Imaging in SARS

IMAGING IN SARS

IMAGING IN SARS

AT Ahuja MBBS(Bom), MD(Bom), FRCR, FHKCR, FHKAM(Radiology) Professor, Department of Diagnostic Radiology and Organ Imaging The Chinese University of Hong Kong, Prince of Wales Hospital, Shatin Hong Kong, China

CGC Ooi MBBCh, BAO Belf, MRCP(UK), FRCR(UK), FHKCR, FHKAM(Radiology) Associate Professor, Department of Diagnostic Radiology The University of Hong Kong, Queen Mary Hospital Hong Kong, China

London • San Francisco

cambridge university press Cambridge, New York, Melbourne, Madrid, Cape Town, Singapore, São Paulo Cambridge University Press The Edinburgh Building, Cambridge cb2 2ru, UK Published in the United States of America by Cambridge University Press, New York www.cambridge.org Information on this title: www.cambridge.org/9781841102191 © Greenwich Medical Media Limited 2004 This publication is in copyright. Subject to statutory exception and to the provision of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published in print format 2004 isbn-13

978-0-511-54534-4

isbn-13 isbn-10

978-1-841-10219-1 hardback 1-841-10219-9 hardback

OCeISBN

Cambridge University Press has no responsibility for the persistence or accuracy of urls for external or third-party internet websites referred to in this publication, and does not guarantee that any content on such websites is, or will remain, accurate or appropriate.

CONTENTS

Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

vii

Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

xi

1

The Epidemiology of Severe Acute Respiratory Syndrome: A Global . . . . . . . . Perspective M Evans

1

2

The Role of Emergency Medicine in Screening SARS Patients . . . . . . . . . . . . . TH Rainer and P Cameron

17

3

Severe Acute Respiratory Syndrome Outbreak in a University Hospital . . . . . in Hong Kong N Lee and JJY Sung

29

4

Imaging of Pneumonias . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . GC Ooi and PL Khong

33

5

The Role of Chest Radiographs in the Diagnosis of SARS . . . . . . . . . . . . . . . . . KT Wong, GE Antonio, EHY Yuen and AT Ahuja

53

6

Chest Radiography: Clinical Correlation and Its Role in the . . . . . . . . . . . . . . . . Management of Severe Acute Respiratory Syndrome DSC Hui, KT Wong, GE Antonio, AT Ahuja and JJY Sung

61

7

The Role of High-Resolution Computed Tomography in . . . . . . . . . . . . . . . . . . Diagnosis of SARS GE Antonio, KT Wong, DSC Hui and AT Ahuja

69

8

The Role of Imaging in the Follow-up of SARS . . . . . . . . . . . . . . . . . . . . . . . . . . GE Antonio, KT Wong, DSC Hui and AT Ahuja

79

9

Treatment of Severe Acute Respiratory Syndrome . . . . . . . . . . . . . . . . . . . . . . . JJY Sung and AK Wu

89

10

SARS in the Intensive Care Unit . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . GM Joynt, GE Antonio and CD Gomersall

99

11

Imaging of Pneumonia in Children . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . WCW Chu

109

12

Imaging and Clinical Management of Paediatric SARS . . . . . . . . . . . . . . . . . . . WCW Chu, EKL Hon, FWT Cheng and TF Fok

121

vi

C O N T E N T S

13

Imaging of SARS in North America . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . NL Müller and H Shulman

131

14

Radiographers’ Perspective in the Outbreak of SARS . . . . . . . . . . . . . . . . . . . . . SSY Ho Implementation of Measures to Prevent the Spread of SARS in a . . . . . . . . . . Radiology Department AD King and ASC Ching Aftermath of SARS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . GE Antonio, JF Griffith and AT Ahuja Update on Severe Acute Respiratory Syndrome . . . . . . . . . . . . . . . . . . . . . . . . . . AT Ahuja and GE Antonio

143

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

175

15

16 17

149

159 165

CONTRIBUTORS

AT Ahuja MBBS(Bom), MD(Bom), FRCR, FHKCR, FHKAM(Radiology) Professor Department of Diagnostic Radiology and Organ Imaging The Chinese University of Hong Kong Prince of Wales Hospital Shatin Hong Kong China GE Antonio MBBS(UNSW), BSc(Med), FRANZCR Assistant Professor Department of Diagnostic Radiology and Organ Imaging The Chinese University of Hong Kong Prince of Wales Hospital Shatin Hong Kong China P Cameron MBBS, FACEM, MD Professor Accident and Emergency Medicine Academic Unit The Chinese University of Hong Kong Prince of Wales Hospital Shatin Hong Kong China FWT Cheng MBChB, MRCPCH Medical Officer Department of Paediatrics Prince of Wales Hospital Shatin Hong Kong China

ASC Ching MBChB, FRCR, FHKCR, FHKAM(Radiology) Senior Medical Officer Department of Diagnostic Radiology and Organ Imaging Prince of Wales Hospital Shatin Hong Kong China WCW Chu MBChB, FRCR, FHKCR, FHKAM(Radiology) Assistant Professor Department of Diagnostic Radiology and Organ Imaging The Chinese University of Hong Kong Prince of Wales Hospital Shatin Hong Kong China MR Evans MBBCh(Wales), MRCP, FRCP, FFPHM Senior Lecturer Department of Epidemiology, Statistics and Public Health University of Wales College of Medicine Honorary Regional Epidemiologist PHLS Communicable Disease Surveillance Centre (Wales) Abton House, Wedal Road, Cardiff CF14 3QX UK TF Fok MD, DCH(London), FRCP(Edin), FHKAM(Paed), FHKC(Paed), FRCPCH Professor Department of Paediatrics The Chinese University of Hong Kong Prince of Wales Hospital

viii

Shatin Hong Kong China CD Gomersall MBBS, BSc, MRCP(UK), FRCA, EDIC, FFICANZCA, FHKAM(Anaesthesiology), FHKCA Assistant Professor Department of Anaesthesia and Intensive Care The Chinese University of Hong Kong Prince of Wales Hospital Shatin Hong Kong China JF Griffith MBBCh, BAO, MRCP(UK), FRCR, FHKCR, FHKAM(Radiology) Professor Department of Diagnostic Radiology and Organ Imaging Prince of Wales Hospital Shatin Hong Kong China SSY Ho PhD, MPhil, BSc(Hon), PDDR, RDMS, RVT Senior Radiographer Department of Diagnostic Radiology and Organ Imaging Prince of Wales Hospital Shatin Hong Kong China EKL Hon FAAP Assistant Professor Department of Paediatrics The Chinese University of Hong Kong Prince of Wales Hospital Shatin Hong Kong China DSC Hui MBBS, MD(UNSW), MRCP(UK), FRACP, FRCP(Lond, Edin, Glasg), FCCP, FHKCP, FHKAM(Medicine) Associate Professor

C O N T R I B U T O R S

Department of Medicine and Therapeutics The Chinese University of Hong Kong Prince of Wales Hospital Shatin Hong Kong China GM Joynt MBBCh(Witwatersrand), FFA(SA)(CritCare), FHKCA, FHKCA(IC), FHKAM(Anaesthesiology), FFICANZCA, FJFICM, FCCP Associate Professor Department of Anaesthesia and Intensive Care The Chinese University of Hong Kong Prince of Wales Hospital Shatin Hong Kong China PL Khong MBBS(Singapore), FRCR(UK), FHKCR, FHKAM(Radiology) Associate Professor Department of Diagnostic Radiology The University of Hong Kong Queen Mary Hospital Hong Kong China AD King MBBCh, MRCP, FRCR, FHKCR, FHKAM(Radiology) Associate Professor Department of Diagnostic Radiology and Organ Imaging The Chinese University of Hong Kong Prince of Wales Hospital Shatin Hong Kong China N Lee MD Medical Officer Department of Medicine and Therapeutics The Chinese University of Hong Kong Prince of Wales Hospital Shatin Hong Kong China

ix

C O N T R I B U T O R S

NL Müller MD, PhD, FRCPC Professor Department of Radiology Faculty of Medicine The University of British Columbia Vancouver, BC Canada CGC Ooi MBBCh, BAO Belf, MRCP(UK), FRCR(UK), FHKCR, FHKAM(Radiology) Associate Professor Department of Diagnostic Radiology The University of Hong Kong Queen Mary Hospital Hong Kong China

JJY Sung MBBS, MD, PhD, FRCP(London), FRCP(Edin), FRACP, FACG, FHKCP, FHKAM(Medicine) Professor Department of Medicine and Therapeutics The Chinese University of Hong Kong Prince of Wales Hospital Shatin Hong Kong China KT Wong MBChB, FRCR, FHKCR, FHKAM(Radiology) Associate Consultant Department of Diagnostic Radiology and Organ Imaging Prince of Wales Hospital Shatin Hong Kong China

TH Rainer BSc, MBBCh, MRCP, MD, FHKCEM, FHKAM Associate Professor Accident and Emergency Medicine Academic Unit The Chinese University of Hong Kong Prince of Wales Hospital Shatin Hong Kong China

AK Wu MRCP(UK) Medical Officer Department of Medicine and Therapeutics Prince of Wales Hospital Shatin Hong Kong China

H Shulman MD, FRCP(C), FACR Professor Department of Radiology University of Toronto Sunnybrook Hospital Toronto Canada

EHY Yuen MBChB, FRCR, FHKCR, FHKAM(Radiology) Senior Medical Officer Department of Diagnostic Radiology and Organ Imaging Prince of Wales Hospital Shatin Hong Kong China

PREFACE

This book has spawned from the recent outbreak of severe acute respiratory syndrome (SARS) that devastated Hong Kong for a 4-month period during the early half of 2003. The objective of this book is to provide the reader with a quick reference guide to the different facets of this newly emerged disease. All the authors and editors of this book have ‘seen SARS in the face’ and were frontline personnel in the management of this outbreak. This book charts the experiences of clinicians, radiologists and radiographers of this new disease entity not only in Hong Kong but also in Canada. Although this book deals primarily with the radiological management of SARS providing the reader with a gamut of radiographical and high-resolution computed tomography imaging features in both adults and children, it also briefly reviews the epidemiology, clinical management and screening for SARS at the accident and emergency department. The book also discusses in detail, the infection control measures that may be required within a radiology department in order to prevent the transmission of the disease to staff and patients.

SARS highlighted the role of radiology in frontline medicine and emphasized the importance of maintaining our expertise with plain radiography. The department of radiology was at the forefront of the ‘battle’ against SARS and radiographers played a leading role. They demonstrated exceptional professionalism and selflessness in discharging their responsibilities while working in an ultra-high-risk environment and have done themselves proud. We owe a large debt of gratitude to our colleagues and staff of our respective hospitals without whose help this book would not have been possible. On a personal level we would like to acknowledge the close support and help provided by our families (during an extremely stressful time in our lives). For Clara this is in particular, her husband James Griffith and daughters Isobel and Olivia. For Anil, his wife Chu Wai Po, daughter Sanjali, mother Laj Ahuja and late father Dr TS Ahuja. Anil Ahuja Clara Ooi January 2004

The Epidemiology of Severe Acute Respiratory Syndrome: A Global Perspective M Evans

Introduction Epidemiological features Clinical features

1 4 8

Introduction Severe acute respiratory syndrome (SARS) is a newly emerged disease caused by a previously unknown coronavirus [1–3]. It joins a long list of emerging infections [4]. However, unlike other contenders such as avian influenza, Nipah virus, Hendra virus or hantaviruses it has established the capacity for efficient human-to-human transmission and thus poses a major threat to international public health. For this reason, the World Health Organisation (WHO) has described SARS as the first serious and readily transmissible disease to emerge in the 21st century [5].

Key points SARS

• Caused by a previously unknown • •

coronavirus. Established efficient human-to-human transmission. First serious readily transmissible disease in 21st century.

The first known cases of SARS occurred in Guangdong Province in southern China in late November 2002. The first official report of an

Incubation, infectivity and transmission Conclusion

CHAPTER

1 10 15

outbreak of atypical pneumonia in the province on 11 February 2003 indicated that the disease had affected 305 persons and caused five deaths, and that around 30% of cases had occurred in health care workers [6]. On 21 February 2003, a medical doctor infected with SARS travelled from Guangzhou, the provincial capital, and stayed one night at a hotel in Hong Kong. He infected at least 16 other guests and visitors in the hotel. Within days, the disease began spreading around the world along international air travel routes as hotel contacts seeded hospital outbreaks in Hong Kong, Vietnam, Singapore and Canada (Figure 1.1) [7]. Hospital staff, unaware that this was a new, highly infectious disease, exposed themselves to the infection without barrier protection. Subsequently, chains of secondary transmission occurred outside the hospital environment. These initial outbreaks were characterised by rapid increases in the numbers of cases, especially in health care workers and their close contacts (Figures 1.2 and 1.3A, B), and prompted the WHO to issue on 12 March a global alert on atypical pneumonia [8]. By 15 March 2003, the WHO had received reports of more than 150 cases of this new disease, which it named Severe Acute Respiratory Syndrome, from several Asian countries, Canada and Germany. The organisation immediately issued emergency travel

2

T H E

E P I D E M I O L O G Y

O F

S A R S

:

A

Two family members

G L O B A L

P E R S P E C T I V E

Two close contacts Four family members

Guangdong Province, China

Four HCWs Hospital 2 Hong kong

F

A A

10 HCWs Canada

F

G† G†

Three HCWs 156 close contacts of HCWs and patients

Hospital 3 Hong Kong

H

H

D

B

C

Four other Hong kong Hospitals

D

E Germany

No HCWs

Singapore

B

HCW

Vietnam 34 HCWs 37 HCWs

HCW

HCW

USA M§

C

B Hospital 4 Hong Kong

I

E

J

Ireland

L§

J

Hospital 1 Hong kong

K†

I

Hotel M, Hong Kong 99 HCWs (includes 17 medical students)

28 HCWs

K†

A

37 close contacts

HCW

Two family members

Unknown number close contacts

Bangkok

Fig. 1.1 Chain of transmission among guests at Hotel M, Hong Kong, 2003. Source: Ref. [7]. HCWs: health care workers. †: All guests except G and K stayed on the ninth floor of the hotel. Guest G stayed on the 14th floor and guest K stayed on the 11th floor. §: Guests L and M (spouses) were not at Hotel M during the same time as index guest A, but were at the hotel during the same time as guests G, H and I, who were ill during that period.

recommendations to alert health authorities, doctors and the travelling public to what was now perceived to be a worldwide threat to health [9]. The diseases spread rapidly, initially in Hong Kong [10], Hanoi [11], Toronto [12] and Singapore [13], and subsequently in the Mainland China and Taiwan [14]. Cumulative number of SARS cases passed 4000 on 23 April, 5000 on 28 April, 6000 on 2 May, 7000 on 8 May and 8000 on 28 May 2003 [15] (Figure 1.4). At the peak of the outbreak, during the beginning of May, more than 200 new cases were being reported each day.

The response to the outbreak was extraordinary. The causative agent was conclusively identified on 17 April following work through global and regional networks of virologists. On 28 April, Vietnam became the first country to stop local transmission of SARS, followed by the Philippines on 20 May and Singapore on 31 May [15]. By June, the number of new cases was gradually dwindling and by the end of June the global SARS epidemic, at least in its initial phase, was under control. There had been over 8000 cases in 30 countries worldwide and over 800 deaths (Table 1.1).

3

I N T R O D U C T I O N

60

Number of cases

50

Community

Health care worker

40

30

20

10

0 1 Nov 15 Nov 29 Nov 13 Dec 27 Dec 10 Jan 24 Jan 7 Feb 21 Feb 7 Mar 21 Mar 4 Apr 18 Apr Date of onset

Fig. 1.2 Probable cases of SARS by date of onset in Guangdong Province, China, 1 November 2002 to 30 April 2003 (n
1454).

Case definition WHO has developed a clinical case definition for SARS in order to describe the epidemiology of the disease, to monitor its spread and to provide the basis for advice on prevention and control [16]. Suspect case 1. A person presenting after 1 November 2002 with history of: – high fever (38°C) and – cough or breathing difficulty and one or more of the following exposures during the 10 days prior to onset of symptoms: – close contact with a person who is a suspect or probable case of SARS; – history of travel, to an area with recent local transmission of SARS; – residing in an area with recent local transmission of SARS. 2. A person with an unexplained acute respiratory illness resulting in death after 1 November 2002, but on whom no autopsy has been performed and one or more of the following exposures during 10 days prior to onset of

symptoms: – close contact with a person who is a suspect or probable case of SARS; – history of travel to an area with recent local transmission of SARS; – residing in an area with recent local transmission of SARS. Probable case 1. A suspect case with radiographical evidence of infiltrates consistent with pneumonia or respiratory distress syndrome (RDS) on chest X-ray (CXR). 2. A suspect case of SARS that is positive for SARS coronavirus by one or more assays. 3. A suspect case with autopsy findings consistent with the pathology of RDS without an identifiable cause. Close contact is defined as having cared for, lived with or had direct contact with respiratory secretions or body fluids of a suspect or probable case of SARS. Exclusion criteria A case should be excluded if an alternative diagnosis can fully explain their illness (but not simply on the basis of a negative test for SARS coronavirus).

4

T H E

E P I D E M I O L O G Y

O F

S A R S

:

A

G L O B A L

P E R S P E C T I V E

120

100

Number of cases

80

60

40

20

0 1 Feb

14 Feb

27 Feb

12 Mar

25 Mar

(a)

7 Apr

20 Apr

3 May

16 May

29 May

11 Jun

Date of onset 14 12

Number of cases

10 8 6 4 2 0 1 Feb

13 Feb 25 Feb

9 Mar

21 Mar

(b)

2 Apr 14 Apr 26 Apr Date of onset

8 May

20 May

1 Jun

13 Jun

Fig. 1.3A Probable cases of SARS by date of onset. (a) Hong Kong SAR, China, 1 February to 16 June 2003 (n
1731; as of 16 June 2003, an additional 24 probable cases of SARS have been reported from Hong Kong SAR, China, for whom no dates of onset are currently available). Source: Department of Health, Hong Kong Special Administrative Region of China. (b) Singapore, 1 February to 16 June 2003 (n
206). Source: Ministry of Health, Singapore, WHO.

Epidemiological features SARS is predominantly a disease of health care workers and younger adults (Table 1.2). There is a slight excess of cases in women, probably due to the predominance of female health workers. Most cases have been in people aged between 25 and 44 years.

The median age of cases was 35 years in Guangdong Province [17], 36 years in Singapore [18], 39 years in Hong Kong [10] and 45 years in Canada [19]. There have been very few cases in children and in older people. However, the highest age-specific incidence is in older people and the relatively small numbers in this age group most likely reflect the

E P I D E M I O L O G I C A L

5

F E A T U R E S

10 9 8

Number of cases

7 6 5 4 3 2 1 0 1 Feb

13 Feb 25 Feb

9 Mar

21 Mar

2 Apr 14 Apr 26 Apr Date of onset

8 May

20 May

1 Jun

13 Jun

9 Mar

21 Mar

2 Apr 14 Apr 26 Apr Date of onset

8 May

20 May

1 Jun

13 Jun

(a) 10 9 8 Number of cases

7 6 5 4 3 2 1 0 1 Feb (b)

13 Feb

25 Feb

Fig. 1.3B Probable cases of SARS by date of onset. (a) Vietnam, 1 February to 16 June 2003 (n
62; as of 16 June 2003 an additional probable case of SARS has been reported from Vietnam for whom no date of onset is currently available). Source: Ministry of Health, Vietnam, WHO. (b) Canada, 1 February to 13 June 2003 (n
242; as of 16 June 2003, one additional probable case of SARS has been reported from Canada for whom no date of onset is available). Source: Health Canada.

younger population profile of countries such as China (Figure 1.5). This means that in countries with a substantial proportion of older people, such as many Western countries, far more cases might be expected in older people if community transmission of SARS occurs. The proportion of cases with

underlying disease is between 10% and 25% depending on the setting in which infection occurs, being higher in situations where nosocomial transmission to other patients occurred as in Canada [19]. The commonest co-morbidity is diabetes or chronic heart disease.

6

T H E

E P I D E M I O L O G Y

O F

S A R S

:

A

G L O B A L

P E R S P E C T I V E

160 140

Number of cases

120 100 80 60 40 20 0 1 Nov 16 Nov 1 Dec 16 Dec 31 Dec 15 Jan 30 Jan 14 Feb 1 Mar 16 Mar 31 Mar 15 Apr 30 Apr 15 May 30 May 14 Jun

(a)

Date of onset 400 350

Number of cases

300 250 200 150 100 50 0 1 Mar (b)

13 Mar

25 Mar

6 Apr

18 Apr

30 Apr

12 May

24 May

5 Jun

17 Jun

Date of report

Fig. 1.4 Probable cases of SARS by date of onset, worldwide (a) 1 November 2002 to 16 June 2003 (n
5923; this graph does not include 2537 probable cases (2522 from Beijing), for whom no dates of onset are currently available). Source: Ministry of Health, China, WHO. (b) 1 March to 16 June 2003 (n
7563; as of 16 June 2003, 8460 probable cases of SARS have been reported to WHO). This graph includes all cases from Hong Kong SAR, Macao SAR and Taiwan, China, but only those cases elsewhere in China reported after 3 April 2003 (1190 cases between 16 November 2002 and 3 April 2003 not shown). Also includes 293 probable cases of SARS who have been discarded and for whom dates of report could not be identified. The USA began reporting probable cases of SARS to WHO on 20 April 2003.

Between 24% and 62% of SARS cases have been in health care workers, the proportion varying by the type of case series and declining as experience in the application of hospital infection control measures

has improved [17–20]. For example, in Guangdong, the proportion of cases in health care workers was 32% in January 2003, declining to 27% in February and 17% by March and April [17] (Figure 1.1).

E P I D E M I O L O G I C A L

7

F E A T U R E S

Table 1.1 Cumulative number of reported probable cases of SARS and deaths, 1 November 2002 to 30 June 2003. Country

Cumulative number of cases

Number of deaths

Date last probable case reported (2003)

Canada China, Mainland China, Hong Kong SAR China, Macao SAR China, Taiwan Europe Philippines Singapore Thailand USA Vietnam Other Asia Other Total

252 5327 1755 1 678 34 14 206 9 73 63 23 12 8449

37 348 298 0 84 0 2 32 2 0 5 2 1 811

25 June 25 June 11 June 21 May 19 June 4 June 15 May 18 May 7 June 23 June 14 April 31 May 9 June 25 June

SAR: Special Administrative Region. Source: WHO.

Table 1.2

Case series describing epidemiological and clinical features (at presentation) of SARS.

Median age (years) Women (%) HCWs (%): Nurses (%) Doctors (%) Medical students (%) Other (%) Hospital exposure (%) Co-morbidity: Diabetes (%) Cardiac disease (%) Other (%) Median incubation (range) (days) Symptoms: Fever (%) Chills (%) Malaise (%) Myalgia (%) Cough (%) Headache (%) Breathlessness (%) Diarrhoea (%) Leucopaenia (%) (3.5  109/L)

Hong Kong (to 25 March) n
138 [20]

Hong Kong (to 28 April) n
1425 [21]

Canada Singapore Guangdong (to 10 April) (to 30 April) (to 30 April) n
144 [19] n
201 [18] n
1454 [17]

39 52 62 40 24 18 18 100

– 57 – – – – – –

45 61 51 40 19 – 41 77

36 66 42 58 15 – 26 76

35 53 24 – – – – –

4 3 6 6 (2–16)

– – – 6.4 (5.3–7.8)a

11 8 8 6 (3–10)b

– – – 5 (1–10)

– – – –

100 73 – 61 57 56 – 20 34

94 65 64 51 50 50 31 27 –

99 28 31 50 69 35 42 24 –

– – – – – – – – –

97 52 42 31 70 40 27 9 14 (continued)

8

T H E

Table 1.2

E P I D E M I O L O G Y

O F

S A R S

:

A

G L O B A L

P E R S P E C T I V E

(continued)

Lymphopaenia (%) (1.0  109/L) CXR: Normal (%) Unilateral infiltrate (%) Bilateral infiltrate (%) Outcome: Intensive care (%) Ventilated (%) Dead (%)

Hong Kong (to 25 March) n
138 [20]

Hong Kong (to 28 April) n
1425 [21]

Canada Singapore Guangdong (to 10 April) (to 30 April) (to 30 April) n
144 [19] n
201 [18] n
1454 [17]

70

54

–

–

–

22 43 36

– – –

25 46 29

– – –

13 – –

23 14 3.6 (day 21)

– – 13.2 estimated (60y)

– – 6.5 (day 21)

– 11 12.5

– – 3.8

HCW: health care workers. a: mid-likelihood estimate and 95% confidence intervals. b: inter-quartile range. Number of cases 5.0 Incidence 4.0

200

3.0 2.0

100 1.0

Cases per 10,000 population

Number of cases

300

0.0

0– 4 5– 10 9 –1 15 4 –1 20 9 –2 25 4 –2 30 9 –3 35 4 –3 40 9 –4 45 4 –4 50 9 –5 55 4 –5 60 9 –6 65 4 –6 9 7 75 0 –74 or ov er

0

Age (years)

Fig. 1.5 Number of SARS cases by age and agespecific incidence, Guangdong Province, China, 1 November 2002 to 30 April 2003 (n
1454).

The largest staff group affected has been nurses, but cases have also occurred in doctors, paramedics, housekeeping staff and medical students. No cases have been reported in laboratory or pathology staff.

Key points SARS: epidemiology

• Predominantly a disease of health care •

workers and young adults. Commonest age group 25–44 years.

• Highest age-specific incidence in older people.

• Therefore, in countries with higher pro-

• • •

portion of older people (as in Western countries), there will be more cases in older people if community transmission of SARS occurs. Proportion of cases with underlying disease 10–25%. Commonest co-morbidity: diabetes and chronic heart disease. 24–62% of SARS cases have been found in health care workers, particularly involving nurses.

Clinical features SARS generally begins with a prodrome of fever (38°C), which is often high, often associated with chills and rigors and sometimes accompanied by other symptoms including headache, malaise and myalgia (Table 1.2). Some cases have mild respiratory symptoms though these are not prominent in the early stage of the illness. A few patients have reported diarrhoea during the febrile prodrome. After 3–7 days, a lower respiratory phase begins with the onset of a dry, non-productive cough or dyspnoea. Symptoms are milder in children [17,21] while in older people respiratory symptoms may be more prominent [17].

C L I N I C A L

9

F E A T U R E S

Key points

Key points

SARS: clinical features

SARS: laboratory tests

• • • • • •

Prodrome of fever 38°C. Chills, rigors. Headache, malaise, myalgia. Lower respiratory phase 3–7 days later. – Dyspnoea, non-productive cough. Symptoms milder in children. Respiratory symptoms more prominent in older people.

CXRs may be normal during the febrile prodrome and in about 20–25% of patients at the time of presentation. Most patients have early focal infiltrates progressing to more generalised, patchy, interstitial infiltrates. These changes are unilateral in around 40–45% of patients at presentation and bilateral in around 30%. Some CXRs from patients in the late stages of SARS have also shown areas of consolidation.

Key points SARS: CXR

• Normal during febrile prodrome in 20–25%.

• 50% have lymphopaenia (1000/L) at presentation.

• 15–35% have leucopaenia (3.5  109/L) at presentation.

• At peak of illness 50% have leucopae• •

nia and thrombocytopaenia (50,000– 150,000/L). In respiratory phase: ↑CPK (3000 IU/L) and hepatic transmission (two to six liver normal). Normal renal function tests in majority.

In 10–20% of cases, the respiratory illness is severe enough to require intubation and mechanical ventilation. The overall case fatality among persons with SARS is estimated to be 14–15%, but ranges from 0% to 50% depending on age. Case fatality is less than 1% in persons aged 24 years or younger, 6% in persons aged 25–44 years, 15% in persons aged 45–64 years and more than 50% in persons aged 65 years and older [21,22,23]. Patients with underlying disease, such as diabetes, are more likely to have a poor outcome [19].

• Early focal infiltrates progressing to patchy interstitial infiltrates.

• Unilateral in 40–45% at presentation, bilateral in 30%.

Key points SARS: clinical features

Early in the course of disease, the absolute lymphocyte count is often decreased. Over half of patients have lymphopaenia (1000/L) at presentation (mean 900/L) [19,20]. Total white cell counts are normal or decreased (mean 5  109/L), but between 15% and 35% have leucopaenia (3.5  109/L). At the peak of the respiratory illness, up to half of patients have leucopaenia and thrombocytopaenia or low-normal platelet counts (50,000–150,000/L). Early in the respiratory phase, elevated creatinine phosphokinase (CPK) levels (up to 3000 IU/L) and hepatic transaminases (2- to 6-times the upper limits of normal) have been noted. Renal function has remained normal in the majority of patients.

• 10–20% require intubation/mechanical ventilation.

• Overall case fatality 14–15% (range: 0–50%).

• Case fatality: – 1% in patients aged 24 years; – 6% in patients between 25 and 44 years of age;

– 15% in patients between 45 and 64 years of age;

– 50% in patients aged 65 years.

• Poorer outcome in patients with underlying disease.

10

T H E

E P I D E M I O L O G Y

O F

Incubation, infectivity and transmission Incubation period Knowing the incubation period of a disease, which is the time from exposure to a causative agent to onset of symptoms, is particularly important as it can help clinicians in making the diagnosis and it forms the basis for many recommended control measures, such as contact tracing and the duration of home isolation of contacts. The incubation period for SARS – based on analysis by WHO of individuals with well-defined single-point exposures in Singapore, Canada and Europe – is usually 2–7 days but may be as long as 10 days [23]. The incubation period will vary from one case to another according to the route by which the person was exposed, the dose of virus received and other factors such as immune status. Although there have been anecdotal reports of incubation periods longer than 10 days, these have not been corroborated.

Period of infectivity

S A R S

:

A

G L O B A L

P E R S P E C T I V E

members, and in health care workers, other patients and visitors inadvertently exposed to a case [10,12]. The period of infectivity appears to be greatest after the person with SARS develops respiratory symptoms, but transmission may occur during the prodromal period [24] (Figure 1.6). Persons with SARS do not appear to be infectious during the incubation period or after febrile symptoms have resolved. Further viral studies will help clarify the stages of the illness when virus shedding is greatest and the concentration of virus in various body fluids.

Key points Incubation period: 2–7 days (may be as long as 10 days). Period of infectivity:

• greatest after patient develops respiratory symptoms;

• may occur during prodrome; • not during incubation period; • not after afebrile symptoms have resolved.

Routes of transmission

The initial rapid spread of SARS in hospitals in Hanoi, Vietnam and in Hong Kong first indicated that the disease was highly contagious. Since then it has become clear that most cases of SARS occur in close contacts of patients, particularly household Encounter Incubation 2–10 days

SARS is spread in the majority of cases through close contact with an infected person. Transmission occurs mainly through exposure to infected large or medium droplets expelled during coughing or sneezing, and Isolation initiated

Recognition

Isolation ended

Period of communicability Period of risk for epidemic propagation

Asymptomatic shedding?

Infection exposure

Time (days) Symptom onset

Seeks care

SARS diagnosis

Infection control and isolation

Discharge

second-degree contacts exposed and infected Contact tracing Public health notified

Fig. 1.6

A contagion model for SARS.

second-degree case ascertainment

I N C U B A T I O N

,

I N F E C T I V I T Y

A N D

probably also occasionally from contact with contaminated fomites [10,12]. Likelihood of contagion seems to be similar to that described with other viral respiratory tract pathogens, such as respiratory syncytial virus and rhinovirus [25]. Shedding of the SARS virus in respiratory secretions, faeces and urine is now well established [24,26]. The role of body fluids (such as saliva, tears and urine) and of faeces in transmission of infection is less clear. In vitro laboratory tests show that the virus can survive in faeces for at least 2 days, and in urine for at least 24 hours [26]. Virus in faeces taken from patients suffering from diarrhoea, which has a lower acidity than normal stools, can survive for 4 days raising the possibility that surfaces contaminated by diarrhoea could be an important source of infection. However, the dose of virus needed to cause infection remains unknown and further studies are needed before the role of faecal–oral transmission can be determined.

Key points Routes of transmission

• Close contact with infected person. • Droplets expelled during coughing/ •

sneezing. Contact with contaminated fomites.

Transmission in hospital Health care workers are at high risk of SARS. Most infections have occurred either before infection control procedures were instituted or where procedures have not been properly followed. Observations from Hong Kong suggest that this risk is greatest when a hospital receives its first admissions of SARS, when patients are admitted to a general ward, and when large numbers of patients are admitted over a short period of time [27]. Nosocomial spread is less likely when patients are admitted directly to a designated ward. One case–control study of risk factors for transmission in hospital staff caring for patients with SARS found that fewer staff who wore masks and gowns, and washed their hands became infected compared with those who did not, though only not wearing a mask was independently associated with increased risk [28]. Investigation of one incident among protected health care workers suggested

T R A N S M I S S I O N

11

that ill-fitting masks or contamination during mask removal were factors in infection [29].

Key points SARS: transmission in hospital Factors involved:

• infection control measures; • admission/cohorting procedures in wards; • use of gowns, masks and hand washing.

Transmission in the community SARS transmits readily within the household setting, although if patients are isolated early in the course of the illness, secondary transmission to other members of the household can be prevented [30]. Other instances of spread in the community are rare with the highest risk probably being in those exposed to symptomatic patients in confined areas such as taxi drivers, and airline staff and passengers. A review by WHO of 35 flights with symptomatic probable SARS cases on board identified four flights during which in-flight transmission may have occurred [31]. One flight was associated with 22 secondary cases including two flight attendants. The affected passengers were seated within seven rows in front and five rows behind the index case suggesting that the infection spread through exposure to respiratory droplets. Cases in other settings are rare though there have been reports of transmission in the banqueting room of a restaurant, in the workplace and at a wholesale market [18]. In contrast to experience in hospitals, there have been very few large outbreaks in the community. One notable exception was a large and sudden cluster of over 300 cases that occurred in residents of the Amoy Gardens housing estate in Hong Kong [32]. Cases associated with this cluster were much more likely to present with diarrhoea and to require intensive care compared with previously documented SARS cases. This raised the possibility of transmission by a different route and infection with a high virus load, such as might happen following exposure to a concentrated environmental source. Subsequent investigations ruled out airborne or

12

T H E

E P I D E M I O L O G Y

O F

:

A

G L O B A L

P E R S P E C T I V E

with SARS who have been implicated in spreading the disease to numerous (10 or more) other individuals, and these cases have been described as ‘superspreaders’ [18,33] (Figure 1.7). Several such individuals have been described particularly during the early days of the SARS outbreak. They include the hotel index case in Hong Kong who triggered the worldwide dissemination of SARS as well as the index case for each of the outbreaks in Vietnam and Singapore [7,11,13]. Superspreaders have also been described with other diseases, such as Ebola [34].

waterborne transmissions and suggested that the outbreak was primarily caused by contaminated sewage entering households through dried U-traps in the bathroom floor drain.

Key points SARS: transmission in community

• Transmits readily in household setting. • Early isolation of patients prevents sec•

S A R S

ondary transmission to other household members. Reports of transmission in taxis, aircraft.

The explanation for the apparent high infectivity of these patients remains unknown. It may be that they were co-infected with another virus and that this caused them to secrete an exceptionally high amount of infectious material or that some other factor, perhaps in the environment, amplified the potential for transmission at some key phase of virus shedding. Alternatively, high secondary attack rates in hospital staff, relatives and other visitors that some superspreaders caused may be a consequence

Superspreaders/superspreading event patient On average, one SARS generates around three secondary cases [30]. However, certain individuals

42 119

86

152 153

34 27

100

78

22 2

85 29

4

24

66

1

7 33

21 20

75

13

89 155

142

94 49

15 14 12

157

8 18

16

30

110

5

9

37

109 120 12021 146 133 113 128 124 126 115 149 138 121 159 151 117 154 122 118 130 139 140 145 173 39 131 161 162 160 136 165 163 158 127 170 137 143

65

111

91

135

166

114 95

108 106

23

76 83 59 25

106

80 70 84 6

90

46 28 45 17 56 51 47 26 52

134

35

61 36 60 70 55 8060 50 435840 71 68 59 93 103 887782 77 01

175

172 164 181 169

177 180 174 179

112

144

123 171

Fig. 1.7 Probable cases of SARS by reported source of infection showing superspreaders (cases 1, 6, 35, 127 and 130), Singapore, 25 February to 30 April 2003. Source: Morb Mortal Wkly Rep 2003; 52: 405–411.

I N C U B A T I O N

,

I N F E C T I V I T Y

A N D

of the absence of stringent hospital infection control measures before the infectivity of SARS was recognised. Isolation of contacts, rapid admission of cases to hospital and improved hospital infection control have all been shown to significantly reduce the average number of new cases of SARS arising from a single-source case [30].

Prevention and control Breaking the chain of transmission At present, the most effective way to control SARS is to break the chain of transmission from infected to healthy people. There are three key actions necessary to achieve this: early case detection, patient isolation and contact tracing (Table 1.3) [35]. Case detection aims to identify a SARS case as soon as possible after the onset of symptoms. Once a case is identified, the next step is to ensure they are isolated promptly and managed according to strict infection control procedures. Finally, it is vital to identify all close contacts of each case and make sure that they are carefully followed up, including daily health checks and voluntary home isolation. These measures limit the daily number of contacts possible for each potentially infectious case. By shortening the amount of time that elapses between onset of illness and isolation of the patient the opportunity for the virus to spread to others is reduced, as is the average number of new cases generated by each case (the effective reproduction number) [21,30] (Figure 1.6). If each new SARS patient infects, on average, less than one person then the outbreak will die out, otherwise the disease will continue to spread. Table 1.3 Detection and protection measures to control SARS.

• • • • • •

Prompt identification of persons with SARS, their movements and contacts Effective isolation of SARS patients in hospitals Appropriate protection of health care workers treating SARS patients Comprehensive identification and isolation of suspected SARS cases Exit screening of international travellers Timely and accurate reporting and sharing of information with other authorities and/or governments

T R A N S M I S S I O N

13

Surveillance and reporting Good surveillance and rapid reporting is the basis of case detection and is required to instigate contact tracing and promptly identify outbreaks. Most surveillance systems rely on notification of infectious diseases by clinical practitioners. In the case of SARS, this has been achieved by voluntary reporting by clinicians of cases meeting the WHO case definition firstly to local public health authorities and then onwards to national ministries of health and the WHO itself. The use of electronic systems can assist in the rapid reporting and response to suspect cases. The current clinical case definition has worked well to control the SARS outbreak but a more precise definition will be needed for longer-term surveillance [36]. Until a highly reliable, sensitive and specific early test is available, diagnosis of SARS will inevitably depend on evaluation of symptoms and history of contact. When local transmission of SARS has ceased, clusters of atypical pneumonia and nosocomial transmission to health care staff or hospital visitors will become important sentinel events for recognising new instances of SARS.

Isolation of patients The earliest possible isolation of all suspect and probable cases of SARS in hospital is vital. A short time between the onset of symptoms and isolation of the patient reduces opportunities for transmission of infection to other people and reduces the number of contacts that require active follow-up [30]. It also gives patients the best chance of receiving lifesaving care, should their condition take a critical course. In an outbreak of SARS every effort should be made to reduce the average time from the onset of symptoms to isolation under 3 days.

Key points SARS: prevention and control Breaking the chain of transmission:

• Early case detection/surveillance and reporting of cases.

• Patient isolation. • Contact tracing.

14

T H E

E P I D E M I O L O G Y

O F

Hospital infection control Health care settings seem to provide an ideal setting for outbreaks of SARS, and have served to amplify and propagate the infection. The reasons are unclear but appear to be a combination of factors such as the potential for aerosol-generating procedures, poor ventilation and air flow patterns, and the presence of vulnerable patients. Inadequate training or compliance with infection control, high workload and cryptic clinical presentations can compound the problem. Meticulous infection control procedures are necessary to prevent nosocomial spread including [37]:

• Comprehensive triage arrangements in emergency departments and clinics.

• Adequate facilities for patient isolation with appropriate air flow control.

• Scrupulous hand hygiene. • Use of appropriate personal protective equipment. • Avoidance of aerosol-generating procedures.

S A R S

G L O B A L

P E R S P E C T I V E

If the source of infection in the index case is not known, a full contact history for the 10 days before the date of onset of symptoms should be obtained. This will include details of all persons who may have had face-to-face contact with the case, as well as information on the workplace, visits to hospitals or other health care facilities, visits to crowded places, details of journeys on public transport, and overnight stays outside the home. Where there are several cases with an unidentified source, it is important to correlate contact histories to identify any common links.

Key points Contact tracing

• The case (if possible) or next-of-kin should

The infection control measures for a radiology department are discussed in a separate chapter towards the end of the book.

• • •

Contact tracing and isolation of contacts

All close contacts should be followed up for 10 days from the last date of contact with the case. Quarantine or home confinement of contacts was employed with considerable success to bring the outbreaks in the Mainland China, Hong Kong and Singapore under control. The temperature of all contacts should be monitored daily, preferably by a health worker. It may be necessary to make arrangements to provide

A

the household with food supplies and other essential commodities during the period of observation. In overcrowded households it may be preferable to relocate the contacts to alternative accommodation to limit the risk of secondary transmission within the household.

SARS has clearly demonstrated that a single case admitted to an unprepared hospital can ignite a new outbreak.

The case should be interviewed by a trained health care worker as soon as possible after the diagnosis of SARS is made, either face-to-face or by telephone. If the patient is too ill to be interviewed a proxy contact history should be obtained from the nextof-kin. The date of onset of symptoms should be corroborated and details of all close contacts since that date obtained. Close contacts are defined as anyone who cared for, lived with or had direct contact with respiratory secretions or body fluids or stool of person with SARS [16].

:

• •

be interviewed by trained health care workers. Date of onset of symptoms should be corroborated. Details of all contacts must be obtained. Contacts should be followed up for 10 days from last date of contact with the case. The temperature of all contacts should be monitored daily. Quarantine of contacts may be necessary.

Management of an outbreak Each hospital, and public health services at each level within the health system, should have an outbreak plan that can be used to guide the response to a SARS incident. It should include the following:

• clear aims and objectives; • a description of the kind of circumstances in • •

which an outbreak should be declared and an Outbreak Control Team (OCT) convened; the terms of reference for the OCT; the proposed membership of the OCT and an outline of the roles and responsibilities of individual members;

15

C O N C L U S I O N

• arrangements for planning and implementing •

•

appropriate control measures, and for securing urgent additional resources, if required; details of lines of communication with partner organisations and agencies, central government, health professionals, the general public and the media; arrangements for declaring the outbreak over and preparing a final written report.

A final report of the outbreak should be prepared and presented to the responsible authority each time an OCT is convened. The plan should regularly updated and revised in the light of changing circumstances.

Public education An outbreak of a serious infectious disease that is poorly understood and for which no treatment is available inevitably produces great public concern. As far as possible, it is important to give the public clear and unambiguous information about the nature of the disease and the risk it poses. Mass-media campaigns to educate the public were an important part of the response to SARS. The wearing of a mask, other than by those in close contact with a suspect or probable case, is unlikely to be of value. However, information about the symptoms of the disease and advice to encourage prompt reporting of symptoms is sensible. In some circumstances, the establishment of fever clinics to screen individuals with symptoms and to relieve pressure on emergency rooms may be helpful [36].

Border control measures and travel precautions In areas where person-to-person transmission of SARS has been documented, screening of travellers and border control measures are indicated [38,39]. All travellers should be made aware of the symptoms of SARS and advised to seek immediate medical attention should symptoms occur. Departing international passengers should be asked to complete a brief questionnaire on any history of contact with a suspect or probable SARS case, or any symptoms of SARS during the previous 48 hours, and be screened for fever by means of a temperature check. Travellers with fever should be requested to postpone travel and to seek medical attention. Persons meeting the SARS case definition should be referred to a health care facility.

Table 1.4

• • • • • • •

In-flight care of a suspected case of SARS.

Isolate passenger, as far as possible, from other passengers and crew Provide passenger with protective mask Identify toilet for exclusive use of the ill passenger Carer(s) should wear protective mask and gloves and wash hands after contact with the ill passenger Captain should radio ahead to alert port health authorities at the destination airport Identify all contacts on board (household members, flight attendants and passengers in the same row or two rows in front or behind) On arrival, place passenger in isolation until assessed by port health authorities

If a passenger on a flight from a SARS-affected area becomes noticeably ill with a fever and respiratory symptoms, the cabin crew are recommended to take basic precautions (Table 1.4) [39]. If medical assessment concludes that the person is a suspect or probable case of SARS all contacts during the flight should be identified and followed up in accordance with the WHO guidance.

Conclusion The speed with which the SARS epidemic was managed demonstrates the decisive power of high-level political commitment to contain an outbreak. Extensive use of measures like patient isolation, contact tracing and follow-up, quarantine, public education and travel precautions have worked to bring the disease under control even in the absence of effective treatments or vaccines [40]. Many unanswered questions about SARS remain, particularly about the origin of the coronavirus. If the disease has, as suspected, a wild animal reservoir, then it may become endemic at low levels or re-emerge as a seasonal or epidemic infection. SARS has features that can thwart even the best plans. The initial symptoms may masquerade as many other diseases, the incubation period is long enough to allow it to spread to any part of the globe without detection, and one single highly infectious case may set off a chain of transmission leading to a hundred or more additional cases. Continued worldwide vigilance, particularly for any hospital-based cluster of febrile patients or health care workers with respiratory symptoms, will therefore be needed for some time to come [15].

16

T H E

E P I D E M I O L O G Y

O F

References 1. Peiris JS, Lai ST, Poon LL et al. Coronavirus as a possible cause of severe acute respiratory syndrome. Lancet 2003; 361: 1319–1325. 2. Ksiazek TG, Erdman D, Goldsmith CS et al. A novel coronavirus associated with severe acute respiratory syndrome New Engl J Med 2003; 348: 1953–1966. 3. Drosten C, Gunther S, Preiser W et al. Identification of a novel coronavirus in patients with severe acute respiratory syndrome. New Engl J Med 2003; 348: 1967–1976. 4. Morse SS. Factors in the emergence of infectious diseases. Emerg Infect Dis 1995; 1: 7–15. 5. World Health Organisation. Severe Acute Respiratory Syndrome (SARS): Status of the Outbreak and Lessons for the Immediate Future. Geneva, World Health Organisation, 20 May 2003. Available at: http://www.who.int/csr/media/sars_wha.pdf (accessed 28 May 2003). 6. Rosling L and Rosling M. Pneumonia causes panic in Guangdong province. Br Med J 2003; 326: 416. 7. Centers for Disease Control and Prevention. Outbreak of severe acute respiratory syndrome – worldwide, 2003. Morb Mortal Wkly Rep 2003; 52: 241–248. 8. World Health Organisation. WHO Issues Global Alert about Cases of Atypical Pneumonia: Cases of Severe Respiratory Illness May Spread to Hospital Staff. Geneva: World Health Organisation, 12 March 2003. Available at: http://www.who.int/ medicentre/releases/2003/pr22/en/ (accessed 14 May 2003). 9. World Health Organisation. Severe Acute Respiratory Syndrome. Available at: http://www.who.int/csr/sarsarchive/2003_ 03_15/en/ (accessed 14 May 2003). 10. Tsang KW, Ho PL, Ooi GC et al. A cluster of cases of severe acute respiratory syndrome in Hong Kong. New Engl J Med 2003; 348: 1977–1985. 11. Vu TH, Cabau JF, Nguyen NT and Lenoir M. SARS in North Vietnam. New Engl J Med 2003; 348: 2035. 12. Poutanen SM, Low DE, Henry B et al. Identification of severe acute respiratory syndrome in Canada. New Engl J Med 2003; 348: 1995–2005. 13. Hsu L-Y, Lee C-C, Green JA et al. Severe acute respiratory syndrome (SARS) in Singapore: clinical features of index patient and initial contacts. Emerg Infect Dis 2003; 6: 713–717. 14. Twu S-J, Chen T-J, Chen C-J et al. Control measures for severe acute respiratory syndrome (SARS) in Taiwan. Emerg Infect Dis 2003; 6: 718–720. 15. World Health Organisation. Severe acute respiratory syndrome (SARS): over 100 days into the outbreak. Wkly Epidemiol Rec 2003; 78: 217–220. 16. World Health Organisation. Case Definitions for Surveillance of Severe Acute Respiratory Syndrome (SARS). Available at: http://www.who.int/csr/sars/casedefinition/en/ (accessed 16 May 2003). 17. Xu R-H, He J-F, Evans MR et al. Epidemiologic clues to the origin of severe acute respiratory syndrome in China (personal communication). 18. World Health Organisation. Severe acute respiratory syndrome – Singapore 2003. Wkly Epidemiol Rec 2003; 78: 157–162. 19. Booth CM, Matukas LM, Tomlinson GA et al. Clinical features and short-term outcomes of 144 patients with SARS in the Greater Toronto area. J Am Med Assoc 2003; 289: 1–9. 20. Lee N, Hui D, Wu A et al. A major outbreak of severe acute respiratory syndrome in Hong Kong. New Engl J Med 2003; 348: 1986–1994. 21. Donnelly C, Ghani AC, Leung GM et al. Epidemiological determinants of spread of causal agent of severe acute respiratory syndrome in Hong Kong. Lancet 2003; 361: 1761–1766. 22. Hon KLE, Leung CW and Cheng WFT. Clinical presentations and outcome of severe acute respiratory syndrome in children. Lancet 2003: 361: 1701–1703.

S A R S

:

A

G L O B A L

P E R S P E C T I V E

23. World Health Organisation. Update 49. SARS Case Fatality Ratio, Incubation Period. Available at: http://www.who.int/ csr/sarsarchive/2003_05_07a/en/ (accessed 14 May 2003). 24. Peiris JSM, Chu CM, Cheng VCC et al. Clinical progression and viral load in a community outbreak of coronavirusassociated SARS pneumonia: a prospective study. Lancet 2003; 361: 1767–1772. 25. Musher DM. How contagious are common respiratory infections? New Engl J Med 2003; 348: 1256–1266. 26. World Health Organisation. Update 47. Studies of SARS Virus Survival, Situation in China. Available at: http://www. who. int/csr/sarsarchive/2003_05_05/en/ (accessed 14 May 2003). 27. Chan-Yeung M, Seto WH, Sung JJ et al. Severe acute respiratory syndrome: patients were epidemiologically linked. Br Med J 2003; 326: 1393. 28. Seto WH, Tsang D, Yung RW et al. Effectiveness of precautions against droplets and contact in prevention of nosocomial transmission of severe acute respiratory syndrome (SARS). Lancet 2003; 361: 1519–1520. 29. Centers for Disease Control and Prevention. Cluster of severe acute respiratory syndrome cases in protected health care workers – Toronto, Canada, April 2003. Morb Mortal Wkly Rep 2003; 52: 433–436. 30. Riley S, Fraser C, Donnelly CA et al. Transmission dynamics of the etiological agent of SARS in Hong Kong: impact of public health interventions. Science 2003; 300: 1961–1966. 31. World Health Organisation. Update 62. More Than 8000 SARS Cases Reported Globally, Situation in Taiwan, Data on In-Flight Transmission, Report on Henan Province, China. Available at: http://www.who.int/csr/sarsarchive/2003_ 05_22/en/ (accessed 23 May 2003). 32. Hong Kong Department of Health. Main Findings of an Investigation into the Outbreak of Severe Acute Respiratory Syndrome at Amoy Gardens. Available at: http://www.info. gov.hk/dh/ap.htm (accessed on 19 April 2003). 33. World Health Organisation. Update 30. Status of Diagnostic Test, Significance of Superspreaders, Situation in China. Available at: http://www.who.int/csr/sarsarchive/2003_04_15/en/ (accessed 14 May 2003). 34. Khan AS et al. The re-emergence of Ebola hemorrhagic fever, Democratic Republic of Congo, 1995. J Infect Dis 1999; 179(Suppl 1): S76–S86. 35. World Health Organisation. Update 54. Outbreaks in the Initial Hot Zones Indicate that SARS Can be Contained. Available at: http://www.who.int/csr/sarsarchive/2003_05_13/en/ (accessed 14 May 2003). 36. Rainer TH, Cameron PA, Smit D et al. Evaluation of WHO criteria for identifying patients with severe acute respiratory syndrome out of hospital: prospective observational study. Br Med J 2003; 326: 1354–1358. 37. World Health Organisation. Hospital Infection Control Guidelines for Severe Acute Respiratory Syndrome. Available at: http://www.who.int/csr/sars/infectioncontrol/en/ (accessed 19 May 2003). 38. World Health Organisation. Update 11. WHO Recommends New Measures to Prevent Travel-Related Spread of SARS. Available at: http://www.who.int/csr/sarsarchive/2003_ 03_27/en/ (accessed 14 May 2003). 39. World Health Organisation. WHO recommended measures for persons undertaking international travel from areas affected by severe acute respiratory syndrome (SARS). Wkly Epidemiol Rec 2003; 78: 97–99. 40. World Health Organisation. Vietnam SARS-free. Wkly Epidemiol Rec 2003; 78: 145–146.

The Role of Emergency Medicine in Screening SARS Patients TH Rainer and P Cameron

Introduction Historical background Questions without answers Definition Clinical features Radiography

17 18 18 19 20 20

Introduction The Health Care System in Hong Kong was completely unprepared for the outbreak of severe acute respiratory syndrome (SARS) which occurred in March 2003. It challenged the entire medical system resulting in the temporary closure of emergency departments (EDs) (Figure 2.1) and hospitals [1–4]. In the early stages, the illness affected an unusually high proportion of health care workers. The health care system was faced with many difficult issues including how and where to screen the hospital workforce and local community, and who should take responsibility for this screening task? Emergency medicine is a systems-based, primary care speciality with skills and training to screen large populations of patients but its primary brief is to deal with acutely ill patients and not to provide a public health service [5]. However, in the acute crisis brought about by the SARS outbreak, it was appropriate for the ED to take short-term responsibility to

CHAPTER

2

Laboratory tests SARS-screening clinic Staff safety Managing patients with suspected SARS Conclusion

22 23 25 26 27

coordinate screening, as there were no other facilities available immediately. This chapter reviews the challenges and responses involved in initiating and developing a SARS-screening clinic in an ED of a University Hospital in Hong Kong. While clinical features aid the assessment of patients suspected of having SARS, radiography supplemented by high-resolution computerized tomography (HRCT) is the cornerstone of SARS assessment.

Key points Accuracy of clinical features and radiology to detect SARS in our experience

• Clinical features have a sensitivity of 90% for identifying SARS.

• Chest radiography (CXR) supplemented with HRCT achieves a sensitivity of 99% for detecting SARS.

18

Fig. 2.1

R O L E

O F

E M E R G E N C Y

M E D I C I N E

:

S C R E E N I N G

P A T I E N T S

An exceptional sign – a closed ED in a closed hospital.

Historical background As mentioned in the previous chapter the first known case of this form of atypical pneumonia was identified in Guangdong Province, South China in November 2002 and in February 2003 the first case, a physician from Guangzhou, was diagnosed in Hong Kong [6,7]. While staying at a local hotel he infected a number of other people, one of whom was subsequently admitted to the Prince of Wales Hospital, New Territories [2]. The index case in the Prince of Wales Hospital was a 26-year-old Chinese man who presented twice to the ED – on 28 February and 4 March – and after the second visit was admitted to a general medical ward [2]. The diagnosis was pneumonia. At the time there were more than 20 other patients in the ward and 30 medical students. In addition to many of the patients and relatives on the ward, 15 medical, 15 nursing, five other ancillary staff and 17 of the medical students fell ill. In response to this disastrous clustered outbreak of SARS in Prince of Wales Hospital, an emergencyscreening clinic was set up to evaluate all staff or immediate contacts [2,8]. The objective of the clinic

was to provide a safe environment in which to screen the staff and relatives of patients in the hospital who had recently been in contact with patients with SARS-pneumonia. The clinic gave us the opportunity to study the clinical response to the virus in a highcontact environment. The global spread of this illness means that our experience is likely to be repeated in many health care settings around the world [9].

Key points Objectives of a SARS-screening clinic

• To provide a safe environment for staff to work.

• To prevent secondary infections. • To accurately and appropriately screen, diagnose and manage suspect SARS cases.

Questions without answers In the early period, health care providers were faced with a severe, life-threatening, infectious disease

19

D E F I N I T I O N

of unknown aetiology, unknown epidemiology, unknown clinical presentation and clinical course, no vaccine, no evidence-based treatment, and an uncertain prognosis. There were numerous reports of health care workers who had contracted the illness [10–12], some of whom were personally known to staff [13], some of whom had died [14]. Apart from the challenge of dealing with patients and their illness, the staff had to deal with their own psychological needs including the fear that they might contract the illness, spread it to their family and the possibility that it could be terminal [15,16].

Clinical assessment The World Health Organisation (WHO), Centers for Disease Communication (CDC) published early guidelines for managing SARS but the evidence base had not been evaluated properly [17–20]. The specific challenges that concerned us were:

assess these cases without secondary spread from suspect SARS cases to the staff and possibly their families or friends? 2. What personal protective equipment (PPE) is necessary and how much training and education is required? Should the assessment area be in the ED or hospital or remote from the hospital? 3. How could we reassure an anxious workforce which was faced with the uncertainty of a potentially deadly illness, for which there was no known cause, no known treatment and no vaccine? 4. What issues affect morale? Not only did these questions require addressing but that in order to provide good answers, it was imperative to start researching the virus and its effects from every possible angle. The four main issues were clinical assessment, appropriate management, staff safety and research.

Definition

1. How reliable and sensitive were current WHO guidelines for screening SARS cases? 2. Advanced cases of SARS usually develop an overt illness with fever, cough, shortness of breath, lymphopaenia, pyrexia (38°C) and thrombocytopaenia but were these the important features early in the illness at the time of screening? 3. Should all suspect cases with any one or more symptoms/signs of a fever, cough, shortness of breath, lymphopaenia, pyrexia (38°C) or thrombocytopaenia have the illness and be admitted to a hospital ward? If so, and they did not have the illness, would they contract it after admission? 4. If subjects were discharged from clinic with plans for daily follow-up, would they return home to infect their families and spread disease further?

At the start of the staff-screening clinic little was known about the illness except that it appeared that a virus was involved. Although the case definition of SARS was clearly defined later during the course of the outbreak (discussed in the previous chapter), in the initial days the case definition itself was vague and not confined. At the time, according to WHO, SARS was suspected in a person with a high fever (38°C), and one or more respiratory symptoms (e.g. cough, shortness of breath or difficulty breathing) and close contact with a person previously diagnosed with SARS [18].

Therefore, how we had to address SARS-pneumonia could best be diagnosed in an anxious, polysymptomatic group of patients who may present to a primary care facility for screening and diagnosis.

Initially in the absence of a reliable test for the virus, the diagnosis of confirmed SARS-pneumonia was made when a person had known contact, documented persistent pyrexia (38°C), and consistent clinical course of the illness, and evidence of pneumonia. Pneumonia was diagnosed either on plain radiography or by CT.

Staff and patient safety

A probable case of SARS was when an individual met the criteria of a suspected case but then developed pneumonic change on CXR.

We also had to consider:

A diagnosis of non-SARS-pneumonia was made if the patient responded well to antibiotics within 48 hours.

1. How should we provide a safe, secure screening environment for medical and nursing staff to

Later, when coronavirus was confirmed as the aetiological pathogen [21–28] and a virological test

20

R O L E

O F

E M E R G E N C Y

became available, the final diagnosis was by a team of medical, respiratory and infectious diseases clinicians in conjunction with positive virological evidence of the presence of coronavirus.

Clinical features The clinical course of the disease is thought to follow an incubation period (⯝2–16 days), with high fevers, chills, rigors and myalgia [8,11]. Respiratory symptoms are not prominent prior to admission to hospital and may be poor discriminators between SARS and non-SARS cases in the early phase. Some patients presented with diarrhoea, abdominal pain and loss of appetite [8]. Although there are reports that SARS patients commonly have basal crackles, these findings are present in admitted patients with advanced illness [11]. In the early assessment phase, very few patients have chest signs [8]. Sore throat, lymphadenopathy and skin rashes are also absent. Basic observations are performed at every follow-up including pulse rate, systolic and diastolic blood pressure, respiratory rate, temperature and oxygen saturation (if available) on room air. Although a pyrexia (38°C) is a characteristic feature of patients with probable SARS after admission to hospital and is a hallmark of WHO guidelines, many patients did not have documented pyrexia when screened in our clinic within a few days of the onset of symptoms [8]. Many of these patients had a good contact history, symptoms of fever, malaise and/or chills and typical radiographical features of pneumonia but in the absence of pyrexia. Therefore, the role of pyrexia in the early assessment of patients with SARS must be viewed with caution.

M E D I C I N E

:

S C R E E N I N G

P A T I E N T S

Other less common features may be:

• • • • •

cough, shortness of breath, abdominal pain, diarrhoea, loss of appetite.

Pyrexia 38°C:

• is often absent in the first few days since symptom onset.

Radiography Our investigations have shown that clinical criteria and temperature alone yield a sensitivity of detecting SARS of only 80% but the addition of imaging (daily CXR and diagnostic HRCT in specific cases) may give a sensitivity of 100% and a positive-predictive value of 60% (personal communication). CXRs requested in the first 100 cases showed a variety of appearances at presentation including unifocal (Figure 2.2), multifocal (Figure 2.3) or diffuse (Figure 2.4) pneumonia [8,11]. HRCT was requested in some

Key points Clinical discriminators for patients with SARS [8] Most discriminating features are:

• • • •

fever, malaise, chills, loss of appetite.

Fig. 2.2 Frontal CXR showing an area of air-space opacification in the right lower zone (arrow).

21

R A D I O G R A P H Y

patients and showed a typical area of ground-glass opacification (Figure 2.5). Our policy in the screening clinic was to request HRCT for patients who had normal CXRs (Figure 2.6a) but persistent pyrexia and symptoms. Some patients with normal radiographs have

retro-cardiac (Figure 2.6b) or retro-diaphragmatic (Figure 2.6c) lesions which are difficult to detect on a frontal CXR but clearly evident on CT. In our experience, the median time from symptom onset to identifying changes on imaging was – 4 days for CXR; – 7 days for HRCT (in patients where initial CXRs were normal). In some rare instances the imaging was positive 21 days later and we believe some of these late presenters may have contacted SARS after being cohorted into a ward which already had other SARS patients.

Key points Early CXR findings in SARS patients

• Unifocal and multifocal pneumonic patches.

• Diffuse or lobar pneumonia. • Median time from symptom onset to positive radiographs was 4 days. Early HRCT findings in SARS patients with normal CXRs Fig. 2.3 Frontal CXR showing confluent consolidation in right upper zone and subtle areas of airspace opacities in both lower zones (arrows).

• Retro-cardiac ground-glass pneumonia. ground-glass • Retro-diaphragmatic pneumonia.

• Median time from symptom onset to positive CT was 7 days.

Fig. 2.4 Frontal CXR showing diffuse areas of airspace opacification in both lower zones.

Fig. 2.5 HRCT showing a small area of groundglass opacification in the apico-posterior segment of the right lower lobe (arrow).

22

R O L E

O F

E M E R G E N C Y

M E D I C I N E

:

S C R E E N I N G

P A T I E N T S

(b)

(a)

(c)

Fig. 2.6 (a) Normal frontal CXR of a patient with clinical signs and symptoms highly suggestive of SARS. (b) HRCT showing consolidation in the anterior basal segment of the left lower lobe (arrow). Note its retrocardiac location which makes it difficult to identify it on a frontal CXR. (c) HRCT showing consolidation in the posterior segment of the right lower lobe (arrow). Note its retro-diaphragmatic location which makes it difficult to identify it on a frontal CXR.

Laboratory tests Laboratory tests, viz. complete blood and differential counts, renal and liver function tests, and coagulation profiles, were requested in our clinic although they offer little advantage over a good clinical assessment and radiography in the early screening phase. Lymphopaenia is a common feature in SARS patients [29,30]. However, many subjects in the screening clinic had profound but transitory lymphopaenias that lasted no more than 24–48 hours. These patients have been found to be coronavirus negative on subsequent serology and so did not have a SARS-related episode. If they had been admitted to a common ward with SARS patients they may well have contracted the illness in hospital. Therefore, lymphopaenia in

the absence of radiographical evidence of pneumonia should not be a criterion for admission to a common area with other possible SARS cases. Specific diagnostic tests are now available but are unlikely to aid physicians working in a screening clinic. At present, these tests become positive late in the course of the disease after patients have developed pneumonia. An immunofluorescence assay for detecting anti-coronavirus IgG antibody is available and is believed to have a sensitivity between 50% and 100% at 7–21 days after the onset of fever. Polymerase chain reaction (PCR) can also be used to detect ribonucleic acid (RNA) from nasopharygeal aspirates, urine and stool, although the interval sensitivity is low [22,24,25].

S A R S

Fig. 2.7

-

S C R E E N I N G

C L I N I C

23

Separate fever area in the ED.

Key points Common tests

• Leucopaenia, lymphopaenia and thrombocytopaenia are common in SARS patients but are also common in coronavirusnegative subjects. Specific tests

• Coronavirus serology has 100% sensitivity at 21 days after fever.

• Viral RNA may be detected in nasopharyngeal aspirates, urine and stool.

SARS-screening clinic In order to address the crisis facing the Prince of Wales Hospital, a SARS clinic based in the ED of the

Prince of Wales Hospital was opened on 12 March 2003 [8]. Initially, the clinic was based in the ED and a SARS, or fever area, was separated away from the rest of the department (Figure 2.7). Staff had to change their PPE when entering and leaving the area. Later the clinic was moved outside the main hospital building. Screening potential patients with this disease is particularly difficult as the signs and symptoms are vague and consistent with virtually any viral illness. Therefore, following up patients over a number of days is the only way of knowing whether they do in fact have SARS. The question of whether to admit all suspicious cases is also an issue. If suspicious cases are admitted, they may actually contract the disease in hospital. If they are discharged they may infect their families and friends.

24

R O L E

O F

E M E R G E N C Y

M E D I C I N E

:

S C R E E N I N G

P A T I E N T S

In our experience of approximately 1000 screening cases, we had over 100 confirmed cases of SARS. We found that there were no cases of secondary spread among the suspicious cases followed at home with strict quarantine instructions.

that clinical examination of the mouth, throat and chest usually revealed no abnormality even in patients with advanced stages of SARS-pneumonia, the importance of regular CXRs with or without CT cannot be over-emphasized.

All cases with confirmed SARS were admitted.

Very few children attended our clinic, but SARS in children has been reported [31] and is discussed in a separate chapter (Chapter 12).

Guidelines for screening high-risk contact and low-risk non-contact subjects have recently been published, although there is currently little published evidence justifying them [8]. We suggest modified guidelines based on our local experience. All of our subjects were likely to have some contact with a SARS case and so, unlike the general population where there may be no known contact, our cohort were at high risk of contracting the disease. In view of the fact that symptoms were often vague, that high-temperature readings may be less common in the early stages,

Guidelines for screening contact subjects For any subject who has been in contact with a person with SARS-pneumonia in the past 10 days and where SARS-pneumonia is a concern, all patients have basic observations, history and examination.

Does the patient have any one of temperature 37.5°C, chills, rigors, myalgia or loss of appetite?

Yes

No

Mx 1. Discharge with hygiene advice and advice to return if symptoms of chills, rigors, myalgia or loss of appetite return or pyrexia 37.5°C.

Signs of consolidation

Mx 2. Admit to SARS-screening ward.

Request CXR

No signs of consolidation

Mx 3. Daily follow-up monitoring for chills, rigors, myalgia or loss of appetite return, pyrexia 37.5°C and CXR.

If symptomatic or pyrexia for 48 hours

If asymptomatic and no pyrexia for 48 hours

Make Mx 1.

Mx 4. Request HRCT.

If HRCT shows no consolidation, make Mx 3.

If HRCT shows consolidation, Make Mx 2.

S T A F F

Staff safety It was clear that hospitals and health care workers are particularly at risk, accounting for 50% of all cases in some early reports. The health care sector has to be particularly prepared as this is most likely where an outbreak may start and restart [32]. Revision of infection control with meticulous attention to detail is important [33–35]. Staff working in, and patients attending, the clinic were issued guidelines regarding personal hygiene. All staff wore PPE (Figure 2.8) and were issued with guidelines regarding dressing and removal of the equipment. There is little evidence base for these procedures and other hospitals which have not been affected with the high number of affected staff and patients have adopted less rigid procedures. However, all institutions emphasize that changing gloves and hand washing after each new patient encounter is vital. Surface cleansing was undertaken with dilute hypochlorite solution four times per day. Staff were also encouraged to take showers after going off duty and in some hospitals they were encouraged to shower after each ‘potential SARS’ case. Some health care epidemiologists may have the view that SARS is no more serious than the usual winter

Fig. 2.8

25

S A F E T Y

Staff wearing PPE.

influenza outbreaks, as many more patients will die from influenza than SARS each year. However, in the last few decades of medicine, we have not seen a health care system paralysed for months by one infectious agent [1]. We therefore re-emphasize that staff safety and strict infection control measures are essential in combating SARS, should the outbreak recur.

Key points Staff safety Essential issues are as follows:

• Staff must wash hands and change gloves • •

after every patient. Staff must shower if they come into contact with vomit, urine or faeces or if a patient directly coughs on them. Staff must wear masks when interviewing patients.

Optional issues are as follows:

• Hats, gowns and N95 masks. • Antiseptic scrubs.

26

R O L E

O F

E M E R G E N C Y

Personal hygiene guidelines for staff working in the screening clinic All staff were advised:

• to wash hands with liquid soap before and after patient contact, and after removing gloves;

• to change gloves between patients and to wash hands;

• to wear a mask in clinic at all times; • to wear a mask out of hospital if they were • • • • • •

• • • • • •

in contact with anyone with respiratory symptoms or fever; to wear gloves for all direct patient contact; to wear a gown in clinic at all times; to wear eye protection (e.g. goggles); to avoid aerosols and use of nebulizers; to clean surfaces regularly with disinfectant; to seek medical protection promptly, if they had symptoms compatible with SARS (e.g. fever, chills, myalgia, shortness of breath and difficulty breathing; to build up good body immunity with proper diet, regular exercise, rest, reduced stress and to avoid smoking; to maintain good ventilation; to avoid crowded places with poor ventilation; to know how to respond when splashed by respiratory secretions (should ask for immediate relief and go washing); that, after hand washing, to use paper towel (not elbow) to turn off the tap; to know their “Shift Infection Control Officer” who would conduct random audits.

Personal hygiene guidelines for patients attending the screening clinic

• To wear a mask in clinic at all times. • To wear a mask out of hospital, if they had

• • •

respiratory symptoms or fever or were in contact with anyone with respiratory symptoms or fever. To wash hands with liquid soap before and after patient contact, and after removing gloves. To clean surfaces regularly with disinfectant. To seek medical protection promptly, if they had symptoms compatible with SARS

M E D I C I N E

• • • •

:

S C R E E N I N G

P A T I E N T S

(e.g. fever, chills, myalgia, shortness of breath and difficulty breathing). To build up good body immunity with proper diet, regular exercise, rest, reduced stress and to avoid smoking To maintain good ventilation. To avoid crowded places with poor ventilation. To pay attention to a hygiene information sheet.

Guidelines on use of PPE When putting on PPE 1. 2. 3. 4. 5.

Wash hands Put on cap/face-shield Put on visor/mask Put on gown Put on gloves

Removing PPE (dirty gown to dispose) 1. 2. 3. 4. 5. 6.

Remove cap/face-shield Remove gown Remove glove Wash hands Remove visor/mask Wash hands

Removing PPE (clean gown to keep for re-use) 1. Remove cap/face-shield 2. Remove glove 3. Remove gown – fold inside-out and keep in plastic bag 4. Wash hands 5. Remove visor/mask 6. Wash hands

Managing patients with suspected SARS One of the first principles of medical practice is ‘do no harm’. It was quickly apparent that hospitals with communal, congested wards were a hazard to any patient or staff member who entered the ward and who did not have the illness. Patients with suspected SARS will not necessarily have the illness and therefore should not be admitted to high-risk communal areas. These patients should be isolated from

27

C O N C L U S I O N

relatives, other members of the community and from patients with SARS. The question is where this can best be achieved? If hospitals can offer protected, single-occupant areas, then suspected cases may be admitted to such areas. But if hospitals cannot offer these areas to each individual then they are highrisk areas as patients are at a risk of contracting the disease in the hospital. Our strategy was to follow up all suspect cases in our screening clinic on a daily basis until either they became probable cases in which case they were admitted or there was a 48-hour period of no symptoms, normal laboratory results and normal CXR. This process may become easier with the development of a more reliable rapid laboratory test. However, sensitivity of a single test is unlikely to be high early in the illness and the principles of managing suspected SARS patients still apply.

Hospital admission criteria At our institution the initial primary criteria for admission to a hospital clearing ward was either a history of close contact with a SARS patient in the previous 10 days, fever and chills and rigors, and documented pyrexia (38°C) or a history of fever, chills and rigors, and any one of four abnormalities: an oxygen saturation of 95%; abnormal CXR; unstable haemodynamics or abnormal blood results. No patients had unstable haemodynamics during the assessment phase and only one had an oxygen saturation of 95%. It also quickly became evident that many patients had brief transient episodes of lymphopaenia or thrombocytopaenia. These patients did not develop SARS-pneumonia and appeared to have a typical viral episode that quickly resolved. Therefore, these cases of suspect SARS were not admitted to hospital but were advised to isolate themselves from their families. Many of these staff moved into hospital accommodation or hotels. Only probable cases of SARS were admitted to a hospital ward.

Follow-up criteria Patients were followed up daily after first attendance if there was a contact history, one or more symptoms (as described in the data collection section below), documented pyrexia (38°C) on at least one occasion,

a normal or indeterminate CXR, or abnormal investigations (e.g. leucopaenia, lymphopaenia, monocytosis or thrombocytosis). Patients were given hygiene advice and a follow-up appointment for the next day.

Discharge criteria Patients were discharged after the first attendance if they had vague or no symptoms, no pyrexia, a normal CXR and normal laboratory investigations. They were given hygiene advice and advised to return if they experienced a fever. Patients who were followed up daily were discharged after 48 hours of remaining asymptomatic, with no documented pyrexia and normal CXRs and laboratory tests.

Key points Managing patients with suspected SARS

• Many patients suspected of SARS do not have SARS.

• Suspected patients should not be isolated in communal areas.

• Suspected SARS patients should be isolated in single-room areas. Admission criteria

• Patients with pneumonia require close observation and isolation. Discharge criteria Patients with the following may be discharged:

• • • •

No symptoms for 48 hours. No pyrexia for 48 hours. Normal CXR. Normal whole blood count.

Conclusion With the global spread of disease, it is likely that other health care settings will also be faced with the dilemma screening symptomatic staff and patients who have close contact with SARS-pneumonia patients after a SARS outbreak [3]. This chapter describes process of care in a screening clinic which provided a safe environment for staff,

28

R O L E

O F

E M E R G E N C Y

patients and their contacts with no secondary infections. We have learnt that hospitals are dangerous places to be in, if you do not have SARS. Unless facilities for mass isolation rooms are available, it may be better to have home quarantine to avoid suspected cases who do not actually have SARS going on to contract the disease after admission to a communal SARS or infectious triage ward. The early phase of the illness is characterized by non-specific symptoms such as fever, myalgia and chills. Documented fever is often not possible despite the presence of pneumonic change on X-ray. In the future, it will be important to have a heightened awareness of the condition, to produce a vaccine, and to immunize the majority of the population. Until this is possible, SARS remains a threat.

References 1. Cameron PA, Rainer TH and Smit D. The SARS epidemic: lessons for Australia. Med J Aus 2003; 178(10): 478–479. 2. Cameron PA and Rainer TH. Update on emerging infections: news from the centers for disease control and prevention. Ann Emerg Med 2003; 42: 1–8. 3. Cameron PA. The plague within: an Australian doctor’s experience of SARS in Hong Kong. Med J Aus 2003; 178 (10): 512–513. 4. Tomlinson B and Cockram C. SARS: experience at Prince of Wales Hospital, Hong Kong. Lancet 2003; 361: 1486–1487. 5. Rainer TH and Smit D. Emergency medicine and trauma systems. Emerg Med 2003; 15: 11–17. 6. WHO. Severe Acute Respiratory Syndrome (SARS): Multicountry Outbreak – Update, 16 March 2003: http://www. who.int/csr/don/2003_03_16/en/ (accessed 7 April 2003). 7. Tsang KW, Ho PL, Ooi GC, Yee WK, Wang T, Chan-Yeung M et al. A cluster of cases of severe acute respiratory syndrome in Hong Kong. New Eng J Med 2003; 348: 1977–1985. 8. Rainer TH, Cameron PA, Smit P deV et al. Evaluation of the WHO criteria for identifying patients with severe acute respiratory syndrome (SARS) pneumonia out of hospital: prospective observational study. Br Med J 2003; 326(7403): 1354–1358. 9. WHO. Cumulative Number of Reported Cases (SARS) from 1st November 2002 to 9th May 2003: http://www.who.int/ csr/sarscountry/2003_05_09/en/ (accessed 9 May 2003). 10. Booth CM, Matukas LM, Tomlinson GA et al. Clinical features and short-term outcomes of 144 patients with SARS in the Greater Toronto area. J Am Med Assoc Exp 2003; 289 (21): 1–9. 11. Lee N, Hui D, Wu A, Chan P, Cameron P, Joynt G et al. A major outbreak of severe acute respiratory syndrome (SARS) in Hong Kong. New Eng J Med 2003; 348: 1986–1994. 12. Poutanen SM, Low DE, Henry B, Finkelstein S, Rose D, Green K et al. Identification of severe acute respiratory syndrome in Canada. New Eng J Med 2003; 348: 1995–2005. 13. Wong RSM. Severe acute respiratory syndrome in a doctor working at the Prince of Wales Hospital. Hong Kong Med J 2003; 9: 202–205. 14. Reilley B, Van Herp M, Sermand D and Dentico N. SARS and Carlo Urbani. New Engl J Med 2003; 348: 1951–1952.

M E D I C I N E

:

S C R E E N I N G

P A T I E N T S

15. Masur H, Emanuel E and Lane HC. Severe acute respiratory syndrome: providing care in the face of uncertainty. J Am Med Assoc 2003; 289: 10–12. 16. Razum O, Becher H, Kapaun A and Junghanss T. SARS, lay epidemiology and fear. Lancet 2003; http://image.thelancet. com/extras/03cor4133web.pdf 17. WHO. Severe acute respiratory syndrome. Wkly Epidemiol Rec No. 14. 2003; 78: 97–120. 18. WHO. Case Definitions for Surveillance of Severe Acute Respiratory Syndrome (SARS): http://www.who.int/csr/sars/ casedefinition/en/ (accessed on 16 May 2003). 19. CDC. Updated Interim US Case Definition for Severe Acute Respiratory Syndrome (SARS): http://www.cdc.gov/ncidod/ sars (accessed 23 May 2003). 20. Ho W, the Hong Kong Hospital Authority Working Group on SARS and the Central Committee on Infection Control. Guideline on management of severe acute respiratory syndrome (SARS). Lancet 2003; 361: 1313. 21. Brown EG and Tetro JA. Comparative analysis of the SARS coronavirus genome: a good start to a long journey. Lancet 2003; http://image.thelancet.com/extras/03cmt124web.pdf 22. Drosten C, Gunther S, Preiser W, van der Werf S, Brodt H-R, Becker S et al. Identification of a novel coronavirus in patients with severe acute respiratory syndrome. New Eng J Med 2003; 348: 1967–1976. 23. Falsey AR, Walsh EE. Novel coronavirus and severe acute respiratory syndrome. Lancet 2003; http://image.thelancet. com/extras/03cmt87web.pdf 24. Ksiazek TG, Erdman D, Goldsmith C, Zaki SR, Peret T, Emery S et al. A novel coronavirus associated with severe acute respiratory syndrome. New Eng J Med 2003; 348: 1953–1966. 25. Peiris JSM, Lai ST, Poon LLM, Guan Y, Yam LYC, Lim W et al. Coronavirus as a possible cause of severe acute respiratory syndrome. Lancet 2003; 361: 1319–1325. 26. Rota PA, Oberste MS, Monroe SS et al. Characterization of a novel coronavirus associated with severe acute respiratory syndrome. Sciencexpress 2003; www.sciencexpress.org/ 1May2003/Page1/10.1126/science.1085952 27. Ruan Y-J, Wei CL, Ee LA et al. Comparative full-length genome sequence analysis of 14 SARS coronavirus isolates and common mutations associated with putative origins of infection. Lancet 2003; http://image.thelancet.com/extras/ 03art4454web.pdf 28. Marra MA, Jones SJM, Astell CR et al. The genome sequence of the SARS-associated coronavirus. Sciencexpress 2003; www.sciencexpress.org/1May2003/Page1/10.1126/ science.1085953 29. Yuen E, Kam CW and Rainer TH. Role of absolute lymphocyte count in the screening of patients with suspected SARS. Emerg Med 2003 15: 395–396. 30. Wong RSM, Wu A, To KF, Lee N, Lam CWK, Wong CK, Chan PKS, Ng MHL, Yu LM, Hui DS, Tam JS, Cheng G, Sung JJY. Haematological manifestations of patients with severe acute respiratory syndrome: retrospective analysis. Br Med J 2003 326(7403): 1358–1362. 31. Hon KLE, Leung CW, Cheng WTF, Chan PKS, Chu WCW, Kwan YW et al. Research letters: clinical presentations and outcome of severe acute respiratory syndrome in children. Lancet 2003; 361: 1701–1703. 32. Kondro. SARS back in Canada. Lancet 2003; 361(9372): 1876. 33. Li TST, Buckley TA, Yap FHY, Sung JJY and Joynt GM. Severe acute respiratory syndrome (SARS): infection control. Lancet 2003; 361: 1386. 34. Seto WH, Tsang D, Yung RWH et al. Effectiveness of precautions against droplets and contact in prevention of nosocomial transmission of severe acute respiratory syndrome (SARS). Lancet 2003; 361: 1519–1520. 35. Yang W. Severe acute respiratory syndrome (SARS): infection control. Lancet 2003; 361: 1386–1387.

Severe Acute Respiratory Syndrome Outbreak in a University Hospital in Hong Kong N Lee and JJY Sung

Epidemiology: University Hospital experience Diagnosis of SARS Clinical features

29 29 30

Epidemiology: University Hospital experience In March 2003, there was an outbreak of atypical pneumonia in Hong Kong [1,2] and our institution was at its epicentre. Epidemiological investigations revealed that the initial outbreak at our institution was related to a single index case admitted to one of our medical wards. In the early phase of the outbreak, the index case infected

• 55 health care workers in the same ward; • 16 medical students who clinically examined the index case;

• 54 patients who were either nursed in the same ward or had visited their relatives. It is postulated that the use of nebulized salbutamol for muco-ciliary clearance may have potentiated its transmission [2].

Diagnosis of SARS At the time of writing, the diagnosis of severe acute respiratory syndrome (SARS) is still based on clinical

CHAPTER

3

Laboratory features Clinical outcomes and prognostic factors Virological testing for SARS

30 31 31

and epidemiological information as in our previously reported cohort [2]. According to the WHO case definition [1], patients are classified as ‘suspect’ or ‘probable’ cases, as discussed in the earlier chapter (Chapter 1). Cases are excluded if an alternative diagnosis can fully explain their illness [3]. However, clinicians are advised that patients should not have their case definition category downgraded while awaiting results of laboratory testing or on the bases of negative results. It must be emphasized that the hallmark of this illness is an initial viral pneumonia followed by a ‘reactive phase’ producing inflammatory changes resembling bronchiolitis obliterans organizing pneumonia (BOOP), subsequently progressing to acute respiratory distress syndrome (RDS)/respiratory failure [4]. Thus, radiological investigations play a major role in the diagnosis and management of this disease.

Key points

• Diagnosis is based on clinical grounds. • Mainly a diagnosis of exclusion (at the time of writing).

30

S A R S

The subsequent part of the chapter will discuss the clinical, laboratory features, and prognostic factors of the initial cohort of confirmed SARS patients admitted to our institution. The current status of virological tests will also be discussed briefly.

Physical examination on admission revealed high body temperature in most patients (median body temperature 38.4°C, range 35–40.3°C). Inspiratory crackles could be heard at the base of the lungs and wheezing was universally absent. Skin rash, lymphadenopathy and purpura were not found. As the disease progressed into 2nd week, high spike of fever, non-productive cough, shortness of breath and in some cases re-emergence of diarrhoea, became the more prominent features [5]. One must note that progression into RDS and respiratory failure may develop during this time.

•

• • •

• •

[2,6]

white cell count 3.5  109/L) in 33.9% of patients. While neutrophil (median 3.45  109/L; range 0.5–11.8  109/L) and monocyte counts were normal in most cases, moderate lymphopaenia (absolute lymphocyte 1.0  109/L) was found in 69.6%. Thrombocytopaenia (platelet count 150  109/L) was documented in 44.8% of patients on presentation. The lymphocyte count continued to drop within the first 2 weeks after admission. Prolonged activated partial thromboplastin time (APTT: 38 seconds) was noted in 42.8%, while the prothrombin time remained normal in most cases. In 45.0% of patients D-dimer levels were also elevated. Reactive lymphocytes were detected in peripheral blood films in 15.2% of cases.

• Serum transaminase levels (alanine aminotrans•

• Fever, chills and/or rigor, myalgia, cough,

•

•

K O N G

Serum chemistry was normal in the majority of cases. There were, however, several abnormalities found in a substantial proportion of patients:

•

•

H O N G

• Initial blood count showed leucopaenia (total

Key points headache and dizziness are the main presenting symptoms in the 1st week of illness. Resurgence of fever, non-productive cough, shortness of breath and diarrhoea are the main features in the 2nd week. Progression into RDS and respiratory failure may occur.

I N

Laboratory features

Clinical features The clinical features of SARS have been remarkably consistent among all reported cohorts [2,5–7]. In brief, the time interval between exposure to the onset of fever ranged from 2 to 16 days. The median incubation period is approximately 6 days. In the cohort of patients at our institution, the most common symptoms at presentation were fever (100%), chills and/or rigor (73.2%), myalgia (60.9%), cough (57.3%), headache (55.8%) and dizziness (42.8%). Less common symptoms included sputum production (29%), sore throat (23.2%), coryza (22.5%), nausea and vomiting (19.6%), and diarrhoea (19.6%).

O U T B R E A K

ferase (ALT) 45 IU/mL) were elevated in 23.4% of patients (mean 60.4  150.4 IU/mL). Creatinine phosphokinase (CPK) levels were elevated in 32.1% of patients (median 126 U/L, range 29–4644). Creatinine kinase MB (isoenzyme) (CK-MB) and troponin-T were assayed in those with elevated CPK levels and none was found to be abnormal, indicating that the source of CPK was unlikely to be from cardiac muscles. Lactate dehydrogenase (LDH) level was elevated in 71.0% of patients. Hyponatraemia (sodium 134 mmol/L) was documented in 20.3% of patients and hypokalaemia (potassium 3.5 mmol/L) in 25.2% of patients.

As the disease progresses, LDH level increased and was believed to correlate well with disease severity. It has been postulated that as the disease progresses, LDH is released into the circulation after pulmonary inflammation and necrosis [2,6]

V I R O L O G I C A L

T E S T I N G

F O R

Key points

• Initial lymphopaenia and its subsequent • •

progression are the main haematological features. LDH level is helpful in the initial diagnosis as well as disease monitoring. Thrombocytopaenia, prolonged APTT, elevated D-dimer, ALT and CPK levels are the other important laboratory features of SARS.

Clinical outcomes and prognostic factors Univariate analysis of our cohort revealed that:

• advanced age, male gender, peak CPK value, LDH

•

31

S A R S

on presentation and its peak value, higher initial absolute neutrophil count and low serum sodium levels were significant predictive factors for intensive care unit (ICU) admission and mortality [2]. The presence of co-morbidities appeared to confer a worse clinical outcome in another cohort [6].

With multivariate analysis, advanced age (odds ratio (for every 10 years)
1.80; 95% confidence interval (CI)
1.16–2.81; P
0.009), high peak LDH level (odds ratio (for every 100 U/L)
2.09; 95% CI
1.28–3.42; P
0.003), and higher absolute neutrophil count on presentation (odds ratio
1.60; 95% CI
1.03–2.50; P
0.04) were independent predictive factor of the adverse outcomes [2].

Key points

suggested that it may be non-human in origin. Now there are several laboratory tests available for the virological diagnosis, namely serology for SARS-CoV (by immuno-flourescent assay (IFA) or enzyme-linked immuno-sorbent assay (ELISA)), virus isolation (cell culture), and reverse transcriptase polymerase chain reaction (RT-PCR) on clinical specimens of different sites (including nasopharyngeal aspirate, nasal swab, throat gargle, sputum, stool, urine, blood, etc.). According to World Health Organization (WHO) [2], positive SARS diagnostic test findings are either confirmed by: 1. RT-PCR on at least two different clinical specimens (e.g. nasopharyngeal and stool) or the same clinical specimen collected on 2 or more days during the course of the illness (e.g. two or more nasopharyngeal aspirates), or two different assays or repeat PCR using the original clinical sample on each occasion of testing. 2. Seroconversion by ELISA or IFA: negative antibody test on acute serum followed by positive antibody test on convalescent serum, or four-fold or greater rise in antibody titre between acute and convalescent phase sera tested in parallel. 3. Virus isolation: isolation in cell culture of SARSCoV from any specimen plus PCR confirmation using a validated method. It is now generally accepted that absence of antibody to SARS-CoV in convalescent serum obtained 21 days after symptom onset should suggest absence of infection. However at the time of writing, there is no properly evaluated rapid diagnostic test for SARS-CoV infection. As a result, clinical, epidemiological and radiological information [11] are crucial in the diagnosis and management of SARS.

Prognostic indicators for adverse clinical outcome:

• advanced age, • high peak LDH level, • high initial neutrophil count.

Virological testing for SARS The cause of SARS is ascribed to a novel coronavirus (SARS-CoV) [2,3,8–10]. It is genetically distinct from other known coronaviruses, and early evidence also

Key points

• Serological testing (IFA or ELISA) for SARS-

•

CoV is generally considered as the ‘gold standard’; cases are excluded when seronegativity is confirmed 3 weeks after symptom onset. RT-PCR offers rapid diagnosis; though multiple clinical specimens are needed to increase the sensitivity and specificity.

32

References 1. WHO. Severe Acute Respiratory Syndrome (SARS): http://www.who.int/csr/sars/en/ (accessed at 14 May 2003). 2. Lee N, Hui DS, Wu A et al. A major outbreak of severe acute respiratory syndrome in Hong Kong. New Engl J Med 2003; 348:1986–1994. 3. CDC. Severe Acute Respiratory Syndrome (SARS) Updated Interim Case Definition: http://www.cdc.gov/ncidod/sars/ casedefinition.htm (accessed at 5 June 2003). 4. Epler GR. Bronchiolitis obliterans organizing pneumonia. Arch Intern Med 2001; 161: 158–164. 5. Peiris JS, Chu CM, Cheng VC et al. Clinical progression and viral load in a community outbreak of coronavirus-associated SARS pneumonia: a prospective study. Lancet 2003; 361: 1767–1772. 6. Booth CM, Matukas LM, Tomlinson GA et al. Clinical features and short-term outcomes of 144 patients with SARS in the greater Toronto area. J Am Med Assoc 2003; 289: 2801–2809.

S A R S

O U T B R E A K

I N

H O N G

K O N G

7. Tsang KW, Ho PL, Ooi GC et al. A cluster of cases of severe acute respiratory syndrome in Hong Kong. New Engl J Med 2003; 348: 1977–1985. 8. Peiris JSM, Lai ST, Poon LLM et al. Coronavirus as a possible cause of severe acute respiratory syndrome. Lancet 2003; 361: 1319–1325. 9. Drosten C, Gunther S, Preiser W et al. Identification of a novel coronavirus in patients with severe acute respiratory syndrome. New Engl J Med 2003; 348: 1967–1976. 10. Ksiazek TG, Erdman D, Goldsmith CS et al. A novel coronavirus associated with severe acute respiratory syndrome. New Engl J Med 2003; 348: 1953–1966. 11. Wong KT, Antonio GE, Hui DS et al. Thin-section CT of severe acute respiratory syndrome: evaluation of 73 patients exposed to or with the disease. Radiology 2003: http://radiology.rsnajnls.org/cgi/content/full/2283030541v1 (accessed 12 May 2003).

CHAPTER

Imaging of Pneumonias GC Ooi and PL Khong

Introduction Radiographic pattern Mycobacterium tuberculosis

33 34 43

Introduction Pneumonia is the sixth most common cause of death, and the leading cause of death from infection in the US [1,2]. The diagnosis of pneumonia requires combination of careful clinical evaluation, appropriate laboratory investigations including microbiological tests and radiological confirmation of pneumonia. The chest radiograph is the cornerstone of radiological diagnosis, and the imaging modality of choice in establishing the presence of pneumonia, including its severity and extent. The efficacy of the clinical examination in detecting pneumonia is acknowledged to be less sensitive, with auscultatory evidence of pneumonia reported to be absent in a quarter of patients [3]. In addition, Osmer and Cole found that stethoscopic findings were often not in concordance with radiographic findings [3]. As this book deals with imaging of severe acute respiratory syndrome (SARS), a predominantly pneumonic illness, this chapter serves to provide a background to the radiological appearances of pneumonias of different aetiology. The main radiographic feature of any pneumonia is consolidation, defined as an opacity that obscures vascular markings, which can range from small ill-defined

Fungal pulmonary infections Utility of CT

4 48 49

areas to larger areas involving one or more lobes. The pulmonary acinus is the smallest airspace unit visible on the radiograph (4–9 mm in diameter) and is that part of the lung distal to the terminal bronchiole including respiratory bronchioles, alveolar ducts and alveolar sacs. Filling of the acini by fluid, whether by inflammatory exudates, transudates or blood will result in patchy areas of consolidation that could either be restricted to the secondary pulmonary lobule appearing as patchy areas of bronchopneumonia, or involve the whole lobe as lobar pneumonia. As connective tissue (interstitium) separates the secondary pulmonary lobule from each other, infective processes that affect the interstitium will produce reticular opacities on the chest radiograph. Similar opacities are however seen with pulmonary oedema and inflammatory processes including the various interstitial pneumonitides that result in lung fibrosis [4,5]. The radiographic appearances of pneumonia can be broadly classified into bronchopneumonia, lobar pneumonia or interstitial pneumonia based on the type of morphologic change at the level of the secondary pulmonary lobule [6]. The radiographic manifestations of an infective process are also affected by a number of factors including age, immunological

34

I M A G I N G

status of the host, and pre- or co-existing lung conditions [7,8]. Hence, the clinical settings under which pneumonias occur are an important consideration in the diagnostic algorithm, and can be categorised as community-acquired infection, nosocomial (hospital-acquired) infection and infection occurring in immuno-compromised hosts [9,10]. In the appropriate clinical setting the radiographic pattern can be helpful at narrowing the diagnosis to a few differential possibilities, although the radiographic features are usually not specific to any aetiological agent as an overlap often occurs.

O F

P N E U M O N I A S

and particularly in Klebsiella pneumoniae infections, voluminous inflammatory exudates result in the expansion of the involved lobe (Figure 4.2), often accompanied by the bulging fissure sign [13].

• Diagnosis of pneumonia requires combina-

•

•

tion of careful clinical evaluation, appropriate laboratory investigations including microbiological tests and radiological confirmation of pneumonia. The chest radiograph is the cornerstone of radiological diagnosis and the imaging modality of choice in establishing the presence of pneumonia, including its severity and extent. The radiographic appearances of pneumonia can be broadly classified into bronchopneumonia, lobar pneumonia or interstitial pneumonia.

Fig. 4.1 Posterior–anterior chest radiograph in a 4-year-old child with Streptococcus pneumoniae. There is homogeneous opacification of the right upper lobe consistent with a lobar pneumonia.

Radiographic pattern Lobar pneumonia Lobar pneumonia is characterised by rapid production of oedema in the distal airspaces (alveoli) with relatively minimal cellular reaction occurring initially in the subpleural regions of the lungs [11]. The fluid spreads from acinus to acinus through the pores of Kohn and canals of Lambert, ultimately to involve the whole lobe. This mechanism of spread respects pleural boundaries, and as the airways are largely spared, with minimal volume loss. Radiographically lobar pneumonia manifests as nonsegmental homogeneous consolidation (Figure 4.1) involving predominantly or exclusively one lobe [12]. As the larger airways remain patent, airbronchograms are common associations. If untreated

Fig. 4.2 One-year-old child with respiratory bronchiolitis secondary to respiratory syncytial virus. He developed superadded bacterial infection. Klebsiella pneumoniae was isolated from his sputum. The posterior–anterior chest radiograph shows right upper lobe consolidation with slight bulging of the horizontal fissure. Note hyperinflation and perihilar peribronchial infiltration consistent with respiratory bronchiolitis.

R A D I O G R A P H I C

35

P A T T E R N

Classically, lobar pneumonia occurs from infections with Streptococcus pneumoniae. Other organisms that also give rise to lobar pneumonias include the aforementioned Klebsiella pneumoniae and Legionella pneumoniae. Less common organisms causing lobar pneumonia include Mycobacterium tuberculosus, Actinomycosis and Nocardia species, Pseudomonas aeruginosa and Escherichia coli.

Legionella pneumophila • Aerobic Gram-negative cocco-bacilli that infects humans from their natural habitat, which is water.

• Outbreaks result from infection via infected water sources such as cooling water towers,

Streptococcus pneumoniae • Gram-positive bacteria are found in 20% of the human population as commensal organisms [14].

• The most common community-acquired bacter-

• •

• •

ial pneumonia [15–17], found in 26–78% of all cases of pneumonia [17], and accounts for up to 40% of all isolated species in hospitalised patients with pneumonia [16]. Cause of death in 15–45% of communityacquired pneumonias that required admission to the intensive care unit (ICU) [18]. Classical radiographic feature is lobar pneumonia (Figures 4.1 and 4.3). Other patterns include bronchopneumonia (20–70%), round pneumonia, and mixed airspace and interstitial opacities (13–22%) [19–23]. Cavitation, pneumatocoele formation and pulmonary gangrene are unusual complications that suggest polymicrobial infection. Parapneumonic effusions are associated with bacteraemia.

(a)

Klebsiella pneumoniae (Friedlander’s pneumonia) • Gram-negative bacteria found in the oral flora. • Immuno-suppressed individuals due to organ

• • • •

transplantation or cytotoxic chemotherapy, chronic debilitating disease and chronic alcoholism are at risk. Nosocomial rather than community acquired. Shares similar radiographic appearances as pneumococcal pneumonia (Figure 4.2). Bulging lobe and fissure now less common in present modern-day antibiotic era. Cavitation (30–50%) [13,24,25], pleural effusions and empyema [24,26] occur more frequently than pneumococcal pneumonia.

(b)

Fig. 4.3 A 48-year-old female with multilobar consolidation due to Streptococcus pneumoniae. (a) Posterior–anterior and (b) lateral chest radiographs showing homogeneous consolidation in the right middle lobe and patchy bronchopneumonia in the right lower lobe.

36

• • •

•

• •

I M A G I N G

air-conditioning systems, shower heads and hot water storage tanks, rather than person-to-person [9,27–29]. Immuno-compromised and elderly male patients are most susceptible [30–32]. Accounts for 2–25% of community-acquired infections that required hospitalisation [33–35], and 1–40% of nosocomial infections [34]. Initial radiographic appearances are similar to those of pneumococcal pneumonia, with patchy consolidation starting at the lung periphery followed by progression to involve the whole lobe [36,37]. Disease progression, despite antimicrobial therapy, is faster than that found in pneumococcal pneumonia [38,39], with multilobar and bilateral lung involvement [37,38,39]. Cavitation is rare in immuno-competent patients [36,37] and more common in immunocompromised hosts [40,41]. Pleural effusions are common as disease progresses in up to two-third of cases [37,40,41].

• Lobar pneumonia manifests as non-

• • •

•

segmental homogeneous consolidation involving predominantly or exclusively one lobe, commonly associated with airbronchograms. Streptococcus pneumoniae, Klebsiella pneumoniae and Legionella pneumoniae are common pathogens causing lobar pneumonia. Streptococcus pneumoniae is the most common community-acquired bacterial pneumonia. Klebsiella pneumoniae causes nosocomial rather than community-acquired pneumonia, and are more associated with cavitations, empyema and effusions than streptococcus pneumoniae. Legionella pneumoniae outbreaks result from water-borne infection rather than personto-person transmission.

Bronchopneumonia In bronchopneumonia, there is relatively intense inflammatory exudate consisting primarily of

O F

P N E U M O N I A S

Fig. 4.4 Anterior–posterior radiograph of a child with bronchopneumonia secondary to Streptococcus pyogenes. There is ill-defined fluffy consolidation in the right lung.

polymorphonuclear leucocytes with oedema centred at the terminal and respiratory bronchioles. This infective process spreads along the intralobular airways until all the pulmonary lobules are involved. This gives rise to the radiographic (Figure 4.4) and high-resolution computed tomography (HRCT) appearance of ill-defined fluffy centrilobular nodular opacities (Figure 4.5), which may coalesce to form a lobar pattern of consolidation, indistinguishable from lobar pneumonia [42,43]. However careful search in other areas of the lungs may reveal areas of volume loss and segmental (lobular) distribution of abnormalities that indicates bronchopneumonia. Bronchiolar involvement can be appreciated on HRCT as branching opacities with tree-in-bud appearance. The quintessential bronchopneumonia is exemplified by Staphylococcus aureus infection. As bronchopneumonia is associated with tissue destruction, complications such as pulmonary abscess, pulmonary gangrene and pneumatocoele formation are quite common. Other organisms that can give rise to bronchopneumonia include most Gram-negative bacteria such as Pseudomonas aeruginosa and Haemophilus influenzae, Streptococcus pyogenes and Escherichia coli. Aspiration pneumonia from anaerobic bacteria such as Bacteroides, Fusobacterium and Actimomyces results in a similar bronchopneumonia

R A D I O G R A P H I C

37

P A T T E R N

(a)

(b)

Fig. 4.5 HRCT scans of a 46-year-old man with Staphylococcal pneumonia. (a) Ill-defined lobular consolidation and ground-glass opacities are noted in the superior segment of the right lower lobe, and also in the subpleural regions of the left upper lobe. (b) At a scan level caudad to (a) in the right lower lobe, a cavitating focal area of consolidation is noted.

pattern with predilection for gravity-dependent areas of the lungs. These areas include the superior segments of the lower lobes and posterior segments of the upper lobes in the recumbent position and the basal segments of the lower lobes in the erect patient (Figure 4.6). There is right lung predominance due to the orientation of the right main bronchus. Cavitation or abscess formation develops in up to 60% of patients [44].

Fig. 4.6 Posterior–anterior chest radiograph of a 67-year-old patient with cerebral infarction, who had difficulty swallowing. There are bilateral illdefined patches of consolidation in the lower lobes, consistent with aspiration pneumonia. Note nasogastric tube in situ.

• Haematogeneous route is another method of

• • • •

Staphylococcus aureus • Gram-positive coagulase-producing bacteria

spread in subacute bacterial endocarditis particularly in intravenous drug abusers (Figure 4.7), septic thrombophlebitis, and staphylococcal infection of indwelling catheters. Bronchopneumonia (Figure 4.5a) is the typical radiographic appearance although lobar pneumonia can sometimes be found. Lung abscesses are common, and if erosion into the bronchial tree occurs, air-fluid levels are seen (Figure 4.5b). Pneumatocoeles are more common in children, and are thought to be due to ball-valve obstruction in small airways. Up to 50% of staphylococcal pneumonia develop either pleural effusion of empyema; the latter is more commonly found in children [48].

which rarely affects healthy adults [16,45].

• Commonly affects debilitated hospitalised patients via aspiration of Staphylococcus aureus from upper respiratory tract [46,47].

Pseudomonas aeruginosa • Gram-negative rods, which are normal commensals in the human intestine and skin.

38

I M A G I N G

O F

P N E U M O N I A S

(a)

Fig. 4.8 A 29-year-old female with AIDS. Posterior– anterior chest radiograph shows multiple peripheral wedge-shaped consolidation. Pseudomonas aeruginosa was cultured from her sputum.

• •

(b)

Fig. 4.7 Posterior–anterior chest radiographs of a male 35-year-old intravenous drug abuser who presented with rigors and fever. (a) Ill-defined patchy consolidation is noted in both lungs with bilateral small effusions. (b) Further evaluation of the patchy consolidation reveals cavitation (arrows) within the consolidation consistent with multiple septic emboli. Staphylococcus pneumoniae was cultured from his blood.

• It is the most lethal form of nosocomial infection, accounting for 20% of nosocomial pneumonia in adult ICU patients [49]. High fatality rates are related to pre-existing diseases such as chronic

obstructive pulmonary disease (COPD), and multiorgan failure. Occasional cause for community-acquired pneumonia in patients with advanced acquired immunodeficiency syndrome (AIDS). Bronchopneumonia is the predominant radiographic pattern. Other patterns include multinodular and reticular patterns [50,51], and occasionally pulmonary infarction resembling invasive aspergillosis (Figure 4.8).

Haemophilus influenzae • Gram-negative cocco-bacilli. • Accounts for between 5% and 20% of communityacquired pneumonias.

• Risk factors include advanced age, pre-existing

•

lung diseases such as COPD, chronic alcoholism, and immuno-compromised patients with diabetes, immunoglobulin defects and AIDS [52–54]. Variable radiographic pattern. Majority (50–60%) present with bronchopneumonia (Figure 4.9), while lobar consolidation is noted in 30–50% of cases either in isolation or in combination with bronchopneumonia [20,55]. Nodular and interstitial patterns are rare, while pleural effusions are found in 50% of cases [55,56].

R A D I O G R A P H I C

39

P A T T E R N

• Organisms that commonly cause broncho-

•

pneumonia include Staphylococcus aureus, Pseudomonas aeruginosa and Haemophilus influenzae, Streptococcus pyogenes and Escherichia coli. Staphylococcus aureus rarely infects healthy individuals, and acquired either haematogeneously or from aspiration of infected upper respiratory tract secretions. Cavitations, pneumatocoeles, pleural effusions and empyemas are common associations.

Infectious interstitial pneumonia

Fig. 4.9 A 36-year-old female with communityacquired Haemophilus influenzae pneumonia. Patchy areas of bronchopneumonia are noted in both lower lobes.

Esherichia coli • Gram-negative bacilli that are commensals in the human small and large bowel.

• Affects debilitated people. Accounts for 5–20% of •

hospital-acquired and nursing-home-acquired pneumonias [57,58]. Multilobar bronchopneumonia with lower lobe predominance and pleural effusions are common, while cavitation is uncommon.

This pattern of pneumonia results from inflammatory process with oedema centred on the interstitium, and bronchiolar and airway walls giving rise to reticular, reticulo-nodular and small nodular appearance. Insidious infections will result in lymphatic infiltration of the alveolar septae without parenchymal abnormality. However with more virulent infections, there is rapid progressive pneumonia that results in diffuse alveolar damage affecting both interstitium and airspaces [59,60]. Organisms that are responsible for this pattern of pneumonia are mainly viruses, Mycoplasma pneumoniae and Pneumocystis carinii.

Viruses • Viral

• In bronchopneumonia, the infection is

•

•

centred on the terminal and respiratory bronchioles with spread along the intralobular airways until all the pulmonary lobules are involved. Typical radiographic and HRCT appearances are ill-defined fluffy centrilobular nodular opacities (Figure 4.5), which may coalesce to form a lobar pattern of consolidation, indistinguishable from lobar pneumonia. Bronchiolar involvement can be appreciated as tree-in-bud opacities on HRCT.

•

pneumonias that afflict immunocompetent hosts are influenza A and B in adults, and respiratory syncytial virus, parainfluenza virus and influenza virus in children. Type A influenza causes epidemics and all pandemics, while type B influenza results in outbreaks and is more common in schoolchildren. Respiratory syncytial virus most commonly affects infants and small children and is normally restricted to an upper respiratory tract infection. Viral pneumonias that affect immunocompromised hosts are cytomegalovirus, herpes simplex type 1 virus, varicella-zoster virus and adenovirus [61]. Cytomegalovirus infection is common in solid-organ and bone marrow transplant recipients (Figure 4.10), while varicella-zoster pneumonia usually occurs as a complication of chickenpox, although predisposing factors include

40

I M A G I N G

(a)

O F

P N E U M O N I A S

(b)

Fig. 4.10 Posterior–anterior chest radiographs of a 43-year-old man with cytomegalovirus infection obtained (a) before and (b) after treatment. Multiple small nodules, almost miliary in distribution and appearance are noted bilaterally in (a), which resolved completely in (b). Note pneumoperitoneum in (b) due to peritoneal dialysis. The patient had underdone renal transplantation 6 months previously, but suffered rejection almost immediately, and had to have the transplanted kidney surgically removed.

•

•

•

neoplastic disease, immune deficiency and pregnancy. Viral pneumonias predominantly infect the terminal and respiratory bronchioles with extension of the inflammatory processes to the adjacent interstitium resulting in an interstitial pneumonia. With more severe inflammation, filling of the alveoli with hyaline membranes and inflammatory exudates, which may be haemorrhagic, causes diffuse alveolar damage. The radiological features in viral pneumonias are variable and overlapping, and do not allow diagnosis of a specific virus. Viral pneumonias can manifest as poorly defined nodules (4–10 mm in diameter), bronchial wall thickening, peribronchial opacities (Figure 4.11), perihilar linear opacities and patchy areas of ground-glass opacities and consolidation [61]. Due to associated bronchiolitis, air-trapping is common particularly in lower respiratory tract respiratory syncytial virus infection (Figure 4.2) [62,63]. With severe fulminant viral pneumonia, there is homogeneous or patchy consolidation and

Fig. 4.11 Posterior–anterior chest radiograph of a 2-year-old boy with metapneumovirus pneumonia showing perihilar peribronchial thickening and infiltrates.

R A D I O G R A P H I C

•

•

•

ground-glass opacities (Figure 4.12), associated with poorly defined centrilobular nodules. Cavitation and pleural effusions are not prominent features. Hilar adenopathy is common with measles and varicella-zoster pneumonias but rare in other viral pneumonias. The above radiological features particularly centrilobular nodules, ground-glass opacities, consolidation (Figure 4.12) and air-trapping are best demonstrated on HRCT or CT scans [64–68]. A predominant CT pattern of consolidation and ground-glass opacities in a lobular pattern reflect diffuse alveolar damage [67,69]. Interstitial inflammatory infiltration is represented by thickened interlobular septae or as ground-glass opacities on HRCT [65,66,68,69]. Areas of inflammatory or haemorrhagic nodules, and organising pneumonia on histopathology correspond to poorly defined centrilobular nodules on HRCT [65,66,68]. Varicella-zoster and influenza pneumonias are associated with the highest frequency of centrilobular nodules on HRCT and chest radiographs, in addition to the ill-defined consolidation and ground-glass opacities. The nodules in varicellazoster infection range from 5 to 10 mm, and in less than 2% of patients these nodules may calcify as diffuse small foci throughout both lungs. Superadded bacterial infection may occur (Figure 4.2), particularly in influenza pneumonia with Streptococcus pneumoniae, Staphylococcus aureus and Haemophilus influenzae.

Mycoplasma pneumoniae • Mycoplasma pneumoniae is a common cause of

•

•

41

P A T T E R N

community-acquired pneumonia, accounting for up to 20% of pneumonias in the general population [70,71]. Its clinical presentation and radiographic appearances resemble viral pneumonias [72]. A mixed pattern of airspace and interstitial opacities, segmental (lobular) consolidation or diffuse reticulo-nodular opacities can be found (Figure 4.13) [73,74]. Pleural effusion and empyema are uncommon. The segmental distribution of consolidation, centrilobular opacities and interstitial opacities are better appreciated on HRCT than on chest radiographs [75].

(a)

(b)

Fig. 4.12 (a) Posterior–anterior chest radiograph of a 6-year-old girl with adenovirus pneumonia showing right lower zone ground-glass opacities and bronchial wall thickening. There is a dense collapse consolidation of the left lower lobe with an airfilled lucency. (b) CT scan of the same patient shows ground-glass opacities and consolidation in the right lower lobe and bronchial wall thickening. There is chronic collapse consolidation in the left lower lobe with a large cavity.

Pneumocystis carinii • Pneumocystis carinii pneumonia (PCP) is now •

considered more closely related to fungi than protozoa by most authorities [76]. It is almost exclusively a pathogen of immunocompromised hosts, accounting for approximately 60% of cases of pneumonia in AIDS patients

42

Fig. 4.13 Posterior–anterior chest radiograph of a 52-year-old male with Mycoplasma pneumonia. There are bilateral basal reticulo-nodular opacities with more confluent consolidation in the left lower zone.

•

•

•

[77], particularly in those with CD4 count of 200 cells/mm3 [78]. It is also prevalent in post-transplant recipients [79], and in patients with malignancies [80] and connective tissue disorders [81]. The classic radiographic appearance is bilateral perihilar reticular, or reticulo-nodular opacities (Figure 4.14) that rapidly progress (3–5 days) to diffuse airspace consolidation involving almost the entire lungs (Figure 4.15) [77], at which time acute respiratory distress syndrome (ARDS) may have supervened. Unusual radiographic appearances include solitary or multiple nodules that cavitate [82], lobar consolidation [77,83], pleural effusions [77,84] and enlarged mediastinal lymph nodes [85]. Pneumatocoeles develop in 10% of cases, rupture of which may lead to spontaneous pneumothorax [86]. On CT and HRCT, the predominant abnormality is ground-glass opacities, which may be diffuse (Figure 4.16) or patchy [75,87,88]. In one series using HRCT for evaluation, consolidation was not found to be a main feature of PCP [75], although in another study using CT for evaluation, consolidation was present in 40% of cases [87].

I M A G I N G

O F

P N E U M O N I A S

Fig. 4.14 A 48-year-old patient with human immunodeficiency virus (HIV) infection, who developed PCP. There are reticulo-nodular opacities around and radiating from the hilar and mediastinum.

• Resolution of ground-glass opacities and consolidation may be incomplete, leaving residual interstitial opacities suggestive of fibrosis [87,89].

• Interstitial pneumonia results from inflam-

• •

• •

matory process centred on the interstitium, and bronchiolar and airway walls giving rise to reticular, reticulo-nodular and small nodular appearance. Viruses, Mycoplasma pneumoniae and Pneumocystis carinii are common causes. Radiological features of viral pneumonias are variable and overlapping: poorly defined nodules (4–10 mm in diameter), bronchial wall thickening, peribronchial opacities, perihilar linear opacities and patchy areas of ground-glass opacities and consolidation. Cavitation, pleural effusions and hilar adenopathy are not common features of viral pneumonias. On HRCT interstitial inflammatory infiltration is represented by thickened interlobular septae or as ground-glass opacities. A predominant pattern of consolidation and

M Y C O B A C T E R I U M

43

T U B E R C U L O S I S

Mycobacterium tuberculosis Mycobacterium tuberculosus bacilli are acid-fast aerobic bacteria. The incidence of pulmonary tuberculosis (TB) was in decline until the mid-1980s when there was resurgence in numbers due to the emergence of AIDS. The radiological manifestations of pulmonary TB are dependent on a number of factors primarily age of the patient, previous exposure to TB and host immune status. In immuno-competent patients there are two distinct patterns of pulmonary TB: primary TB in individuals without previous exposure and post-primary TB in individuals with prior exposure who have acquired specific immunity.

Primary pulmonary TB Fig. 4.15 PCP in a 32-year-old man who had to be mechanically ventilated. There is diffuse consolidation affecting both the lungs, with air-bronchograms. This appearance is indistinguishable from ARDS.

• Droplet infection is the mode of transmission, with

•

•

Fig. 4.16 HRCT scan showing diffuse groundglass opacification in a patient with PCP.

•

ground-glass opacities in a lobular pattern reflects diffuse alveolar damage. Pneumocystis carinii is almost exclusively a pathogen of immuno-compromised hosts. It classically appears as bilateral perihilar reticular, or reticulo-nodular opacities which rapidly progress to diffuse airspace consolidation.

• • •

droplets containing Mycobacterium tuberculosus, infecting the gravity-dependent regions of the lungs from which infection disseminates. Primary pulmonary TB can involve the tracheobronchial tree, lung parenchyma, mediastinal and hilar lymph nodes or pleura. Children with primary pulmonary TB may be clinically asymptomatic and without radiographic abnormalities while adults are usually symptomatic. The radiographic appearance of the primary focus (Ghon focus) is non-specific and can range from ill-defined airspace opacities to bronchopneumonia (Figure 4.17) and lobar consolidation [90]. Right upper lobe is the most commonly involved, while the right middle lobe is the least commonly involved [91]. It is multifocal in 25% of cases and bilateral in 10% [91,92] (Figure 4.18). Pleural effusion is more common in adults, while enlarged mediastinal and hilar lymph nodes are more frequently present in children [91–93]. On CT, peripheral rim enhancement of the lymph nodes is present. The central low attenuation areas represent caseation necrosis. Cavitation is found in 8–29% of primary TB (Figure 4.19) [91,94]. Endobronchial spread occur after breakdown of a lobar infection or rupture of infected lymph node into a bronchus. Radiographically, it is noted as ill-defined 5–10 mm centrilobular nodules (Figure 4.19) [95], while on HRCT, poorly defined

44

I M A G I N G

O F

P N E U M O N I A S

(a)

Fig. 4.17 This is an anterior–posterior chest radiograph of a 12-year-old child showing ill-defined areas of consolidation. Her sputum was positive for Mycobacterium tuberculosis.

•

centrilobular nodules (2–4 mm in diameter), branching centrilobular opacities with tree-inbud configuration (Figure 4.20), acinar shadows (4–10 mm) and bronchopneumonia are seen [90,96,97]. Haematogenous spread results in diffuse miliary nodules (3 mm diameter) (Figure 4.21).

Post-primary TB • Reactivation of previously dormant primary

•

•

infection accounts for 90% of cases, while exogenous re-infection is rare [95]. Risk factors for re-infection include immuno-suppression, malnutrition, old age and debilitation. Typical radiographic appearances are focal illdefined consolidation with adjacent satellite nodules, cavitation in single or multiple sites, upper lobe predominance (apical and posterior segments) and absence of lymphadenopathy [90,94] (Figures 4.22 and 4.23). Endobronchial spread is the commonest method of spread to other regions of the lungs [90,95]. Typical HRCT features include centrilobular

(b)

Fig. 4.18 (a) Posterior–anterior and (b) lateral chest radiographs of a 27-year-old female with primary pulmonary TB showing cavitating focal consolidation in the superior segment of the left lower lobe, and more unusually another focal area of consolidation in the anterior segment of the right upper lobe.

nodules with tree-in-bud appearance (Figure 4.23b), corresponding to caseous necrosis within and around the terminal and respiratory bronchioles at pathology [90]. Other HRCT features

M Y C O B A C T E R I U M

45

T U B E R C U L O S I S

(b) (a)

Fig. 4.19 A 24-year-old female with endobronchial spread in primary pulmonary TB. (a) Posterior–anterior chest radiograph showing an area of consolidation in the right lower zone, with (b) multiple small ill-defined nodules (2–4 mm in diameter) seen predominantly in the right lung.

• •

•

include bronchial wall thickening, lobular consolidation and bronchiectasis (Figure 4.23) [98]. Miliary TB is denoted on HRCT as tiny (4 mm) well-defined nodules with random distribution throughout the lungs (Figure 4.24). Tuberculomas are common. These are solitary nodules, 1–4 cm in diameter with smooth margins, although spiculated and lobulated margins can be found. Satellite lesions are commonly associated, and calcification occurs with time. Prolonged infection results in severe upper lobe fibrosis with bronchiectasis and tracheomegaly in both primary and post-primary TB.

Non-tuberculosis mycobacteria • Commonly caused by Mycobacterium aviumintracellulare and M. kansasii [94,99].

• M. xenopi, M. fortuitum and M. chelonae are Fig. 4.20 HRCT scan showing centrilobular nodules with tree-in-bud (arrows) configuration in the superior segment of the right lower lobe in a patient with endobronchial TB.

uncommon pathogens for pulmonary infections although they tend to cause a spectrum of pulmonary, cervical lymph node, cutaneous and soft tissue infections.

46

I M A G I N G

O F

P N E U M O N I A S

(a)

(a)

(b)

Fig. 4.22 A 65-year-old male with post-primary TB in the upper lobes. (a) HRCT shows multiple cavitating lesions (arrows), and smaller nodules (arrow heads) in both lung apices. Reticulation in the right lung suggests mild fibrosis. (b) Caudad to (a), areas of bronchopneumonia (asterisk) are noted together with bronchial wall thickening and dilatation (arrow heads).

• Pre-existing lung disease is often present, particu•

•

• (b)

Fig. 4.21 (a) Chest radiograph of a 45-year-old female with miliary TB, shown more clearly in (b) which is a magnified view of the right lower lobe.

larly in elderly white Caucasian males but not in women [99,100]. There is a considerable overlap in radiographic features between non-tuberculosis mycobacteria (NTMB) and Mycobacterium tuberculosus infections. Characteristic radiographic appearances in males include bronchiectasis, cavitation (80–95%), pleural thickening (37–56%) and upper lobe fibrosis [94,99]. Endobronchial spread is found in 40–70% of cases [99] with typical features of branching centrilobular nodules on HRCT [100]. Characteristic radiographic appearances in women are bronchiectasis and centrilobular nodules (Figure 4.25) without upper lobe predilection, although the bronchiectasis is ‘typically’ found in the right middle lobe and lingula [94,99]. Pleural effusions and mediastinal lymphadenopathy are rare.

M Y C O B A C T E R I U M

47

T U B E R C U L O S I S

(a)

Fig. 4.25 HRCT scan showing centrilobular nodules with cylindrical bronchiectasis (arrows) in NTMB. Mycobacterium avium-intracellulare was cultured from sputum.

• Primary TB lesions can range from ill-defined

(b)

Fig. 4.23 HRCT scans of a 46-year-old man with reactivation of TB. (a) Cavitating area of consolidation is noted in the posterior segment of the right upper lobe. (b) Centrilobular and tree-in-bud opacities (arrows) are noted in the anterior segment of the right upper lobe.

•

•

• • Fig. 4.24 HRCT scan showing tiny miliary nodules in miliary TB.

airspace opacities to bronchopneumonia and lobar consolidation. Necrotic lymph nodes and cavitation are common features. Post-primary TB is characterised by focal ill-defined consolidation with adjacent satellite nodules, cavitation in single or multiple sites, upper lobe predominance (apical and posterior segments) and absence of lymphadenopathy. In both primary and post-primary TB, endobronchial spread and military TB can occur. Endobronchial spread is characterised by the presence of ill-defined centrilobular nodules (2–4 mm in diameter), branching centrilobular opacities with tree-in-bud configuration (Figure 4.20), acinar shadows (4–10 mm) and bronchopneumonia on HRCT. NTMB infection is commonly caused by mycobacterium avium-intracellulare and M. kansasii. There is a considerable overlap in radiographic features between NTMB and Mycobacterium tuberculosus infections.

48

I M A G I N G

O F

P N E U M O N I A S

Fungal pulmonary infections • Fungi can be categorised into two groups: primary

• • •

•

•

•

pathogens that infect healthy individuals and opportunistic pathogens that affect immuno-compromised individuals or pre-existing lung conditions. Histoplasma capsulatum, Coccidioides immitis and Blastomyces dermatitidis represent the first group of pathogens and are usually saprophytes. Aspergillus and Candida species form the opportunistic group. In acute histoplasmosis (Histoplasma capsulatum) chest radiographs can be normal although ill-defined consolidations with hilar lymphadenopathy are common. Miliary nodules (3–4 mm) resembling miliary TB are seen in the disseminated histoplasmosis [101]. In chronic histoplasmosis, the radiographic appearances are indistinguishable from post-primary TB. In coccidioidomycosis (Coccidioides immitis), the chest radiograph may be normal, or show segmental consolidations with lower lobe predominance, which may resolve and reappear elsewhere in the lung. Prolonged infection is characterised by peripheral lung nodules (5 mm to 5 cm) in the upper and middle lung zones, with cavitation in 10–15% of cases [102]. Chronic infection results in cavities and fibrosis. Aspergillosis is usually caused by Aspergillus fumigatus and A. niger. Aspergillosis can manifest in three ways: (a) as a saprophytic fungal ball in immuno-competent host with pre-existing lung conditions such as asthma, COPD, bronchiectasis or cavitatory lung disease without invasion of host tissue; (b) as allergic aspergillosis and (c) as invasive aspergillosis. In allergic aspergillosis, mucoid impaction of ectactic bronchi are typically seen with finger-in-glove or cluster of grapes appearance while invasive aspergillosis can manifest as bronchopneumonia (Figure 4.26), an angio-invasive process (Figure 4.27), acute tracheobronchitis or chronic necrotising infection. Angio-invasive aspergillosis is the most common, characterised by focal cavitating opacities (Figure 4.27), and on CT or HRCT a halo sign may be found [103]. Cryptococcus (Cryptococcus neoformans), infects both immuno-competent and immunocompromised individuals with AIDS or lymphoma [104]. The most common radiographic manifestation is single or multiple nodules (5 mm

(a)

(b)

Fig. 4.26 A 42-year-old man with acute myeloid leukaemia with Aspergillus fumigatus pneumonia. (a) Posterior–anterior chest radiograph showing illdefined opacities in both lungs. Note Hickman line in situ. (b) HRCT scan confirms the presence of illdefined areas of bronchopneumonia mainly in the subpleural regions of the lungs.

to 4 cm) usually peripheral in site (Figure 4.28), followed by ill-defined consolidation. Cavitation, enlarged mediastinal and hilar lymph nodes, and disseminated disease including multiple or miliary nodules are found in immuno-compromised hosts [105,106].

• Histoplasma capsulatum, Coccidioides immitis and Blastomyces dermatitidis are fungi that infect immuno-competent individuals as saprophytes.

U T I L I T Y

O F

49

C T

(a)

Fig. 4.27 Posterior–anterior chest radiograph in a 60-year-old man with acute myeloid leukaemia and invasive aspergillosis. Cavitation is noted within the area of consolidation in the right lower lobe.

• Aspergillus and Candida species are • •

opportunistic fungi that infect immunocompromised hosts. Both acute and chronic histoplasmosis infection can mimic mycobacterial infections. Aspergillosis can manifests in three clinical syndromes: as saprophytic fungal ball, allergic aspergillosis and invasive aspergillosis.

(b)

Fig. 4.28 HRCT scans in a 47-year-old female with cryptococcus infection. (a) There is a large subpleural area of consolidation in the superior segment of the left lower lobe with surrounding ground-glass opacity and satellite nodules. (b) Smaller nodules associated with ground-glass opacities are also found in both the upper lobes and in the left lower lobe.

Utility of CT CT or HRCT evaluation of pneumonias serves as a useful adjunct to chest radiographs in selected circumstances when the chest radiograph is nonrevealing or non-diagnostic [107,108] and to evaluate complications of pneumonia such as abscess formation (Figure 4.29), empyema, parapneumonic effusions and bronchopleural fistulas (Figure 4.30) [108–113]. CT and HRCT provide greater anatomic detail and superior morphologic evaluation of opacities including their pattern and extent. Groundglass opacities, consolidation, centrilobular nodules, peribronchial distribution, air-bronchograms, septal thickening and bronchial wall thickening are all better appreciated on CT and HRCT than chest radiographs [64–68,75,89,95–98,114]. Identification of

Fig. 4.29 Contrast-enhanced CT scan showing consolidation in the anterior segment of the right middle lobe with central area of cavitation and airfluid level. Right pleural effusion is also present.

50

I M A G I N G

(a)

(b)

Fig. 4.30 Bronchopleural fistula demonstrated on (a) axial CT section and (b) two-dimensional reformation showing a large air-filled pleural cavity in direct communication with the apico-posterior segmental bronchus of the left upper lobe (arrows). The patient had history of TB infection that had destroyed part of the left upper lobe.

pleural, pericardial or lung complications in pneumonias directs appropriate and early management of these complications [108–111,113]. Several studies have also reported the utility of HRCT in the early detection of pneumonia in immuno-compromised patients with neutropaenic fever and normal chest radiograph [114,115].

• The role of CT or HRCT in the imaging of pneumonias is to (a) further characterise lung lesions or detect pneumonia when the chest radiograph is non-revealing or nondiagnostic and (b) evaluate complications of pneumonia such as abscess formation, empyema, parapneumonic effusions and bronchopleural fistulas.

References 1. Niederman MS, Bass Jr JB, Campbell GD et al. American Thoracic Society: Guidelines for the initial management of adults with community-acquired pneumonia: diagnosis, assessment of severity, and initial antimicrobial therapy. Am Rev Respir Dis 1993; 148: 1418–1426.

O F

P N E U M O N I A S

2. Mandell LA. Community acquired pneumonia – aetiology, epidemiology and treatment. Chest 1995; 108: 35–42. 3. Osmer JC and Cole BK. The stethoscope and roentgenogram in acute pneumonia. South Med J 1966; 59: 75–77. 4. Heitzman ER, Markarian B, Berger I and Dailey E. The secondary pulmonary lobule: a practical concept for interpretation of chest radiographs. I. Roentgen anatomy of the normal secondary pulmonary lobule. Radiology 1969; 93: 507–512. 5. Heitzman ER, Markarian B, Berger I and Dailey E. The secondary pulmonary lobule: a practical concept for interpretation of chest radiographs. II. Application of the anatomic concept to an understanding of roentgen pattern in disease states. Radiology 1969; 93: 513–519. 6. Heitzman ER. The radiological diagnosis of pneumonia in the adult: a commentary. Semin Roentgenol 1989; 24: 212–217. 7. Goodman LR, Goren RA and Teprick SK. The radiographic evaluation of pulmonary infections. Med Clin N Am 1980; 64: 553–574. 8. Zornosa L, Goldman M, Wallace S, Valdivieso M and Bodey GP. Radiologic features of Gram-negative pneumonias in the neutropenic patient. Am J Roentgenol 1976; 127: 989–996. 9. Fraser DW, Tsai TR, Orenstein W et al. Legionnaires’ disease: description of an epidemic of pneumonia. New Engl J Med 1977; 297: 1189–1197. 10. Sarosi GA. Atypical pneumonia. Postgrad Med 1999; 105: 131–138. 11. Genereux GP and Stilwell GA. The acute bacterial pneumonias. Semin Roentgenol 1980; 15: 9–16. 12. Fraser RG and Wortzman G. Acute pneumococcal lobar pneumonia: the significance of nonsegmental distribution. J Can Assoc Radiol 1959; 10: 37–42. 13. Barnes DJ, Naraqi S and Igo JD. The diagnostic and prognostic significance of bulging fissures in acute lobar pneumonia. Aust NZ J Med 1988; 18: 130–136. 14. George WL and Finegold SM. Bacterial infections of the lung. Chest 1982; 81: 502–507. 15. Smith CB and Overall JC. Clinical and epidemiological clues to the diagnosis of respiratory infections. Radiol Clin N Am 1973; 11: 261–278. 16. MacFarlane J. An overview of community-acquired pneumonia with lessons learned from the British Thoracic Society Study. Semin Respir Infect 1994; 9: 153–165. 17. Mufson MA. Pneumococcal infections. J Am Med Assoc 1981; 246: 1942–1948. 18. Leeper Jr KV. Severe community acquired pneumonia. Semin Respir Infect 1996; 11: 96–108. 19. Levy M, Dromer F, Brion N, Leturdu F and Carbon C. Community-acquired pneumonia: importance of initial non-invasive bacteriologic and radiographic investigations. Chest 1988; 93: 43–48. 20. Moine P, Vercken JB, Chevret S, Chastang C and Gajdos P. Severe community-acquired pneumonia: etiology, epidemiology, and prognostic factors. Chest 1994; 105: 1487–1495. 21. Kantor HG. The many radiologic faces of pneumococcal pneumonia. Am J Roentgenol 1981; 137: 1213–1220. 22. Ort S, Ryan JL, Barden G and D’Esopo N. Pneumococcal pneumonia in hospitalized patients: clinical and radiological presentations. J Am Med Assoc 1983; 249: 214–218. 23. MacFarlane JT, Miller AC, Roderick Smith WH, Morris AH and Rose DH. Comparative radiographic features of community acquired Legionnaires’ disease, pneumococcal pneumonia, mycoplasma pneumonia, and psittacosis. Thorax 1984; 39: 28–33. 24. Felson B, Rosenberg LS and Hamburger Jr M. Roentgen findings in acute Friedlander’s pneumonia. Radiology 1949; 53: 559–563. 25. Pierce AK and Stanford JP. Aerobic Gram-negative bacillary pneumonias. Am Rev Respir Dis 1974; 110: 647–658.

R E F E R E N C E S

26. Moon WK, Im JG, Yeon KM and Han MC. Complications of Klebsiella pneumonia: CT evaluation. J Comput Assist Tomogr 1995; 19: 176–181. 27. Nguyen MLT and Yu VL. Legionella infection. Clin Chest Med 1991; 12: 257–268. 28. Band JD, La Venture M, Davis JP et al. Epidemic Legionnaires’ disease: airborne transmission down a chimney. J Am Med Assoc 1981; 245: 2404–2407. 29. Stout J, Yu VL, Vickers RM and Shonnard J. Portable water supply as the hospital reservoir for Pittsburgh pneumonia agent. Lancet 1982; 1: 471–472. 30. Davis GS, Winn Jr WC and Beaty HN. Legionnaires’ disease: infections caused by Legionella pneumophilia and Legionellalike organisms. Clin Chest Med 1981; 2: 145–166. 31. Kirby BD, Snyder KM, Meyer RD and Finegold SM. Legionnaires’ disease: report of sixty-five nosocomially acquired cases and review of the literature. Medicine 1980; 59: 188–205. 32. Prodinger WM, Bonatti H, Allerberger F et al. Legionella pneumonia in transplant recipients: a cluster of cases of eight years duration. J Hosp Infect 1994; 26: 191–202. 33. Neil AM, Martin IR, Weir R et al. Community-acquired pneumonia: aetiology and usefulness of severity criteria on admission. Thorax 1996; 51: 1010–1016. 34. Roig J, Domingo C and Morera J. Legionnaires’ disease. Chest 1994; 105: 1817–1825. 35. Lieberman D, Porath A, Schlaeffer F, Lieberman D and Boldur I. Legionella species community-acquired pneumonia – a review of 56 hospitalized adult patients. Chest 1996; 109: 1243–1249. 36. Fairbank JT, Patel MM and Dietrich PA. Legionnaries’ disease. J Thorac Imag 1991; 6: 6–13. 37. Tan MJ, Tan JS, Hamor RH, File Jr TM and Breiman RF. The radiologic manifestations of Legionnaires’ disease. Chest 2000; 117: 398–403. 38. Dietrich PA, Johnson RD, Fairbank JT and Walke JS. The chest radiograph in Legionnaires’ disease. Radiology 1978; 127: 577–582. 39. Kroboth FJ, Yu VL, Reddy SC and Yu AC. Clinicoradiographic correlation with the extent of Legionnaires’ disease. Am J Roentgenol 1983; 141: 263–268. 40. Meenhorst PL and Mulder JD. The chest X-ray in Legionalla pneumonia (Legionnaires’ disease). Eur J Radiol 1983; 3: 180–186. 41. Mirich D, Gray R and Hyland R. Legionella lung cavitation. J Can Assoc Radiol 1990; 41: 100–102. 42. Itoh H, Tokunaga S, Asamoto H et al. Radiologic–pathologic correlation of small lung nodules with special reference to peribronchiolar nodules. Am J Roentgenol 1978; 130: 223–231. 43. Muller NL and Miller RR. State-of-the-art: diseases of the bronchioles: CT and histopathologic findings. Radiology 1993; 186: 653–658. 44. Landay MJ, Christensen EE, Bynum LJ and Goodman C. Anaerobic pleural and pulmonary infections. Am J Roentgenol 1980; 134: 233–240. 45. Ruiz M, Weig S, Marcos MA et al. Etiology of community acquired pneumonia: impact of age, comorbidity, and severity. Am J Respir Crit Care Med 1999; 160: 397–405. 46. al-Ujayli B, Nafziger DA and Saravolatz L. Pneumonia due to Staphylococcus aureus infection. Clin Chest Med 1995; 16: 111–120. 47. George DL, Falks PS, Wunderink RG et al. Epidemiology of community-acquired pneumonia. Am J Respir Crit Care Med 1999; 160: 397–401. 48. MacFarlane J and Rose D. Radiographic features of staphylococcal pneumonia in adults and children. Thorax 1996; 51: 539–540. 49. Dunn M, Wunderink RG. Ventilator-associated pneumonia caused by Pseudomonas infection. Clin Chest Med 1995; 16: 95–109.

51

50. Winer-Muram HT, Jennings SG, Wunderink RG, Jones CB and Leeper K. Ventilator associated Pseudomonas aeruginosa pneumonia: radiographic findings. Radiology 1995; 195: 247–252. 51. Iannini PB, Claffey T and Quintiliani R. Bacteremic Pseudomonas pneumonia. J Am Med Assoc 1974; 230: 558–561. 52. Boschini A, Smacchia C, DiFine M et al. Community acquired pneumonia in a cohort of former injection drug users with and without human immunodeficiency virus infection: incidence, etiologies and clinical aspects. Clin Infect Dis 1996; 23: 107–113. 53. Falco V, Fernandez de Savilla T, Alegre J et al. Bacterial pneumonia in HIV-infected patients: prospective study of 68 episodes. Eur Respir J 1994; 7: 235–239. 54. Trollfors B, Claesson B, Lagergard T and Sandberg T. Incidence, predisposing factors and manifestations of invasive Haemophilus influenzae infections in adults. Eur J Clin Microbiol 1984; 3: 180–184. 55. Pearlberg J, Hagger AM, Saravolatz L, Beute GH and Popovich J. Hemophilus influenzae pneumonia in the adult: radiographic appearance with clinical correlation. Radiology 1984; 151: 23–26. 56. Wallace Jr RJ, Musher DM and Martin RR. Haemophilus influenzae in adults. Am J Med 1978; 64: 87–93. 57. Crossley KB and Thurn JR. Nursing home-acquired pneumonia. Semin Respir Infect 1989; 4: 64–72. 58. Carson MJ, Chadwick DL, Brubaker CA et al. Thirteen boys with progressive septic granulomatosis. Pediatrics 1965; 35: 405–408. 59. Feldman PS, Cohan MA and Hierholzer Jr WJ. Fatal Hong Kong influenza: a clinical microbiological and pathological analysis of none cases. Yale J Biol Med 1972; 45: 49–63. 60. Yeldandi AV and Colby TV. Pathologic features of lung biopsy specimens form influenza pneumonia cases. Hum Pathol 1994; 25: 47–53. 61. Sullivan CJ and Jordan MC. Diagnosis of viral pneumonia. Semin Respir Infect 1988; 3: 148–161. 62. Han BK, Son JA, Hoon HK and Lee SI. Epidemic adenoviral lower respiratory tract infection in pediatric patients: radiographic and clinical characteristics. Am J Roentgenol 1998; 170: 1077–1080. 63. Osborne D. Radiologic appearance of viral disease of the lower respiratory tract in infants and children. Am J Roentgenol 1978; 130: 29–33. 64. Kim JS, Ryu CW, Lee SI, Sung DW and Park CK. High resolution CT findings of varicella-zoster pneumonia. Am J Roentgenol 1999; 172: 113–116. 65. Kang EY, Patz EF and Muller NL. Cytomegalovirus pneumonia in transplant patients: CT findings. J Comput Assist Tomogr 1996: 20: 295–299. 66. Moon JH, Kim EA, Lee KS, Kim TS, Jung KJ and Song JH. Cytomegalovirus pneumonia high-resolution CT findings in ten non-AIDS immunocompromised patients. Kor J Radiol 2000; 1: 73–78. 67. Tanaka N, Matsumoto T, Kuramitsu T et al. High resolution CT findings in community-acquired pneumonia. J Comput Assist Tomogr 1996; 20: 600–608. 68. McGuinness G, Scholas JV, Garay SM, Leitman BS, McCauley DI and Naidich DP. Cytomegalovirus pneumonitis: spectrum of parenchymal CT findings with pathologic correlation in 21 AIDS patients. Radiology 1994; 192: 451–459. 69. Kim EA, Lee KS, Primack SL et al. Viral pneumonias in adults: radiologic and pathologic findings. Radiographics 2002; 22: S137–S149. 70. Ortqvist A, Hedlund A, Grillner L et al. Aetiology, outcome and prognostic factors in community-acquired pneumonia requiring hospitalization. Eur Respir J 1990; 3: 1105–1113. 71. Almirall J, Morato I, Riera F et al. Incidence of communityacquired pneumonia and Chlymadia pneumoniae infection: a prospective multicentre study. Eur Respir J 1993; 6: 14–18.

52

72. Rosmus HH, Pare JA, Masson AM and Fraser RG. Roentgenographic patterns of acute mycoplasma and viral pneumonitis. J Can Assoc Radiol 1968; 19: 74–77. 73. Putman CE, Curtis Am, Simeon JF and Jensen P. Mycoplasma pneumonia: clinical and roentgenographic patterns. Am J Roentgenol 1975; 124: 417–422. 74. Cameron DC, Borthwick RN and Philp T. The radiographic patterns of acute mycoplasma pneumonia. Clin Radiol 1977; 28: 173–180. 75. Reittner P, Muller NL, Heyneman L et al. Mycoplasma pneumoniae pneumonia: radiographic and high-resolution CT features in 28 patients. Am J Roentgenol 2000; 174: 37–41. 76. Smulian AG and Walzer PD. The biology of Pneumocystis carinii. Crit Rev Microbiol 1992; 18: 191–216. 77. Naidich DP, Garay SM, Leitman BS and McCauley DI. Radiographic manifestations of pulmonary disease in the acquired immunodeficiency syndrome (AIDS). Semin Roentgenol 1987; 22: 14–30. 78. Elvin K, Lidman C, Tynell E, Linder E and Bjorkman A. Natural history of asymptomatic and symptomatic Pneumocystis carinii infection in HIV infected patients. Scan J Infect Dis 1994; 26: 643–651. 79. Ballardie FW, Winearls CG, Cohen J, Carr DH, Rees AJ and Williams G. Pneumocystis carinii pneumonia in renal transplant recipients: clinical and radiographic features, diagnosis and complications of treatment. Quart J Med 1985; 57: 729–747. 80. Varthalitis I, Aoun M, Daneau D and Meunier F. Pneumocystis carinii pneumonia in patients with cancer: an increasing incidence. Cancer 1993; 71: 481–485. 81. Godeau B, Coutant-Perronne V, Le Thi Huong D et al. Pneumocystis carinii pneumonia in the course of connective tissue disease: report of 34 cases. J Rheumatol 1994; 21: 246–251. 82. Barrio JL, Suarez M, Rodriguez JL, Saldana MJ and Pitchenik AE. Pneumocystis carinii pneumonia presenting as cavitating and noncavitating solitary pulmonary nodules in patients with the acquired immunodeficiency syndrome. Am Rev Respir Dis 1986; 134: 1094–1096. 83. Forrest JV. Radiographic findings in Pneumocystis carinii pneumonia. Radiology 1972; 103: 539–544. 84. Lubat E, Megibow AJ, Balthazar EJ, Goldenberg AS, Birnbaum BA and Bosniak MA. Extrapulmonary Pneumocystis carinii infection in AIDS: CT findings. Radiology 1990; 174: 157–160. 85. Mayor B, Schnyder P, Giron J, Landry M, Duvoisin B and Fournier D. Mediastinal and hilar lymphadenopathy due to Pneumocystis carinii infection in AIDS patients: CT features. J Comput Assist Tomogr 1994; 18: 408–411. 86. Sandhu JS and Goodman PC. Pulmonary cysts associated with Pneumocystis carinii pneumonia in patients with AIDS. Radiology 1989; 173: 33–35. 87. Hartman TE, Primack SL, Muller NL and Staples CA. Diagnosis of thoracic complications in AIDS: accuracy of CT. Am J Roentgenol 1994; 162: 547–553. 88. Kuhlman JE, Kavuru M, Fishman EK and Siegelman SS. Pneumocystis carinii pneumonia: spectrum of parenchymal CT findings. Radiology 1990; 175: 711–714. 89. Naidich DP and McGuinness G. Pulmonary manifestations of AIDS: CT and radiographic correlations. Radiol Clin N Am 1991; 29: 999–1017. 90. Lee KS and Im JG. CT in adults with tuberculosis of the chest. Characteristic findings and role in management. Am J Roentgenol 1995; 164: 1361–1367. 91. Choyke PL, Sostman HD, Curtis AM et al. Adult onset tuberculosis. Radiology 1983; 148: 357–362. 92. Leung AN, Muller NL, Pineda PR and FritzGerald JM. Primary tuberculosis in childhood: radiographic manifestations. Radiology 1992; 182: 87–91.

I M A G I N G

O F

P N E U M O N I A S

93. Weber AL, Bird KT and Janower ML. Primary tuberculosis in childhood with particular emphasis on changes affecting the tracheobronchial tree. Am J Roentgenol 1968; 103: 123–132. 94. Woodring JH, Vandiviere HM, Fried AM, Dillon ML, Williams TD and Melvin IG. Update: the radiographic features of pulmonary tuberculosis. Am J Roentgenol 1986; 146: 497–506. 95. Im JG, Itoh H and Han MC. CT of pulmonary tuberculosis. Semin Ultrasound CT MR 1995; 26: 420–434. 96. Lee KS, Song KS, Lim TH, Kim PN, Kim IY and Lee BH. Adult onset pulmonary tuberculosis: findings on chest radiographs and CT scans. Am J Roentgenol 1993; 160: 753–758. 97. Im JG, Itoh H, Shim YS et al. Pulmonary tuberculosis: CT findings – early active disease and sequential change with antituberculosis therapy. Radiology 1993; 186: 653–660. 98. Lee KS, Hwang JW, Chung MP, Kim H and Kwon OJ. Utility of CT in the evaluation of pulmonary tuberculosis in patients without AIDS. Chest 1996: 110: 977–984. 99. Miller Jr WT. Spectrum of pulmonary nontuberculous mycobacterial infection. Radiology 1994; 191: 343–350. 100. Moore EH. Atypical mycobacterial infection in the lung: pulmonary CT appearance. Radiology 1993; 187: 777–782. 101. Kauffman CA, Israel KS, Smith JW, White AC, Schwarz J and Brooks GF. Histoplasmosis in immunosuppressed patients. Am J Med 1978; 64: 923–932. 102. Greendyke WH, Resnick DL and Harvey WC. The varied Roentgen manifestations of primary coccidioidomycosis. Am J Roentgenol Radium Ther Nucl Med 1970; 109: 491–499. 103. Hruban RH, Meziane MA, Zerhouni EA, Wheeler PS, Dumler JS and Hutchins GM. Radiologic–pathologic correlation of the CT halo sign in invasive pulmonary aspergillosis. J Comput Assist Tomogr 1987; 11: 534–536. 104. Kerkering TM, Duma RJ and Shadomy S. The evolution of pulmonary cryptococcosis: clinical implications from a study of 41 patients with and without compromising host factors. Ann Intern Med 1981; 94: 611–616. 105. Patz Jr EF and Goodman PC. Pulmonary cryptococcosis. J Thorac Imag 1992; 7: 51–55. 106. Woodring JH, Ciporkin G, Lee C, Worm B and Woolley S. Pulmonary cryptococcosis. Semin Roentgenol 1996; 31: 67–75. 107. Janzen DL, Padley SPG, Adler BD and Muller NL. Acute pulmonary complications in immunocompromised non-AIDS patients: comparison of diagnostic accuracy of CT and chest radiography. Clin Radiol 1993; 47: 159–165. 108. Donnelly LF and Klosterman LA. The yield of CT of children who have complicated pneumonia and noncontributory chest radiography. Am J Roentgenol 1998; 170: 1627–1631. 109. Stark DD, Federle MP, Goodman PC, Podrasky AE and Webb WR. Differentiating lung abscesses and empyema: radiography and computed tomography. Am J Roentgenol 1983; 141: 163–167. 110. Baber CD, Hedlund LW, Oddson TA and Putman CE. Differentiating empyemas and peripheral pulmonary abscess: the value of computed tomography. Radiology 1980; 135: 755–758. 111. Klein JS, Schultz S and Heffner JE. Interventional radiology of the chest: image-guided percutaneous drainage of pleural effusions, lung abscess, and pneumothorax. Am J Roentgenol 1995; 164: 581–588. 112. Stern EJ, Sun H and Haramati LB. Peripheral bronchopleural fistulas: CT imaging features. Am J Roentgenol 1996; 167: 117–120. 113. Ricci ZJ, Haramati LB, Rosenbaum AT and Liebling MS. Role of computed tomography in guiding the management of peripheral bronchopleural fistula. J Thorac Imag 2002; 17: 214–218.

The Role of Chest Radiographs in the Diagnosis of SARS KT Wong, GE Antonio, EHY Yuen and AT Ahuja

Introduction Pathological considerations Role of the CXR in the diagnosis of SARS Protocol for diagnostic imaging Digital radiography and picture archiving and communication system

53 53 54 54

CHAPTER

5

CXR appearances Differential diagnosis Blind spots for CXRs Conclusion

56 58 59 59

55

Introduction At the onset of the severe acute respiratory syndrome (SARS) crisis, the majority of patients were presented with respiratory symptoms. As the epidemic progressed, either due to a different mode of transmission or a mutation of the virus, some SARS patients presented with minor or no respiratory symptoms but diarrhoea [1]. Understandably, this created a problem with case definition and diagnosis, and the lack of a reliable and rapid biochemical test for SARS placed more emphasis on chest imaging findings for diagnosis of the disease. The wide availability, speed and inexpensive nature of the chest radiograph (CXR) has made it the firstline imaging investigation when faced with a respiratory complaint. It is only fitting that the initial imaging investigation of SARS also starts here. This chapter presents the radiographic features of SARS and the differential diagnosis.

Key point

• SARS patient may not present with respiratory symptoms.

Pathological considerations Viral infection of the respiratory tract may involve the upper system [2], from the common cold (rhinoviruses and coronaviruses), larynx (respiratory syncitial virus), trachea and bronchi (herpes simplex type 1) to the lung parenchyma (influenza). The initial phase in viral lung parenchymal involvement is called a pneumonitis. A local inflammatory response is directed towards the offending virus, an inflammatory cocktail of cells and fluid accumulate in the alveolar interstitium of the lung parenchyma. In bacterial infections this exudate spills over into the airspace and results in the classic consolidation.

54

R O L E

This flooding of the alveolar space occasionally occurs in viral pulmonary infections. Coronavirus infection of the respiratory tract is not new. It is previously well known for causing upper respiratory tract infections, such as coryza and pharyngitis. There has been scanty documentation of lung involvement [3] and none on the imaging appearance of it. This has changed with SARS, and the imaging appearances (on CXR and high-resolution computed tomography (HRCT)) of SARS-pneumonia are well documented. Similarities between other viral infection and SARS include the following: 1. Mild or patchy consolidation may occur in both instances, the consolidation may initially be focal and may progress to a multi-focal involvement in the later stages. 2. In severe, later stages of a viral infection of the lung, changes of adult respiratory distress syndrome (ARDS) may occur in the lung [4]. This is also seen in patients with SARS and has been confirmed by post-mortem in patients who have succumbed to the disease [5]. Thus two broad categories of lung pathology may occur in SARS and be seen on presentation, viz. an earlier focal or multi-focal pneumonitis and a later extensive consolidation resembling ARDS. Many variations will exist between these two extremes, at presentation and during the course of the patient’s illness. The imaging of SARS reflects this spectrum of pathology.

O F

C X R

I N

D I A G N O S I S

O F

S A R S

with respiratory symptoms, the immediate role of the CXR is to establish the presence of intra-thoracic disease and to exclude a surgically correctable cause such as a pneumothorax, large pleural effusion/empyema. When a patient presents with less severe symptoms but has respiratory problems and is febrile, a CXR is performed as a routine workup in most centres. Differentiation of SARS from other causes of respiratory distress requires knowledge of the radiographic appearance of SARS and the radiographic appearances of other pulmonary diseases, particularly other pneumonias, which often overlap among themselves (see Chapter 3). The lack of a reliable, rapid and widely available test, at the time of writing this chapter, has further compounded the issue. Several investigators [6–8] have reported that the CXR is abnormal in approximately 80% of the initial radiograph of SARS patients. However, one must view this figure in the correct perspective. These CXRs were performed during the course of an epidemic, in patients with significant clinical signs and symptoms, and a history of contact. This is probably the reason for the high sensitivity of CXRs in identifying SARS. At the end of the epidemic, when the number of patients was few, signs and symptoms variable and the history of contact difficult to verify, the accuracy of the CXR is likely to be lower. In the authors’ experience, under these circumstances, the initial CXR is often normal even in later confirmed cases of SARS. The diagnosis is made a couple of days later when laboratory results are available. It is during this time when the risk of transmission of infection in the medical staff and community is the highest, particularly if the patients are housed in the general wards and not in separate isolation rooms.

Key points Two types of pathology 1. Focal or multi-focal pneumonitis (early stage disease) 2. Confluent consolidation sometimes resembling ARDS

Role of the CXR in the diagnosis of SARS This is the first-line imaging tool for all respiratory complaints and remains so for SARS. In routine clinical practice, when confronted with a very sick patient

Protocol for diagnostic imaging It is therefore imperative that the protocol for diagnostic imaging be reviewed in its proper perspective: 1. during the initial outbreak; 2. as the situation stands now at the end of the epidemic.

During the initial outbreak During the initial outbreak/epidemic it was essential to make a prompt diagnosis as: (a) it allowed for

D I G I T A L

R A D I O G R A P H Y

A N D

more timely medical treatment which would affect prognosis and (b) it allowed prompt isolation of infected patients to protect other in-patients, healthcare workers and the community. The protocol used at our institution at the time was:

• Patient suspicious of SARS satisfying World

• •

55

P A C S

Health Organization (WHO) criteria, underwent a frontal CXR. If this showed pneumonic changes, the patient was treated as a SARS patient, i.e. admitted to SARS isolation ward and treated. Patient suspicious of SARS satisfying WHO criteria, had a frontal CXR. If this was normal, an HRCT scan was performed. Patients with respiratory symptoms, low suspicion for SARS and not fulfilling WHO criteria had a frontal CXR. If this was negative, the patient is considered as not having SARS and treated in a non-SARS ward.

Thus, an HRCT was performed only in patients with a strong clinical suspicion and a negative CXR.

End of the epidemic By the end of the epidemic, the situation was remarkably different. The patients who were ultimately diagnosed to have SARS were often the frail elderly from old age/convalescent homes. They presented with variable symptoms, often, non-respiratory in nature. The CXR in these patients was often normal at presentation and became positive a few days later (around the same time as the laboratory results return). Therefore, the early use of a CXR was probably not beneficial and may provide a false sense of security. The authors believe that in this category of patients the threshold for doing an HRCT should be low as it may detect lung abnormalities early in the course of the disease. The HRCT is obviously interpreted based on the clinical findings and hopefully as the radiologists and clinicians acquire more experience with SARS, one may be able to make an early diagnosis. This is probably a stopgap measure as eventually accurate laboratory tests will become available and one may not need to resort to imaging for diagnosis. For SARS patients, the initial radiographic appearance will also act as a baseline for the progression of disease. It is useful as a guide for tailoring treatment [5] and may potentially have some value in terms of an outcome predictor.

Key points Roles of CXR in diagnosis of SARS

• Confirm presence of lung disease • Limit differential diagnosis • Act as baseline for disease progression •

monitoring Outcome predictor

Digital radiography and picture archiving and communication system Digital radiography and picture archiving and communication system (PACS) deserve special mention for their contribution in a crisis like SARS. With a wide dynamic range and the power of post-processing, digital radiography minimizes the number of repeat CXRs. Not only does this decrease the radiation dose to the patient (which is substantial due to the long hospital stay and frequent radiographs), it also reduces exposure of radiographers to the infected patients and areas. On the other hand, a PACS system protects the radiologists and clerical staff in that it stops film and envelop transaction between the isolation wards or clinics and the radiology department. PACS also allows instantaneous reporting of the radiograph and remote consultation with clinicians without needing the radiologist to be at the ward or clinic. Reporting images while wearing full-protection clothing is uncomfortable (hot and sweaty), inefficient (thick gloves and muffled voice) and probably less accurate (looking through the visor or face-shield). One can only speculate the psychological effects on the radiologist and his reading accuracy when one has to report films in an ultra-high-risk environment. With either piece of equipment, the window levels could be adjusted so that progress changes can be more easily evaluated.

Key points Digital radiography for SARS

• Minimized repeat radiographs •

– less time with infected patients – less radiation for patients Uniform images for progress review.

56

R O L E

Key points PACS for SARS

• No physical transfer of film or envelops • Almost instantaneous reports • Tele-consultation of images

CXR appearances The presenting CXR of patients suspected of SARS, using WHO criteria was abnormal in 78.3% in a study by Wong et al. [6] and 80% in a study by Tsang et al. [7]. Hon et al. found similarly high sensitivity of 90% in children [8].

O F

C X R

I N

D I A G N O S I S

O F

S A R S

reaction [7]. The opacities have ill-defined margins and no evidence of cavitation or calcification. There is no reticular or nodular pattern and there was no hilar or mediastinal enlargement or pleural effusion [6,7,10]. There is no evidence of significant volume loss or increase. Two radiographic patterns of distribution of the lesions are seen in SARS at presentation: (a) unifocal, peripheral airspace opacification (Figures 5.1 and 5.2) and (b) lobar or extensive involvement, sometimes bilateral and occasionally resembling ARDS [6] (Figure 5.3). The two patterns represent the extremes of a spectrum of appearances and may reflect either the stage of the disease or the extent of host reaction to the infection.

The most representative radiographic abnormality seen in SARS patient is airspace opacification. This opacification may be of ground-glass opacity (Figure 5.1) suggesting an alveolitis or less commonly consolidation with air-bronchograms (Figure 5.2). The two types may represent different stages of the disease or different degrees of host

The majority of the opacities are found in the lower to mid-zones of the lungs. In one study of 108 patients [6], the lower or the mid-zones (64.8% and 52.8%, respectively) were involved (Figures 5.4 and 5.5). Only 16.7% of patients had upper zone involvement. The disease appears to start in one lung, unilateral involvement was seen in 82% of patients. The right lung was slightly more commonly

Fig. 5.1 A 40-year-old female, CXR on day 4 since the onset of fever and respiratory symptoms. There is an area of ground-glass opacification in the right lower zone peripherally. Vascular markings are not obscured. No air-bronchograms present. There is no lymphadenopathy or pleural effusion.

Fig. 5.2 A 26-year-old female, CXR on day 5 since the onset of symptoms. There is an area of consolidation in the anterior segment of the right upper lobe. Vascular markings are obscured and air-bronchograms are present. There is no lymphadenopathy or pleural effusion.

C X R

57

A P P E A R A N C E S

involved than the left (75.9% on the right and 62.0% on the left). A solitary lesion (seen in 54.6% of patients) was slightly more common than multifocal involvement (45.4%) (Figures 5.6 and 5.7).

In the majority of patients, the lesions involved the peripheral or subpleural lung parenchymal (88%) and only a small proportion of lesions (12%) did not involve the peripheral lung (Figure 5.8).

Key points Airspace opacification

• Ground-glass density or consolidation • No volume loss or increase • No cavitation or calcification

Fig. 5.3 A 44-year-old male, CXR on day 5 since the onset of symptoms. There are bilateral widespread areas of mixed ground-glass and consolidative opacification. There is no cardiomegaly, lymphadenopathy or pleural effusion.

Total 12%

Total

8%

17%

39%

32%

53%

48%

38%

65%

76%

Fig. 5.5 A 25-year-old male, CXR on day 7 since the onset of symptoms. There is an area of consolidation in the periphery, between the lower and the mid-zones on the right. There is no lymphadenopathy or pleural effusion.

62%

Fig. 5.4 Location of opacities on presenting radiograph (n
108). Percentage of patients with abnormalities in this zone. The height of each zone is defined as occupying one-third of the cranio-caudal distance of the lung.

Uni-focal (55%)

Fig. 5.6

Multi-focal unilateral (7%)

Number of lesion(s).

Bilateral (38%)

58

R O L E

O F

C X R

I N

D I A G N O S I S

O F

S A R S

Key points Location

• • • •

Unilateral involvement more common Right slightly more than left Lower or mid-zone predominance Peripheral/subpleural involvement in the majority of lesions

Differential diagnosis Although the imaging appearances of various pneumonias have been discussed in a separate chapter, the following paragraphs re-emphasize a few salient features.

Fig. 5.7 A 25-year-old female, CXR on day 5 since the onset of symptoms. There are two areas of opacification, in the right upper zone and the right lower zone. There is no lymphadenopathy or pleural effusion.

Peripheral (75%)

Fig. 5.8

Central (12%)

Peripheral and central (13%)

Location of lesion(s).

Key points Two types of distribution

• Focal (slightly more common) or multifocal

• Extensive lobar, sometimes bilateral resembling ARDS

The differential diagnosis for focal or limited multi-focal peripheral hazy opacification are: other forms of atypical pneumonia, bronchopneumonia, bronchiolitis obliterans organizing pneumonia (BOOP), chronic eosinophilic pneumonia and acute extrinsic allergic alveolitis. When the radiographic features are combined with clinical information, these differential diagnoses (except for other forms of atypical pneumonia) are virtually ruled out. The radiographic features of the more common types of atypical pneumonia (chlamydia, mycoplasma and influenza) share many features with SARS [6]. Influenza pneumonia also demonstrates airspace opacification that involves the lower zones and may be either patchy or homogeneous, unilateral or bilateral. Extensive bilateral involvement resembling pulmonary oedema may also occur. Pleural effusion is rare [4,11]. Mycoplasma pneumonia may present with either airspace or reticular opacification and involves the lower zones predominantly. It may produce a segmental involvement [12,13]. Hilar lymphadenopathy may occur especially in children. Pleural effusions occur in 20% of patients. Chlamydial pneumonia similarly shows either airspace or reticular opacification, may be uni-focal or multi-focal, unilateral or bilateral. Pleural effusions may occur [14]. In summary, the radiographic

59

R E F E R E N C E S

features of SARS overlap with those of the different causes of atypical pneumonia and are thus non-specific. The differential diagnosis for lobar or bilateral confluent consolidation, sometimes resembling ARDS, are other causes of ARDS, which include sepsis, shock, inhalational injury, aspiration, narcotics, pancreatitis, etc. Again, the features seen in SARS are indistinguishable from these other causes radiologically but the clinical history should be helpful in ruling out other causes.

Key points Blind spots

• • • •

Behind breast shadows Retrocardiac Paraspinal Posterior costophrenic angles

Conclusion Key points Differential diagnosis

• • • • • •

Atypical pneumonia Bronchopneumonia BOOP Chronic eosinophilic pneumonia Acute extrinsic allergic alveolitis ARDS Clinical history rules out most of the differential diagnoses.

Key points General features of atypical pneumonia (non-specific), viz. influenza, mycoplasma and chlamydia

• Lower lobe predominance • Airspace opacification • Rarely pleural effusion

Blind spots for CXRs The blind spots for radiographic abnormalities in the detection of SARS are those inherent in using a single frontal CXR. Lesions that were detected by HRCT on SARS patients with normal initial CXRs reveal that the hidden lesions were behind dense breast shadows, retrocardiac, paraspinal regions and posterior costophrenic angles [5,6,9,10]. If HRCT is not a feasible option, a lateral CXR may be helpful.

The radiographic appearance of SARS on presentation are airspace opacification of the lung periphery and the lower zone, and absence of cavitation, hilar lymphadenopathy or pleural effusion. The roles of the CXR in SARS are to rule out a surgical cause in an acutely unwell patient, to confirm the presence of lung disease in suspected cases, to limit the differential diagnosis and to act as a baseline for disease progression monitoring. The limitations of the CXR in diagnosing SARS lies in both the nonspecific appearance of the lesions and the poor ability to detect small lesions in ‘blind spots’. The former can be partly resolved with the help of the clinical history and the latter by the use of HRCT.

References 1. World Health Organization. Update 47 – Studies of SARS Virus Survival, Situation in China. Online document at http://www.who.int/csr/sarsarchive/2003_05_05/en/ (accessed 5 May 2003). 2. Yun BY, Kim MR, Park JY, Choi EH, Lee HJ and Yun CK. Viral etiology and epidemiology of acute lower respiratory tract infections in Korean children. Pediatr Infect Dis J 1995; 14(12): 1054–1059. 3. Vabret A, Mourez T, Gouarin S, Petitjean J and Freymuth F. An outbreak of coronavirus OC43 respiratory infection in Normandy, France. Clin Infect Dis 2003; 36(8): 985–989. 4. Razer RS, Muller NL, Colman N and Pare PD. Viruses, mycoplasmas, chlamydiae, and rickettsiae. In: Fraser and Pare (Eds) Diagnosis of Diseases of the Chest, 4th edition. W. B. Saunders Company, Philadelphia, PA. 1999, 979–1032. 5. Lee N, Hui D, Wu A, Chan P, Cameron P, Joynt GM, Ahuja A, Yung MY, Leung CB, To KF, Lui SF, Szeto CC, Chung S and Sung JJY. A major outbreak of severe acute respiratory syndrome in Hong Kong. New Engl J Med 2003; 20(348): 1986–1994. 6. Wong KT, Antonio GE, Hui DS, et al. Severe acute respiratory syndrome: radiographic appearances and pattern of progression in 138 patients. Radiology 2003; 228: 401–406. 7. Tsang KW, Ho PL, Ooi GC, Yee WK, Wang T, Chan-Yeung M, Lam WK, Seto WH, Yam LY, Cheung TM, Wong PC, Lam B, Ip MS, Chan J, Yuen KY and Lai KN. A cluster of cases of

60

R O L E

severe acute respiratory syndrome in Hong Kong. Published online at www.nejm.org on 31 March 2003 (10.1056/ NEJ Moa030666) and to appear in the 15 May 2003 issue of the New Engl J Med. 8. Hon KLE, Leung CW, Cheng WTF et al. Clinical presentations and outcome of severe acute respiratory syndrome in children. Lancet, Published online 29 April 2003. 9. Wong KT, Antonio GE, Hui DSC, Lee N, Yuen EH, Wu A, Leung CB, Rainer TH, Cameron P, Chung SS, Sung JJ and Ahuja AT. Thin-section CT of severe acute respiratory syndrome: evaluation of 73 patients exposed to or with the disease. Radiology, Published online at http://radiology. rsnajnls. org/cgi/content/full/2283030541v1 (accessed 8 May 2003). 10. Wong KT, Antonio GE, Hui DSC, Lee N, Yuen EH, Wu A, Leung CB, Rainer TH, Cameron P, Chung SS, Sung JJ and Ahuja AT. Radiological appearances of severe acute respiratory syndrome. J Hong Kong Coll Radiol 2003; 6: 4–6.

O F

C X R

I N

D I A G N O S I S

O F

S A R S

11. Kim EA, Lee KS, Primack SL, Yoon HK, Byun HS, Kim TS, Suh GY, Kwon OJ and Han J. Viral pneumonias in adults: radiologic and pathologic findings. Radiographics 2002; 22: S137–S1349. 12. Putman CE, Curtis AM, Simeone JF and Jensen P. Mycoplasma pneumonia. Clinical and Roentgenographic patterns. Am J Roentgenol 1975; 124(3): 417–422. 13. Reittner P, Muller NL, Heyneman L, Johkoh T, Park JS, Lee KS, Honda O and Tomiyama N. Mycoplasma pneumoniae pneumonia: radiographic and high-resolution CT features in 28 patients. Am J Roentgenol 2000; 174(1): 37–41. 14. McConnell Jr CT, Plouffe JF, File TM, Mueller CF, Wong KH, Skelton SK, Marston BJ and Breiman RF. Radiographic appearance of Chlamydia pneumoniae (TWAR strain) respiratory infections. CBPIS Study Group. Community-Based Pneumonia Incidence Study. Radiology 1994; 192(3): 819–824.

Chest Radiography: Clinical Correlation and Its Role in the Management of Severe Acute Respiratory Syndrome

CHAPTER

DSC Hui, KT Wong, GE Antonio, AT Ahuja and JJY Sung

Introduction Treatment protocol Radiological and clinical patterns during treatment

61 61 62

Introduction Chest radiography (CXR) not only plays an important role in the diagnosis of severe acute respiratory syndrome (SARS), it is crucial in the management of these patients. During treatment there are variable clinical and radiological responses in different patients and serial CXR help in deciding whether escalation to more aggressive treatment is necessary. Based on our preliminary experience, we believe that changes on serial radiographs is also an important prognostic indicator. This chapter aims to examine the correlation between the clinical course and the radiological features, and the role of CXR in the management of SARS.

Clinico-radiological correlation Role of CXR in management of SARS Conclusion

6 65 67 67

correlation one must be familiar with the basic treatment principles. These are therefore discussed briefly in the following paragraph. Patients were treated for the first 2 days with broadspectrum antibiotics for community-acquired pneumonia according to the American Thoracic Society Guidelines [1]. Our initial treatment consisted of intravenous (IV) cefotaxime 1 g every 6 hours and oral clarithromycin 500 mg twice daily (or oral levofloxacin 500 mg daily for those who could not tolerate clarithromycin). Clinical symptoms, arterial blood oxygen saturation and CXR were assessed daily.

Treatment protocol

If fever persisted, patients were given a combination of ribavirin and ‘low-dose’ corticosteroid therapy commencing on days 3–4 (oral ribavirin as a loading dose of 2.4 g stat followed by 1.2 g three times daily and prednisolone 0.5–1 mg/kg body weight per day).

The treatment of SARS patients is discussed in detail in a separate chapter (see Chapter 9). However, in order to better understand the clinical and radiological

Those with dyspnoea at presentation were treated with IV ribavirin (400 mg every 8 hours) combined with hydrocortisone (100 mg every 6 hours).

62

C X R

:

C L I N I C A L

C O R R E L A T I O N

A N D

Pulses of high-dose methylprednisolone (0.5 g IV infusion for 3 consecutive days) were given as a response to persistent fever, radiographic progression of lung opacity and hypoxaemia despite combination therapy. Further pulses of methylprednisolone were given as deemed necessary, up to a total of 3 g.

The intention was to continue with the combination of ribavirin and ‘low-dose’ corticosteroid for up to 12 days. Those who became afebrile but with incomplete radiological resolution were given oral ribavirin 600 mg t.i.d. and prednisolone 0.5 mg/kg body weight per day for at least further 1 week [2].

M A N A G E M E N T

Type 1

25%

Percentage area of lung involvement

Patients who continued to deteriorate despite these measures were given convalescent plasma.

S A R S

Type 2 25%

Type 3

25%

Type 4

Radiological and clinical patterns during treatment Based on our experience treating over 300 SARS patients, there are certain basic radiological and clinical patterns that we have observed.

Radiological pattern during treatment (Figure 6.1) It is clear that serial CXRs contribute significantly in the daily management of patients with SARS. Based on the serial CXRs from the initial 138 SARS patients, we have observed four radiographic progression patterns during treatment [3]: Type 1: Initial radiographic deterioration to peak level followed by radiographic improvement, with maximum radiographic change 25% (Figure 6.2). Type 2: Fluctuating radiographic changes, with at least two radiographic peaks and an intervening trough which differed by 25%. Type 3: Static radiographic changes, with no discernible radiographic peak or change of involvement area 25% for more than 10 days. Type 4: Progressive radiographic deterioration (Figure 6.3).

Time

Fig. 6.1 Schematic diagram showing four radiographic progression patterns of SARS during treatment in hospital.

Key points Four radiographic progression patterns during treatment 1. Initial deterioration followed by improvement (commonest) (70.3%). 2. Fluctuating radiographic changes (17.4%). 3. Static radiographic changes (7.2%). 4. Progressive deterioration (least common, poor clinical outcome) (5.1%).

Clinical pattern during treatment The clinical course of SARS appears to follow a triphasic pattern [4]: Phase 1: This phase (viral replication) is associated with increasing viral load and clinically characterized by fever, myalgia and other

R A D I O L O G I C A L

A N D

C L I N I C A L

(a)

(c)

systemic symptoms that generally improve after a few days. Phase 2: This phase (immuno-pathological damage) is characterized by recurrence of fever, oxygen desaturation and radiological

P A T T E R N S

63

(b)

Fig. 6.2 Type 1 radiographic progression pattern. Serial CXRs in a 28-year-old man with SARS. (a) Frontal CXR at presentation showing ill-defined consolidation in left paracardiac area. (b) Follow-up CXR 7 days later showing progressive airspace lesions involving both lower zones. (c) Subsequent follow-up CXR another 6 days later shows radiographic improvement.

progression of pneumonia with a fall in viral load. The majority of patients will respond to treatment with a combination of ribavirin and IV steroid, whereas 20% of patients may progress into the next phase.

64

C X R

:

C L I N I C A L

C O R R E L A T I O N

(a)

(c)

Phase 3: This phase is characterized by acute respiratory distress syndrome (ARDS) necessitating ventilatory support. Among these 138 cases, 36 (26.1%) were admitted to the intensive care unit (ICU) because of respiratory

A N D

S A R S

M A N A G E M E N T

(b)

Fig. 6.3 Type 4 radiographic progression pattern. Serial CXRs in a 74-year-old woman with SARS. (a) Frontal CXR at presentation shows ill-defined consolidation in right lower zone. (b) Follow-up CXR after 3 days showing radiographic progression with more extensive involvement of right upper and middle zones. New finding of ill-defined airspace opacification involving left upper and middle zones is noted. (c) Follow-up CXR after another 3 days shows progressive radiographic deterioration. The patient succumbed on the day after the last radiograph.

failure. In the first 4 weeks of this outbreak, there were eight mortalities (crude mortality rate
5.8%). All eight cases were originally admitted for major medical conditions. Two patients suffered from myelodysplastic syndrome, four with cardiac diseases

C L I N I C O

-

R A D I O L O G I C A L

C O R R E L A T I O N

(one with congestive heart failure, two with ischaemic heart disease and one with rheumatic heart disease), one with alcoholic liver cirrhosis and one with hepatitis B reactivation.

Key points

• Triphasic clinical pattern of SARS during

•

treatment: – Viral replication phase – Immuno-pathological damage – Acute respiratory syndrome Over 25% of patients develop respiratory failure requiring intensive care.

Radiographic changes and correlation with clinical course Phase 1 The initial CXR was performed on average 2.5 days after onset of fever (range 0–10 days). On presentation, the majority (78.3%) of our patients with SARS had evidence of airspace consolidation on their CXRs. One must note that based on a single CXR the radiographic appearances of SARS cannot be distinguished from other causes of pneumonia. The initial CXR may appear normal up to 25% of cases [5] but serial CXR invariably demonstrates abnormal airspace disease after 1–7 days (median 3 days). The opacities occupy a peripheral or mixed peripheral and axial location in 88% of patients [3]. The rapid radiographic progression, predominant involvement of lung periphery and the lower zone, in addition to the absence of cavitation, hilar lymphadenopathy or pleural effusion, are the more distinctive radiographic features of SARS [2,3].

65

mean 8.6  3.1 days from fever onset in about 80% of our patients, at almost the same time as clinical deterioration with recurrence of fever and respiratory failure. Even when there is only 10% of pneumonic changes noted in the lung field, approximately 40% of patients would require supplementary oxygen to maintain satisfactory oxygenation. Of the patterns of radiographic progression, type 1 was the commonest (70.3%) followed by type 2 (17.4%), type 3 (7.2%) and type 4 (5.1%) patterns [4]. Radiographic improvement following pulse steroid treatment is commonly encountered towards the end of phase 2 [2,3].

Phase 3 The significant morbidity of SARS was reflected by the wide spectrum of clinical symptoms followed by progression of consolidation leading to respiratory failure, intensive care admission, ARDS and death in some cases. In addition, there appears to be a high incidence of pneumo-mediastinum/ pneumothorax (Figures 6.4 and 6.5) causing worsening of hypoxaemia, which may occur either spontaneously or be related to barotrauma in SARS [4,5]. These complications are probably due to poor lung compliance.

Key points

• The initial CXR may appear normal up to •

•

25% of cases. SARS predominantly involves lung periphery and the lower zone with absence of cavitation, hilar lymphadenopathy or pleural effusion. Radiographic progression occurs in line with clinical deterioration at approximately the beginning of the 2nd week with a high incidence of spontaneous pneumomediastinum and/or pneumothorax.

Phase 2 Radiographic progression from unilateral focal airspace opacity to either multi-focal or bilateral involvement occurs during the 2nd week of the disease. Fifty-nine patients (54.6%) had focal unilateral opacity, while 49 (45.4%) had unilateral multi-focal or bilateral involvement. Radiographic progression of consolidation occurred at a peak of

Clinico-radiological correlation Since SARS is a new disease, little is known about its indicators of poor prognosis. We have examined the

66

C X R

:

C L I N I C A L

C O R R E L A T I O N

A N D

S A R S

M A N A G E M E N T

Fig. 6.4 Frontal CXR in a 46-year-old man with SARS showing the presence of pneumothorax and loculated pleural effusion during in-patient treatment. A chest drain and pig-tail catheter have been inserted for drainage.

Fig. 6.5 Frontal CXR in a 33-year-old man with SARS shows complication of pneumo-mediastinum and surgical emphysema.

role of imaging and various biochemical parameters which may predict poor outcome [2].

with ICU admission/death than those with 1 zone involvement (P  0.001 for both).

Radiographic changes and clinical outcome

Patients with bilateral pneumonic changes on the initial CXR were more likely to require ICU care or have a fatal outcome in comparison to those with unilateral pneumonia on admission (P  0.001).

The extent of pneumonia on presentation appears to correlate with adverse clinical outcome in SARS. Those who either required ICU care or died had more extensive radiological evidence of pneumonia (%) on the initial CXR (median 3.30, interquartile range. 1.70–7.93 versus 1.70, interquartile range 0–3.30, P  0.001) and day 7 CXR from fever on set (median 15.00, interquartile range 6.48–28.73 versus 5.00, interquartile range 2.50–7.50, P  0.001) compared to those survivors not requiring ICU care. Those with consolidation 1 zone on the initial CXR and day 7 CXR were significantly more associated

Our multivariable analysis has shown that more than one CXR zone involvement on presentation (odds ratio
3.16; 95% confidence interval (CI)
1.07–9.32; P
0.037) is an independent predictor of adverse outcome after adjusting for high baseline lactate dehydrogenase (LDH), advanced age and a high neutrophil count. In addition, the pattern of radiographic progression also appears to correlate well with clinical outcome. Among the 97 patients with type 1 pattern on serial CXR, only 17 (17.5%) were either admitted to ICU or dead whereas 21 out of 41 patients (51.2%) with CXR category other than type 1 were either admitted to

67

R E F E R E N C E S

ICU or dead (P  0.001). Of these 21 patients, 14 had type 2, while seven had type 4 pattern. Thus patients with type 1 pattern (initial radiographic progression followed by improvement) on serial CXR seemed to have a more favourable outcome, whereas patients with type 4 pattern (progressive deterioration) had an adverse clinical outcome.

Key points Radiological predictors of poor clinical outcome

• more extensive radiographic disease (%) at presentation,

• more than one zone involvement on presentation,

• type 4 progression pattern.

Radiographic changes and laboratory features Among all the laboratory parameters, there was a positive correlation between CXR trend and the rate of change of LDH, a marker of tissue damage. The rate of change of LDH (units/day) significantly correlated with the rate of change of CXR % involvement (Spearman rs
0.399, P
0.014). We believe that LDH reflects the extent of lung injury in this setting, and both serial CXRs and LDH levels are important in the management of SARS.

lung changes by at least 10% or development of new or contralateral lung lesions) and/or hypoxaemia, we initiated treatment with IV pulse methylprednisolone to prevent immuno-pathological lung injury, on the rationale that progression of the pulmonary disease may be mediated by the host inflammatory response [4]. The peak of the extent of pneumonic changes has corresponded to the time of commencement of pulse IV methylprednisolone treatment. The median time of starting the first pulse of methylprednisolone was 8 days (interquartile range 6–9 days). Almost 90% of our patients have shown favourable response with defervescence, resolution of radiographic changes by at least 25%, and oxygen independence after steroid treatment. Serial CXRs are essential especially in the intensive care setting to detect complications such as pneumothorax and pneumo-mediastinum. In a case series, 12% and 20% of patients developed spontaneous pneumo-mediastinum and evidence of ARDS, respectively, over a period of 3 weeks [4].

Key points Serial CXR is essential in

• monitoring disease progression, • response to treatment, • development of pulmonary complications.

Conclusion CXR is a useful modality in the diagnosis and management of SARS.

Key point Radiographic progression correlates with LDH, a marker of lung injury.

More extensive radiographic involvement on presentation is associated with adverse clinical outcome. Serial CXRs are essential for guiding appropriate medical treatment, monitoring response to treatment and detecting development of complications such as pneumo-mediastinum and pneumothorax.

Role of CXR in management of SARS

References

Serial CXRs are most useful in guiding treatment of disease progression. Based on the evidence of radiographic progression (worsening of pre-existing

1. Guidelines for the management of adults with communityacquired pneumonia: diagnosis, assessment of severity, antimicrobial therapy and prevention. Am J Respir Crit Care Med 2001; 163: 1730–1754.

68

C X R

:

C L I N I C A L

C O R R E L A T I O N

2. Lee N, Hui DS, Wu A et al. A major outbreak of severe acute respiratory syndrome in Hong Kong. New Engl J Med 2003; 348. 3. Wong KT, Antonio GE, Hui DS, et al. Severe acute respiratory syndrome: radiographic appearances and pattern of progression in 138 patients. Radiology 2003; 228: 401–406. 4. Peiris JS, Chu CM, Cheng VC et al. Clinical progression and viral load in a community out break of coronavirus-associated

A N D

S A R S

M A N A G E M E N T

SARS pneumonia: a prospective study. Lancet 2003; 361: 1767–1772. 5. Booth CM, Matukas LM, Tomlinson GA et al. Clinical features and short-term outcomes of 144 patients with SARS in the greater Toronto area. J Am Med Assoc 2003; 289: 2801–2809.

The Role of High-Resolution Computed Tomography in Diagnosis of SARS

CHAPTER

7

GE Antonio, KT Wong, DSC Hui and AT Ahuja

Introduction Scanning technique Indication Imaging features on HRCT

69 70 70 72

Differential diagnosis Diagnostic pitfalls Conclusion

Introduction Plain radiography and high-resolution computed tomography (HRCT) are the cornerstones for imaging the lungs. HRCT is capable of imaging the lungs with excellent spatial resolution, providing anatomical detail similar to that available from gross pathological specimens or lung slices [1–3]. It is especially good for the early detection and characterization of localized or diffused lung parenchymal abnormalities. However HRCT involves a high-radiation dose, is not readily available and therefore may not be suitable as the first line of investigation for suspected severe acute respiratory syndrome (SARS) patients or in a screening role in endemic/pandemic situation. In such a situation, HRCT should be reserved for selected group of patients with good clinical indication and non-diagnostic chest radiograph (CXR). The indications should be more relaxed with sporadic cases. The diagnostic protocol for imaging and the use of CXR and HRCT have been discussed previously (Figure 7.1). Apart from diagnosis, HRCT also plays an important role to monitor progress and response to treatment and for follow-up. These will be dealt with separately in later chapters. This chapter aims to give the reader an insight about the role of HRCT in diagnosis of

74 76 76

Suspected SARS

CXR Positive

Negative (if strong clinical suspicion)

Treatment

HRCT Positive

Negative

Treatment

Follow-up

Fig. 7.1 Imaging protocol for suspected SARS patients in our institution.

SARS and to describe various radiological appearances on HRCT.

Key points Introduction

• Currently there is no simple diagnostic test that can reliably diagnose SARS.

70

R O L E

• HRCT provides excellent spatial resolution • •

and anatomical detail of the lungs. Due to radiation hazard it cannot be used as first-line investigation or screening tool in an epidemic. Indications for scanning may be more relaxed in sporadic cases.

O F

H R C T

I N

D I A G N O S I S

O F

S A R S

Key points HRCT imaging technique

• Patient in supine position, full inspiration • 1-mm collimation, 6-mm inter-slice gap • 120 kV, 140 mA, with a scan time of • •

1 second Close patient monitoring during scanning Infection control guidelines to be strictly followed.

Scanning technique The imaging technique for suspected SARS patients is similar to that of other pulmonary diseases [4]. As there was no literature on the imaging features of SARS to consult in the initial phase of the outbreak, Wong et al. performed both conventional CT (without intravenous contrast) and HRCT for delineation of lung parenchymal, mediastinal and pleural abnormalities [6]. As experience about this disease increased with time and with an initial observations of absence of lymphadenopathy and pleural abnormality, HRCT alone was performed for suspected SARS patient. The examination is done with the patient in supine position at full inspiration. Scanning parameters are standardized at 1-mm collimation, 6-mm inter-slice gap, 120 kV, 140 mA, scan time of 1 second for each axial slice. The scanning procedure of the entire thorax takes about approximately 1 minute. For patients in respiratory distress, scanning in shallow breathing is performed. Images are viewed at lung window settings (window level: 700 Hounsfield units (HU); window width: 1500 HU) after image reconstruction using a highspatial frequency algorithm. Close monitoring of vital signs including SaO2, pulse and blood pressure should not be neglected especially for critically ill patients. Meticulous attention should be placed on infection control measures in the CT suite. The measures are designed to protect: 1. staff working in the CT suite; 2. other patients who would be examined using the same scanners. The infection control measures are discussed in detail in a separate chapter towards the end of this book (see Chapter 15).

Indication Whether during the early phase of an epidemic or later when dealing with sporadic cases, HRCT has a higher sensitivity in detecting lung abnormalities than CXR in diagnosing SARS. As discussed earlier, this sensitivity difference should be larger in sporadic cases than earlier in an epidemic. In the initial phase of the epidemic, the majority (78%) of patients with SARS had abnormal CXR on clinical presentation [5]. In the presence of relevant clinical context and laboratory findings, the presence of airspace opacification on CXR helped to confirm the diagnosis of SARS. In the remaining 22% of patients, the initial CXR at presentation was negative [5]. In those patients with strong clinical suspicion (including positive contact history, high fever, lymphopaenia, thrombocytopaenia, elevated lactate dehydrogenase (LDH) level, etc.) and negative CXR, HRCT played an important role in the detection of early lung parenchymal changes to support the clinical diagnosis. For patients with strong clinical suspicion and CXR finding of airspace opacification, HRCT is usually not necessary as the diagnosis is already supported by the CXR findings. HRCT does not add further information in terms of diagnosis, though in some patients additional lung parenchymal involvement can be demonstrated apart from those revealed by CXR. The importance of good clinical assessment for the level of suspicion cannot be overemphasized. In a study by Wong et al. [6], all (17 out of 17) patients with high clinical index of suspicion and negative CXR had lung parenchymal changes detectable on HRCT whereas all 34 patients with minor symptoms

71

I N D I C A T I O N

Fig. 7.2 HRCT of a 50-year-old female SARS patient showing small patch of ground-glass opacification in antero-lateral basal segment of left lower lobe. Subtle areas of ground-glass opacification are present in the rest of both lower lobes (arrows). Note CXR on the same day is clear.

(a)

or low level of clinical suspicion had normal HRCT. Therefore in our institution, HRCT was performed only in patients with high clinical index of suspicion and a negative initial CXR. HRCT was also used in suspected cases with subtle or equivocal findings on CXR (Figure 7.2). However, one must note that in the absence of an epidemic setting, the algorithm for the use of HRCT should be revisited. In such circumstances, the patients are usually frail and old with no obvious history of contact with SARS patients. Many of these patients have pre-existing disease and CXR alone may not be very useful. Therefore the threshold for indication of HRCT may be lowered.

Key points

(b)

Fig. 7.3 A 43-year-old female SARS patients: (a) Postero-anterior CXR showing a small area of increased opacity in left lower zone (arrow). The finding is equivocal, especially in female patient with dense breast shadows. (b) HRCT showing small area of ground-glass opacification in posterior basal segment of left lower lobe corresponding to the CXR change.

HRCT not recommended for:

Indication for HRCT in patients with suspected SARS:

• all patients with low clinical index of

•

•

•

high clinical index of suspicion with negative CXR; high clinical index of suspicion with equivocal CXR findings (especially in females with dense breasts) (Figure 7.3).

suspicion; high clinical index of suspicion with definite CXR changes.

Note: Indications more relaxed at a late stage of the epidemic/with sporadic cases.

72

R O L E

O F

H R C T

I N

D I A G N O S I S

O F

S A R S

Imaging features on HRCT The radiological features of SARS on HRCT are variable, depending on the stage of the disease at presentation [6]. For patients presenting during the early phase of the disease, the HRCT abnormalities are less extensive than those with late presentation. It is important to recognize that there is no single radiological sign to definitely diagnose SARS without knowing the patients’ clinical information.

Appearances In the author’s experience, the most common appearances of lesions in patients with SARS are the following: 1. Ground-glass opacification, which is defined as increased lung parenchymal attenuation without obscuring the underlying vascular architecture [7]. This appearance accounts for 68% of all 149 lesions seen in our initial series of 40 patients with SARS [8]. 2. Consolidation, defined as opacification where the underlying vasculature is obscured [7], is seen in 32%, half of these are admixed with areas of ground-glass opacification (Figures 7.4–7.8). 3. Within areas of ground-glass opacification, there are associated abnormalities including thickened intralobular interstitium (32%) and interlobular septa (24%) [6]. A ‘crazy-paving’ appearance is noted if these interstitial changes are marked (Figures 7.9 and 7.10).

Fig. 7.4 HRCT of a 51-year-old female with SARS showing focal area of ground-glass opacification involving posterior and lateral basal segments of left lower lobe (arrow).

Fig. 7.5 HRCT of a 29-year-old male with SARS showing multiple peripheral subpleural ground-glass opacifications in both lower lobes.

Fig. 7.6 HRCT of a 50-year-old male with SARS showing more extensive consolidation and groundglass opacification in both lower lobes.

Fig. 7.7 HRCT of a 61-year-old male with SARS showing mixed consolidation (arrow) and groundglass opacification in right middle lobe.

I M A G I N G

F E A T U R E S

O N

73

H R C T

4. Bronchial dilatation is present in 7% of lesions and affects the segmental bronchi supplying the area of parenchymal opacification [6]. 5. There is no peri-bronchovascular interstitial thickening, mass or nodule, emphysema, cavitation or pleural effusion encountered in our series.

Location 1. The lesions tend to be peripheral (72%) or both central and peripheral (20%) while pure central location is uncommon (8%) [6]. 2. Although all segments of the lung can be involved, there is a slight predominance of lower lobe involvement. 3. Bilateral involvement occurs in about half of patients, which is more common in the advanced cases (61% versus 18%).

Key points HRCT features of SARS

• Ground-glass opacification  consolidation

• Thickened intralobular interstitium and interlobular septa

• Peripheral/subpleural in location • Unilateral/bilateral (depends on the stage of disease at presentation). Cavitation, lymphadenopathy and pleural effusion are not imaging features Fig. 7.8 HRCT of a 52-year-old patient with SARS showing pure consolidation in right middle lobe.

Fig. 7.9 HRCT of a 33-year-old patient with SARS showing interlobular septal thickening (arrow heads) within the ground-glass opacity in left lower lobe.

Fig. 7.10 HRCT of a 25-year-old female with SARS showing markedly thickened intralobular interstitium and interlobular septa within ground-glass opacity with ‘crazy-paving’ appearance.

74

R O L E

O F

H R C T

I N

D I A G N O S I S

O F

S A R S

HRCT is especially useful for detection of lung parenchymal changes of SARS in certain radiographical blind spots (Figure 7.11) which are not easily assessed by plain radiograph. These include retrocardiac area (Figure 7.12), paraspinal region, retrodiaphragmatic area over posterior costophrenic sulcus [8] (Figure 7.13). Early disease with small areas of ground-glass opacification is also commonly missed on plain radiograph but is readily demonstrated by HRCT.

Fig. 7.11 HRCT of a 30-year-old female with SARS showing focal consolidation/ground-glass opacification in paraspinal area of left upper lobe (arrow). This is a common radiographical blind spot on CXR.

Key points Blind spots on CXR are best assessed by HRCT

• Retrocardiac area • Paraspinal area • Posterior costophrenic sulcus

Differential diagnosis

Fig. 7.12 HRCT of a 28-year-old male with SARS showing focal consolidation in left retrocardiac area (arrow). The corresponding CXR on the same date is normal.

Fig. 7.13 HRCT of a 33-year-old male with SARS showing area of ground-glass opacification in right posterior costophrenic sulcus which is not detectable on CXR.

It should be stated from the beginning that none of the CT features of SARS are themselves specific or diagnostic. The differential diagnosis in terms of the CT findings includes atypical pneumonia caused by other infective agents, bronchiolitis obliterans organizing pneumonia (BOOP) (Figure 7.14) and chronic oeosinophilic pneumonia (CEP) (Figure 7.15). Atypical pneumonia is most commonly caused by Mycoplasma, Chlamydia, Legionella and virus (such as influenza virus) (Figure 7.16). Centrilobular opacities, acinar shadows, airspace consolidation and ground-glass opacity with a lobular distribution are the most common HRCT features of atypical pneumonia [9]. However, in our experience, these are not consistent findings in patients with SARS. Mycoplasma pneumonia is a common cause of community-acquired pneumonia, accounting for up to 30% of all pneumonias in the general population [9,10]. The most commonly described HRCT findings in patients with serologically proven mycoplasma pneumonia are [11]:

• areas of ground-glass opacification (85%) and airspace consolidation (78%);

D I F F E R E N T I A L

D I A G N O S I S

Fig. 7.14 HRCT of a 44-year-old female with BOOP showing multiple peripheral subpleural consolidation (black arrows), peribronchial consolidation (white arrow) and thickened bronchovascular interstitium (arrowheads).

75

Fig. 7.16 HRCT of a 17-year-old patient with influenza pneumonia showing multiple small acinar/ centrilobular nodules (arrowheads) and consolidation (arrow) in right upper lobe. Occasional ‘tree-in-bud’ appearances are present.

The radiological findings reflect the variable extents of the histopathological features: diffuse alveolar damage (intra-alveolar oedema, fibrin and variable cellular infiltrates with a hyaline membrane), intraalveolar haemorrhage and interstitial (intrapulmonary or airway) inflammatory cell infiltration. BOOP is a disease of unknown cause characterized by the presence of granulation tissue polyps within lumina of bronchioles and alveolar ducts and patchy areas of organizing pneumonia. Typical HRCT features include the following [13–15]: Fig. 7.15 Spiral CT of a 28-year-old female with CEP with multiple peripheral consolidations in both lungs.

• lobular distribution of consolidation (59%); • nodules in 89% patients, predominantly centrilobular in distribution;

• thickening of the bronchovascular bundles in 82% of patients. Influenza virus types A and B cause most cases of viral pneumonia in immunocompetent adults. Common HRCT findings include the following [12]: 1. poorly defined centrilobular nodules; 2. ground-glass attenuation with a lobular distribution; 3. segmental consolidation; 4. diffuse ground-glass attenuation with thickened interlobular septa.

1. patchy consolidation or ground-glass opacification in subpleural and/or peribronchial distribution; 2. small ill-defined nodules that may be peribronchial or peribronchiolar; 3. large nodules or mass; 4. bronchial wall thickening or dilatation in abnormal lung regions; 5. crazy-paving, with superimposition of groundglass opacification and interlobular septal thickening may also be seen in patients with BOOP; 6. lower lung zones are involved more commonly than upper lung zones. CEP is an idiopathic condition characterized by extensive filling of alveoli by mixed inflammatory infiltrates consisting primarily of oeosinophils. Common HRCT findings [16,17] include: 1. consolidation in lung periphery and patchy in distribution;

76

R O L E

O F

H R C T

I N

D I A G N O S I S

O F

S A R S

2. patchy or peripheral ground-glass opacification, sometimes associated with crazy-paving; 3. linear or band-like opacities; 4. an upper lobe predominance of abnormalities. Since there is much overlap in radiological appearances of SARS and these pulmonary disorders, clinical information is indispensable for accurate diagnosis. In patients with clinical presentation of high fever, chills and rigor with recent contact history and laboratory findings of lymphopaenia and thrombocytopaenia, the presence of lung parenchymal abnormality strongly supports the diagnosis of SARS.

Key points

Fig. 7.17 Dependent densities (arrows) mimicking consolidation. The characteristic distribution in dependent positions and symmetrical in nature help differentiate from genuine pulmonary lesion.

Differential diagnosis of HRCT findings

• Other atypical pneumonias (including mycoplasma and viral pneumonia)

• BOOP • CEP

Diagnostic pitfalls With use of modern CT scanners and meticulous imaging techniques, HRCT helps in the diagnosis of early cases of SARS. However, several confusing artefacts may be seen on HRCT, which may pose diagnostic difficulty, especially for the inexperienced. Familiarity with their appearances should eliminate potential misdiagnosis. Atelectasis is commonly seen in dependent lung in both normal and abnormal subjects, resulting in a so-called dependent density [18] (Figure 7.17). This normal finding can closely mimic the appearances of consolidation or early lung fibrosis. It can be easily distinguished from true pathology by obtaining scans in both supine and prone positions. In our experience, it was seldom required given the specific appearances of this artefact. Motion artefact due to respiratory movement is especially important to note in patients with respiratory distress (Figure 7.18). Pulsation artefact from adjacent heart and major thoracic vessels are also commonly encountered. These could potentially create

Fig. 7.18 Pulsation artefacts from adjacent cardiac motion (arrow) may mimic ground-glass opacification. Note the presence of ground-glass opacification/ consolidation in periphery of right lower lobe.

areas of increased attenuation mimicking lung parenchymal abnormalities.

Conclusion HRCT is a useful imaging tool for early diagnosis of patients with SARS. It is especially useful for patients with high clinical suspicion and negative CXR at initial presentation. In our experience, patchy ground-glass opacification with or without consolidation in a peripheral distribution were the most typical appearances of SARS on HRCT.

R E F E R E N C E S

References 1. Muller NL and Miller RR. Computed tomography of chronic diffuse infiltrative lung disease: Part 1. Am Rev Respir Dis 1990; 142: 1206–1215. 2. Muller NL and Miller RR. Computed tomography of chronic diffuse infiltrative lung disease: Part 2. Am Rev Respir Dis 1990; 142: 1440–1448. 3. Colby TV and Swensen SJ. Anatomic distribution and histopathologic patterns in diffuse lung disease: correlation with HRCT. J Thorac Imaging 1996; 11(1): 1–26. 4. Webb WR, Muller NL and Naidich DP. Technical aspects of high-resolution computed tomography. In: High-Resolution CT of the Lung, 3rd edition. Lippincott Williams & Wilkins, Philadelphia, USA. 2001, 1–48. 5. Wong KT, Antonio GE, Hui DS, Lee N, Yuen EH, Wu A, Leung CB, Rainer TH, Cameron P, Chung SS, Sung JJ, Ahuja AT. Severe acute respiratory syndrome: radiographic appearances and pattern of progression in 138 patients. Radiology 2003 Aug; 228(2): 401–406. Epub 2003 May 20. 6. Wong KT, Antonio GE, Hui DS, Lee N, Yuen EH, Wu A, Leung CB, Rainer TH, Cameron P, Chung SS, Sung JJ, Ahuja AT. Thin-section CT of severe acute respiratory syndrome: evaluation of 73 patients exposed to or with the disease. Radiology 2003 Aug; 228(2): 395–400. Epub 2003 May 08. 7. Austin JHM, Muller NL, Friedman PJ et al. Glossary of terms of CT of the lungs: Recommendations of the nomenclature committee of the Fleischner Society. Radiology 1996; 200: 327–331. 8. Antonio GE, Wong KT, Hui DS, Lee N, Yuen EH, Wu A, Chung SS, Sung JJ, Ahuja AT. Imaging of severe acute respiratory syndrome in Hong Kong. Am J Roentgenol 2003 Jul; 181(1): 11–17.

77

9. Tanaka N, Matsumoto T, Kuramitsu T et al. High resolution CT findings in community-acquired pneumonia. J Comput Assist Tomogr 1996; 20: 600–608. 10. John SD, Ramanathan J and Swischuk LE. Spectrum of clinical and radiographic findings in pediatric mycoplasma pneumonia. Radiographics 2001; 21(1): 121–131. 11. Reittner P, Muller NL, Heyneman L et al. Mycoplasma pneumoniae pneumonia: radiographic and high-resolution CT features in 28 patients. Am J Roentgenol 2000; 174: 37–41. 12. Kim EA, Lee KS, Primack SL et al. Viral pneumonias in adults: radiologic and pathologic findings. Radiographics 2002; 22(suppl): 137S–149S. 13. Arakawa H, Kurihara Y, Niimi H, Nakajima Y, Johkoh T and Nakamura H. Bronchiolitis obliterans with organizing pneumonia versus chronic eosinophilic pneumonia: highresolution CT findings in 81 patients. Am J Roentgenol 2001; 176(4): 1053–1058. 14. Muller NL, Staples CA and Miller RR. Bronchiolitis obliterans organizing pneumonia: CT features in 14 patients. Am J Roentgenol 1990; 154: 983–987. 15. Bouchardy LM, Kuhlman JE, Ball WC et al. CT findings in bronchiolitis obliterans organizing pneumonia (BOOP) with radiographic, clinical, and histologic correlation. J Comput Assist Tomogr 1993; 17: 352–357. 16. Johkoh T, Muller NL, Akira M et al. Eosinophilic lung diseases: diagnostic accuracy of thin-section CT in 111 patients. Radiology 2000; 216(3): 773–780. 17. Ebara H, Ikezoe J, Johkoh et al. Chronic eosinophilic pneumonia: evolution of chest radiograms and CT features. J Comput Assist Tomogr 1994; 18: 737–744. 18. Primack SL, Remy-Jardin M, Remy J et al. High-resolution CT of the lung: pitfalls in the diagnosis of infiltrative lung disease. Am J Roentgenol 1996; 167: 413–418.

The Role of Imaging in the Follow-up of SARS

CHAPTER

GE Antonio, KT Wong, DSC Hui and AT Ahuja

Introduction Follow-up presentation of SARS patients Pathological considerations Role of imaging in the follow-up of SARS

79 79 79

8

Imaging appearances Protocol for follow-up imaging Conclusion

81 87 87

80

Introduction Severe acute respiratory syndrome (SARS) has shown itself to be different from most other forms of viral pneumonia in its infectivity, clinical course, predilection for affecting health care workers, and high rates of mortality and morbidity [1]. During the acute phase of the epidemic the imaging characteristics of SARS during the acute phase been investigated, but its post-treatment sequelae are only just becoming apparent as they surface in the imaging of patients attending follow-up. In line with the acute stages of this disease, the recovery also appears to be punctuated with an exaggeration of the host response, with patients developing residual disease or early signs of fibrosis in affected areas of the lungs. With this in mind, the follow-up of these patients will require close clinical and radiological monitoring. This chapter shall present the appearances and role of imaging in the follow-up of SARS.

Follow-up presentation of SARS patients Follow-up is usually uneventful for most other types of viral pneumonia in adults. However, while a portion

of treated and discharged SARS patients may be completely asymptomatic, a significant number have residual symptoms. It has been reported that 46% of discharged patients complained of exertional dyspnoea at 1-month follow-up [2].This was not restricted to elderly patients but also affected patients in their 30s, resulting in the limitation of their daily activities.

Key points Post-treatment SARS patients 1. Asymptomatic (54%) 2. Dyspnoeic (46%)

Pathological considerations For the more common viral pneumonia in adults (such as the influenza virus, adenovirus, herpes simplex I and varicella-zoster viruses) despite substantial morbidity, most cases are associated with complete clinical recovery. On the whole, viral pneumonia usually resolves without significant clinical or radiological sequelae [3]. Poor outcome is usually found only at the extremes of age [4] and among the

80

R O L E

O F

immunocompromised. There are rare exceptions and the consequences range from poor lung function after varicella pneumonia [5], adenovirus pneumonia causing bronchiolitis obliterans [6], to lung fibrosis seen in influenza pneumonia [7]. To these complications we now add SARS-related pulmonary fibrosis. For SARS, over half (62.5%) of the patients had architectural distortion and other signs of possible fibrosis on early follow-up [2]. While this may represent early scarring as a result of the viral infection itself, patients with such changes are found to be those who had a more stormy clinical course and required more intense therapy. The latter suggests that damage may be due to lung inflammation by an exaggerated cell-mediated host immune response elicited by viral antigen [8], a phenomenon that is apparent during the acute stages of the disease. As a comparison, an exaggerated host response is seen in other complications related to viral infections such as bronchiolitis obliterans organizing pneumonia (BOOP) and adult respiratory distress syndrome (ARDS). SARS shares some features with both of these diseases:

• The acute radiological features of SARS, espe•

• •

cially those on computed tomography (CT), are similar to those seen in BOOP [1,8,9]. Some of the SARS patients, especially those with a more stormy in-patient course, develop ARDSlike radiographical appearance during the peak of their acute illness [10]. Post-mortem of patients who have succumbed to SARS, show changes consistent with ARDS are present in the lungs [8]. Pulmonary fibrosis is a known complication of ARDS.

With the above in mind, the possible early fibrosis seen in SARS may indeed be a modified expression of BOOP. Finally, treatment, part of which remains on a trial basis, may also play a significant role on the eventual lung damage.

Key points

I M A G I N G

I N

F O L L O W

-

U P

O F

S A R S

2. Over-activity of host immune response (SARS, BOOP and ARDS) 3. Treatment side effect

Role of imaging in the follow-up of SARS Chest radiographs and high-resolution CT The chest radiograph examination is a major diagnostic component endorsed by the World Health Organization (WHO) and Center of Disease Control and Prevention (CDC) in their guidelines [11,12]. Thin-section or high-resolution CT (HRCT) has been helpful in diagnosing the more difficult and probably earlier SARS cases where the chest radiograph is normal [1,13]. Both modalities have shown themselves useful for monitoring progress and complications during treatment. As a continuation of their role during the acute illness, chest radiographs and fine cut or HRCT are the mainstay for the follow-up for SARS patients. A study has shown that HRCT was rarely (4.2%) normal in follow-up patients with dyspnoea [2]. The same study has also shown how soon (mean follow-up period of 36.5 days after hospital admission and 17.8 days after discharge) HRCT architectural distortion and suggestions of fibrosis may begin in patients with SARS. How much these changes will resolve in future is unknown, although it is unlikely that the areas of severe architectural distortion (Figure 8.1) will completely disappear. In the more pronounced cases of residual disease, further follow-up by radiographs should be sufficient. Finally, a very small proportion of patients may develop acute symptoms of fever, malaise and increased dyspnoea after discharge. These may be suggestive of incomplete resolution or reactivation of disease. Imaging plays a role in monitoring the lung involvement in these patients in the same manner as for the acute stages of the disease.

Pulmonary fibrosis post-SARS

Other types of imaging

Possible causes: 1. Sequelae damage

of

infection

causing

lung

Magnetic resonance imaging (MRI) has so far only had a minor role to play. In our experience,

I M A G I N G

81

A P P E A R A N C E S

Fig. 8.1 A 37-year-old male, treated and discharge for SARS. Follow-up HRCT on day 47 (from the 1st day of symptoms). The right hemithorax is smaller than the left at this level. There are signs of fibrosis such as irregular interfaces, traction bronchiectasis (arrow), parenchymal band (arrowhead), pleural and fissural distortion (open arrow). Ground-glass opacification is also present and may represent residual inflammation or due to fibrosis.

Fig. 8.2 A 36-year-old female with no previous illness. Discharged after treatment for SARS. MRI was performed on day 81 (from onset of SARS symptoms). Coronal T1-weighted (T1W) image of both hips showing avascular necrosis affecting the subchondral bone of both femoral heads.

a few patients developed confusion during treatment, and we have performed both CT and MRI of the brain on them. Neither modality showed any abnormality, the confusion in these patients was thus considered to be a side effect of the corticosteroid therapy. In a few treated patients (four patients at the time of writing this chapter) who developed bone or joint pain during or after their SARS treatment, MRI was performed on the areas of complaint. These were normal except in one patient, where widespread areas of avascular necrosis were present involving both lower limbs (Figures 8.2 and 8.3).

Key points

Fig. 8.3 Same patient as Figure 8.2. Coronal T1W image of the right knee showing avascular necrosis affecting the marrow of the distal tibial and proximal fibula. Similar changes were present in the other knee.

Role of imaging in follow-up 1. Monitor imaging progress of patients dyspnoeic on follow-up 2. Detection of complication (fibrosis) 3. Monitor response or complications to drugs 4. Monitor reactivation of the disease

Imaging appearances Chest radiograph The chest radiograph helps in the diagnosis by demonstrating lung opacities and plays an important role in evaluating the progress of disease and

82

R O L E

O F

I M A G I N G

I N

F O L L O W

-

U P

O F

S A R S

response to treatment [1,13]. These opacities are usually still present to a variable extent at hospital discharge [10], keeping in mind that hospital discharge criteria used in our institution are based on:

• being afebrile for at least 96 hours after the last dose of steroid;

• resolving respiratory symptoms and oxygen independence;

• radiological improvement (based on serial chest radiographs);

• improving laboratory parameters. Evidently, complete resolution of imaging abnormalities is not requisite. For patients complaining of dyspnoea and exercise intolerance at 1-month follow-up, 83.3% had residual changes on their earlier chest radiograph on day of discharge. But the percentage of abnormals dropped to 62.5% for the radiographs taken at the time of the 1st month follow-up. When comparing the two sets of radiographs, two-thirds of the radiographs taken on the day of discharge showed an improvement at the time of the 1-month follow-up film, while one-third of the abnormal radiographs showed no worsening [2].

Fig. 8.4 Same patient as the one in Figure 8.1, follow-up chest radiograph on day 59 (from 1st day of symptoms). There is diffuse ground-glass opacification, linear markings and fissural elevation (arrowheads) in the right upper and mid-zone. Elevation of the right hilum indicates volume loss.

The radiographical abnormalities seen on follow-up radiography include (Figure 8.4):

• ground-glass opacification, • linear opacification, • pleural/fissural tethering and evidence of volume loss (seen as shifting of the mediastinal or hilar structures). The findings are non-specific and if read in isolation could represent a combination of residual inflammation, atelectasis or fibrosis.

Key points Follow-up CXR appearances in treated SARS 1. Complete resolution 2. Ground-glass opacification 3. Evidence of fibrosis: linear opacification, volume loss and pleural tethering 4. Lack of temporal change or resolution.

HRCT As previously described, the HRCT findings in SARS for initial diagnosis included:

• ground-glass opacification and/or consolidative • • •

opacification; thickening of interlobular septa when present is superimposed on a ground-glass opacification, giving a ‘crazy-paving’ pattern [9,13] (Figure 8.5); intralobular interstitium thickening; bronchial dilatation within areas of consolidation (suggesting that it is a form of reactive/ respiratory bronchial dilatation).

For patients who have both an initial and follow-up HRCT for comparison progress changes can be expected. There will be a variable degree of resolution of the ground-glass opacification and thickened interlobular septae between the initial and followup scan. This can be considered to represent imaging

I M A G I N G

83

A P P E A R A N C E S

Fig. 8.5 A 25-year-old female, initial HRCT on day 7 (from 1st day of symptoms). There is ground-glass opacification with superimposed smooth thickening of the interlobular septae giving rise to a ‘crazy-paving’ pattern (arrow).

Fig. 8.6 Same patient as in Figure 8.5, follow-up HRCT of the same level on day 43 (from 1st day of symptoms). There is minor residual ground-glass opacification and increased reticular markings (irregular, thickened interlobular septae) (arrow).

evidence of improvement (Figure 8.6). Resolution of the ground-glass opacification suggests the initial lung parenchymal changes are of inflammatory nature and improve after successful therapy.

Fig. 8.7 A 51-year-old female treated and discharged. Follow-up HRCT on day 38 (from 1st day of symptoms). There is patchy residual consolidation and ground-glass opacification, irregular septal thickening. There are secondary lobules (arrow) of normal density within the area of abnormality. This is different from the uniform opacification of the affected area commonly seen at presentation (Figure 8.3).

However, in one study ground-glass opacification was present on most (95.8%) follow-up HRCTs of dyspnoeic patients [1]. Instead of the uniform density ground glass-seen in the initial HRCT, there was variable density between different secondary lobules (Figure 8.7). This may partly be due to different rates of resolution of the initial fluid/inflammation or may be related to distortion caused by fibrosis. In addition, there was clearing of the subpleural 5 mm of the lung, a picture similar to ARDS or pulmonary oedema. The crazy-paving pattern (Figure 8.5), so commonly seen in the initial HRCT [1], is not common in the follow-up HRCT. If thickened septae are present, they are, thinner, distorted and irregular, an appearance more akin to idiopathic pulmonary fibrosis (Figure 8.8). On follow-up HRCT, signs of fibrosis are (Figures 8.1 and 8.9):

• parenchymal bands, • irregular interfaces (bronchovascular, pleural or mediastinal)

• traction bronchiectasis.

84

R O L E

O F

Fig. 8.8 A 32-year-old male treated and discharged. Follow-up HRCT on day 46 (from 1st day of symptoms). There is residual ground-glass opacification with some clearing of the subpleural 5 mm. Superimposed thinner, distorted and irregular interlobular septae thickening (arrow) (as compared to that seen in the acute disease) are more in keeping with the appearances seen in pulmonary fibrosis.

I M A G I N G

I N

F O L L O W

-

U P

O F

S A R S

Fig. 8.10 A 36-year-old male treated and discharged. Follow-up HRCT on day 44 (from 1st day of symptoms). There is peri-bronchovascular thickening and patchy areas of ground-glass opacification (arrow) and irregular interlobular septal thickening (arrowheads).

Thickened interlobular septae or intralobular interstitium should not be used as signs of fibrosis on early follow-up as they may represent unresolved interstitial inflammation. The other signs mentioned are however, not present in the initial HRCT used for diagnosis and are thus more reliable. There were signs of fibrosis (parenchymal band, irregular interface and traction bronchiectasis) and peri-bronchovascular interstitial thickening (Figure 8.1) in 62.5% of dyspnoeic patients on early follow-up [2]. These were associated with architectural distortion resulting in rotation of the fissures and bronchovasculature. The ground-glass opacification in these patients surrounded the areas of fibrosis.

Fig. 8.9 A 33-year-old male treated and discharged. Follow-up HRCT on day 49 (from 1st day of symptoms). There are signs of fibrosis such as parenchymal bands and irregular interfaces (bronchovascular, pleural or mediastinal) (arrows).

Consolidation is not a common feature of SARS on follow-up HRCT. Only very small areas of consolidation were seen in follow-up HRCT and when present are around thickened bronchi (Figure 8.10). There are also small patches of consolidation in the centre of the fibrotic areas, adjacent to the bronchi. There were no masses or nodules, emphysema, cavitation or calcification present.

I M A G I N G

Key points Follow-up HRCT appearances in treated SARS 1. Complete resolution 2. Patchy ground-glass opacification 3. Evidence of fibrosis: architectural distortion, parenchymal bands, irregular interfaces (bronchovascular, pleural or mediastinal) and traction bronchiectasis.

Key points Comparison between initial and follow-up HRCT Initial HRCT Ground-glass Uniform opacification density Septal thickening Architectural distortion Parenchymal bands Irregular interfaces Bronchial dilatation

85

A P P E A R A N C E S

Follow-up HRCT

Uniform and thick Absent

Variable density and subpleural clearing Irregular and thin Present

Absent

Present

Absent

Pleural, mediastinal bronchovascular Respiratory Traction

In SARS patients with follow-up HRCT evidence of fibrosis, the significance of the concomitant groundglass opacification is not clear. If BOOP or forms of idiopathic interstitial pneumonia are used as a frame of reference [14,15], these changes may represent persistent inflammation that is potentially reversible upon treatment. Currently, treatment with corticosteroid or other steroid-sparing immuno-modulating agents (such as cyclophosphamide) have been used in BOOP [16,17]. These are being tried for SARSinduced fibrosis. Hence, the role of HRCT in follow-up of SARS patients is to assess the extent of long-term

lung parenchymal injury/fibrosis and to identify these potentially reversible components early so that appropriate treatment may be instituted to prevent further lung damage. It also plays a role in monitoring the disease response and development of complications in drug trials. The early abnormalities detected on radiographs and HRCT must be interpreted with caution. While some of the changes obviously represent parenchymal fibrosis, others may represent potentially atelectasis and inflammatory exudate. Although the absolute identity of these lesions is as yet unknown, the possibility that there may be a reversible component warrants continuation of treatment to halt or minimize fibrosis and to monitor progress. This is where follow-up imaging is mandatory in the group of patients developing short- to medium-term symptoms. More long-term studies are needed to properly define fibrosis in SARS and the ultimate appearance.

Patients with major complications Intensive care patients A minority of patients survive a very stormy course in the intensive care unit (ICU) before recovery. These patients understandably have more abnormalities on their discharge chest radiographs and HRCT (Figure 8.11). In addition, some of these SARS patients have progressed to ARDS during the peak of their illness, as may occur in other types of viral pneumonia [18]. Fibrosis, to the extent of honeycombing, is known to occur in patients who have survived ARDS and there is evidence that the same may occur in some of the SARS patients. In particular, cystic changes are seen in some of the SARS patients who have developed ARDS during their acute illness (Figures 8.12 and 8.13). These cysts are similar to those seen after ARDS of other causes [19] and are not a feature in the HRCTs from other SARS patients who did not develop ARDS during their acute illness [2].

SARS patients with evidence of lung fibrosis Dyspnoeic patients with follow-up HRCT early evidence of fibrosis were relatively older males [2]. These patients also possess the worst initial HRCTs and the worst appearing chest radiographs during

86

R O L E

O F

I M A G I N G

I N

F O L L O W

-

U P

O F

S A R S

Fig. 8.13 A 54-year-old male in the ICU, same patient as Figure 8.11. Subpleural and parenchymal cysts are present (arrows). Fig. 8.11 A 54-year-old male in the ICU. HRCT on day 67 from 1st day of symptoms (day 53 from the onset of ARDS). Loculated pneumothorax was present (arrow). There are parenchymal bands and irregular interfaces (arrowheads).

treatment. These most likely reflect that these patients suffered a more severe course than those who did not develop evidence of early fibrosis. Patients with HRCT evidence of fibrosis also had a higher requirement of pulse intravenous methylprednisolone during treatment. This is on top of the combination of oral ribavirin and oral corticosteroids used on all patients. High-dose corticosteroid in the form of pulse therapy was given to patients not responding favourably to the standard combination. The need of pulse steroid therapy gives a reflection of the magnitude of the cytokine storm elicited by the viral antigen, which in fact may be the underlying pathogenesis of lung damage and subsequent development of fibrosis. The peak lactate dehydrogenase (LDH) level is also higher in these patients. LDH is an indicator of tissue destruction (presumably lung tissue in SARS) and has shown to be a good independent predictor of worse clinical outcome [8].

Key points SARS fibrosis patient profile

Fig. 8.12 A 33-year-old male from the ICU. HRCT on day 56 from 1st day of symptoms (day 45 from the onset of ARDS). There are areas suggestive of fibrosis within which large (arrow) and small (arrowheads) subpleural cysts are present.

1. 2. 3. 4. 5. 6.

Older (average age 45 years) Majority of worst cases involve males Worst initial HRCT Worst serial chest radiographs Higher pulse steroid requirement Higher peak LDH

R E F E R E N C E S

Protocol for follow-up imaging Chest radiographs could be performed on standard equipment using the established routine. For HRCT we use 1-mm thick slices with 6 mm gap, scanning the patient supine and inspiration. Exposure parameters were set at 1 second, 120 kV and 140 mA: 1. A baseline chest radiograph should be performed on hospital discharge. HRCT may be added on discharge if there are marked radiographical abnormalities, if the patient is breathless or if the clinical course was severe. 2. For dyspnoeic patients, monthly chest radiographs performed just prior to follow-up clinic. This may need to be more frequent warranted by the patients’ condition or if trial treatment is commenced. 3. HRCT at 6-month follow-up, if a discharge HRCT was required. 4. Further imaging follow-up, if clinically required.

Conclusion Pulmonary fibrosis may develop early in a substantial proportion of SARS patients who have been discharged after treatment. Patients who are older and have more severe disease during treatment are more likely to develop HRCT findings of fibrosis. The role of imaging in the follow-up of SARS patients should be focused on these patients. Imaging follow-up will provide information on the progress of disease, a guide to treatment response and demonstrate complications that may arise.

References 1. Wong KT, Antonio GE, Hui DS, Lee N, Yuen EH, Wu A, Leung CB, Rainer TH, Cameron P, Chung SS, Sung JJ, Ahuja AT. Severe acute respiratory syndrome: radiographic appearances and pattern of progression in 138 patients. Radiology 2003 Aug; 228(2): 401–406. Epub 2003 May 20. 2. Antonio GE, Wong KT, Hui DS, Wu A, Lee N, Yuen EH, Leung CB, Rainer TH, Cameron P, Chung SS, Sung JJ, Ahuja AT. Thin-section CT in patients with severe acute respiratory syndrome following hospital discharge: preliminary experience. Radiology 2003 Sep; 228(3): 810–815. Epub 2003 Jun 12.

87

3. Razer RS, Muller NL, Colman N and Pare PD. Viruses, mycoplasmas, chlamydiae, and rickettsiae. In: Fraser and Pare (Eds) Diagnosis of Diseases of the Chest, 4th edition. W. B. Saunders, Philadelphia, PA. 1999, 979–1032. 4. Tillett HE, Smith JW and Gooch CD. Excess deaths attributable to influenza in England and Wales: age at death and certified cause. Int J Epidemiol 1983; 12(3): 344–352. 5. Mohsen AH, Peck RJ, Mason Z, Mattock L and McKendrick MW. Lung function tests and risk factors for pneumonia in adults with chickenpox. Thorax 2001; 56(10): 796–799. 6. Becroft DM. Bronchiolitis obliterans, bronchiectasis, and other sequelae of adenovirus type 21 infection in young children. J Clin Pathol 1971; 24(1): 72–82. 7. Winterbauer RH, Ludwig WR and Hammar SP. Clinical course, management, and long-term sequelae of respiratory failure due to influenza viral pneumonia. Johns Hopkins Med J 1977; 141(3): 148–155. 8. Lee N, Hui D, Wu A, Chan P, Cameron P, Joynt GM, Ahuja A, Yung MY, Leung CB, To KF, Lui SF, Szeto CC, Chung S and Sung JJY. A major outbreak of severe acute respiratory syndrome in Hong Kong. New Engl J Med 2003; 20(348): 1986–1994. 9. Wong KT, Antonio GE, Hui DSC, Lee N, Yuen EHY, Wu A, Leung CB, Rainer TH, Cameron P, Chung SSC, Sung JJY and Ahuja AT. Thin-section CT of severe acute respiratory syndrome: evaluation of 73 patients exposed to or with the disease. Radiology Published online at http://radiology.rsnajnls. org/cgi/content/full/2283030541v1 before print (accessed 8 May 2003). 10. Tsang KW, Ho PL, Ooi GC, Yee WK, Wang T, Chan-Yeung M, Lam WK, Seto WH, Yam LY, Cheung TM, Wong PC, Lam B, Ip MS, Chan J, Yuen KY and Lai KN. A cluster of cases of severe acute respiratory syndrome in Hong Kong. New Engl J Med 2003; 20(348): 1977–1985. 11. World Health Organization. Online document. Preliminary Clinical Description of Severe Acute Respiratory Syndrome. http://www.who.int/csr/sars/clinical/en/ (accessed 21 March 2003). 12. Center for Disease Control and Prevention, USA. Online document. Diagnosis/Evaluation for SARS. http://www.cdc.gov/ ncidod/sars/diagnosis.htm (accessed 7 April 2003, 3:00 PM EDT). 13. Antonio GE, Wong KT, Hui DS, Lee N, Yuen EH, Wu A, Chung SS, Sung JJ, Ahuja AT. Imaging of severe acute respiratory syndrome in Hong Kong. Am J Roentgenol 2003 Jul; 181(1): 11–17. 14. Lee KS, Kullnig P, Hartman TE and Muller NL. Cryptogenic organizing pneumonia: CT findings in 43 patients. Am J Roentgenol 1994; 162(3): 543–546. 15. Kim EY, Lee KS, Chung MP, Kwon OJ, Kim TS and Hwang JH. Nonspecific interstitial pneumonia with fibrosis: serial highresolution CT findings with functional correlation. Am J Roentgenol 1999; 173(4): 949–953. 16. King Jr TE and Mortenson RL. Cryptogenic organizing pneumonitis. The North American experience. Chest 1992; 102(1 Suppl): 8S–13S. 17. Purcell IF, Bourke SJ and Marshall SM. Cyclophosphamide in severe steroid-resistant bronchiolitis obliterans organizing pneumonia. Respir Med 1997; 91(3): 175–177. 18. Ferstenfeld JE, Schlueter DP, Rytel MW and Molloy RP. Recognition and treatment of adult respiratory distress syndrome secondary to viral interstitial pneumonia. Am J Med 1975; 58(5): 709–718. 19. Gattinoni L, Bombino M, Pelosi P, Lissoni A, Pesenti A, Fumagalli R and Tagliabue M. Lung structure and function in different stages of severe adult respiratory distress syndrome. J Am Med Assoc 1994; 271: 1772–1779.

Treatment of Severe Acute Respiratory Syndrome JJY Sung and AK Wu

Background General approach Clinical outcome Antiviral agents Immuno-modulators

89 89 91 92 94

Background Severe acute respiratory syndrome (SARS) is a newly emerged disease and the epidemic in Hong Kong came as a crisis. The clinical course of SARS appears to follow a triphasic pattern [1,2]: phase I is clinically characterized by fever, myalgia and other systemic symptoms that generally improve after a few days. This is the phase when active viral replication occurs. Phase II is characterized by recurrence of fever, oxygen desaturation and radiological progression of pneumonia. The clinical progression during phase II appears to be related to immuno-pathological damage. The majority of patients recovered spontaneously but in some the disease progressed into phase III, characterized by acute respiratory distress syndrome (ARDS) necessitating ventilatory support (Figure 9.1). Reports show that with the development of respiratory failure and ARDS, 15–30% of patients will require intensive care admission [3]. Histological examination shows the presence of coronavirus particles in the alveoli of the infected lungs. Histopathology of post-mortem cases also reveal diffuse alveolar damage, pulmonary oedema, hyaline membrane formation and highly activated

Convalescent plasma Ventilatory support New treatment Conclusion

CHAPTER

9 95 96 96 97

macrophages with haemophagocytosis. Thus, the treatment modalities should include antivirals, immuno-modulators and respiratory support at the different stages of the diseases [3,4].

Key points Triphasic clinical pattern 1. Viral replication: fever, myalgia and other systemic symptoms that generally improve after a few days. 2. Immuno-pathological damage: recurrence of fever, oxygen desaturation and radiological progression of pneumonia. 3. Recovery (most patients) or progression to ARDS.

General approach The treatment protocol used in Hong Kong included the use of broad-spectrum antibiotics. Initial treatment usually consists of intravenous (IV) cefelosporin in combination with macrolides or quinolones.

90

T R E A T M E N T

Immune-hyperactive phase

Maximum daily body temperature (°C)

Viral-replicative phase

S A R S

Pulmonary-destruction phase

40

39

38

37

36 1

Fig. 9.1

O F

2

3

4

5

6

7

8 9 10 11 12 13 14 15 16 17 18 19 20 21 Days after onset of disease

A triphasic presentation of SARS.

A combination of ribavirin with or without ‘low-dose’ corticosteroid therapy is commenced when patients fail to respond to antibiotics treatment for 2 days. Pulses of high-dose methylprednisolone are given as a response to persistence or recurrence of fever and radiographic progression of lung opacity  hypoxaemia despite initial combination therapy. Further pulses of methylprednisolone can be given, if there is no clinical or radiological improvement. Patients who develop hypoxaemia are given supplemental oxygen therapy. Patients would be admitted to the intensive care unit (ICU) when severe respiratory failure develops as evidenced by: 1. failure to maintain an arterial oxygen saturation of at least 90%, while receiving supplemental oxygen of 50% and/or 2. respiratory rate greater than 35 breaths per minute. Non-invasive positive-pressure ventilation is used by some centres but avoided in the others because of the fear of viral transmission potentially resulting from mask leakage and flow compensation. Criteria for intubation and positive-pressure ventilation are, in general: 1. persistent failure to achieve arterial oxygen saturation of 90% while receiving 100% oxygen via a non-rebreathing mask and/or

2. onset of respiratory muscle fatigue as evidenced by an increase in the partial pressure of carbon dioxide (PaCO2), sweating, tachycardia and/or a subjective feeling of exhaustion. Mechanical ventilation with synchronized intermittent mandatory ventilation (SIMV) or pressure control ventilation are often instituted. Figure 9.2 summarizes the treatment protocol adopted at the Prince of Wales Hospital in Hong Kong.

Key points Treatment protocol 1. Broad-spectrum antibiotics. 2. If no response, then change to ribavirin with/without corticosteroids. 3. If there is persistence or recurrence of fever and radiographic progression of lung opacity  hypoxaemia, then give pulses of high-dose methylprednisolone. 4. If hypoxaemic, then give supplemental oxygen therapy.

C L I N I C A L

91

O U T C O M E

Fever, lung consolidation, exposure to SARS

IV cefotaxime, oral clarithromycin and oseltamivir Fever persisted and lung shadow increased

Admission to ICU for oxygen therapy and considered for PEEP ventilation

Hypoxaemia

No dyspnoea

Dyspnoea

Oral ribavirin, oral prednisolone

IV ribavirin, IV hydrocortisone

Hypoxaemia Fever persisted and lung shadow increased IV methylprednisolone, for three consecutive doses Hypoxaemia

Fever persisted and lung shadow increased IV methylprednisolone, for up to a maximum of six

Hypoxaemia

Continue ribavirin and corticosteroid for a total of 12 days or until lung shadows totally subsided

Fig. 9.2

Treatment protocol for SARS.

Key points ICU admission 1. Failure to maintain an arterial oxygen saturation of at least 90% while receiving supplemental oxygen of 50% and/or 2. Respiratory rate greater than 35 breaths per minute.

Key points Intubation criteria 1. Persistent failure to achieve arterial oxygen saturation of 90% while receiving 100% oxygen via a non-rebreathing mask and/or

2. Onset of respiratory muscle fatigue as evidenced by an increase in PaCO2, sweating, tachycardia and/or a subjective feeling of exhaustion.

Clinical outcome The clinical response to treatment can be objectively assessed by changes in body temperature, resolution of radiological lesions and oxygen requirement to maintain arterial oxygen saturation. At the Prince of Wales Hospital, sustained response to therapy is defined as: 1. defervescence (daily peak temperature 37.5°C) for at least 4 consecutive days,

92

T R E A T M E N T

2. radiological improvement, as assessed by three radiologists blinded to the clinical data, of more than 25% and 3. oxygen independence as assessed by pulse oximetry (oxygen saturation 95% on room air) on the 4th afebrile day. Patients with defervescence who achieved either resolution of lung consolidation or oxygen independence, but not both, are classified as showing a partial response. Patients who fall short of criteria 2 and 3 above are classified as non-responders to therapy.

• Before the sensitivity of SARS-associated coron-

•

Clinical response

Antiviral agents Genomic analysis identified two types of targets for antiviral therapy. The surface targets for cell entry and the enzymatic targets for viral replication, i.e. the RNA replicase and the protease (Figure 9.3).

S A R S

Ribavirin is an inhibitor of replicase. The choice of ribavirin in the treatment of SARS was based on the following reasons:

Key points 1. 37.5°C for at least 4 consecutive days 2. Radiological improvement 3. Oxygen saturation 95% on room air on the 4th afebrile day.

O F

•

avirus (SARS-CoV) was known, ribavirin was chosen because of its broad-spectrum antiviral activity for both RNA and DNA viruses (respiratory syncytial virus, influenza A and B, measles and parainfluenza as well as Lassa fever). In vitro study using plague reduction assay showed that ribavirin has a modest activity against SARS-CoV at the concentration of 50 g/mL [5]. Unfortunately, more recent study revealed that ribavirin has no significant in vitro activity against this novel coronavirus, believed to be responsible for SARS [6]. Besides a mild antiviral activity, ribavirin has been shown, in a coronavirus hepatitis murine model, to have a modest immuno-modulatory effect. Ribavirin has been shown to inhibit viral-induced macrophage production of pro-inflammatory cytokines and T-helper 2 cells (Th2) cytokines. As immunological reaction is believed to play a part in the pathogenesis of pulmonary injury, ribavirin may have some beneficial effect also in this aspect.

In fact, reviewing our data on ribavirin and low-dose steroid combination, the treatment has not produced any significant benefit in the treatment of SARS.

Lipid bilayer or membrane

Antifusion peptide

Protease (lopinavir) Internal core structure RNA Replicase (ribavirin)

Fig. 9.3 Possible targets for coronavirus are surface target for cell entry and enzymatic targets for cell replication.

Based on the results of our cohort of 138 patients, favourable response to ribavirin was found in a minority of patients. Ninety-four patients received oral ribavirin and prednisolone. Among them, there were 14 sustained responders and nine partial responders. These 23 patients were discharged uneventfully. Two patients died in the early phase of the disease before additional therapy could be given. Forty-four patients received IV ribavirin and hydrocortisone and, among them, only two had a sustained response whereas four patients died (Figure 9.4). This combination therapy failed to show any appreciable response in the remaining 107 patients (Table 9.1). With the current dose of ribavirin used, we observed the modest degree of anaemia in most patients (59% dropping haemoglobin (Hb) by 2 g/dL), probably the result of haemolysis. A much higher dose of ribavirin, based on the dosage for treatment of haemorrhagic fever

A N T I V I R A L

93

A G E N T S

viruses, has been reported to be associated with more significant toxicity. In a report from the Toronto group [7], haemolysis was reported in 76% and a decrease in Hb of 2 g/dL in 49%, elevated transaminases in 40% and bradycardia in 14% of SARS patients.

Based on these results, ribavirin cannot be recommended as a first-line therapy for coronavirus infection. The other antiviral therapy that has been put to test is lopinavir. Lopinavir is a protease inhibitor used in the SARS (n
138)

Antibiotic therapy (n
138)

SR
14 PR
9 NR
2

Oral ribavirin (n
94)

SR and PR all discharged, NR all died PR
52 (one died,31 discharged)

SR
0

SR
2 NR
4

IV ribavirin (n
44)

NR
107 IV methylprednisolone

three doses (n
107)

SR all discharged, NR all died SR
45 (all discharged)

NR
10, one died, nine received further methylprednisolone

n
20

IV methylprednisolone  three doses (n
29)

PR
13 (one died,12 discharged)

SR
5 (all discharged)

NR
11 (six died, one in ICU, one on medical ward and three discharged) Summary: 15 (10.7%) died, Summary 121 (87.7%) discharged home and 2 (1.4%) remained in hospital

Fig. 9.4 Clinical outcome of 138 patients with SARS. SR: sustained response; PR: partial response; NR: no response. Table 9.1

SR PR NR a

Clinical response to therapy. Broad-spectrum antimicrobiala (%), n
138

Ribavirin  corticosteroidb (%), n
138

IV methylprednisolonec (%), n
107

0 (0) 0 (0) 138 (100)

16 (11.6) 9 (6.5) 113 (81.9)

50 (46.7) 45 (42.1) 12 (11.2)

Antimicrobials included cefotaxime and clarithromycin (or levofloxacin) plus oseltamivir. Ribavirin (oral or IV) plus oral prednisolone or IV hydrocortisone. c IV methylprednisolone up to 3 g in total. Clinical outcome definitions: (1) afebrile (daily peak temperature 37.5°C) for at least 4 consecutive days; (2) resolution of chest radiograph consolidation by 25% (comparing film of maximal consolidation and that on the 4th afebrile day) and (3) oxygen independence (oxygen saturation 95% on room air) on the 4th afebrile day. Sustained response (SR): 1  2  3; partial response (PR): 1  2 or 3 and no response (NR): fail to fulfil the criteria of SR and PR. b

94

T R E A T M E N T

treatment of human immunodeficiency virus (HIV). Lopinavir is combined with ritonavir (as Kaletra™) to reduce its metabolism in the body. In vitro data suggest that lopinavir has a much-augmented activity against SARS-CoV. The 50% inhibitory activity of lopinavir is around 4 g/mL, around 10-fold higher than that of ribavirin [5]. In a pilot study of using Kaletra™ as initial treatment of SARS and compared to historic control of ribavirin-treated (age- and sexmatched) patients, the oxygen desaturation rate, requirement of intubation and mechanical ventilation as well as mortality of the former was significantly reduced. These results, however, are retrospective and uncontrolled. Interpretation must be taken with caution.

Key points Ribavirin

• Inhibitor of replicase • Antiviral activity for both RNA and DNA viruses

• ? in vitro activity against SARS-CoV • Modest immuno-modulatory effects in coronavirus hepatitis murine model

• Cohort of 138 patients showed favourable response in the minority of patients Ribavirin side effects

• • • •

Haemolysis (76%) Decrease in Hb of 2 g/dL (49%) Elevated transaminases (40%) Bradycardia (14%)

Ribavirin cannot be recommended as a firstline therapy for coronavirus infection.

Immuno-modulators Previous studies have shown that in acute viral respiratory infections, large amounts of early-response cytokines, such as interferon alpha (IFN ), tumor necrosis factor alpha (TNF ), interleukin (IL)-1 and IL-6 are produced. These cytokines mediate antiviral activities but at the same time may contribute to tissue injury. The finding of activated macrophage in the lung, haemophagocytosis and overproduction

O F

S A R S

of cytokines in patients with SARS have prompted the idea of using immuno-modulators to suppress over-reaction of the body immune system. The most commonly used immuno-modulators are corticosteroids. In our cohort of 138 cases at the Prince of Wales Hospital, IV pulse therapy with high-dose methylprednisolone was given to 107 patients who did not respond to ribavirin and ‘low-dose’ corticosteroid therapy. After three infusions of 0.5 g methylprednisolone, 45 patients (42.1%) showed a sustained response and recovered from the disease. Fifty-two patients (48.6%) demonstrated a partial response to the therapy. Among those with a partial response, 31 recovered and were discharged from hospital, one died, whereas 20 required further pulses of high-dose methylprednisolone. There were 10 non-responders, and among them one died. Among the partial responders and non-responders, 29 received further doses of IV methylprednisolone for up to 3 g in total. Sustained response was reported in five and partial response in 13. Eleven patients (median age 55 years, range 33–82 years) failed to show any response to more than three pulses of high-dose methylprednisolone. Among them, six patients died, one remained in the ICU, one remained on medical ward, while three were discharged home (Figure 9.4). The overall success rate of high-dose methylprednisolone therapy was 88.8% (Table 9.1). The side effects of high-dose corticosteroids are well known. In this cohort, hyperglycaemia (plasma spot glucose 11.0 mmol/L) was detected in 21.5% of patients and hypokalaemia in 15%. These metabolic derangements were easily corrected when IV high-dose methylprednisolone was discontinued. Two patients developed transient confusion, delusion and anxiety which subsided after discontinuation of steroid. The risk of nosocomial infection is reckoned with the use of high-dose steroid. In our series, however, secondary bacterial or fungal infection was reported in 11 (10.2%) of patients. Following high-dose methylprednisolone therapy, rapid resolution of lung opacity is usually followed by improvement of hypoxaemia. Most patients responded after receiving three doses of high-dose methylprednisolone (up to 1.5 g in total). Less than 30% of cases required additional doses. The timing of administration of high-dose methylprednisolone is important.

C O N V A L E S C E N T

It should be administered only during phase II when radiological progression of consolidation and increasing hypoxaemia were documented. In most cases, high-dose methylprednisolone was given at the end of the 1st week. We have avoided high-dose methylprednisolone in the early phase of SARS, as viral clearance by host immunity might be hampered. It must be emphasized that high-dose methylprednisolone should not be used only to control fever. In some of our patients, the lung opacities continued to deteriorate even after defervescence. In these patients, the benefit of high-dose methylprednisolone in reversing radiological progression is also seen. While we recognize that the benefit of high-dose methylprednisolone cannot be confirmed without a control group, the use of high-dose corticosteroid in the treatment of SARS warrants further investigation. Other immuno-modulating agents that have been used included IV immunoglobulin (IVIG), pentaglobulin, azathioprine and anti-TNF in small number of patients. The numbers of cases were small and as experience was anecdotal, it is difficult to confirm the efficacy of these treatments. In vitro tests have also indicated that IFN has antiviral activity against SARS-CoV. IFN has been used in the treatment of viral infections. However, to date, there is no clinical data on its use in the treatment of SARS. There are concerns that IFN might aggravate the injurious effects of cytokines.

Key points IV pulse therapy with high-dose methylprednisolone

• Given to patients who did not respond • •

95

P L A S M A

to ribavirin and ‘low-dose’ corticosteroid therapy Overall success rate of 88.8% Administered only during phase II, when there is radiological progression and increasing hypoxaemia

Side effects of high-dose methylprednisolone

• Hyperglycaemia (21.5%) • Hypokalaemia (15%) • Transient confusion, delusion and anxiety

Convalescent plasma The Prince of Wales Hospital was the first to use convalescent plasma for the treatment of SARS. Convalescent plasma was obtained from patients who recovered from the illness. These patients

• • • •

were afebrile for at least 7 consecutive days, had radiographic improvement by at least 25%, no further need of oxygen supplement, passed 14 days since onset of symptoms.

All donors had to screen negative for hepatitis B, C, HIV and veneral disease research laboratory slide test (VDRL), and had to be confirmed to be seropositive for SARS-CoV. Apharesis was performed using a cell separator. Blood volume that was processed ranged from 2000 to 2500 mL. An average of 600–900 mL of serum was harvested per patient. Normal saline was used for replacement of fluid volume. Calcium gluconate (10% solution, 10 mL/1000 mL serum extracted) was given to the donor as replacement. At the Prince of Wales Hospital cohort, 40 patients had progressive disease after three doses (500 mg each) of pulsed methylprednisolone. Nineteen patients received convalescent plasma after the three doses of pulsed methylprednisolone, two of whom received further pulsed methylprednisolone after plasma infusion. They were compared to 21 patients who received only pulsed methylprednisolone. Seventy-four per cent of the patients who received convalescent plasma were discharged by day 22 as compared with 19% in the group that received steroid alone (P
0.001). There were no differences between age, sex and admission lactate dehydrogenase (LDH) between the convalescent plasma group and steroid group (Table 9.2). There were five deaths in this cohort study, all occurring in patients receiving steroids only, as compared with no death in the serum group (P
0.049). Hospital stay was significantly longer in those who received steroid alone. Our preliminary results with convalescent plasma indicate that it might be beneficial in ‘neutralizing’ the virus in the infected host. Yet, to achieve the maximum benefit, convalescent plasma should be given early. This promising result of convalescent plasma also prompts the development of hyperimmune globulin (monoclonal antibody) as a therapeutic agent in the future.

96

T R E A T M E N T

O F

S A R S

Table 9.2 Comparison of treatment outcome between those who received convalescent plasma (after failed response to corticosteroid) and those who received corticosteroid alone.

Number of patients Age LDH (IU/L) on admission Patients discharge by day 22 Patients discharged by day 22 after adjustment of co-morbidities Mortality rate

Convalescent plasma

Corticosteroid

P

19 38.7 256.1 73.4% (n
14) 77.8% (14/18)

21 47.9 247.7 19% (n
4) 23% (3/13)

0.087 0.7 0.001 0.004

0%

23.8% (n
5)

0.049

Key points Convalescent plasma

• Should be given early in course of disease • Results in earlier hospital discharge • Less deaths

Ventilatory support Patients who developed hypoxaemia were given supplemental oxygen therapy. Oxygen was delivered by nasal catheters or in combination with oxygen mask. A surgical mask was applied, if the patient was using nasal catheter alone. Use of high-flow Venturi-type masks should be avoided to avoid dissemination of droplets if patient cough. Nebulization should be avoided for the same reason. Patients were admitted to the ICU when severe respiratory failure developed as evidenced by: 1. failure to maintain an arterial oxygen saturation of at least 90% while receiving supplemental oxygen of 50% and/or 2. respiratory rate greater than 35 breaths per minute. Criteria for intubation and positive-pressure ventilation were: 1. persistent failure to achieve arterial oxygen saturation of 90% while receiving 100% oxygen via a non-rebreathing mask and/or 2. onset of respiratory muscle fatigue as evidenced by an increase in PaCO2, sweating, tachycardia and/or a subjective feeling of exhaustion.

Mechanical ventilation with SIMV, or pressure control ventilation, was instituted. Positive end-expiratory pressure (PEEP) and inspired oxygen concentration was titrated to achieve an arterial saturation of 90–95%. Tidal volume should be maintained at 6–8 mL/kg estimated body weight and plateau pressure maintained at 30 cmH2O or less. PaCO2 is allowed to rise provided the pH was greater than 7.15. Patients unable to meet the above parameters can be ventilated in the prone position. Non-invasive positive-pressure ventilation was avoided because of the risk of viral transmission potentially resulting from mask leakage and flow compensation, possibly causing wide dispersion of contaminated aerosol. Yet, experience from China has alluded that if low pressure ventilation was used in a room with good ventilation, dissemination of droplet and crossinfection would not be a major problem.

Key points Supplemental oxygen

• Nasal catheters or in combination with oxy• •

gen mask Surgical mask applied, if using nasal catheter alone High-flow Venturi-type masks or nebulization should be avoided

New treatment Recently, there has been interest in the use of herbal medicine against SARS. Glycyrrhizin, an active component of liquorice roots, for instance, has been

97

R E F E R E N C E S

recently shown to be active in vitro against 2SARS-CoV [8]. Other agents that have been tried in SARS patients include immuno-modulators such as IVIG and pentaglobin (IgM-enriched IGs). It has been postulated that these compounds may act via different mechanisms in the modulation of the systemic sepsis response, including neutralizing endotoxins and exotoxins, and scavenging active complement components and lipopolysaccharides. These compounds have been used in SARS patients who have failed conventional therapy (e.g. IVIG 0.4 g/kg for 5 days, or pentaglobin 300 mL IV over 12 hours for 3 days). Their efficacy and safety, as well as other novel treatment strategies in SARS patients, remain to be determined; and no formal recommendations could be given for their use at this stage.

Key points New treatment

• Antiviral and immuno-modulatory agents • Lack of evidence excludes recommenda-

The use of specific antiviral and immuno-modulatory therapies directed against the SARS-CoV such as ribavirin and corticosteroids, remain experimental and controversial at this stage. Randomized controlled studies will be required to evaluate the efficacy and best timing for high-dose methylprednisolone therapy.

References 1. 2.

3. 4. 5. 6. 7.

tion at this stage 8.

Conclusion At present, the most efficacious treatment regime for SARS is still not known. There is no formal treatment recommended except for meticulous supportive care.

Sung JJY. Severe acute respiratory syndrome: What do we know about this disease? Hong Kong Med Diary 2003; 8: 15–16. Peiris JS, Chu CM, Cheng VC et al. Clinical progression and viral load in a community outbreak of coronavirus-associated SARS pneumonia: a prospective study. Lancet 2003; 361(9371): 1767–1772. Lee N, Hui D, Wu A et al. A major outbreak of severe acute respiratory syndrome in Hong Kong. New Engl J Med 2003; 348(20): 1986–1994. Epub 7 April 2003. Nicholls JM, Poon LL, Lee KC et al. Lung pathology of fatal severe acute respiratory syndrome. Lancet 2003; 361: 1773–1778. Personal communication with Prof. KY Yuen. Cyranoski D. Critics slam treatment for SARS as infective and perhaps dangerous. Nature 2003; 423: 4. Booth CM, Matukas LM, Tomlinson GA et al. Clinical features and short-term outcomes of 144 patients with SARS in the greater Toronto area. J Am Med Assoc 2003; 289(21): 2801–2809. Epub 6 May 2003. Cinatl J, Morgenstern B, Bauer G et al. Glycyrrhizin, an active component of liquorice roots, and replication of SARSassociated coronavirus. Lancet 2003; 361: 2045–2046.

SARS in the Intensive Care Unit

CHAPTER

10

GM Joynt, GE Antonio and CD Gomersall

Introduction ICU admission ICU management and progress

99 99 101

Introduction Severe acute respiratory syndrome (SARS) is clinically severe with a high proportion of cases, approximately 20%, requiring intensive care unit (ICU) admission [1]. The provision of organ support in the ICU therefore plays a potentially important role in reducing mortality, which may be as high as 10% for younger patients and 50% for patients older than 60 years [2]. Radiological imaging of the chest is important because of the overriding importance of respiratory failure in determining the management and outcome of SARS. At the time of writing there were little published data detailing the ICU management and outcome of SARS, and much the information that follow are based on the observational data derived from our institution.

ICU outcome Infection control Conclusion

103 105 106

(LDH) concentration, lymphopaenia, hypocalcaemia and moderate thrombocytopaenia [1,3]. SARS is a slowly progressive disease and the average interval from the onset of symptoms to requirement for ICU admission is approximately 10 days. Clinical deterioration of cases admitted to the ward is manifested by progressive hypoxia and dyspnoea, and is accompanied by progression of pulmonary infiltrates on chest radiograph [1]. Close monitoring of disease progress in the general wards is therefore important to detect deterioration in those patients who will be admitted to ICU. As clinical deterioration appears to be closely mirrored by the development of progressively worsening radiographical opacity, chest radiographs may serve to be a useful objective predictor of disease progression.

Key points

• Severe SARS patients usually develop

ICU admission Patients generally present to the hospital with fever, chills, rigors, myalgia, headache and a nonproductive cough. Common laboratory features include an elevated serum lactate dehydrogenase

progressive dyspnoea and hypoxia over about 10 days prior to ICU admission. Admission to ICU is invariably a consequence of progressive, severe respiratory failure unresponsive to administration of moderate concentrations of

100

S A R S

Fig. 10.1 Chest radiograph of a 46-year-old male on the day of admission to ICU, 9 days after the onset of symptoms. The patient required an inspired oxygen concentration of 100% to maintain arterial oxygen saturation at 95%. There are multiple areas of opacification in both lungs, worse on the right, some of these are of ground-glass density and others are consolidative. There is no pleural effusion or lymphadenopathy.

inspired oxygen. In general, patients were admitted following: 1. failure to maintain an arterial oxygen saturation of at least 90% while receiving supplemental oxygen of 50% and/or 2. respiratory rate greater than 35 breaths per minute. Although chest radiograph features of bilateral, diffuse consolidation and/or ground-glass opacification were present in all but one patient admitted to the ICU, radiograph appearance by itself was not sufficient to warrant ICU admission in the absence of the patient meeting the above pathophysiological criteria. Typical radiological features seen on admission are areas of consolidation with ill-defined borders, mostly affecting the mid- and lower zones. These tend to be bilateral in distribution. There is no evidence of hilar or mediastinal lymphadenopathy, cardiomegaly (unless pre-existing), pleural effusion,

I N

T H E

I N T E N S I V E

C A R E

U N I T

Fig. 10.2 Chest radiograph of a 47-year-old male on the day of admission to ICU, 9 days after the onset of symptoms. The patient required an inspired oxygen concentration of 100% to maintain arterial oxygen saturation at 90%. There is a small pneumothorax present on the right side, despite the absence of established risk factors such as previous mechanical ventilation or central venous catheter insertion. There are also multiple areas of opacification. There is no pleural effusion or lymphadenopathy.

cavitation or calcification. The radiographical appearances are non-specific on their own and simulate severe cases of other types of pneumonia, especially bronchopneumonia (Figure 10.1). Surprisingly, some patients were noted to develop spontaneous pneumothorax and/or pneumo-mediastinum prior to mechanical ventilation, some even prior to ICU admission (Figure 10.2). The reasons for this observation are unclear.

Key points

• Admission to ICU is precipitated by respira•

tory failure, which is usually the only organ failure present. Bilateral diffuse infiltrates are typically seen on chest radiographs.

I C U

M A N A G E M E N T

A N D

101

P R O G R E S S

The demographical profile of patients admitted to ICU is as follows. The average age of ICU admissions with SARS is approximately 50 years (range 23–81 years), with an approximately equal number of males and females admitted. All patients admitted to our ICU had severe respiratory failure and most meet the basic criteria for acute respiratory distress syndrome (ARDS). A typical admission Acute Physiology And Chronic Health Evaluation (APACHE) II score is a moderate 10–12, indicating a relative lack of associated deranged physiology.

Key points

sepsis in the first 3–5 days following high-dose methylprednisolone therapy. Suspected infection is treated early with empirical broad-spectrum antibiotics and if necessary, antifungal agents. The antibiotic regimen is modified according to the results of bacterial culture of sputum, tracheal aspirate and blood.

Key points

• High-dose pulse methylprednisolone is administered to try to suppress the immuno-inflammatory process that apparently affects the lungs in the later, severe stages of the disease.

ICU patient demographics

• Average age of 50 years • Approximately equal number of males and females

• All patients admitted had severe respiratory failure

• Most meet the basic criteria ARDS • Typical admission APACHE II score is 10–12

ICU management and progress Medical therapy for SARS is evolving and is an extension of the protocol utilized in the general ward [1]. Currently this includes the use of broadspectrum antibiotics (to treat the common causes of atypical pneumonia), ribavirin and low-dose corticosteroids. In addition, methylprednisolone 500 mg to 1 g daily for 2–3 days is used in an attempt to dampen the inflammatory response in those patients who continue to deteriorate. Deterioration is generally manifested by worsening hypoxia, respiratory distress and radiological evidence of pulmonary deterioration. High-dose methylprednisolone may be repeated to a total dosage of up to 5 g in the most severe cases. Some of these cases may also receive convalescent plasma donated by patients who had recovered from SARS and/or IgM-enriched immunoglobulin. Broadspectrum antibiotics to cover typical nosocomial organisms are administered at the time of institution of high-dose pulse methylprednisolone therapy, but are withdrawn in the absence of obvious clinical

Supportive management in the ICU focuses on oxygen supplementation and, when absolutely necessary, mechanical ventilation. Oxygen supplementation is provided by the use of nasal cannulae, and where necessary Hudson-type masks. The use of entrainment or Venturi-type masks is avoided as the high gas-flows generated might encourage the dispersal of contaminated droplets during coughing or sneezing. Non-invasive positive-pressure ventilation is also avoided because of the risk of viral transmission potentially resulting from mask leakage and high gas-flow compensation, possibly causing wide dispersion of contaminated aerosol.

Key points

• Avoid devices that produce high gas-flows Although radiological changes mirror clinical deterioration in the majority of cases, as with the criteria for ICU admission, the decision to initiate intubation and mechanical ventilation is primarily a clinical one. Criteria for intubation and positive-pressure ventilation are: 1. persistent failure to achieve arterial oxygen saturation of 90% while receiving 100% oxygen via a non-rebreathing mask and/or 2. onset of respiratory muscle fatigue as evidenced by an increase in arterial carbon dioxide tension (PaCO2), sweating, tachycardia and/or a subjective feeling of exhaustion.

102

S A R S

While these indications lead to intubation, relatively later than might be usually expected, mechanical ventilation is required in 50–60% of patients admitted to the ICU.

Key points Criteria for intubation and positive-pressure ventilation 1. Persistent failure to achieve arterial oxygen saturation of 90% while receiving 100% oxygen via a non-rebreathing mask and/or 2. Onset of respiratory muscle fatigue Usually, mechanical ventilation with synchronized intermittent mandatory ventilation (SIMV), or pressure control ventilation is instituted. Positive endexpiratory pressure (PEEP) and inspired oxygen concentration are titrated to achieve an arterial saturation of 90–95%. Tidal volume is maintained at 6–8 mL/kg estimated body weight and plateau pressure maintained at 30 cmH2O or less. PaCO2 was allowed to rise provided the pH was greater than 7.15 [4]. A small number of patients unable to meet the above parameters were ventilated in the prone position.

I N

T H E

I N T E N S I V E

C A R E

U N I T

The vast majority of patients admitted with SARS meet the criteria for ARDS during the ICU stay [5]. The plain radiograph features of established SARS in the ICU are indistinguishable from those of ARDS (Figure 10.3). The typical features of ARDS include bilateral widespread, confluent opacification, with the lung periphery being denser or more extensively involved than the perihilar regions. A ‘bat’s wing’ appearance, where the perihilar regions are more densely opacified, is not a feature of ARDS in SARS but more commonly seen in cardiogenic (left heart failure) pulmonary oedema or that due to renal disease. There is no cardiomegaly, upper lobe pulmonary venous dilatation, peribronchial cuffing, septal lines or pleural effusion. The findings are nonspecific for SARS but resemble ARDS from most other non-cardiac causes. It is possible that the

Key points

• Controlled mechanical ventilation with low tidal volumes and pressures to protect the patient from barotrauma is required in the most severe, progressive cases. Fluid intake and losses are strictly controlled to maintain an intake/output balance of approximately nil. Vasopressors at small to moderate doses are used to maintain adequate blood pressure in preference to the use of bolus fluid infusion. Patients are otherwise managed according to standard ICU organ support protocols.

Key points

• Excess fluid administration is avoided to prevent pulmonary venous hypertension and potential fluid leakage into the lung.

Fig. 10.3 Chest radiograph of a 44-year-old male on day 9 after admission to ICU, 14 days after the onset of symptoms. He was receiving mechanical ventilation by pressure control mode, with an inspired oxygen concentration of 60%, a peak pressure of 34 cmH2O, a tidal volume of 360 mL and PEEP of 15 cmH2O. There are bilateral almost symmetrical areas of consolidative opacification. This gives an appearance similar to ARDS. A right jugular central venous catheter is present.

I C U

103

O U T C O M E

pathogenesis of SARS-related ARDS results in specific morphological changes that might be demonstrated on computed tomography (CT) scan, as typical changes have been described in acute lung injury (ALI)/ARDS from different aetiologies [6]. There are also known to be changes in CT findings over time, particularly related to the duration of mechanical ventilation and the natural evolution of ARDS [7]. Unfortunately, no CT images were available for patients in the early acute stage of ARDS, however post-mortem histology obtained from patients who died in the early stages of ARDS demonstrated changes consistent with the early and organizing phase of diffuse alveolar damage. The early phase is characterized by pulmonary oedema with hyaline membrane formation suggestive of the acute stage of ARDS and cellular fibromyxoid organizing exudates in the airspaces indicates an organizing phase that follows alveolar damage [1]. Multinucleated pneumocytes are common. SARS is also associated with epithelial-cell proliferation and an increase in macrophages in the alveoli and the interstitium of the lung [8].

excess pressure and volume during mechanical ventilation, the incidence of barotrauma appears to be high. So far the pneumothorax rate in ventilated patients is approximately 20%. This is substantially higher than that reported previously in ventilated patients with ARDS [9,10]. At present we have no explanation for this observation, but the relatively high rate of pneumothorax in mechanically ventilated cases, coupled with the occurrence of barotrauma in non-ventilated cases suggests that care needs to be taken to avoid circumstances that might exacerbate the risk of barotrauma. Avoiding mechanical ventilation as much as possible and if required, utilizing low-volume, low-pressure ventilation would seem prudent. Routine daily chest radiographs are recommended to assist early detection of barotrauma and avoid progression to complications such as tension pneumothorax, particularly in mechanically ventilated patients.

Key points

• Barotrauma appears to be frequent, despite attempts to reduce the incidence.

Key points X-ray features

• Indistinguishable from those of ARDS • Bilateral widespread, confluent opacifica-

• •

tion, with the lung periphery being denser or more extensively involved than the perihilar regions The ‘bat’s wing’ appearance is not a feature No cardiomegaly, upper lobe pulmonary venous dilatation, peribronchial cuffing, septal lines or pleural effusion

Key points SARS with ARDS

• Clinical features, chest radiograph, histological and CT findings are similar to those seen in ARDS from other causes. Despite relatively late intubation and mechanical ventilation and the close attention to limitation of

ICU outcome It is not yet clear what the mortality rate of patients admitted to the ICU will be, but based on the outcomes achieved in our cohort so far is expected to be about 30–35%. The average length of stay in the ICU is about 10 days. A remarkable feature of ICU patients with SARS is the apparent limitation of organ failure to the respiratory system. The median maximal multiple organ dysfunction (MOD) score during the ICU stay in our patients is a moderate to low score of five, with the majority of the score being made up of the respiratory component, again indicating the relative lack of organ failure outside the respiratory system. Nosocomial infection rates appear unusually high. Common organisms include Staphylococcus aureus, Stenotrophomonas maltophilia and Candida albicans. Common sites of infection were the lungs and urinary tract. The high incidence of nosocomial infection may be caused by the use of high-dose steroid therapy, or may be a consequence of the immunosuppressive effects of the disease itself. The ultimate cause of death is usually the result of oxygenation

104

S A R S

Fig. 10.4 High-resolution CT (HRCT) of a 33-year-old man 38 days after ICU admission and 46 days after the onset of symptoms. This patient had received high concentrations of oxygen and mechanical ventilation in the acute phase. Of special note is a 2-cm thick-walled cyst in the middle lobe. Multiple smaller, subpleural cysts are present in the right lower lobe anteriorly with thickened interlobular septae. There are architectural distortion, volume loss, bronchiectasis, parenchymal bands and irregularly thickened interlobular septae in the left lower lobe.

failure, organ failure as a consequence of nosocomial sepsis or complications of pre-existing comorbid disease.

I N

T H E

I N T E N S I V E

C A R E

U N I T

(a)

Fig. 10.5a HRCT of a 73-year-old man taken 29 days after the onset of symptoms and 23 days after ICU admission and mechanical ventilation. Mechanical ventilation was predominantly achieved by pressure control mode, with an inspired oxygen concentration of 60–100%, an average peak pressure of 28 cmH2O, an average tidal volume of 6.4 mL/kg estimated lean body mass and a PEEP range of 8–15 cmH2O. There is consolidation in the dependent regions of both lungs. The non-dependent regions show mainly ground-glass opacification and thickened interlobular septae. This distribution is typical for ARDS. There is a tiny right-sided pneumothorax.

Key points

• The mortality of patients admitted to ICU •

is about 30%. Nosocomial infection, particularly pneumonia, is common.

Those patients who require mechanical ventilation remain in the ICU much longer than those who do not, and the average duration of ventilation is currently about 2 weeks. A large number of mechanically ventilated patients therefore progress beyond the acute stage and into the chronic stages of ARDS [7]. In general, the CT features of late-stage ARDS caused by SARS are similar to those seen in latestage ARDS from other causes [11,12]. CT scans of a number of patients with late-stage ARDS were performed (Figures 10.4, 10.5a and 10.5b). Patients had bilateral segmental or sub-segmental areas of ground-glass opacification, which involved most of

the lung segments. Consolidative changes were generally of small volume and only a minority of the lesions were segmental. Segmental lesions tended to be in dependent regions. Septal thickening was evident. Irregular interfaces, parenchymal bands and traction bronchiectasis indicated fibrosis in these patients. Patients with fibrosis also demonstrated irregular septal thickening. Small, thick-walled pulmonary cysts (less than 1 cm in diameter), were commonly seen. Larger cysts, also thick walled and somewhat distorted in shape, were less commonly seen. Both types of cysts were found in the dependent and non-dependent segments and in areas of architectural distortion suggesting fibrosis. Interestingly, the duration of mechanical ventilation does not appear to have a major impact on the disease progression as measured by CT findings. Severe CT changes of late-stage ARDS may be present even in patients not mechanically ventilated (Figure 10.6).

I N F E C T I O N

105

C O N T R O L

At this stage the outcome of patients with late-stage ARDS from SARS is unknown.

Key points CT features of late-stage ARDS caused by SARS

• Similar to those seen in late-stage ARDS • • (b)

Fig. 10.5b HRCT showing features from the same patient as in Figure 10.6. Small emphysematous bullae are present in the medial aspect of the right upper lobe. The amount of inflation of the secondary lobules is variable giving rise to the cyst-like (hyperinflated) lobules mixed among lobules with groundglass opacification.

• • •

from other causes Ground-glass opacification: bilateral segmental or sub-segmental, involving most of the lung segments Consolidative changes: usual subsegmental, larger in dependent regions Septal thickening Irregular interfaces, parenchymal bands and traction bronchiectasis Cysts with thick walls

Infection control SARS is readily transmissible and high viral RNA concentrations have been detected in respiratory secretions and faeces [13]. Spread probably occurs most frequently via droplets and aerosols, which may be enhanced by the use of nebulizers or similar devices [14]. The virus is stable on surfaces for days after shedding and so contact with infected surfaces could also be a possible source of contamination. In ICUs, where patients with SARS may be clustered together, the concentration of virus in the environment can be expected to be particularly high.

Fig. 10.6 HRCT of the patient whose early chest radiograph is shown in Figure 10.1, performed 35 days after the onset of symptoms of SARS and in the late stage of ARDS. This patient received high concentrations of inspired oxygen (inspired oxygen concentration 80%) for more than 1 week, and oxygen supplementation for more than 1 month, but was never mechanically ventilated. There is bilateral widespread ground-glass opacification with thickened interlobular septae. There is clearing of the opacification in the subpleural 5 mm of the lung. A loculated hydro-pneumothorax is present in the left major fissure with a chest drain in situ.

In high-risk environments like the ICU, strict adherence to infection control procedure is critical to prevent transmission. Details of infection control procedures used in the ICU can be obtained online [15], but some of the most important issues are summarized here. The ICU should only be accessible to staff directly involved in patient care to prevent unnecessary exposure. Visitors should only be allowed under exceptional circumstances. Personal protective equipment such as N95 mask respirators, caps, goggles or full face-shields, disposable gowns and gloves should be readily available to all staff (including visiting staff such as radiographers). All staff should undergo proper training and close supervision

106

S A R S

during preparation and when leaving the ICU in designated ‘gown-up’ and ‘gown-down’ areas [16]. Initial, formal ‘fit-testing’ with a commercially available kit should be performed to ensure adequate mask size and fit for each individual expected to enter the ICU environment.

Key points

• ICU should only be accessible to staff •

•

directly involved in patient care. Personal protective equipment such as N95 mask respirators, caps, goggles or full face-shields, disposable gowns and gloves should be readily available. All staff should undergo proper training in putting on protective gear.

Infection control behaviour in the ICU must be monitored and repeatedly enforced. Regular hand cleansing and glove changing between patient contacts are essential. This aspect of infection control is important for radiographers and assisting technicians from the radiology department to remember as they are frequently required to move from one patient to another while completing daily routine radiograph exposures. When exiting the ICU the ‘gown-down’ area should also be regulated and monitored. Inanimate objects must either be placed in a protective covering or be properly cleansed when leaving highrisk area. All clinical areas and equipment (ultrasound or radiograph units) should be disinfected regularly and thoroughly with chlorine or hypochlorite solutions, especially prior to leaving a high-risk area [17]. Extra care needs to be taken if it is necessary to come into contact with patients receiving high flow oxygen or nebulization. Nebulization, oxygen delivery by Venturi masks and non-invasive positive-pressure ventilation, is avoided; if possible, to minimize dissemination of contaminated aerosols by these high flow and pressure-generating devices. Oxygen should be delivered by nasal catheters, Hudson mask or non-rebreathing mask, when possible. Infection control precautions must be maintained during transport of patients outside the ICU, for example in the radiology suite during specialized imaging procedures such as CT scan. For example,

I N

T H E

I N T E N S I V E

C A R E

U N I T

when transporting mechanically ventilated patients, a high efficiency heat and moisture exchanger filter at the circuit Y-piece and a viral filter at the ventilator expiratory port should be incorporated in the breathing circuit to minimize viral contamination of the environment [16]. Of course, all ICU transport and receiving radiology suite staff should be attired in full personal protective clothing and equipment.

Key points

• Infection control is essential to protect staff and other patients in high-risk areas. Education of staff, availability of personal protective equipment and strict enforcement of infection control protocols are key goals.

Conclusion SARS is a serious infection that causes predominantly severe respiratory failure, with little other organ failure. Admission rates to ICU are high and the morbidity and mortality are significant. There appears to be a high incidence of barotrauma, particularly among those patients requiring mechanical ventilation. Nosocomial infection rates appear high, and may be disease or therapy related. Radiological imaging forms an important part of monitoring this condition and may have the ability to prevent complications and be predictive of clinical progress. Formal and detailed radiological imaging of the respiratory system in particular may increase our understanding of the pathological nature of this new condition.

References 1. Lee N, Hui D, Wu A, Chan P, Cameron P, Joynt GM et al. A major outbreak of severe acute respiratory syndrome in Hong Kong. New Engl J Med 2003; 348: 1986–1994. 2. Donnelly CA, Ghani AC, Leung GM, Hedley AJ, Fraser C, Riley S et al. Epidemiological determinants of spread of causal agent of severe acute respiratory syndrome. Lancet 2003; 361: 1761–1766. 3. Booth CM, Matukas LM, Tomlinson GA, Rachlis AR, Rose DB, Dwosh HA et al. Clinical features and short-term outcomes of

R E F E R E N C E S

4.

5.

6.

7.

8. 9.

144 patients with SARS in the greater Toronto area. J Am Med Assoc 2003; 289: 2801–2809. The Acute Respiratory Distress Syndrome Network. Ventilation with lower tidal volume as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome. New Engl J Med 2000; 342: 1301–1308. Bernard GR, Artigas A, Brigham KL, Carlet J, Falke K, Hudson L et al. The American–European Consensus Conference on ARDS. Definitions, mechanisms, relevant outcomes, and clinical trial coordination. Am J Respir Crit Care Med 1994; 149(3 Pt 1): 818–824. Goodman LR, Fumagalli R, Tagliabue P, Tagliabue M, Ferrario M, Gattinoni L et al. Adult respiratory distress syndrome due to pulmonary and extrapulmonary causes: CT, clinical, and functional correlations. Radiology 1999; 213: 545–552. Gattinoni L, Bombino M, Pelosi P, Lissoni A, Pesenti A, Fumagalli R and Tagliabue M. Lung structure and function in different stages of severe adult respiratory distress syndrome. J Am Med Assoc 1994; 271: 1772–1779. Nicholls JM, Poon LL, Lee KC, Ng WF, Lai ST, Leung CY et al. Lung pathology of fatal severe acute respiratory syndrome. Lancet 2003; 361: 1773–1778. Weg JG, Anzueto A, Balk RA, Wiedemann HP, Pattishall EN, Schork MA et al. The relations of pneumothorax and other air leaks to mortality in the acute respiratory distress syndrome. New Engl J Med 1998; 338: 341–346.

107

10. Stewart TE, Meade MO, Cook DJ, Granton JT, Hodder RV, Lapinsky SE et al. Evaluation of a ventilation strategy to prevent barotrauma in patients at high risk for acute respiratory distress syndrome. New Engl J Med 1998; 338: 355–361. 11. Gattinoni L, Caironi P, Pelosi P and Goodman LR. What has computed tomography taught us about the acute respiratory distress syndrome? Am J Respir Crit Care Med 2001; 164: 1701–1711. 12. Rouby J, Puybasset L and Nieszkowska A. Acute respiratory distress syndrome: lessons from computed tomography of the whole lung. Crit Care Med 2003; 31(4 suppl): S285–S295. 13. Drosten C, Gunther S, Preiser W, van der WS, Brodt HR, Becker S et al. Identification of a novel coronavirus in patients with severe acute respiratory syndrome. New Engl J Med 2003; 348: 1967–1976. 14. Tomlinson B and Cockram C. SARS: experience at Prince of Wales Hospital, Hong Kong. Lancet 2003; 361: 1486–1487. 15. Joynt GM and Gomersall CD. Severe acute respiratory syndrome (SARS). http://www.aic.cuhk.edu.hk/web8/sudden_ acute_respiratory_syndrom.htm (accessed 13 January 2004). 16. Li TS, Buckley TA, Yap FH, Sung JJ and Joynt GM. Severe acute respiratory syndrome (SARS): infection control. Lancet 2003; 361: 1386. 17. Hospital infection control guidance for severe acute respiratory syndrome (SARS) http://www.who.int/csr/sars/ infectioncontrol/en (accessed 11 April 2003).

Imaging of Pneumonia in Children WCW Chu

Introduction Chest X-ray and common infective agents CT and common infective agents Evaluation of persistent or recurrent pneumonia

109 109 112

CHAPTER

11

Complications of pneumonia and role of CT Chronic sequelae and role of CT Conclusion

115 118 120

113

Introduction

The other indications of imaging, including CT, in children presenting with chest infections are:

Pneumonia is one of the most common serious infections of childhood. Plain chest radiographs remain the diagnostic mainstay in childhood pneumonia and chest computed tomography (CT) is rarely required in immunocompetent children presenting with symptoms and signs typical of chest infection. The appearance of different kinds of the childhood pneumonia is well established and fully described in the early literature of 1970s and 1980s. Since then, there is no significant further update on the radiographic aspects of childhood pneumonia. The relatively old references are therefore omitted in this chapter.

• to predict or suggest the nature of the infectious

The major indication of chest radiograph in children presenting with chest infection is to confirm or exclude the presence of pneumonia. A follow-up chest X-ray is not routine in the management of children who have uneventful recovery, as postobstruction pneumonia (secondary to pulmonary carcinoma), which occurs in adults, is not a concern in paediatric age group.

agent;

• to look for any underlying developmental anomaly • •

that predisposes a child to persistent or recurrent pneumonia; to assess acute complications and to guide management; to evaluate the sequelae of respiratory infection.

Chest X-ray and common infective agents The common aetiological agents that cause lower respiratory tract infection in children vary with age.

Infants and preschool children Causes In infants and preschool-age children, viruses are the major cause of respiratory tract infections.

110

I M A G I N G

The common viruses include adenovirus, respiratory syncytial virus, parainfluenza, influenza, measles or herpes virus.

Radiological findings Viral pneumonia The most common Roentgenographical findings in viral chest infection are as follows:

O F

P N E U M O N I A

I N

C H I L D R E N

more frequently encountered in paediatric age group. Hilar adenopathy is sometimes seen, while the following Roentgenographical findings are uncommon:

• • • • •

diffuse interstitial infiltrates, significant pleural effusion, pneumothorax, pneumatocoele, lung abscess.

• Parahilar peribronchial infiltrates: viral infections

• • •

predominantly involve respiratory mucosa of the airway, hence peribronchial inflammation and oedema are demonstrated as increased peribronchial opacities radiating from the hila on radiography (Figure 11.1). Bronchial wall thickening. Hyper-expansion of lungs. Segmental or lobar atelectasis: this can be explained by the small calibre of airways in children. The presence of minor amounts of oedema, mucus or inflammatory debris may compromise or occlude bronchi or bronchioles. In infants, collateral pathways of ventilation are less well developed. Their airways are also more collapsible. All these factors contribute to the observation that atelectasis is

Key points In viral pneumonia, the most common radiographic features are parahilar peribronchial infiltrates, bronchial wall thickening, hyperexpansion, segmental or lobar atelectasis. Significant pleural effusion, pneumothorax, pneumatocoele and lung abscess are uncommon.

School-age children Causes In school-age children, though viral agents remain the most common cause of lower respiratory tract infections, there is an increased incidence of bacterial infection, such as Streptococcus pneumoniae, Staphylococcus aureus, Haemophilus influenzae as well as Mycoplasma pneumoniae [1].

Radiographic findings It is difficult to differentiate between typical (i.e. bacterial) pneumonia and atypical (i.e. viral or mycoplasmal) pneumonia both clinically [2] and radiologically [3].

Bacterial pneumonia The classical radiographic appearance is localized airspace consolidation with or without airbronchogram:

• Caused by inflammatory exudate and oedema Fig. 11.1 Typical acute respiratory syncytial viral pneumonia in a 3-year-old child. There is pulmonary over-inflation and parahilar peribronchial shadowing (arrow heads).

within the acini.

• Typical distribution is lobar or segmental. • Associated pleural effusions are not uncommon [4] (Figure 11.2).

C H E S T

X

-

R A Y

A N D

C O M M O N

I N F E C T I V E

• Sometimes, pneumonia in children may have

•

a ‘round’ appearance, simulating intrathoracic mass in both antero-posterior and lateral views. Although it may closely resemble a tumour on the initial study, it changes rapidly after appropriate antibiotic therapy (Figure 11.3). Round pneumonia is associated with pneumococcal, staphylococcal or Klebsiella infection.

•

A G E N T S

111

of consolidation with or without airbronchograms, typically segmental or lobar distribution. Pleural effusion is more common. In primary tuberculosis, the most common radiographic features are hilar or mediastinal lymphadenopathy, with or without opacities in the lung.

Tuberculosis pneumonia Primary tuberculosis is the commonest form encountered in children [5]. Radiographic features include: – Hilar or mediastinal lymphadenopathy, with or without opacities in the lung [6]. – Occasionally, cavitation may be seen within the consolidation (Figure 11.4).

Key points

• In bacterial pneumonia, the most common radiographic features are areas

(a)

(b)

Fig. 11.2 Streptococcal pneumonia in a 5-year-old child. There is broncho-pneumonia in the right lower lobe associated with moderate amount of pleural effusion.

Fig. 11.3 Round pneumonia in a 3-year-old child. (a) Three rounded areas of consolidation are present (arrow heads) in the right lung, which resemble tumours. (b) Three days after initiation of appropriate antibiotic therapy, there is change in configuration of the consolidation.

112

I M A G I N G

O F

P N E U M O N I A

I N

C H I L D R E N

Fig. 11.5 Bacterial pneumonia in a 2-year-old child. CT showing segmental airspace consolidation with airbronchograms (arrow heads) in the right lower lobe.

• tends to be located at the middle and outer zones [7]. Fig. 11.4 Chest radiograph of an 8-year-old boy with tuberculous infection. There are hazy infiltrates in both apical regions with cavitation (arrow head) on the right side. There is also mild prominence of the right hilar region suggestive of presence of lymphadenopathy (arrow).

Consolidation on CT is defined as areas of increased pulmonary opacity with obscuration of underlying bronchovascular structures [8].

Viral infection • The commonest CT feature is peribronchial thick-

CT and common infective agents Why is it sparingly used in kids? Due to the added radiation, CT is rarely indicated in the primary assessment of uncomplicated respiratory infections in the immunocompetent child. Like chest radiography, CT features of typical (bacterial) and atypical (viral or mycoplasmal) pneumonia do overlap and some features are more frequently visualized in a certain group of pneumonia [7].

•

ening [9], reflecting the inflammatory changes and oedema of bronchial mucosa. The presence of ground-glass attenuation (defined as hazy increased attenuation without obscuration of bronchovascular structures [8]) without associated consolidation, a lobular distribution, at the inner layer of the lung in addition to the middle and outer layers, are more in favour of a viral pneumonia [7], i.e. no zonal predominance.

Tuberculosis In children with primary tuberculosis, the presence of CT features such as:

CT features

• low-attenuation lymph nodes and lymph node

Bacterial infection

• branching centrilobular nodules (‘tree-in-bud’

The most common CT manifestations are:

• areas of consolidation with or without airbronchograms (Figure 11.5);

• typically with a segmental or lobar distribution;

calcifications (Figure 11.6), appearance) (Figure 11.7) and

• miliary nodules (Figure 11.8) are helpful in suggesting the diagnosis in cases where the radiograph is normal or equivocal [10].

P E R S I S T E N T

O R

R E C U R R E N T

P N E U M O N I A

Fig. 11.6 CT of an 8-year-old boy with tuberculous infection. Note the calcified lymph nodes (arrows) in both hilar regions. A central venous catheter is in situ.

113

Fig. 11.8 HRCT of a 16-year-old boy with miliary tuberculosis. There are well-defined 1- to 2-mm nodules disseminated throughout both the lungs (arrow heads).

Key points

• In bacterial infection, the most common •

•

(a)

CT features are areas of consolidation with or without air-bronchograms. In viral infection, the suggestive CT features are peribronchial thickening, presence of ground-glass attenuation without associated consolidation, a lobular and inner distribution. In tuberculosis infection, the suggestive CT features are low attenuation lymph nodes, lymph node calcifications, branching centrilobular nodules (‘tree-in-bud’ appearance) and miliary nodules.

Evaluation of persistent or recurrent pneumonia Underlying predisposing conditions

(b)

Fig. 11.7 HRCT of a 17-year-old boy with pulmonary tuberculosis. Note numerous centrilobular nodules (arrow heads in (a)) and linear branching structure (‘tree-in-bud’ appearance, arrow in (b)) highly suggestive of endobronchial spread of tuberculosis.

Children with persistent or recurrent respiratory tract infection may have an underlying condition that predisposes them to the susceptibility of pneumonia, such as:

• • • • •

immunodeficiency, gastro-oesophageal reflux, repeated aspiration, inhalation of foreign bodies, underlying bronchiectasis such as cystic fibrosis (Figure 11.9).

114

I M A G I N G

O F

P N E U M O N I A

I N

C H I L D R E N

(a) (a)

(b)

Fig. 11.9 An 8-year-old Jewish boy with cystic fibrosis. (a) HRCT showing cylindrical bronchiectasis (arrows) of bilateral upper lobes. (b) Chest radiograph taken during acute infective exacerbation showing peribronchial thickening and patchy infiltrates in both upper zones as well as nodular shadowing in the lower zones.

Developmental abnormalities If the above predisposing conditions are excluded, developmental abnormalities of the lung should always be considered, such as:

• Pulmonary sequestration: demonstration of a systemic arterial supply to the consolidated lung tissue on contrast-enhanced CT is diagnostic of the condition [11] (Figure 11.10).

(b)

Fig. 11.10 Pulmonary sequestration in a 11-yearold boy who suffers from recurrent left lower lobe consolidation. (a) Contrast-enhanced CT showing an aberrant artery (arrow) arising from the thoracic aorta, supplying the pulmonary tissue in the left lower lobe. (b) Three-dimensional CT reconstruction showing the relationship of the aberrant artery (arrows) with adjacent vertebrae and ribs.

• Cystic adenomatoid malformations: these are char•

acterized by multi-septated air- and fluid-filled cysts [12] (Figure 11.11). Bronchogenic cyst: usually presents as an ovoid or round lesion of water or soft tissue attenuation (Figure 11.12). Sometimes the cyst may appear hyperdense due to intracystic haemorrhage, protein content or calcium of milk. The bronchogenic cyst may get infected or cause compression onto the adjacent bronchus.

C O M P L I C A T I O N S

O F

P N E U M O N I A

A N D

R O L E

O F

C T

115

• Immunodeficiency • Underlying bronchiectasis • Developmental lung masses – Pulmonary sequestration – Cystic adenomatoid malformation – Bronchogenic cyst

Complications of pneumonia and role of CT CT is useful for evaluation of complications related to community-acquired pneumonia and helpful in guiding the management [13]. Fig. 11.11 Congenital cystic adenomatoid malformation in an infant. CT showing multiple thin-walled cystic lesions with surrounding consolidation in the right lung.

Parapneumonic effusion/empyema Parapneumonic effusions (Figure 11.13) occur commonly in children with bacterial pneumonia. Plain radiograph is good enough for initial diagnosis. However, if the effusion is large, loculated, has delayed appearance, or clinically not responsive to antibiotic therapy alone, an empyema should be suspected. Contrast-enhanced CT is advocated in differentiating empyema and transudative parapneumonic effusion. A combination of the following features is highly suggestive of empyema [14,15] (Figure 11.14):

• enhancement and thickening of the parietal and visceral pleura,

• thickening of extrapleural subcostal tissues, • increased attenuation of the extrapleural subFig. 11.12 Bronchogenic cyst in an 8-year-old boy. CT showing a classical appearance of an ovoid lesion with soft tissue density (arrow) in the right paravertebral region.

costal fat,

• adjacent chest wall oedema. A more aggressive therapy is therefore indicated in this circumstance such as chest tube placement, thrombolytic therapy [16], thoracoscopy and debridement [17] for management of empyema.

Key points Causes of persistent and recurrent chest infection

Necrotizing pneumonia

• Gastro-oesophageal reflux • Aspiration • Foreign body inhalation

Underlying suppurative parenchymal complications should be suspected in children with persistent fever and sepsis despite appropriate medical treatment of pneumonia.

116

I M A G I N G

Fig. 11.13 Bacterial pneumonia in a 6-year-old child. There is presence of a large right parapneumonic effusion. Contrast-enhanced CT showing homogeneous enhancement of the consolidated lung (arrow) indicating that the lung parenchyma is noncomplicated.

O F

P N E U M O N I A

I N

C H I L D R E N

Fig. 11.15 Necrotizing pneumonia in a 6-yearold boy. There are areas of decreased enhancement in the consolidated lung (arrow) and multiple tiny air cavities (arrow head) indicating cavitary necrosis of the underlying lung parenchyma. There is also presence of pleural effusion.

Fig. 11.16 Lung abscess in a 15-year-old boy. CT showing a cavitary lesion containing air-fluid level (arrow). The luminal margin is characteristically thick and irregular. Fig. 11.14 Empyema complicating Steptococcus pneumoniae pneumonia in a 3-year-old girl. Contrastenhanced CT showing enhancement and thickening of the pleura (black arrows), increased attenuation of the extrapleural subcostal space (arrow heads) and oedema of the chest wall (white arrow).

• In addition, the necrotic lung tissue may become liquefied thus forming multiple thin-walled cavities containing air or fluid but without enhancing border [19](Figure 11.15).

CT features of compromised and non-compromised lung

Abscess

In necrotizing pneumonia:

Lung abscesses are characterized by fluid- or airfilled cavities with enhancing wall (Figure 11.16). The surrounding lung shows no evidence of necrosis.

• There are areas of decreased or without enhancement in the consolidated lung (Figure 11.15) indicating parenchymal ischaemia or impending infarction [18]. This is in contrast to the diffuse enhancement of non-compromised lung parenchyma consolidated with pneumonia (Figure 11.13).

It is important to differentiate lung abscesses and necrotizing pneumonia. The former requires aspiration or drainage if there is poor response to medical therapy, whereas necrotizing pneumonia does not require invasive treatment, which may

C O M P L I C A T I O N S

O F

P N E U M O N I A

Fig. 11.17 Pneumatocoele in a 2-year-old child. CT showing a thin-walled cavity (arrow) within the consolidated right lung. Air-bronchograms are evident in both consolidated lungs.

even be harmful to the patient resulting in complications such as bronchopleural fistula [20].

A N D

R O L E

O F

C T

117

Fig. 11.18 This is the follow-up CT of the same patient in Figure 11.14. Despite antibiotic treatment, the child still has persistent fever. There is a residual loculated collection (arrow) in the right lung despite resolution of the consolidation and pleural effusion.

Pneumatocoele The presence of air cavities within the consolidation does not always point to severe stage of complicated pneumonia. Pneumatocoeles are thin-walled cysts, which represent a stage of resolving or healing necrosis (Figure 11.17). The wall of the pneumatocoele does not enhance and the surrounding consolidated lung does not demonstrate evidence of necrosis.

Other roles of CT Besides its diagnostic role, CT is also useful in monitoring treatment progress and guiding therapy:

• detection of inadequately drained effusion, • detection of loculated collection (Figure 11.18), • identify malpositioned chest tube (Figure 11.19)

Fig. 11.19 CT showing sub-optimal positioning of the chest tube (arrow) in relation to the more posteriorly located empyema.

which requires re-adjustment,

• guide the aspiration or drainage procedure for abscesses not responding to medical therapy.

• • • •

empyema, necrotizing pneumonia, lung abscess, pneumatocoele.

Key points

CT is also useful in guiding management of

CT is useful for the evaluation of acute complications related to

• acute pulmonary or pleural complications, • placement of chest tube, • guided aspiration of loculated effusion or

• community-acquired pneumonia, • parapneumonic effusion,

abscess.

118

I M A G I N G

Chronic sequelae and role of CT

O F

P N E U M O N I A

I N

C H I L D R E N

Swyer–James syndrome

There are a number of complications that are commonly associated with chest infection in children.

Swyer–James or Macleod’s syndrome is a variant of post-infectious constrictive bronchiolitis after respiratory chest infection in early childhood [23].

Brochiolitis obliterans

The cardinal sign on chest radiograph is unilateral hyper-transradiancy (Figure 11.21a).

Causes

HRCT shows unilateral hyper-lucency and decreased pulmonary vascularity (Figure 11.21b).

Bronchiolitis obliterans is observed in children following a chest infection, in particular adenoviral or mycoplasma infection. It is characterized by inflammatory damage to the small airways, resulting in bronchiole wall thickening, mucostasis, progressive bronchiole narrowing and distortion [21].

Radiological findings Chest radiographs are usually normal, although hyper-aeration and vascular attenuation are sometimes seen. High-resolution CT (HRCT) demonstrates a mosaic perfusion pattern due to oligaemia and air-trapping. There are areas of decreased parenchymal attenuation in the affected lung segments as compared with the higher attenuation in the normal parenchyma (Figure 11.20). The mosaic attenuation pattern is further exaggerated on an expiratory scan [22].

(a)

(b)

Fig. 11.20 HRCT of a 5-year-old child with postviral bronchiolitis obliterans. Note the difference in attenuation of the lung parenchyma (mosaic attenuation pattern). The affected lung segments (arrows) showing a lower attenuation than the normal parenchyma (arrowheads).

Fig. 11.21 Swyer–James syndrome in a 12-year-old girl. (a) Chest radiograph showing the cardinal sign of unilateral hyper-transradiancy of the right lung. (b) HRCT showing reduced attenuation and paucity of bronchovascular markings in the right lung.

C H R O N I C

S E Q U E L A E

A N D

R O L E

Fibrosis Results of remodelling of pulmonary architecture post-infection include parenchymal scarring or fibrosis, distortion of bronchovascular bundles (Figure 11.22) and bronchial wall thickening (Figure 11.23).

Bronchiectasis Bronchiectasis (Figure 11.24) is one of the most common chronic complications of childhood pneumonia. HRCT is more sensitive than chest radiograph in making the diagnosis.

O F

C T

119

HRCT features of bronchiectasis are [24]:

• bronchial dilatation in relation to the accompanying pulmonary artery (‘signet ring’ sign);

• bronchial wall thickening; • visualization of airways more distally than • •

usual, within 1 cm of costal or paravertebral pleura [25]; crowding of airways; absence of normal tapering of airways.

Chronic bronchiectasis most commonly occurs secondary to adenovirus, bacterial and tuberculosis infection [26] (Figure 11.25).

Fig. 11.22 Post-infective pulmonary fibrosis in a 5-year-old boy. There is evidence of parenchyma bands (arrow) and minor bronchovascular distortion in the right posterior lung compatible with scars.

Fig. 11.24 Bronchiectasis in a 17-year-old boy with viral pneumonia at early childhood. HRCT showing dilatation of the bronchioles in the left lower lobe. The calibre of the bronchiole is much larger than its accompanying vessel, giving rise to the ‘signet ring’ sign (arrow).

Fig. 11.23 Post-infective bronchial wall thickening in a 6-year-old boy. There is diffuse bronchial wall thickening (arrowheads) in the right lung. Minor parenchymal band is also present.

Fig. 11.25 Persistent bronchiectasis in a 10-yearold girl post-tuberculous infection. HRCT showing bronchiectasis (arrow) in the chronically consolidated right middle lobe.

120

I M A G I N G

Key points Chronic sequelae of respiratory chest infection is well demonstrated by HRCT. These include the following:

• • • • • •

Bronchiolitis obliterans Swyer–James syndrome Parenchymal scarring/fibrosis Bronchovascular bundle distortion Bronchial wall thickening Bronchiectasis

Conclusion Plain radiograph is the primary imaging modality for diagnosing pneumonia in children. It is usually not possible to make a definitive diagnosis as to the pathogen responsible for a lower respiratory infection using either plain radiographs or CT; however, certain radiological features are relatively more frequent in a certain group of pathogens. CT including conventional CT or HRCT has superior imaging accuracy than plain radiographs and is indicated in the following circumstances: 1. to exclude an underlying abnormality in unresolved and recurrent infections; 2. when a complication is suspected; 3. to assess the sequelae of respiratory infection.

References 1. McIntosh K. Current concepts: community-acquired pneumonia in children. New Engl J Med 2002; 346: 429–437. 2. Forgie IM, O’Neill KP, Lloyd-Evans N et al. Etiology of acute lower respiratory tract infections in Gambian children. I. Acute lower respiratory tract infections in infants presenting at the hospital. Pediatr Infect Dis J 1991; 10: 33–41. 3. Courtoy I, Lande AE and Turner RB. Accuracy of radiographic differentiation of bacterial from nonbacterial pneumonia. Clin Pediatr (Phila) 1989; 28: 261–264. 4. Condon VR. Pneumonia in children. J Thorac Imag 1991; 6: 31–44. 5. Stansberry SD. Tuberculosis in infants and children. J Thorac Imag 1990; 5: 17–27.

O F

P N E U M O N I A

I N

C H I L D R E N

6. Leung AN, Muller NL, Pineda PR and FitzGerald JM. Primary tuberculosis in childhood: radiographic manifestations. Radiology 1992; 182: 87–91. 7. Tanaka N, Matsumoto T, Kuramitsu T et al. High resolution CT findings in community-acquired pneumonia. J Comput Assist Tomogr 1996; 20: 600–608. 8. Austin JH, Muller NL, Friedman PJ et al. Glossary of terms for CT of the lungs: recommendations of the Nomenclature Committee of the Fleischner Society. Radiology 1996; 200: 327–331. 9. Kuhn JP. High-resolution computed tomography of pediatric pulmonary parenchymal disorders. Radiol Clin N Am 1993; 31: 533–551. 10. Kim WS, Moon WK, Kim IO et al. Pulmonary tuberculosis in children; evaluation with CT. Am J Roentgenol 1997; 168: 1005–1009. 11. Frush DP and Donelly LF. Pulmonary sequestration spectrum: a new spin with helical CT. Am J Roentgenol 1997; 169: 679–682. 12. Kim WS, Lee KS, Kim IO et al. Congenital cystic adenomatoid malformation of the lung. CT–pathologic correlation. Am J Roentgenol 1997; 168: 47–53. 13. Donnelly LF. Practical issues concerning imaging of pulmonary infection in children. J Thorac Imaging 2001 Oct; 16(4): 238–250. 14. Muller NL. Imaging of pleura. Radiology 1993; 186: 297–309. 15. Waite RJ, Carbonneau RJ, Balikian JP et al. Parietal pleural changes in empyema: appearance at CT. Radiology 1990; 175: 145–150. 16. Moulton JS, Benkert RE, Weisiger KH et al. Treatment of complicated pleural fluid collections with image-guided drainage and intracavitary urokinase. Chest 1995; 108: 1252–1259. 17. Silen ML and Weber TR. Thoracoscopic debridement of loculated empyema thoracis in children. Ann Thorac Surg 1995; 59: 1166–1168. 18. Donnelly LE and Klosterman LA. Pneumonia in children; decreased parenchymal contrast enhancement-CT sign of intense illness and impending cavitary necrosis. Radiology 1997; 205: 817–820. 19. Donnelly LF and Klosterman LA. Cavitary necrosis complicating pneumonia in children: sequential findings on chest radiography. Am J Roentgenol 1998; 171: 253–256. 20. Hoffer FA, Bloom DA, Colin AA and Fishman SJ. Lung abscess versus necrotizing pneumonia: implications for interventional therapy. Pediatr Radiol 1999; 29: 87–91. 21. Colby TV. Bronchiolitis. Pathologic considerations. Am J Clin Pathol 1998; 109: 101–109. 22. Hansell DM, Ruben MB, Padley SP and Wells AU. Obliterative bronchiolitis. Individual CT signs of small airways disease and functional correlation. Radiology 1997; 203: 721–726. 23. Marti-Bonmati L, Ruiz Perales F, Catala F et al. CT findings in Swyer–James syndrome. Radiology 1989; 172: 477–480. 24. Kornreich L, Horev G, Ziv N and Grunebaum M. Bronchiectasis in children: assessment by CT. Pediatr Radiol 1993; 23: 120–123. 25. Kim JS, Muller NL, Park CS et al. Cylindrical bronchiectasis; diagnostic findings on thin-section CT. Am J Roentgenol 1997; 168: 751–754. 26. Donnelly LF. CT of acute pulmonary infection/trauma. In: J Lucaya and JL Strife (Eds) Pediatric Chest Imaging. Springer-Verlag, Berlin, Heidelberg. 2002.

Imaging and Clinical Management of Paediatric SARS WCW Chu, EKL Hon, FWT Cheng and TF Fok

Introduction Clinical presentation Plain radiography High-resolution computed tomography

121 121 122

CHAPTER

12

Management and outcome Infection control Conclusion

128 129 130

126

Introduction

Clinical presentation

Although severe acute respiratory syndrome (SARS) has wreaked havoc in south-east Asia and other parts of the world, it appears to be a disease that predominantly affects adults. Less than 10% of the infected population in Hong Kong were children (http://www.info.gov.hk/). Among these infected children, only 5% required care in the intensive care unit (ICU) and less than 1% required mechanical ventilation based on the local experience from Hong Kong hospitals. In contrast to its adult counterpart:

Most children with SARS have either been in close contact with infected adults, as a household contact or in a health care setting. These are believed to be the important routes of transmission that put children at a particular risk. Surprisingly, in Hong Kong there has been no major spread of the disease among classmates in schools. This may partly be explained by the early strict hygiene precautions undertaken by schools following a large-scale educational programme conducted by the local government.

• clinical course of affected children was usually milder [1];

• duration for resolution was shorter [1]; • potential of children to infect others was lower [2,3]. This chapter will discuss the clinical features, radiological presentation, management and outcome of children suffering from SARS based on our institutional experience.

There are two distinct patterns of clinical presentation among children in different age groups:

• Teenage patients presented with symptoms of •

malaise, myalgia, chills and rigor similar to presenting symptoms in adults [2,3]. Younger children presented mainly with cough and runny nose, and none had chills, rigor or myalgia [1].

122

C L I N I C A L

M A N A G E M E N T

100 80 %

P A E D I A T R I C

S A R S

change. Therefore, in our experience it is not logical to differentiate suspected SARS from probable SARS based on radiographical changes alone. Furthermore, any child residing in an endemic area who has a ‘cold’ with fever and cough would be diagnosed as suspected SARS by WHO definition!

120

60 40 20

Fe ve C r ou M gh ya lg C i hi lls a R /r un ig ny or D nos ys e So pno re ea th H ro a ea da t D che iz zi Fe n br M e ss ile al co ais nv e ul si on

0

Fig. 12.1 children.

O F

WHO has since revised the definition of a probable case to include a suspected case of SARS that is positive for SARS coronavirus [5].

Incidence of presenting symptoms of SARS

Key points 1. Contacts at home and in health care setting are important routes of transmission of SARS to children. 2. Young children have a milder form of disease with a less-aggressive clinical course than adults. 3. Teenagers may mimic the disease pattern of adult and can have a longer and more severe form of the disease.

The clinical course is much milder and shorter among younger patients except infants. The average hospital stay is 2–4 weeks in all these children. The incidence of clinical presentation of children with suspected SARS is shown in Figure 12.1.

Problems with World Health Organization definition • Although we follow World Health Organization

•

(WHO) case definition for SARS in the diagnosis of paediatric patients who suffer from the disease, we find ‘SARS’ a misnomer for children since the clinical features in majority of them are neither ‘severe’ nor ‘respiratory’ in nature [4]. The distinction between suspected case and probable case based on radiographic changes alone is also not very helpful in the clinical management. WHO defines SARS as either suspected or probable. The case definition of suspected SARS is (i) fever; (ii) respiratory symptoms including cough and difficulty breathing and (iii) close contact with SARS patients or (iv) history of travel to an epidemic area. Probable SARS is a suspected case with radiographic evidence of pneumonia or respiratory distress syndrome. Thus, probable SARS is a suspected SARS with chest radiographic changes. Aetiological diagnosis is not required in these clinical definitions. However, it is neither practical not to have chest radiographs performed on a child suspected as having SARS nor sensible to label a child as having suspected SARS in the absence of chest radiographical

Plain radiography A chest radiograph is the primary initial imaging modality for SARS, as in the majority of acute chest infections in children. It provides supportive information on the diagnosis of SARS in addition to the WHO definition of surveillance case. Most children have radiographical abnormalities on presentation to the hospital. However, the radiographical appearance itself are non-specific. Based on our institutional experience:

• The primary radiographical finding in paediatric

•

•

patients with SARS is airspace opacification, which can be unilateral focal (62%) or unilateral multiple/ bilateral (38%) (Figure 12.2). Children with younger age, and those with mild symptoms usually present with unilateral focal consolidation (Figure 12.3), while multi-focal and bilateral involvement (Figure 12.4) occurs in a few teenage patients who present with a more severe disease and require supportive oxygen therapy as well as a longer hospital stay [1]. There is a higher prevalence of consolidation in the lower lung zone (63%) compared with the

P L A I N

123

R A D I O G R A P H Y

25%

62% 13%

focal

multi-focal

bilateral

Fig. 12.2 Pattern of airspace opacification on chest radiograph in SARS children.

12%

Fig. 12.4 Chest radiograph of a 15-year-old boy, who presented on day 3 with fever, chills and myalgia. There is a unilateral focal airspace consolidation in the left lower zone (arrows).

25%

6%

57% Upper zone

Lower zone

Middle zone

Upper and lower

Fig. 12.3 Zonal distribution of airspace opacification on chest radiograph in SARS children.

upper lung zone (31%), while only 6% involve the mid-zone (Figure 12.5). Although a coronavirus has been implicated as the causative agent of SARS [6], the radiological presentation of SARS in children is different from the most common radiological appearance of viral disease of the lower respiratory tract in children. In viral infection such as respiratory syncytial viral infection of children, the most common radiological findings on chest radiographs are:

• peribronchial shadowing, bronchial wall thickening and perihilar linearity [7];

• air-trapping and atelectasis are often seen [8]; • coalescent airspace consolidation is less common.

Fig. 12.5 Chest radiograph of an 8-year-old boy, who presented on day 7 with fever, chills and rigor. There is ill-defined airspace consolidation in the left lower zone, with loss of the left heart silhouette. Another smaller are of airspace opacification (arrows) is present in the right upper zone.

Localized airspace consolidation with lobar or segmental distribution is a more classical radiographical presentation of bacterial pneumonia [9], though there is significant overlap of radiological appearance between viral and bacterial infection [10,11].

124

C L I N I C A L

M A N A G E M E N T

(a)

O F

P A E D I A T R I C

S A R S

(b)

Fig. 12.6 Chest radiographs (a) on admission, (b) 4 days later of a 15-year-old boy, who presented with fever, rough and myalgia. There is a small focal area of airspace opacification with air-bronchogram in the right upper zone (arrows in a), which enlarges and becomes more apparent on the subsequent radiograph (b).

(a)

(b)

Fig. 12.7 Chest radiographs (a) on day 3 and (b) day 10 after the onset of fever in a 15-year-old girl presenting with sore throat, myalgia and dyspnoea. Initial radiograph (a) shows unilateral focal airspace opacification in the left lower zone (arrows) which progresses to coalescent airspace consolidation of bilateral lower zones (b). Courtesy: Dr CW Leung and Dr MC Chiu, Princess Margaret Hospital, Hong Kong.

The primary radiological feature of SARS therefore resembles that of bacterial pneumonia with airspace consolidation; however, there is a striking absence of pleural effusion, which is commonly encountered

in bacterial chest infection [12]. Cavitation that occurs in pneumococcal and anaerobe pneumonia [13,14] is also not a feature of SARS. Unlike, tuberculosis [15], another common disease in children

P L A I N

125

R A D I O G R A P H Y

transmitted via contact, lymphadenopathy is not a feature of SARS. In our patients with SARS, the airspace opacification usually became worse on days 5–7 after the onset of the fever. Unlike adults who usually progressed to multiple areas of involvement, majority of our children only showed increase in extent of airspace opacification in the same lung zone where consolidation was first identified (Figure 12.6). Progression to multi-focal bilateral lung infiltrates was only observed in a few teenage patients who run a more-aggressive course of the disease (Figure 12.7). The mean duration of time required for complete resolution of the consolidation on radiograph was 16 days (range 8–30 days) from our institutional experience. No definite scarring, volume loss, bronchial thickening or bronchiectasis is identified in the follow-up radiographs of our paediatric patients who have recovered from the illness (Figure 12.8). Again this is in contrary to the initial report from adults that pulmonary complications in the form of pulmonary fibrosis and bronchiectasis may be as high as 20% [16,17].

Fig. 12.8 Chest radiograph of the same patient in Figure 12.3, 1 month after the acute infection, shows complete resolution of radiological abnormality. No evidence of residual parenchymal density, bronchiectasis or volume loss.

Key points Comparison of radiographical features in children and adults Children

Adults

The most common radiological finding is airspace consolidation Younger children and those with mild symptoms usually present with unilateral focal consolidation; teenage patients with a more severe disease present with multi-focal and bilateral involvement Majority show focal progression of airspace opacification without bilateral lung involvement Lymphadenopathy, pleural effusion and cavitation are not seen Radiological abnormality becomes worse on days 5–7 Complete resolution is achieved around 2 weeks

Similar to children Most adults also present with initial unilateral focal consolidation but with rapid progression

Progression with extensive bilateral involvement is common Similar to children Radiological abnormality becomes worse on day 7 It takes a longer time for resolution and many have long-term residual changes

126

C L I N I C A L

M A N A G E M E N T

High-resolution computed tomography Chest computed tomography (CT) is usually not required for the initial diagnosis of paediatric SARS because majority of the patients show radiographical abnormality at presentation. However, CT plays a role in the detection and exclusion of SARS in highly suspected cases with clinical symptoms and contact history but negative radiographical findings. The recognition and treatment of paediatric SARS are particularly important not only because of the impact on the personal health of infected children, but also because children with primary infection may become the reservoir from which future cases will emerge. We therefore advocate the use of high-resolution CT (HRCT) in making prompt diagnosis in highly suspected cases with a non-contributory chest radiograph. Not uncommonly, the number of areas of abnormality detected on CT is much higher than the number detected on chest radiograph.

O F

P A E D I A T R I C

S A R S

consolidation are the two predominant features on HRCT (Figure 12.11). The former refers to hazy increased attenuation of the lung with preservation of the bronchial and vascular markings, caused by partial filling of the airspaces or partial collapse of alveoli. The latter refers to a homogeneous increase in pulmonary parenchymal attenuation that obscures the margins of vessels and airway walls. It is common to find a combination of both findings in SARS. In patients with multi-focal disease, a mosaic pattern of lung attenuation with ground-glass and airspace infiltrates is observed simulating the appearance of bronchiolitis obliterans organizing pneumonia (BOOP). Again the above radiological appearances are non-specific. Both ground-glass opacity and consolidation attenuation are common findings in children suffering from pneumonia of any aetiology [19]. Pulmonary nodules, septal thickening, pleural effusion and lymphadenopathy are not the features of SARS.

The fast scan times of modern CT scanners permit performing HRCT in children without the need for sedation. The radiation dose in children is reduced using a lower milliampere technique (50–80 mA) without compromising the diagnostic value of the scan [18] (Figure 12.9). Similar to the findings of plain radiography, most patients present with milder form of the disease and show focal segmental airspace disease on HRCT (Figure 12.10). Ground-glass opacification and

Fig. 12.9 HRCT of a 4-month-old baby, who presented with fever and history of contact with SARSinfected adults. Using only 50 mA and non-breath-hold technique, there is satisfactory image quality with identification of airspace consolidation and air-bronchogram in the superior segment of left lower lobe (arrows).

Fig. 12.10 HRCT of a 15-year-old boy presenting with fever, rough and myalgia. There is a large area of mixed ground-glass opacification and airspace consolidation in the right upper lobe.

Fig. 12.11 HRCT of a 14-year-old girl, who presented with persistent fever for 1 week with chills and rigor, runny nose and myalgia. There is mixed airspace consolidation and ground-glass opacity in the left lower lobe.

H I G H

-

R E S O L U T I O N

C O M P U T E D

There is no specific preference of distribution of the disease in children. We have observed an approximately equal involvement of subpleural and peribronchial regions in children presenting with either segmental or multi-focal disease (Figure 12.12) whereas peripheral distribution is a predominant feature in the adult cases [20].

T O M O G R A P H Y

127

initial children who have recovered from SARS:

• Despite no significant change on radiograph,

•

HRCT was performed for children with prolonged course of the disease. In the limited follow-up of our

HRCT shows evidence of partial resolution of the airspace consolidation (Figure 12.13) and therefore provides more accurate assessment of the response to the treatment. Radiological findings on HRCT show no evidence of bronchial dilatation, fibrous scarring, airtrapping or emphysematous changes.

Key points 1. HRCT is recommended to aid diagnosis in children with strong clinical suspicion of SARS but noncontributory radiographical findings. 2. A lower milliampere technique (50–80 mA) is recommended for HRCT in paediatric patients to reduce the radiation dose. Comparison of HRCT features in children and adults Children

Adults

Ground-glass opacification and consolidation are the two predominant features Pulmonary nodules, septal thickening, pleural effusion and lymphadenopathy are not seen

Same as children

Equal involvement of subpleural and peri-bronchial regions Bronchial dilatation, fibrous scarring, air-trapping or emphysematous changes not seen in the follow-up HRCT of recovered children

(a)

Interlobular septal and intralobular interstitial thickening are sometimes seen giving rise to crazy-paving pattern. Pleural effusion and lymphadenopathy are not seen Lesions are predominantly peripheral and subpleural in location Parenchymal bands, bronchovascular distortion, irregular interface and traction bronchiectasis are seen in some adult patients on follow-up

(b)

Fig. 12.12 HRCT of an 8-year-old boy, who presented with day 5 fever with chills and rigor. There are multifocal areas of consolidation in the central region of left upper lobe (a) and the peripheral region of right upper lobe (b). Courtesy: Dr PS Kan, Alice Ho Miu Ling Nethersole Hospital, Hong Kong.

128

C L I N I C A L

M A N A G E M E N T

(a)

(b)

(c)

(d)

O F

P A E D I A T R I C

S A R S

Fig. 12.13 Chest radiographs of a 14-year-old girl presenting with prolonged course of symptoms. There is no significant change in the extent of ground-glass opacification in the left lower zone on admission (a) and 10 days later (b). HRCT however, shows evidence of partial resolution of the ground-glass opacification in the left lower lobe 10 days after the onset of fever (d) as compared with the initial study (c).

Management and outcome The first 10 children reported in our initial cohort received a treatment regimen similar to the regimen of adult SARS in Hong Kong [1]. Currently, intravenous cefotaxime (25–50 mg/kg/dose, every 6 or 8 hours), oral clarithromycin (15 mg/kg/day, to a maximum dose of 250 mg twice daily) and oral ribavirin (40 mg/kg/day, in two or three divided doses) are started if a clinical diagnosis of SARS is suspected on admission. Oral prednisolone (0.5–2 mg/ kg/day) is added if there is no decrease in fever or improvement in the general well-being of the patient within 48 hours. If the child is admitted with

moderately severe symptoms of high swinging fever and marked malaise, then intravenous ribavirin (40–60 mg/kg/day, in three divided doses) and intravenous hydrocortisone (2 mg/kg/dose, every 6 hours) in addition to antibacterial therapy are administered immediately after admission. For patients with persistent fever and progressive clinical or radiological deterioration, pulse intravenous methylprednisolone (10–20 mg/kg/dose) is administered. The decision to give further pulsed treatment is based on clinical response. Antibacterial agents are discontinued 5 days after defervesence. Ribavirin is administered for 1–2 weeks, and corticosteroid is tapered over the course of 2–4 weeks.

I N F E C T I O N

129

C O N T R O L

The management regimen is likely to undergo revisions as more is known about SARS. There have been concerns with regard to the use of corticosteroids in the treatment of SARS. The indication for corticosteroid use in severe infectious diseases is the need to decrease inflammation caused by excessive cytokine response by the host. Since all patients with SARS have pneumonia, one might assume that ribavirin treatment would be more effective if administered by small particle aerosol generator (SPAG), as this method has been recommended previously for the treatment of respiratory syncytial viral infections in children. However, this method of administration is not considered because the first major outbreak of SARS appeared to have been accentuated by the use of a nebulizer on a general ward [2]. The overall prognosis in children with SARS appears to be good. However, a small number of children have required care in the ICU and mechanical ventilation. In accordance to our institutional policy, all patients with probable or suspected SARS are hospitalized for 21 days.

Key points Current medical treatment for paediatric patients with SARS at our institution Suspected paediatric SARS

Infection control Risk stratification and guidelines for paediatric patients All children with fever and pneumonic changes on chest radiograph are admitted to the ultra-high-risk (UHR) wards, either UHR-S area (with SARS contact) or UHR-I area (without SARS contact). This policy is adopted to separate patients with probable or suspected SARS from those who are admitted for afebrile illness to prevent the possibility of cross infection. Patients with respiratory failure are admitted to the ICU based on the prevailing selection criteria. During their hospitalization, all children are required to wear surgical masks at all times. Upon discharge from hospital, parents of the children are given an information sheet detailing the precautions to be taken at home. In particular, they are reminded that their body secretions and excretions might still contain the pathogen [15]. These children should limit interaction outside the home (i.e. not go to school or other public areas) for at least 10 days following discharge.

SARS precautions in paediatric wards The precaution measures taken by health care workers are the same as in the adult wards. Below are the highlights of some important issues:

Mild symptoms

Moderately severe symptoms + high swinging fever

(i) Cefotaxime i.v. (ii) Clarithromycin p.o. (iii) Ribavarin p.o.

(i) Ribavirin i.v. (ii) Hydrocortisone i.v./ prednisolone p.o. (iii) Cefotaxime i.v. (iv) Clarithromycin p.o.

No improvement

Persistent fever, clinical deterioration

+ Prednisolone p.o.

+ Pulse methylprednisolone i.v. No improvement + Pulse methylprednisolone i.v.

• Designated places and instructions are available • •

• •

for putting on and removing personal protective equipment. Routine thorough cleansing of the ward is carried out at least three times per day using hypochlorite solution. There is a strict control on patient visits to reduce the risk of cross infection. Visitors are not allowed in UHR-S area, and only one parent is allowed to visit each patient for at most 2 hours everyday. Visitors must follow the dress codes appropriate to the risk stratification of the areas. ‘Police nurses’ are present at the entrance to ensure that the above steps are strictly followed.

130

C L I N I C A L

M A N A G E M E N T

Conclusion Based on our institutional experience in imaging children with SARS, we advocate chest radiography to be the initial imaging modality for all patients suspected of SARS. HRCT is reserved for those cases with high index of suspicion but negative radiographical findings. HRCT is also useful in monitoring the treatment response in children with unexpectedly long course of the disease. Like any other chest infection of children, SARS may give rise to long-term complications though significant chest radiographical changes are not evident in the initial cohort of our patients. We speculate that those children without obvious extensive BOOP pattern would do well, since children have a lot of pulmonary reserve. The impact of SARS on a growing child may become clearer with long-term follow-up of this group of patients in the future.

References 1. Hon KL, Leung CW, Cheng WT, Chan PK, Chu WC, Kwan YW, Li AM, Fong NC, Ng PC, Chiu MC, Li CK, Tam JS, Fok TF. Clinical presentations and outcome of severe acute respiratory syndrome in children. Lancet 2003 May 17; 361(9370): 1701–1703. 2. Lee N, Hui D, Wu A, Chan P, Cameron P, Joynt GM, Ahuja A, Yung MY, Leung CB, To KF, Lui SF, Szeto CC, Chung S, Sung JJ. A major outbreak of severe acute respiratory syndrome in Hong Kong. N Engl J Med 2003 May 15; 348(20): 1986–1994. Epub 2003 Apr 07. 3. Tsang KW, Ho PL, Ooi GC, Yee WK, Wang T, Chan-Yeung M, Lam WK, Seto WH, Yam LY, Cheung TM, Wong PC, Lam B, Ip MS, Chan J, Yuen KY, Lai KN. A cluster of cases of severe acute respiratory syndrome in Hong Kong. N Engl J Med 2003 May 15; 348(20): 1977–1985. Epub 2003 Mar 31. 4. Hon KL, Li AM, Cheng FW, Leung TF, Ng PC. Personal view of SARS: confusing definition, confusing diagnoses. Lancet 2003 Jun 7; 361(9373): 1984–1985. 5. World Health Organization, Geneva. Case Definition for Surveillance of Severe Acute Respiratory Syndrome, SARS. http://www.who.int/csr/sars/casedefinition (accessed 1 May 2003).

O F

P A E D I A T R I C

S A R S

6. Centers for Disease Control and Prevention. Diagnosis/ Evaluation for SARS. www.cdc.gov/ncidod/sars/diagnosis.htm (accessed 2003). 7. Osborne D. Radiologic appearance of viral disease of the lower respiratory tract in infants and children. Am J Roentgenol 1978; 130: 29–33. 8. Kirkpatrick JA. Pneumonia in children as it differs from adult pneumonia. Semin Roentgenol 1980; 15: 96–103. 9. Condon VR. Pneumonia in children. J Thorac Imag 1991; 6: 31–44. 10. Han BK, Son JA, Yoon HK et al. Epidemic adenoviral lower respiratory tract infection in pediatric patients: radiographic and clinical characteristics. Am J Roentgenol 1998; 170: 1077–1078. 11. Conte P, Heitzman ER, Markarian B. Viral pneumonia. Roentgen pathological correlations. Radiology 1970 May; 95(2): 267–272. 12. Donnelly LF. Practical issues concerning imaging of pulmonary infection in children. J Thorac Imag 2001; 16: 238–250. 13. Leatherman JW, Iber C and Davies SF. Cavitation in bacteremic pneumococcal pneumonia. Am Rev Respir Dis 1984; 129: 317–321. 14. Donnelly LF and Klosterman LA. Cavitary necrosis complicating pneumonia in children: sequential findings on chest radiography. Am J Roentgenol 1998; 171: 253–256. 15. Leung AN, Muller NL, Pineda PR and FitzGerald JM. Primary tuberculosis in childhood: radiographic manifestations. Radiology 1992; 182: 87–91. 16. Peiris JS, Chu CM, Cheng VC, Chan KS, Hung IF, Poon LL, Law KI, Tang BS, Hon TY, Chan CS, Chan KH, Ng JS, Zheng BJ, Ng WL, Lai RW, Guan Y, Yuen KY; HKU/UCH SARS Study Group. Clinical progression and viral load in a community outbreak of coronavirus-associated SARS pneumonia: a prospective study. Lancet 2003 May 24; 361(9371): 1767–1772. 17. Antonio GE, Wong KT, Hui DS, Wu A, Lee N, Yuen EH, Leung CB, Rainer TH, Cameron P, Chung SS, Sung JJ, Ahuja AT. Thin-section CT in patients with severe acute respiratory syndrome following hospital discharge: preliminary experience. Radiology 2003 Sep; 228(3): 810–815. Epub 2003 Jun 12. 18. Amorosa NM, Genieser NB, Rocke KJ et al. Feasibility of highresolution, low-dose chest CT in evaluation of the pediatric chest. Pediatr Radiol 1994; 26: 6–10. 19. Lucaya J and Le Pointe HD. High-resolution CT of the lung in children. In: J Lucaya and JL Strife (Eds) Pediatric Chest Imaging. Springer, Berlin 55–91. 20. Wong KT, Antonio GE, Hui DS, Lee N, Yuen EH, Wu A, Leung CB, Rainer TH, Cameron P, Chung SS, Sung JJ, Ahuja AT. Thin-section CT of severe acute respiratory syndrome: evaluation of 73 patients exposed to or with the disease. Radiology 2003 Aug; 228(2): 395–400. Epub 2003 May 08.

Imaging of SARS in North America NL Müller and H Shulman

Introduction Clinical manifestations of SARS in North America Radiographic manifestations of SARS in North America

131

CHAPTER

13 High-resolution CT findings of SARS in North America

136

132 132

Introduction The first diagnosis of severe acute respiratory syndrome (SARS) in North America was made in Toronto on 13 March 2003 [1]. This was the first recognized case of the disease outside of Asia [2]. By 2 June, 198 cases of probable SARS were reported in Canada, 30 (15%) of whom had died [3]. In the US at the same time 66 cases of probable SARS had been reported, none of whom had died [3]. The first patient with SARS in North America was a 78-year-old woman who returned home to Toronto on 23 February 2003 after a visit to relatives in Hong Kong [1]. Two days later she developed fever, myalgia, sore throat and mild non-productive cough. Five days later she developed increasing cough and dyspnoea. She died 3 days later, on 5 March, at home, 9 days after the onset of her illness. The diagnosis of SARS was only made in retrospect. The index patient’s 43-year-old son developed fever and sweating on 27 February, 2 days after his mother first noted the symptoms [1]. He subsequently developed non-productive cough, chest pain, and dyspnoea and eventually high fever (temperature, 39.8°C).

A chest radiograph revealed bilateral lower lobe consolidation. He was admitted to the hospital with a diagnosis of community-acquired pneumonia. On the 2nd day after admission, he developed respiratory failure, was intubated and received mechanical ventilation. He died on 13 March 2003, 15 days after becoming ill. On 8 and 9 March, because of concern about possible tuberculosis in the family, the remaining five adult family members and their three children, who had all been exposed to the index patient, underwent screening chest radiography [1]. All had fever, cough or dyspnoea, and abnormal chest radiographs, except for the three children and one of the adults. One of the adults met the criteria for suspected SARS, and three met the criteria for probable SARS. All four were admitted to the hospital, three of them to intensive care units (ICUs); one patient required mechanical ventilation. Unfortunately, as spread to contacts had already occurred before the first patients presented to hospital, the SARS outbreak spread rapidly. By 31 March, contact tracing had identified an additional 100 patients as having probable or suspected SARS in the Greater Toronto area. Transmission was limited

132

I M A G I N G

to close contacts of patients, including household members, health care workers or other patients who were not protected with contact or respiratory precautions. Furthermore, case-finding measures identified additional individuals who had developed SARS after returning from travel to areas in Asia where there had been documented transmission of SARS. The only other cluster of cases in North America occurred in Vancouver. The first patient was a 55-year-old man who presented to the Emergency Department at Vancouver General Hospital on 7 March 2003 with a history of recent travel from Hong Kong and symptoms of pneumonia [2]. He had arrived from Hong Kong on 6 March. He had little contact outside of the immediate family. He was provided with a mask within minutes after arriving at the emergency department. Shortly thereafter, he was admitted into full-respiratory isolation. Owing to these circumstances he did not spread the disease. Subsequently only three more cases were observed in Vancouver, two in individuals with recent travel to Asia and one in a health care worker exposed to one of these patients.

Key points Initial experience: North America

• Case clusters had traceable contact with

O F

S A R S

I N

N O R T H

A M E R I C A

suspected or probable SARS in the Greater Toronto area [4]. The study included patients who had fever, a known exposure to SARS, and respiratory symptoms or parenchymal abnormalities seen on chest radiograph. Patients were excluded if an alternative diagnosis was determined. Of the 144 patients, 111 (77%) were exposed to SARS in the hospital setting. The most common clinical manifestations included fever (99%), non-productive cough (69%), myalgia (49%) and dyspnoea (42%). Common laboratory features included elevated LDH (87%), hypocalcaemia (60%) and lymphopaenia (54%). Twenty-nine patients (20%) were admitted to the ICU with or without mechanical ventilation and eight patients died (21-day mortality, 6.5%). Multivariable analysis showed that the presence of diabetes and other co-morbid conditions were independently associated with poor outcome. All four patients with proven SARS in Vancouver had temperature greater than 100.4°F (greater than 38°C) and one or more clinical findings of respiratory illness including cough, shortness of breath, difficulty breathing or hypoxia. All had lymphopaenia (absolute lymphocyte count was less than 1000/mm3) and elevated serum liver transaminases (aspartate aminotransferase and alanine aminotransferase). One of the four patients required mechanical ventilation. All four patients survived.

Hong Kong or Asia.

Key points

Clinical manifestations of SARS in North America The clinical manifestations of the first 10 patients with SARS in North America, including nine from Toronto and one from Vancouver were described by Poutanem et al. [1]. The patients ranged from 24 to 78 years in age. The main presenting symptoms included fever (in 100% of cases), non-productive cough (in 100%) and dyspnoea. Commonly seen laboratory findings included lymphopaenia (in 89% of cases), elevated lactate dehydrogenase (LDH) levels (in 80%) and elevated aspartate aminotransferase levels (in 78%). Five of the 10 patients required mechanical ventilation and three died. Booth et al. reviewed the clinical findings and shortterm outcomes of 144 patients with a diagnosis of

• Clinical presentation: fever, non-productive cough, myalgia and dyspnoea.

• Laboratory findings: increased LDH, hypocalcaemia and lymphopaenia.

Radiographic manifestations of SARS in North America The radiographic findings of SARS at presentation include unilateral or bilateral areas of consolidation (Figure 13.1) or poorly defined hazy increased opacities without obscuration of underlying vascular margins (ground-glass opacities) (Figure 13.2) [5–7]. These findings may involve any or all lung zones and be random in distribution, but tend to involve

R A D I O G R A P H I C

M A N I F E S T A T I O N S

mainly the lower lung zones and the outer third of the lungs. In a small percentage of symptomatic patients the chest radiograph may be normal at presentation, but unilateral or bilateral consolidation is usually evident in these patients on radiographs performed 24–48 hours later [6–8]. In patients who present with focal consolidation, the consolidation may remain unchanged for several days and then clear [8]. However, more commonly the consolidation remains focal, but increases in extent and then gradually clears [8]. In patients with more severe symptoms the consolidation can progress

Fig. 13.1 A 29-year-old man with SARS. Chest radiograph at presentation demonstrating patchy bilateral areas of consolidation.

(a)

O F

S A R S

133

to multi-focal patchy or confluent bilateral consolidation. Patients with multi-focal unilateral or bilateral disease at presentation often develop more extensive disease after admission and tend to have a more protracted clinical course (Figure 13.3) [8]. Several groups of investigators have reviewed the radiographic manifestations of SARS seen in Toronto and Vancouver [1,7–10]. Grinblat et al. reviewed the radiographic findings in 40 patients with SARS seen at a University of Toronto teaching hospital [8]. The patients ranged from 17 to 73 years of age; 55% were female. Fifty-eight per cent (23/40) of patients presented with unilateral or bilateral areas of consolidation and 42% (17/40) presented with a normal chest radiograph. All patients with initially normal radiograph developed focal unilateral (12/17, 71%) or bilateral consolidation (5/17, 29%) within 24–48 hours. Overall, the mean duration from the date of exposure to the first abnormal chest radiograph in the 40 patients was 12 days (range 4–26). The mean time period from onset of fever to an abnormal radiograph was 5 days (range 1–19) [8]. Twenty of the 40 (50%) patients in the study by Grinblat et al. had focal consolidation and 20 had multi-focal or bilateral consolidation (Figure 13.4) [8]. In all cases the consolidation had poorly defined margins. In 26 (52%) patients the consolidation had a predominantly peripheral distribution; in the remaining cases there was no apparent central or peripheral predominance. In 70% (28/40) of patients

(b)

Fig. 13.2 A 29-year-old woman with SARS. (a) Chest radiograph on admission showing ill-defined hazy increased density (ground-glass opacity) in right middle lung zone. (b) Chest radiograph 24 hours later showing dense focal consolidation (reprinted with permission from the American Journal of Roentgenology from Ref. [7]).

134

I M A G I N G

(a)

O F

S A R S

I N

N O R T H

A M E R I C A

(b)

Fig. 13.3 A 64-year-old woman with SARS. (a) Initial chest radiograph showing area of consolidation in right perihilar region and ground-glass opacities in right middle and lower lung zones. (b) Chest radiograph performed the following day demonstrating extensive consolidation in right lung and focal consolidation in left lung (reprinted with permission from the American Journal of Roentgenology from Ref. [7]).

(a)

(b)

Fig. 13.4 A 44-year-old man with SARS. (a) Chest radiograph at hospital admission showing focal consolidation in the region of the right costophrenic sulcus. (b) Chest radiograph 3 days later demonstrating increased right lower lobe consolidation and patchy areas of consolidation in the left upper and lower lobes.

the consolidation involved the middle or lower lung zones, and 30% (12/40) the middle or upper lung zones. Seventy-five per cent of the patients (15/20) with focal consolidation as the initial abnormality did not worsen or cleared completely on subsequent radiographs. The remaining 25% (5/20) of patients with focal consolidation progressed to bilateral disease within a mean of 2.2 days (range 1–4 days) of the initial focal findings (Figure 13.4). Patients with

multi-focal unilateral or bilateral disease at presentation often developed more extensive disease after admission and frequently had a protracted clinical course [8]. Radiographs in patients with residual disease after several weeks often show a predominantly reticular pattern suggestive of fibrosis (Figure 13.5). Nicolaou et al. reported the radiographic and computed tomography (CT) findings in the first case of

R A D I O G R A P H I C

M A N I F E S T A T I O N S

(a)

O F

S A R S

135

(b)

Fig. 13.5 A 62-year-old man with SARS. (a) Chest radiograph performed 6 days after hospital admission showing extensive bilateral consolidation. (b) Chest radiograph performed 7 weeks after hospital admission demonstrating coarse reticular opacities involving mainly the middle and upper lung zones. A tracheostomy tube is in place.

SARS seen in Vancouver [9]. The patient was a 55year-old previously healthy man who had travelled to Hong Kong and presented with a 2-week history of fever, headache, malaise, dyspnoea and cough. Bedside, computed radiograph demonstrated extensive bilateral ground-glass opacities and dependent areas of consolidation (Figure 13.6). Müller et al. reviewed the radiographic findings in 12 patients with SARS, including five from Vancouver and seven from Hong Kong [7]. The main radiographic findings at presentation consisted of unilateral or bilateral ground-glass opacities (n
5), focal unilateral or bilateral areas of consolidation (n
5) and diffuse small nodular opacities (n
1) (Figures 13.2 and 13.3). In one patient the admission chest radiograph was normal. In one other patient the chest radiograph had been prospectively interpreted as negative, although in retrospect subtle bilateral ground-glass opacities could be seen. Ground-glass opacities were bilateral, extensive and fairly symmetric in three patients, limited to one lung in two patients, limited to the lower lobes in one patient, and bilateral and asymmetric in one patient. The areas of consolidation involved mainly the upper lung zones in two patients, the lower zones in two patients and the middle lung zones in one patient [7]. Eight of the 10 patients who were hospitalized and had follow-up chest radiographs performed within 24 hours of presentation demonstrated progression of disease. In these eight patients follow-up radiographs

demonstrated extensive unilateral (n
2) or bilateral areas of consolidation (n
6) regardless of the initial radiographic pattern [7].

Key points Chest radiograph manifestations May be initially normal in a small number of patients

• Opacities

•

– Unilateral of bilateral areas of consolidation or – Poorly defined haze increased opacities without obscuration of underlying vascular margins (ground-glass opacities). Location of lesions – Random involvement – Mainly the lower lung zones and the outer third of the lungs.

Radiographic progression

• Focal consolidation at presentation – Consolidation increases in extent and then gradually clears – Patients with more severe symptoms, may progress to multi-focal patchy or confluent bilateral consolidation.

136

I M A G I N G

• Multi-focal disease at presentation – Often develop more extensive disease after admission and tend to have a more protracted clinical course.

O F

S A R S

I N

N O R T H

A M E R I C A

High-resolution CT Findings of SARS in North America The high-resolution CT (HRCT) manifestations of SARS consist of unilateral or bilateral ground-glass

(a)

(b)

(c)

(d)

Fig. 13.6 A 55-year-old previously healthy man with recent travel history to Hong Kong. (a) Bedside anteroposterior computed radiograph obtained with patient upright showing extensive bilateral ground-glass opacities and poorly defined nodular pattern. The abnormalities are diffuse in the right lung, but the radiograph showing relative sparing of the left lung apex. Mild airspace consolidation is seen in the retrocardiac region of the right lower lobe. Also noted is mild cardiomegaly. (b) Bedside antero-posterior computed radiograph obtained with the patient supine 12 hours after the initial radiograph (a) showing diffuse bilateral airspace consolidation. Note prominent air-bronchograms, low position of endotracheal tube and gaseous distension of the stomach. The radiographic findings and the rapid progression are consistent with adult respiratory distress syndrome (ARDS). (c) CT image (5-mm collimation) obtained at the level of the right upper lobe bronchus showing diffuse bilateral areas of ground-glass attenuation and dependent areas of consolidation. (d) CT image at the level of the lower lobe bronchi, showing similar findings to those in (b) (reprinted with permission from the American Journal of Roentgenology from Ref. [9]).

H I G H

-

R E S O L U T I O N

C T

F I N D I N G S

opacities and/or unilateral or bilateral areas of consolidation (Figure 13.7) [7,11,12]. Smooth thickening of interlobular septa and smooth intralobular lines are often present superimposed on areas of

(a)

(b)

Fig. 13.7 A 30-year-old woman with SARS. (a) Chest radiograph showing haziness (ground-glass opacity) over the lateral aspect of the right lung base with some loss of definition of the right hemidiaphragm. (b) HRCT demonstrating bilateral focal areas of consolidation.

O F

S A R S

137

ground-glass attenuation. The parenchymal abnormalities tend to involve mainly the lower lobes. The findings can be diffuse or random in distribution but tend to involve mainly the outer third of the lungs [11,12]. In all cases reported so far HRCT has been abnormal at initial presentation even when concurrent radiographs are normal or show only questionable abnormalities (Figures 13.7 and 13.8) [7,11,12]. Furthermore, in the majority of patients HRCT shows more extensive disease than apparent on the radiograph [12]. Müller et al. reviewed the initial CT findings seen at presentation in five patients from Vancouver and seven from Hong Kong [7]. One of the patients had normal chest radiograph, one had radiograph prospectively interpreted as normal but in retrospect demonstrated bilateral ground-glass opacities, two had bilateral ground-glass opacities and one had focal consolidation seen on the radiograph. HRCT in one patient with normal chest radiograph showed patchy bilateral areas of ground-glass attenuation and focal consolidation in the superior segment of the left lower lobe. HRCT in the three patients with ground-glass opacities on the radiograph demonstrated extensive bilateral areas of ground-glass attenuation (n
3) and small focal areas of consolidation (n
2) (Figure 13.9). The areas of consolidation involved mainly the dorsal lung regions. HRCT in the remaining patient showed focal consolidation in the right lower lobe. In a subsequent study, Müller et al. reviewed the HRCT findings in 29 patients with SARS, including four from Vancouver and 25 from Hong Kong [12]. The patients included 16 men and 13 women ranging from 25 to 82 years of age. Twelve of the 29 patients had CT performed at presentation or within 12 hours after hospital admission. All patients had parenchymal abnormalities on initial HRCT, including eight patients who had normal concurrent chest radiographs. The predominant HRCT findings at presentation consisted of unilateral (n
6) or bilateral (n
2) ground-glass opacities or focal unilateral (n
2) or bilateral (n
2) areas of consolidation. Four of the 12 (33%) patients had associated mild thickening of the interlobular septa within the areas of ground-glass attenuation or adjacent to areas of consolidation. The abnormalities involved predominately or exclusively the lower lung zones in five patients, the middle lung zones in five patients and upper lung zones in two patients. A predominantly

138

I M A G I N G

O F

S A R S

I N

N O R T H

A M E R I C A

(a)

(b)

(c)

subpleural distribution was evident on HRCT in eight patients, a patchy random distribution in three, and diffuse abnormalities in one patient. None of the patients had had evidence of hilar or mediastinal lymphadenopathy or pleural effusion at presentation. Twenty-five patients in the study by Müller et al. had HRCT scans performed 2–27 days after hospital admission (median 9 days) [12]. The predominant HRCT findings in hospitalized patients consisted of

Fig. 13.8 A 66-year-old woman with SARS. (a) Chest radiograph was interpreted by the emergency room physician as being normal. The radiologist reviewing the radiograph the following morning noted a vague opacity overlying the anterior right second rib and recommended HRCT for further evaluation. (b) and (c) HRCT images of the right upper lobe demonstrating focal area of consolidation, ground-glass opacities, mild septal thickening and smooth intralobular lines.

unilateral (n
2) or bilateral ground-glass opacities (n
13) or unilateral (n
2) or bilateral consolidation (n
5), or a mixed bilateral pattern of groundglass attenuation, consolidation and reticulation (n
3). Common findings seen in association with ground-glass opacities included mild smooth thickening of interlobular septa and smooth intralobular lines. Reticulation with associated irregular interfaces, architectural distortion and mild traction bronchiectasis was present in eight patients (32%).

H I G H

-

R E S O L U T I O N

C T

F I N D I N G S

O F

S A R S

139

Other findings seen on follow-up HRCT included pneumothorax (n
2), pneumo-mediastinum (n
3) and small pleural effusions (n
2).

(a)

(b)

The unilateral and bilateral areas of consolidation seen on HRCT in patients with SARS are similar to those seen in a variety of bacterial, fungal and viral pneumonias [11]. The findings in patients with SARS differ, however, from those described in other viral pneumonias, by the absence of centrilobular nodular opacities. Centrilobular nodules and branching centrilobular opacities resulting in a ‘treein-bud’ pattern are commonly seen in patients with bacterial, viral and mycoplasma pneumonia [13–15]. Reittner reviewed the HRCT findings in 114 patients with different types of pneumonia [13]. In their study seven of nine (78%) patients with viral pneumonia had centrilobular nodules. Septal thickening was seen at presentation in three (33%) patients with viral pneumonia in the study by Reittner et al. and in three of twelve (25%) patients with SARS in the current study. In patients with a protracted clinical course a reticular pattern is commonly seen on the radiograph and HRCT 2 weeks or more after hospital admission (Figure 13.10) [6,8,11,12]. On HRCT the reticular pattern is often associated with irregular interfaces and mild traction bronchiectasis. These findings suggest the presence of fibrosis [16]. However, longterm follow-up will be required to determine whether these changes resolve over time or whether they represent irreversible fibrosis.

Key points HRCT manifestations

• Location of lesions (c)

Fig. 13.9 A 48-year-old man with SARS. (a) Chest radiograph performed 12 hours after hospital admission showing subtle bilateral ground-glass opacities with relative sparing of left upper lobe. Radiographic findings were similar to those seen just prior to admission. (b) and (c) HRCT performed concomitant with chest radiograph demonstrating extensive bilateral ground-glass opacities (reprinted with permission from the American Journal of Roentgenology from Ref. [7]).

•

– Unilateral or bilateral distribution – Lower lobe predominance – Diffuse or random in distribution – Mainly the outer third of the lungs. Lesion appearance – Ground-glass opacities and/or consolidation – Smooth thickening of interlobular septa and smooth intralobular lines, often present superimposed on ground-glass opacities

140

I M A G I N G

O F

(a)

(b)

(c)

(d)

S A R S

I N

N O R T H

A M E R I C A

Fig. 13.10 A 48-year-old woman with SARS. The admission chest radiograph was normal. The radiographs remained normal until 4 days after admission when minimal consolidation was evident. The consolidation progressed rapidly; the patient developed respiratory failure and required intubation and mechanical ventilation. (a) Chest radiograph performed 7 days after hospital admission showing extensive bilateral consolidation. (b) Chest radiograph performed 1 month after hospital admission showing small foci of consolidation and extensive bilateral coarse reticular pattern. (c) HRCT at the level of the left main bronchus demonstrating bilateral ground-glass opacities, irregular linear opacities and distortion of architecture. (d) HRCT at the level of the right inferior pulmonary vein demonstrating predominantly right-sided abnormalities.

– Absence of centrilobular nodular opacities or a ‘tree-in-bud’ pattern – No hilar or mediastinal lymphadenopathy or pleural effusion at presentation.

• Pneumothorax • Pneumo-mediastinum • Small pleural effusion Sensitivity of HRCT

Follow-up HRCT findings

• Reticulation with irregular interfaces, architectural distortion and mild traction bronchiectasis (32%)

• Abnormal on patients with normal initial radiograph

• Showed more extensive disease than apparent on the radiograph

R E F E R E N C E S

References 1. Poutanen SM, Low DE, Henry B et al. Identification of severe acute respiratory syndrome in Canada. New Engl J Med 2003; 348(20): 1995–2005. 2. Patrick DM. The race to outpace severe acute respiratory syndrome (SARS). Can Med Assoc J 2003; 168: 1265–1266. 3. World Health Organization. Cumulative number of reported probable cases of SARS. http://www.who.int/csr/sars/ country/2003_06_02/en/ (accessed 2 June 2003). 4. Booth CM, Matukas LM, Tomlinson GA et al. Clinical features and short-term outcomes of 144 patients with SARS in the Greater Toronto area. J Am Med Assoc (Epub ahead of print) Available at http://www.ncbi.nlm.nih.gov/entrez/query.fcgi? cmd
Retrieve&db
PubMed&list_uids
12734147& dopt
Abstract (accessed 6 May 2003). 5. Lee N, Hui D, Wu A et al. A major outbreak of severe acute respiratory syndrome in Hong Kong. New Engl J Med (Epub ahead of print) Available from: URL: http://content. nejm.org/ cgi/reprint/NEJMoa03068 5v2.pdf (accessed 14 April 2003). 6. Wong KT, Antonio GE, Hui DS et al. Severe acute respiratory syndrome: radiographic appearances and pattern of progression in 138 patients. Radiology (Epub ahead of print) from URL: http://radiology.rsnajnls.org/cgi/content/full/ 2282030593v1 (accessed 20 May 2003). 7. Muller NL, Ooi GC, Khong PL, Nicolaou S. Severe acute respiratory syndrome: radiographic and CT findings. AJR Am J Roentgenol 2003 Jul; 181(1): 3–8. 8. Grinblat L, Shulman H, Glickman A, Matukas L, Paul N. Severe acute respiratory syndrome: radiographic review of 40 probable cases in Toronto, Canada. Radiology 2003 Sep; 228(3): 802–809.

141

9. Nicolaou S, Al-Nakshabandi NA and Muller NL. SARS: imaging of severe acute respiratory syndrome. Am J Roentgenol 2003; 180: 1247–1249. 10. Nicolaou S, Al-Nakshabandi NA and Müller NL. Images in clinical medicine. Radiologic manifestations of severe acute respiratory syndrome. New Engl J Med 2003; 348: 2006. 11. Wong KT, Antonio GE, Hui DS et al. Thin-section CT of severe acute respiratory syndrome: evaluation of 74 patients exposed to or with the disease. Radiology (Epub ahead of print) from: URL: http://radiology.rsnajnls. org/cgi/content/full/2283030541v1 (accessed 8 May 2003). 12. Muller NL, Ooi GC, Khong PL, Zhou LJ, Tsang KW, Nicolaou S. High-resolution CT findings of severe acute respiratory syndrome at presentation and after admission. AJR Am J Roentgenol 2004 Jan; 182(1): 39–44. 13. Reittner P, Ward S, Heyneman L, Johkoh T and Müller NL. Pneumonia: high-resolution CT findings in 114 patients. Eur Radiol 2003; 13: 515–521. 14. Gruden JF, Webb WR, Naidich DP and McGuinness G. Multinodular disease: anatomic localization at thin-section CT – multireader evaluation of a simple algorithm. Radiology 1999; 210: 711–720. 15. Collins J, Blankenbaker D and Stern EJ. CT patterns of bronchiolar disease: what is ‘tree-in-bud’? Am J Roentgenol 1998; 171: 365–370. 16. Remy-Jardin M, Giraud F, Remy J, Copin MC, Gosselin B and Duhamel A. Importance of ground-glass attenuation in chronic diffuse infiltrative lung disease: pathologic–CT correlation. Radiology 1993; 189: 693–698.

Radiographers’ Perspective in the Outbreak of SARS SSY Ho

Introduction Modification of routine radiographic practice Emotional impact of SARS outbreak on radiographers

143

CHAPTER

14

Radiographers’ perspective in the outbreak of SARS

148

144 146

Introduction Heath care workers are routinely exposed in a variety of diseases during their daily work. Although most health care workers would consider this as a part of their job, the general public is unaware of the risks taken by health care workers and the sacrifices they routinely make. The arrival of severe acute respiratory syndrome (SARS) changed that. Attention was focused on the health care workers since prior to SARS there was no precedent for an infectious disease that struck down hundreds of health care workers in a short period of time in a hospital setting. Hospitals were suddenly considered to be dangerous places to be in. As very little was known about this disease there was significant fear and anxiety among the hospital staff, and emotional disturbances in health care workers was inevitable. During the SARS crisis, radiographers were indispensable in imaging patients for the diagnosis, management and follow-up of SARS patients. Similar to other medical staff, frontline radiographers needed to

overcome their emotional stress and apprehension in order to deliver high-quality radiographic service to the SARS patients efficiently and safely. Radiographic practice had to be modified to minimize the risk of cross infection within the hospital and the radiology department. At short notice, radiographers at all levels of seniority had to familiarize themselves with infection control measures and the correct use of personal protective equipment within the radiology department, the hospital and other departments they served.

Key points Introduction During the SARS crisis, radiographers were

• • • •

indispensible in imaging patients; required to overcome emotional stress; needed to modify radiographic practice; required to familiarize with infection control measures.

144

Modification of routine radiographic practice Plain chest radiography and computed tomography (CT) of the thorax were the two major services requested for patients with suspected or confirmed SARS.

Plain radiography During the SARS epidemic, plain chest radiography was primarily performed for three categories of patients: patients with clinical signs and symptoms suspected of or confirmed with SARS, hospital staff and their family members for excluding SARS, patients recovered from SARS for follow-up. In routine practice, plain radiography is undertaken primarily in the radiology department. Bedside or portable radiography is provided in the wards for patients who are clinically unfit to be brought down to the radiology department. During the SARS outbreak, this practice was changed to cover both non-ambulatory patients and ambulatory patients with suspected or confirmed SARS to reduce the risk of cross infection within the hospital. Plain radiography was hence performed at sites outside the radiology department. Satellite X-ray units utilizing portable X-ray machines, lead screens and chest stands were set up at different sites within the hospital to provide safe and readily accessible chest radiography for different categories of patients (Figure 14.1). The satellite X-ray units were set up in the vicinity of the SARS wards and in clinics where patients’ waiting area was spacious with good ventilation. Meticulous attention was paid to radiation protection and infection control measures. To ensure smooth running of the service in the satellite X-ray units, senior radiographers co-ordinated the radiographic service with the involvement of the medical officers in charge of SARS patients, the nursing staff in the SARS wards and clinic, and the hospital administration. They also undertook duties in the SARS wards so that they could respond rapidly to problems and changes arising from the service. As the demand of plain chest radiography was high during the SARS outbreak and the service was scattered in different locations within the hospital, radiographer’ workforce was reorganized to cover

R A D I O G R A P H E R S

’

P E R S P E C T I V E

duties in these locations. This was possible as during the outbreak most routine clinical and radiological services were suspended. All radiographers in the department took turn to perform their duties in the satellite X-ray units for SARS patients. This duty arrangement was aimed to avoid overloading or overstressing radiographers or team of radiographers that normally performed general and portable radiography. Pregnant female radiographers were exempted from all duties that possessed any potential risk of SARS and the hospital management granted them special early maternity leave till the 14th week of their gestation. All student radiographers were immediately withdrawn from clinical attachment early during the course of the outbreak. Some radiographers chose to quarantine themselves during and after their rotation through the SARS duty to reduce the possibility of transmitting the infection to their families. The early notification of the SARS duty roster to all frontline radiographers helped them to prepare and make necessary arrangement for self-quarantine. It is absolutely essential for the staff to strictly follow infection control guidelines, particularly while performing bedside radiography in the SARS wards and intensive care units. The risk of contracting the disease is high in this environment as the patients are quite sick and there is close contact with the patients during the procedure. Therefore it was recommended that radiographers work in pairs when undertaking bedside radiography for SARS patients in order to smoothen the radiographic procedure and hence reduce patient contact time and stay in the wards. Another advantage of working in pairs was that they could remind each other of complying with stringent infection control measures during the procedure. Whenever possible, digital radiography should be used to minimize ‘retake’ due to inaccurate exposure settings. To ensure uninterrupted service in the satellite X-ray units, ample supply of film cassettes is necessary. Designated porters are necessary to transport contaminated exposed film cassettes back to the radiology department for decontamination before film processing, and to return the decontaminated unexposed film cassettes with processed films to the

M O D I F I A T I O N

O F

R O U T I N E

R A D I O G R A P H I C

P R A C T I C E

145

(a)

(b)

Fig. 14.1

(a) Satelitte X-ray unit in a cubicle next to (a) SARS ward and (b) SARS clinic.

satellite X-ray units. With this arrangement, chest radiographs taken for the SARS patients were promptly available for review by the medical officers during their ward rounds. During the outbreak, radiographers were encouraged to acquire basic knowledge of the radiographic

appearances of SARS and pay additional attention to all chest radiographs for any evidence of pneumonic change when they checked the films. Suspicious radiographs were sent for immediate reporting by radiologists and the patients managed appropriately.

146

R A D I O G R A P H E R S

Key points Modification of routine radiographic practice For plain radiography

• Bedside or portable radiography pro• • • • • • •

vided for both non-ambulatory and ambulatory SARS patients Satellite X-ray units set up near SARS wards and clinics Radiographers took turn to perform SARS duty and worked in pairs Film cassettes decontaminated before film processing Radiographers checked films carefully for evidence of pneumonic change Pregnant female radiographers granted with special early maternity leave Routine clinical and radiological services suspended Student radiographers clinical attachment withheld

CT Since the CT of the thorax is more sensitive than plain chest radiography to identify early pneumonic change in SARS, CT became an important diagnostic tool in patients with high index of clinical suspicion of SARS but no radiological abnormality on a chest radiograph [1]. In contrast to plain radiography, patients requiring high-resolution CT (HRCT) had to come to the radiology department for the procedure. Therefore, stringent infection control measures had to be implemented in CT service. Designated hours during the day and sessions outside office hours were arranged for SARS patients. During these sessions, radiographers put up warning signs outside the CT scanning suite and only restricted staff with proper personal protective apparel were permitted to the scanning room. Whenever possible, radiographers worked in pairs in the CT suite, one responsible for patient positioning in the scanning room and the other concentrating on manipulating the control panel inside the control room.

’

P E R S P E C T I V E

As CT service for SARS patients was also offered outside office hours, CT radiographers on evening or night shift have to familiarize themselves with the appropriate infection control measures and to provide assistance to the medical officers escorting the patients for CT scans in case of emergency. They also had to instruct and supervise staff for room cleansing after SARS sessions. The room is thoroughly cleaned and disinfected before it is opened for use for non-SARS patients.

Key points For CT

• Stringent infection control measures • • •

implemented Designated hours and sessions for suspected/confirmed SARS patients Radiographers in CT worked in pairs Thorough cleansing of CT room before open for use for non-SARS patients

Emotional impact of SARS outbreak on radiographers In the early days of the outbreak When faced with an infectious, potentially lethal, previously unknown disease, it is understandable that frontline health care workers would be under enormous stress leading to emotional disturbance/outbursts. At the time, we believed the best way to deal with this was to provide support and empathy but more importantly make sure that the staff were well equipped and protected in their working environment. Therefore the radiology management adopted and implemented a set of infection control measures with the help of the hospital infection control unit that were applicable to different radiological practices within the department. Radiographers were provided with infection control training with continual update of their knowledge on the infection control measures. Radiographic practice was closely monitored by the infection control officers in the department and amply supply of working

E M O T I O N A L

I M P A C T

O F

S A R S

O N

clothes and personal protective equipment was provided for each radiographer.

R A D I O G R A P H E R S

147

This created a vicious cycle leading to a significant drop in staff morale. Open and direct communication between senior radiographers and their frontline staff was especially essential at this time to address the immediate need and grievances of the frontline staff, and to convey important information from hospital management. Messages of appreciation from the general public expressing their gratitude to the frontline health care workers provided a boost to the sinking morale.

During the outbreak After a short period of anxiety and panic, radiographers were acclimatized to their precarious working environment and became more cautious but calm. They were fully aware of the significance of infection control measures and accepted that future radiographic practice would change. Teamwork with frontline staff of different disciplines was essential to combat the disease and at the same time it was important to pass useful feedback to their seniors to streamline service delivery.

Towards the end of the outbreak After a 3-month battle with SARS, things began to improve. As the number of cases fell, radiographers became relaxed and felt blessed they could survive this ‘battle’. They also realized they were in a unique position to significantly help the community and offer support to the patients who had unfortunately contracted the disease and had been quarantined in the hospital for a long time. This episode of SARS reminded the radiographers that their role is not only to provide radiographic service but also to demonstrate a caring attitude and offer comfort to the patients.

Due to the heavy workload and underlying emotional stress, more radiographers reported sick and were unable to work. Those who remained had to shoulder the additional burden in their workplace. The situation was further aggravated in some radiographers during the SARS outbreak, by selfimposed quarantine removing themselves from emotional and social support from family and friends that might have helped diffuse the stress.

Key points Emotional impact of SARS outbreak on radiographers SARS outbreak

Remedies

In the early days • Radiographers were emotionally disturbed

• To provide support and empathy • To provide infection control training and supply of personal protective equipment

• To closely monitor radiographic practice During • Radiographers were calm but cautious • Staff morale was sinking

• To have open and direct communication • To address immediate need and grievances •

Towards the end

• Radiographers were relaxed and realized their unique role in health care provision

of staff To convey important information from hospital management

148

R A D I O G R A P H E R S

Radiographers’ perspective in the outbreak of SARS Although our experience during SARS crisis was bitter and stressful, it was valuable and revealing. It stimulated us to re-evaluate our professionalism and our role in the health care provision. In the early days of the outbreak, it exposed our inadequacy in professional training, especially in the aspects of infection control and risk management. The difficult working conditions made us realize the importance of teamwork and good communication between frontline staff and senior radiographers. The radiographers’ commitment and professionalism during the SARS outbreak earned appreciation and recognition from the public and the community we serve. This increased our self-esteem as medical professionals. Most importantly, after the SARS outbreak, we have adopted a safer radiographic practice and are better equipped to face future challenges.

’

P E R S P E C T I V E

Key points Radiographers’ perspective in the outbreak of SARS After the outbreak, radiographers were

• stimulated to re-evaluate their professionalism and role;

• aware of their inadequacy in professional training;

• having more self-esteem as medical professionals;

• better equipped to face future challenges.

Reference 1.

Ho SSY, Chan PL, Wong PK, Antonio GE, Wong KT, Lyon DJ, Fung KSC, Li CK, Cheng AFB, Ahuja AT. Perspective. Eye of the storm: the roles of a radiology department in the outbreak of severe acute respiratory syndrome. Am J Roentgenol 2003; 181: 19–24.

Implementation of Measures to Prevent the Spread of SARS in a Radiology Department* AD King and ASC Ching

Introduction Infection control measures to be taken by staff Infection control measures to be taken by patients

149 150 151

Introduction Patients infected by severe acute respiratory syndrome (SARS) require imaging during the course of their disease. Plain radiography and computed tomography (CT) will be employed routinely, although during an epidemic all imaging modalities may be requested at some point. Infection control is concerned with protecting the individual against infection and containing an outbreak, at the same time as providing medical care for those patients with SARS. In a radiology department it is, of course, of utmost importance that the staff are protected, but for those centres that also have to maintain essential services for patients without SARS, there is the additional consideration of preventing cross *At the time of writing this chapter our department has been imaging patients with SARS for 4 months. Unfortunately, one member of our staff, a radiographer working for cardiology in the cardiac catheter laboratory, was in the first wave of medical staff to contract the disease. This occurred before we were aware of the very existence of SARS. Since that time with good fortune and the implementation of strict infection control measures no other staff in the radiology department has contracted the disease at work, and to our knowledge no patients have contracted SARS in the radiology department.

CHAPTER

15

General measures to be taken by managers in the radiology department 152 Infection control measures for specific modalities 155 Conclusion 157

infection between patients. When planning control measures against any infection, details of the mode of infection.Transmission must be taken into account, for the SARS coronavirus the important points to take into consideration are shown in Table 15.1. The format of this chapter may appear laborious,

Table 15.1

SARS virus.

1. Found in the respiratory secretions, saliva, blood, urine and faeces of patients. 2. Spread mainly by droplets and aerosolized respiratory secretions, droplet infection usually occurring within 3 feet of a patient. 3. Spread also through direct contact with patient’s secretions and excreta. 4. Relatively robust, surviving on surfaces sometimes for more than 24 hours, so that infection can then be transferred inadvertently from surfaces when staff touch their eyes, mouth and other mucosal areas. 5. Symptoms of SARS may be mild and nonspecific, especially in the elderly population. A high index of suspicion and stringent infection control measures should therefore be applied to all patients.

150

C O N T R O L

M E A S U R E S

O F

but it is designed to provide practical checklists covering some of the most important issues that need to be considered when preparing a department for the battle against infection.

S A R S

:

R A D I O L O G Y

D E P A R T M E N T

Cap Face-shield N95 mask

Infection control measures to be taken by staff Staff education

Lead coat

This is one of the most important aspects of infection control:

• All staff must be fully aware of the infection con-

• •

trol guidelines of the department and hospital, and should be trained in infection control measures before entering any high-risk area. Personal hygiene cannot be overemphasized, especially the obsessive washing of hands and avoidance of touching the mouth, eyes or masks [1]. There are general measures that can be taken to reduce infection. These include minimizing personal accessories brought to work, avoiding visits to other staff during imaging sessions, covering pagers with a disposable plastic bag and refraining from eating or drinking in scanning rooms. At meal breaks while eating and drinking staff should not talk and should face away from colleagues. Infection control measures need to be continued at home to reduce the potential risk of spreading infection in family members. These include changing clothes and taking a shower immediately on returning home, avoiding close contact with family members, wearing a mask, avoiding sharing food, eating utensils and towels, frequent hand washing for all the family using liquid soap. All staff are responsible for monitoring their own health closely. If they develop a febrile illness they must take immediate sick leave and attend a screening clinic in the hospital.

Gown Gloves

Fig. 15.1 Standard personal protective apparel for staff in the high-risk patient areas in a radiology department. Table 15.2

Personal protective apparel for staff.

Personal protective apparel

1. N95 mask (this must be tested to ensure it fits properly, a surgical mask is not acceptable in areas where there are respiratory aerosols). 2. Disposable isolation gown (water-proof or water-repellent gown according to the nature of activity and risk of exposure, gowns should be disposed of when soiled or when leaving high-risk areas). 3. Latex gloves to be changed between each patient (plus, of course, hand washing), ensuring there is no gap between the gloves and gown. 4. Disposable cap. 5. Visor/goggles/face-shield (for high-risk procedures goggles should be worn in addition to the face-shield). 6. Work shoes (covers are optional but shoes should not have laces).

• The amount of personnel protective apparel that is

Plus additional personal protection apparel according to risk.

•

•

worn will depend upon the level of risk, remembering that both under-protection and over-protection are hazardous. For contact with patients with suspected or confirmed SARS the personal protective apparel is shown in Figure 15.1 and Table 15.2.

• Changing into protective apparel should be done in designated areas using the correct sequence for ‘putting on’ and ‘taking off ’ personal protective apparel is shown in Figure 15.2. It is essential

I N F E C T I O N

C O N T R O L

M E A S U R E S

:

B Y

(b)

(a)

Fig. 15.2

• •

151

P A T I E N T S

1

2

1

2

3

4

3

4

5

6

5

6

7

8

7

8

Sequential steps for (a)’putting on’ and (b) ‘taking off’ personal protective apparel.

that meticulous care is taken when ‘taking off ’ the apparel to avoid self-contamination. Once the protective apparel is discarded it must be placed in a designated bag in a bin with a lid. High-dose contamination requires immediate decontamination in a shower.

Key points 1. Staff education for infection control at work and at home 2. Use of personal protective apparel Wash hands obsessively! Do not touch mouth, eyes or masks!

Infection control measures to be taken by patients The measures will be determined by patient’s risk group, for this purpose patients are divided into four groups: 1. Outpatients without suspected SARS – Use a questionnaire on arrival to screen for SARS, any patient with suspected SARS should have the examination postponed and asked to attend a screening clinic. – Instruct patients to wash their hands before and after attending the radiology department and wear a mask. 2. Outpatients with suspected SARS undergoing screening chest X-ray

152

C O N T R O L

Table 15.3 1. 2. 3. 4.

M E A S U R E S

O F

S A R S

Personal protective apparel for patients.

Key points

• Divide patients into different risk categories, screen outpatients and check the category has not changed before calling an inpatient to the department.

R A D I O L O G Y

D E P A R T M E N T

• Patients with SARS and suspected SARS

Clean gown or isolation gown Cap Surgical mask Hand washing is mandatory, gloves are optional

– Instruct patients to wash their hands before and after attending the radiology department and wear a mask. 3. Inpatients without suspected or confirmed SARS – All inpatients with a febrile illness should be suspected of having SARS and the procedures listed below should be taken. All other inpatients are potentially at risk of SARS because of cross infection within hospital and the level of protective measures will have to depend upon resources and risk of the procedure. – Before calling any non-SARS patient to radiology, the SARS status should be checked again to ensure it has not changed since the examination was requested. 4. Inpatients with suspected or confirmed SARS – Change the patient into the protective apparel shown in Table 15.3 before entering the radiology department. – Request the ward staff to site and remove intravenous lines for contrast injection on the ward and where possible obtain signed consent forms on the ward and fax to the department. Patients who require oxygen should have a nasal cannula rather than an oxygen mask and high-flow oxygen supply to these patients should be avoided. – Leave clinical records and packets of radiographs on the patient trolley and handle only if absolutely essential. – Minimize the chance of contracting the disease from patients, by keeping the time spent in direct contact to a minimum, the distance from a patient to a maximum and the number of working staff to a minimum, while maintaining good quality care.

:

•

should wear personal protective apparel and avoid wearing oxygen masks. Keep contact with SARS and suspected SARS patients to a minimum, site intravenous lines on the ward, fax signed consent forms to the department and avoid handling ward notes and X-ray packets.

General measures to be taken by managers in the radiology department There are many issues to take into account when reorganizing a radiology department and these do not always fit into neatly organized categories. Below is a list of some of the major points that need to be taken into consideration.

Patient segregation Some departments have to continue providing a radiology service to patients with non-SARS-related illnesses at the same time as imaging those patients with suspected or confirmed SARS. In this scenario patients have to be segregated according to risk and special consideration has to be given to non-SARS patients who are especially vulnerable to infection, such as patients who are pregnant, immunosuppressed or the newborn:

• The best way to achieve segregation is to site

•

equipment for imaging SARS patients away from the main radiology department, preferably close to the infectious disease wards or screening clinics. Some equipments, such as CT may not be moved and patients with SARS will have to attend the main radiology department. The only option in this circumstance is to segregate patients by time. The allocation of bookings should be from low to high-risk; outpatients without suspected SARS (low-risk), inpatients without a febrile illness or suspected/confirmed SARS (moderate risk because of risk of cross infection), inpatients with a febrile illness (high risk) and inpatients with suspected/confirmed SARS (ultra-high-risk).

G E N E R A L

M E A S U R E S

T O

B E

T A K E N

B Y

M A N A G E R S

153

• Waiting areas for patients have to be segregated

• Set up a contingency plan for a sudden surge

and clearly marked, where possible there should be different routes for access. Do not forget that patients with SARS need to be segregated on their way to the department by using designated lifts. In order for patient segregation to be successful, the different modalities in radiology must coordinate the time slots for the different categories of patients, and outpatients should be instructed to attend their appointments on time to ensure there is no overcrowding and no overlap with higher-risk patients. To prevent unnecessary waiting in the department, the examination room should be ready to accept high- and ultra-high-risk patients as soon as they arrive and an efficient portering system is required to ensure the patient returns to the ward promptly. Patients’ relatives or visitors should not be allowed in the department unless it is absolutely necessary.

in SARS patients and massive high-dose contamination.

• •

•

•

Designate areas for staff to change into personal protective apparel • Set up designated areas close to the scanning rooms.

• Keep the areas stocked with personal protective apparel and avoid contamination during storage.

• Standardize the layout of these areas and restrict • •

•

Staff education and enforcement of infection control measures

the number of staff changing at any one time to prevent overcrowding. Post instructions showing the correct sequence for ‘putting on’ and ‘taking off ’ apparel on the walls of the changing areas (Figure 15.2) [2–4]. Arrange bins for the disposal of contaminated and non-contaminated items; these bins should have a lid and be emptied regularly before they become full. Divide the department into clean areas (i.e. offices) and dirty areas (i.e. scanning rooms), and ensure that contaminated apparel is removed before entering clean areas.

• Set up a team (preferably including a manager,

•

•

•

nurse, radiographer, radiologist, clerk and cleaner) to introduce guidelines, educate staff, provide regular updates, monitor and audit measures. Appoint one team member as the infection control officer with overall responsibility for control within the department, as well as being the designated person to receive and implement hospital guidelines. Training and access to advice are of the utmost importance for infection control, all members of staff must be included. Emergency advice should be available 24 hours a day. Set up an effective method for disseminating information. Dissemination of information is particularly important as measures often change daily, especially at the beginning of an outbreak. E-mail is a good way to provide this information but many staff members may not have access, so noticeboards or a cascade system of word of mouth may need to be introduced. Encourage all staff to monitor their colleagues and correct any mistakes immediately.

Showering system • Ensure there is a showering system available

•

within the department for emergency use by any staff member who is contaminated by patient’s secretions or who has performed cardiopulmonary resuscitation. Equip with an emergency kit containing shampoo, soap, towels, a clean change of clothes and disposable bags for contaminated items.

Appointments • Decrease appointments to allow sufficient time to •

carry out all infection control measures and discontinue non-essential services. Hospitals that do not have electronic means for receiving examination requests should institute a fax system for requests. In this way clinicians are discouraged from visiting the department

154

•

C O N T R O L

M E A S U R E S

O F

and potentially contaminated request forms are not sent from the wards. All requests for patients with suspected or confirmed SARS must be clearly marked. Clinicians must be prudent in their requests ensuring that any examination for higher-risk patients will have a direct impact on patient management, and these examinations should be performed by experienced staff.

S A R S

• Those areas which have a high throughput of •

•

•

SARS patients, such as general radiography and CT, will need special attention. Cleaning can be facilitated by removing all unnecessary clutter and by drawing up a list of items, such as door handles, soap dispensers and computers, that may be easily overlooked in the cleaning process. Recommendations for cleaning are shown in Table 15.4. Provide adequate instruction and supervision of cleaners, especially in handling patient’s excreta and cleaning toilets. Items in a room that cannot be easily cleaned, such as soft furnishings and computers, can be removed or covered with disposable coverings. Couch linen needs to be changed regularly and always after high- and ultra-high-risk patients are scanned.

R A D I O L O G Y

D E P A R T M E N T

Table 15.4 General principles on cleaning and disinfection in the radiology department during a SARS outbreak.

• •

Cleaning and disinfection Complete cleaning and disinfection of the working environment and equipment is mandatory. As a result cleaning of all rooms and equipment in the department will need to be increased which will require extra manpower. This area must be given a high priority and if necessary non-cleaning staff may have to assist if there is a shortage of manpower:

:

• •

• •

Cleaners must receive full training in infection control and wear protective apparel and wash hands frequently. Cleaning will need to be more frequent and more thorough and should follow a schedule together with extra cleaning after a room has been used by a patient/batch of patients with SARS. After the examination of any patients with severe coughing, the room and especially nearby surfaces should be thoroughly disinfected before the next patient is examined. Before cleaning, one must ensure that the imaging machine and devices are suitable for general cleaning and disinfection. In addition to the usual surfaces that need cleaning there should be a list of items that require special attention and may be normally overlooked, i.e. handles, soap dispensers, telephone, cabinets, intercom, keyboards and mouse of computer workstations. Computers and workstations may have to be cleaned with special agents such as ethyl alcohol. Ventilation fans need regular cleaning.

Resuscitation Resuscitation is a very high-risk procedure!

• Fully stock the emergency resuscitation trolley

• •

with protective apparel. The resuscitator bags must have viral filters and the resuscitator bag-valvefilter mask should fit tightly to avoid air leakage. Arrange practice drills and ensure that all staff know they must be fully protected before starting resuscitation. Avoid intubation by inexperienced staff.

Staffing arrangements Key points

• Rotate staff through the high-risk areas to reduce

• •

their viral load but at the same time ensure that staff working in the higher-risk areas are experienced in working in this environment. Provide staff with adequate rest to prevent mistakes in infection control procedures. Pregnant staff should not be put in a high-risk environment.

1. Segregate different risk groups of patients – Preferably by relocating equipment to SARS areas, but if this is not possible segregate patients by time – Do not forget to segregate waiting areas and lift access – Restrict relatives and visitors

C O N T R O L

M E A S U R E S

F O R

S P E C I F I C

2. Staff education and enforcement of infection control measures – Set up an infection control team – System to disseminate information quickly and efficiently – Continual training, updating and monitoring 3. Designate areas for staff to change into personal protective apparel – Clear instructions for putting on and taking off apparel – System for disposal of contaminated items 4. Showers – Ensure showering system for highdose decontamination 5. Appointments – Reduce number of appointments – Fax system for request forms 6. Cleaning and disinfection – Increase resources 7. Staffing – Rotate staff through high-risk areas and allow sufficient rest – Move pregnant staff to low-risk areas 8. Resuscitation – Fully stocked emergency trolley with protective apparel and resuscitator bags with viral filters. Avoid intubation by inexperienced staff.

Infection control measures for specific modalities General radiography This is the area that will see the greatest increase in workload and will cause a great deal of stress and anxiety among radiographers in the initial stages of an outbreak. It is very important that staff are well educated in infection control measures before undertaking these duties and are provided with full personal protective apparel. Satellite X-ray rooms should be set up for screening and a specific room should be designated in the accident and emergency department for X-raying suspected cases. Equipment including portable

M O D A L I T I E S

155

X-ray machines will require frequent cleaning. There will be a great increase in the demand for portable X-rays on the SARS wards and on intensive care units (ICUs). Radiographers may have to spend prolonged periods of time in very close contact with SARS patients as they position the X-ray cassettes. Radiographers may also encounter practical difficulties including a limitation of space in which to manoeuvre the portable machines and a lack of cassettes. To prevent radiographers having to repeatedly go between the wards and the main department extra cassettes will have to be purchased. When cassettes are brought back to the department they must be protected by a disposable cover and after use all cassettes from the contaminated areas should be disinfected. Where possible two radiographers should be used: one ‘dirty’ radiographer to perform the positioning and one ‘clean’ radiographer to operate the control panels. During an outbreak all patients having a chest radiograph, irrespective of the provisional diagnosis, should be considered as high risk and procedures taken accordingly. There should be access to an instant reporting system for all patients with suspicious chest X-rays in screening clinics and routine radiography sessions. If the likelihood of SARS is high, then the room should be cleaned before the next patient is examined.

Key points

• High throughput of patients will be stress• • • •

ful for radiographers who will require priority in training Set up satellite X-ray equipment for screening and examining SARS patients Use ‘paired’ radiographers Increased demand for cassettes; cassettes require disinfection Access to instant film reporting service for screening films

CT This is the second modality that will encounter the greatest number of patients with SARS. Fortunately most patients require a high-resolution CT (HRCT) which is a quick examination that does not require intravenous contrast, so direct contact can be limited

156

C O N T R O L

M E A S U R E S

O F

to the short time it takes to position the patient on the scanner. One ‘clean’ radiographer should remain in the control room to operate the scanner while a second ‘dirty’ radiographer performs the positioning. Equipment that is not in use, such as the injection pump, should be moved away and covered. The room and the equipment should be thoroughly cleaned after the last batch of higher-risk patients.

S A R S

:

R A D I O L O G Y

D E P A R T M E N T

dressings or colostomy bags on the abdomen. When performing abdominal ultrasound deep breathing should to be avoided or performed with the patient turned on their side facing away from the sonographer. Gloves must be changed after each patient and hands washed. The transducer should be cleaned after each patient and covered with disposable covers for all inpatients.

Key points

Key points

• High throughput of patients with SARS, so

• High risk because of prolonged period of

• • • • •

priority should be given to training these staff HRCT is a quick investigation that requires no intravenous contrast, so contact time with the patient is short One CT scanner may have to scan both high- and low-risk patients, so patients will need to be segregated by time High input of resources and stringent measures for cleaning and disinfection Use ‘paired’ radiographers Cover equipment that is not being used such as the contrast pump

Ultrasound Ultrasound is potentially a high-risk procedure because of the prolonged contact with patients at close quarters. Some patients with a febrile illness may undergo abdominal ultrasound with a provisional diagnosis of a non-SARS-related illness such as a postoperative collection or cholangitis, only to be diagnosed with SARS later on. Therefore, a high index of suspicion should be maintained for ultrasound examinations. Rooms and ultrasound machines should be designated for high- and ultrahigh-risk patients and a portable machine should be left on the ICU. If equipment allows it may also be prudent to allocate a machine for non-SARS patients who are at particular risk of infection (obstetrics, neonatal unit or bone marrow transplant unit). The ultrasound examination should be kept as short as possible to answer the clinical question and if appropriate a CT scan should be considered as an alternative examination, especially if there are wound

• • • •

contact at close quarters, consider CT as an alternative High index of suspicion of SARS is required Designate US machines for different risk levels of patients Avoid asking patient to take deep breaths Disinfect transducers after each patient

Fluoroscopy/contrast studies Many of these examinations, such as a barium enema, small bowel enema, sialogram and dactocystogram carry a potentially high risk to staff and other patients. Where possible the examination should be avoided all together in those patients with suspected/confirmed SARS. Fortunately these examinations are rarely required for patients with SARS, although they may be performed inadvertently on patients in whom SARS is not initially suspected, especially the elderly with bowel problems. Therefore, the highest level of infection control should be taken in all cases and staff should ensure that their personal protective apparel includes a face-shield. There must be a designated place and procedure for safe disposal of all fluids.

Key points

• Most fluoroscopic examinations are high • •

risk due to close contact, and handling patients excreta and secretions System for safe disposal of bodily fluids Be aware that patients with SARS may present with diarrhoea

157

R E F E R E N C E S

Angiography and interventional radiology These are also high-risk procedures because of contact with patients’ fluids together with a prolonged time in close contact with the patient [5]. Therefore similar points to those for fluoroscopy above are applied. Where possible a dedicated room and ultrasound machine should be used for suspected and confirmed SARS cases.

Magnetic resonance imaging All metallic items and false teeth should be removed on the ward and the patient’s mask should not have a metallic bar. All inpatients with suspected or confirmed SARS should have the safety questionnaire filled out on the ward and faxed to the department, this will ensure that patients in whom magnetic resonance imaging (MRI) is contraindicated are not brought to the scanner. Once again two radiographers should be employed one ‘clean’ radiographer to stay in the control room and one ‘dirty’ radiographer to position the patient. The call bell must be covered with a disposable plastic bag. Potentially MRI possesses a higher risk of cross infection because of the prolonged period of time in which a patient lies in very close proximity to the equipment, therefore thorough cleaning of MRI machine, coils, injector and room must be performed especially following the examination of any inpatients.

Key points

• Fill out MRI safety questionnaire before arrival in MRI department

• The patient’s mask should not have a metallic bar

• Use ‘paired’ radiographers • Thorough disinfection of the MRI to prevent cross infection

aerosal-generating procedures such as pulmonary ventilation scans should not be performed. Unfortunately, recent information suggests that some patients with SARS may have pulmonary embolism or thromboembolic phenomena. In these cases the perfusion study alone may be insufficient because it can show defects due to the pneumonia. Therefore the role of nuclear medicine may be limited for the investigation of suspected pulmonary embolism and other imaging modalities will have to be considered such as contrast-enhanced helical CT.

Key points

• Avoid ventilation lung scan using aerosalgenerated procedures

• Modify diagnostic algorithm for acute pulmonary embolism

Conclusion Even in the most difficult of circumstances it is possible to run a radiology service during an outbreak of SARS while protecting staff and preventing cross infection between patients. However, it requires meticulous planning, great effort to maintain high vigilance and a very strict adherence to infection control measures. As we continue to increase our knowledge of how the infection spreads and improve our understanding of SARS, we are able to better tailor our infection control measures.

Acknowledgements We would like to acknowledge the input from our whole department, our hospital infection control team, the management at the Prince of Wales Hospital, and the Hong Kong Health Authority who issued guidelines for us to follow (which were posted on their web page) and gave their support in these difficult times.

Nuclear medicine

References

Staff should take extra precautions in handling patients with SARS and in particular

1. King AD, Ching AS, Chan PL, Cheng AY, Wong PK, Ho SS, Griffith JF, Lyon DJ, Fung KS, Choi P, Li CK, Cheng AF and Ahuja AT. Severe acute respiratory syndrome: avoiding the

158

C O N T R O L

M E A S U R E S

O F

spread of infection in a radiology department. Am J Roentgenol 2003;181(1): 25–27. 2. Department of Diagnostic Radiology and Organ Imaging page. The Chinese University of Hong Kong web site. Available at: www.droid.cuhk.edu.hk (accessed 13 June 2003). 3. Hong Kong Hospital Authority web page. Hospital Authority Guidelines on Severe Acute Respiratory Syndrome web site. Available at: http://ha.home/ho/ps/sars_infection_control.htm (accessed 20 June 2003).

S A R S

:

R A D I O L O G Y

D E P A R T M E N T

4. First data on stability and resistance of SARS coronavirus compiled by members of WHO laboratory network. Available at: http://www.who.int/csr/SARS/survival_2003_05_04/en/ind ex.html (accessed 30 June 2003). 5. Hong Kong SAR Government, Department of Health web page. Health advice: for people who have been in close contact with patients. Available at: http://www.info.gov.hk/dh (accessed 24 May 2003).

CHAPTER

Aftermath of SARS GE Antonio, JF Griffith and AT Ahuja

Introduction Audits

159 159

16

Post-SARS sequelae Conclusion

159 163

Introduction

The following recommendations were made:

This chapter briefly examines:

1. Effective surveillance, data collection and sharing 2. High level of awareness and implementation of effective infection control measures. 3. Rapid and comprehensive contact tracing. 4. Timely declaration and enforcement of isolation and quarantine measures.

• the weaknesses in the health care systems exposed •

by the severe acute respiratory syndrome (SARS) epidemic in Hong Kong [1], the sequelae of the disease.

Audits Two audit committees, set up by the Hong Kong Government and Hospital Authority respectively, reviewed the response of the health systems to the SARS epidemic [2]. The findings of both audits were similar. The main criticism were that: 1. Government agencies were not prepared for an outbreak of such magnitude. Procedural mechanisms were not in place beforehand so agencies were always trying to catch up. 2. Information about an unusual viral infection in early February 2003 only had the status of rumour and early ‘soft’ evidence was neglected. 3. Not enough was done early in the epidemic to alert other hospitals and health care workers of potential risks. And when information was disseminated it was not very clearly communicated.

Both the audits were comprehensive, clearly identified the deficiencies and made suitable recommendations for future improvement.

Key points Recommendations of audits

• Improved surveillance • Increased awareness • Quicker response

Post-SARS sequelae Following recovery from SARS, patients developed additional problems during convalescence related to the musculoskeletal system, adrenal insufficiency and psychological well-being.

160

Musculoskeletal problems Many patients experienced generalised muscular weakness, arthralgia and lethargy during the convalescent period. Towards the end of the March to June 2003 SARS epidemic, the authors were presented with their first post-SARS patient (a young female health care worker) with severe hip and knee joint pain that had developed during her hospital stay and persisted since hospital discharge. Magnetic resonance imaging (MRI) of the hips and knees revealed avascular necrosis of the femoral heads, femoral and tibial condyles and extensive intra-medullary bone infarcts in the femoral and tibial shafts.

A F T E R M A T H

O F

S A R S

steroids administered over a longer period to predisposed patients (patients with systemic lupus erythematosus, rheumatoid arthritis, malignancy or following organ transplantation) have a dose-related risk of osteonecrosis varying from 4% to 52% [7,8]. Using a multiple multinomial logistic regression model for the results, it emerged that the cumulative dose of corticosteroid was the most important risk factor for predicting osteonecrosis. In our experience, no osteonecrosis was observed in patients who received less than 3 g prednisolone equivalent,

A prospective study was therefore performed by the authors’ institution in September 2003. This showed avascular necrosis to be present in just under 5% of patients (all stage I disease using the University of Pennsylvania classification) at 6-month follow-up with MRI [3] (Figure 16.1). This bone necrosis may be a complication of the infection or of the treatment (particularly corticosteroids) or both. It may be difficult to prove which factor is the major or sole cause since corticosteroids were almost uniformly used to treat SARS patients. Autopsy results on SARS patients have found fibrin thrombi in pulmonary vessels or intimal swelling of pulmonary vessels [4]. It is conceivable that viralinduced vascular damage or thrombosis may occur in other parts of the body and if present in bones may partially explain the high incidence of avascular necrosis and bone infarcts in the SARS patients. Corticosteroids are postulated to induce osteonecrosis by a decrease in regional blood flow. Steroids may decrease blood flow by inducing (a) formation of lipid emboli and lipid-loaded fibrin–platelet thrombi which occlude the subchondral arterioles and capillaries; or (b) marrow fat hypertrophy which leads to increased intra-osseous pressure, compression of sinusoidal channels and impairment of venous flow [5]. It is known that high-dose steroids administered over a short period to patients otherwise not predisposed to osteonecrosis seem to confer little or no risk of osteonecrosis [6]. On the other hand, high-dose

Fig. 16.1 A 34-year-old female previously treated with steroids for SARS. Sagittal T2-weighted images of knee showing severe avascular necrosis involving the subchondral bone on the mid- to posterior aspect of the medial femoral condyle (small arrows). No articular surface collapse is present. There is also a medullary infarct in the distal femoral diaphysis and the posterior aspect of the medial tibial plateau (large arrows).

P O S T

-

S A R S

S E Q U E L A E

161

1. Subchondral bone marrow abnormality (28% of patients) which ranged from 3 to 26 mm (mean 6.5 mm) in size (Figure 16.2). Some of

these patients had associated joint degenerative change. The MRI appearances of these subchondral marrow abnormalities was comparable to that observed in severe degenerative disease [9] and in keeping with this likelihood they did tend to occur in a relatively older age group to those with normal MRI examinations or osteonecrosis. Another probable cause of subchondral marrow abnormality is early subchondral osteonecrosis. The distribution of the marrow abnormalities around the knee was similar to that observed with osteonecrosis. 2. Intra-medullary bone marrow abnormality (17% patients), which ranged in size from 3 to 20 mm (mean 5.9 mm) (Figure 16.3). The incidence of intra-medullary abnormalities in the normal population is not known. Although these lesions may potentially represent very small areas of osteonecrosis, they are of small size (mostly several millimetres) and by virtue or their location

Fig. 16.2 A 31-year-old female following steriod treatment for SARS. T1-weighted coronal image of both femora. There are medullary infarcts in the distal femoral diaphyses bilaterally (arrows). No subchondral avascular necrosis is evident.

Fig. 16.3 A 54-year-old female previously treated for SARS. T1-weighted oblique coronal image of the left hip. There is a small area of subchondral avascular necrosis involving the superior aspect of the femoral head. No articular surface collapse is present.

while 12.5% of patients who received more than 3 g developed osteonecrosis. Unlike many of the other side effects of corticosteroid therapy (such as immunosuppression, myopathy and reduced bone density), osteonecrosis, once established, will not recover on discontinuation of steroid therapy. Given the implications of avascular necrosis and the possible subsequent joint destruction and disability, this side effect/complication and its possible association with SARS treatment needs to be fully explained to the patient as part of informed consent before commencing treatment, if SARS does recur. In addition to avascular necrosis, non-specific bone marrow abnormalities were present in the MRI of 34% of patients. These were as follows:

162

A F T E R M A T H

O F

S A R S

(not subchondral) unlikely to cause symptoms, structural weakening or long-term problems. In our experience, more than 50% of patients experienced large joint pain following recovery from SARS. The vast majority of these joints show no abnormality on MRI examination. Joint pain following other viral infections is not uncommon. It is recognised that there are ‘arthritogenic’ viruses, including hepatitis C, rubella and human T-cell lymphotrophic virus type 1 (HTLV-I) [10]. An important implication is that joint pain cannot be used as a reliable clinical indicator when screening for osteonecrosis following steroid treatment for SARS.

Key points Avascular necrosis is a sequelae of SARS

• effect of infection/illness and/or treatment • Some relationship between steroid dose and avascular necrosis

• Non-specific abnormalities detected with •

MRI of unknown significance Pain not a sensitive sign of avascular necrosis

Adrenal insufficiency As a result of steroid therapy, suppression of normal adrenal steroid production is to be expected. Many patients were still on steroid replacement therapy months after hospital discharge. They continued to fail steroid challenge tests and were thus not weaned off exogeneous corticosteroids.

Psychological scars During the post-epidemic period, it became apparent that significant psychological effects aside from physical scars were present in patients following the initial SARS outbreak. Many patients were still recovering from the initial trauma of the infection and ‘near death’ experience 6 months after the event. In the authors’ institution, patients who developed psychiatric symptoms since their SARS infection underwent MRI of the brain after their hospital

Fig. 16.4 A 34-year-old male patient with psychiatric symptoms 6 months after treatment of SARS. Axial T2 FLAIR MRI scan demonstrating no focal abnormality.

discharge. None of the MRIs demonstrated an abnormality (Figure 16.4) [11,12]. In a study by Chan et al. [13], patients who have recovered from SARS show symptoms of psychological trauma. In the early recovery phase, about 5 weeks from onset of SARS, 26% (27 of 101) of inpatients showed moderate to severe degrees of anxiety and 16% (16 of 101) of inpatients showed moderate to severe degrees of depression. It was assumed that this psychological aftermath will probably improve over time. The same study reported data from another series of 75 patients who were evaluated at 1–2 months after hospital discharge. Only 5% of these patients were reported to have moderate to severe anxiety and depressive symptoms in this later stage. Other than anxiety or depression, post-SARS patients suffered from some impairment of health-related quality of life. Using the validated Medical Outcomes Survey (MOS) 36-items Short Form Health Survey, this second series showed a decrease in health-related quality-of-life scores, particularly in the domains of physical and social functioning and bodily pain.

R E F E R E N C E S

Some post-SARS-infected health care workers have found it difficult to return to the workplace (e.g. a hospital ward) or their accommodation, which they associate with the source of their infection. This appears to be a form of post-traumatic stress disorder and may require long-term rehabilitation. Will these health care workers, many of whom young nurses, be able to return to full-time work? Only time will tell.

Key points

• Psychological trauma may occur early and some persist after physical recovery

• Most appear to improve with time.

Conclusion Information from retrospective studies and reviews continue to stream in regarding various aspects of SARS. New protocols and practices are being instigated as new information become available. It is hoped that enough has now been learnt to affect a significant positive difference if and when the next outbreak occurs.

References 1. SARS Expert Committee. SARS in Hong Kong: from experience to action. Available online at www.sars-expertcom.gov.hk (accessed 8 September 2003).

163

2. Hospital Authority Review Panel on the SARS Outbreak. Report of the Hospital Authority Review Panel on the SARS Outbreak. September 2003. www.ha.org.hk (accessed 14 November 2003). 3. Griffith JF, Antonio GE, Kumta SM, Hui DSC, Wong KT, Joynt GM, Wu A, Cheung AYK, Chiu KH, Chan KM, Leung PC and Ahuja AT. Osteonecrosis in SARS patients treated with steroids. Radiology (submitted). 4. Nicholls JM, Poon LLM, Lee KC et al. Lung pathology of fatal severe acute respiratory syndrome. Lancet 2003; 361: 1773–1778. 5. Beltran J, Knight CT, Zuelzer WA et al. Core decompression for avascular necrosis of the femoral head: correlation between long-term results and preoperative MR staging. Radiology 1990; 175: 533–536. 6. Wing PC, Nance P, Connell DG and Gagnon F. Risk of avascular necrosis following short term megadose methylprednisolone treatment. Spinal Cord 1998; 36: 633–636. 7. Cook AM, Dzik-Jurasz AS, Padhani AR, Norman A and Huddart RA. The prevalence of avascular necrosis in patients treated with chemotherapy for testicular tumours. Br J Cancer 2001; 85: 1624–1626. 8. Zizic TM, Marcoux C, Hungerford DS, Dansereau JV and Stevens MB. Corticosteroid therapy associated with ischemic necrosis of bone in systemic lupus erythematosus. Am J Med 1985; 79: 596–604. 9. Zanetti M, Bruder E, Romero J and Hodler J. Bone marrow edema pattern in osteoarthritic knees: correlation between MR imaging and histologic findings. Radiology 2000; 215(3): 835–840. 10. Masuko-Hongo K, Kato T and Nishioka K. Virus-associated arthritis. Best Pract Res Clin Rheumatol 2003; 17: 309–318. 11. Maunder R, Hunter J, Vincent L et al. The immediate psychological and occupational impact of the 2003 SARS outbreak in a teaching hospital. Can Med Assoc J 2003; 168(10): 1245–1251. 12. McAlonan GM, Chua SE, Cheung V, Cheung C, Wong JGWS, Choy KM, Wong MMC and Tsang KWT. Psychological effects of SARS on health-care workers in Hong Kong. Response to Hui 2003. Br Med J Online (5 June 2003) at http://bmj. bmjjournals.com/cgi/eletters/326/7398/1067#33006 13. Chan KS, Zheng JP, Mok YW, Li YM, Liu YN, Chu CM and Ip MS. SARS: prognosis, outcome and sequelae. Respirology 2003 November; 8(suppl. 1): S36.

Update on Severe Acute Respiratory Syndrome AT Ahuja and GE Antonio

Introduction Diagnosis and classification Prominence of radiology

165 166 168

Introduction The passage of time has healed some of the wounds inflicted by the deadly epidemic. The economy in Hong Kong and most of Asia has picked up and the communities have gone back to their daily activities.The masks and the gloves are off and people have gone back to their old habits with respect to personal hygiene and work practices. After all, old habits die hard. However, severe acute respiratory syndrome (SARS) continues to lurk in the background and still causes a diagnostic dilemma. By the end of 2003/early 2004 (at the time of writing this chapter), there were five sporadic cases of SARS, two involving laboratory researchers and three outside the laboratory in Guangdong Province in China. Although the two laboratory cases (Taiwan, Singapore [1]) were readily identified, the cases in Guangdong were definitively verified more than a week after the suspicion was raised. The clinical parameters, laboratory tests and history of contact in this patient were initially nebulous, highlighting the difficulty in the early diagnosis of this potentially fatal disease. Despite detailed contact, tracing the source of the infection was not definitively traced and it has been postulated that the source might have been civet cats as unpublished reports from Guangdong and Hong Kong have

Clinical trials Plans for the future Conclusion

CHAPTER

17 168 169 173

found similarities in the genetic sequences of the virus in the SARS patient and civet cats, which is a gourmet delicacy in southern China. Based on this assumption the local authorities in Guangdong undertook a controversial cull of 10,000 civet cats. The World Health Organization (WHO) cautioned against the move saying there was no convincing evidence that the patients had been in direct contact with wild animals (one of the patients worked in a restaurant which had civet cats on its menu). The WHO has also expressed concerns that the mass slaughter of civet cats could destroy evidence of origin in tracing the SARS virus. Reports from China have also found DNA traces of the SARS in rats and local authorities in Guangdong are planning a large extermination programme against rats in the city. In all the cases, the patient and their contacts were swiftly isolated and quarantined by the local public health authorities. As a result no further transmission appears to have occurred and the local authorities were commended for their efforts in limiting the disease spread by the WHO. These cases highlight several issues regarding SARS in the post-epidemic period: 1. The difficulty in early diagnosis of the disease in the post-epidemic period. During the previous

166

U P D A T E

outbreak, the diagnosis was predominantly clinical. If a patient presented with clinical symptoms, positive chest X-ray and history of contact with SARS patient, then the diagnosis was SARS. However, in the post-outbreak situation the diagnosis is based predominantly on a series of laboratory tests which are time consuming and by the time the definitive diagnosis is established patient become asymptomatic. 2. Health care workers will continue to be at risk of infection. 3. Infection control and safe practice are of prime importance. 4. A well-planned protocol for disease surveillance and notification, public health alert system and contact isolation/quarantine can work very well in limiting the disease.

Key points

• Sporadic cases were occurring after the end of the epidemic

• Some cases had no apparent source • Laboratory tests take time for confirmation • Health care workers remain at risk of infections

On the other hand, in view of issues arising from the after-effects of SARS (lung scarring, avascular necrosis of bones, psychological trauma to health care workers, patients and their families), a wellthought-out future treatment plan and research protocols would also help to reduce future damage should we have to deal with the disease again. If one is to learn from past experiences, in addition to early and accurate diagnosis, the following issues will have to be addressed:

• Randomized controlled trial(s) on different diag•

nostic tests (including radiology) and treatment regimes. Plan of action for the community (isolation, quarantine and immigration) and hospitals.

This chapter hopes to update the readers with the new case definition of SARS and the procedures put in place to deal with any recurrence of SARS on a global scale.

O N

S A R S

Diagnosis and classification CDC case definition Less than a week prior to the sporadic case in Taiwan, the Centers for Disease Control and Prevention of the United States (CDC) issued an updated interim US case definition for SARS [2].

Clinical criteria • Early illness: Presence of two or more of the fol-

•

•

lowing features: fever (might be subjective), chills, rigors, myalgia, headache, diarrhoea, sore throat or rhinorrhoea. Mild-to-moderate respiratory illness: Temperature of 100.4°F (38°C) and one or more clinical findings of lower respiratory illness (e.g. cough, shortness of breath or difficulty breathing). Severe respiratory illness: Meets clinical criteria of mild-to-moderate respiratory illness and one or more of the following findings: – radiographical evidence of pneumonia, – acute respiratory distress syndrome (RDS) or – autopsy findings consistent with pneumonia, or acute RDS, without an identifiable cause.

Epidemiological criteria Possible exposure to SARS-associated coronavirus (SARS-CoV) One or more of the following exposures in the 10 days before onset of symptoms: 1. Travel to a foreign or domestic location with documented or suspected recent transmission of SARS-CoV or 2. Close contact with a person with mild-to-moderate or severe respiratory illness and history of travel in the 10 days before onset of symptoms to a foreign or domestic location with documented or suspected recent transmission of SARS-CoV.

Likely exposure to SARS-CoV One or more of the following exposures in the 10 days before onset of symptoms: 1. Close contact with a person with confirmed SARS-CoV disease or

D I A G N O S I S

A N D

167

C L A S S I F I C A T I O N

2. Close contact with a person with mild-to-moderate or severe respiratory illness for whom a chain of transmission can be linked to a confirmed case of SARS-CoV disease in the 10 days before onset of symptoms.

Laboratory criteria Tests to detect SARS-CoV are being refined and their performance characteristics assessed; therefore, criteria for laboratory diagnosis of SARS-CoV are changing. The following are general criteria for laboratory confirmation of SARS-CoV:

• •

•

•

• detection of serum antibody to SARS-CoV by a test validated by CDC (e.g. enzyme immunoassay),

• isolation in cell culture of SARS-CoV from a •

clinical specimen or detection of SARS-CoV RNA by a reverse transcription polymerase chain reaction test validated by CDC and with subsequent confirmation in a reference laboratory (e.g. CDC).

SARS-CoV disease • Probable case of SARS-CoV disease: meets the

•

Exclusion criteria

SARS-CoV disease or places with known ongoing transmission of SARS-CoV. Reports in persons from areas where SARS activity is occurring. SARS RUI-2: Cases meeting the clinical criteria for mild-to-moderate illness and the epidemiological criteria for possible exposure (spring 2003 CDC definition for suspect cases). SARS RUI-3: Cases meeting the clinical criteria for severe illness and the epidemiological criteria for possible exposure (Spring 2003 CDC definition for probable cases). SARS RUI-4: Cases meeting the clinical criteria for early or mild-to-moderate illness and the epidemiological criteria for likely exposure to SARS-CoV.

clinical criteria for severe respiratory illness and the epidemiological criteria for likely exposure to SARS-CoV. Confirmed case of SARS-CoV disease: clinically compatible illness (i.e. early, mild-to-moderate or severe) that is laboratory confirmed.

A case may be excluded as a SARS report under investigation (SARS RUI), including as a CDC-defined probable SARS-CoV case, if any of the following apply:

Further definition of the above terms is available at the CDC web site [2].

• an alternative diagnosis can explain the illness

WHO case definition

• •

fully, antibody to SARS-CoV is undetectable in a serum specimen obtained more than 28 days after onset of illness or the case was reported on the basis of contact with a person who was excluded subsequently as a case of SARS-CoV disease; then the reported case also is excluded, provided other epidemiological or laboratory criteria are not present.

Prior to the sporadic case in Singapore, on 14 August 2003, the WHO also modified their case definition for SARS in an article titled ‘Alert, verification and public health management of SARS in the post-outbreak period’ [3]. This modified clinical case definitions is as follows:

• A person with a history of fever (38°C) and

Case classification SARS RUI Reports in persons from areas where SARS is not known to be active

• One or more symptoms of lower respiratory tract illness (cough, difficulty breathing and shortness of breath) and

• SARS RUI-1: Cases compatible with SARS in

• Radiographical evidence of lung infiltrates con-

groups likely to be first affected by SARS-CoV if SARS-CoV is introduced from a person without clear epidemiological links to known cases of

• Autopsy findings consistent with the pathology of

sistent with pneumonia or RDS or pneumonia or RDS without an identifiable cause.

168

U P D A T E

and

• No alternative diagnosis can fully explain the illness. For a sporadic outbreak or the early days of an epidemic, it may be difficult to use the epidemiological criteria for the diagnosis of SARS since sporadic cases are unlikely to provide history of contact. On the other hand, the lack of a rapid and accurate biochemical test for the detection of SARS-CoV in the first few days of the illness [2,4] excludes the use of the laboratory criteria early in the disease. Thus, one is again left with relying on the clinical criteria for early diagnosis.

O N

S A R S

sense that it does not include patients who present with no fever or with minimal chest symptoms, or with gastrointestinal symptoms (see Amoy Gardens SARS outbreak [7]) as compared to the broader CDC clinical definition [2]. On the other hand, an all-encompassing case definition may increase the number of false positives and place a considerable load on health resources and increase the public anxiety.

Key points

• Radiology continues to be a major diagnostic tool

• CT should be performed early when in Key points

doubt

• New case definitions released by the CDC • •

and WHO in the post-epidemic period. Clinical signs and radiographical confirmation constitute the main elements in early diagnosis. CDC definition is extensive and covered various different situations.

Prominence of radiology In both revised case definitions, radiographical findings have been included as part of the clinical case definition and diagnostic workup by the CDC and WHO [2,4,5]. Such reliance on imaging demands a high sensitivity from whichever imaging modality is chosen. At present, computed tomography (CT) has the highest sensitivity in the demonstration of lung abnormalities. A study by Wong et al. [6] has shown that this high sensitivity was also true for the early detection of ‘occult’ lung abnormalities in SARS patients. Therefore, CT should be used as a close second-line investigation after an initial normal chest radiograph in the workup of a suspected SARS patient as recommended by the CDC [4]. The significance of an early and accurate diagnosis in the quarantine of disease and effecting early treatment cannot be overstated. The revised WHO clinical case definition is narrower in scope compared to its previous definition in the

Clinical trials During the epidemic, radiology played an important role in the initial diagnosis, in progress monitoring and assessment of treatment effect, and in the detection of complications and treatment side effects. All of these parameters were retrospectively analysed for the patients after the epidemic and useful results have emerged. For the next outbreak, prospective trials have to be considered and forward planning is necessary. Similarly, the treatment of SARS has been controversial and well-designed clinical trials are essential to settle the differences.

Treatment trials Two major agents have been used in the treatment of SARS during the epidemic, namely antivirals and corticosteroids. The treatment remained controversial particularly due to the lack of a randomized controlled trial using either agent. The emergence of possible side effects (psychosis and cardiac dysrhythmia) and complications (secondary infections and avascular necrosis) had made the issue of treatment choice even more difficult. On the other hand, other drugs such as Kaletra show some promising results. Therefore, it is essential that a double-blinded, randomized controlled trial be in place for the next outbreak.

P L A N S

F O R

T H E

169

F U T U R E

Imaging trials

emergency. Therefore surveillance is of utmost importance.

The sensitivity and specificity of imaging in the diagnosis of SARS have not been established. This is probably the most important contribution imaging could provide for the management of the disease. Future studies on this will have to establish this in a double-blind trial where all patients with fever and lower respiratory tract symptoms (as per the current WHO case definition) are imaged and diagnosed to have SARS or not.

The WHO has designated three levels of surveillance for SARS for different regions in the world based on their previous level of exposure [3]. These are defined as:

The use of the early imaging signs as predictors of prognosis is an attractive option. Previous studies have suggested that the percentage of lung volume involvement and the number of lesions may be useful predictors. Although the results appear intuitive, the study was done retrospectively and suffers from treatment bias (progressive chest radiograph deterioration would have caused the clinicians to step up treatment). A future trial should put this into practice prospectively and a blinded randomized controlled trial on treatment should take this into account as one of the parameters for assessment.

1. Potential zone of re-emergence of SARS-CoV. Identified as source(s) of the previous outbreak in November 2002 or areas with an increased likelihood of animal to human transmission of SARS-CoV infection. 2. Nodal areas where sustained local transmission experienced during the previous outbreak or entry of large numbers of persons from the potential zone of re-emergence of SARS-CoV. 3. Low-risk areas. Never reported cases, reported only imported cases or experienced only limited local transmission during the previous outbreak. Based on this, WHO recommends the staged approach to surveillance, which summarized in Table 17.1.

The SARS alert Key points

• Clinical trials valuable and essential if • •

SARS recurs Both treatment and imaging efficacy need to be evaluated Imaging may provide an early indication to prognosis

For the purposes of SARS surveillance (expedite diagnosis, step up infection control, activate the public health response and raising a global alert), the WHO has devised the SARS alert [3]. When this alert is present, appropriate infection control and public health measures against SARS should be implemented until SARS has been ruled out as a cause of the atypical pneumonia or RDS. Table 17.1

Staged approach to surveillance.

Plans for the future For sporadic cases, time is of the essence to avoid an outbreak. The diagnosis needs to be established rapidly to allow confinement of the contacts and isolation of the patient. Hospitals and health organizations have been encouraged by the WHO to set up surveillance systems and early warning protocols [3].

Plans on a global scale The reappearance of SARS in the human population would be considered a global public health

Potential Nodal Low-risk zone of areas areas re-emergence SARS alert Enhanced surveillance for SARS Special studies for SARS-CoV infections in animal and human populations

Y Y

Y

Y Y

Y

170

The SARS alert is defined as: 1. Two or more health care workers in the same health care unit fulfilling the clinical case definition of SARS (see new case definition described above) and with onset of illness in the same 10-day period or 2. Hospital-acquired illness in three or more persons (health care workers and/or other hospital staff and/or patients and/or visitors) in the same health care unit fulfilling the clinical case definition of SARS and with the onset of illness in the same 10-day period.

Plans for the community For each region, the WHO recommends a plan for infection surveillance and control to be adapted based on its guidelines [3]. The following should be included in such a plan: 1. 2. 3. 4. 5.

rapid diagnosis; immediate isolation of the infected patient(s); proper isolation facilities; contact tracing and quarantine; surveillance of high-risk institutions/populations (hospitals, nursing homes, etc.).

In line with the WHO recommendations, the Hong Kong Government has developed a three-level response system [8]:

• Alert Level Response: activated when there is a • •

laboratory-confirmed SARS case outside Hong Kong; or a SARS alert in Hong Kong. Level 1 Response: activated when there is one or more laboratory-confirmed SARS cases in Hong Kong occurring in a sporadic manner. Level 2 Response: activated when there are signs of local transmission of the disease in Hong Kong.

In addition, the following measures will be undertaken: 1. A steering committee will be set up to oversee disease control and coordinate measures to limit the socio-economic effects of an epidemic. Disaster drills will be conducted to test the surveillance system and response strategies. 2. A centre for disease control will be set up to speed up laboratory tests, strengthen contact tracing

U P D A T E

O N

S A R S

and disease investigations. Communication is another important aspect of disease control. In Hong Kong, the public health authority and the government will keep close contact with Mainland China and neighbouring health authorities to ensure information exchange. 3. As in other international ports, temperature screening with infra-red scanners and health declarations will continue to be operated at immigration checkpoints. Aside from providing a robust system of disease surveillance and a clearly defined response system, the WHO has recommended vaccination against influenza be encouraged to minimize this common disease from the differential diagnosis of SARS [9]. Health care workers in particular should all be vaccinated and those who do develop the ‘flu’ should be encouraged to stay at home, rather than to soldier on, to avoid spreading this mimicker of SARS. While the WHO and regional public health authorities will continue to monitor the disease and coordinate communication, there is another very effective channel of communication among medics, health care workers and the public. The Internet has been an invaluable platform in disseminating news and information to the whole world. Our radiology department has had some success with our SARS web page in providing radiographical images of SARS [10]. The World Wide Web will continue to play an important role in the fight against SARS. Television advertising campaigns have been continually launched by the government to: 1. promote cleanliness and vigilance; 2. keep the public informed of preparations being made against a second outbreak; 3. advertise to visitors that the city is once again safe.

Plans for the hospitals Isolation wards, and in some centres infectious disease wings, have been built since the end of the epidemic. Additional 530 isolation rooms with 1280 beds in total were added in nine major hospitals in Hong Kong. Different levels of infection control measures are in place to be stepped up to meet the level of suspicion (Figures 17.1 and 17.2). Protective clothing and resources have been replenished to meet a drawn out epidemic (3 months supply).

P L A N S

F O R

T H E

F U T U R E

171

Fig. 17.1 A security guard and a health care personnel at the entrance of outpatient clinics supervising hand-cleaning and mask-wearing for all visitors before entering the clinics.

Fig. 17.2

Visitor and health care personnel cleaning hands at the main hospital entrance.

172

U P D A T E

O N

S A R S

Protection during transport of patients is being put into place. These are all essential preparations that need to be laid down before the next outbreak. In Hong Kong, a three-tier response plan has been established by the hospital authority [11] to be used in the event of a major outbreak of infectious disease, including SARS:

• Green Alert (prior to Alert Level Response): Hospital

•

•

alerted to an abnormal pattern of infectious disease in the community or inside the hospital system and when there are existing guidelines and knowledge on treatment and control, and local action is judged to be adequate. Yellow Alert (Level 1 Response): The hospital is alerted to an abnormal pattern of infectious disease, which may have territory-wide implications, or require a health authority-wide response in: – providing central coordination in data collection and interpretation of the epidemiological pattern; – refining clinical management or infection control guidelines; – mounting a territory-wide response in service management and resource deployment. Red Alert (Level 2 Response): Higher-level interdepartmental response will be required and the government may activate the Inter-departmental Action Coordinating Committee. In a major disaster situation, there will be a need for strategic command to effect prompt and decisive response. There will need to be a mechanism to mobilize resources (supplies and manpower) in an efficient manner to ensure that the corporate interest and the health of the population are protected.

The plan provides a detailed description of the response actions, staff deployment, patient allocation and distribution of resources [11]. A system to monitor the sick leave of the health care workers in all hospitals is in place to monitor for unusual patterns. All staff are required to report fever and flu-like symptoms to hospital management. E-SARS, an electronic database will be expanded for use in surveillance and monitoring of infectious diseases.

Plans for radiology departments The chest radiograph with its inherent blind areas will not be able to pick up all early cases [6].

Fig. 17.3 Nursing staff in protective gear, at the entrance of radiology department registering outpatient attending for imaging.

Thin-section CT will have to be performed on these suspicious patients if the chest radiograph is negative. This is particularly important as the laboratory tests have yet to achieve a high sensitivity in the early stages of the disease. Given that sporadic cases are likely to herald an outbreak and that these cases will be radiographed with other lower respiratory tract infection patients, work practices may have to be modified. Specific time slots should be cleared for imaging these suspicious cases, preferably at the end of the day when the department is not crowded and more time and attention could be spent on infection control. Different levels of infection control have to be devised to correspond to the different levels of SARS alert. Education on and supervision of infection control measures should be continued. Reporting practices may also have to be modified. Unusual radiographical trends (atypical pneumonia) need to be monitored as a possible indicator of

173

R E F E R E N C E S

Conclusion The knowledge on SARS is continuously evolving and being updated on a daily basis. It is essential that health care workers are familiar with the updated information and the knowledge acquired should be put into use if a future outbreak is to occur. Prospective clinical trials and public health protocols need to be planned taking past experience into account. Radiology will continue to play a major role in the diagnosis and assist in the treatment of patients.

References

Fig. 17.4 Nursing staff in protective gear, taking temperature of out-patient at the entrance of radiology department.

disease outbreak. Decreasing film transport, limiting handling/transporting of ward examination requests and patient would help reduce contamination during the epidemic (Figures 17.3 and 17.4). Obviously a picture archiving and communication system (PACS) system solves a lot of these problems, but with limited resources, the problems could still be overcome with some planning, e.g. immediate reporting, faxing of requests, mobile/ satellite radiographical examination units.

Key points

• Plans were drawn up for all levels of society • Effectiveness of various plans rest on individuals carrying out their designated roles

• Daily practices for all need to be changed to complement the plans

• Education, surveillance and communication are important

1. World Health Organization, Communicable Disease Surveillance and Response (CSR). Severe acute respiratory syndrome (SARS) in Singapore – update 2: SARS case in Singapore linked to accidental laboratory contamination. Downloadable at http://www.who.int/csr/don/2003_09_ 24/en/ (accessed 24 September 2003). 2. Centers for Disease Control and Prevention of the United States (CDC): Updated interim US case definition for severe acute respiratory syndrome (SARS). Available online at http://www.cdc.gov/ncidod/sars/casedefinition.htm (accessed 12 December 2003). 3. World Health Organization. Communicable Disease Surveillance and Response (CSR). Alert, verification and public health management of SARS in the post-outbreak period. Downloadable at http://www.who.int/csr/sars/ postoutbreak/en/ (accessed 30 September 2003). 4. Centers for Disease Control and Prevention of the United States (CDC): Clinical guidance on the identification and evaluation of possible SARS-CoV disease among persons presenting with community-acquired illness. Available online at http://www.cdc.gov/ncidod/sars/clinicalguidance. htm (accessed 18 December 2003). 5. World Health Organization. Alert, verification and public health management of SARS in the post-outbreak period. Available online at http://www.who.int/csr/sars/postoutbreak/en/ (accessed 14 August 2003). 6. Wong KT, Antonio GE, Hui DS et al. Thin-section CT of severe acute respiratory syndrome: evaluation of 73 patients exposed to or with the disease. Radiology 2003; 228(2): 395–400 (Epub May 2003). 7. Rainer TH, Cameron PA, Smit D et al. Evaluation of WHO criteria for identifying patients with severe acute respiratory syndrome out of hospital: prospective observational study. Br Med J 2003; 326(7403): 1354–1358. 8. LegCo Panel on Health Services Subcommittee. Contingency mechanism to deal with possible resurgence of SARS. Available online at http://www.legco.gov.hk/yr03-04/english/panels/ hs/hs_sars/papers/hs_sars1117cb2-339-1e.pdf (accessed 26 November 2003). 9. World Health Organization. Influenza vaccination for the 2003–2004 season: recommendations in the context of concern about SARS. Downloadable at http://www.who. int/csr/disease/influenza/sars/en/print.html (accessed 2 September 2003). 10. Griffith J, Antonio G and Ahuja A. SARS and the modern day pony express (the World Wide Web). Am J Roentgenol 2003; 180(6): 1736. 11. Hospital Authority’s Response Plan for Infectious Disease Outbreaks (3/10/2003). Available online at http://ha.home/ ho/ps/HA_response.htm

Index

SARS
severe acute respiratory syndrome Page numbers in italics refer to figures; but note that figures are only indicated when they are separated from their text references.

abdominal pain 20 abscess, pulmonary see pulmonary abscess Actinomyces 35, 36–37 activated partial thromboplastin time (APTT) 30 Acute Physiology and Chronic Health Evaluation (APACHE) II score 101 acute respiratory distress syndrome (ARDS) differential diagnosis 59 Pneumocystis carinii pneumonia 42, 43 SARS-related 54, 63–64, 89 chest X-ray features 65, 66, 102–103 follow-up imaging 85, 86 management 101, 103 similarity to SARS 56, 80 acute respiratory phase of SARS (phase 3) 63–64, 89, 90 see also acute respiratory distress syndrome; respiratory failure adenovirus pneumonia 39, 41 admission criteria hospital 27 intensive care unit see intensive care, indications adolescents 121, 122 adrenal insufficiency 162 adult respiratory distress syndrome see acute respiratory distress syndrome advertising campaigns 15, 170 aftermath of SARS 159–163

age intensive care and 101 prognosis of SARS and 31 SARS cases 4–5, 7, 8 AIDS 38, 41–42, 43 air travel 1 air-bronchograms 56, 112, 124 aircraft in-flight care of suspected case 15 transmission within 11 air-trapping, viral pneumonias 40, 41 alanine aminotransferase (ALT) 30 alert, SARS 169–170 Alert Level Response 170 alveolitis, acute extrinsic allergic 58 Amoy Gardens housing estate, Hong Kong 11–12 anaemia, ribavirin-associated 92–93 angiography, infection control measures 157 antibiotics 61, 89, 101 efficacy 93 paediatric SARS 128 anti-tumour necrosis factor (anti-TNF) 95 antiviral agents 92–94 see also ribavirin anxiety post-SARS 162 in radiographers 146–147 APACHE II score 101 appointments systems, radiology 153–154 ARDS see acute respiratory distress syndrome aspartate aminotransferase 132

aspergillosis 48, 49 allergic 48 invasive 48, 49 Aspergillus 48 aspiration pneumonia 36–37 atelectasis 76, 110 audits 159 avascular necrosis 81, 160–161, 162 azathioprine 95 Bacteroides 36–37 barotrauma 103 Blastomyces dermatitidis 48 blind spots, radiological 59, 74 body fluids 11 bone, avascular necrosis 81, 160–161, 162 bone marrow abnormalities 161–162 border control measures 15, 170 brain imaging 81, 162 breathlessness see dyspnoea bronchial dilatation 73, 82 bronchial wall thickening 119 bronchiectasis complicating childhood pneumonia 113, 114, 119 non-tuberculosis mycobacterial infection 46, 47 traction 81, 83, 84, 138, 139 tuberculosis 45 bronchiolitis obliterans 118 bronchiolitis obliterans organizing pneumonia (BOOP) 58, 75 similarity to SARS 80, 85, 126 bronchogenic cyst 114, 115 bronchopleural fistula 49, 50

176

bronchopneumonia 33, 36–39, 58 bulging fissure sign 34, 35 Canada 1, 2, 5, 7, 131–132 see also North America Candida 48, 49 Candida albicans 103 carbon dioxide tension, arterial (PaCO2) 102 cardiac disease 63–64 case definition CDC 166–167 case classification 167 clinical criteria 166 epidemiological criteria 166–167 exclusion criteria 167 laboratory criteria 167 role of radiology 168 WHO 3, 19, 29 efficacy 13 modified 167–168 paediatric patients 122 see also diagnosis of SARS case detection 13 case fatality 9 cavitation bacterial pneumonias 35, 36, 37 tuberculosis 43, 44, 45, 47, 111 viral pneumonia 41 cefotaxime 61, 128 Centers for Disease Control and Prevention (CDC) case definition 166–167 chest X-ray (CXR) 8, 9, 61–68 childhood pneumonia 109–111 clinical course and 65, 66 clinical outcome and 66–67 clinico-radiological correlation 65–67 diagnostic role 53–60, 69 appearances 56–58 blind spots 59, 74 differential diagnosis 58–59 at end of epidemic 55 initial outbreak 54–55 pathological aspects 53–54 emergency department screening 20–21, 22 follow-up of SARS 80, 81–82, 87 in future planning 172

I N D E X

infection control measures 155 intensive care and 67, 100 laboratory features and 67 modification of routine practice 144–146 in North America 132–136 paediatric SARS 122–125 patterns during treatment 62, 63, 64 pneumonias 33, 34–49 role in SARS management 67 SARS-related ARDS 65, 66, 102–103 chickenpox pneumonia 39–40 children pneumonia 109–120 chest X-ray 109–111 chronic sequelae/role of CT 118–120 common causes 109–111 complications/role of CT 115–117 CT scanning 112–113 developmental abnormalities 114–115 differential diagnosis 123–125 persistent or recurrent 113–115 predisposing conditions 113, 114 viral 39, 109–110 SARS 121–130 clinical presentation 8, 121–122 HRCT 125–127, 128 infection control 129 management and outcome 128–129 plain chest X-rays 122–125 risk stratification 129 vs adult SARS 125, 127 tuberculosis 43, 111, 112, 113 chills 7, 8 China 1, 2, 3, 6, 7 see also Guangdong Province, China; Hong Kong; Taiwan chlamydial pneumonia 58–59 chronic eosinophilic pneumonia (CEP) 58, 75–76 chronic obstructive pulmonary disease (COPD) 38 civet cats 165 clarithromycin 61, 128

cleaning CT suite 146 radiology department 154 surfaces 25, 26, 154 clinic, SARS screening 23–24 clinical course chest X-ray changes during 64 paediatric SARS 121, 122 response to treatment 91–92, 93 triphasic pattern 62–65, 90 clinical features of SARS 7, 8–9, 20, 30 children 8, 121–122 in North America 132 post-recovery 79, 159–163 clinical trials 166, 168–169 imaging 169 treatment 166 clinico-radiological correlation 65–67 Coccidioides immitis (coccidioidomycosis) 48 community future plans 170 transmission in 11–12 co-morbidity 5, 7 complications of SARS follow-up imaging 85, 86 post-recovery 159–163 see also pulmonary fibrosis, post-SARS computed tomography (CT) childhood pneumonias 112–113 chronic sequelae 118–120 complications 115–117 follow-up of SARS 80, 81 in future planning 172 high resolution see high-resolution computed tomography infection control measures 70, 146, 155–156 in North America 134–135, 136 pneumonias 49–50 role in case definition 168 SARS-related ARDS 103 tuberculosis 43 viral pneumonia 41 confirmed cases CDC-defined 167 infection control measures 152 WHO-defined 19

177

I N D E X

confusion 81, 94 consolidation, pulmonary 33 childhood pneumonias 112 lobar pneumonia 34 Pneumocystis carinii pneumonia 42, 43 SARS pneumonia 54, 56, 57, 65 in children 122–123, 126 follow-up 84 HRCT 72, 73 in North America 133–134, 135, 136, 137–139, 140 SARS-related ARDS 104 tuberculosis 43, 44 viral pneumonias 40–41 contacts definition 3, 14 isolation 14 screening guidelines 24–25 tracing 13, 14 contagion model 10 contrast radiography, infection control measures 156 convalescent plasma 62, 95–96, 101 coronaviruses respiratory tract infection 54 SARS see SARS coronavirus corticosteroid therapy 61–62, 90, 94–95, 101 adrenal insufficiency after 162 avascular necrosis and 160–161 efficacy 92–93 in paediatric SARS 128, 129 post-SARS lung fibrosis 85 radiological guidance 67 side effects 94, 95 cough 7, 8, 132 ‘crazy-paving’ appearance 72, 73 follow-up 82, 83 creatinine phosphokinase (CPK) 9, 30 Cryptococcus neoformans 48, 49 CT scanning see computed tomography cystic adenomatoid malformations, congenital 114, 115 cystic changes, pulmonary post-SARS 85, 86, 104 cytokines 94 cytomegalovirus pneumonia 39, 40

D-dimers 30 deaths intensive care patients 103–104 numbers 2, 7 radiological predictors 66–67 dependent densities 76 depression, post-SARS 162 developmental abnormalities, lung 114–115 diagnosis of SARS 3, 13, 19–20, 29 chest radiography 53–60, 69 HRCT 55, 69–77 imaging protocol 69 post-epidemic period 165–166 updates 166–168 see also case definition diarrhoea 7, 8, 20 virus transmission and 11 digital radiography 55, 144 discharge criteria, suspect cases 27 disinfection, radiology department 154 droplet transmission 10–11 dyspnoea (breathlessness) 7, 8, 132 post-treatment 79, 80, 82, 87 progressive 99 education public 15 staff 150, 153 emergency department (ED) 17–28 challenge of initial SARS outbreak 18–19 clinical assessment 19 clinical case definition 19–20 clinical features 20 closures 17, 18 historical background 18 laboratory tests 22–23 management of suspect cases 26–27 radiography 20–21, 22 screening clinic 18, 23–24 staff and patient safety 19, 25–26 emotional impact, on radiographers 146–147 emphysema, surgical 66 empyema 35, 37, 49 in children 115, 116 endobronchial spread non-tuberculosis mycobacteria 46 tuberculosis 43–44, 45, 47

enzyme-linked immunosorbent assay (ELISA), SARS virus antibodies 31 epidemiology of SARS 1–8 chain of transmission 1, 2 features 4–8 Hong Kong University Hospital outbreak 29 in North America 131–132 numbers of cases 2, 3, 4, 5, 6, 7 E-SARS 172 Escherichia coli pneumonia 35, 36, 39 Europe 7 examination, clinical 33 exclusion criteria, SARS 3, 167 exercise intolerance, post-SARS 82 extrinsic allergic alveolitis, acute 58 eye protection 26, 150 faeces 11 fatality, case 9 fever 7, 8, 30, 132 role in early detection 20 screening travellers 15 fibrosis, lung see pulmonary fibrosis film cassettes, radiographic 144–145, 155 fluid balance 102 fluoroscopy, infection control measures 156 follow-up 79–87 clinical presentation at 79 criteria, suspect cases 27 imaging 80–87 appearances 81–86 protocol 87 role 80–81 paediatric patients 127 pathological aspects 79–80 fomites 11 Friedlander’s pneumonia 35 fungal pulmonary infections 48–49 Fusobacterium 36–37 gangrene, pulmonary 35, 36 gender differences 4, 7 Germany 1, 2 Ghon focus 43 gloves 25, 26, 150 glycyrrhizin 96–97

178

gowns 26, 150 Green Alert 172 ground-glass opacities Pneumocystis carinii pneumonia 42, 43 SARS 56, 71, 72 associated abnormalities 72, 73, 138–139 in children 126 follow-up 82–83, 84 in North America 132, 133, 135, 136–138, 139 SARS-related ARDS 104, 105 viral pneumonias 40, 41, 112 Guangdong Province, China post-outbreak cases 165 SARS outbreak 1, 3, 6, 18 haemolysis, ribavirin-associated 92–93 Haemophilus influenzae pneumonia 36, 38, 39, 110 hand washing 25, 26, 150, 151 headache 7, 8 health care workers (HCWs) emergency department 19 influenza vaccination 170 post-epidemic risks 166 psychological impact of SARS 143, 163 routes of transmission 11 SARS cases 1, 2, 6–8 SARS screening clinic 25–26 see also radiographers health systems, audit of response to SARS 159 herbal medicine 96–97 herpes simplex type 1 virus 39 high-resolution computed tomography (HRCT) bronchopneumonia 36, 37 chronic sequelae of childhood pneumonia 118, 119–120 diagnosis of SARS 55, 69–77 differential diagnosis 74–76 indications 70–71 pitfalls 76 emergency department screening 20–21, 22 features of SARS 72–74

I N D E X

appearances 72–73 location of lesions 73–74 in North America 136–140 paediatric patients 126–127, 128 follow-up of SARS 80, 82–85, 87 fungal pulmonary infections 48, 49 infection control measures 146, 155–156 intensive care patients 104–105 Pneumocystis carinii pneumonia 42, 43 pneumonias 49–50 scanning technique 70 tuberculosis 43–45, 46, 47 viral pneumonias 41 hilar lymphadenopathy fungal pulmonary infections 48 tuberculosis 43, 111, 112 viral pneumonias 41, 110 hip joint pain 160 Histoplasma capsulatum (histoplasmosis) 48, 49 home confinement, contacts 14 Hong Kong audit of health system response 159 emergency department screening 17–28 future plans 170–172 index case 1, 18, 29 routes of transmission 11–12 SARS outbreak 1, 2, 4, 7 University Hospital SARS outbreak 29–31 hospital admission criteria 27 future plans 170–172 infection control 14 isolation of patients 13 outbreak management 14–15 transmission in 11 hospital-acquired see nosocomial hotels 1, 2 households, transmission within 11 HRCT see high-resolution computed tomography hydrocortisone 61, 128 hygiene, personal patients attending screening clinic 26 radiology staff 150

screening clinic staff 25, 26 hyperglycaemia, corticosteroidinduced 94 hypocalcaemia 132 hypokalaemia 30, 94 hyponatraemia 30 hypoxaemia/hypoxia progressive 99 treatment options 90, 91, 96 ICU care see intensive care imaging clinical trials 169 follow-up 80–87 in North America 131–141 pneumonias 33–52 protocol, diagnosis of SARS 69 role in case definition 168 see also specific modalities immune response, post-SARS pathological changes and 80 immunocompromised patients bacterial pneumonias 35, 38 fungal pulmonary infections 48, 49 Pneumocystis carinii pneumonia 41–42 viral pneumonias 39–40 immunofluorescence assay (IFA), SARS virus antibodies 22, 31 immunoglobulin, intravenous (IVIG) 95, 97 immuno-modulators 94–95, 97 immuno-pathological damage phase (phase 2) 63, 89, 90 chest X-ray features 65 incubation period 10, 30 infection control CT scanning suite 70, 146, 155–156 emotional impact on staff 146–147 enforcement 153 future plans 170–172 hospital 14 intensive care unit 105–106 paediatric SARS 129 plain radiography 144–146 post-epidemic period 166 radiology department 149–158 screening clinic staff 25–26 infectivity, period of 10

179

I N D E X

influenza pneumonia 39, 41, 58, 75 vaccination 170 information dissemination 153 intensive care 63–64, 99–107 follow-up imaging 85, 86 indications 90, 91, 96, 99–101 infection control 105–106 management and progress 101–103 outcome 103–105 patient demographics 101 radiological predictors 66–67 role of chest X-rays 67, 100 interferon (IFN) 95 internet 170 interstitial thickening, intralobular 72, 73, 82 interventional radiology, infection control measures 157 intravenous drug abuse 37, 38 intravenous immunoglobulin (IVIG) 95, 97 intubation, endotracheal 90, 91, 96, 101–102 isolation contacts 14 facilities 170 patients 13, 26–27 joint problems, post-SARS 81, 160–162 Kaletra™ 94 Klebsiella pneumoniae pneumonia 34, 35, 36 knee joint pain 160 laboratory tests CDC case definition 167 chest X-ray changes and 67 emergency department screening 22–23 epidemiological features 7–8, 9 Hong Kong University Hospital outbreak 30–31 in North America 132 lactate dehydrogenase (LDH), serum 30, 132 post-SARS lung fibrosis and 86

prognosis of SARS and 31 radiological progression and 67 Legionella pneumoniae 35–36 leucopaenia 7, 9, 30 Level 1 Response 170 Level 2 Response 170 levofloxacin 61 lobar pneumonia 33, 34–36 lopinavir 93–94 lymph nodes fungal pulmonary infections 48 tuberculosis 43, 111, 112, 113 lymphocyte count 9, 30 lymphocytes, reactive 30 lymphopaenia 8, 9, 22, 30, 132 Macao 7 Macleod’s syndrome 118 magnetic resonance imaging (MRI) brain 81, 163 follow-up of SARS 80–81 infection control measures 157 musculoskeletal sequelae 81, 160, 161–162 malaise 7, 8 managers, radiology department 152–155 masks 11 members of public 15 radiology staff 150 screening clinic staff 25, 26 meals, staff 150 measles 41 media campaigns 15, 170 mediastinal lymph node enlargement 42, 43, 48, 111 metapneumovirus pneumonia 40 methylprednisolone 62, 90, 94–95, 101 clinical response 93, 94–95 paediatric SARS 128 radiological guidance 67 side effects 94, 95 monocyte counts 30 motion artefacts, HRCT 76 MRI see magnetic resonance imaging multiple organ dysfunction (MOD) 103 musculoskeletal problems, post-SARS 81, 160–162

myalgia 7, 8, 132 mycobacterial infections, non-tuberculosis (NTMB) 45–46, 47 Mycobacterium tuberculosis 35, 43–45 see also tuberculosis Mycoplasma pneumoniae pneumonia 41, 42, 58, 74–75 paediatric patients 110 myelodysplastic syndrome 63–64 necrotizing pneumonia 115–116 neutrophil counts 30, 31 Nocardia 35 non-invasive positive-pressure ventilation 90, 96, 101 North America 131–141 clinical features of SARS 132 HRCT findings in SARS 136–140 radiographic features of SARS 132–136 nosocomial infections in SARS 103 see also under pneumonia nosocomial transmission of SARS 11 nuclear medicine, infection control measures 157 nurses 8 older people 4–5, 8 intensive care 99 prognosis 31 symptoms 8 opportunistic pathogens 48 osteonecrosis, post-SARS 81, 160–161, 162 Outbreak Control Team (OCT) 14 outbreaks future plans see plans, future management 14–15 outcomes of SARS 8 after intensive care 103–105 chest X-ray changes and 67–68 in children 128–129 Hong Kong University Hospital 31 oxygen saturation, arterial 90, 91, 100, 101 oxygen therapy, supplemental 90, 96, 101, 106 infection control aspects 152

180

paediatric patients see children parainfluenza virus 39 parapneumonic effusions childhood pneumonia 115, 116 pathology, lung 53–54, 89 on follow-up 79–80 patients infection control measures 151–152 safety guidelines 26 segregation 152–153 superspreaders/superspreading event 12–13 pentaglobin 95, 97 personal protective equipment (PPE) changing areas 153 changing into/out of 150–151 disposal of used 151, 153 emergency department staff 19 guidelines on use 26 intensive care unit 105–106 patients 152 radiology staff 150–151 SARS screening clinic 25, 26 Philippines 2, 7 picture archiving and communication system (PACS) 55–56, 173 plans, future 169–173 community 170 global scale 169–170 hospitals 170–172 radiology departments 172–173 plasma, convalescent 62, 95–96, 101 platelet counts 9, 30 pleural effusions 58–59 bronchopneumonia with 37, 38, 39 childhood pneumonias 110, 111 lobar pneumonia with 35, 36 SARS in North America 138, 139 tuberculosis 43 viral pneumonias 41 pneumatocoele 35, 36, 37, 42 in children 117 pneumococcal pneumonia see Streptococcus pneumoniae pneumonia Pneumocystis carinii pneumonia (PCP) 41–42, 43, 43 pneumo-mediastinum 65, 66, 67, 100

I N D E X

pneumonia aspiration 36–37 atypical 1, 58–59, 74–75 in children see under children chronic eosinophilic (CEP) 58, 75–76 classification 33–34 community-acquired 35, 36 differential diagnosis 19, 58–59 imaging 33–52 immunosuppressed patients 35, 38 interstitial 33, 42 interstitial infectious 39–43 lobar 33, 34–36 necrotizing 115–116 nosocomial (hospital-acquired) 35, 36, 38, 103 pathology 53–54 radiographic pattern 34–43 round 111 SARS clinico-radiological correlation 65, 66–67 intensive care 100 radiological diagnosis 54, 56–57, 58 viral see viral pneumonias pneumonitis SARS 54 viral 53–54 pneumothorax spontaneous 65, 66, 67, 100 ventilated patients 103 polymerase chain reaction (PCR) 22, 31 positive end-expiratory pressure (PEEP) 96, 102 post-SARS sequelae 159–163 post-traumatic stress disorder 163 potassium, serum 30 PPE see personal protective equipment prednisolone 61, 62 paediatric SARS 128 pregnant radiographers 144, 154 pressure control ventilation 90, 96, 102 prevention and control 13–15 probable cases CDC definition 167 in children 122

infection control measures 152 management 27 WHO definition 3, 19 prodrome, febrile 8, 9 prognostic factors future trial needs 169 Hong Kong University Hospital 31 Pseudomonas aeruginosa pneumonia 35, 36, 37–38 psychological sequelae 162–163 public education 15 public health services future plans 169–170 outbreak management 14–15 pulmonary abscess 36, 37, 49 in children 116–117 pulmonary embolism, ventilation scanning 157 pulmonary fibrosis after childhood pneumonia 119 post-SARS 80 ARDS 104 chest X-ray features 82 HRCT appearances 80, 81, 83–84, 85 in North America 139 patient characteristics 85–86 pulmonary sequestration 114 pulsation artefacts, HRCT 76 quality of life, post-SARS 162 quarantine contacts 14 radiographers 144, 147 radiographers 143–148 emotional impact of SARS 146–147 infection control measures 150–151 modification of routine practice 144–146 perspective on SARS outbreak 148 SARS infection 149 staffing arrangements 154 radiography digital 55, 144 emergency department screening 20–21, 22 infection control measures 144–146, 155 see also chest X-ray

181

I N D E X

radiology see imaging radiology department future plans 172–173 infection control measures 149–158 emotional impact on staff 146–147 for managers 152–155 for patients 151–152 for specific modalities 155–157 for staff 150–151 randomized controlled trials see clinical trials rats 165 Red Alert 172 reporting 13 respiratory failure 30, 63–64 intensive care 99–100, 101 treatment 90, 96 see also acute respiratory phase respiratory rate, increased 90, 91, 96, 100 respiratory secretions 11 respiratory symptoms/signs 8, 20, 30 respiratory syncytial virus infection 34, 39, 40, 110, 123 response to therapy, clinical 91–92 resuscitation, infection control measures 154 reverse transcriptase polymerase chain reaction (RT-PCR), SARS coronavirus 31, 167 ribavirin 61, 62, 90, 94 efficacy 92, 93 paediatric SARS 128 toxicity 92–93, 94 rigors 8 risk stratification paediatric patients 129 patients 152 ritonavir 94 routes of transmission 10–11 salbutamol, nebulized 29 SARS coronavirus (SARS-CoV) antiviral agents active against 92–94 diagnostic testing methods 22, 31 isolation in cell culture 31, 167 origin 15, 165 RNA detection/RT-PCR 22, 167

serological (antibody) tests 22, 31, 167 transmission see transmission SARS report under investigation (SARS RUI) 167 satellite X-ray units 144–145, 155 screening clinic 23–24 contact 24 emergency department 17–28 staff safety 25–26 septal thickening, interlobular 72, 73 follow-up 82–83, 84 North American patients 137 SARS-related ARDS 104, 105 viral pneumonias 139 sequelae, post-SARS 159–163 sequestration, pulmonary 114 shoes 150 showering facilities, radiology department 153 screening clinic staff 25, 26 ‘signet ring’ sign 119 Singapore 1, 2, 4, 7 small particle aerosol generator (SPAG) 129 sodium, serum 30 staff, health care see health care workers Staphylococcus aureus nosocomial, in SARS 103 pneumonia 36, 37, 38, 39, 110 Stenotrophomonas maltophila 103 Streptococcus pneumoniae (pneumococcal) pneumonia 34, 35, 36 in children 110, 111 complications 116 Streptococcus pyogenes pneumonia 36 stress, in radiographers 146–147 student radiographers 144 superspreaders/superspreading event patient 12–13 surfaces, cleaning 25, 26, 154 surveillance 13, 169–170 staged approach 169 suspect cases in children 122 indications for HRCT 70–71 infection control measures 151–152

management 26–27 WHO definition 3, 19 Swyer–James syndrome 118 symptoms 7, 8, 30 at follow-up 79 synchronized intermittent mandatory ventilation (SIMV) 90, 96, 102 Taiwan 2, 7 taxis, transmission within 11 teenagers 121, 122 Thailand 7 thrombocytopaenia 9, 30 tidal volume 102 training, staff 153 transaminases, serum 9, 30, 132 transmission 10–13, 149 breaking the chain 13 in community 11–12 in hospital 11 prevention and control 13–15 routes 10–11 transport, SARS patients 106 travel emergency recommendations 1–2 precautions 15 treatment of SARS 89–97 antiviral agents 92–94 in children 128–129 clinical patterns during 62–65 clinical response 91–92, 93 clinical trials 168 convalescent plasma 95–96 general approach 89–91 immuno-modulators 94–95 intensive care unit 101–103 new 96 predictors of post-SARS lung fibrosis 86 protocol 61–62, 89–90, 91 radiological patterns during 62, 63, 64 role of chest X-ray 67 ventilatory support 96 tree-in-bud appearance 139 tuberculosis 44, 45, 47, 112, 113 tuberculomas 45 tuberculosis (TB) 43–45, 47 in children 43, 111, 112, 113 miliary 44, 45, 46, 47, 112, 113

182

tuberculosis (Continued) post-primary 44–45 primary pulmonary 43–44 vs non-tuberculosis mycobacteria 46, 47 ultra-high-risk (UHR) wards paediatric patients 129 ultrasound, infection control measures 156 United States (USA) 7, 131 see also North America urine 11 varicella-zoster virus pneumonia 39–40, 41

I N D E X

vasopressors 102 ventilation lung scanning 157 ventilatory support 96 indications 90, 91, 96, 101–102 outcomes 104 Vietnam 1, 2, 5, 7 viral pneumonias 39–41, 42, 75 children 39, 109–110, 112 differential diagnosis 123, 139 pathology 53–54, 79–80 viral replication phase (phase 1) 62–63, 89, 90 chest X-ray features 65 virological testing 22, 31 virus, SARS see SARS coronavirus visitors 129

waiting areas, patient 153 white cell counts 7, 9, 30 World Health Organization (WHO) 1–2 case definition see case definition, WHO early guidelines 19 on origin of SARS virus 165 SARS surveillance 169–170 World Wide Web 170 worldwide number of cases 6 X-ray machines, portable 155 X-ray units, satellite 144–145, 155 Yellow Alert 172