 2001;15: 46–49. 8. Guermazi A, Feger C, Rousselot P, et al. Granulocytic sarcoma (chloroma): imaging findings in adults and children. AJR Am J Roentgenol 2002;178:319–325. 9. Sano Y, Shimizu F, Nakayama H, et al. Endothelial cells constituting blood-nerve barrier have highly specialized characteristics as barrier-forming cells. Cell Struct Funct 2007;2:139–147. 10. Meller I, Alkalay D, Mozes M, Geffen DB, Ferit T. Isolated metastases to peripheral nerves. Cancer 1995;76: 1829–1832.
Correspondence
J. CLIFFORD RICHARDSON AND 50 YEARS OF PROGRESSIVE SUPRANUCLEAR PALSY
To the Editor: The historical neurology article concerning J. Clifford Richardson and progressive supranuclear palsy (PSP) was a delightful account of the early history of this clinical syndrome.1 However, the authors’ conclusion that the clinical findings of the described syndrome be designated “Richardson disease” is a step backward for 21st century neurologists. Despite our specialty’s fondness for eponyms, we must keep pace and move forward with the current and future findings from molecular biology. Although I share the desire to honor outstanding clinical neurologists who identified new and different clinical constellations of symptoms and signs in their time, it does not seem to me that we have furthered the education of neurologists in training or advanced our knowledge by promoting these ultimately false classifications. Although far less romantic, it would be more accurate to describe “Richardson syndrome” and all of the other subgroupings of PSP by their ultimate— yet to be discovered— genetic classifications. William J. Weiner, Baltimore, MD Disclosure: The author reports no disclosures.
Reply from the Authors: Eponyms are about knowledge. They honor the past, train us in observation, define our practice, and enrich the art of medicine. No wonder neurologists have a fondness for them. For example, patients agree that “Parkinson disease” is a better term for their disease than “the shaking palsy.” And what greater understanding or comfort could “synucleinopathy” bring to a family stricken with it? For such understanding, better it be known as “the shaking palsy.” But best of all is Parkinson disease, a unifying designation for an idiopathic illness that affects many patients and is the research passion of diverse neuroscientists. It is a comfortable term for a dreadful disease, and it permits all scientific disciplines to refine it by their separate observations and interpretation. There is no better designation for this disease at this time. The same is true for diseases experienced by Lou Gehrig
and described by Alzheimer and Pick, diseases which also await definition by molecular biologists and geneticists. Fifty years ago, J. Clifford Richardson, a Canadian neurologist in Toronto, recognized a handful of patients with a unique clinical syndrome as British neurologist James Parkinson had identified in London many years before. Richardson’s observation in 1963— coupled with further laboratory findings— have elucidated the clinical syndrome that he first identified as universal and common and its disease may include phenotypes with parkinsonism and freezing. In 1963, Dr. Jerzy Olszewski, a master of neuroanatomy, and I (his resident at the time) identified the unique neuropathology of Richardson syndrome and each of us revealed this clinical–pathological syndrome to our colleagues.2– 4 Olszewski and Richardson agreed that I should be the first author of the article that we published together in 1964 as PSP.5 The order of authorship led Andre Barbeau in 1965 to suggest that it also be known as the SteeleRichardson-Olszewski syndrome (SRO).6 That was an honor for me but it missed the point that the original observation that led to our description was by Richardson. By our recent article and this letter, I appreciate the opportunity to suggest that PSP also be known as Richardson disease. I have no doubt Professor Olszewski would agree and understand that we will be equally remembered by history. And I hope others— including Professor Weiner—will also agree and adopt this worthy eponym to honor J. Clifford Richardson’s seminal observation and the knowledge it brings to neurology. John C. Steele, David R. Williams, Andrew J. Lees, John R. Wherrett, Tamuning, Guam Disclosure: The authors report no disclosures. Copyright © 2009 by AAN Enterprises, Inc. 1.
2.
Williams DR, Lees AJ, Wherrett JR, Steele JC. J. Clifford Richardson and 50 years of progressive supranuclear palsy. Neurology 2008;70:566 –573. Richardson JC, Steele J, Olszewski J. Supranuclear ophthalmoplegia, pseudobulbar palsy, nuchal dystonia and dementia: a clinical report on eight cases of “heterogeneous
Neurology 72
January 13, 2009
199
3.
4.
5.
6.
system degeneration.” Trans Am Neurol Assoc 1963;8: 25–29. Steele J, Richardson JC, Olszewski J. Heterogeneous system degeneration: a clinical report of eight cases. Can Neurol Assoc Prog 1963;14 –15. Olszewski J, Steele J, Richardson JC. Pathological report on six cases of heterogeneous system degeneration. J Neuropathol Exp Neurol 1963;23:87–188. Steele JC, Richardson JC, Olszewski J. Progressive supranuclear palsy; a heterogeneous degeneration involving the brain stem, basal ganglia and cerebellum with vertical gaze and pseudobulbar palsy, nuchal dystonia and dementia. Arch Neurol 1964;10:333–359. Barbeau A. Degenerescence plurisystematisee du nevraxe. Syndrome de Steele-Richardson-Olszewski. Un Med Can 1965;94:15–718.
REVERSAL OF TRANSTENTORIAL HERNIATION WITH HYPERTONIC SALINE
To the Editor: Over the years, The Hopkins Neurocritical Care group has carefully studied the effect of hypertonic solutions clinically and experimentally. The recent report by Koenig et al.1 invites the following comments. Hypertonic saline or mannitol could improve neurologic examination in a patient with a new mass. We and others have observed rapid dramatic changes with osmotic agents. For example, we have observed dilated pupils and extensor responses at baseline, but soon after injection, pupils normalize and improvement to localization to noxious stimuli is seen. Koenig et al. and the editorialists suggest the immediacy of such a response is reversal of transtentorial herniation (TTH), but evidence of that is not provided; there are few clinical details except a GCS sumscore and no CT or MR studies. We are not certain that the clinical improvement is due to decompression or dislodging of tissue squeezed through the tentorium. In fact, when the effects are studied in mannitol, change in midline shift is not observed as changes are too minimal or almost too subtle to catch the eye on serial MRI.2 Many consider that osmotic diuresis cannot physiologically explain the rapid improvement (within 30 minutes or less in many patients). Alternatively, sudden hypertonicity may cause vasodilatation of the cerebral resistance vessels and with an intact autoregulation result in vasoconstriction, reduced cerebral volume, and decrease in intracranial pressure. Even that explanation—albeit plausible—may be based on some assumptions including increased intracranial pressure and intact autoregulation. Perhaps a simpler explanation is a suddenly improved blood flow and oxygenation to the displaced ischemic upper brainstem.3,4 Cerebral perfusion improved substantially within 30 minutes in subarachnoid hemorrhage in a recent study followed by later 200
Neurology 72
January 13, 2009
reduction of ICP, although admittedly a larger saline bolus was used in that study.5 Koenig et al. excluded 62 patients with no serial neurologic examination available for review who were treated with hypertonic saline. It would be interesting to know their outcome, and if poor, could in some way further diminish their claim that hypertonic saline results in successful reversal in a high proportion of patients. Documentation in these cases may have been less detailed because patients did not initially do well in the first place. Finally, we would like to point out that hypertonic saline—typically injected with a 30 cc syringe—requires a central venous access catheter because severe thrombophlebitis occurs with peripheral infusion. This could delay emergent management and its placement has the potential for complications such as pneumothorax. This is an issue when such an intervention is considered in patients in stroke units, in the emergency department, or even in those en route to the hospital. Alexander Y. Zubkov, MD, PhD, Eelco F.M. Wijdicks, MD, Rochester, MN Disclosure: The authors report no disclosures.
Reply from the Authors: We appreciate the observations of Drs. Wijdicks and Zubkov. TTH refers both to a clinical syndrome and an anatomic entity. Radiographic evidence of TTH may be observed in the absence of clinical signs and vice versa.6,7 We evaluated patients with a clinical diagnosis of TTH, defined as the sudden onset of a dilated, unresponsive pupil and decline in consciousness, which are widely accepted criteria.8,9 While we found that 23.4% saline was associated with reversal of clinical evidence of TTH, our data do not elucidate the mechanisms underlying these changes. Given the diverse effects of hypertonic saline on intracranial physiology, it is possible that factors other than mechanical decompression contributed to the observed effects. We measured displacement of midline structures on CT scans obtained before and after TTH. However, we did not include these data because few scans were performed in the interval after herniation was recognized but before administration of 23.4% saline. In order to evaluate the effect of 23.4% saline on TTH, we selected patients who met strict clinical criteria and who had adequate documentation of the neurological examination before and after administration. We excluded 62 patients with documentation that was insufficient to determine whether TTH occurred or was reversed. Our colleagues suggest that documentation in the excluded patients might have been inadequate because these patients were refrac-
Correspondence
J. CLIFFORD RICHARDSON AND 50 YEARS OF PROGRESSIVE SUPRANUCLEAR PALSY
To the Editor: The historical neurology article concerning J. Clifford Richardson and progressive supranuclear palsy (PSP) was a delightful account of the early history of this clinical syndrome.1 However, the authors’ conclusion that the clinical findings of the described syndrome be designated “Richardson disease” is a step backward for 21st century neurologists. Despite our specialty’s fondness for eponyms, we must keep pace and move forward with the current and future findings from molecular biology. Although I share the desire to honor outstanding clinical neurologists who identified new and different clinical constellations of symptoms and signs in their time, it does not seem to me that we have furthered the education of neurologists in training or advanced our knowledge by promoting these ultimately false classifications. Although far less romantic, it would be more accurate to describe “Richardson syndrome” and all of the other subgroupings of PSP by their ultimate— yet to be discovered— genetic classifications. William J. Weiner, Baltimore, MD Disclosure: The author reports no disclosures.
Reply from the Authors: Eponyms are about knowledge. They honor the past, train us in observation, define our practice, and enrich the art of medicine. No wonder neurologists have a fondness for them. For example, patients agree that “Parkinson disease” is a better term for their disease than “the shaking palsy.” And what greater understanding or comfort could “synucleinopathy” bring to a family stricken with it? For such understanding, better it be known as “the shaking palsy.” But best of all is Parkinson disease, a unifying designation for an idiopathic illness that affects many patients and is the research passion of diverse neuroscientists. It is a comfortable term for a dreadful disease, and it permits all scientific disciplines to refine it by their separate observations and interpretation. There is no better designation for this disease at this time. The same is true for diseases experienced by Lou Gehrig
and described by Alzheimer and Pick, diseases which also await definition by molecular biologists and geneticists. Fifty years ago, J. Clifford Richardson, a Canadian neurologist in Toronto, recognized a handful of patients with a unique clinical syndrome as British neurologist James Parkinson had identified in London many years before. Richardson’s observation in 1963— coupled with further laboratory findings— have elucidated the clinical syndrome that he first identified as universal and common and its disease may include phenotypes with parkinsonism and freezing. In 1963, Dr. Jerzy Olszewski, a master of neuroanatomy, and I (his resident at the time) identified the unique neuropathology of Richardson syndrome and each of us revealed this clinical–pathological syndrome to our colleagues.2– 4 Olszewski and Richardson agreed that I should be the first author of the article that we published together in 1964 as PSP.5 The order of authorship led Andre Barbeau in 1965 to suggest that it also be known as the SteeleRichardson-Olszewski syndrome (SRO).6 That was an honor for me but it missed the point that the original observation that led to our description was by Richardson. By our recent article and this letter, I appreciate the opportunity to suggest that PSP also be known as Richardson disease. I have no doubt Professor Olszewski would agree and understand that we will be equally remembered by history. And I hope others— including Professor Weiner—will also agree and adopt this worthy eponym to honor J. Clifford Richardson’s seminal observation and the knowledge it brings to neurology. John C. Steele, David R. Williams, Andrew J. Lees, John R. Wherrett, Tamuning, Guam Disclosure: The authors report no disclosures. Copyright © 2009 by AAN Enterprises, Inc. 1.
2.
Williams DR, Lees AJ, Wherrett JR, Steele JC. J. Clifford Richardson and 50 years of progressive supranuclear palsy. Neurology 2008;70:566 –573. Richardson JC, Steele J, Olszewski J. Supranuclear ophthalmoplegia, pseudobulbar palsy, nuchal dystonia and dementia: a clinical report on eight cases of “heterogeneous
Neurology 72
January 13, 2009
199
3.
4.
5.
6.
system degeneration.” Trans Am Neurol Assoc 1963;8: 25–29. Steele J, Richardson JC, Olszewski J. Heterogeneous system degeneration: a clinical report of eight cases. Can Neurol Assoc Prog 1963;14 –15. Olszewski J, Steele J, Richardson JC. Pathological report on six cases of heterogeneous system degeneration. J Neuropathol Exp Neurol 1963;23:87–188. Steele JC, Richardson JC, Olszewski J. Progressive supranuclear palsy; a heterogeneous degeneration involving the brain stem, basal ganglia and cerebellum with vertical gaze and pseudobulbar palsy, nuchal dystonia and dementia. Arch Neurol 1964;10:333–359. Barbeau A. Degenerescence plurisystematisee du nevraxe. Syndrome de Steele-Richardson-Olszewski. Un Med Can 1965;94:15–718.
REVERSAL OF TRANSTENTORIAL HERNIATION WITH HYPERTONIC SALINE
To the Editor: Over the years, The Hopkins Neurocritical Care group has carefully studied the effect of hypertonic solutions clinically and experimentally. The recent report by Koenig et al.1 invites the following comments. Hypertonic saline or mannitol could improve neurologic examination in a patient with a new mass. We and others have observed rapid dramatic changes with osmotic agents. For example, we have observed dilated pupils and extensor responses at baseline, but soon after injection, pupils normalize and improvement to localization to noxious stimuli is seen. Koenig et al. and the editorialists suggest the immediacy of such a response is reversal of transtentorial herniation (TTH), but evidence of that is not provided; there are few clinical details except a GCS sumscore and no CT or MR studies. We are not certain that the clinical improvement is due to decompression or dislodging of tissue squeezed through the tentorium. In fact, when the effects are studied in mannitol, change in midline shift is not observed as changes are too minimal or almost too subtle to catch the eye on serial MRI.2 Many consider that osmotic diuresis cannot physiologically explain the rapid improvement (within 30 minutes or less in many patients). Alternatively, sudden hypertonicity may cause vasodilatation of the cerebral resistance vessels and with an intact autoregulation result in vasoconstriction, reduced cerebral volume, and decrease in intracranial pressure. Even that explanation—albeit plausible—may be based on some assumptions including increased intracranial pressure and intact autoregulation. Perhaps a simpler explanation is a suddenly improved blood flow and oxygenation to the displaced ischemic upper brainstem.3,4 Cerebral perfusion improved substantially within 30 minutes in subarachnoid hemorrhage in a recent study followed by later 200
Neurology 72
January 13, 2009
reduction of ICP, although admittedly a larger saline bolus was used in that study.5 Koenig et al. excluded 62 patients with no serial neurologic examination available for review who were treated with hypertonic saline. It would be interesting to know their outcome, and if poor, could in some way further diminish their claim that hypertonic saline results in successful reversal in a high proportion of patients. Documentation in these cases may have been less detailed because patients did not initially do well in the first place. Finally, we would like to point out that hypertonic saline—typically injected with a 30 cc syringe—requires a central venous access catheter because severe thrombophlebitis occurs with peripheral infusion. This could delay emergent management and its placement has the potential for complications such as pneumothorax. This is an issue when such an intervention is considered in patients in stroke units, in the emergency department, or even in those en route to the hospital. Alexander Y. Zubkov, MD, PhD, Eelco F.M. Wijdicks, MD, Rochester, MN Disclosure: The authors report no disclosures.
Reply from the Authors: We appreciate the observations of Drs. Wijdicks and Zubkov. TTH refers both to a clinical syndrome and an anatomic entity. Radiographic evidence of TTH may be observed in the absence of clinical signs and vice versa.6,7 We evaluated patients with a clinical diagnosis of TTH, defined as the sudden onset of a dilated, unresponsive pupil and decline in consciousness, which are widely accepted criteria.8,9 While we found that 23.4% saline was associated with reversal of clinical evidence of TTH, our data do not elucidate the mechanisms underlying these changes. Given the diverse effects of hypertonic saline on intracranial physiology, it is possible that factors other than mechanical decompression contributed to the observed effects. We measured displacement of midline structures on CT scans obtained before and after TTH. However, we did not include these data because few scans were performed in the interval after herniation was recognized but before administration of 23.4% saline. In order to evaluate the effect of 23.4% saline on TTH, we selected patients who met strict clinical criteria and who had adequate documentation of the neurological examination before and after administration. We excluded 62 patients with documentation that was insufficient to determine whether TTH occurred or was reversed. Our colleagues suggest that documentation in the excluded patients might have been inadequate because these patients were refrac-
tory to treatment, thus introducing a selection bias. To address this concern, we reanalyzed our data to determine the outcomes of the 62 excluded patients. If the excluded patients were treatment failures, one would expect a higher mortality rate than in the analyzed group. However, the in-hospital mortality for the excluded patients was 60% (37/62), which was slightly lower than the population included in the study (68%), arguing against therapeutic failure in the patients with incomplete documentation. Brain herniation is a medical emergency analogous to cardiopulmonary arrest and delaying treatment in order to obtain central venous access is unacceptable. Thus, 23.4% may only be used as a first-line hyperosmolar agent in patients with a preexisting central venous catheter. In others, a peripherally compatible hyperosmolar agent such as mannitol should be given initially, until central access has been established. In our Neurosciences Critical Care Unit this has rarely been an issue, since a high proportion of patients who experience herniation already have central venous access.
1.
Robert D. Stevens, Matthew A. Koenig, Baltimore, MD
8.
Disclosure: The authors report no disclosures. Copyright © 2009 by AAN Enterprises, Inc.
2.
3.
4.
5.
6.
7.
9.
Koenig MA, Bryan M, Lewin JL 3rd, et al. Reversal of transtentorial herniation with hypertonic saline. Neurology 2008;70:1023–1029. Videen TO, Zazulia AR, Manno EM. Mannitol bolus preferentially shrinks noninfarcted brain in patients with ischemic stroke. Neurology 2001:57:2120 –2122. Ritter A, Muizelaar J, Barnes T, et al. Brain stem blood flow, pupillary response, and outcome in patients with severe head injuries. Neurosurgery 1999;44:941–948. Wijdicks EFM. Acute brainstem displacement without uncal herniation and posterior cerebral artery injury. J Neurosurg Neurol Psychiatry 2008 (in press). Tseng MY, Al-Rawi PG, Czosnyka M, et al. Enhancement of cerebral blood flow using systemic hypertonic saline therapy improves outcome with poor-grade spontaneous subarachnoid hemorrhage. J Neurosurg 2007: 274 –284. Reich JB, Sierra J, Camp W, Zanzonico P, Deck MD, Plum F. Magnetic resonance imaging measurements and clinical changes accompanying transtentorial and foramen magnum brain herniation. Ann Neurol 1993;33:159 – 170. Ropper AH. Lateral displacement of the brain and level of consciousness in patients with an acute hemispheral mass. N Engl J Med 1986;314:953–958. Meyer A. Herniation of the brain. Arch Neurol Psychiatry 1920;4:387– 400. Posner JB, Plum F. Plum and Posner’s Diagnosis of Stupor and Coma. 4th ed. New York: Oxford University Press; 2007.
CORRECTION Lamotrigine extended-release as adjunctive therapy for partial seizures In the article “Lamotrigine extended-release as adjunctive therapy for partial seizures” by D.K. Naritoku et al. (Neurology® 2007;69:1610 –1618), as a result of changes to the database from audits conducted across all participating study sites, some of the data reported in the manuscript have changed slightly. These audits were conducted by the study sponsor, GlaxoSmithKline, in preparation for response to the FDA approvable letter for the Lamictal XR New Drug Application. The outcome of the audits did not change the overall conclusions of the study, but did lead to small changes in some of the data in the manuscript. The new data have replaced the previous numbers published in the abstract, now in brackets: ABSTRACT Objective: To evaluate the efficacy and tolerability of once-daily adjunctive lamotrigine extended-release (XR) for partial seizures in epilepsy. Methods: Patients more than 12 years old diagnosed with epilepsy with partial seizures and taking one to two baseline antiepileptic drugs were randomized to adjunctive once-daily lamotrigine XR or placebo in a double-blind, parallel-group trial. The study comprised a baseline phase, a 7-week double-blind escalation phase, and a 12-week double-blind maintenance phase during which doses of study medication and concomitant antiepileptic drugs were maintained. Results: Of the 243 randomized patients, 239 (118 lamotrigine XR, 121 placebo) entered the escalation phase and received study medication. Lamotrigine XR was more effective than placebo with respect to median percent reduction from baseline in weekly partial seizure frequency (primary endpoint— entire 19-week treatment phase: 46.6% [was 46.1%] vs 24.5% [was 24.2%], p ⫽ 0.0001 [was 0.0004] via Wilcoxon test; escalation phase: 29.8% [was 28.0%] vs 15.6% [was 16.3%], p ⫽ 0.027 [was 0.028]; maintenance phase: 58.4% [was 58.0%] vs 26.8% [was 26.7%], p ⬍ 0.0001). The percentage of patients with ⱖ50% reduction in partial seizure frequency (44.0% [was 42.2%] vs 20.8% [was 24.2%], p ⫽ 0.0002 [was 0.0037]) and time to ⱖ50% reduction in partial seizure frequency ( p ⫽ 0.0001) (was 0.0007) also favored lamotrigine XR over placebo. A similar pattern of results was observed for secondarily generalized seizures. The most common adverse events were headache (lamotrigine XR 16% [was 17%], placebo 18% [was 15%]) and dizziness (lamotrigine XR 19% [was 18%], placebo 5%). Differences between lamotrigine XR and placebo on health outcomes measures were not significant. Conclusions: Once-daily adjunctive lamotrigine extended-release compared with placebo effectively reduced partial seizure frequency and was well tolerated in this double-blind study. Results support the clinical utility of this new once-daily formulation.
Neurology 72
January 13, 2009
201
Section Editors Christopher J. Boes, MD Kenneth J. Mack, MD, PhD
Book Review
BEHAVIORAL ASPECTS OF EPILEPSY: PRINCIPLES AND PRACTICE
by Steven C. Schachter, Gregory Holmes, and Dorothee Kasteleijn-nolse, 519 pp., New York, Demos, 2007, $119 The behavioral, social, and psychiatric disorders seen with epilepsy are receiving increasing attention. Providing holistic care for patients with epilepsy extends beyond just the treatment of seizures, to include assessment and therapy for these other comorbidities. Behavioral Aspects of Epilepsy: Principles and Practice is authored by an international panel of experts including physicians, psychologists, and behavioral scientists who are well recognized for their expertise in this area. One of the major strengths of this book is its benchto-bedside translational approach. The first part is focused on basic science. A section on animal models reviews the neurobehavioral data from animals with seizures, addressing the cognitive and behavioral impact of seizures on both the developing and mature brain. The impact of antiepileptic medications on attention, working memory, and learning in the animal model is also addressed. A section on mechanisms underlying epilepsy and behavior explores the neurophysiology of seizures and the impact of epileptogenesis, sprouting, and neuronal excitability on behavior.
The larger part of this book is devoted to clinical science, and includes concise reviews of memory, cognition, and language in persons with epilepsy and the impact of epilepsy surgery on neuropsychological outcome. The section on neuropsychiatric disorders is extremely well written, providing thorough reviews on diagnostic and therapeutic options for anxiety, depression, psychosis, and personality disorders in epilepsy. The common and often complex issue of psychogenic nonepileptic events is reviewed, and a practical approach to this problem is proposed. One specific section is devoted to pediatric and adolescent epilepsy and reviews key topics such as the impact of specific epilepsy syndromes or antiepileptic drugs on behavior, the effect of family factors, and the impact of epilepsy on academic and social competence. This well-written, comprehensive book provides an authoritative summary of the important advances in both basic and clinical science that have been made in this area, and is a “must-read” for physicians, psychologists, and other clinicians who care for patients with epilepsy. Reviewed by Elaine Wirrell, MD Disclosure: The author reports no disclosures. Copyright © 2009 by AAN Enterprises, Inc.
Did You Know. . . . . .you can browse by subspecialty topics on www.neurology.org? Go to: http://www.neurology.org/collections and click on the specific topic for a complete list of articles.
202
Neurology 72
January 13, 2009
Calendar
2009 Neurology® publishes short announcements of meetings and courses related to the field. Items must be received at least 6 weeks before the first day of the month in which the initial notice is to appear. Send Calendar submissions to Calendar, Editorial Office, Neurology®, Suite 214, 20 SW 2nd Ave., P.O. Box 178, Rochester, MN 55903
[email protected]
JAN. 16 –18 AAN Winter Conference will be held at Disney Contemporary Resort in Orlando, FL. American Academy of Neurology: tel (800) 879-1960; www.aan.com/winter. FEB. 9 –11 Case Studies in Epilepsy Surgery will be held at the Silver Tree and Snowmass Conference Center in Snowmass, CO. Contact Martha Tobin at (216) 445-3449 or (800) 2232273, ext 53449, or at
[email protected] for seminar details. FEB. 9 –13 The 22nd Annual Practicing Physician’s Approach to the Difficult Headache Patient will be held at the Camelback Inn, Scottsdale, AZ. Approved for AMA PRA Category 1 credit. Diamond Headache Clinic Research & Educational Foundation: tel (877) 706-6363 or (733) 883-2062;
[email protected]; www.dhc-fdn.org. FEB. 16 –17 Fifth Annual Update Symposium on Clinical Neurology and Neurophysiology will be held in Tel Aviv, Israel. Presented by Weill Cornell Medical College, Department of Neurology, and Tel Aviv University, Adams Brain Supercenter. www.neurophysiology-symposium.com. FEB. 20 –22 International Symposium on Stereotactic Body Radiation Therapy and Stereotactic Radiosurgery will be held at the Floridian Resort & Spa in Lake Buena Vista, FL. Contact Martha Tobin at (216) 445-3449 or (800) 223-2273, ext 53449, or at
[email protected] for seminar details. APR. 2– 4 The Innsbruck Colloquium on Status Epilepticus 2009 will be held at the Congress Innsbruck, Austria.
[email protected]; www.innsbruck-SE2009.eu. APR. 3 5th Annual Contemporary Issues in Pituitary: Casebase Management Update will be held at the Cleveland Clinic Lerner Research Institute in Cleveland, OH. Contact Martha Tobin at (216) 445-3449 or (800) 223-2273, ext 53449, or at
[email protected] for seminar details. APR. 20 –22 Leksell Gamma Knife® Perfexion™ Upgrade Course will be held at the Gamma Knife Center in Cleveland, OH. Contact Martha Tobin at (216) 445-3449 or (800) 2232273, ext 53449, or at
[email protected] for seminar details. APR. 25–MAY 2 AAN Annual Meeting will be held in Seattle, Washington State Convention & Trade Center, WA. American Academy of Neurology: tel (800) 879-1960; www.aan.com/am. MAY 3– 6 2nd International Epilepsy Colloquium, Pediatric Epilepsy Surgery Cite´ Internationale will be held in Lyon, France. http://epilepsycolloquium2009ams.fr.
MAY 6 –10 International SFEMG Course and Xth Quantitative EMG conference will be held in Venice, Italy. tel 39041-951112;
[email protected]; www.congressvenezia.it. MAY 8 The Office of Continuing Medical Education at the University of Michigan Medical School is sponsoring a CME conference entitled: Movement Disorders: A Practical Approach. It is located at The Inn at St. John’s in Plymouth, Michigan. tel (734) 763-1400; fax (734) 936-1641. MAY 11–12 Music and the Brain will be held at the InterContinental Hotel & Bank of America Conference Center in Cleveland, OH. Contact Martha Tobin at (216) 445-3449 or (800) 223-2273, ext 53449, or at
[email protected] for seminar details. MAY 15–17 The Fifth International Conference on Alzheimer’s Disease and Related Disorders in the Middle East will be held in Limassol, Cyprus. www.worldeventsforum.com/alz. MAY 28 –30 6th International Headache Seminary. Focus on Headaches: New Frontier in Mechanisms and Management will be held at the Grand Hotel des Iles Borromees in Stresa (Italy); tel/fax 02 7063 8067;
[email protected]. JUN. 8 –12 Leksell Gamma Knife® Perfexion™ Introductory Course will be held at the Gamma Knife Center in Cleveland, OH. Contact Martha Tobin at (216) 445-3449 or (800) 223-2273, ext 53449, or at
[email protected] for seminar details. JUN. 12 Mellen Center Regional Symposium on Multiple Sclerosis will be held at the InterContinental Hotel & Bank of America Conference Center in Cleveland, OH. Contact Martha Tobin at (216) 445-3449 or (800) 223-2273, ext 53449, or at
[email protected] for seminar details. JUN. 19 –24 Epileptology Symposium will be held at the InterContinental Hotel & Bank of America Conference Center, in Cleveland, OH. Contact Martha Tobin at (216) 445-3449 or (800) 223-2273, ext 53449, or at
[email protected] for seminar details. JUL. 7–10 SickKids Centre for Brain & Behaviour International Symposium.
[email protected]; www.sickkids.ca/ learninginstitute. JUL. 16 –18 Mayo Clinic Neurology in Clinical Practice2009 will be held at the InterContinental Hotel, Chicago, IL. Mayo CME: tel: (800) 323-2688;
[email protected]; http:// www.mayo.edu/cme/neurology-neurologic-surgery.html. SEP. 12–15 13th Congress of the European Federation of Neurological Societies will be held in Florence, Italy. For more information: tel ⫹41 22 908 0488; http://www.kenes.com/efns2009/ index.asp;
[email protected]. Neurology 72
January 13, 2009
203
SEP. 25 Practical Pearls in Neuro-Ophthalmology–International Symposium in Honour of Dr. James Sharpe will be held on September 25, 2009 at the University of Toronto Conference Centre, Toronto, Ontario. For further information contact the Office of Continuing Education & Professional Development, Faculty of Medicine, University of Toronto: tel (416) 978-2719; (888) 5128173; fax (416) 946-7028;
[email protected]; http:// events.cmetoronto.ca/website/index/OPT0907.
OCT. 8 –11 The Third World Congress on Controversies in Neurology. Full information is available at: ComtecMed - Med-
ical Congresses, PO Box 68, Tel-Aviv, 61000 Israel; tel ⫹972– 3-5666166; fax ⫹972–3-5666177; cony@comtecmed. com; www.comtecmed.com/cony. OCT. 24 –30 19th World Congress of Neurology, WCN 2009, will be held in Bangkok, Thailand. www.wcn2009bangkok.com. NOV. 19 –22 The Sixth International Congress on Vascular Dementia will be held Barcelona, Spain. For further details, please contact: Kenes International 17 Rue du Cendrier, P.O. Box 1726, CH-1211, Geneva 1, Switzerland; tel ⫹41 22 908 0488; fax ⫹41 22 732 2850;
[email protected]; http://www. kenes.com/vascular.
Save These Dates for AAN CME Opportunities! Mark these upcoming dates on your calendar for these exciting continuing education opportunities, where you can catch up on the latest neurology information. AAN Annual Meetings ● April 25—May 2, 2009, Seattle, Washington State Convention & Trade Center ● April 10 –17, 2010, Toronto, Ontario, Canada AAN Regional Conference ● January 16 –18, 2009, Disney Contemporary Resort in Orlando, FL
204
Neurology 72
January 13, 2009
In the next issue of Neurology® Volume 72, Number 3, January 20, 2009 www.neurology.org THE MOST WIDELY READ AND HIGHLY CITED PEER-REVIEWED NEUROLOGY JOURNAL
THIS WEEK IN Neurology®
205
268
Highlights of the January 20 issue
EDITORIALS
206
208
Epilepsy surgery patients with cortical dysplasia: Present and future therapeutic challenges Gary W. Mathern
210
Sydney S. Schochet, Jr., MD (1937–2008) Ludwig Gutmann and Jack E. Riggs
ARTICLES
211
Predictors of surgical outcome and pathologic considerations in focal cortical dysplasia D.W. Kim, S.K. Lee, K. Chu, K.I. Park, et al.
217
Incomplete resection of focal cortical dysplasia is the main predictor of poor postsurgical outcome P. Krsek, B. Maton, P. Jayakar, P. Dean, et al.
224
Identification of a possible pathogenic link between congenital long QT syndrome and epilepsy J.N. Johnson, N. Hofman, C.M. Haglund, et al.
232
VIEWS & REVIEWS
273
Seizures and arrhythmias: Differing phenotypes of a common channelopathy? Jill V. Hunter and Arthur J. Moss
IN MEMORIAM
Incidence of acquired demyelination of the CNS in Canadian children B. Banwell, J. Kennedy, D. Sadovnick, et al.
240
FBXO7 mutations cause autosomal recessive, early-onset parkinsonian-pyramidal syndrome A. Di Fonzo, M.C.J. Dekker, P. Montagna, et al.
246
X-linked distal hereditary motor neuropathy maps to the DSMAX locus on chromosome Xq13.1-q21 M. Kennerson, G. Nicholson, B. Kowalski, et al.
Plasma A, homocysteine, and cognition: The Vitamin Intervention for Stroke Prevention (VISP) trial A. Viswanathan, S. Raj, S.M. Greenberg, et al.
Genetics of epilepsy syndromes starting in the first year of life L. Deprez, A. Jansen, and P. De Jonghe
CLINICAL IMPLICATIONS OF NEUROSCIENCE RESEARCH
282
␥-Hydroxybutyric acid and its relevance in neurology Eduardo E. Benarroch
CLINICAL/SCIENTIFIC NOTES
287
A case of ALS-FTD in a large FALS pedigree with a K17I ANG mutation M.A. van Es, F.P. Diekstra, J.H. Veldink, F. Baas, et al.
289
Inflammatory pseudotumor associated with HIV, JCV, and immune reconstitution syndrome A. Gonzalez-Duarte, S. Sullivan, G.J. Sips , et al.
VIDEO NEUROIMAGES
291
Ocular flutter as the first manifestation of Lyme disease Jesper Gyllenborg and Dan Milea
RESIDENT & FELLOW SECTION
e11
Clinical Reasoning: A case of Wegener granulomatosis complicated by seizures and headaches: Curiouser and curiouser G. Gorman, M. Hutchinson, and N. Tubridy
CORRESPONDENCE
292
Transmissible spongiform encephalopathy Intranasal insulin in early AD
253
Personality and lifestyle in relation to dementia incidence H.-X. Wang, A. Karp, A. Herlitz, M. Crowe, et al.
292
260
Autosomal dominant subcortical gliosis presenting as frontotemporal dementia R.H. Swerdlow, B.B. Miller, M.B.S. Lopes, et al.
FUTURE ISSUES
Abstracts In the Next Issue of Neurology®
Subject to change.
THE OFFICIAL JOURNAL OF THE AMERICAN ACADEMY OF NEUROLOGY
