Development of Normal Fetal Movements
Alessandra Piontelli
Development of Normal Fetal Movements The First 25 Weeks of Gestation
13
ALESSANDRA PIONTELLI Department of Maternal/Fetal Medicine Clinica Mangiagalli University of Milan Milan, Italy E-mail:
[email protected]
With the assistance of: Luisa Bocconi, Chiara Boschetto, Elena Caravelli, Florinda Ceriani, Isabella Fabietti, Roberto Fogliani, Alessandra Kustermann, Umberto Nicolini✝, Sarah Salmona, Beatrice Tassis, Laura Villa, Cinzia Zoppini. Department of Maternal/Fetal Medicine, University of Milan, Clinica Mangiagalli - Fondazione IRCCS Ca’ Granda, Ospedale Maggiore, Policlinico di Milano, Italy
ISBN 978 88 470 1401 5
e ISBN 978 88 470 1402 2
DOI 10.1007/978 88 470 1402 2 Library of Congress Control Number: 2010923295 © Springer Verlag Italia 2010 This work is subject to copyright. All rights are reserved, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilm or in any other way, and storage in databanks. Duplication of this publication or parts thereof is permitted only under the provisions of the Italian Copyright Law in its current version, and permission for use must always be obtained from Springer. Violations are liable to prosecution under the Italian Copyright Law. The use of general descriptive names, registered names, trademarks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. Product liability:The publishers cannot guarantee the accuracy of any information about dosage and application contained in this book. In every individual case the user must check such information by consulting the relevant literature. Cover illustration: Ikona S.r.l., Milan, Italy Typesetting: Ikona S.r.l., Milan, Italy Printing and binding: Grafiche Porpora S.r.l., Segrate, Milano Printed in Italy Springer Verlag Italia S.r.l. Via Decembrio 28 I 20137 Milan Springer is a part of Springer Science+Business Media
To my mother and to all our mothers
Preface
This work sees the light for various reasons. There is a general lack of detailed information about the earliest stages of human motor development. The reasons for this are explained more fully in the Introduction; here we may simply state that, apart from their intrinsic interest, earlier phenomena are fundamental to the comprehension of later phenomena rooted in them, whether pathological or normal. This is especially so in the rapidly developing young organism. At birth the neonate is catapulted into a profoundly different physical and social environment requiring extremely diverse functioning: suffice it to mention aerial respiration, no longer being fed through the placenta and the cord, and the full impact of gravity on neonatal movements. The neonate generally adapts smoothly to the transition, as it has been equipped to do so during the 9 months of pregnancy. However, the study of the early stages of fetal motor development should not be exclusively directed towards the understanding of functioning in the neonate. Fetuses undergo constant and very rapid changes throughout pregnancy. Equally, the intrauterine environment varies continuously. The young organism is continually reshaped in more or less subtle ways and its functions modify, develop, become redundant or are incorporated almost beyond recognition into subsequent ones. Fetuses are perfectly adapted to each step of their rapid development. Only some functions are fully and wholly anticipatory and uniquely geared toward postnatal life. In other words, at any given gestational age fetuses are to be considered in themselves and in relation to their environmental conditions. On this basis, preparatory or mixed functions and their evolution can be teased out. Besides trying to bring to light a severely neglected topic, some other reasons motivated me to write this work. Relatively recent technologies such as 4D ultrasonography once seemed to hold the promise to revolutionize fetal studies; however, 4D ultrasound produces not real-time images, but computerized reconstructions of the fetus in motion. As such, it fails to capture essential features of most fetal movements. On the other hand, 4D ultrasound does offer some sensational and easily readable images. Researchers with little or no experience in the field are increasingly drawn to the 4D technique by its deceptive accessibility and apparent ease of interpretation. This often results in far-fetched conclusions being reached over very thin ice. In a few years’ time new technologies will certainly bring about a true revolution in fetal studies. However, future studies will also always require a thorough basic knowledge of fetal functioning and how it changes. For the moment this basic knowledge can only be acquired by dipping into a mixture of techniques, using the best each has to offer. With the recent invasion of ultrasound pictures, and ‘special effects’ images which have been totally artificially constructed, fetuses and their behaviour and development are often referred to as ‘wonders of nature’. Almost nothing makes us feel that we are witnessing
VII
Preface
a miracle of nature like watching a new life unfold. Others regard the miracle of life as a gift from God. This latter stance, legitimate as it may be, belongs to religion, not to the study of nature. The sense of wonder, whether based upon compelling religious beliefs or sentimental and emotional reactions, impinges strongly on society’s attitudes, on decisions about the fetus, and ultimately on the lives of many women deemed to be mere containers of the ‘miraculously’ unfolding life. Fetal behaviour is fascinating, but it can be analysed. By relying on observational data, this work hopes to re-establish a balance in favour of evidence. Having spent many years working in this field I felt that the time had come to gather together all the accumulated knowledge about the first 25 weeks of pregnancy, the time span on which I have always focused my attention. My professional background is somewhat multi-faceted. After graduating in medicine I specialized in psychiatry and neurology, trained amongst other things as a ‘baby-watcher’ following the principles of ethology in England, but then went back to medicine and, albeit not formally, became an expert in obstetrics and fetal behaviour by pursuing my research in the main maternity hospital in Italy, my native country, for almost 20 years. By propounding a fresh look at our first movements and the rapid changes they undergo, I hope to elicit renewed interest in the reader in this vastly unexplored but fascinating field. I also hope this will result in further research bringing about true advances in our still rudimentary understanding of this neglected area of knowledge. Milan, April 2010
Alessandra Piontelli
Acknowledgements
The help and support of many individuals were critical in making this work a reality. These include former and current staff, students, and nurses of the Department of Maternal Fetal Medicine at the Clinica Mangiagalli of the University of Milan. I am greatly indebted to Professor Fedele for allowing me full access to the Clinic. This book could not have seen the light of day without the help and the teachings of the late Umberto Nicolini, of Alessandra Kustermann, and of, in alphabetical order: Stefano Acerboni, Luisa Bocconi, Chiara Boschetto, Elena Caravelli, Florinda Ceriani, Isabella Fabietti, Roberto Fogliani, Leo Gallo, Sarah Salmona, Beatrice Tassis, Laura Villa, and Cinzia Zoppini. Most of these persons will be further acknowledged in each chapter for helping me collect the material and for independently reviewing it. Thank you Lucy for always being so available and kind. The pregnant mothers who so generously agreed to be part of all these studies have obviously been essential to this work. I am greatly indebted to all of them. This work could not have been made possible without the help of Carlo Castellano and his enormous generosity in allowing Esaote S.p.A. (Genoa, Italy) to lend me the equipment specifically chosen for the various requirements of this work. Carlo and Ileana, you have been marvellous friends. All the technicians and local directors of Esaote have been excellent in solving the many problems of ultrasonographic research. A special thank to Dr Testa and Dr Schiavi, and to Emilio Busato for technical help. In their special training courses, Heinz Prechtl, Christa Einspieler, Giovanni Cioni, and Paolo Ferrari have been fundamental teachers for learning to observe fetal and neonatal motions. Heinz and Christa were so kind as to ask me to speak at to two meetings in Graz. These symposia were invaluable for meeting extraordinary people, and for learning and testing out some of my thoughts. I am particularly grateful to Peter Wolff for some long discussions in Graz, Boston, and Milan. Many other special colleagues and friends have been at my side at various stages of my work. Particular thanks go to Antonio d’Elia, Daniel Stern, Colwyn Trevarthen, Elizabeth Spillius, Jerry Bruner, the late Mauro Mancia and the late Elizabeth Bryan, and to Diane Garcia and all my friends in Los Angeles. Giannis Kougioumouzakis in Crete has always provided a wonderful arena to test out my ideas and share them with him and with an exceptional audience. I am also grateful to the anonymous reviewers of the manuscript for their valuable suggestions. A very special thank you goes to my editor Donatella Rizza for being so accessible, helpful, and encouraging. Working with her and with all the staff at Springer-Verlag Italia has been a real pleasure. Chris Benton as usual has carefully and patiently revised my English. Andrea Sommaruga
X
has helped me with figures and computer work, and Sergio Belfiore with photographic material. Luigi with Filippo and Roberto have been always lovingly and unfailingly by my side. Ultimately, of course, the responsibility for what I say in this book remains mine alone.
Acknowledgements
Contents
1
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
1
General Movements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
7 8 11 12 16 16
1.1 1.2 1.3 1.4 1.5
2
Startles, Twitches and Clonuses. . . . . . . . . . . . . . . . . . . . . . . . . . . 19 2.1 2.2 2.3
3
19 25 27
Hiccups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Yawning. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Gasping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
29 33 36
Fetal Breathing Movements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 4.1 4.2 4.3 4.4 4.5
5
Startles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Twitches. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Clonuses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Hiccups, Yawning and Gasping . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 3.1 3.2 3.3
4
General Movements: 7 16 weeks . . . . . . . . . . . . . . . . . . . . . . . . . Length of the Feet and Epidermal Ridges . . . . . . . . . . . . . . . . . . . General Movements: 17 25 weeks . . . . . . . . . . . . . . . . . . . . . . . . General Movements: Frequency and Duration . . . . . . . . . . . . . . . Central Pattern Generators. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Fetal Breathing Movements: General Features . . . . . . . . . . . . . . . Fetal Breathing Movements: Non-Coincidence with other Behavioural Events . . . . . . . . . . . . . . . . . . . . . . . . . . . Apnoeic Pauses . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Possible Functional Significance. . . . . . . . . . . . . . . . . . . . . . . . . . Neurological Substrate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
40 41 42 44 45
Swallowing, Sucking, and Handedness as Inferred from Fetal Thumb Sucking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 5.1 5.2 5.3 5.4 5.5
Swallowing and Sucking: General Features . . . . . . . . . . . . . . . . . Swallowing: Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Fetal Swallowing: Possible Functions. . . . . . . . . . . . . . . . . . . . . . Swallowing: Possible Regulation . . . . . . . . . . . . . . . . . . . . . . . . . Handedness in the Human Fetus as Assessed by Thumb-Sucking . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
49 50 53 54 54
XI
Contents
6
Localized Movements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59 6.1 6.2
7
Basic Emotions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Cross-Modal Integration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Preparing for Post-Natal Communications . . . . . . . . . . . . . . . . . . Parental Reactions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Yawning: a Form of Communicating? . . . . . . . . . . . . . . . . . . . . .
Sleep in Children . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Behavioural States in Premature Infants and Mature Fetuses . . . Early Fetal Functioning: Rest-Activity Cycles and Clusters of Activities. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8.3.1 Ontogeny of Sleep and its Possible Precursors . . . . . . . . . . . . . . .
88 88 89 92
Twin Fetuses and Twin Myths . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97 9.1 9.2 9.3 9.4 9.5 9.6 9.7 9.8
10
78 80 82 83 85
Rest-Activity Cycles, Clusters and the Ontogeny of Sleep . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 87 8.1 8.2 8.3
9
61 71
Facial Expressions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 77 7.1 7.2 7.3 7.4 7.5
8
Hand and Arm Movements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Leg Movements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .
Beginnings of Intrapair Stimulation and its Relevance for Our Knowledge of the Sensory Capacities of All Fetuses. . . . Features of Rest Cycles Revealed by Twins . . . . . . . . . . . . . . . . . Similarities and Differences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Behavioural Individuality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Universal Myths . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Twins: Open to Mutual Communication . . . . . . . . . . . . . . . . . . . . Maternal Emotions and their Impact on the Twin Fetus . . . . . . . . Bereavement in the Twin Fetus . . . . . . . . . . . . . . . . . . . . . . . . . . .
97 99 100 104 104 105 105 106
Conclusions. Movement is Life . . . . . . . . . . . . . . . . . . . . . . . . . . . 107 10.1 10.2 10.3 10.4 10.5
Fetal Movements: Varied and Varying Functions . . . . . . . . . . . . . Shaping a Sense of our Boundaries. . . . . . . . . . . . . . . . . . . . . . . . Building a Body Schema and a Proto-Sense of Self . . . . . . . . . . . Forming the Cortical Homunculus and its Curious Layout?. . . . . Building on Expressive Repertoire . . . . . . . . . . . . . . . . . . . . . . . .
107 109 110 111 112
Glossary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115 Subject Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131
Introduction
Keywords Fetus • Neonate • Movements • Gestation • Premature • Ultrasound
Normal human fetal movements during the first 25 weeks of gestation have generally been overlooked. The beginnings of fetal activities have in the main been regarded as chaotic and disorganized, and as such not worthy of detailed investigation. In reality, fetal movements during the first 25 weeks of pregnancy are simply organized differently to those in later periods, and are in any case functional to the various stages of development of that time span. However, quite apart from the neglect suffered by this specific period, all research on human fetal movements seems to have come to a standstill in recent years, whilst myths surrounding our first movements have flourished. Our origins have always fascinated us, and the origins of our movements are particularly fascinating as they initiate many vital phenomena. Historically, Aristotle can be considered the father of embryology. He was the first scholar to challenge the traditional view of existence as beginning only at birth; his investigations began at or soon after conception and were always grounded in direct observation. Based as it is on the observation of movements from their very beginning, in this particular sense this work could be regarded as ‘Aristotelian’. Aristotle dissected many animals, possibly including human embryos, and observed their changes throughout the various stages of pregnancy. He attributed increasing degrees of animation to the growing embryo. According to his theories, the embryo started its development in a ‘vegetative’ state. A ‘sentient’ phase then followed until 16 20 weeks, when ‘quickening’, the stage of pregnancy at which the mother begins to sense fetal movements, began and the embryo was thought to have irrevocably achieved a ‘raA. Piontelli, Development of Normal Fetal Movements. © Springer Verlag Italia 2010
tional soul’. Following Aristotle, many scholars continued to regard quickening as the sign that the fetus had finally emerged from the chain of its former vegetable and animal hazy states and attained ‘ensoulment’ or ‘animation’, a term derived from the Latin anima (soul). Other scholars, mainly basing their interpretations on analysis of the Old Testament, considered the first breath as the true beginning of human life. Still others regarded viability as the indication that the fetus had fully entered the human world. Opponents of all these theories of ‘delayed ensoulment’ contended that the embryo possessed a soul from conception and was fully human long before it quickened and long before it took its first aerial breath. These concepts were debated for centuries amongst various religious communities, and similar arguments, though refined and updated, are still largely at the basis of pro-choice and pro-life debates. When the fetus attains a ‘rational soul’ or, as we would call now it, ‘awareness’ and ‘consciousness’, continues to be an unanswered question which will not be touched upon in this book. The very concepts of consciousness and awareness are far from being clarified and still await both differentiation and convincing consensual definitions. Furthermore, the attainment of consciousness is not an either/or phenomenon, but one that is built gradually and takes on different gradations and shades during its development. It may possibly be easier to state when a fetus cannot sustain consciousness than when it can. It was only when scientists stopped looking for the ‘soul’ that a truly observational and scientific interest in the fetus began. Towards the turn of the nineteenth century, several disciplines split from philosophy, making it
2
possible for proper fetal research grounded in reality to start. William Preyer, a German embryologist and physiologist, can be considered the true father of fetal studies. Preyer wrote in 1885 a seminal treatise, ‘Spezielle Physiologie des Embryo’(‘Special Physiology of the Embryo’), in which he dealt with the sensorimotor functions of the human fetus, giving expression to many far-reaching intuitions [1]. However, the means available to early scientists were very limited. They could only observe the excursions of the maternal abdominal wall or try to infer the behaviour of the fetus from that of premature infants. After Preyer’s pioneering work, the subject stagnated for about 50 years, and it was only between the 1920s and the 1940s that fetal studies saw a remarkable resurgence. Several scholars observed fetuses after spontaneous miscarriage or abortion by caesarean section. Significant contributions were made by the Swiss neurologist and psychiatrist Myeczyslaw Minkowski [2], who observed the behaviour of surgically removed fetuses; by Davenport Hooker [3], an American anatomist, who added cinematic documentation and tactile stimulation to the observation of moribund human aborted fetuses; and by Hooker’s collaborator Tryphena Humphrey [4]. Their investigations, pursued by testing and directly observing the early human fetus, provided numerous basic concepts, many of which are still fundamental today. Another historical classic is Arnold Gesell’s The Embryology of Behavior, published in 1945, which describes and illustrates the physical and behavioural development in the human from embryo to fetus, and from fetus to neonate [5]. Other relevant contributions came from scientists observing and experimenting with various animals. However, the majority of the studies performed in these years were strongly influenced by the otherwise remarkable work of Sherrington and Pavlov centring on reflexes, and as a consequence of this, a ‘reflexogenic’ view of fetal motions became prevalent whereby they were considered to be invariably generated by unidentified stimuli. After this extraordinary blossoming, the subject was largely abandoned until the introduction of ultrasound in the mid-late 1970s. Ultrasound, by offering the unprecedented possibility of observing the undisturbed and unharmed fetus in its natural environment, opened a completely new era for the study of fetal activities, which not surprisingly underwent an unparalleled renaissance. The belief in the reflexogenic nature of human fetal movements was promptly and fully dispelled. Ultrasonographic investigations of human fetal motor development were pioneered by Birnholz and colleagues [6], Ianniruberto
Introduction
and Tajani [7], and deVries, Visser and Prechtl [8 10]. The studies of Prechtl and his followers have been particularly influential. Prechtl, a developmental neurologist, was initially a student of Konrad Lorenz, one of the founders of ethology. Ethology is the branch of zoology studying animal behaviour within its natural environment as well as in the laboratory. The pictures of Lorenz immersed in a pond observing the greylag goose or walking about followed by a young ‘imprinted’ duck have become icons of naturalistic studies. Prechtl applied many of the principles of ethology to fetal studies. Amongst other things, he furnished an accurate account of the evolution of fetal movements throughout pregnancy and a so-called ‘ethogram’, the description and definition of various fetal movements. This classification, based as it was on terminology used for the premature and the neonate, has been almost universally adopted, fostering mutual understanding amongst researchers in the field. Grounded as they are in the observation of the human fetus within its natural milieu, studies of fetal movements, including the present volume, are largely rooted in ethology. Niko Tinbergen, the other founder of ethology, who won the Nobel Prize for Medicine in 1974 together with Konrad Lorenz and Karl von Frisch, also had a significant influence on fetal studies. Tinbergen thought that scientists always needed to pay attention to four fundamental kinds of explanation when faced with any behaviour: function, causation, development, and evolutionary history [11]. These questions are still of primary relevance for those attempting to study human fetal behaviour. Whenever feasible each motor phenomenon described in this book will be examined in the light of Tinbergen’s four queries. However, to return to where we started: despite the importance of Prechtl’s leading work and perhaps because of its intimidating significance except for a few valuable additions, the study of human fetal motor functioning, in particular during the first 25 weeks of pregnancy, seems to have come to a halt. According to deVries, one of the pioneers in the field, human fetal motor research peaked between 1980 and 1990 and declined thereafter [12]. Even during this ‘golden age’ the first half of pregnancy was usually overlooked. Out of 109 relevant articles examined by deVries, 83% dealt with the second half of pregnancy only. As a result of this stagnation, the same data are quoted repeatedly and fairly uncritically, and our knowledge of the origins of fetal movements can be said to be still at the embryonic stage. Apparently contrasting with this stagnation in fetal
3
studies, a great wave of visually enthralling depictions of the ‘marvels’ of the womb have contributed to make many pseudo-experts believe that all the mysteries have been solved. Not only that, but since early fetuses move a lot, neglecting this period has fostered an ever growing body of myths and legends. Early fetuses are often regarded as highly evolved creatures functioning even better than neonates. Several reasons lie behind this state of affairs. Other branches of science investigating prenatal life have advanced enormously, but these advancements have barely touched on fetal movements. Embryologists have given us invaluable insights into the earliest stages of development by describing molecular, cellular and structural bodily changes and their evolution with minute accuracy. However, embryology focuses primarily on the initial phase of life the so-called embryonic period, which ends between 7 and 8 weeks of gestation. Furthermore, embryologists work mainly at a molecular and cellular level or on anatomical preparations derived from dead animals, including aborted human fetuses. Because of this methodological tendency, most embryologists, of necessity, do not take into account a relevant aspect of all embryos and fetuses: their frequent movements. Unlike embryologists, obstetricians are directly involved with the live fetus and its maternal environment. However, obstetricians are understandably not much interested in fetal movements, a topic which they tend to regard as secondary, especially during a pre-viable phase of development. Up until mid-gestation, obstetricians focus primarily on a variety of crucial matters ranging from ruling out ectopic implantation, establishing pregnancy dating, chromosomal testing and checking fetal anatomical morphology. Most of these concerns have an important bearing on clinical and parental decision making, and on determining the type and frequency of subsequent examinations. The time span from 22 to 25 weeks’ gestation is a grey area where survival ex utero is becoming increasingly possible, albeit often at high cost. Clearly, the survival rates of extremely premature infants (defined as infants born before 28 completed weeks of gestation) increase with gestational age. Currently the chances of survival are almost nil at 22 weeks and approximately 60% at 25 26 weeks [13]. Despite enormous medical progress, surviving infants born at less than 26 weeks’ gestation still show very high rates of life-long consequences including cerebral palsy (up to 36%), mental retardation (up to 47%), chronic lung disease (up to 61%),
blindness (up to 25%), and deafness (up to 7%) [14]. Neonatologists caring for these infants predominantly direct their clinical and research efforts on trying to keep these infants alive and avoiding all the above-mentioned complications. Motor functioning in the premature infant is looked at with a clinical eye as an indicator of whether current and future functioning will be normal or pathological. In addition, despite the many continuities between prenatal and postnatal life, obstetrics and neonatology have remained largely separate fields. Perinatology, the branch of medicine devoted to the perinatal period, should in theory have bridged the gap. As pointed out by the first professor of child development, the psychologist Arnold Gesell in 1945, extremely premature infants, born before 28 weeks, are to be considered ‘fetal infants’ with a dual nature, being at once extrauterine fetuses and prenatal neonates [15]. However, obstetricians specializing in perinatal medicine care for high-risk pregnancies, and perinatologists working in the intensive care unit deal with high-risk infants. While the division is no longer absolute, motor development once more understandably tends to be overlooked as of secondary concern. Developmentalists, on the other hand, have studied in depth the behaviour of the premature infant, trying to infer fetal functioning from it. Unlike the fetus, the premature infant can be directly and continuously observed. Various parameters and functions can be checked and verified. Nevertheless, extremely premature infants are not equivalent or wholly comparable to fetuses in utero. Although such infants largely remain ‘true to their fetality’ by adhering to their natural sequence of maturation, they live in a very different external environment that impacts profoundly upon their functioning [4]. Moreover, developmentalists, by starting the description of the motor and emotional behaviour of the infant from birth, tend to ignore the prenatal history of many neonatal phenomena with which they are not acquainted. The prevailing view of development also stresses a linear progression. Following this outlook the fetus is regarded solely as preparing to become a neonate, just as the neonate is regarded as practising to become an adult, in a linear sequence. Attention to preparatory and anticipatory functions generally prevails. This outlook does not take into account that fetuses inhabit a very different environment from the neonate and even more so from the adult. Furthermore, the intrauterine environment is subject to changes to which the fetus has to adapt. Such adaptations, named ‘ontogenetic adaptations’ by Oppenheim [16], may in-
4
volve particular morphological, biochemical, physiological and behavioural mechanisms which are different from those of the neonate and even more so from those of the adult. All these may require modification or even abandonment before the neonatal stage is attained. Wellorganized neonatal functioning does not spring into action without preceding steps, nor are all fetal activities geared towards postnatal existence. Many are simply functional to a particular phase, and still others have both a preparatory and an adaptive component. On the other hand, those researchers who study fetal behaviour prefer to focus on the movements of the fetus approaching birth and on how they link with those of the neonate. At this stage the physician has several possibilities of intervention, including inducing birth. Research and clinical efforts usually concentrate on detecting signs of distress and alarm. In addition, fetal movements during the early stages of pregnancy are commonly considered chaotic and ‘disorganized’ and as such have been generally disregarded. In contrast with this, all movements to be found in the near-term fetus are widely considered to be already present by 16 weeks’ gestation. Whilst it is true that the overall gestalt of fetal motions allows a particular movement to be identified and classified using the same designation throughout, such motions are nevertheless executed in varying ways at different gestational ages. During pregnancy, fetal dimensions, proportions, neurological and generally speaking physiological functioning, as well as intrauterine factors all change at a dazzling pace. Growing fetal constriction is just one example of macroscopic change. If only for this reason, movements cannot be performed uniformly at 10 weeks’ and at 25 weeks’ gestation. Another important reason for the lack of advances in the study of early fetal movements has been the wait for new technological developments. So far, dynamic ultrasonography offered a two-dimensional (2D) and only partial picture of the fetal body, especially in the second half of pregnancy. Three-dimensional and, more recently, 4D ultrasonography seemed to hold great promise for fetal studies, and researchers have been waiting for these and other new refinements. Three-dimensional ultrasonography can give us beautifully detailed static pictures of the fetal body. These images have enhanced our knowledge of fetal anatomy, as well as providing a valuable tool for detecting some anatomical malformations. However, good 3D pictures are not always easy to obtain. For instance, when the uterus becomes crowded it becomes almost impossible to visualize the fetal face if the cord or
Introduction
the hands are in the way. In other words, the target of a study is easily lost. Four-dimensional ultrasonography has added the dimension of movement to the images the fourth dimension referred to in the name. Four-dimensional ultrasound has shown that fetal movements start earlier than was thought, and allow more precise reading of some movements. The work of Asim Kurjak has been particularly influential in this field [17,18]. However, 4D images (like 3D images) are computerized reconstructions of fetal motions, and as such do not allow us to capture many of the features of real-time movements, such as speed, quality, tempo and precise sequence, or to visualize very fast movements such as breathing, startles or hiccups. These are not secondary issues for the study of movements. Nor are fast movements of secondary relevance for fetal functioning. Sensational 3D and 4D images are beginning to invade all sorts of publications ranging from obstetrics treatises to magazines. Frequently these are touched up using Photoshop or similar programs in order to make them look more appealing. A true revolution in fetal studies will be brought about by the advent of real-time 4D ultrasonography. In the observations reported in this volume these new technologies have been used alongside traditional 2D realtime ultrasonography. The most rewarding application of 4D ultrasonography turned out to be the investigation of the beginnings and evolution of hand movements and of various facial expressions. In contrast to the relative decline of human fetal studies, those of animal fetuses, because they are open to various kinds of experimentation, have continued to flourish and give us fundamental information which for obvious reasons cannot be obtained from the human fetus. The leap from other species to humans is not always pertinent; however, whenever appropriate, animal studies will be referred to throughout this book. The studies of developmental psychobiologists an interdisciplinary field encompassing developmental psychology, biological psychology, neuroscience and many other areas of biology too numerous to mention have all been particularly influential. It must suffice to mention here almost at random the fundamental work of just a few, such as Jeffrey Alberts [18], Myron Hofer [19], George Michael [20, 21], Celia Moore [20], Peter Hepper [22], Scott Robinson [23], Carolyn Rovee-Collier [24] and William Smotherman [23] and their colleagues, who, along with many others, through the use of experimentation
References
have all given us essential understandings into fetal life. Enormous advances have also been made in disparate fields ranging from genetics to molecular biology and histology. While some reference will be made to these, it is beyond the scope of this book, which deals with the macroscopic phenomenon of fetal movements, to discuss in detail the findings of these highly specialized fields. Neuroimaging, particularly using magnetic resonance imaging (MRI), is also contributing a great deal of information on the growth and development of the fetal body in general and of the central nervous system in particular. However, a thorough discussion of the rapidly accumulating data is also beyond the scope of this book. This volume aims to serve other researchers in the field, clinicians, developmentalists, and ultimately parents if possible. By understanding normal motor functioning, obstetricians may become alert to unusual phenomena which could in turn lead to further, lengthier and longitudinal observation. Neurologists may be encouraged to derive from it meticulous fetal ‘developmental milestones’ that might eventually lead to a long-awaited neurological examination in utero. A thorough knowledge of normal early fetal functioning may assist neonatologists and paediatricians in understanding phenomena that turn out to be pathological in the premature infant and, hopefully, may lead to the devising of new ways to cope with them. It may aid developmentalists to judge which phenomena are rooted in the physiology of our prenatal past and to distinguish these from other phenomena that arise ex novo or belong to the pathology of the present. Finally, parents of severely premature infants may also be helped to look at their tiny babies in a different way. Knowing that fetuses of the same gestational age are barely social creatures may help them to feel less frustrated and at the same time less guilty for not providing ‘adequate’ care and stimulation to their fetal infants. They may also start regarding some phenomena such as gaze avoidance or hiccups not as signs of ‘stress’, ‘avoidance’ or ‘refusal to relate’, portending psychological catastrophes later on, but as perfectly functional phenomena linked to the true age of their infants. Since this book is addressed principally to specialists in various fields other than fetal studies, the language has been kept as plain and as descriptive as possible, and a glossary has been added at the end. All observations discussed in this book are derived from both general and targeted investigations. Each observation (no matter whether general or targeted) lasted an hour, took place at the same time of the day after a
5
standard meal, was recorded on DVD or tape, and was later analysed off-line by two or more researchers. Save for twins, a cross-sectional approach was utilized, and each week 30 subjects were investigated for a general view and 30 more for a targeted one in relation to fetal breathing movements, leg and arm movements, and sucking and swallowing motions. Thirty pairs of twins were observed longitudinally twice a month from week 10 till the end of pregnancy. Facial expressions were studied cross-sectionally in 10 women from week 10 to week 25 with 3D and 4D ultrasounds. Throughout this book ‘weeks’ refer to gestational age calculated from the date of the last menstrual period not to conceptional age. As to the organization of this book, Chapter 1 deals with the most noticeable motor phenomena arising first, general movements. Chapter 2 deals principally with startles, another frequent movement which starts during early prenatal life, and with twitches and clonuses, in order of decreasing frequency. Chapter 3 deals with three phenomena whose functional significance is still largely obscure: hiccups, yawning, and gasping; all three, however, have been linked to some extent with fetal breathing. Chapter 4 describes fetal breathing motions. Chapter 5 describes swallowing and sucking motions and discusses handedness as inferred from fetal thumb sucking. Chapter 6 deals with localized motions, focusing in particular on arms and hand movements as well as those of the legs. Chapter 7 describes the beginning of facial expressions and their evolution. Chapter 8 deals with rest activity cycles, clusters of various motions, and with some hypotheses on the ontogeny of sleep. Chapter 9 describes differences and similarities in the motions of twin fetuses and discusses their meaning for all fetuses. In addition, the host of ‘fetal myths’ that burden twin fetuses are analysed and dispelled. Finally, in Chapter 10 some considerations of a more speculative nature are discussed.
References 1. Preyer W (1885) Spezielle Physiologie des Embryo. Grieben, Leipzig 2. Minkowski M (1922) Über frühzeitige Bewegungen, Reflex und muskuläre Reaktionen beim menschlichen Fötus and ihre Beziehungen zum fötalen Nerven and Muskelsystem. Schweiz Med Jahrbuch 52:721 724 and 751 755 3. Hooker D (1952) The prenatal origin of behavior. University of Kansas Press, Lawrence, Kansas 4. Humphrey T (1978) Function of the nervous system during prenatal life. In: Stave U (ed) Perinatal physiology, 3rd edn.
6 Plenum, New York 5. Gesell A (1945) The embryology of behavior. In: Classics in de velopmental medicine, 2nd edn (1988). Lippincott, Philadelphia 6. Birnholz JC, Stephens JD, Faria M (1978) Fetal movement patterns: a possible means of defining neurologic develop mental milestones in utero. Am J Roentgenol 130: 537 540 7. Ianniruberto A, Tajani E (1981) Ultrasonographic study of fetal movements. Semin Perinatol 5:175 181 8. de Vries JIP, Visser GHA, Prechtl HFR (1982) The emergence of fetal behaviour. 1: Qualitative aspects. Early Hum Dev 7:301 322 9. de Vries JIP, Visser GHA, Prechtl HFR (1984) Fetal motility in the first half of pregnancy. In: HFR Prechtl (ed) Continuity of neural functions from prenatal to postnatal life. Spastics International Medical Publications, London 10. de Vries JIP, Visser GHA, Prechtl HFR (1988) The emergence of fetal behaviour. 3: Individual differences and consistencies. Early Hum Dev 16:85 103 11. Tinbergen N (1951) The study of instinct. Oxford University Press, Oxford 12. de Vries JIP, Fong BF (2006) Normal fetal motility: an overview. Ultrasound Obstet Gynecol 27:701 711 13. Brodsky D, Oullette MA (2008) Transition of the premature infant from hospital to home. In: Brodsky D, Oullette MA (eds) Primary care of the premature infant, chap 1. Saunders, Philadelphia, pp 1 8 14. Wilson Costello DE, Hack M (2006) Follow up for high risk neonates. In: Martin RJ, Fanaroff AA, Walsh MC (eds) Care of the premature infant, vol 2, 8th edn, chap 39. Mosby, Philadelphia, pp 1035 1043
Introduction 15. Gesell A (1945) The embryology of behaviour: the beginnings of the human mind (1988 edn). Mac Keith Press, London 15. Oppenheim RW (1984) Ontogenetic adaptations in neural development: towards a more ‘ecological’ developmental psychobiology. In: HFR Prechtl (ed) Continuity of neural functions from prenatal to postnatal life. Spastics International Medical Publications, London 16. Jackson D, Kurjak A (eds) (2004) An atlas of 3D and 4D sonog raphy in obstetrics and gynecology. Informa Healthcare, London 17. Kurjak A, Azumedi G (eds) (2007) The fetus in three dimen sions: imaging, embryology and fetoscopy. Informa Health care, London 18. Ronca AE, Alberts JR (2000) Physiology of a microgravity environment. Selected contribution: effects of spaceflight during pregnancy on labor and birth at 1 G. J Appl Physiol 89:849 854 19. Hofer MA (1995) Roots of human behavior. WH Freeman Company, Plymouth, Michigan 20. Michel GF, Moore CL (1995) Developmental psychobiology: an interdisciplinary science. MIT Press, Cambridge, Massa chusetts 21. Hopkins B, Barr RG, Michel GF et al (eds) The Cambridge encyclopedia of child development. Cambridge University Press, Cambridge, UK 22. Hepper PG (ed) (1991) Kin recognition. Cambridge Univer sity Press, Cambridge, UK 23. Smotherman WP, Robinson SR (1990) Behaviour of the fetus. CRC Press, Boca Raton 24. Rovee Collier C, Lipsitt LP, Hayne H (eds) (2000) Progress in infancy research, vol 1. Lawrence Erlbaum, Philadelphia
General Movements
1
With the assistance of Florinda Ceriani, Isabella Fabietti, Roberto Fogliani and Alessandra Kustermann
Keywords General movements • Central nervous system • Central pattern generators • Motor patterns • Brainstem • Lordosis • Opisthotonus • Kyphotic • Volar pads • Popliteal angle • Hypotonia • Myotubes • Primitive stepping • Supine kicking • Nuchal tone
General movements, also called total pattern or holokinetic movements (from the Greek holos meaning whole and kinema meaning motion) emerge as periodic bursts of whole-body activity and are one of the earliest and most dramatic forms of fetal movement. In 1929, the American anatomist Coghill, studying the aquatic stage of development of the amphibian lizard Amblystoma, was the first to describe and distinguish between ‘total and partial patterns of motion’. Coghill postulated that the same distinction could apply to human fetuses, which he viewed as similar to amphibians: creatures living in an aquatic medium, but preparing to enter a terrestrial one [1]. Then, in the early 1950s, Hooker noted that local movements appeared later in development than ‘total pattern’ ones [2]. Because his observations were performed directly on human fetuses, Hooker’s remarks are still widely quoted in reference to the emergence of particular fetal sensitivities. However, it was only with the advent of ultrasonography that naturally occurring general movements were observed in the healthy fetus living within its natural environment and differentiated from localized motions. Using ultrasound, the Austrian physician Reinold [3] was the first to distinguish strong and brisk movements involving the entire body from movements ‘confined to fetal parts’. Similar descriptions were given by other physicians, especially Juppila [4] and Van Dongen and Goudie [5]. During the 1980s, detailed classifications of various movement patterns were expounded. AlA. Piontelli, Development of Normal Fetal Movements. © Springer Verlag Italia 2010
though these taxonomies were similar, the one developed by Prechtl and his collaborators [6] eventually became the one universally adopted in scientific circles, due partly to its particular accuracy and partly to Prechtl’s own renown in the field of child neurology. General movements were described as non-stereotyped motions in which ‘the whole body is moved but no distinctive pattern or sequence of body parts can be recognized… Movements of limbs, trunk and head are rapid, but smooth in appearance.’ Prechtl and his collaborators noticed the first appearance of general movements at between 8 and 9 weeks, and described their qualitative changes in time as follows. At 8 9 weeks general movements were said to be slow and limited in amplitude. At 10 12 weeks they were described as more forceful and rapid, with limbs, trunk and head all involved in the motion. After 12 weeks general movements became ‘more variable in speed and amplitude’, and their duration was noted to vary from about 1 to 4 min. As Prechtl said, general movements ‘wax and wane’, and, ‘however variable these movements are, they are always graceful in character’ [6]. Besides these qualitative aspects, quantitative characteristics and individual consistencies and variations were described. In parallel with this widely quoted and authoritative work, interpretations attaching an emotional meaning to fetal movements have continued to proliferate. The most frequent comment one hears from parents and obstetricians alike when a fetus starts a burst of generalized motion
8
is: ‘Look, it is waking up.’ Given their dramatic quality, general movements are also often interpreted as signs of hyperactivity and anxiety, whereas, in fact, brief episodes of wakefulness are a late acquisition in pregnancy, and anxiety-driven states of hyperactivity only belong to life after birth. Technical refinements, in particular transvaginal and 4D ultrasonography, have shown general movements to start about 1 week earlier than was previously thought, right at the start of the fetal period, between 7 and 8 weeks [7]. From the very beginning general movements are not stereotypical, but performed in unpredictable combinations, constantly adapted to the ever changing internal and external environmental conditions. Even during the initial, greatly ‘simplified’ stages, general movements are never exactly alike. The fetus can start coiling slightly on one side of the body rather than the other; the upper or the lower segment of the trunk can be involved; a slight extension of the head, a little stretching of the spine or a rudimentary movement of one limb can follow. All combinations are different. Twins, because they allow simultaneous observation of two fetuses using one or the other co-twin as a control, highlight beautifully the non-stereotyped quality of general movements [8]. With advancing gestation, the complexity and variation of movement increase enormously. Body proportions, muscle length and composition, ossification (from the Latin ossae meaning bones), connections with and within the central nervous system, and spatial constrictions, to name but a few factors, change at a dazzling pace, allowing fetuses to perform new modes of movement. Adaptation and calibration of motor patterns takes place as the fetus grows and develops. At the same time, some forms of movement become redundant or impossible.
1.1 General Movements: 7-16 weeks Before 10 weeks, general movements are barely perceptible and have an oscillatory, swimming-like quality. Operators and parents alike frequently comment on these motions, saying that the fetus looks like a ‘little fish’. The head and the trunk periodically coil moderately on one side or the other, and the spine occasionally arches back, frequently accompanied by a slight extension of the head. At this early stage, general movements do not cause positional changes and limbs hardly
1 General Movements
participate in their execution. By the late embryonic stage, corresponding to 7 8 weeks, limb buds have changed into differentiated limbs, and ‘human’ hands and feet with separate fingers and toes have formed [9]. However, in terms of length and size, the limbs are small relative to the head and trunk. Hip and shoulder joints are not yet fully independent and mobile, while trunk muscles are more developed than limb muscles (Fig. 1.1). Between 10 and 13 weeks general movements are propelled by the axial component of the body, the head and the trunk. Shoulders and trunk act in synchrony. The spine stretches back, provoking a marked arching of the entire body followed by rotation of one shoulder and finally of the head. The head, by pointing into the uterine wall, can occasionally act as a pin around which the whole body moves. Head rotations on the other hand are very limited both in frequency and in amplitude, as the neck has not yet reached full anatomical autonomy. Though some independent arm and legs movements can be noted, generally the limbs follow the axial component fairly passively. Stretching of the spine continues to be important, and will remain relevant throughout. At 14 weeks, when stretched backwards the spine displays extreme lordosis or arching (Fig. 1.2). When not arched back the spine is wholly kyphotic (meaning flexed forward or gibbous). Although later in pregnancy the spine becomes more flexible and articulate, only after birth will its alignment develop with cervical and lumbar lordosis, allowing erect stance. At around 15 16 weeks the head can rotate and extend fully. Given its relative proportions and weight it can set the whole body in motion by itself. A rotation or a stretch of the head can unbalance the rest of the body and trigger a cascade of motions. However, from 14 weeks onwards legs and feet increasingly participate in setting the whole body in motion. Legs no longer move only simultaneously: alternate movements have started. The speed and amplitude of alternate movements is not comparable to those achieved a few weeks later, but from now on legs become essential pivots and driving forces of motion. Predominantly legs are extended till the feet touch the uterine wall, push against it and impart a thrust to the whole body (Fig. 1.3). At this stage the arms play a less relevant role in general motions. On the other hand, arms and hands have become a principal component of localized motions. Besides alternating movements, legs can now trigger
1.1 General Movements: 7-16 weeks
9
a
b
c
d
a
Fig. 1.1 Early anatomy and bodily proportions. a, b Fetuses at 8 weeks’ and 9 weeks’ gestational age, re spectively. The head and trunk are extremely large relative to the limbs in length and size. Hip and shoulder joints are not independent of the axial com ponent of the body, and the neck is not anatomically independent of the trunk. The earliest general move ments can only consist in slight stretches and coils of the head and the trunk. c, d Fetus at 11 weeks’ gestational age. The limbs have become longer, the head has become slightly smaller compared to the trunk, and the neck has lengthened and attained partial anatomical independence. Trunk muscles begin to be developed. Although limbs start to accomplish some independent movements, they clearly cannot drive the entire body, but rather are trailed by it
b
Fig. 1.2 Curvature of the spine. a Fetus at 14 weeks’ gestation. The spine displays extreme lordosis, almost giving the impression of an opisthotonic posture a posture characterized by hyperextension of the back and neck muscles, and arching forward of the trunk. In life after birth opisthotonus is seen especially in severe cases of meningitis and decerebration. b Fetus at 20 weeks’ gestation. The spine still displays considerable lordosis. However, a slight curvature begins to be noticed at the level of the lumbar tract
10
1 General Movements
a
b
c
d
Fig. 1.3 Alternate and articulated leg movements. a d Fetus at 16 weeks’ gestation. Alternate and articulated leg movements have become an essential component of general movements. The fetus points its legs (a); alternate leg motions impart a thrust (b); the fetus has turned round, and well articulated leg motions contribute to bring about further positional changes (c, d)
a
b
Fig. 1.4 ‘Locust’ jump. Fetus at 16 weeks’ gestation. a The fetus is ‘sitting’ with its legs bent to the limit. The femur is almost in contact with the tibia (each seen in white, like all ‘bony’ parts) and presumably with the fibula (not visualized here). b The legs are suddenly re leased and stretched, imparting a jump like thrust to the whole body
1.2 Length of the Feet and Epidermal Ridges
general movements by acting in a spring-like manner. Fetuses frequently ‘sit’ in the ‘fetal position’ with the spine flexed forwards and bend their legs to the limit till these and the forelegs are in tight contact with each other as well as with the lower abdomen. Extreme leg bending is probably favoured by composite factors ranging from laxity of ligaments to sparse ossification, allowing a degree of bone flexibility that will no longer be possible a few weeks later. All these factors change rapidly. For instance, bone formation is not a once and for all phenomenon, but more commonly starts from so-called ossification nuclei, small areas of bony tissue, and spreads from there to the rest of the limb or spine. Ossification is not wholly complete at birth [9]. Spatial constrictions, too, now favour the loading of the axial body on the legs. Additionally, during this kind of movement the hands are almost constantly placed on the knees, both stabilizing and locking them, while at the same time increasing the pressure exerted by the legs and feet against the uterine wall. While crouched in this way fetuses push forcefully and simultaneously with both feet against the uterine wall for 6 10 s. Their feet are then suddenly released and their legs elongated, thrusting the whole body with force (Fig. 1.4). This kind of thrust can be compared to the jumping of a locust, although the locust has a totally different structure of the legs, which are also folded and positioned in a completely different way. As with lo-
11
custs, however, fetal jumping may be dependent on the storage and subsequent rapid release of energy to make the leap possible [10 12].
1.2 Length of the Feet and Epidermal Ridges Interestingly, fetuses have ‘big feet’. Up until 25 weeks, fetal feet are almost as long as the femur (Fig. 1.5). Big feet may, amongst other things, have the function of maximising the push, with a consequent increase of stored energy. After 16 weeks, ‘locust jumping’ continues to occur occasionally, but spatial constrictions do not make it easy for the fetuses to ‘jump’. Another factor may be of relevance. By 14 weeks epidermal ridges have formed on the soles of the feet as well as on the palms of the hands. In 1929 Harold Cummins, an American professor of microscopic anatomy, published a paper, ‘The topographic history of the volar pads in the human embryo,’ in which he described how epidermal ridges are preceded by the formation of the so-called volar pads, a swelling in the mesenchyme, or connective tissue, of the palms and soles. Initially the ridges are not fixed because ‘the skin possesses the capacity to form ridges, but the alignment of these ridges are as responsive to stresses in growth
Fig. 1.5 Foot and femur: rela tive proportions. At 12 13 weeks the length of the foot almost equals that of the fe mur. A slight (less than 0.5 cm) discrepancy can be noted between weeks 13 and 16. From week 17 to week 19, foot and femur are the same length, and at week 20 the foot is minimally longer
12
1 General Movements
a
b
c
d
Fig. 1.6 Sliding motions. Fetus at 17 weeks’ gestation. a By pointing its feet and bending its legs, the fetus lifts up the rest of its body. b The feet are suddenly released and the fetus slides downwards along the surface of the uterus. c, d With some variation in the pointing of the legs causing an arching of the lower spine and a forward bending of the head, a similar sliding motion is finally achieved
as are the alignments of sand to sweeping by wind or wave’ [13]. In addition to their relevance to forensic investigation and palmistry, ridges have the other function of causing a friction between the ridged surface and any other surface with which it may come in contact namely, the fairly slippery placenta and/or uterine wall. In other words, ridged feet can push forcefully against the uterine wall without easily sliding away. In fact, other non-ridged surfaces like the back of the body are used to perform sliding motions along the uterine wall which well-anchored feet make possible (Fig. 1.6). Feet remain relevant for the mechanics of motions up to 25 week and beyond. Though less so, even at birth they are still particularly long. From 14-15 weeks alternating leg movements performed in increasingly different combinations are used to cause a rotation of the entire fetal body. Generally speaking, extension of the legs is now largely replacing flexion in initiating general movements.
1.3 General Movements: 17-25 weeks Some other broad modes seem important. Fetuses stretch their legs and arch their spines backwards until the head and both feet are firmly pressed against the wall. By using these two hinges fetuses then turn their shoulders, thorax, and finally pelvis, rotating their body with their legs following last of all. Occasionally knees can also be used as hinges. In contrast to legs and feet, up to 18 weeks the arms seem to act mainly as balancers or equalizers, with limited active participation in the thrust that sets general movements in motion. This balancing function possibly allows the body more flexibility of movement in different planes whilst also minimising its effort. By carefully reducing, correcting and compensating the turbulent motions of other bodily parts, arms provide
1.3 General Movements: 17-25 weeks
continuous adjustment to the moving organism. Until the end of pregnancy the arms are actually longer than the legs, favouring this ‘balancing’ role (Fig. 1.7). From 18 20 weeks, however, the arms too begin to participate increasingly in the initiation and execution of general movements. During this period, fetal constraint has become ever more relevant. Fetuses still move a lot, but all bodily parts are now more and more in contact with the placenta and the uterine wall. Besides the legs, the spine, the head, elbows, knees and shoulders all become important points of support and thrust. At around 18 20 weeks a particular form of motion begins to be observed. The growing uterus is hardly an even surface. Lumps, protuberances, small gorges, temporary and presumably hard areas of contraction can all be noted. Fetuses start using these bumpy areas as support planes. They lean their forearms and hands, trunk or abdomen on the surface. The head, aided by the decreased pull of gravity, is raised, but is encased between the shoulders, which are regularly and markedly elevated. On the whole the image of the upper part of the body seen in profile reminds us of a gibbous animal such as the wild bison or the African gnu. With the rest of the body supported by one of these surfaces, the legs display a vigorous, fast trotting motion (Fig. 1.8). If this kind of motion slows down it is reminiscent of some crawling patterns seen in the infant: the abdomen in contact with the supporting surface while the body and legs are pulled along by the arms only. It is also especially reminiscent of so-called ‘bunny hopping’ and is possibly a forerunner of these movements [14, 15]. The spine has by now acquired more flexibility of movement and various regions can display at least temporarily different curvatures. By leaning against a ‘sur-
13
face’ with the abdomen while keeping a near-horizontal spine and raising their head, fetuses achieve almost complete extension of the legs, as if practising to stand up. This motion too is reminiscent of the movements of infants trying to raise themselves up by leaning against a surface, be it a chair or their mother’s legs. Elongation and semi-flexion of the lower limbs increasingly become main thrusting modes. Extreme folding of the legs is usually noted when fetuses are in a cycle of rest. Legs can also be fully extended above the head with a minimal (<60°) popliteal angle, as observed in severely premature infants. During such extreme extension, the ‘long feet’ of the fetus, by touching and pushing against the wall, can impart a somersault (Fig. 1.9). After 20 weeks complete turning round becomes rarer. The fetus now begins to ‘put on weight’. Bones and musculature have changed profoundly. For instance, principles such as neuronal selection and cell death apply to the innervation of skeletal muscles. Initially muscles have a ‘polyneuronal’ innervation, each muscle fibre being in contact with several nerve fibres arising from one neuron whose cell body is located in the ventral horn of the spinal cord. By now, after 20 weeks, each muscle fibre is in contact with just a few nerve fibres. The adult modality of one muscle fibre connecting with just one nerve fibre will be achieved only after birth. This selection, just like neuronal selection in the brain, is activity-dependent. Fetal muscle fibres, or so-called myotubes, also begin to change their composition. Ossification nuclei have expanded, and so on [16, 17]. As a result of these and many other changes, fetal motions have become more forceful and ‘quickening’ is felt by all pregnant women. Heads are pushed very
Fig. 1.7 Arms as balancers. Fetus at 18 weeks’ gestation. In the course of a general movement, the fetus utilizes its ‘long’ arm to balance and adjust its body
14
1 General Movements
a
b
c
Fig. 1.8 ‘Trotting’ and crawling motions. a c Fetuses at 20 weeks’ gestation. By leaning on a surface (b) or being supported by the head (a and c), these fetuses can start ‘trotting’ motions, particularly evident in c. d Fetus at 22 weeks’ gestation. Crawling pattern. The head and abdomen are in contact with and supported by the uterine wall. The legs are free to be moved and trailed along, albeit within the very limited intrauterine space
vigorously against the uterine wall, similarly to the way neonates push their heads against the walls of their cots. The vigour of the push can be such that the mother may notice a bald patch on the head of the neonate. Arms have fully joined in as propellers during generalized motions. However, up until 22 23 weeks hands are hardly used for support or propulsive purposes. Only from 23 25 weeks does one begin to notice a contribution from the hands, especially the palms. As to legs and feet, although alternate movements continue to be prevalent, supine kicking and pedalling, and, when leaning on a surface, even alternate hopping movements can be observed. Due to crushing gravitational forces, all these leg patterns are not performed even by the neonate, let alone by the premature infant. However, one can see in the alternate stepping of the fetus the emergence of the so-called stepping reflex of the full-term neonate. When
the newborn is supported and the soles of the feet are made to touch a surface, the infant stepping motions start. Such ‘reflex’ actions were thought to disappear within the first 2 3 months. The late Esther Thelen, a professor of psychology at Indiana University and an influential researcher on motor development, argued along with her colleagues against the disappearance of ‘primitive stepping’ as well as supine kicking. Thelen regarded development as a series of evolving and dissolving patterns of varying dynamic stability rather than an inevitable march toward maturity. She showed that these patterns are simply masked for a while by the fact that legs gain weight before they gain muscular strength. In the alternate stepping, kicking, cycling, and even hopping motions of the 24- to 25-week fetus one can glimpse the forerunners of much later movements [18] (Fig. 1.10).
1.3 General Movements: 17-25 weeks
15
a
a
b
b
c
c
Fig. 1.9 a-c Extreme extension of the legs with the feet imparting a thrust. Fetuses at 20 weeks’ gestation. The thigh is flexed on the abdomen and straightened, and the feet are brought into contact and pointed against the uterine surface. Robust legs and long feet can now impart a somersault. The popliteal angle, which is gener ally evaluated in severely premature infants, is an assessment of the tone of the hamstring muscles, and below 90 degrees it indicates severe hypotonia. All these fetuses display a minimal popliteal angle; however, this is perfectly physiological for their gestational age, and may even be functional in performing major positional changes in a narrow space
Fig. 1.10 Forerunners of hopping and standing? a Fetus at 18 weeks’ gestation. By pointing its head and supported by it, this fetus performs hopping motions. As in hopping motions, one leg is semi flexed, and the other is straightened. b c Fetus at 20 weeks’ gestation. By pointing its head into the uterine wall and supported by it, the fetus is capable of reaching a semi standing position and performs tentative stepping motions
16
1 General Movements
1.4 General Movements: Frequency and Duration In addition to their mode of execution, the frequency and duration of general movements also change with increasing gestation. Up to 13 weeks general movements prevail over all other types of motion save heartbeats and startles. After that they decrease steadily until 15 weeks and then maintain a fairly constant level; they are still present in the term fetus. During the early stages of pregnancy, general movements surface as short, fragmentary events. The fetus moves for a few seconds, pauses, and then resumes a generalized motion until the cycle is complete. The fragments can be set only a few or many seconds apart. Before 10 weeks the duration of each bout of generalized motion rarely exceeds 6 s, and bouts tend to appear as isolated events in an otherwise quiescent organism. As gestation progresses, fetuses gradually tend to perform longer and uninterrupted bursts of motion. Nevertheless, by 25 weeks fragmentary and short general movements have not wholly disappeared (Fig. 1.11). It could rightly be argued that fragments might belong to the same general movement which ‘waxes and wanes’ during its execution. The fact that the duration of general movements increases with advancing gestation seems to corroborate this view. The growing organism could be progressively more able to support prolonged physical activity. The duration of general movements has in fact been evaluated in two different ways. In this chapter the duration of each single episode has been calculated. In Chapter 8, when describing rest activity cycles, single episodes of general movements appearing in succession have instead been considered as clusters or temporal areas where general movements surface as opposed to the temporal areas where they do not.
1.5 Central Pattern Generators When general movements start, the fetal central nervous system is very rudimentary. In order for movement to begin, at least minimal connections between neurons and innervation of muscular fibres must have formed. The first synapses originate in the spinal cord. Shortly after, innervation of muscle fibres begins, and with it the first detectable motions guided by neural activity originating from spinal motor neurons also start [19 20].
a
b
Fig. 1.11 General movements (GMs): developmental trends and varying duration. a General movements peak at 10 11 weeks, start decreasing at 12 weeks, decrease sharply after 14 weeks, and subsequently continue on an even slope. b The duration of general movements varies according to gestational age. Initially, fragmented and short (<60 s) general movements are prevalent. After 15 weeks the prevalent duration ranges between 60 and 90 s
At around 7 weeks the brainstem, also called truncus encephalicus or encephalic trunk the portion of the brain connecting the cerebral hemispheres with the spinal cord and comprising the mesencephalon, pons and medulla oblongata begins to form and gradually takes over the majority of fetal motions from purely spinal control. Subdivisions of the brainstem as well as the spinal cord will continue to be in charge of most forms of fetal motility until mid-pregnancy at the least. Cortical control is non-existent at this stage [21, 22].
1.5 Central Pattern Generators
Simplified organisms ranging from the lizard Amblystoma to fishes and other aquatic creatures have all shown that even complex motions can be supported by fairly simple anatomical structures of the spinal cord and the brainstem. The concept of central pattern generators is fundamental to an understanding of fetal motions. Central pattern generators are defined as ‘neural networks that can endogenously (i.e. without rhythmic sensory or central input) produce rhythmic patterned outputs; these networks underlie the production of most rhythmic motor patterns’ [23, 24]. Rhythmic motor patterns control a large part of behaviour, prenatal or otherwise. The first evidence that rhythmic motor patterns are centrally generated came from the locust. Wilson, in 1961, severed and isolated the nervous system of the insect from the rest of its body. Working on this socalled deafferented preparation an insect (or animal) whose sensory input from a portion of the body is lost due to surgical interruption of the peripheral sensory fibres he observed that the isolated nervous system could produce rhythmic outputs similar to those shown during the flight of the intact animal [25]. Subsequent work showed that, in a wide variety of animals, nervous systems isolated from sensory feedback could produce rhythmic outputs resembling those observed during motor pattern production. As Hooper says, all this work has confirmed that ‘rhythmic pattern generation does not depend on the nervous system acting as a whole, but that central pattern generators are instead small and autonomous neural networks’ [26]. Recently, however, a sensory input has come increasingly to be considered as essential for the working of central pattern generators. Studies on locusts have shown important differences between the flying motions produced by intact and deafferented insects. As Grillner and Wallen claim, central pattern generators are functional units within the central nervous system whose activity is modulated by sensory feedback from peripheral receptors to control the frequency and amplitude of centrally generated motor patterns [27]. Central pattern generators circuits have an adaptable nature, as the organism must constantly adjust its behaviour to meet the requirements of its changing internal and external environment. Network flexibility appears to be a common feature in vertebrates and invertebrates. The activity of quite complex sets of pattern-generating neurons can be completely rearranged by sensory inputs operating ei-
17
ther independent of, or in conjunction with, central nervous system activity. In the mollusc Tritonia diomedea the swimming central pattern generator can produce reflexive withdrawal in response to weak sensory input, escape swimming in response to strong sensory input, and crawling after escape swimming has ceased. A single neuronal network, such as a central pattern generator, can be modulated from moment to moment to produce several different physical actions depending on the needs of the animal [27]. In vertebrates, electrical stimulation of the mesencephalic or midbrain locomotor region can produce different locomotor gaits in the high-decerebrate cat, whose cerebral function has been eliminated principally by removal of the brain or transaction of the brainstem. Low-strength stimulation induces slow walking and progressively stronger stimulation produces fast walking, then trotting and finally galloping [28]. Fetal general movements can be considered to be driven by a central pattern generator whose time-dependent intrinsic properties regulate their surfacing. Sensory feedback is an essential component of general movements in many and different ways. General movements are not just storms of motion, but sensorimotor storms. Sensory feedback, for one, gives the possibility of performing a so-called ‘corrective input’, the possibility of altering movements to adapt to environmental variations. This would also provide a convincing explanation for the non-stereotypical nature of general movements right from the start. New types of motion are rapidly attained during prenatal life. Research in vertebrates indicates that new rhythmic motor displays arise by changes in the central pattern generators that produce earlier motions. Data also suggest that fundamental central pattern generator properties are innately established, and the acquisition of new rhythmic motor patterns occurs by the central pattern generator becoming increasingly multifunctional, presumably as a result of increasing synaptic and cellular complexity of the central pattern generator, and additional descending input form outside the central pattern generator. The ability to produce motor patterns that are expressed at only one developmental stage is not lost as the animal matures, but can be re-induced by applying the proper sensory input at more mature stages [28]. The anatomical and neural substrate of movements as well as various characteristics of the environment all change at a dramatic rate. For instance, relative fetal
18
proportions vary rapidly, as does body weight. Cutaneous, muscular and skeletal composition, joint mobility and innervation all change. A shift from almost purely spinal control to increasing cerebellar and cortical organization and regulation also takes place. Environmental components such as the quantity and viscosity of the amniotic fluid and increasing constraint on the growing fetal body all vary, and entail consequent variations in fetal movements. The functional significance of generalized movements is also likely to change with increasing gestational age. Despite all this, only too often the same characteristics and functions are implicitly attributed to general movements from their start during the first trimester of pregnancy to the initial months of postnatal life.
References 1. Coghill GE (1929) Anatomy and the problem of behaviour. Macmillan, New York 2. Hooker D (1952) The prenatal origin of behavior. University of Kansas Press, Lawrence 3. Reinold E (1971) Beobachtung fötaler Aktivität in der ersten Hälfte der Gravidität mit dem Ultraschall. Padiatr Padol 6:274 279 4. Juppila P (1976) Fetal movements diagnosed by ultrasounds in early pregnancy. Acta Obstetr Gynecol Scand 55:131 135 5. Van Dongen LGR, Goudie EG (1980) Fetal movement patterns in the first trimester of pregnancy. Br J Obstet Gynaecol 87:191 193 6. de Vries JIP, Visser GHA, Prechtl HFR (1984) Fetal motility in the first half of pregnancy. In: HFR Prechtl (ed) Continuity of neural functions from prenatal to postnatal life. Spastics In ternational Medical Publications, London, pp 46 64 7. Kurjak A, Azumendi G (2007) The fetus in three dimensions. Informa Healthcare, London 8. Piontelli A (2002) Twins from fetus to child. Routledge, London 9. Jirasek JE (2004) An atlas of human prenatal developmental mechanics: anatomy and staging. Informa Healthcare, London 10. Gabriel JM (1985) The development of the locust jumping mechanism. II: Energy storage and muscle mechanics. J Exp Biol 118:327 340 11. Scott J (2005) The locust jump: an integrated laboratory in vestigation. Adv Physiol Educ 29:21 26
1 General Movements 12. Sutton JP, Burrows M (2008) The mechanics of elevation control in locust jumping. J Comp Physiol 194:557 563 13. Malawala J (2005) Harold Cummins and the birth, growth and development of dermatoglyphics. Am J Phys Anthropol 42:177 181 14. Seyfarth A, Geyer H, Blickhan R et al (2006) Running and walking with compliant legs. In: Diehl M, Mombaur K (eds) Fast motions in biomechanics and robotics, chap 18. LNCIS vol 340. Springer, Berlin/Heidelberg, pp 383 401 15. Patrick SK, Noah J, Yang JF (2009) Interlimb coordination in human crawling reveals similarities in development and neural control with quadrupeds. J Neurophysiol 101:603 613 16. Rees S, Walker D (2001) Nervous and neuromuscular systems. In: Harding R, Bocking AL (eds) Fetal growth and development. Cambridge University Press, Cambridge, UK 17. Chamley C, Cason P, Randall D (2005) Developmental anatomy and physiology of children. Churchill Livingstone, London 18. Thelen E, Smith LB (1996) A dynamic system approach to the development of cognition and action. MIT Press, Cambridge, Massachusetts 19. Strafstrom CE, Johnston D, Wehner JM et al (1980) Spontaneous neural activity in fetal brain reaggregate culture. Neuroscience 5:1681 1689 20. Streit J (1993) Regular oscillations of synaptic activity in spinal networks in vitro. J Neurophysiol 70:871 878 21. Royal College of Obstetricians and Gynaecologists (1998) Fetal awareness: report of a working party. RCOG Press, London 22. Lagercrantz H (2008) Development of consciousness: fetal, neonatal and maternal interactions. In: Levene MI, FA Chevernak (eds) Fetal and neonatal neurology and neurosurgery, 4th edn. Churchill Livingstone, Philadelphia 23. Stein PSG, Grillner S, Selverston AI et al (1999) Neurons, networks, and motor behavior. MIT Press, Cambridge, Mas sachusetts 24. Marder E, Calabrese R (1996) Principles of rhythmic motor pattern generation. Physiol Rev 76:687 717 25. Wilson D (1961) The central nervous control of locust flight. J Exp Biol 38:471 490 26. Hooper SL (2000) Central pattern generators. Curr Biol 10:176 177 27. Grillner S, Wallen P (2004) Innate versus learned movements a false dichotomy? In: S Mori, Stuart DG, Wiesendanger W (eds) Brain mechanisms for the integration of posture and movement. Progress in Brain Research, vol 143. Elsevier, Amsterdam 28. Clarca F, Brocard F, Vinay L (2004) The maturation of locomotor networks. In: S Mori, Stuart DG, Wiesendanger W (eds) Brain mechanisms for the integration of posture and movement. Progress in Brain Research, vol 143. Elsevier, Amsterdam
Startles, Twitches and Clonuses
2
With the assistance of Luisa Bocconi, Chiara Boschetto, Florinda Ceriani and Alessandra Kustermann
Keywords Startles • Twitches • Clonuses • Epilepsy • Hyperexplexia • Moro reflex • Crown rump length • Central pattern generator • Clusters • Hiccups • Dreaming • REM sleep
2.1 Startles In everyday language the term ‘startle’ suggests a behavioural and emotional reaction of surprise, shock and fear to a sudden, unexpected event. Startles are also considered as a motor response to noxious and painful stimuli aimed at withdrawing from a threat [1]. However, the word ‘startle’ is not solely linked with fear and pain, as one can also be pleasantly startled by an unexpected but agreeable experience. In medical terminology the word ‘startle’ can take on many additional meanings, some of which are directly or indirectly connected with prenatal life. Generally speaking, startles designate involuntary, spasmodic movements of the limbs and often of the head and face, linked with a variety of phenomena ranging from epilepsy to sleep. Startle epilepsy is a form of ‘reflex’ epilepsy characterized by seizures triggered by unexpected sensory stimuli. Generally, patients are sensitive to only one sensory modality, usually the auditory one; however, the unexpected nature of the stimulus rather than the sensory modality is the distinguishing feature of the condition. Most patients have intractable seizures, non-progressive encephalopathy, and neurological deficits principally infantile hemiplegia and cortical dysplastic lesions with corresponding abnormal imaging results. The onset is generally considered to be pre- or perinatal and typically manifests itself within the first 2 years of life [2]. Startle epilepsy is to be distinguished from startle disease or hyperexplexia (from the Greek hyper meanA. Piontelli, Development of Normal Fetal Movements. © Springer Verlag Italia 2010
ing exaggerated and ekplexia meaning panic), also commonly called ‘stiff baby syndrome’ a non-epileptic, predominantly genetic disorder characterized by hypertonia, nocturnal myoclonus, generalized stiffness and hyperreflexia marked by brief muscle jerks in response to unexpected auditory, somatosensory and visual stimuli. The tonic spasms may mimic generalized tonic seizures, and can on rare occasions cause apnoea and death. Hyperexplexia usually presents in the neonatal period, but normally decreases within a few months and disappears by 2 3 years of age [3]. Startles are also connected with sleep. In the initial stages of falling asleep, adults and children alike can occasionally be shaken suddenly by one or more endogenously generated startles. These have no emotional meaning nor do they markedly affect the process of falling asleep. The individual hardly awakens and after one or more startles soon plunges back into sleep. Henry Gastaut, the famous French researcher into epilepsy and these forms of disorders, and the founder of the legendary Marseille school of epileptology, located the source of startles as being in the brainstem [4]. In the newborn the word ‘startle’ indicates yet another two phenomena. The first is also linked with sleep. Sudden, spontaneous and generalized jerks similar to those of the drowsy adult or child can be noted in the dormant neonate during quiet sleep. Infants too are not aroused by startles [5]. However, the newborn also shows the so-called ‘startle’ or ‘Moro’ reflex, named after the German paediatrician who first described it. The Moro reflex is observed in normal infants
20
from birth to the first 4 5 months of postnatal life. It can be elicited by altering the equilibrium of the infant in various ways, by making it feel it is falling, or by exposing it to a loud noise. The newborn reacts to the manoeuvre with a startle accompanied by an abduction of all the extremities, especially the arms. The spreading out of the arms is followed by their flexion and adduction as if in an embrace, and is generally accompanied by crying. The primary importance of this reflex for the physician is that it allows evaluation of the integration and intactness of the central nervous system. Consistent failure to respond to the Moro manoeuvre may indicate diffuse central nervous system damage, while asymmetric responses are seen in palsies of both central and peripheral origin. On the other hand, persistence of the reflex after 6 months of age usually indicates cerebral cortical disturbance. The Moro reflex is generally believed to be the only unlearned expression of fear in the human newborn [6]. Purely motor startle reactions supported by simple neural networks and reflex pathways can be found even in lower organisms, such as invertebrates. These simplified reactions are generally considered to be synonymous of escape/withdrawal responses. The study of startle reactions in lower organisms such as the cockroach, the crayfish or the locust have all given us important information on the underlying neural circuits and have led to important discoveries about the functioning of the nervous system [7]. The work of Eric Kandel on the analysis of withdrawal/escape reactions in the marine mollusc Aplysia californica opened up a whole new field of research on the physiological basis of memory storage in neurons. Kandel used the intensification of this simple protective reflex evoked by particular stimuli to study molecular mechanisms at the centre of the formation of basic learning processes such as habituation, sensitization, short-term memory formation and conditioning [8, 9]. In humans startles are a frequent and prominent phenomenon during fetal life (Fig. 2.1). Up to 9 weeks they represent the most conspicuous form of fetal motion. Their frequency peaks at 13 weeks, and gradually decreases thereafter. However, startles continue to be relevant components of the motor behaviour of the fetus throughout pregnancy. They are not reflexive in nature, but spontaneously and endogenously generated phenomena which surface as abrupt, shock-like jerks of the entire fetal body lasting about 1 s each. Despite their high rate of occurrence, researchers
2 Startles, Twitches and Clonuses
Fig. 2.1 Startles (St): developmental trends. Startles increase rapidly from week 10 to week 13, when they reach a peak. They then de crease rapidly between weeks 13 and 14, and this decrease con tinues gradually until week 20. From then on they continue to sur face at an even rate
have generally paid scant attention to startles in the human fetus, implicitly assuming them to be mere epiphenomena of the maturing embryonic nervous system. However, when one observes the natural sequence of emergence of all motor events, an interesting phenomenon can be noticed. Early in pregnancy startles appear to set in motion general movements. Up to 13 weeks, general movements show an absolute dependence on startles for activation as each general movement is immediately (1 2 s) preceded by a startle. Dependence remains significant between weeks 14 and 16. By week 17, independent general movements become more frequent than activated ones, although up to 25 weeks some general movements continue to be activated by startles (Fig. 2.2). Thus, startles are not just vestigial manifestations linked to the immaturity of the nascent nervous system, but have the temporary and adaptive function of activating and promoting general movements during early prenatal life. Considering the proportions and structure of the fetal body, one could imagine how initially fetuses may need this kind of ‘push’ not only to start, but also to execute general movements. As already said, up to 11 weeks the fetal head is disproportionately big and heavy compared to the body, equalling roughly half of the crown rump length (CRL), which is the measurement
2.1 Startles
21
a
Fig. 2.2 Startles and general movements. General movements (GMs) show an absolute dependence on startles for activation until week 13. Dependence remains significant until week 16, when a crossover point between elicited and non elicited general move ments is reached. After 16 weeks’ gestation elicited general move ments drop rapidly. From 19 weeks only an insignificant number of general movements is occasionally elicited by startles
of the length of human embryos and fetuses from the top of the head (crown) to the bottom of the buttocks (rump). Between 12 and 13 weeks the growth of the head begins to slow down. However, up until 16 17 weeks the head is elongated and heavy compared to the limbs. At this stage the muscles of the limbs are also less developed than those of the thorax, and the abdomen and its content are quite heavy in contrast to these thin limbs [10]. Up to 13 weeks startles generally cause a pronounced upward thrust of the entire fetal body. Such massive displacement unbalances the body and precipitates a chain of counter-reactive movements, facilitating the setting in motion of general movements (Fig. 2.3). Although startles progressively lose their power to cause a pronounced displacement of the increasingly heavier fetal body, dependence on the ‘propulsive’ mode is still relevant up to 16 weeks. Activation, however, is not a one-to-one phenomenon. Not all startles are ‘activators’ or ‘pacemakers’ setting general movements in motion during early prenatal life. Startles also surface independently of general movements and outside phases when these emerge, including periods of ‘rest’ (Fig. 2.4). Both startles and general movements are likely to be regulated by central pattern generators. At least up to mid-pregnancy, the central pattern generator responsible for startles seems to exert a direct influence on the central pattern generator responsible for general
b
c
Fig. 2.3 Thrust of the fetal body provoked by startles. a, b Fetuses at 13 weeks’ gestation, c fetus at 14 weeks’ gestation. Save for the ‘heavy’ head (and a minimal portion of the rump in fetus a), the whole fetal body is dislodged and considerably thrust upwards by startles
movements. However, the central pattern generator presiding over general movements is not coupled in a fixed way to the one generating startles. One could say that startles cannot act on the central pattern generator re-
22
2 Startles, Twitches and Clonuses
St Activators Non-Activators
a
Generall M Movements Locali lized d Moti tions Rest S chr Syn chrono onous us Mot Motion ionss
Startles l Hiccup Hi ps Twitches
Fetall Breathi hing M Movements Swall llowing i g Reggular Fetal Breathing g
b
Fig. 2.4 a Startles (St): t activators and non activators. Only a minority of startles are ‘activators’, especially until 13 weeks’ gestation. Most startles do not elicit any motion. From 13 weeks non activator startles follow a trend which is very similar to that of all startles, whether activators or not. By 20 weeks activators and non activators nearly coincide. This trend is in line with the minimal number of general movements elicited from 20 weeks’ gestation. b Startles: 10 weeks’ gestation. Example of how an observational session was graphically represented. The plot corresponding to the first 30 minutes of a typical observation recorded at 15 weeks’ gestation is shown. The vertical axis represents 30 min of observation, with the progression of time plotted from top to bottom. The horizontal axis represents the 60 s within each minute, with time increasing from left to right. Each small square represents 1 s. Each fetal movement is represented by a bar of a specific colour. The length of each bar represents the duration of the fetal movement in seconds. All general movements (red) are activated by a startle (yellow), but startles also fire independently of these. Some (5) short (2 3 s) localized movements (green) and 1 short (5 s) breathing movement (blue) can be noted. None of these are elicited
sponsible for general movements when this is not ready to set a general movement in motion. As Marder and Calabrese [11] say, ‘Electrical coupling among pattern-generating elements is common.
It has been long appreciated that electrical coupling tends to synchronize the activity of the coupled neurons. However, there are increasing numbers of examples in which electrically coupled neurons do not fire synchro-
2.1 Startles
nously and may even fire out of phase.’ Fetal startles continuing to fire independently during periods when general movements are unreceptive to being set in motion could be considered a fitting example of this. Startles may also have the additional function of acting as a ‘reinforcement’ for general movements as they often fire during their execution. Albeit much less frequently, startles can also fire during the execution of other events. However, unlike general movements, other forms of motion display no significant dependence on startles. Any motor pattern can be preceded by one or more startles, but most are not. Compared to other motions, localized movements, especially of the limbs, are more frequently influenced by startles, but not regularly nor significantly. Startles easily provoke a displacement of the fetal limbs as these are particularly ‘light’ and easily shifted in the relative weightlessness of the amniotic fluid. Such dislocation can occasionally trigger a counter-reactive movement, or it can set off a brand new pattern of motion in a previously motionless limb: for instance, a hand may start to touch the face, or a leg may change its position, be flexed or extended, or both. Contrary to what is generally believed, up until 25 weeks startles do not cause the spreading of the arms considered the most typical component of the Moro reflex. Spreading of the arms is a very rare, occasional event. On the other hand, between 24 and 25 weeks startles can cause a raising of the arms accompanied by a facial expression which in life after birth we would interpret as fear (Fig. 2.5). In the premature the Moro reflex begins to be observed at 28 weeks in an incomplete form with spreading of the arms followed by a recoil along the midline. Its complete form (including crying) can only be elicited from 34 weeks. The link between startles and general movements does not rule out that startles may have other, unknown additional functions. Such functions may also change over time. In fact, startles continue to fire well past the time when they trigger general movements. Another feature characterizes startles. As we have said, startles continue to fire outside those clusters when general movements surface. From the beginning, periods in which fetuses are reactive to startles and perform general movements rapidly alternate with cycles of non-reactivity. After general movements have stopped surfacing, fetuses switch over to another cluster, and begin to perform fetal breathing movements and sucking and swallowing motions. As will be explained later
23
in Chapter 8, until 18 20 weeks these clusters are separate. Startles tend to accumulate at both ends of the general movements phase. Before fetuses fully enter into a period of absence of general movements, startles may still trigger one or more very short (4 8 s) general movements. Increasingly, however, fetuses do not respond to them and instead start displaying the abovementioned activities. When these too are over, the fetus enters into a phase of near-absolute rest. Occasional startles fire even during this phase, however. Nevertheless, at mid-cycle they do not cause any perturbation save occasional and very brief (2 6 s) localized movements. Towards the end of the cycle of rest, startles begin to increase again in numbers heralding or precipitating a change of phase in which fetuses again breathe and swallow before starting yet another cluster of general movements. Taking a dynamic systems approach could be of help in the understanding of fetal cycling. Dynamic systems can be described in a very simplified manner as systems whose state evolves in time. However, within these systems fixed points can be found or steady states which will not change over time, and periodic points states of the system which repeat themselves after several time-steps. Some of these fixed and periodic points are ‘attractive’, meaning that if the system starts out in a nearby state, it will converge towards the fixed point or return to a periodic one after completing its cyclic time-steps [12]. The periods of non-reactivity to the abrupt and massive disruption presented by startles within a cycle of rest could be regarded as strong ‘attractors’ towards which fetuses spontaneously shift following a cycle in which general movements have been present. At midcycle of rest fetuses are impervious to this form of perturbation and these impervious time-regions can be considered fixed points. Only towards the end of a cycle of rest do startles increasingly become strong ‘perturbators’ again, hastening once more a shift to another phase of general movements. Startles could thus be considered as natural ‘perturbators’, heralding a change and hastening the beginning of a whole new loop of recurring activities [12]. Hiccups, on the other hand, despite producing a jerk of the fetal body which can be quite strong, do not hasten or herald any change when they surface within periods of rest. One could say that hiccups do not act as a perturber, whereas startles can have such a function towards the end of phase shifts.
24
2 Startles, Twitches and Clonuses
a
b
c
d
Fig. 2.5 Reaction to startles at 24 25 weeks. Fetus at 24.5 weeks’ gestation. a The fetus is resting with its hand leaning on its forehead. The tubular ‘spongy’ structure in front of its face is the umbilical cord. b A startle surfaces and displaces the hand. Only the thumb is still in contact with the forehead and the other fingers are spread out. The other hand (not visible in the first image) is brought to cover one ear. c, d Another startle fires. The fingers are spread out further and the ‘startled’ fetus displays a facial expression that we would classify as fear. This image could have been captured by 4D ultrasonography as two startles fired in close succession
Although at around 18 20 weeks the picture begins to change, startles continue to fire well beyond 25 weeks. In the infant the so-called ‘benign neonatal sleep myoclonus’ is a very common, but transient phenomenon. Coulter and Allen, in 1982, were the first to describe the startles, twitches and rare clonuses which usually accompany neonatal sleep [13]. These authors did not differentiate between all these phenomena. Startles generally disappear during the first few weeks of life, or anyway become a very rare occurrence. Twitches, by contrast, will continue to surface for a
few months. Startles do not cause arousal in the neonate, just as fetal startles do not cause the ‘arousing’ of a burst of generalized motion at mid-cycle of rest. Possibly traces of this fetal phenomenon can still be found in the above-mentioned occasional startles accompanying the beginnings of falling asleep at later stages in life. In the adult, startles linked with sleep, called ‘physiologic sleep myoclonus’ or ‘hypnic jerks’ (from the Greek word hypnos meaning sleep) are considered non-pathological phenomena associated with the transition from wakefulness or arousal to sleep.
2.2 Twitches
However, years, or, in the case of the neonate, weeks, separate these sleep-linked phenomena and we cannot prove that the phenomena are the same. One can only simply assume that neonatal and adult startles heralding sleep may have their roots in our prenatal past. Some generalized seizures may be linked with spontaneous startles. Cases of generalized seizures triggered by startles have been observed in utero in the late stages of pregnancy. Whether this pathological phenomenon could be due to the remains of some neuronal aggregates that normally die or change their composition and function during late prenatal and postnatal life is all yet to be demonstrated, as is whether the ‘error’ that causes startles to trigger generalized epileptic seizures instead of general movements could be caused by noxious events acting during early development, by certain genetic components or by both. Investigators in the field of central pattern generators have considered the destiny of central pattern generator-driven, but developmentally discarded motor patterns. According to Grillner and collaborators, later rhythmic motor patterns arise through modification of the central pattern generators that generated earlier patterns, and the ability to produce motor patterns that are expressed at only one developmental stage is not lost as the animal matures, but can be reintroduced by applying the proper sensory input at more mature stages [14]. As Hooper, another researcher who worked on the central pattern generators guiding the lobster somatogastric system, says, ‘All the neurons of this system are present in the early embryo, and at this stage, the system produces a single, unified, rhythmic output. As the animal matures, subsets of the system’s neurons segregate into different functional networks until, in the adult, the system contains central pattern generators that produce different motor outputs. However, in response to appropriate sensory output, these networks can again fuse into a single network that produces a unified output. This system not only shows the innate development and gradual increase in complexity observed in vertebrates, but may also show the retention of developmentally early motor patterns observed in them’ [15]. In response to appropriate sensory outputs, startles capable of triggering general movement-like fits of motion could also be revived at later stages of human pregnancy or in the human neonate.
25
2.2 Twitches Twitches are brief and small involuntary contractions of a muscle or a group of muscles. ‘Twitches’ is sometimes used synonymously with ‘startles’. However, twitches can easily be differentiated from startles as they do not involve the entire fetal body, nor do they cause it to undergo pronounced thrusts. The part of the body involved jerks momentarily with minimal displacement. In the human fetus, twitches surface later than startles, at around 10 12 weeks. Initially twitches are very infrequent, but from 15 to 16 weeks they increase considerably (Fig. 2.6). From the very beginning twitches surface mainly outside periods when general movements emerge and abound during clusters when fetuses ‘breathe’ and swallow, and within cycles of rest. Unlike startles, twitches do not accumulate at the extreme ends of cycles of rest, but emerge especially at mid-cycle without causing any motor reaction. All these characteristics seem to indicate that twitches and startles are different and independent phenomena. In the human fetus twitches can affect many muscular zones. The areas most frequently involved are the face, the head, the limbs, the thorax and the diaphragm. Up to 25 weeks, twitches of small areas such as the eyelids, the lips, a finger or the nose, which are frequently experienced at later stages, cannot be noted. This may possibly be due to the limitations of current
Fig. 2.6 Startles (St), t twitches (Tw), and clonuses (Cl): develop mental trends. Twitches are negligible before 12 weeks’ and in crease sharply between 13 and 15 weeks’ gestation, just when startles begin their sharp decrease. Subsequently twitches continue to increase very gradually. Clonuses are always negligible, one off events throughout gestation
26
equipment, but it could also depend upon other reasons as yet unknown. Fetal breathing movements seem to be somehow correlated with twitches, as the latter often emerge during their execution. However, given the frequent unfavourable positioning of the fetal body, it is impossible to establish whether this association is significant or not. Twitches may, however, serve important developmental functions. Petersson and collaborators in Sweden suggest that muscle twitches provide vital sensory information which is initially recorded in the spinal cord and eventually in the brain [16]. Rustem Khazipov and his team in France believe that twitches may help the nervous system to organize itself [17]. The brain may record information learned from muscle twitching, such as the weight of the limbs or their position relative to other parts of the body. Eventually, this information is recorded in the somatosensory cortex and may result in the brain’s map of the body [17]. Apparently trivial phenomena such as twitches could thus prove to be significant in building our body scheme. In addition, twitches and startle reactions have been taken into account as possible indicators of pain in the fetus and in the neonate [18]. Invasive medical procedures are increasingly being performed not only on the premature, but also on fetuses in order to save their lives and to grant them a better future quality of life. Most paediatricians already use anaesthetics when performing major surgery on extremely premature infants, but it has also become important to know whether and when fetuses may start to feel pain [19]. The practice of fetal analgesia in utero meets with some resistance for fear of upsetting or damaging fetal development, yet many physicians go as far as to suggest that anaesthesia should be administered to early fetuses before an abortion in order to make the whole process more humane. Twitches are also linked with sleep, and as such have been taken as a possible window into it. Infants and young children twitch frequently during active sleep [20]. However, the main studies on the relationship between twitches and sleep have been performed on animals. Just like startles, twitches do not belong solely to the human race, but are widespread in the animal kingdom and have been observed in many sleeping animals ranging from elephants to cats and dogs [21]. Scientists pondered for centuries as to whether twitches and other forms of motion observed in sleeping animals could be connected with dreaming. Darwin interpreted
2 Startles, Twitches and Clonuses
the motions and the cries let out by animals during sleep as a sign of dreaming, and on this basis he granted dogs and cats with ‘some power of imagination’ [21]. In the adult mammal, REM sleep can be recognized following a series of behavioural and electrophysiological criteria, including muscular atonia, rapid eye movements (hence the term ‘REM sleep’), muscular twitches and desynchronized electroencephalogram (EEG), a rapid low-voltage electrical activity similar in appearance to that recorded during wakefulness. At birth the human neonate spends roughly 75% of its time asleep. Gradually over the first 3 months this reduces to 50% [22]. Severely premature infants spend all their time in a sleep-like state during which, just like fetuses, they alternate between periods of intense and turbulent motion and periods of relative rest in which twitches surface. However, other parameters generally used to distinguish sleep and its phases have not yet emerged by 25 or even by 28 weeks. For example, eye movements are scarce and the EEG is only erratically recordable and undifferentiated [23]. Other altricial neonates (i.e. neonates born at a still immature and dependent stage, like the human infant) do not exhibit, even at birth, a distinct REM state as judged by adult norms. Their EEG is less cohesive, their postural control is minimal, and rapid eye movements are infrequent [23]. Yet, newborns spend 90% of their sleeping time in a sleep state characterized by frequent twitches, occasional REM and irregular breathing patterns, all of which resemble components of adult REM sleep. These analogies have driven several scientists to consider the prevalent behavioural state of the newborn as REM sleep. On the other hand, the limited indicators of this state present in the neonate have pushed them to consider twitches as markers of an immature form of the REM phase of sleep. By analogy, twitching in the premature human infant has been taken to be an indicator of phases of REM sleep. Other scientists, however, dispute the view that neonatal sleep and even much less so sleep in the premature is equivalent to REM sleep [24]. Blumberg and Lucas in particular discounted a link between REM sleep and twitching by showing that the signals that cause muscle twitches originate from the spinal cord, not the brain [21]. Twitches may guide us to understand other phenomena, but can hardly be considered a good indicator of REM sleep.
2.3 Clonuses
2.3 Clonuses The word clonus derives from the Greek klonus, meaning unrest and tumult. Clonuses are a series of rapid, rhythmic, alternating contractions and relaxations of one or more muscles. In medical dictionaries the word ‘clonus’ is almost invariably associated with the attributive adjective ‘abnormal’ or ‘pathological’ and regarded as symptomatic of an affection of the brain or spinal cord. Disorders presenting clonuses amongst other symptoms are wide-ranging and vary from multiple sclerosis to Parkinson’s disease and some forms of epilepsy. However, except for epilepsy, most of these pathologies do not affect infants. Clonuses can be spontaneous or elicited, as during a neurological examination. In order to trigger a clonus, the physician usually flexes the knee of the patient slightly, then jerks the foot upward and a little outward whilst continuing to keep finger pressure on the sole of the foot. Spontaneous clonuses are generally triggered by pushing hard against the sole. Fetuses frequently push hard with the soles of their feet against the uterine wall, and clonuses can be noted occasionally from 16 to 18 weeks. Fetal clonuses are almost exclusively ankle clonuses, meaning a movement of the foot. Wrist clonus a spasmodic movement of the hand initiated at the wrist was noted just once in a 24-week fetus who was pushing its palm against the placenta. All 20 clonuses which were observed were characterized by less than 5 6 beats. All fetuses observed in our various studies were seen at a 6-month follow-up and parental reports were obtained after 1 and 2 years. None had developed a manifest neurological condition. However, frequent clonuses have been described in fetuses whose mothers suffered from continued substance abuse, especially alcohol and cocaine [25]. Clonuses in infants are also regarded as possible indicators of cerebral palsy, and of spasticity in particular, and therefore the children presenting them are subject to repeated follow-ups [26]. Special attention ought to be paid by physicians examining fetuses with ultrasound if more frequent clonuses are observed. Subtle seizures are another area that may warrant attention. Usually neonatal seizures are subtle. In the almost universally accepted scheme proposed by Volpe, subtle seizures represent 50% of all neonatal seizures [27]. These include clonic movements of a limb. However, general movements triggered by startles past the first
27
20 25 weeks of pregnancy may also warrant a special awareness, especially if they lack the fluency and elegance described by Prechtl and collaborators [28]. All the above-mentioned phenomena, if observed with undue frequency or linked with other motor displays past their time, may be pointers to some form of pathology in the fetus. Obviously one would need to observe much larger populations before establishing well-defined norms and deviations which could be considered pathological. Sigmund Freud’s name is almost invariably quoted on treatises dealing with cerebral palsy. Freud started his career as a neurologist, and neurology continued to interest and attract him right to the end of his life. Before shifting his interests towards psychoanalysis, he was the first to point out, in 1897, that cerebral palsies, then called cerebral paralyses, could be developed in utero. At the time a theory first propounded by an English surgeon, Dr William Little, in 1861, was widely accepted and continued to be so for more than a century. Dr Little was the first to describe cerebral palsy, which he believed to be due to birth asphyxia. The causal theory of birth asphyxia was only challenged in the 1980s when a large study found that only a very small percentage of cerebral palsy cases could be attributed to difficult births [29]. We are now much more cognisant of neurological impairment starting in utero and many of its causes, ranging from genetic defects to viral and bacterial infections. Even apparently insignificant motions such as startles, twitches and clonuses may well in future become clinical indicators of neurological disfunctioning in utero.
References 1. Chokroverty S, Walczak T, Hening H (1992) Human startle reflex: technique and criteria for abnormal response. Elec troencephalogr Clin Neurophysiol 85:236 242 2. Rennie JM (1996) Neonatal seizures. Eur J Pediatr 156:83 87 3. Shahar E, Raviv R (2004) Sporadic major hyperekplexia in neonates and infants: clinical manifestations and outcome. Pediatr Neurol 31:30 34 4. Alajuanine T, Gastaut H (1955) La syncinésie sursaut et l’épilépsie sursaut à déclanchement sensoriel or sensitif inopiné. I. Les faits anatomocliniques (15 observations). Rev Neurol 93:29 41 5. Wolff PH (1966) The causes, controls and organization of be havior in the newborn. Psychol Issues, monograph 5, n 17 6. Popper A, Samuels M (2009) Normal development and deviations in development of the nervous system. In: Adams
28 and Victor’s principles of neurology, part 3, chap 28. Mc Graw Hill Professional, New York, pp 493 418 7. Eaton RC (1984) Neural mechanisms of startle behavior. Springer, New York 8. Kandel ER, Schwartz JH (1982) Molecular biology of learning. Modulation of transmitter release. Science 218:433 443 9. Kandel ER, Abramst T, Bernier L et al (1983) Classical con ditioning and sensitization share aspects of the same molecular cascade in Aplysia. Cold Spring Harb Symp Quant Biol 48:821 830 10. Jirasek JE (2004) An atlas of human prenatal developmental mechanics: anatomy and staging. Informa Healthcare, London 11. Marder E, Calabrese RL (1996) Principles of rhythmic motor pattern generation. Physiol Rev 76:687 717 12. Scott Kelso JA (1995) Dynamic patterns: the self organization of brain and behavior. MIT Press, Cambridge, Massachusetts 13. Coulter DL, Allen RJ (1983) Benign sleep myoclonus. Arch Neurol 39:191 192 14. Grillner S, Georgopoulus AP, Jordan LP (1997) Selection and initiation of motor behavior. In: Stein PSG, Grillner S, Selverston AI et al (eds) Neurons, networks, and motor behavior. MIT Press, Cambridge, Massachusetts 15. Hooper SL (2001) Central pattern generators. Encyclopedia of life sciences. John Wiley, Hoboken 16. Petersson P, Waldenström A, Fåhraeus C, Schouenborg J (2003) Spontaneous muscle twitches during sleep guide spinal self organization. Nature 424:72 75 17. Khazipov R, Sirota A, Leinekugel X et al (2004) Early motor activity drives spindle bursts in the developing so matosensory cortex. Nature 432:758 761 18. Anand KJS, Hickey PR (1987) Pain and its effects in the human neonate and fetus. N Engl J Med 317:1321 1329
2 Startles, Twitches and Clonuses 19. Grunau RE, Holsti L, Whitfield M et al (2000) Are twitches, startles, and body movements pain indicators in extremely low birth weight infants? Clin J Pain 161:37 45 20. Curzi Dascalova L, Giganti F, Salzarulo P (2008) Neuro physiological basis and behavior of early sleep development. In: Marcus CL, Carrol JL, Donnelly DF et al (eds) Sleep in children: developmental changes in sleep pattern, 2nd edn. Informa Healthcare, London 21. Blumberg MS, Lucas DE (1994) Dual mechanism of twitching during sleep in neonatal rats. Behav Neurosci 1086:1196 1202 22. Sadeh A (2008) Maturation of sleep patterns during infancy and childhood. In: Marcus CL, Carrol JL, Donnelly DF et al (eds) Sleep in children: developmental changes in sleep pattern, 2nd edn. Informa Healthcare, London 23. Salzarulo P (2003) Il primo sonno: sviluppo dei ritmi sonno veglia nel bambino. Bollati Boringhieri, Torino 24. Frank MG, Heller HC (2003) The ontogeny of mammalian sleep: a reappraisal of alternative hypotheses. J Sleep Res 22:25 34 25. Baraban SC, McCarthy EB, Schwartzkroin PA (1997) Evidence for increased seizure susceptibility in rats exposed to cocaine in utero. Brain Res Dev Brain Res 102:189 196 26. Futagy Y, Otani K, Goto M (1997) Prognosis of infants with ankle clonus within the first year of life. Brain Dev 19:50 54 27. Volpe JJ (2008) Neonatal seizures. In: Neurology of the newborn, 5th edn. Saunders, Philadelphia 28. de Vries JIP, Visser GHA, Prechtl HFR (1982) The emergence of fetal behaviour. 1: Qualitative aspects. Early Hum Dev 7:301 322 29. National Institute of Neurological Disorders and Stroke (NINDS) http:/www.ninds.nih.gov/disorders/cerebral palsy/de tail cerebral palsy.htm. Accessed 27 Jan 2010
Hiccups, Yawning and Gasping
3
With the assistance of Florinda Ceriani, Roberto Fogliani and Alessandra Kustermann
Keywords Hiccups • Yawning • Gasping • Singultus • Synchronous diaphragmatic flutter • Sneeze • Phrenic nerve • Central pattern generators • Anencephali • Apnoea • Hypoxia • SIDS • Meconium
The anatomical basis and, in particular, the functional significance of the three phenomena discussed in this chapter is still largely obscure. The main link uniting these phenomena is that they all begin to emerge during prenatal life. Additionally, despite being disparate phenomena, hiccups, yawning and gasping have all been associated with breathing and with some form of ‘primitive’ or ‘emergency’ respiration.
3.1 Hiccups According Dorland’s Medical Dictionary, hiccups, or hiccoughs, also called singultus or synchronous diaphragmatic flutter, are ‘produced by a sudden and forceful contraction of the diaphragm followed by a closure of the epiglottis. The vigorous closing of the epiglottis causes the characteristic ‘hic’ sound reflected in the onomatopoeic name hiccup’ [1]. Hiccups also cause a lowering of the diaphragm, or more usually a hemidiaphragm, and a consequential large decrease in intrathoracic pressure. Hiccups are a common event which almost everyone experiences once in a while. Since time immemorial this widespread phenomenon has elicited the mirth of observers as well as the interest of physicians, philosophers, and the general public alike. In his dialogue ‘The Symposium’, a discussion on the nature of love in the form of a series of speeches, the Greek philosopher Plato describes one of the participants, Aristophanes, the greatest comic A. Piontelli, Development of Normal Fetal Movements. © Springer Verlag Italia 2010
poet of Athens, hiccupping embarrassingly when launching in an eulogy of love. Aristophanes himself states how absurd this is while others are discussing lofty questions of love. His physician finally intervenes, telling him: ‘Hold your breath, and if after you have done so for some time the hiccup is no better, then gargle with a little water, and if it still continues, tickle your nose with something and sneeze, and if you sneeze once or twice, even the most violent hiccup is sure to go.’ To this day people still try to provoke sneezing in order to stop hiccupping. Hippocrates, the father of medicine, associated hiccups with liver inflammation. He also regarded hiccups supervening on an abundant discharge of blood, an ileus, or vomiting as an ominous sign. According to him, hiccups, like convulsions, occurred when ‘depletion or repletion’, an excessive reduction or an excessive accumulation of ‘substances’ including drink and food, took place, and he too considered sneezing as a sure remedy. Galen, on the other hand, believed hiccups to be caused by violent emotions arising in the stomach. He suggested giving the patient a fright as a curative measure. Down the centuries numerous homespun remedies have been suggested to stop this noisy, awkward, and disagreeable occurrence. Some of these remedies are still popular today and can occasionally be effective in stopping transient bouts of hiccups. We have all tried lemon juice, or going back to Galen’s theory and giving someone hiccupping a fright. Recently, innovative and sometimes bizarre treatments have been proposed for adults, such as smoking marijuana, having sexual intercourse and
30
even a cure that was granted the 2006 Ig Nobel Medicine Prize [2 4]. Only in 1833 did the Edinburgh physician Shortt recognize an association between phrenic nerve irritation and hiccups [5]. As a remedy for severe cases Shortt recommended blistering and burning the skin along the course of this nerve, which innervates the diaphragm. Such drastic measures were soon dropped, although in cases of intractable hiccups the phrenic nerve may be surgically severed to relieve the patient’s distress. In 1943 Bailey, another physician, postulated the existence of a ‘hiccup centre’ located in the brainstem, and of a hiccup reflex arc, a neural pathway that mediates a reflex action by connecting afferent (sensory) and efferent (motor) signals [6]. Hiccups do in fact originate in the brainstem. Cats, and other mammals stimulated electrically in a small and still ill-defined area of the brainstem start hiccupping [7]. However, the central coordination of hiccups still needs further elucidation and precise location. The brainstem controls other rhythmic, absolutely vital functions, such as breathing and swallowing. In animals these functions also originate from central pattern generators located in the brainstem that send well-designed impulses to the relevant nerves innervating specific muscles, thus triggering the patterned sequence of breathing and of other phenomena [8]. After birth a conscious influence can be exerted on these largely automatic functions. For instance, during wakefulness breathing can be controlled in a wide variety of circumstances ranging from athletes trying to breathe optimally to obtain better performances, to people taking a deep breath before facing a difficult task. Hiccups, however, elude conscious, upper brain control. Unlike the central control, the peripheral hiccup reflex arc is largely, but not fully clarified. In the adult the afferent pathways of the reflex arc consist principally of the vagus or tenth cranial nerve, the phrenic nerve, the so-called thoracic sympathetic chain, a series of nerves leaving the spinal cord between T6 and T12 (the 6th and the 12th thoracic vertebrae) and innervating intercostal muscles, and by the pharyngeal plexus, a network of nerves connected with the pharynx [9, 10]. The efferent pathways pass mainly through the vagus and the phrenic nerve [9, 10]. Given this long and tortuous pathway of the reflex arc, anything that interferes with any of these nerves, from tumours affecting the peritoneal viscera to stomach distension, can cause a spasm and trigger more or less prolonged hiccupping.
3 Hiccups, Yawning and Gasping
Hiccups often occur spontaneously, but many circumstances, such as a cold drink, spicy food, or getting drunk, can act as ‘irritants’ and have long been recognized to trigger them. However, hiccups can have a central origin and be symptomatic of various conditions affecting the central nervous system at various levels of the brain, the brainstem, the cervical cord, and the root entry zones of the last cranial nerves. Persistent hiccups can also indicate a variety of other diseases ranging from peritonitis, cirrhosis and heart attack to intestinal obstruction. A distinction has in fact been made between transitory bouts (lasting no more than 48 h), persistent episodes (longer than 48 h but less than a month) and intractable hiccups (longer than a month). As already said, persistent and intractable hiccups are generally caused by a variety of pathological conditions. Hiccups can be isolated events one can have a single singultus but more frequently hiccups occur in bouts, with each hiccup following another at a regular frequency. The rate of occurrence has been calculated to vary on average between 4 and 12 spasms a minute. Just occasionally the incidence can be much higher, reaching up to 60 spasms per minute in exceptional cases [11]. Despite the clarifications, the exact cause of hiccups is still hypothetical and most think of hiccups as having no functional purpose [12]. However, it seems strange that such a complex and finely tuned reflex should not serve any purpose at all. Hiccups can be noted at birth. The healthy infant frequently hiccups, especially during wakefulness. In infants hiccups generally stops by suckling, either by breastfeeding or simply inserting a finger, the teat, or a dummy into the baby’s mouth. Clinicians have in fact suggested that in some cases hiccups may protect the respiratory tract from gastro-oesophageal reflux [13]. An old French saying, ‘enfant hoquetant, enfant bien portant’ (a child who hiccups is a healthy child) seems to recognize this theory. Feeding, possibly by causing stomach distension and secondary pressure-irritation of the phrenic nerve, is an especially frequent trigger. Parents of young babies are familiar with the fact that hiccups commonly follows a good feed. However, unlike the adult, the newborn does not seem to be distressed by it. Severely preterm infants also hiccup frequently. They have been observed to spend 2.5% of their time hiccupping [14]. In the premature, hiccups are often taken as an indicator of an ill-defined state of ‘stress’. Hiccups per se are not pathological even in the preterm. Secondarily, however, they can cause potentially pathologic conse-
3.1 Hiccups
31
quences such as gastro-oesophageal reflux. Hiccups start long before birth [15, 16]. They are one of the earliest patterns of movement. Fetuses start hiccupping at 9 weeks, and up until 12 weeks hiccups are a conspicuous, frequent motion. After peaking between 10 and 12 weeks, hiccups decrease, but up to 25 weeks and beyond bouts of hiccups continue to occur regularly (Fig. 3.1). In the fetus a hiccup can occur as a one-off event, but much more frequent are bouts of hiccupping of various duration. Whilst the number of the bouts does not increase with gestational age, their duration does. The rate of occurrence of hiccups is high, reaching 60 spasms per minute. Hiccups are also ordered, as variability is seldom observed. Hiccups can be easily distinguished from fetal breathing by the larger, briefer, more uniform and regular changes they induce in thoracic dimensions. When fetuses hiccup they do not open their mouth, nor does the operator watching a bout of hiccups using ultrasound hear the characteristic ‘hic’ sound. On the other hand, once quickening has started, mothers accurately perceive when fetuses hiccup. Most, given the regularity of the motion, realize that the fetus is hiccupping; others worry and think that the fetus may be having a prolonged seizure. Up to 12 13 weeks, just like startles, hiccups cause an upward thrust of the entire fetal body. From week 13 onwards the thrust is minimal (4 6 mm). However, even initially hiccups can be easily distinguished from startles as the thrust caused by startles is more pronounced, and startles do not fire at regular intervals (Fig. 3.2).
Hiccups are observed more frequently, but not exclusively, during cycles of activity (63% v. 37%). During cycles of rest, hiccups can stimulate occasional and brief (1 6 s) isolated limb movements, but otherwise, as in the neonate, they do not seem to upset quietness. Unlike startles, a hiccup occurring during a cycle of rest does not precipitate a change of condition to a cycle of activity. Neither does it herald such a change by being placed more frequently at the end of cycles of rest as startles are. Hiccups can occur simultaneously with other motor activities, but not with fetal breathing and swallowing. Up to 25 weeks hiccups, breathing and swallowing appear to be either/or events. When fetuses hiccup, they stop breathing or swallowing and vice versa, nor do they initiate these activities simultaneously (Fig. 3.3). Hiccups are a significant phenomenon during prenatal life and as such they may have different causes and functions at different gestational ages. Few hypotheses have been put forward to explain their possible functional significance in the fetus [17, 18]. In the fetus they are not connected with any obvious pathological condition. Initially, at least, no irritative-compressive causation seems to be operative in the fetus. Sparse swallowing movements begin to surface at 12 weeks, but do not cause visible and considerable stomach distension until 2 weeks later. This does not rule out the possibility that some bouts of hiccupping may be triggered by stomach distension or other ‘irritants’ in utero, especially at later stages. According to one hypothesis, by stretching the di-
Fig. 3.1 Hiccups (Hic): developmental trends. Hiccups are partic ularly pronounced between 10 and 13 weeks’ gestation, reaching a peak at 11 weeks. From 14 weeks onwards hiccups follows a regular trend
Fig. 3.2 Thrust of the fetal body provoked by hiccups. Fetus at 13 weeks’ gestation. Hiccups provokes a small thrust of the fetal body. Only the back is moved slightly upwards; the head and the legs remain in contact with the uterine surface. Compare this with the thrust provoked by startles in Fig. 2.3
32
3 Hiccups, Yawning and Gasping
General Movements Localized Motions Re t Res Sy chr Syn chrono onous us Mot M ion ionss
Startles Hiccup Hic cupss cup Twitch tches es
Fetall Breathi thing gM Movements Swallowingg Reggula ular Feta etal Brea reathi thing ng
Fig. 3.3 Hiccups: compatibilities. See Fig. 2.4 legend for an explanation of the chart design. At 14 weeks hiccups are compatible with general movements. The two phenomena can surface synchronously. However, when a breathing bout emerges at 11 min 11 s, the emergence of hiccups interrupts it. Breathing starts again after hiccups have stopped. In this chart, save for general movements, hiccups do not surface synchronously with any other behavioural event
aphragmatic muscles hiccups ease fetal respiration, and also exert a preparatory function for postnatal breathing. According to this theory, these ‘exercises’ would be specially suited to the fetus as the amniotic fluid, being much more viscous than air, requires particular effort from the diaphragmatic muscles to expand. The fact that hiccups decrease rapidly just when fetal breathing movements begin to emerge consistently at 12 weeks would seem to favour this hypothesis, corroborating the view of a ‘facilitatory’ function for the onset of fetal respiration (Fig. 3.4). However, hiccups do not cease at 12 weeks. This could be interpreted as a continuing ‘facilitatory’ role for prenatal breathing, as well as an anticipatory and equally facilitatory one for postnatal respiration. The two roles could co-exist. In addition, it is unclear why hiccups, breathing and swallowing are not compatible during the first half of pregnancy and beyond. Nor is it clear in what way the fetal thorax, which, being bell-shaped, has a different shape not only from the adult thorax but also from the neonatal one, would need this kind of stretching exercise in order to allow pre- and postnatal breathing. Given the unchanging nature of hiccups, it has been postulated that beyond the initial weeks of gestation they
Fig. 3.4 Hiccups and breathing. Fetal breathing movements (FB ) start to increase sharply at 12 weeks’ gestation, just when hiccups (Hic) begin their sharp decease. Hiccups and breathing also reach a crossover point at 12 weeks’ gestation
take on an almost vestigial pattern of occurrence and possibly no longer have any role at all [17]. Hiccups is a widespread phenomenon that does not belong to the human race alone; many other mammals and many amphibians hiccup. Hiccups could perhaps just be a vestige
3.2 Yawning
of our ancestral past. Other scientists link hiccups to our phylogenesis, and in particular to amphibian respiration and more specifically to tadpole respiration [19]. Amphibians, especially tadpoles, use both lungs and branchiae (also commonly called gills) to breathe switching to one mode or the other according to the environment where they happen to be. When in a watery medium, tadpoles use their branchiae and start hiccupping. Hiccupping allows water to enter their throat, whilst the almost simultaneous closure of the epiglottis (causing the ‘hic’ sound) prevents water from flooding their lungs. According to this theory the same happens in the fetus, a watery creature preparing to enter the world of aerial breathing. Once born (the argument runs), infants have to breathe air but must also drink, and later eat, while continuing to protect their lungs from being flooded. Hiccupping could help us learn to not choke: in fetal hiccupping, fetuses would gulp the amniotic fluid, but the closure of the epiglottis prevents the fluid from entering their lungs. However, it has been objected that fetuses never develop branchiae, and the analogy may be too farfetched. During bouts of hiccupping, fetuses do not open their mouth, nor do they swallow when hiccupping. Therefore, hiccupping does not prevent flooding of the lungs nor do fetuses run the risk of choking during prenatal life. Hiccups would only prevent choking during postnatal life. The epiglottis would start practicing closing when swallowing during fetal life. Prenatal hiccups could be yet another exercise (like coughing) preparing us to dislodge foreign pieces of food which have become stuck in the oesophagus, or which are too large and descending too slowly. When a piece of food is swallowed that is too large for the natural peristalsis of the oesophagus to move the food quickly into the stomach, it applies pressure on the phrenic nerve invoking the hiccup reflex. This anticipatory role could be important enough; however, the ‘non-choking’, ‘dislodging’ theory is difficult to reconcile with the fact that fetuses hiccup more frequently at the beginning of the fetal period when they do not even swallow. However, after the publication of a very successful book, Your Inner Fish, this hypothesis is gaining increasing popularity amongst the public at large [20]. Linking our prenatal past with phylogenesis is becoming increasingly trendy and is applied to almost all sorts of phenomena. This goes back to the so-called theory of recapitulation, which states that ‘ontogeny recapitulates phylogeny’. This theory was mainly propounded by Ernst
33
Haeckel, a German zoologist, in 1866. Haeckel believed that the development of an organism (its ontogeny) followed the evolutionary history of its species (its phylogeny) and sometimes that of other species as well. Although some vestiges of our animal past are still with us, the uncontrolled use of the idea has been widely criticized by the scientific community. Despite this, the theory is still very popular. For some reason, being compared to tadpoles is a fascinating idea for many. Whichever hypothesis may be true, the functional role of the hiccup, or one of its functional roles if any might be sought in the initial phases of our prenatal past.
3.2 Yawning According to Dorland’s Medical Dictionary, ‘A yawn is a deep, involuntary inhalation with the mouth widely open, often accompanied by the act of stretching’ [1]. Yawning is composed roughly of an active wide opening of the mouth followed simultaneously by a deep inhalation, a dilatation of the pharynx, larynx and thorax, and by a lowering of the diaphragm. The oral cavity becomes amply visible, the tongue is retroverted, the eyes halfclose, the nostrils dilate, the eyebrows rise and the forehead wrinkles slightly. Finally, the mouth passively closes again with a deep expiration. Fetuses yawn. Some authors claim that they do so from 12 weeks [15]. However, we did not detect the first yawns earlier than 15 weeks. Fetal yawns do not correspond completely to the above description (Fig. 3.5). Up until 25 weeks the eyes are kept closed, and the dilation of the nostrils as well as the lowering of the diaphragm are too difficult to evaluate. At around 25 weeks fetuses open their mouth widely, blink, raise their eyebrows, wrinkle their forehead and retrovert their tongue (Fig. 3.6). Dilation of the larynx, pharynx and thorax have been ascertained in fetal animals such as sheeps. Yawns last on average 8 10 s and occur principally as occasional and isolated events. However, in three cases (at 20, 21 and 25 weeks respectively), one fetus yawned in a series of five and two in a series of six yawns with an interval of 8 10 s between the closure of the mouth and the next yawn. Most, but not all yawns emerge before a cycle of rest. However when fetuses start yawning, they stop all previous activities, and only resume any other motion when the yawn or yawns are over.
34
3 Hiccups, Yawning and Gasping
a
a
b
b
c
c
Fig. 3.5 2D yawns. Fetuses at a 18 and b, c 20 weeks’ gestation yawning. The pictures were taken using traditional 2D ultrasonog raphy. At 20 weeks (b c) yawning has ‘matured’. The mouth is open more widely, the head is retroflexed, and the tongue is slightly arched with the tip pushing against the mandible
Fig. 3.6 4D yawn. Fetus at 24 weeks’ gestation yawning. The face is visualized in great detail. a The fetus is at rest with four digits (the thumb cannot be seen) touching the face. One finger touches an eye. When the fetus starts yawning (b, c), the fingers are moved slightly to the side, uncovering the eye but touching the nose. However, the fast arching back of the head cannot be captured, nor can the tongue be visualized
3.2 Yawning
Yawning fetuses do not display apparent signs of distress. Cerebral blood flows were checked during and immediately after yawning, but these were well within the normal range. On the whole, we are still far from understanding the functional significance of yawning, whether before or after birth. Nevertheless, one cannot but wonder why fetuses yawn. In life after birth, yawning is mostly connected with tiredness, sleepiness, boredom, irritation, interruption of sleep and rousing. However, yawning can also be related to pleasure and relaxation as well as hunger [21]. It is difficult to ascribe complex emotions such as boredom or irritation to fetuses. Hunger is another stimulus which may hardly be experienced in utero, save perhaps in extreme circumstances. Hedonic feelings are hard to demonstrate at this early stage, too, though fetuses may begin to function on a pleasure/displeasure base. Some motions and positions are ‘preferred’ to others, but one cannot go much further beyond this simple fact. Anyway, even assuming fetuses could begin to experience some form of pleasure or displeasure after the first half of pregnancy, why should this be expressed through yawns? In life after birth yawning has a ‘contagious’, social nature. Seeing someone yawning triggers a yawn [22]. Yawning also has an ideational component as even just thinking of yawning makes one yawn. None of this, however, can be associated with fetal life. Fetuses can hardly be regarded as social beings, though they increasingly prepare to enter a social world. In utero they clearly never see someone yawning, nor were the fetal yawns we studied preceded or followed by their mothers’ yawning. All mothers (and fathers) were very attentive and laughed during and after the fetal yawn. It is also difficult to assume that fetuses could have such complex forms of ideation as to think about a yawn, and start yawning as a consequence. After birth, yawning is not a necessary event, in the sense that it does not invariably accompany any of the above-mentioned states, and displays large variations both within the same individual and amongst different individuals. The same quality of apparent non-necessity applies in fetal life. After birth, frequent yawning is linked with a variety of medical conditions. Neurological conditions are often associated with frequent yawning. Viral infections, diseases directly affecting the trunk or its compression due to intracranial hypertension, diseases of the thalamus, of the region of the hypophysis, brain tumours, cerebral haemorrhage, multiple sclerosis, myasthenia
35
gravis and more are all coupled with excessive yawning. All sorts of non-neurological conditions can also cause excessive yawning. These range from diabetes to profuse bleeding, liver cirrhosis and heart attack. However, nonexcessive yawning is a perfectly normal phenomenon in man during all phases of life, and fetuses can hardly be defined as excessive yawners. Furthermore, all fetuses who yawned were perfectly healthy at birth and at subsequent follow-ups. Though yawning is readily and easily recognizable at all ages, including the fetal stage, various modes of yawning and their possible significance may change with age. In the neonate, yawning is almost invariably accompanied by stretching. In the adult, stretching especially occurs when yawning is associated with awakening, and as such has been related with a kind of re-setting in motion of the organism after the relative stillness of sleep. In the fetus, some tentative form of stretching can be observed only from 24 to 25 weeks. One or both arms can be slightly stretched outwards and the forearm stretched either upwards or downwards. However, at this stage stretching is barely perceptible. The neural mechanisms controlling yawning have been to some extent clarified. A primary ‘yawning centre’ is located in the bulbopontine regions of the cerebral trunk. Anencephali children born with the cerebral hemispheres missing or reduced to small masses attached to the base of the cranium, and without the cerebellum and the flat bones of the skull yawn and stretch just like all of us. In other words, we don’t need a cortex in order to yawn [23]. In man and in some mammals a second pathway is thought to be represented by the ill-defined limbic system, an assemblage of interrelated, phylogenetically old, deep brain structures commonly regarded as involved in emotions, motivation, processing of sensory and motor functions, memory and even cognitive information. Given its link with sleep, and especially with the transitions between sleep and wakefulness and vice versa, yawning has elicited a special curiosity in those involved in sleep research. The main experts in the field can be considered to be the American psychologist Robert Provine and the neurologist and psychiatrist Piero Salzarulo in Italy [24, 25]. The association between yawning and sleepiness has been explained by the anatomical proximity between the bulbopontine region involved in yawning and the ascending reticular formation discovered in 1949 by Moruzzi and Magoun. This latter system acts
36
diffusely on the cerebral cortex and activates it, regulating arousal and sleep [26]. Awakening is very rare before 34 36 weeks, and fetal sleep prior to this age is considered by many to be a preparatory function only akin to sleep. Fetuses do not yawn because they feel sleepy. Besides in man, yawning is widespread throughout the animal kingdom. Darwin in his famous book The expression of emotions in man and animals mentioned yawns in various species of monkeys, including the baboon, the macaque and the cercopithecus. Following his lead, other scientists have investigated yawning in animals. When monkeys yawn, they display acuminate teeth. ‘Alpha males’ have been found to yawn more frequently not only than females, but also than less high-ranking males. For this reason, this display and consequently yawning have been taken to indicate dominance, aggression and territorial defence [25]. Besides monkeys and mice, all sorts of animals, including snakes, fish, penguins, crocodiles, and parrots yawn [27]. Given its widespread nature within the animal kingdom, yawning could have different functions in different species. All sorts of explanations have been given to explain yawning in man, ranging from oxygenation to brain cooling. Yawning has been suggested to foster wakefulness, or to communicate relaxation after a period of high vigilance. By being contagious, yawning indicates empathy, appreciation of other people’s behavioural and physiological states. Many more explanations have been offered. However, none of the above could apply to the fetus. A possible function of yawning during fetal life will be explored together with facial expressions in Chapter 8.
3.3 Gasping Apnoea is an almost universal phenomenon in mammals, before birth and with dying, just as gasping is nearly universal at the beginning and the end of life [28]. In the adult, gasping is a particular form of emergency respiration noted especially during choking, hypoglycaemia, neuromuscular failure and cardiac arrest. The patient takes one or more deep, frantic breaths with a full open mouth, literally gasping for air and for breath. Gasping is thus associated with potentially deadly conditions, and especially with being on the point of death, as reflected in the idiomatic expression ‘the last gasp’. In the neonate, gasping is a vital form of emergency respiration and
3 Hiccups, Yawning and Gasping
arousal permitting successful self-resuscitation and eventually the setting in motion of normal or so-called eupnoeic respiration. The human neonate has a particular resistance to hypoxia, but when hypoxia reaches a critical level, apnoea intervenes and gasping fires up. The infant revives, re-starting a benign cycle of respiration. However, the protective function of gasping and the elevated resistance to hypoxia only last for the neonatal period, defined as the first 28 days after birth, or slightly beyond. Sudden infant death syndrome (SIDS), or cot death, is rarely seen in the first 3 or 4 weeks of life; an incapacity to gasp has been regarded as a possible cause of this syndrome [28]. The fact that gasping persists for longer in the neonate was first demonstrated in 1812 by the physician Legallois in a range of animals which he drowned. Compared to older animals of the same species, the drowning newborn took longer to die and gasped for longer. In 1923 Thomas Lumsden was the first to demonstrate, by performing transections at various levels in animals, that the centres responsible for respiration are different from those presiding over gasping, and that gasping is generated by different mechanisms. Respiratory centres are located in the brainstem, especially the pons, and the central pattern generator guiding gasping is situated in a lower area at the junction between the pons and the first cervical level of the spinal cord [29]. Early in life gasping represents a simpler and coarser form of respiration, critical as a back-up for the neonate to re-start, but also to initiate eupnoeic breathing at birth. At or around the time of birth, gasping is important for two reasons. One has to do with pathology, and specifically with the so-called meconium aspiration syndrome, also called neonatal aspiration of meconium. Meconium is the first stool of the infant, composed of substances swallowed during fetal life such as dermal cells, intestinal epithelial cells, mucus, amniotic fluid, and vernix caseosa. Meconium generally is not passed until after birth, and unlike faeces is odourless and sterile. However, in 5 20% of all births, and especially in those past term, meconium is found in the amniotic fluid. Only 5% of these infants will develop the meconium aspiration syndrome. Fetal distress during labour triggers intestinal contractions and a relaxation of the anal sphincter with consequent so-called ‘staining’ of the amniotic fluid, which takes on a characteristic greenish colour. If the infant inhales the stained fluid, this may obstruct the airways, impeding the start of breathing, or by acting as an irritant it can cause chemical pneumonia [29].
References
However, at birth gasping is also vital for the shift to aerial respiration. In a very sketchy and simplified way one could say that during the delivery the lifeline of the umbilical cord and the placenta providing oxygen to the fetus are severed. The neonate starts gasping, with gasping the lungs expand and become air-filled, and breathing begins. Fetuses clearly do not gasp for air; however, occasionally they can be observed to perform ‘breathing’ efforts very similar to postnatal gasping. Hypoxia usually inhibits fetal breathing movements. This is thought to be a ‘sparing’ mechanism, saving precious oxygen for vulnerable regions such as the brain. However, fetal gasping can be induced by severe asphyxia. If fetal arterial PO2 falls to very low levels, gasping starts. Gasping causes intense activation of most inspiratory muscles and is composed of individual breathing efforts, separated from each other by respiratory arrest of varying duration. The gasping breath is considerably shorter and deeper than other forms of breathing, and can be easily distinguished from fetal breathing movements by the opening of the mouth, dropping of the jaw, and backward jerking of the head [29]. In the fetus, gasping has been generally associated with impending death due to late-pregnancy asphyxia and to events such as cord strangulation and meconium aspiration. Occasionally, however, even fetuses at earlier stages of pregnancy can be seen to gasp. In all our studies gasping was noted in three fetuses, at 15, 16, and 20 weeks. One fetus (20 weeks) had the umbilical cord around its neck, as ascertained by colour-Doppler ultrasonography. However, all quickly (less than 15 s) resumed normal activities and/or respiration. Interestingly, the gasping efforts were accompanied and immediately followed by some apparent vomiting efforts. One can only wonder if these could be a means of expelling some amniotic fluid inhaled through gasping. No ill effects could be detected in these fetuses at neonatal and 6-month checks, and at 1- and 2-year telephone follow-up. This could be explained by the wellknown capacity of newborns to withstand asphyxia better than adults. Hypoxia normally inhibits fetal breathing movements, but severe hypoxia may initiate gasping, a train of deep inspiratory efforts. Gasping was investigated in animals by Dawes and collaborators, who measured the time to the last gasp in fetal lambs and guinea pigs after tying the umbilical cord. Having found that the times to the last gasp were significantly greater at early gestational ages than at full term,
37
these authors concluded that post-asphyxial recovery occurred more rapidly and effectively in the early fetus. At this stage arterial chemoreceptors are not yet functioning, and gasping is probably solely triggered by a central effect. Unaided by other sensors, the respiratory centre itself may be more susceptible to asphyxia, yet more liable to recover from it [30]. Earlier in gestation the capacity to withstand hypoxia could be even greater, and gasping, albeit warranting prolonged monitoring, would not necessarily imply impending fetal death.
References 1. Dorland’s Illustrated Medical Dictionary (2007) 31st edn. Saunders, Philadelphia 2. Gilson I, Busalacchi M (1998) Marijuana for intractable hiccups. Lancet 351:267 3. Peleg R, Peleg A (2000) Case report: sexual intercourse as potential treatment for intractable hiccup. Can Fam Physician 46:1631 1632 4. Fesmire FM (1988) Termination of intractable hiccups with digital rectal massage. Ann Emerg Med 17:872 5. Shortt T (1934) Hiccup. Can Med Assoc J 31:38 41 6. Bayley H (1943) Persistent hiccup. Practitioner 150:173 177 7. Newson DJ (1970) An experimental study of hiccup. Brain 93:851 872 8. Park MH, Koh SB, Park MK, Lee DH (2005) Lesional location of lateral medullary infarction presenting hiccups (singultus). J Neurol Neurosurg Psychiatr 76:95 98 9. Golomb B (1990) Hiccup for hiccups. Nature London 345:774 10. Askenazy JJM (1992) About the mechanisms of hiccup. Eur Neurol 32:159 163 11. Marsot Dupuch K, Bousson V, Cabane J, Tubiana JM (1995) Intractable hiccups. Am J Neuroradiol 16:2093 2100 12. Kahrilas PJ, Guoxiang SHI (1997) Why do we hiccup? Gut 41:712 713 13. Launois S, Bizec JL, Whitelaw WA et al (1993) Hiccup in adults: an overview. Eur Respir J 6:563 575 14. Brouillette RT, Thatch BT, Abu Psba YK, Wilson SL (1980) Hiccups in infants: characteristics and effects on ventilation. J Pediatr 96:219 225 15. Vries JIP de, Visser GHA, Prechtl HFR (1982) The emergence of fetal behaviour I. Qualitative aspects. Early Hum Dev 7:301 322 16. Pillai M, James D (1990) Hiccups and breathing in human fetuses. Arch Dis Child 65:1072 1075 17. Stark RI, Myers MM (1995) Breathing and hiccups in the fetal baboon. In: Lecanuet JP, Fifer WP, Krasengor NA, Smotherman WP (eds) Fetal development: a psychobiological perspective. Lawrence Erlebaum, Hillsdale 18. Popescu EA, Popescu M, Bennett TL et al (2007) Magnetographic assessment of fetal hiccups and their effect on fetal heart rhythm. Physiol Meas 28:665 676 19. Straus C, Vasliakos K, Wilson RJA et al (2003) A phylogenetic hypothesis for the origin of hiccough. BioEssays 25:182 188
38 20. Shubin N (2008) Your inner fish. Allen Lane, London 21. Baenninger R, Binkley S, Baenninger M (1994) Field observation of yawning and activity in humans. Physiol Behav 55:412 425 22. Platek SM, Critton SR, Myers TE, Gallup GG Jr (2003) Contagious yawning: the role of self awareness and mental state attribution. Cogn Brain Res 17:223 227 23. Visser GHA, Laurini RN, Vries JIP (1985) Abnormal motor be haviour in anencephalic fetuses. Early Hum Dev 12:173 183 24. Deputte BL (1994) Ethological study of yawning in primates. Ethology 98:221 245 25. Ficca G, Salzaruolo P (2002) Lo sbadiglio dello struzzo. Psicologia e biologia dello sbadiglio. Bollati Boringhieri, Torino
3 Hiccups, Yawning and Gasping 26. Provine RR (2005) Yawning. American Scientist 93:532 539 27. Guntheroth WG, Kawabori I (1975) Hypoxic apnea and gasping. J Clin Invest 56:1371 1377 28. St John WM (1990) Neurogenesis, control, and functional sig nificance of gasping. Brief Review. J Appl Physiol 1305 1315 29. Pollack JA, Moise KJ Jr, Tyson WR, Galan HL (2003) The role of fetal breathing motions compared with gasping motions in pulmonary uptake of intra amniotic iron dextran. Am J Obstet Gynecol 189:958 962 30. Dawes GS, Fox HE, Leduc BM et al (1972) Respiratory move ments and rapid eye movement sleep in the foetal lamb. J Physiol 220:119 143
Fetal Breathing Movements
4
With the assistance of Luisa Bocconi, Chiara Boschetto, Florinda Ceriani and Alessandra Kustermann
Keywords Arterial chemoreceptors • Wakefulness • Ultrasound • Tachypnoeic • Rapid eye movement • Lung liquid • Pulmonary hypoplasia • Regular fetal breathing
Although aerial respiration only begins at birth, it is currently widely recognized that breathing has a long preparatory history during the 9 months preceding parturition. In describing the changes occurring at birth, most authors purposely omit the idiomatic expression ‘first breath’ and talk instead about a shift from periodic to continuous respiration. Prenatal breathing is a universal phenomenon amongst all animals that rely on aerial respiration after birth. Fetal breathing movements, which will also be referred to simply as breathing movements or breathing, had long been postulated to exist, but they only began to be investigated towards the turn of the last century. In 1905 von Ahlfeld made the first recordings applying a pressure sensor to the maternal abdominal wall, and described fetal breathing movements as periodic in nature and independent of the mother’s pulse and respiration [1]. Contemporary critics, however, attributed these movements to maternal motions. Later observations were made principally on preagonic human and animal fetuses extracted from the uterus, but still attached to the umbilical cord. Given their dying condition, only isolated and sporadic chest contractions could be observed prior to 18.5 weeks’ gestation. At later gestational ages no more than a few spontaneous breathing efforts were occasionally noted. These observations were interpreted as evidence of absence of spontaneous breathing in the healthy fetus in utero. Particularly influential was the work of Sir Joseph Barcroft and Don Barron in England during the late 1930s and 1940s. They delivered fetal lambs under A. Piontelli, Development of Normal Fetal Movements. © Springer Verlag Italia 2010
maternal spinal anaesthesia in a bath of warm saline solution, concluding that fetuses breathed at early stages of gestation, but that their breathing efforts ceased thereafter [2]. In 1970 Dawes and his co-workers in Oxford finally proved beyond doubt the existence of breathing movements in the fetal sheep by measuring tracheal pressure variations as an indication of fetal breathing. The periodic nature of breathing movements was also confirmed, together with the detection of their link with sleep, their anatomical substrate in the brainstem, and their independence from arterial chemoreceptors, which are principally but not only sense organs, located in the carotid, which are sensitive to chemical changes in the bloodstream and play a role in the regulation of respiration [3, 4]. Following this work, breathing in the human fetus was once again accepted and definitely recognized. Animal preparations, particularly sheep preparations, are still the principal means of investigation into breathing movements. Especially notable was the work of Henrique Rigatto and his collaborators, who inserted a double Plexiglas window into the abdomen of the pregnant sheep hence called ‘fenestrated’ through which fetal behaviour could be observed directly. This ingenious technique was devised principally for studying wakefulness in utero. Wakefulness, defined by opened eyes and purposeful head movements, was not noted in the fetal lamb even in the latest stages of pregnancy. On the other hand, breathing movements were observed to be a consistent feature of fetal life [5].
40
A further essential impetus in the study of breathing movements came from the advent of ultrasonography towards the end of the 1970s. Researchers concentrated mainly on the later stages of pregnancy, trying to detect whether breathing movements could be useful indicators of fetal compromise and/or well-being. As a consequence, a considerable body of knowledge on breathing movements during the advanced stages of pregnancy has accumulated [6, 7]. However, breathing movements turned out to be of scant clinical usefulness. Their rate of occurrence during a short observation time is so wide in both normal and pathological pregnancies as to make reliable predictions unfeasible. In the last 10 weeks of pregnancy it is actually impossible to distinguish within a reasonably short span whether the absence of breathing movements is due to physiological apnoea or to pathological causes. Observations lasting up to 2 4 hours would be necessary to settle this doubt, but these are clearly clinically impractical. A fairly reliable ‘fetal biophysical profile’, a test scoring of fetal well-being or suffering, was first developed by Frank Manning and his colleagues in 1980. The fetal biophysical profile combines the monitoring of breathing movements jointly with other biophysical variables such as fetal movement, fetal tone, amniotic fluid volume, and fetal heart rate. The latter is assessed with a non-stress test estimating fetal heart rate variability in relation to fetal movements. However, besides being lengthy, the fetal biophysical profile cannot be employed before the third trimester of pregnancy. Nevertheless some significant clinical observations have been made [8 10]. When fetuses become hypoxic, for whatever reason, they cease their ‘breathing’ efforts and redirect the diminished afflux of blood and oxygen towards more ‘sensitive’ and ‘higher ranking’ areas of the fetal body such as the central nervous system, the heart and the adrenals. Doppler colour flows show an increase of blood flow in such areas, thus alerting the obstetrician to a hypoxic problem [11]. Towards the end of pregnancy breathing movements are also known to increase 2 3 hours after maternal meals and maternal glucose intake [12]. Drinking alcohol and smoking, on the other hand, both cause a decrease of breathing movements in late gestation [13 15]. However, unlike breathing movements in late pregnancy, breathing movements in the pre-viable fetus have been little investigated. So far technological ad-
4 Fetal Breathing Movements
vances such as 4D ultrasonography have not added to our knowledge of fetal breathing movements, as they are too fast to be captured by computerized reconstruction. Real-time ultrasonography is still the elective method for studying breathing movements. This has shown that the start of breathing movements does not to coincide with the onset of movement at around 7 weeks’ gestational age. Breathing movements begin to be observed sporadically at 10 weeks and increase progressively with advancing gestation [16, 17]. Breathing movements are ‘paradoxical’ in their nature, consisting of vigorous downward movements of the diaphragm, simultaneously accompanied by a slight inward motion of the thorax (2 5 mm) and a greater (3 8 mm) outward movement of the abdomen, especially at the level of the umbilicus (Fig. 4.1). Like the neonate, who relies principally on nasal respiration, prior to 21 weeks fetuses do not open their mouth when breathing, and up to week 25 mouth opening during breathing is still a very occasional occurrence.
4.1 Fetal Breathing Movements: General Features From the start, breathing movements are not continuous, but occur in bouts of varying duration. The number of bouts of breathing movements increases steadily with advancing gestation. The duration of bouts varies with gestational age. Medium to short episodes (10 30 s) increase, while very short bouts (<10 s), which are prevalent at earlier stages, progressively decrease. Longer episodes (>30 s) are always rare and do not appear to be linked with fetal growth (Fig. 4.2). Up to 13 weeks fetuses show a mean breath-tobreath interval of 2 s. Subsequently breath-to-breath intervals show large variations. This variance can be found both between separate episodes, but also within the same bout. On average fetuses are fairly tachypnoeic, meaning they breathe rapidly, displaying a mean breath-to-breath frequency of 0.8 s (range: 0.5 120 s). Bouts of breathing movements show an increasing tendency to appear in sequence. Parallel to this, isolated bouts progressively decrease. Sequential episodes are separated from each other by ‘apnoeic’ intervals. Apnoeic spells are considered such if two bouts of breathing movements are separated by an interval that does not include other motions, save occasional startles or brief (2 8 s) isolated limb movements. Very short intervals (<10 s),
4.2 Fetal Breathing Movements: Non-Coincidence with other Behavioural Events
41
a
a
b
b
c
Fig. 4.1 Fetal breathing movements. Fetus 20 weeks’ gestation. a The fetus has not started a breathing movement. Seen in profile the thorax and the abdomen appear aligned. b The breathing movement has now started. The thorax has shifted slightly in wards and the abdomen is bulging out. c Near the end of the breathing movement thorax and abdomen are almost in full align ment again
prevalent at early stages, progressively decrease while longer apnoeic pauses (>40 s) increase with advancing gestation (Fig. 4.3).
Fig. 4.2 Fetal breathing movements (FBM): M developmental trends and duration of the bouts. a Developmental trends. Fetal breathing movements increase steadily and rapidly from 11 weeks’ gestation. The slight fall seen in the curve at week 17 could be due to the great variation in the number of bouts within an hour. b Duration of bouts. Short bouts (5 15 s) emerge throughout, but are prevalent till 16 18 weeks’ gestation. From then on, 15 to 25 s and 25 to 35 s bouts become prevalent. Bouts of a longer duration (35 45 s and >45 s), almost non existent at earlier ages, become consistent at 19 20 weeks and 23 24 weeks respectively
4.2 Fetal Breathing Movements: Non-Coincidence with other Behavioural Events Save for occasional startles or brief (2 6 s) isolated limb movements, breathing movements are independent and non-coincident with all other behavioural patterns. At 21 weeks an occasional and short coincidence between
42
4 Fetal Breathing Movements
weeks (g a.)
a
solely in an either/or mode (Fig. 4.4). However, though largely continuing to surface within the same area of activity, when fetuses breathe, swallowing movements do not emerge or overlap, and are possibly inhibited. The fine-tuning of these activities, which are needed for feeding after birth, is not yet established by 25 weeks. Breathing movements are also separated from hiccups. Contrary to breathing movements, hiccups appear to be compatible with movement from 10 weeks. Although some bouts of breathing movements are immediately (1 3 s) preceded by startles, most are not. This occurs more frequently up to 12 13 weeks, when startles reach a peak, and decreases thereafter. These observations seem to indicate that during the first half of pregnancy the respiratory central pattern generators, which have been found to be spread over multiple sites in the brainstem of the fetal mouse [18], work autonomously and do not act in concert with other variables. After 20 weeks the central pattern generators presiding over various activities begin to act in concert, and clusters of activities start to merge. However, up to 25 weeks fetuses still largely function in an either/or mode. This way of operating can have a clinical significance for extremely premature infants.
4.3 Apnoeic Pauses b
Fig. 4.3 Fetal breathing movements: isolated and sequential bouts and duration of apnoeic pauses. a Isolated fetal breathing move ments are prevalent up until 12 weeks. After this, sequential bouts largely prevail. b Short (<10 s) pauses between breathing bouts prevail until 16 weeks. Pauses lasting 10 20 s are relevant through out. Pauses lasting more than 40 s increase sharply from 16 weeks, and pauses of 20 30 s from 18 20 weeks
breathing movements and general movements begins to be noted. However, such coincidence remains a rare occurrence lasting just a few seconds (range: 3 9 s). Fetuses momentarily start both breathing and moving, but then either stop breathing and continue moving, or simultaneously stop both activities. Possibly the potential to breathe while moving is established by then, but is not functional to the intrauterine requirements. Up to 25 weeks breathing and swallowing are also discrete events. Up to 20 weeks breathing and swallowing occur within the same cluster. At around 20 weeks clusters begin to dissolve, and fetuses no longer function
As said, unlike in postnatal life, fetal breathing movements are not continuous, but periodic in their nature. Periodic breathing is still a benign and frequent respiratory pattern in the premature and even in the young term newborn. The infant breathes regularly for 10 18 s. Breathing is then interrupted by pauses of at least 3 s, and the whole cycle lasts for about 2 min. Severely premature infants in particular, however, can suffer from apnoeas, respiratory pauses lasting 20 s or more, accompanied by significant consequences such as bradycardia (less than 100 beats/minute), hypoxia, cyanosis, or pallor, bradyarrhythmia, or abnormally slowed heart rhythm, and a fall in blood pressure. Apnoeic spells are totally physiologic in the fetus, but acquire clinical relevance in the severely premature infant whose respiration, as in the fetal mode, becomes interspersed by life-threatening apnoeas no longer adapted to the extrauterine conditions [19]. The non-conjugation between breathing movements, sucking and swallowing during fetal life is also relevant for clinical practice. The premature infant is no longer
4.3 Apnoeic Pauses
General Movements Locali lized d Moti tions Rest Sy chr Syn chrono onous us Mot Motion ionss
43
St rtles Sta tl Hiccup ps Twitches
Fetall Breathi thing gM Movements Swallowingg Reggular Feta etal Brea reaathing ng
Fig. 4.4 Fetal breathing movements and general movements at 20 weeks’ gestation. See Fig. 2.4 legend for an explanation of the chart design. The division between general movements (red) and breathing movements (dark blue) begins to be no longer absolute. Although division is still prevalent, fetuses start being able to perform generalized motions and to breathe. The same cannot said for swallowing (light pink), k which continues to be a separate event
dependent on the placenta and the umbilical cord to fulfil most of its nutritive and oxygen needs, but has to breathe autonomously as well as to feed without suspending breathing for unduly prolonged spells, and without choking. Severely premature infants often have trouble coordinating these activities. Feeding can cause suspension of breathing, or so-called feeding-induced apnoea, just as fetal sucking and swallowing cause suspension of breathing during fetal life. Many of these infants in fact have to be tube-fed. Coordination between breathing, sucking, and swallowing which allows oral feeding usually starts at around 32 35 weeks [20]. Equally, the increasing duration of apnoeic intervals with advancing gestation is of clinical relevance. The early fetus mainly performs short bouts of breathing movements interspersed by equally short pauses of apnoea. Growth brings with it an increasing capacity to sustain prolonged and less fragmented episodes of breathing movements. Parallel to this, however, the intervals between the bouts also increase in duration, and around mid-pregnancy long apnoeic spells become a consistent feature of fetal breathing. This physiological characteristic can also acquire pathological significance in the premature, and occasionally also in the neonate, which
under certain circumstances could revert to a fetal form of breathing with long apnoeas separating breathing spells. In order to prevent respiratory arrest, most severely premature infants have to be kept on a ventilator. Furthermore, the suppression of breathing during hypoxia is functional to the fetal stage. This efficient ‘sparing’ method, present during prenatal life, can have dire consequences if triggered in the premature or neonate. Hypoxia of whatever origin could expose some vulnerable neonate to the danger of reverting to a fetal form of breathing. Hypoxia and obstruction of the upper airways (with consequent hypoxia) have been postulated to be at the basis of some cases of sudden infant death syndrome (SIDS) [21]. The weak link between general movements and breathing could be of clinical relevance. Behavioural states comparable to the various phases of sleep in the neonate only emerge towards the end of pregnancy, at around 36 weeks. These states have been defined according to the criterion that during them several variables have to recur in specific, fixed, and temporally stable combinations. In late pregnancy, breathing movements have been found to be concomitant to state 2F as described by Prechtl and his co-workers. State 2F (a
44
stage of sleep) is accompanied by frequent and periodic gross body movements, mostly stretches and retroflexions, and by movements of the extremities. Eye movements occur continuously, and the fetal heart rate shows wide oscillations with frequent accelerations in association with movements [22]. Other authors, using different terminology, note a coincidence between breathing movements and REM or active sleep in the advanced stages of gestation. As will be explained, the terminology ascribed to fetal sleep is fairly confused and confusing. During active sleep fetuses approaching term display both paradoxical abdominal respiration and irregular respiration, move, and occasionally perform sucking and swallowing motions [22]. Observation of the noncoupling between breathing movements, swallowing and hiccups, and the limited coupling between breathing movements and general movements up to 25 weeks, could alert the clinician to regard active sleep as a particularly sensitive state during which the premature (but also the neonate) could be more prone to function again as a fetus by suspending breathing when performing a variety of activities that are not compatible with each other during fetal life.
4.4 Possible Functional Significance The functional significance of breathing movements is not yet fully understood. However, much essential information is now known. The most obvious and major feature differentiating fetal from postnatal breathing is that fetuses do not breathe air and are dependent largely on the mother for oxygen supply. During fetal life the future air spaces are filled with a particular liquid that is produced by the lung’s epithelium [23]. Early debates about the fetal lung revolved around the origins of this liquid. The preferred hypothesis was that the liquid was composed of amniotic fluid aspirated by breathing movements. The amniotic origin of lung liquid was disproved in the 1940s by Jost and Pollicard, who observed progressive lung distension in fetal rabbits whose tracheas had been ligated [24]. Further evidence of the local production of fetal lung liquid came from the observation that the composition of fetal lung liquid differed from that of amniotic fluid in fetal sheep. The source of lung liquid was shown to derive from an active secretory process by the epithelium lining the developing lung. This liquid plays a crucial role in
4 Fetal Breathing Movements
fetal lung development. By maintaining the future air spaces in a distended state, it restricts the entry into the lungs of amniotic fluid, which can potentially have damaging effects. The volume of the liquid within the future air spaces and its flux to and from the lower airways are influenced by fetal muscular activity. By maintaining the lung in a distended state, fetal liquid serves as an internal ‘splint’ around which the distal air spaces of the lung develop. Without this underlying degree of expansion, the fetal lung is unable to grow and structurally mature [23]. Although it is likely that breathing movements are important for normal lung growth, they do not influence the rate of liquid secretion or the net movement of lung liquid out of the lung. The secreted liquid is passively expelled into the trachea and into the oropharynx, where it is either swallowed or expelled into the amniotic cavity. The net flow is out of the lungs. This essentially unidirectional flux maintains a constant chemical environment within the developing air spaces, thus restricting the entry of potentially harmful substances in the amniotic fluid such as meconium during the late stages of pregnancy [23]. Expansion of the fetal lung is crucial for lung development. Lung expansion is maintained by and dependent on a series of activities, all set in motion during breathing movements. Factors that reduce or inhibit these muscular activities result in a reduction in fetal lung expansion which is currently recognized as underlying various disorders, especially the so-called pulmonary hypoplasia in human infants [25]. Fetal lung development is dependent on complicated interactions of several endocrine, metabolic, and mechanical factors consisting especially of breathing movements. During fetal life the future airways of the lung are liquid-filled and gas exchange occurs across the placenta. At birth, the lung must rapidly assume a role that it has not previously performed, and in most cases the transition to an air-breathing existence is rapid and uneventful. The lungs must be structurally and biochemically mature so as to inflate with ease and not collapse during expiration [25]. To exchange respiratory gases successfully, the lung must have developed millions of thin-walled alveoli perfused by an extensive vascular network and the airways must be cleared of liquid. Lung development goes through constant changes during fetal life. At least five distinct stages can be schematically distinguished. During the embryonic stage (3.5 7 weeks), the trachea separates from the
4.5 Neurological Substrate
45
oesophagus and begins to be formed. Buds of the lung, of primary bronchi, and of major airways develop. A so-called pseudoglandular stage follows between 5 and 17 weeks. During this stage future pulmonary tubules and acini start to develop, and the capillary vascular bed increases. Cells responsible for surfactant production also start to differentiate. Starting at 24 and up until 38 weeks, alveolar ducts and air sacs are formed and surfactant is increasingly produced. Finally, during the alveolar stage, which last until 2 years, alveoli are formed and the capillary bed fuses into a single network [26].
4.5 Neurological Substrate Respiratory gas exchange after birth largely depends upon the ability of respiratory muscles to move air into and from the lungs. After birth the respiratory muscles are activated in a rhythmic, coordinated manner via the activation of brainstem neuronal networks. Until late in pregnancy breathing movements are solely caused by an output from the brainstem. No evidence of central or peripheral mechanisms regulating breathing movements has been found. In animal preparations, vagotomy and denervation of carotid and aortic chemoreceptors did not modify the incidence of breathing activity [27]. Most afferent postnatal stimuli are also absent during fetal life. The airways are full of lung fluid, pulmonary circu-
lation is minimal, and due to constant gas exchanges through the placenta, blood gases show little variation. The principal muscle activated at all gestational ages is the diaphragm. The diaphragm is innervated principally by the phrenic nerve, arising in the mid-cervical spinal cord. Up to mid-gestation, in synchrony with the inspiratory phase of the respiratory cycle, bursts of intercostal muscle activity can be noted in animals. Later these are possibly inhibited by descending neural influences. Muscles of the upper respiratory tract such as those of the larynx, nostrils, and tongue are all activated. The dilator muscle of the larynx is innervated by a branch of the vagus, the recurrent laryngeal nerve, originating from the brainstem. The contraction of the respiratory muscles causes an alteration of the shape of the thorax and lungs and is stopped by section of the phrenic nerve [28]. During fetal life, breathing movements have an essential and unique role in promoting lung growth, development, and functioning. However, they are hardly at all involved in respiration as we mean it after birth. Breathing movements start at 10 weeks. At 18 20 weeks another type of ‘breathing’ emerges. Regular fetal breathing is a form of superficial, regular and nonparadoxical breathing characterized by low synchronized outward excursions of the thorax and the abdomen. Up to 25 weeks the episodes of regular fetal breathing are isolated, brief events and their frequency remains low. These episodes surface during cycles of rest (Fig. 4.5).
a
Fig. 4.5 Fetal breathing movements (FBM) M and regular fetal breathing movements (R FBM): M developmental trends and emergence. a Regular fetal breathing movements start emerging at 15 16 weeks. From then on, regular breathing continues on an even slope. However, the quota of regular breathing movement continues to be quite low up to 25 weeks. (cont.)
46
4 Fetal Breathing Movements
(continua)
General Movementts Localized Motions Rest Synchr Syn chrono onous us Mot Motion ionss
St rtles Sta tl Hiccup Hic cupss Twitch Twi tches es
Fetall Breathi thing gM Movements t Swallo Swa llowin wingg R ular Fetal Breathi Reg hing ng
b
Fig. 4.5 b Fetal breathing movements and regular fetal breathing at 18 weeks. See Fig. 2.4 legend for an explanation of the chart design. Only one bout of regular breathing (light blue) can be noted. The bout emerges within an epoch of relative rest lasting 7.52 min, only interrupted by sparse (5) and short (range 5 12 s) localized motions, and equally sparse startles (5) and twitches (2). On the other hand, eight bouts of breathing movements (dark blue) surface before the period of rest
Given its scarce relevance during the first 25 weeks which form the theme of this work, the possible significance of this form of breathing will be considered at some length when rest-activity cycles and the origins of sleep are discussed in Chapter 8.
References 1. Ahlfeld von F (1988) In: Windle WF (1940) Physiology of the fetus. Saunders, Philadelphia 2. Barcroft J, Barron DH (1937) Movements in midfoetal life in the sheep embryo. J Physiol 91:329 351 3. Dawes GS, Fox HE, Leduc BM (1970) Respiratory movements and rapid eye movement sleep in fetal lambs. J Physiol 210:77 4. Dawes GS (1982) The central control of fetal breathing and skeletal muscle movements. Review Lecture. J Physiol 346:1 18 5. Rigatto H, Moore M, Cates D (1986) Fetal breathing and be haviour measured through a double Plexiglas window in sheep. J Appl Physiol 61:160 164 6. Haddad G (1991) Developing neurobiology of breathing. CRC Press, Boca Raton 7. Hanson MA, Spencer JAD, Rodeck CH, Walters D (eds) (1994) Breathing. Fetus and neonate: physiology and clinical
applications, vol 2. Cambridge University Press, Cambridge, UK 8. Manning FA, Platt LD, Sipos L (1980) Antepartum fetal evaluation: development of a fetal biophysical score. Am J Obstet Gynecol 151:343 347 9. Manning FA (1995) Fetal biophysical profile scoring. In: FA Manning (ed) Fetal medicine: principles and practice. Appleton and Lange, Norwalk 10. Patrick GE, Campbell K, Charmichael L (1980) Patterns of human fetal breathing during the last ten weeks of pregnancy. Obstet Gynecol 56:24 30 11. Beckdam DJ, Visser GHA (1985) Effects of hypoxemic events on breathing, body movements and heart rate variation. Am J Obstet Gynecol 153:52 56 12. Patrick NR, Richardson B (1978) Effects of human maternal plasma glucose concentrations on fetal breathing movements. Am J Obstet Gynecol 132:36 41 13. Fox HE, Steinbrecher M, Pessel D et al (1978) Maternal ethanol ingestion and the occurrence of human fetal breathing movements. Am J Obstet Gynecol 132:354 358 14. Mulder EJ, Morssnik LP, van der Schee T, Visser GH (1998) Acute maternal alcohol consumption disrupts behavioural state organization in the near term fetus. Pediatr Res 44:774 749 15. Gennser G, Marsal K, Brantmark B (1975) Maternal smoking and fetal breathing movements. Am J Obstet Gynecol 123:861 867
References 16. de Vries JIP, Visser GHA, Prechtl HFR (1982) The emergence of fetal behaviour. I. Qualitative aspects. Early Hum Dev 7:301 322 17. de Vries JIP, Visser GHA, Prechtl HFR (1984) Fetal motility in the first half of pregnancy. In: HFR Prechtl (ed) Continuity of neural functions from prenatal to postnatal life. Clinics in Developmental Medicine, no 94. Mac Keith Press, London, pp 46 64 18. Eugenin G, Nicholls JG, Cohen LB, Muller KJ (2006) Optical recording from respiratory pattern generator in the fetal mouse brainstem reveals a distributed network. Neuroscience 137:1221 1228 19. Miller MJ, Martin RJ (2004) Pathophysiology of apnoea of prematurity. In: Polin RA, Fox WW, Abman SH (eds) Fetal and neonatal physiology, vol 2, 3rd edn. Saunders, Philadelphia, pp 905 918 20. Lau C, Sheena H, Shulman R, Schanler R (1997) Oral feeding in low birth weight infants. J Pediatr 130:561 569 21. DeCristofor JJ (2005) Apnoea of prematurity and apparent life threatening events. In: Spitzer AR (ed) Intensive care of the fetus and neonate, 2nd edn. Elsevier, Philadelphia, pp 541 556 22. Njijhuis JG, Prechtl HFR, Martin CB, Bots RSGM (1982)
47 Are there behavioural states in the human fetus? Early Hum Dev 6:177 195 23. Barker PM, Wolfson MR (2006) Regulation of liquid secretion and absorption by fetal and neonatal lung. In: Polin RA, Fox WW, Abman SH (eds) Fetal and neonatal physiology, vol 2, 3rd edn. Saunders, Philadelphia, pp 905 918 24. Jost A, Policard A (1948) Contribution expérimentale à l’étude du développement prénatal du poumon chez le lapin. Arch Anat Microsc 37:323 332 25. Hilsop A (2003) Fetal and postnatal anatomical development. In: Greenough A, Milner AD (eds) Neonatal respiratory dis orders, 2nd edn. Arnold, London, pp 1 11 26. Burri PH (1997) Structural aspects of prenatal and postnatal development and growth of the lung. In: McDonald JA (ed) Lung growth and development. Marcel Dekker, New York, pp 1 12 27. Rigatto H (2006) Control of breathing in fetal life and onset and control of breathing in the neonate. In: Polin RA, Fox WW, Abman SH (eds) Fetal and neonatal physiology, vol 2, 3rd edn. Saunders, Philadelphia, pp 890 900 28. Fewell JE, Lee CC, Kitterman JA (1981) Effects of phrenic nerve section on the respiratory system of fetal lambs. J Appl Physiol 51:293 297
Swallowing, Sucking, and Handedness as Inferred from Fetal Thumb Sucking
5
With the assistance of Chiara Boschetto, Florinda Ceriani, Isabella Fabietti, Roberto Fogliani and Alessandra Kustermann
Keywords Swallowing • Sucking • Fetal thumb sucking • Handedness • Feeding • Hypotonic urine • Oligohydramnios • Polyhydramnios
Wide-ranging theories have flourished around fetal nutrition for centuries. The Greeks were somehow more accurate in their intuitions than many of those who followed them. Democritus and Epicurus thought that the unformed fetus ate and drank ‘per os’, through the mouth [1]. Hippocrates rightly assumed that the maternal blood flow nourished the embryo and even respiration emerged from the cord. Subsequently an almost ‘agricultural’ view of nutrition prevailed. Fetuses were regarded as passive creatures which simply absorbed nutrients from their mothers, who were equated to the earth and its fertile soil. Albertus Magnus thought that embryos absorbed nutrients ‘like a sponge’. Others, like Hildegard of Bingen, believed that retained menstrual blood was the primary source of fetal nourishment [1]. Menstruation ceased in the pregnant woman, and this was taken as a sign that menstrual blood was the fetus’s main food. Still others imagined the fetus branching out many vessels into the placenta, the socalled matrix, through which the nourishment was sucked in as from a fertile terrain. Some allowed the fetus hunger, but they also believed its appetite to be immediately satisfied by the life-giving maternal womb. Throughout the centuries the mother’s contribution to the act of creation was endlessly debated and most often hotly denied, but her nutritive functions were not [2]. However, the uterus was not a safe haven: it also had the power of harming the unborn with toxic products, including mental ones, which could foster or thwart its healthy growth. Maternal thoughts and fantasies were deemed to be passed on to the fetus by a A. Piontelli, Development of Normal Fetal Movements. © Springer Verlag Italia 2010
kind of thinking uterus, and the fetus could be forever imprinted by them. A similar philosophy continues to this day when the fetus is thought to be passively soaking up even unconscious states of mind, particularly anxiety, from the mother. However, apart from lingering beliefs such as this, we now know a lot about fetal nutritional physiology. Much of our data comes from fetal sheep, but new technologies increasingly contribute to make even the human fetus a subject of investigation. Fetuses start sucking long before they attach to the breast, and swallowing long before they start to drink their mother’s milk. However, achieving effective and safe feeding from the breast and even more so the bottle takes the whole of pregnancy and beyond. Neonatal feeding requires a coordinated, rhythmic pattern of sucking, swallowing and breathing involving a fine-tuned interaction between the lips, jaw, tongue, palate, pharynx, larynx and oesophagus [3]. The pharynx serves as a conduit for food passing into the oesophagus as well as for air passing into the larynx, and precise coordination of breathing, sucking and swallowing must be achieved before the neonate can start feeding safely through the mouth [4].
5.1 Swallowing and Sucking: General Features Fetal swallowing begins before sucking. This latter is not fully developed until after birth. Fetal (and neonatal) sucking involves drawing a liquid into the mouth by
50
5 Swallowing, Sucking, and Handedness as Inferred from Fetal Thumb Sucking
creating a partial vacuum through the motions of the mouth, tongue, and lips. Swallowing, on the other hand, implies taking in a substance a fluid in the case of the fetus and the neonate through the mouth and pharynx, past the upper oesophageal (also called cricopharyngeal) sphincter, which contracts to prevent food from returning to the pharynx and the mouth, and finally through the oesophagus and into the stomach [5]. Swallowing activities achieve stability of form and rhythm earlier than sucking activities. At 32 34 weeks an immature suck swallow pattern consisting of swallowing before or after short sucking bursts emerges. A second pattern is seen at 34 36 weeks, consisting of longer suck bursts with swallowing integrated into the bursts [6, 7]. Additionally, sucking has two components: expression and suction. The positive pressure exerted by the squeezing of the nipple between the hard palate and tongue leads to milk expression. After expression, the negative intraoral pressure favours suction of the milk into the oral cavity. Initially expression predominates and is coordinated with swallowing. As the infant matures, suction and expression begin to be coordinated into an alternating pattern, and both become increasingly fine-tuned with swallowing. This long maturational process is not learnt, but is linked to gestational age. In the premature, oral feeding is generally not attempted before 32 weeks. However, most infants cannot achieve oral nutrition before 34 36 weeks [8]. Despite being quite different, and complex activities, the terms sucking and swallowing are often used interchangeably. Fetuses start swallowing at 11 weeks [9] (Fig. 5.1). At 10 11 weeks the palate fuses, forming the hard palate. The soft palate, originating from both sides of the adjacent pharyngeal mesenchymal tissue (an embryonic connective tissue), also begins to form and fuse. The formation of the palate separates the oral cavity from the nasal cavities [10]. By 10 11 weeks the naturally prolapsed intestinal loops also return into the abdominal cavity [10]. Simultaneously with these anatomical changes, swallowing begins. Nevertheless, prior to 15 16 weeks not all fetuses will be observed to swallow during half-hour observations. Only at 18 20 weeks does swallowing become a consistent and repeated feature of each observation. However, contrary to breathing, swallowing cannot be visualized at all times. Fetuses may turn in an unfavourable position, impeding the view of their face. For this reason, the true rate of occurrence of the phenomenon is quite probably underestimated.
Fig. 5.1 Swallowing (Sw): developmental trends. Swallowing starts at 12 weeks’ gestation and increases steadily until 25 weeks. The decrease observed at 23 and 24 weeks may be due to the fact that swallowing cannot be visualized at all times. Head rotations, which are particularly frequent after the first half of pregnancy, make true evaluation of swallowing impossible
5.2 Swallowing: Development The first swallowing motions emerge as isolated bouts lasting a few seconds each (range 4 6 s). Up to 13 weeks swallowing bouts occur as sporadic, one-off events. From 15 to 16 weeks bouts show a tendency to occur in a series of three to four episodes interspersed with more or less long pauses (range 8 20 s). Parallel to this, the duration of the bouts increases (range 8 15 s). From 16 to 25 weeks both the number (range 4 8) of consecutive bouts and their duration (range 10 20 s) increase. Up to 20 weeks, swallowing and breathing largely take place within the same cluster when general movements are not present, and before a period of quiet starts. However, until 25 weeks swallowing remains a discrete event. The interval between bouts of swallowing and bouts of breathing is progressively reduced, but the two do not begin to intermingle, nor does swallowing occur in synchrony with any other motor event, save occasional and brief (2 4 s) localized movements (Fig. 5.2). The mechanics of swallowing change with increasing gestation. Up until 15 weeks fetuses open and close their mouth in a sequence similar to some fishes. Closure of the lips is not necessarily complete, as the lips can either seal tight or remain slightly open. Nor is the
5.2 Swallowing: Development
General Movementts Locali lized dM Moti tions Rest Synchr Syn chrono onous us Mot Motion i s ion
51
St rtles Sta tl Hiccup cup ps Twitches
Fetall Breathi thing M Movements t Swallowingg Reggular Fetal Breathingg
Fig. 5.2 Swallowing patterns at 18 weeks’ gestation. See Fig. 2.4 legend for an explanation of the chart design. Swallowing bouts (light pink) tend to occur in sequence. However, save for two bouts showing a brief (3 5 s) coincidence with localized motions (green), swal lowing does not occur synchronously with other motor events. In particular, breathing (blue) and swallowing are still separate and non coordinated
mouth inevitably opened wide. Regular sealing, necessary to latch the nipple firmly into place together with tongue motions is not noted during swallowing before 22 24 weeks. This activity causes the amniotic fluid to be inhaled into the oral cavity. Once the fluid has been captured within it, slight and repeated movements of the tongue direct the fluid into the pharynges. Minimal cupping of the tongue, which is not observed prior to 15 16 weeks, seems to act like a straw, helping to draw the fluid into the mouth. As pregnancy progresses, cupping becomes increasingly pronounced [11]. These data have been confirmed by additional overlapping of colour-Doppler ultrasound to demonstrate swallowing of the amniotic fluid. Besides swallowing through the mouth, some fluid is inhaled via the nostrils, and some is regurgitated [12]. From 18 weeks, movements of the tongue become more vigorous, and cupping and opening of the mouth more pronounced (Fig. 5.3). After 20 weeks, strong anterior and posterior motions of the tongue propel the fluid into the pharynx after each episode of mouth opening and closing. Between 20 and 25 weeks the tongue increasingly rolls backwards, nearly touching the palate. Fetuses have ‘big tongues’; by 25 weeks
the size of the fetal tongue is half the size of that of an adolescent, and a big tongue will help to latch the nipple firmly against the palate [13]. In the neonate, during feeding, the tongue is primarily used to push the nipple upwards and forwards, latching and squeezing it [14]. Slight protrusion of the tongue, anticipatory of the necessary forwards movement, can be noted from 16 weeks. Tongue protrusion becomes more marked at 20 weeks. Fetuses can be seen licking components of their bodies, especially their hands. Between 23 and 25 weeks full protrusion as well as lateral movements of the tongue are exhibited, and fetuses start licking components of the intrauterine environment such as the placenta and the cord (Fig. 5.4). Random laryngeal contractions in the absence of swallowing accompanied by occasional diaphragmatic contractions are noted after 18 20 weeks and consistently between 24 and 25 weeks. Like Miller and collaborators [11], we interpret these as possibly being pre-phonatory manifestations. From this time on, fetuses may start preparing to utter vocal sounds. As can be seen in Figure 5.5, even clearer preparatory signs are revealed by facial expressions. Like breathing, fetal swallowing is inhibited during
52
5 Swallowing, Sucking, and Handedness as Inferred from Fetal Thumb Sucking
a
b
c
d
Fig. 5.3 Swallowing: changes with increasing gestation. a Fetus at 12 weeks’ gestation. The mouth is opened in a fish like manner, and the tongue cannot be seen. b Fetus at 13 weeks’ gestation. The mouth is only slightly open, and the tongue cannot be seen. c Fetus at 24 weeks’ gestation. The mouth is wide open, but the tongue cannot be visualized in this picture. d Fetus at 24 weeks’ ges tation. The mouth is wide open. The voluminous tongue shows a slight cupping, indicating its initial participation in swallowing motions
hypoxia. Inhibition is considered a ‘sparing’ mechanism, whereby valuable oxygen, as indicated by blood flows, is directed towards more fragile and ‘higherranking’ regions of the body, such as the brain, the heart, and the adrenals. At around mid-pregnancy the stomach can be seen to fill after a few sustained bouts of swallowing. Equally the bladder voids. In the human fetus, hypotonic urine probably enters the amniotic sac at around 10 weeks. Urine entry may be continuous at this early age as the
urinary bladder and its sphincters are not fully developed. During the later stages of gestation, the bladder fills every 20 30 min and is emptied by active contraction of the bladder wall. Urine leaves the bladder through the urethra and is excreted into the amniotic fluid. Fetal urine is different from neonatal urine, and even more so from adult urine. We know very little of its function in fetal swallowing. However, if fetal urine does not enter the amniotic sac regularly, severe oligohydramnios develops [15].
5.3 Fetal Swallowing: Possible Functions
53
a a
b
b
Fig. 5 5 Preparing to utter vowels. a Fetus at 25 weeks’ gestation. b Fetus at 23 weeks’ gestation. Both fetuses have their mouth open. However, the shaping of the mouth does not indicate swal lowing, and is very similar to the shaping of the mouth seen in the neonate when uttering vowels. Further observation (not displayed here) confirmed that these two fetuses were not swallowing
5.3 Fetal Swallowing: Possible Functions
c
Fig. 5.4 Beginnings of licking motions and considerable size of the tongue. Fetuses at 25 weeks’ gestation. a Fetus licking the umbilical cord. b Fetus licking the placenta. c The fetus is not licking, only preparing to do so, but the tongue is noticeable in its remarkable size
Although the placenta takes care of the main nutritional needs of the fetus, in the human fetus 60 70% of the protein in the amniotic fluid is turned over every day by fetal swallowing. Studies in sheep suggest that the input of protein from the amniotic fluid alone could account for 15 20% of total body protein deposition. By 16 weeks a sheep fetus swallows 2 6 ml of amniotic fluid per day, increasing to an average of 450 ml per
54
5 Swallowing, Sucking, and Handedness as Inferred from Fetal Thumb Sucking
day near term. About 20% of the fluid swallowed by the fetus is lung fluid, not amniotic fluid [16]. Swallowing is important for amniotic fluid homeostasis, regulation of amniotic fluid volume, gastrointestinal development and somatic growth. Failure of the fetus to swallow amniotic fluid is associated with gastrointestinal obstruction and polyhydramnios. The human fetus swallows large quantities of fluid, estimated at 300 1000 ml per day. Swallowed fluid contains contributions from amniotic fluid, including fetal urine, lung liquid and oral secretions. Further secretions are added from the gut, liver, and pancreas. The intestinal lining is severely impaired and the gut develops abnormally in the absence of swallowed fluid input. Studies of sheep indicate that swallowing may be important for intestinal development in many ways, including the composition of fluid, the ingested volume, the frequency of ingestion, and the presence of specific factors ranging from hormones to lung liquid and growth factors. It is likely that the swallowed fluid also distends the gut tube. Gut tissues are severely growth-retarded and develop abnormally in the absence of swallowed fluid input. Furthermore, intestinal growth correlates well with liver growth when the fetal oesophagus is ligated. The correlation shifts to favour more intestinal growth when swallowing is restored after oesophageal ligation. Even more surprising is the fact that when the composition of the swallowed fluid is altered, the body allometry the relative proportions of various organs and often their shape is also changed. This indicates that fetal organ growth is a highly coordinated process exhibiting an extremely high degree of whole-body cooperation and communication [17].
5.4 Swallowing: Possible Regulation In the fetal sheep, about two to seven swallowing episodes occur per day; however, it is not clear how well regulated fetal swallowing actually is. Nor it is known what triggers the urge for fetal ingestion [17]. Interesting findings could come from newborn rats. These seem not to experience ‘appetite’ as adult or even slightly older rats do. Young pups spend the majority of their time attached to the nipple, even when sleeping. Milk availability which is not continuous, rather than internal cues linked to hydration, gastric fullness, or metabolic state, would appear to regulate the intake. If experimentally provided with infused milk continuously, these rat pups could literally drown in
food. However, an important regulator seems to be the sleepiness caused by gastric distension. Like human neonates, young pups become drowsy and soon plunge into sleep once their stomachs are full. Once asleep, though still attached to the nipple, they stop sucking and swallowing. In other words, the pup’s behavioural state would be the main regulator of food ingestion [16]. Interestingly, human fetuses concentrate their swallowing before starting a cycle of rest, and, conversely, rest starts once fetuses have stopped swallowing and thus have distended stomachs. Initially and up to midpregnancy, swallowing and breathing take place within the same cluster, when general movements have ceased and prolonged rest has not started. With advancing pregnancy, breathing and intense moving begin to overlap. However, bouts of swallowing continue to occur just before periods of rest, and cease once these start. The analogy may be far fetched; nevertheless, the human fetus’ ingestive behaviour also seems to be regulated by states, albeit not yet mature ones. In other words, swallowing bouts precede rest cycles, just as months later the neonate tends to fall into a state of drowsiness after a good feed. It has also been suggested that swallowing may be important in fetal thirst and appetite programming [18]. The neural mechanisms involved in swallowing involve a number of nerves. The most important are the ninth and tenth pairs of cranial nerves (glossopharyngeal and vagus). A branch of the vagus nerve carries important sensory input from the larynx, epiglottis and vallecular storage area, an area that is present in infants but not in older children and adults. Swallowing can be elicited reflexively by fluid in the vallecular space even when there are no connections from higher parts of the brain above the brainstem, as anencephalic infants have demonstrated. All the necessary neural components for swallowing are present below the level of the mid brain or mesencephalon, where the central pattern generator is presumably located [19 21]. Only growth will bring with it a progressive ‘corticalization’ of swallowing after birth.
5.5 Handedness in the Human Fetus as Assessed by Thumb-Sucking As we have said, coordinated sucking is not observed by 25 weeks, but emerges between 30 to 40 weeks, and continues to be perfected after birth.
5.5 Handedness in the Human Fetus as Assessed by Thumb-Sucking
55
a
b
c
d
e
f
Fig. 5.6 Is thumb sucking observed before 25 weeks? a c Fetus at 14 weeks’ gestation. a The fetus has the mouth slightly open some could infer in anticipation. Three digits can be seen in the lower part of the picture. b The digits are now moved to touch the face. The thumb in particular is touching the mouth. However, the mouth has not changed its shape as if preparing for insertion. Only the head is bent slightly forwards. c A moment later the digits are moved from the face, and the mouth has not changed its shape, but looks again as in a. The head has remained slightly bent. d, e Fetus at 18 weeks’ gestation. The mouth of the fetus is open and its hand can be seen near it. Two seconds later the hand is moved away to touch the placenta, and the mouth is closed. f Fetus at 22 weeks’ gestation. The head is bent backwards. The hand is near the face and the thumb touches the mouth. However, no shaping of the mouth can be seen, and the thumb (not shown here) was moved away after 1 s. The head remained bent
56
5 Swallowing, Sucking, and Handedness as Inferred from Fetal Thumb Sucking
Proper thumb-sucking, so often mentioned as one of the ‘extraordinary’ capacities of the fetus, is not frequent. During a hand face or a hand mouth contact, fetuses may touch their open lips with one hand. The thumb can be inserted momentarily (1 2 s) into the mouth, but the mouth does not close to latch it, nor does the tongue cup around it, and the thumb consequently slips out. Fetuses do not display any attempt to recapture it (Fig. 5.6). One wonders whether proper, sustained and frequent thumb sucking can occur before sucking is established. Some authors have described the emergence of handedness in the fetus on the basis of thumb-sucking described to occur at 10 weeks [22 24]. Handedness, meaning why some individuals are lefthanded and most others right-handed, is still an unsolved problem. Handedness has been linked with brain lateralization, a fact discovered by the French neurosurgeon Paul Broca in the mid-1800s and confirmed shortly after by the German neurologist Carl Wernicke. Broca had a patient who suffered from aphasia and could only articulate a few words. When the patient died his autopsy revealed a profound lesion in the left frontal lobe of his brain which became known as Broca’s area and was recognized as an important region for speech articulation. Wernicke found that lesions of the left posterior temporal gyrus (hence called Wernicke’s area) caused an inability to understand rather than to utter language. Broca and Wernicke were mainly concerned with language, but handedness too has been linked with socalled hemisphere dominance, the dominance of one hemisphere over the other in regard to some functions, and with the possibility that our cerebral hemispheres have different specializations [25, 26]. Since Broca and Wernicke’s work, the two cerebral hemispheres, separated by a long fissure and connected by the so-called corpus callosum, have been largely regarded as performing different functions, including being responsible for handedness. As we have said, some authors have recently suggested that handedness can already be detected in the fetus on the basis of a judgement on an alleged preference for sucking the right or left thumb from 10 weeks. Those caring for premature infants well know that a firm grip on the nipple or the teat is laboriously achieved, and many premature infants have to be tube-fed. Leaving aside the doubtful capacity of the fetus to
Fig. 5.7 Beginnings of thumb sucking at 25 weeks’. Fetus at 25 weeks’ gestation. The thumb is partly inserted into the mouth, which shows some shaping. However, the tongue is moved side ways and does not capture the thumb. The fetus was in fact ob served licking the thumb a few seconds later
perform thumb-sucking at this early stage, the works in question did not specify the criteria used to distinguish the right from the left hand [22 24]. The only criterion one can use to distinguish the right from the left arm approaching the face is their position relative to the heart. Excluding dextrocardia (a rare condition in which the heart is located in the right sector of the chest), one can say that the left hand is the one on the side of the heart. However, simultaneous visualisation of one hand and the heart is rarely possible. It is even more difficult to observe both hands and the heart. Even assuming that the fetus was sucking the right thumb, it would be difficult to know what the other hand was doing. Possibly 4D ultrasonography may help us to understand how and when lateralization can be said to emerge. However, thumb-sucking, especially if considered as the main indicator for handedness before 25 weeks, may not be the ideal parameter for judging its emergence (Fig. 5.7).
References
References 1. Gupta D, Datta B (1988) The cultural and historical evolution of medicine and psychological ideas concerning conception and embryo development. In: Fedor Freybergh PG, Vanessa Vogel ML (eds) Prenatal and perinatal psychology and medicine. Parthenon Publishing Group, Park Ridge, pp 514 531 2. Piontelli A (2008) Twins in the world. MacMillan Palgrave, New York 3. Mizuno K, Ueda A (2003) The maturation and coordination of sucking, swallowing, and respiration in preterm infants. J Pediatr 142:36 40 4. Koenig JS, Davies MA, Thach BT (1990) Coordination of breathing, sucking, and swallowing during bottle feedings in human infants. J Appl Physiol 60:1623 1629 5. Lau C, Sheena HR, Schulman RJ, Schanler RJ (1997) Oral feeding in low birth weight infants. J Pediatr 130:561 569 6. Gewolb IH, Vice FL, Scweitzer Kenney EL et al (2001) De velopmental patterns of rhythmic suck and swallow in preterm infants. Dev Med Child Neurol 43:22 27 7. Dowling DeMonterice D (1999) Physiologic responses of preterm infants to breastfeeding and bottle feeding. Nurs Res 48:78 85 8. Lau C, Alagugurusamy R, Schanler RJ et al (2000) Character ization of the developmental stages of sucking in preterm infants during bottle feeding. Acta Paediatr 89:846 852 9. de Vries JIP, Visser GHA, Prechtl HFR (1984) Fetal motility in the first half of pregnancy. In: Prechtl HFR (ed) Continuity of neural functions from prenatal to postnatal life. Spastics In ternational Medical Publications, London, pp 46 64 10. Jirasek JE (2004) An atlas of human prenatal developmental mechanics: anatomy and staging. Informa Healthcare, Lon don 11. Miller JL, Sonies BC, Macedonia C (2002) Emergence of oropharyngeal, laryngeal and swallowing activity in the de veloping fetal upper aerodigestive tract: an ultrasound evaluation. Early Hum Dev 71:61 87 12. Richards DS, Farah LA (1994) Sonographic visualization of the fetal upper airway. Obstet Gynecol 4:21 23 13. Achiron R, Ben Arie A, Gabbay U et al (1997) Development of the fetal tongue between 14 and 26 weeks gestation: in
57 utero ultrasongraphic measurements. Obstet Gynecol 9:39 41 14. Riordan J, Wambach K (2009) Breastfeeding and human lac tation, 4th edn. Jones and Bartlett, Sudbury, Massachusetts 15. Brace RA (2001) Fluid balance. In: Harding R, Bocking AD (eds) Fetal growth and development. Cambridge University Press, Cambridge, UK 16. Rinaman L (2004) Postnatal development of central feeding circuits. In: Stricker E, Woods S (eds) Neurobiology of food and fluid intake. Handbook of behavioral neurobiology, vol. 14, 2nd edn. Plenum, New York 17. Trahair J (2001) Digestive system. In: Harding R, Bocking AD (eds) Fetal growth and development. Cambridge University Press, Cambridge, UK 18. Tucker Blackburn S (2007) Maternal, fetal, and neonatal phys iology: a clinical perspective. Saunders, St. Louis, Missouri 19. Abadie V, Champagnat J, Fortin G, Couly G (1999) Succion déglutition ventilation et genes du développement du tronc cerebral. Arch Pédiatr 6:1043 1047 20. da Costa SP, van den Engel Hoek L, Bos AF (2008) Sucking and swallowing in infants and diagnostic tools. J Perinatol 28:247 257 21. Barlow SM (2009) Central pattern generation involved in oral respiratory control for feeding in the term infant. Curr Opin Otolaryngol Head Neck Surg 17:187 193 22. Hepper PG, Shadidullah S, White R (1991) Handedness in the human fetus. Neuropsychologia 29:1107 1111 23. Hepper PG, McCartney GR, Shannon EA (1998) Lateralised behaviour in first trimester foetuses. Neuropsychologia 36:531 534 24. McCartney G, Hepper PG (19999) Development of lateralized behaviour in the human fetus from 12 to 27 weeks’ gestation. Dev Med Child Neurol 41:83 86 25. Wallesch CW, Herrman M, Bartels C ((2001) Wernicke’s cases of conduction aphasia. In: Code C (ed) Classic cases in neuropsychology II: Brain damage, behaviour, and cognition. Psychology Press, Philadelphia 26. Code C, Joanette Y (2001) The control of speech in the adult brain. The disconnected right hemispheres of PS, VP, and JW. In: Code C (ed) Classic cases in neuropsychology II: Brain damage, behaviour, and cognition. Psychology Press, Philadelphia
Localized Movements
6
With the assistance of Chiara Boschetto, Florinda Ceriani and Alessandra Kustermann
Keywords Hand movements • Arm movements • Myotubes • Hyperplasia • Hypertrophy • Corticalization • Hand-shaping • Bimanual prehension • Foot positioning • Athetosis • Scratching • Cranial nerves • Proprioception • Cerebral palsy
Localized movements, also called isolated, partial or ideokinetic movements, involve only segments of the fetal body. This definition is somewhat arbitrary as bodily movements are hardly ever isolated phenomena, but emerge from the confluence of multiple subsystems. The amplitude of isolated finger movements is relatively minute, and yet they set in motion minor adjustments spreading to other fingers, the wrist and the arm. Facial expressions may seem limited to a small area of the face, yet even an apparently simple frown or a smile involves the activation of a group of facial muscles and brings about slight positional changes of the head, the neck or the shoulders. Movements of the legs, trunk and spine or ample sways of the arms cause major changes in the balance of the body, which produce passive displacement of other parts of the fetal body and activate a re-balancing of the same and of other body parts. In the end almost the entire body is drawn in. In other words, patterns of behaviour can emerge spontaneously from the cooperation of multiple subsystems or components and are complex, varied, open to information flow, and constantly adapt to the current status of the performer and the task. Sensory feedback plays a pivotal role in the execution of the movement. Cutaneous, joint and muscle afferents respond to movements, influence perception and evoke reflex changes during movement [1 3]. As we have said, general movements are ontogenetically ‘older’ than localized ones. General movements start at around 7 weeks’ gestation whereas local A. Piontelli, Development of Normal Fetal Movements. © Springer Verlag Italia 2010
movements surface at about 10 weeks as various fetal parts have not yet reached full anatomical independence during this age span. At the beginning of the fetal period limb buds have developed into well-segmented upper limbs. the humerus, ulna and radius are differentiated, as are the hands. The fingers no longer look like palmipedal appendixes, but are well separated. Some ossification nuclei have begun to develop. Muscles have migrated to their place of destination and have started to form. At this stage muscles are called myotubes and are constituted by spindle-shaped cells with a single nucleus containing clusters of myofilaments. Actin and myosin, the two proteins indispensable for muscle contraction, have begun to be synthesized [4]. However, muscles, bones and all other bodily components will undergo many changes before and after birth. For example, schematically speaking, throughout the fetal stage each muscle fibre is poly-innervated, meaning innervated by more than one nerve, and only after birth will each muscle be innervated by just one nerve. Innervation undergoes other changes during gestation and beyond, involving increasing complexity and elongation of the nerve fibres to match the lengthening of the muscle spindles [4]. Muscle fibres grow by hyperplasia (an increase in the number of muscle cells) and hypertrophy (an increase in muscle cell size) during the fetal stage. After birth, muscle growth occurs overwhelmingly by hypertrophy. The type of muscle fibres also changes. Adult muscles are formed by two
60
main types of fibres: slow-twitch, adapted for endurance activities, and fast-twitch, suited to short-duration intense activity. At birth, fast-twitch units are prevalent, and slow-twitch ones develop only gradually [5]. Bone also undergoes remodelling throughout the life span. Old bone is regularly replaced by new bone, but the formation and resorption rate change profoundly with age. Bone composition also changes over the life span. The percentages of organic and inorganic components are very different in an old person compared to a child and even more so to a fetus. In the neonate, most ossification centres have not yet fused to form complete bones [5]. In other words, when talking of bones, muscles or any other tissues, components and functions in the fetus, one must be aware that we are often using the same terms to designate what may be profoundly different phenomena from the neonatal stage, and even more so from those of the adult. Generally when watching ultrasound live or analysing a DVD or a tape, differentiating between general and localized movements is done on subjective gestalt/perceptual grounds which come easily to the expert eye. However, localized movements can be distinguished from general movements on the basis of several non-subjective characteristics. The first is the rate of occurrence. To start with, general movements are more frequent than localized ones. However, localized movements rapidly and progressively increase, and by 14 weeks they outnumber general ones (Fig. 6.1). General movements, on the other hand, continue at an even rate almost until the end of gestation. Another feature is duration. On average, localized movements last a shorter time than general ones (3 10 s). Only towards the second half of pregnancy does the duration of some localized movements increase. From then on localized movements are of varying length: some continue to be short whilst others become longer (>20 s). A third characteristic is ‘compatibility’, which is composed of two distinct phenomena. The first implies simultaneity and the second chronological surfacing. Local movements can be performed simultaneously with all other movements and in this sense are ‘compatible’ with them. A fetus may be performing breathing movements while at the same time touching its face, or it can be swallowing and simultaneously rub its feet. Breathing movements, because they move the thorax downwards, frequently cause leg displacement and con-
6 Localized Movements
a
b
Fig. 6.1 Localized movements (LMs): developmental trends. a Lo calized movements start to be noticed at 10 weeks’ gestation but up until 12 weeks are very infrequent. At 13 weeks they increase rapidly. After this the increase is fairly steady, but not so pronounced. b Comparing localized movements with generalized ones (GMs), an opposite development trend can be noticed. Localized movements increase whilst generalized ones decrease almost at the same rate, reaching a crossover point at 13 weeks. The opposing tendencies are one of the indicators suggesting that these movements may have different origins, and may ultimately have different developmental outcomes
sequently also initiate short leg movements. Initially, general movements are not compatible with either swallowing or breathing. A slight overlap between breathing and general movements lasting only a few seconds (range 2 6 s) begins to be observed towards mid-pregnancy. However, up to 25 weeks this form of
6.1 Hand and Arm Movements
overlap does not increase. Unlike breathing movements, swallowing and general movements still emerge only separately up until 25 weeks. Chronological surfacing relates the fact that up to 20 22 weeks fetuses display three fairly distinct behavioural aggregates or clusters which emerge cyclically and separately in time. Up to mid-pregnancy, fetuses largely function in an either/or way. Specifically, a breathing and swallowing phase, a general movements phase and a rest phase can be distinguished. Clusters contain only particular functions, to the exclusion of others. Unlike general movements, local ones are compatible with and emerge in all clusters [6]. Behavioural aggregates will be examined in detail in Chapter 8. Finally, local movements appear to be ‘goal-directed’ or ‘targeted’. Goal-directed movements do not indicate the involvement of the cortex at these early stages. Corticalization will bring about proper intentionality and complex motions such as hand-shaping, proper bimanual prehension or foot positioning for walking [7]. In utero only very coarse hand-shaping or foot positioning can be noted before the ‘target’ is reached. However, rubbing the eyelids, scratching the head or performing alternating leg movements are all accurate, targeted motions. General movements, in contrast to localized movements, do not appear to be ‘goal-directed’, but look like storms of turbulent motion without a specific target. Thus, the distinction between localized and generalized movements seems to involve a divergence between two different kinds of movement, possibly distinct control, and a possibly different maturational outcome. Isolated head and spine motions are considered as localized. However, this chapter will focus on the movements of the appendicular body only, head and spine motions having been described in some detail in the discussion of general movements (see Chapter 1).
6.1 Hand and Arm Movements The hand is a most important organ; through it we perceive, explore and recognize our environment. Hands are also the main organ for acting on our environment and manipulating it with gross and fine motor skills. In addition, hands perform important expressive activity. Gestures are a major form of communication both in ordinary social life and in sign language. These activities are largely vision-guided after birth, but commu-
61
nicative purposes and vision-guided activities do not pertain to prenatal life [8]. Fetuses use hands mainly, if not exclusively, as perceptual tools. Manipulative skills are very limited and possibly of little use during prenatal life. Hence the focus here will be on tactile information received through the hands. The glabrous (smooth and hairless) surface of the palm and even more so the fingertips are densely innervated by sensory fibres. The fingertips can be said to be the richest source of tactile sensation of the entire human body. Before 10 weeks hands can hardly be said to perform localized movements. Up to 12 weeks all fingers are kept semi-flexed, as in the so-called ‘simian’ or ‘monkey’ hand found both in monkeys (hence the name) and in individuals with lesions of the median and ulnar nerves, two of the three main nerves innervating the hand. The thumb at this stage is kept in full opposition. The thumb does not share muscles with any other fingers, and its muscular apparatus and kinematic repertoire are different from those of other fingers, allowing amongst other things its opposition, as well as other important hand skills [9]. The opposition of the thumb is no longer regarded as uniquely human, as Darwin and many of his followers believed. Other animals with opposable thumbs have been discovered. What makes the human hand unique is ‘ulnar opposition’: the capacity of the small and ring fingers to rotate across the palm to meet the thumb, which adds matchless capabilities to the human hand [10, 11]. During early fetal life, save for the thumb, which initially hardly participates in any movement at all, all fingers move synchronously [12]. In later life, lesions of the primary motor cortex or of the corticospinal tract, which connects the motor cortex with the spinal cord and is involved predominantly in finely coordinated voluntary movements, can result in a similar inability to move one finger at a time [12]. Both the motor cortex and the corticospinal tract have yet to be formed when finger movements begin. Up to mid-pregnancy and beyond, fetal motions are overwhelmingly guided by subcortical circuits. In addition, at this stage flexor muscles prevail over the extensors which allow fingers to be straightened out. The semi-flexed ‘simian’ position of the digits could be explained by this. Around 13 to 14 weeks all fingers, including the thumb, begin to be spread, but they still tend to act synchronously (Fig. 6.2). Separate finger motions of a particular kind are noted
62
6 Localized Movements
a
b
c
Fig. 6.2 Finger movements. a Fetus at 13 weeks’ gestation. The fingers are kept semi flexed and the thumb is slightly bent, but no longer in full opposition. b Fetus at 13 weeks’ gestation. The hands touch each other. The fingers are all semi flexed and kept in the same position. At this stage fingers do not act independently. c Fetus at 15 weeks’ gestation. The fingers are no longer kept semi flexed, but can spread out
from 15/16 weeks. Fetuses making the V for victory sign or even some ‘obscene’ gesture with their middle finger often make the headlines. Up until 25 weeks most of these individual finger movements can be seen during cycles of rest and have an ‘athetoid’ quality. Fingers move slowly, sinuously and in a worm-like way, like the tentacles of an octopus. Individual fingers extend, bend and spread, apparently with very little vigour, and the individual position can be maintained for some length of time. After birth, athetosis (from the Greek word athetos meaning non-fixed) results from various injuries of the brain ranging from lesions of the basal ganglia a group of interconnected nuclei in the upper brainstem and deep in the cerebral hemispheres involved in some movement disorders, but whose function is not yet fully clarified to a range of forms of encephalopathy, and cerebral haemorrhages. None of these lesions is present in the normal fetus. Transient fetal ‘athetosis’ could be due to various factors acting in combination or alone, such as motion in a fluid, decreased muscular tone during a cycle of rest, or immaturity of some structures of the central nervous system. Since perinatal insults are often mentioned as one of the main causes of postnatal athetosis, it may be worth investigating when movements of this kind normally cease, and to note deviations from the norm from then on. Athetoid finger motions are still frequent at 25 weeks (Fig. 6.3). Up until 20 22 weeks both hands act independently. Only after 23 weeks can hands be seen to touch each other seemingly purposefully. However, proper bimanual cooperation, prehension, grasping and holding cannot be observed nor, for that matter, are single-hand prehension, grasping or holding present. All these acts involve functioning cortical areas and kinematic and prehensile synergies [12] (Fig. 6.4). Newborn arm movements are asymmetrical and random. The first bilateral motions are the extension and raising of the arms, observed at approximately 2 months of age, and only at approximately 4.5 months do infants start reaching for objects with both arms [13, 14]. Fetal use of hands for ‘playing’ or ‘exploring’ the environment, and even more so for manipulating it, are frequently mentioned especially by popular books, manuals and magazines in a sensational tone, but cannot be observed. Hands do, however, fulfil important ‘exploratory’ and preparatory functions during prenatal life, being used principally for eliciting and receiving sensations which may be essential for fetal development in several ways.
6.1 Hand and Arm Movements
63
a
b
c
d
Fig. 6.3 Athetoid finger motions. a Fetus at 15 weeks’ gestation. Athetoid finger motions begin to be noticed. Two fingers are spread out, looking like small tentacles. The time dimension cannot be captured by a single picture, but the fingers were kept in the same position for about 10 s. b, c Fetuses at 18 weeks’ gestation. Two very popular finger motions: V for victory and one finger sticking out as in an obscene gesture. In fact neither gesture has a meaning or is intentional; they simply show the athetoid quality of finger movements seen during cycles of rest during the 16 25 weeks age span. d Fetus at 20 weeks’ gestation. Another athetoid ‘tentacle like’ finger movement
One of the first, frequent hand movements fetuses perform is scratching the temples or the parietal region of their heads with their fingers moving concomitantly, as monkeys do (Fig. 6.5). Scratching is usually associated with the removal of an annoying sensation, primarily itchiness. Recently, an itch gene has even been identified. An occasional itch may well be experienced by the fetus. However, it would be difficult to postulate that all fetuses not only itch, but also itch frequently. In everyday bodily language, scratching the parietal
region of the head even symbolically indicates puzzlement, embarrassment or having a problem which gives us a headache. This range of emotions can hardly be ascribed to fetuses. Disturbed adults trying to emerge from a feeling of emotional numbness scratch themselves until they bleed in order to ‘feel alive’ or ‘feel something’. Unlike adults, fetuses start scratching their heads when their nails have not begun to emerge from their nail beds. However, even without nails, the simple act of scratching elicits fairly strong sensations.
64
6 Localized Movements
a
c
Interestingly, the cranium is rarely scratched. The cranium is barely innervated by sensory fibres, granting it a kind of natural anaesthesia when contractions during parturition push and squeeze it through the birth canal. The fetal skull includes the so-called fontanelles (from the Latin for small fountains): soft, membranous, nonsensitive areas which allow some flexibility of the skull and its bony parts during the delivery. Additionally, the fontanelles allow pre- and postnatal enlargement of the brain; they seal at about 18 months after birth. Early sealing of the fontanelles, or craniosynostosis, can lead to severe problems ranging from visual impairment to breathing problems. On the other hand, bulging
b
Fig. 6.4 Bimanual activities, not bimanual cooperation. a, b Fetuses at 18 and 16 weeks’ gestation respectively. In both pictures the hands perform a similar movement: in a the hands are kept in a fist under the chin, in b both fists cover the eyes. Despite the similar attitude, one cannot talk of bimanual cooperation, which refers only to actions accomplished through the joint synergy and collaborative effort of both hands, as in bimanual prehension. c Fetus at 23 weeks’ gestation. The hands touch and overlap. These may possibly be the precursors of bimanual cooperation, which only begins several months after birth and is not perfected until at least school age
fontanelles and an enlarged cranium indicate an undue accumulation of cerebrospinal fluid (hydrocephalus). In contrast to these insensitive areas, the posterior part of the head (the occiput) and the nucha (the nape), which are not insensitive, are scratched (Fig. 6.6). Other hand movements indicate that sensation-eliciting activities may be important for the fetus. Fetuses engage frequently in so-called hand face contacts, i.e. they repeatedly and protractedly touch their face. Recently, the obstetrician Asim Kurjak and his collaborators, using 4D ultrasound, have divided hand face contacts into several subcategories: hand to head, hand to mouth, hand near mouth, hand to face, hand near face,
6.1 Hand and Arm Movements
65
a
b
c
d
Fig. 6.5 Scratching. a d Fetuses at 10, 11, 12, and 13 weeks’ gestation respectively. All are scratching their head or forehead, but none is scratching the cranium, which is barely innervated by sensory fibres. The fetus in c can be seen to limit its scratching efforts at the border with the fontanelles (in black) k
hand to eye, and hand to ear motions [15]. However, these subcategories give little insight into why fetuses engage in these activities. The cranial nerves are amongst the first to develop and to reach their target. The trigeminal or fifth cranial nerve, infamous for causing trigeminal neuralgia, is particularly relevant for the sensitivity of the face. The trigeminal is a mixed nerve with a predominant sensory component that innervates most of the cutaneous surface of the face through its three major branches. The trigeminal is the main source of sensations (tactile, pro-
prioceptive and nociceptive or pain-sensing) of the face and mouth. Besides scratching their forehead (an area reached by the ophthalmic branch of the trigeminal nerve), from about 12 weeks fetuses frequently touch their face, pressing their fists just in front of the crus elix, the part of the external ear located in front of the acoustic meatus on the temporal bone. In doing so they look like Edvard Munch’s ‘scream’ portraits of individuals expressing angst by screaming and pushing their fists onto their ears. Angst cannot be attributed to fetuses; however,
66
Fig. 6.6 Touching the occiput. Fetus at 15 weeks’ gestation. This fetus is touching the region between the parietal bones and the oc ciput. None of these areas are innervated by the trigeminal or fifth cranial nerve. Parietal bones in particular are barely innervated by sensory fibres at this stage, in order to favour painless squeezing in the passage through the birth canal. The occiput, although scant ily innervated by sensory fibres compared to the face, is less in sensitive than the parietal region. Possibly this fetus (like the one in Fig. 6.5c) is trying out different sensations, one stronger, the other weaker, arising from differently innervated areas of the skull
this gesture seems particularly apt to elicit strong sensations, including possibly a tinge of mild pain. This area is also innervated by the trigeminal nerve and its branches (Fig. 6.7). After 13 weeks, hand face contacts consist predominantly in touching the face and the mouth with the palms and the back of the hand. The mouth is also frequently rubbed. Only between 3 and 4 months do infants become more consistent at bringing the hand to the mouth rather than to other parts of the face, and at 5 months they begin to open the mouth in anticipation of the ‘arrival’ [16, 17]. None of this can be observed in the fetus. The glabrous or hairless surface of the hand is innervated by copious sensory fibres, the back of the hand much less so. We all tend to touch predominantly our faces, but also other surfaces and objects, with the palm rather than with the back of our hands. In contrast to postnatal life, fetuses use both sides of their hands for touching their face. By doing so they may elicit varied sensations, the back of the hand being harder than the palm (Fig. 6.8). The soles of the feet are another glabrous area of the skin which, like all glabrous areas, are especially
6 Localized Movements
a
b
c
Fig. 6.7 a c Fetus at 18 weeks’ gestation. In all pictures the same fetus can be seen pressing its fingers against the ears and the parietal bone. The slight frowning of the forehead and mouth add to the impression of ‘angst’ and to the likeness to Munch’s portraits
6.1 Hand and Arm Movements
67
a
b
c
Fig. 6.8 Touching the face with the back of the hands. a c Fetuses at 16, 18 and 24 weeks’ gestation. All can be seen touching their face with the back of the hand. Fetuses also touch their face with the palm of their hands: both ways of touching are used almost equally. This gesture is slightly unusual in life after birth, as we all tend to touch our face more with the palm rather than the back of the hand. Possibly in doing this fetuses are trying out different sen sations, the palm being softer than the back of the hand
sensitive to tactile stimuli. Fetal eyes are barely touched before eye motions begin to be noticed occasionally at 16 18 weeks. From then on fists are frequently pressed against the orbits. Fetal eyes are closed at this stage as the palpebrae are still fused. They will not be able to open before 23 24 weeks. Even then eye opening will continue to be an exceptional, very brief occurrence until approaching birth. On the other hand, retinal development is fairly well advanced by mid-gestation. By pressing their fists forcefully against their eyes, fetuses may exert pressure on the photoreceptors of the retina eliciting those flashes of light called ‘pressure phosphenes’ (from the Greek phos meaning light and phainein meaning to show). This does not mean that by 16 18 weeks fetal vision is ready to meet the requirements of the external world. However, one could postulate that phosphenes may be our very first experience of light (Fig. 6.9). Fetuses rarely stroke other areas of their body less sensitive to touch before 20 25 weeks. The abdomen, for instance, is rarely touched before then, as are the thorax, the buttocks and the back. By contrast, when erections begin to be noticed, by 18 20 weeks, male fetuses occasionally direct their hands over their penises. However, the study of penile erections is fraught with difficulties. Unless male fetuses keep their legs well open one cannot tell what is going on down there. Until the end of pregnancy arms are longer than legs. Long arms, besides other possible functions, allow the reaching of bodily parts which would otherwise be difficult to touch. Before 20 weeks, fetuses often touch the soles of their feet without having to bend forward to their limit as we all have to do during gymnastics. After 20 weeks, all the above-mentioned areas of the body also begin to be reached by the hands and touched. Many apes have long arms and use them for prehension or climbing trees. However, the long arms of the human fetus do not seem to have the same function. Only too often the dictum ‘ontogeny recapitulates phylogeny’ continues to be stretched to fit many features of the human fetus. Adherents to the evolutionary theory claim that remnants of our animal past can be found in some of the so-called primitive reflexes. Primitive reflexes (also called infantile, newborn or infant reflexes) are exhibited by normal infants in response to particular stimuli and disappear or are inhibited within the first few months of postnatal life when motility begins to
68
6 Localized Movements
a
b
c
d
Fig. 6.9 Eliciting pressure phosphenes? Fetuses at 23 (a), 25 (b), and 20 weeks’ gestation (c, d). All can be seen pressing a fist against one eye. By pressing fists against our eyes we can elicit particular flashes of light called pressure phosphenes. Could fetuses do the same? If so, this would be their very first sensation of light
be guided by cortical control. These reflexes can reappear in neurologically impaired adults. As for arms and hands, the remnants of our simian past are possibly to be found in the so-called palmar grasp reflex. The palmar grasp reflex is the name given to the phenomenon by which, when an object is placed in the infant’s hand and touches its palm, the fingers close and grasp it. This reflex would allow infant monkeys to grasp and hold on tightly to their mother’s fur
when in danger of falling down during her often boisterous wanderings in the jungle. One often hears of fetuses ‘playing’ with the umbilical cord, but up to 25 weeks, and beyond, fetuses do not reach out actively to grasp the cord, for which the acquisition of a firm grip is needed. Neonates begin to grasp objects at about 1 month of age. In early grasping the infant holds an object against the palm without the thumb providing opposition. Eventually the infant uses the thumb in op-
6.1 Hand and Arm Movements
position but still holds the object against the palm. Both types of grips are called ‘power grips’. They display the so-called ‘precision grip’ only after 9 months of age, holding the object between the thumb and one or more fingers [18]. As the uterus becomes more crowded, the cord is almost inevitably in contact with and coiled around many parts of the fetal body including the face, the hands and the neck. The cord is very elastic and its blood vessels (normally two arteries and one vein) are covered by a thick layer of a slippery and flexible substance called Wharton’s jelly which favours quick disentanglement and impedes tightening. When the cord accidentally touches the palm forcefully, the fingers close clenching into a fist and grasp it. However, initially the grip only lasts for a few seconds (range 4 6 s). This reaction, which begins to be displayed between 24 and 25 weeks, could be considered as the prenatal emergence of the palmar grasp reflex (Fig. 6.10). When the most relevant parts of the fetal body have been repeatedly and thoroughly touched by its long arms, the fetus begins to actively and consistently reach out for and touch its environment. So-called ‘explorations’ of the placenta and proper touching of the cord are exceptionally performed before 16 18 weeks. Even then fetuses continue to be more prone to touch themselves (Fig. 6.11). Neonates are frequently born with scratches on their face, and mothers often worry about these. The preoccupation with having to cut neonatal nails to avoid scratching of the face is very widespread. Nails have grown by birth, and scratches indicate that fetuses approaching birth and neonates continue to touch their face, testifying to the importance of tactile stimulation for both the fetus and the neonate.
69
Apart from active touching with hands, general movements provide a wealth of contacts even on the most inaccessible parts of the body and of the uterine environment. Loss of sensitivity to touch is considered agonizing and worse than deafness or blindness by those who suffer from it. Numbness in the sense of absence of tactile stimulation may be detrimental for the fetus. One could postulate that afferent bodily sensations elicited by touch are important building blocks for the development of the central nervous system and for the beginnings of a kind of somatosensory demarcation, an unconscious proto-sense of one’s sense of body. Jean Piaget’s description of the initial phases of the sensorimotor stage, which he applied to the neonate, seem more fitting for the fetal stage than for the neonatal one. As described by Piaget, movements are initially non-targeted, as general movements are for the fetus. Gradually, however, the first ‘motor schemata’, albeit totally non-consciously, begin to function and tend to be assimilated. The subject tends to generate already performed acts which give it particular sensations. In the case of the fetus these would correspond to localized movements. Each fetus, within the limits imposed by its capabilities, has its own ‘style’, and thus begins to form a ‘personal’ sense of body. As Shaun Gallagher pointed out, body schema should not be confused with body image. ‘Body image’ is ‘an internal representation in the conscious experience of visual, tactile and motor information of corporal origin’ and implies a conscious awareness of one’s body. ‘Body schema’ implies a non-conscious performance of the body against which all subsequent changes of posture are measured [19]. And as Paillard said, ‘A
Fig. 6.10 Beginnings of the palmar grasp reflex. Fetus at 24 weeks’ gestation. The palmar grasp reflex is one of the so called ‘primitive reflexes’ present at birth and persists until 5 or 6 months of age. When an object is placed in the infant’s hand, striking the palm, the fingers close and grasp it. The grip is strong and may be able to support the infant’s weight. The fetus in its now fairly cramped en vironment often touches or is touched by the umbilical cord. How ever, only when the umbilical cord touches the palm do its fingers close and grasp it. A tentative grasp can be seen in this picture. The cord has touched the palm and the fingers semi close around it
70
6 Localized Movements
a
d
b
e
c
f
Fig. 6.11 Touching the environment. Fetuses at 16 (a, b), 18 (c) and 22 (d ff) weeks’ gestation. All fetuses are seen touching the wall of the uterus and the placenta. Touching seems less tentative and more ‘evolved’ in the fetus in d f. Hand positioning is noticed (d), both hands are used (e, ff) and the entire sequence lasts approximately 1 min. The contacts in a c from the 16 weeks fetus lasted only a few seconds
6.2 Leg Movements
71
distinction between body image and body schema is to distinguish between a conscious awareness of one’s body and a nonconscious performance of the body.’ According to Paillard, the non-conscious sensory mode of processing of body schema is more primitive in the evolution of the human brain than the more conscious representational mode of processing of body image [20]. Rossetti and his collaborators suggest that an evolutionarily later ‘semantic’ system is able to override the earlier non-conscious pragmatic system underlying the body schema [21, 22]. A body schema is progressively built through movement during prenatal life (Fig. 6.12). a
6.2 Leg Movements Strictly, the term ‘leg’ should denote the portion of the lower limb between the knee and the ankle. However, in common language the word leg is generally used for the entire lower limb. For reasons of simplicity, unless specified, this term will be used in its common sense. As is the case for arms and hands, at the beginning of the fetal period well-segmented lower limbs have formed, with differentiated femur, tibia and fibula. The feet are also differentiated and toes have separated, but most features, from bone and muscle formation to innervation, the relative proportions of the legs and many more, will undergo numerous changes that will not be completed until well after birth. In the human, whilst the arms and hands are principally specialized for manipulation, the legs and feet are adapted for bipedal locomotion. Besides walking, the legs are used for other bipedal movements such as standing, jumping, running and hopping. The feet support the weight of the body and allow all these movements. During fetal life the legs are widely used for movement, but clearly not for locomotion, although steps towards locomotion begin to be paved. Leg movements start at 10 11 weeks (Fig. 6.13). At 10 11 weeks conjugated leg movements are largely prevalent, as legs move predominantly together. However, some alternate movements can be noticed even at this initial stage. Conjugated leg movements are executed with greater speed (1 2 s) than alternate ones, which on average are executed in 5 6 s. Full extension of the legs, as seen in the decerebrate posture, and full flexion with the soles of the feet in contact with each other are often observed. In the 10 11 weeks time span
b
c
Fig. 6.12 Developing a proto sense or a body schema? Fetuses at 22 (a), 18 (b) and 20 (c) weeks’ gestation. These pictures all illus trate the importance of touch, and especially of touching particularly sensitive areas of the body such as the face. It is possible that, in addition to the sensory feedback received through movements, all this touching is one of the building blocks through which fetuses begin to form a proto sense or a body schema
72
6 Localized Movements
a
b
d
e
Fig. 6.13 Leg movements. In all these drawings the fetal body is represented schematically and not necessarily corresponding to the real alignment between the body and the legs. When the legs are focused on, the ultrasonographic picture of the rest of the body is most often lost or blurred. For this reason, the legs should be the main consideration in these pictures. a Fetuses display fairly ‘decerebrated’ postures. The uterus (represented as the outline of a quarter of a circle) impinges little on the legs, which mainly float or in any case are little in contact with it. At 12 weeks the wall of the uterus starts to become a relevant point of contact. b Some decerebrated postures can still be seen. However, independent leg movements have become relevant, as is leg folding and extension. The wall of the uterus is nearly always touched, representing an important pivot and plane of support. c Alternate movements have begun, and the range of movements shows greatly increased variation. The uterus is beginning to offer two points of support to the fetal body. d The fetal legs are almost invariably in contact with the uterus. Alternate stepping can be executed in a range of ways, and ‘decerebrated’ postures are no longer noted. e The space has now become cramped. The legs are rarely extended, and one or both feet are almost invariably in contact with the uterine w ll. Amongst other things this favours a great variety of stances. Alternate movements are largely prevalent
6.2 Leg Movements
the variety of leg movements is also limited. Most fetuses perform similar, rather monotonous though not identical movements. The feet are rarely in contact with the uterine wall. From 12 to 13 weeks conjugated leg movements no longer prevail, and alternate movements acquire speed (3 4 s) and variety. Full extension of the legs is no longer noted. Legs also flex less, and when flexed the soles of the feet are no longer in contact with each other. Feet are predominantly in contact with the uterine wall. By 14 15 weeks alternation occurs rapidly (2 3 s), and alternate stepping movements pushing the feet against the uterine wall are clearly observed. Equally, cycling movements begin to be seen. At 16 18 weeks alternate stepping movements become even more rapid (1 s) (Fig. 6.14). Full extension of the legs over the fetal body with a minimal popliteal angle (>60°) is observed and continues to be so until 23 25 weeks. Heel-to-head contact, which could be noted previously, is no longer observed. Both parameters are indicative of muscular tone and are evaluated in severely premature and premature infants to determine their gestational age as well as being signs of prognostic value. From 18 to 25 weeks leg movements are carried out with increasing speed, variety and skill. However, environmental constrictions are also increasing and flexion largely prevails over extension. The feet in particular can dorsiflex to the limit until they touch the leg. So-called ‘primitive reflexes’ involving leg movements which are noticed at birth in the healthy neonate have a long prenatal history. Stepping motions, as seen in the ‘stepping reflex’, are frequently performed. When stepping, fetuses clearly are not supported by an adult. However, within the uterus they execute stepping motions with environmental support, which is sometimes minimal. Stepping movements occur mostly when fetuses are lying on their backs. However, support can also be found the other way round when they lay their torso on some uterine protuberance as they would on a chair or on their mother’s lap after birth. From 23 25 weeks the forearms alone can sustain stepping motions. Besides these general norms, the intrauterine watery medium allows postures and movements which are not to be seen for quite a while in the neonate and the child living under the full impact of gravity after birth. Albeit for a very brief lapse of time (a few seconds), fetuses can perform ‘mature’ movements or take on ‘mature’ postures. They can sit unsupported with their heads
73
a
b
Fig. 6.14 a Fetus at 13 weeks’ gestation. The first alternate leg movements can be noted. b Fetus at 15 weeks’ gestation. Using the head as a support, the fetus performs some semi standing al ternate stepping motions. Besides neurological and extensive bodily changes, the watery medium in which the movements are performed seems to have a big impact on the variety and range of the movements
raised, they can hop on one leg once or twice, and they can attain an almost erect stance with their knees slightly bent. However, this does not mean that fetuses are more ‘mature’ than neonates, as is frequently implied when the ‘miraculous’, ‘wondrous’ nature of their activities is mentioned and illustrated in glossy images. Simply, the watery medium allows them to execute and prepare for activities which will be carried out in a steady and sustained way in life after birth. Aquatherapy, also called water or swimming therapy, or hydrotherapy, is increasingly used to help patients suffering from infantile cerebral palsy. As defined by Dorland’s Medical Dictionary, the term ‘cerebral palsy’ comprises a group of motor disorders characterized by ‘delayed or abnormal motor development’ [23]. These
74
motor problems, besides affecting the autonomy of the patient, can also cause serious muscular damage due to spasticity (or stiffness) in the muscles. When immersed in water, patients are less affected in their movement by their hypertonus or increased muscular tone, improve their coordination and endurance, and can perform movements such as walking, kicking and jumping which they are not able to carry out once back in their normal environment. Although cerebral palsy is predominantly caused by brain damage suffered during delivery and, especially, during intrauterine life, the overwhelming majority of fetuses do not suffer from cerebral palsy and spasticity. Like spastic children, however, within the amniotic fluid fetuses can perform, albeit momentarily, movements that they could not carry out in a terrestrial environment. Premature infants born at the same gestational age are totally unable to move in the same way. Severely premature infants barely move at all. Fetal leg ‘exercise’, easily carried out in the amniotic fluid, is vital for muscular and bone development, as is shown by the contractures and skeletal deformities found in fetuses suffering from prolonged and severe oligohydramnios (from the Greek words oligos meaning scarce and hudor meaning water). Club feet are just one of these deformities. Severe cases of oligohydramnios may be treated with amnio-infusion, the introduction of particular solutions into the amniotic sac to prevent deformities as well as other complications. Besides muscle and bone development, and a preparatory function for movements to be performed after birth, leg movements may have other important roles. Localized movements do not solely elicit tactile sensations, they are also an important source of proprioceptive information (from the Latin word proprius meaning one’s own, and perception) the information that we receive about the position, balance and movement of our body or any of its parts, the relative position of the body parts to each other, the position of the body in space, and the nature of the objects with which the body comes in contact. Proprioception allows us to adjust our stance constantly and reflexively to ongoing movements (Fig. 6.15). The terms ‘proprioception’ and ‘kinaesthesia’ are often used interchangeably, and the definition of and distinction between these terms are hotly debated. As a term, ‘kinaesthesia’ (or movement sense) places a greater emphasis on motion and relates to the capacity to locate the body and its related segments in space by
6 Localized Movements
a
b
c
Fig. 6.15 Legs as a source of proprioceptive and tactile feedback. a c Fetus at 20 weeks’ gestation. The fetus is observed in different planes. The feet are clearly an important source of tactile feed back. However, each single movement (and not only of the legs) is also a source of proprioceptive feedback. Both tactile and pro prioceptive feedback could be considered the initial ground on which a proto self begins to be based
sensing movement, weight and position. The term ‘proprioception’, on the other hand, refers mainly to a feedback mechanism. When the body moves or is moved,
References
information about this motion is returned to the central nervous system, which makes continuous adjustments in the movements. To complicate matters further, the term ‘haptic’ is also frequently used. ‘Haptic’ relates to the sense of touch in all its forms. For reasons of simplicity, the more commonly used terms ‘proprioceptive’ and ‘proprioception’ are used in this book. Legs movements, like all movements, are an important source of proprioceptive feedback information. Proprioceptive information is derived from so-called proprioceptors, sensory receptors located in the muscles, tendons and joints, and integrated from information arising from vestibular receptors, as well as from visual, auditory and tactile receptors. Visual and auditory information are clearly absent in the period considered in this work. Vestibular feedback, on the other hand, is most likely to be functional. Vestibular feedback pertains to the perception of balance, head position, acceleration and deceleration. Vestibular information is obtained from the semicircular canals in the inner ear. However, since vestibular perception includes the perception of gravity, fetal vestibular perception is likely to be different to that of the postnatal stage and adapted to the different requirements and environmental conditions characterizing fetal life. Sensorimotor demarcation and a proto-sense of self are possibly built progressively during prenatal life through tactile and proprioceptive information as well as vestibular feedback. This point will be explained in detail in the concluding chapter.
References 1. Ulrich BD (1997) Dynamic systems theory and skill development in infants and children. In: Connolly KJ, Forssberg H (eds) Neurophysiology and neuropsychology of motor development. Mac Keith Press, London 2. Bizzi E, Hogan N, Mussa Ivaldi FA, Giszter S (1994) Does the nervous system use equilibrium point control to guide single and multiple joint movements? In: Cordo P, Hranad S (eds) Movement control. Cambridge University Press, New York 3. Gandevia SC, Burke D (1994) Does the nervous system depend on kinaesthetic information to control natural limb movements? In: Cordo P, Hranad S (eds) Movement control. Cambridge University Press, New York 4. Jirasek JE, Keith LG (2001) An atlas of the human embryo
75 and fetus. Informa Healthcare, London 5. Haywood KM, Getchell N (2001) Life span motor development, chap 4: Development and aging of the body systems. Human Kinetics, Champaign, Illinois, pp 63 82 6. Piontelli A (2006) On the onset of human fetal behavior. In: Mancia M (ed) Psychoanalysis and neuroscience, chap 15. Springer, Milano Berlin Heidelberg, pp 415 442 7. Haywood KM, Getchell N (2001) Life span motor development, chap 5: Early motor development. Human Kinetics, Champaign, Illinois, pp 85 100 8. Lederman SJ, Klatzy RL (1998) The hand as a perceptual system. In: Connolly KJ (ed) The psychobiology of the hand. Chapter 2, Mac Keith Press, London, pp 16 35 9. Newell KM, McDonald PV (1997) The development of grip patterns in infancy. In: Connolly KJ (ed) The psychobiology of the hand, chap 12. Mac Keith Press, London, pp 232 256 10. Napier J, Tuttle RH (1993) Hands. Princeton University Press, Princeton, New Jersey 11. Jones LA, Lederman SJ (2006) Human hand function. Oxford University Press, Oxford 12. Latash ML (2008) Neurophysiological basis of movement, 2nd edn, chap 24: Prehension. Human Kinetics, Champaign, Illinois, pp 241 248 13. Wilson FR (1999) The hand. Vintage Books, London 14. Rocaht P (1993) Hand mouth coordination in the newborn: morphology, determinants, and early development of a basic act. In: Savelsbergh GJP (ed) The development of coordination in infancy, chap 10. Elsevier, Amsterdam, pp 265 288 15. Kurjak A, Azumedi G (2003) Fetal hand movements and facial expressions in normal pregnancy studied by four dimensional sonography. J Perinat Med 31:496 508 16. Dodwell PC, Muir DW, DiFranco D (1976) Responses of infants to visually presented objects. Science 194:209 211 17. Bard C, Fleury M, Gagnon M (1990) Coincidence anticipation timing. An age related perspective. In: Brad C, Fleury M, Hay L (eds) Development of eye hand coordination across the life span. University of South Carolina Press, Columbia, pp 283 305 18. Butterworth G, Verweij E, Hopkin B (1997) The development of prehension in infants. Br J Dev Psychol 15:223 236 19. Gallagher S (2005) How the body shapes the mind. Oxford University Press, Oxford 20. Paillard J (1999) Body schema and body image: a double dis sociation in deafferentiated patients. In: Gantchev GN, Mori S, Massions J (eds) Motor control today and tomorrow. Akademico Izdatelstvo, Sofia 21. Rossetti YG, Rode G, Boisson D (1995) Implicit processing of somaesthetic information: a dissociation between where and how? Neuroreport 6:506 510 22. Mishara AL (2005) Body self and its narrative representation in schizophrenia. In: De Preester H, Knockaert V (eds) Body image and body schema. John Benjamins, Amsterdam 23. Dorland’s Medical Dictionary (2007) 31st edn. Saunders, Philadelphia
Facial Expressions
7
With the assistance of Florinda Ceriani, Roberto Fogliani and Alessandra Kustermann
Keywords Physiognomy • Vocalizations • Anger • Disgust • Fear • Happiness • Sadness • Surprise • Tongue protrusion • Cross-modal integration • Mirror neurons • Autism
The introduction of 4D ultrasonography has made it possible to study the ontogeny and development of facial expressions in utero, albeit with some limitations. Interest in facial expressions dates back to antiquity and for centuries had to do with reading facial morphology or so-called ‘physiognomy’ (from the Greek words physys meaning nature and gnosis meaning knowledge). Aristotle wrote the first treatise on physiognomy in which he compared physiognomic traits of various people with animals. Such traits were assumed to indicate various tendencies ranging from stupidity to courage. Still today we say that so-and-so has an equine or horse-like face or an aquiline nose (from the Latin aquila meaning eagle). Pythagoras, Hippocrates and Galen used physiognomy as an important medical investigative tool for diagnosing pathology (from the Greek pathos meaning to suffer or display an emotion, and the Latin patior indicating suffering and affliction). If someone had a bilious face or was a lymphatic type, it was considered to indicate a possible weakness of the underlying apparatuses. In the Middle Ages facial morphology shifted from providing clues about temperament to indicating fate. We still say that someone has a ‘ominous face’ [1, 2]. In the eighteenth century, under the influence of a Swiss protestant pastor, Johannes Caspar Lavater, people started looking for traces of sanctity and marks of God in the eyes and creases and lines of the face. Cesare Lombroso (1835 1909), an Italian jurist and physician and the founder of ‘criminal anthropology’, introduced yet another change by A. Piontelli, Development of Normal Fetal Movements. © Springer Verlag Italia 2010
looking for criminal tendencies in those unfortunate individuals who carried marks of having never progressed from animals or from ‘savages’ [1]. The first to fully introduce emotions in face-reading was the French neurologist Guillaume-Benjamin Duchenne de Boulogne (1806 1875) who, in taking up Galvani’s studies on muscular contraction applying electrical stimuli to frogs’ limbs, applied small electric shocks to the facial muscles of his patients to understand and demonstrate the functioning of facial expressions. His distinction between a felt and a false smile was particularly important for later studies. The differentiation between felt versus false facial expressions has been the subject of innumerable researches and is still hotly debated. According to Duchenne, the ‘felt’ smile is accompanied by ‘wincing’ the lateral contraction of a facial muscle called orbicularis oculi, which regulates the closure of the eyelids. A portion of the muscle is not controlled by volition. Its involuntary contraction would thus reveal the spontaneity or otherwise of a socalled ‘Duchenne’ or ‘non-Duchenne’ smile [3]. Although not universally accepted, this distinction is still widely used. Duchenne’s work had a great influence on Darwin, who is universally credited with having regarded facial motion as displays of emotions in his work ‘Expression of the emotions in man and animals’. The word ‘expression’ is derived from the Latin exprimere meaning representing and showing with clarity. Darwin started observing facial expressions in animals and in the hu-
78
7 Facial Expressions
man newborn, namely his cousin’s son and his own children. Darwin wrote that facial expressions are universal, not learnt, but biologically determined, and are the product of man’s evolution. Darwin also claimed that there are specific inborn emotions which are expressed even in the neonate and serve a basic survival function. However, Darwin did not regard expressions as having a communicative function, nor were facial expressions of particular interest to him. Following Darwin’s work the study of facial expressions has become a central theme in many disciplines ranging from psychology to genetics [4].
7.1 Basic Emotions The communicative and emotional sides of facial expressions, originally neglected by Darwin, have become particularly hot topics. Especially relevant is the work of Ekman and Friesen, who in 1971 distinguished six
primary emotions each accompanied by a distinct facial expression. These so-called basic emotions are anger, disgust, fear, happiness, sadness and surprise [5] (Fig. 7.1). Although mixed emotions and correlated mixed expressions are increasingly being recognized, Ekman and Friesen’s classification is still widely used. Currently debates and research centre on many topics. Those relevant to the fetus are the innateness of facial expressions, their universality, communicative function, and changeability through experience [6 10]. To solve these and other basic questions, scientists have been searching for nocturnal non-vocalizing primates, or for blind and possibly deaf neonates. Grimacing without the possibility of using vision and audition would imply a genetically programmed innate reflex motion, without any imitational or learnt component. So far nobody has been able to find these ‘experiments in nature [6]. Yet the human fetus studied longitudinally could be such an ‘ideal’ creature. Audition and, especially, vision begin to function late in the fetus, and
a
b
c
d
e
f
Fig. 7.1 From 25 weeks’ gestation fetuses can display the facial expressions accompanying so called ‘basic emotion’. a c, e g Fetuses at 25 weeks’ gestation. d Fetus at 20 weeks’ gestation. All basic emotions are displayed by the facial expressions of the fetuses. a The expression displayed by this fetus could be classified as anger. The cavity under the nose is not due to a malformation, but to the current limitations of our equipment. b The expression displayed by this fetus could be classified as disgust. c The expression displayed by this fetus could be classified as fear. d The expression displayed by this fetus could be classified as happiness. Fetus d is younger than the others, the smile being the first facial expression to be noted in fetuses. e The expression displayed by this fetus could be classified as sadness. f The expression displayed by this fetus could be classified as surprise
7.1 Basic Emotions
79
a a
b
Fig. 7.2 Blank expression in the early fetus. a, b Fetuses at 13 weeks’ gestation. At this early stage fetuses display a static, blank expression. Muscle, skin and bone formation as well as innerva tion are all still in the early stages and do not allow more varied facial expressions
vocalizations only start after birth. Before 14 16 weeks the fetal face generally looks impassively blank and it would probably be classified as ‘static’ by those trying to study facial expression in the adult (Fig. 7.2). In 1971 Van Hoof observed smiling in several primates, and since then smiling has been considered amongst the most universal and phylogenetically old signals [11]. Anatomically speaking, a smile is the simplest of all expressions, as it requires in its simplest form the contraction of only one facial muscle, the greater zygomatic, innervated like the overwhelming majority of muscles involved in facial expressions by the facial or seventh cranial nerve and its branches. Central or peripheral palsy of the facial nerve impinges deeply on facial expressions on the affected side. The smile is the first facial expression which appears in the fetus. Occasional smiles can be noted from around 15 to 16 weeks. Fetuses begin to smile slightly more consistently between 18 20 weeks and they do so predominantly during cycles of rest (Fig. 7.3). According to
b
c
Fig. 7.3 Early smiles. a Fetus at 20 weeks’ gestation. b, c Fetuses at 17 weeks’ gestation. All fetuses smile. Smiling is the simplest of all facial expressions and the first facial expression to be dis played clearly
Duchenne’s classification, up to 25 weeks fetal smiles would be categorized as ‘false’ [3], not only as the eyes
80
7 Facial Expressions
a
necessarily do not participate in the expression, but also because the involvement of the lower face and especially of the forehead in the smile is limited. One could say that initially fetuses barely smile. By 25 weeks fetal smiles have evolved to big smiles (Fig. 7.4). From mid-pregnancy facial expressions begin to appear consistently, including ‘negative emotions’ which require the involvement and coordination of more than one muscle. Both positive and negative expressions surface in particular during cycles of rest. Surprise is hardly noted before 25 weeks. However, surprise is the briefest of our facial expressions, lasting only a fraction of a second [3]. It could be that 4D ultrasonography cannot capture it as it cannot capture other fast movements such as startles or breathing. Disgust, on the other hand, can occasionally be noted. Tongue protrusion per se is not properly a facial expression. However, in many cultures it indicates varying degrees of hostility towards the addressee. In utero tongue protrusion cannot have a derogatory connotation, although many parents interpret it as such and laugh. Tongue protrusion begins to be observed at about 18 20 weeks. Tongue protrusion also evolves: it starts as a straight sticking out of the tongue, and develops into more complex lateral motions (Fig. 7.5). Some opening movements of the mouth, apparently non-classifiable as expressions, begin to surface between 24 and 25 weeks. Comparing these with the facial display of neonatal vocalizations, the similarity is striking. Neonatal vocalizations are overwhelmingly composed of vowels. Clearly fetuses do not vocalize in utero; however, they may be preparing to do so after birth (Fig. 7.6). All these are unlearnt movements.
7.2 Cross-Modal Integration
b
Fig. 7.4 Full smile. Fetus at 25 weeks’ gestation. The fetus can now display a full smile. In addition to the wide opening of the mouth, ‘wincing’ can also be noted. This smile lasted more than 10 s
In 1977 Andrew Meltzoff and Keith Moore published a ground-breaking paper, ‘Imitation of facial and manual gestures by human neonates,’ in which they described how infants between 12 and 21 days old can imitate adult facial (and manual) gestures. Subsequently infants were found capable of imitation from birth [12]. Early facial imitation had been previously regarded as impossible. For instance, Piaget thought that facial imitation did not occur before 8 12 months [13]. The research had widespread implications for a range of studies
7.2 Cross-Modal Integration
81
b
a
c
Fig. 7.5 Tongue protrusion. a c Fetuses at 25 weeks’ gestation. All fetuses protrude their tongue. However, tongue protrusion is executed in different ways and indicates different facial expressions. a Tongue protrusion is accompanied by wide opening of the mouth, corrugation of the forehead and tightening of the eyelids. The fetus seems to be crying angrily. b The opening of the mouth is minimal, with just the tongue sticking out. The tip of the tongue is slightly curled and the face rather impassive. The fetus seems to be ready to lick something. c Tongue protrusion is minimal, with wide opening of the mouth. The eyelids are slightly raised and the forehead distended. The etus seems to be ready to utter a sound
a
b
c
d
Fig. 7.6 Preparing to utter vowels. a Fetus at 20 weeks’ gestation. b Fetus at 25 weeks’ gestation. Both fetuses look as if they could utter a vowel, possibly an O or an E. c, d Fetus at 22 weeks’ gestation. The mouth can be seen as a black hole. The nostrils are clearly visible in c and the nose appears as an oblong protuberance in d. Both mouth openings were of short duration, unlike yawns, nor was mouth opening repeated as in swallowing. In c the fetus seems to be uttering an A, and in d an O
82
and topics. Relevant to fetuses and to neonates is socalled cross-modal integration, the integration of information from different sensory modalities. Fetuses do not use vision while in utero and yet they do imitate some facial expression in the adult soon after birth. How can they do so? In the Meltzoff and Moore experiment the assumption was that in order to imitate a facial gesture such as a tongue protrusion or mouth opening it was necessary for the infant to translate its experience from one sense modality, vision in the infant very close to its first vision into another sense modality, proprioception [14]. The question, which became known as the Molyneux question, was first raised by the Irish scientist and politician William Molyneux in a famous letter dated 7 July 1688 to the philosopher John Locke, and reformulated in a second letter dated 2 March 1693 as follows: Suppose a man born blind, and now adult, and taught by his touch to distinguish between a cube and a sphere of the same metal, and nighly of the same bigness, so as to tell, when he felt one and the other, which is the cube, which is the sphere. Suppose then the cube and the sphere placed on a table, and the blind man be made to see: quaere, whether by his sight, before he touched them, he could now distinguish and tell which is the globe, which is the cube? [15, 16]. Basically Molyneux was asking whether tactile sensation could be transferred to vision in someone who had been born blind. Historically, answers to the Molyneux problem have been related to the issue of the first perception in the newborn. The Meltzoff and Moore experiment showed that perception is intermodal from the start. Experience in one sense ‘educates’ other sense modalities [14, 17 19]. Fetuses start practising facial gestures and expressions close to mid-pregnancy. From then on fetuses display facial expressions that we would classify as smile, or cry, or even silent vocalizations, and they certainly stick out their tongues. One could thus turn the matter round and say that facial expressions which have long been practised in utero are transferred to the visual modality at birth. This response would thus not be pure imitation but some form of recognition. Infants ‘recognize’ with sight what they have long rehearsed in utero. The work of Rizzolati and his colleagues seems to hold a promising response to Molyneux’s query, with the discovery of so-called mirror neurons. Mirror neurons would provide, amongst other things, an ‘action recognition’ mechanism, meaning the capacity to rec-
7 Facial Expressions
ognize the action and the intention in the other. The mirror neuron system is most likely to be the substrate not only of action understanding the understanding of other’s intentions but possibly of emotion reading and empathy [20, 21]. Humans share mirror neurons with monkeys; however, it is understandable that little research has been conducted so far with humans and infants. Nevertheless, it could be postulated that the nascent mirror neurons of the neonate would allow it to recognize the action of the other and respond by, for example, opening its mouth wide or sticking out its tongue. Other researchers working with infants have postulated that they can appreciate correspondence between their own actions and those of others who are interacting with them [22].
7.3 Preparing for Post-Natal Communications The practising of facial expression and silent vocalizations in utero may be relevant because of another implication. Through facial expressions we can read the intentions and states of the other. If neonates were born without an expressive repertoire, their caregivers would be at a loss in trying to understand and respond to their needs. The pre-verbal child needs to be able to produce facial expressions in order to express its requirements (Fig. 7.7). If an infant cries, the mother takes an action and picks it up, cradles, feeds or puts it down to sleep according to various characteristics associated with the cry. The infant smile is a powerful tool to captivate the attention of the caregiver and elicit a sympathetic response, driving him or her to become involved in an interaction. The same can be said of infant vocalizations, which stir the mother to engage in a so-called proto-conversation. The term ‘proto-conversation’ was first coined by Elizabeth Bates in the 1970s to describe how infants respond to their mothers’ talk with appropriately timed smiles or coos, in give-and-take dialogue-like pattern [23]. Tactile and kinetic (meaning stimulation by motion or motions) modes can be added to the proto-conversation. In this dialogue, mothers talk in ‘baby talk’ or ‘mother-ese’, more properly called ‘child-directed speech’. The intonation of baby talk is different from that of ordinary adult speech, being high pitched, with short and simple words, repetitions, and many glissando variations. Baby talk contributes to
7.4 Parental Reactions
83
a
b
c
d
e
f
Fig. 7.7 a-f Relaxed and sound asleep. All these fetuses look relaxed and sound asleep. The images were taken during a cycle of rest. However, wakefulness is only noted towards the end of gestation. Mothers looking at infants with a ‘sound asleep’ facial expression would not normally attempt to wake them. Even an ‘asleep’ expression (accompanied by other signs after birth) does thus communicate a need which the caregiver is able to understand, in this case the need to sleep. ‘Sound asleep’ is quite different from the blank expression seen in Fig. 7.2, which disquiets mothers as they cannot read it
mental development and to the development of speech [23]. Although fetuses are clearly a long way away from speech, when engaged in displaying facial expressions they may be paving the way for future language as well as for other future social interactions. In contrast to many other movements, facial expressions practised in utero may have an entirely anticipatory role, and a primary one for the start of intersubjectivity, the entry in a social world. During the fetal stage facial expressions do not have a direct and immediate communicative purpose. Fetuses prepare to communicate and by birth are equipped to do so, but within the uterus they have nobody to communicate with.
7.4 Parental Reactions Having said this, it is interesting to hear and see parental reactions to facial expressions during a scan. The impassive, blank 3D or 4D image of a fetal face before 16 weeks is perceived as strange and disquieting by
most mothers, who comment, ‘It looks like an alien’ or ‘It is almost frightening’. The face is supposed to be communicative, emotional and relational par excellence. A blank face elicits unsettling and alarmed feelings. On the other hand, whenever fetuses display a facial expression, an emotional state is immediately attributed to whatever the expression is. When negative expressions are displayed, some go as far as wanting to soothe the unborn. These same expressions will play a very important role in eliciting parental care after birth, and especially so during the pre-verbal period. However, had we not unveiled the expression with the ultrasound scan, no mother would find the ‘blank’ fetus frightening, just as no mother would think of soothing the fetus because at that moment it looks as though it is crying. Facial expressions in utero last only 1 3 s (Fig. 7.8). The ‘crying’ fetus spontaneously ceases to do so after 2 3 s. Prolonged ‘crying’ is not observed. Nevertheless, 4D ultrasonography, by reconstructing and ‘freezing’ the image of the crying fetus, also freezes time and
84
7 Facial Expressions
a
b
c
d
Fig. 7.8 a-d Rapidly changing expressions. Fetus at 20 weeks’ gestation. These four images were captured only seconds apart from each other. Facial displays rapidly change even minimally in utero as in life after birth
speed. Mothers seeing the frozen image dilate time and worry that the cry may be long-lasting. One should ask whether these facial displays may have some hedonic tone. The question has been thoroughly investigated à propos of pain. The Royal College of Obstetricians and Gynaecologists set up a working party on fetal awareness and issued a report in October 1997 [24]. On pages 15 and 16 the report states: ‘In conclusion, the thalamus begins to mature earlier than the cortex but its function beyond a simple relay depends on connections with higher levels of the nervous system which do not begin before 22 weeks’ gestation.
A functional cortex is essential if the fetus is to be aware or to perceive external events… Relatively little is known about human cortical development but a critical fact is that thalamocortical connections are first observed penetrating the frontal cortical plate at 26 34 weeks’ gestation. Therefore, before that time there is no sensory input to the cortex’ [24]. Up until 25 weeks, just as for pain, a hedonic tone is hard to postulate, and even harder to demonstrate. Matters could be different for a slightly older fetus, and even very different for a fetus approaching birth. Communicative purposes are not immediately rele-
7.5 Yawning: a Form of Communicating?
vant during intrauterine life. Fetuses may be preparing for communication, but they certainly do not need to communicate their emotional states in utero. During the first half of pregnancy the substrate that could support emotions or even perceive sensations is simply not there. As Peter Hepper says, sensing is not to be equated with perceiving a more complex operation involving the interpretation of the sensations to give them meaning [25]. We just do not know when perceiving starts. Possibly, just as with consciousness, perceiving is not an either/or phenomenon but one that is gradually built up with varying degrees and shades. Probably a 20- to 25-week fetus can only sense what a mature neonate can perceive, and up to 25 weeks when fetuses smile they are not expressing a deep inner state of joy. They may possibly sense pleasurable or unpleasurable perceptions. However, this is difficult to demonstrate.
7.5 Yawning: a Form of Communicating? Finally, a few additional words on yawning. Provine, the researcher who studied many of its facets, considered it a facial expression with a communicative function after birth. Jean Piaget was the first to make an important observation in 1951: that children started yawning in response to seeing a yawn only during the 2nd year of life [13]. Subsequently, Provine [26] and Anderson and Meno [27] proved experimentally that contagiousness of yawning did not start reliably in children under 6 years old. This finding aroused a lot of interest from two main points of view. Following the seminal studies of Meltzoff and Moore [12], it has long been known that newborns are capable of recognizing
85
and imitating most facial expressions. Many scientists wonder why they do not do the same with such a dramatic facial expression as a yawn. Recently, yawning has met with renewed interested linked with studies of mirror neurons and autism. So-called autistic spectrum disorders are characterized by a more or less profound impairment in communicative capacities and shared social interactions. Autistic individuals have problems in relating to others and in reading other people’s emotional states. In particular, they are unable to display empathic reactions when others show pleasure, fear or pain. Empathy is connected with the capacity to recognize, understand and share the emotions of others [28]. A number of anatomical and functional studies all seem to point to disfunctioning of the above-mentioned mirror neuron system in autism [29, 30]. The fetus and the neonate generally are not ‘autistic’, yet, as Piaget pointed out, they cannot imitate and possibly decode yawns in others [13]. This is just an hypothesis, but what if one turned the matter round and saw yawning as an albeit unconscious form of communicating various states to the caregiver. When mothers see a neonate yawning, they assume, depending on the context, that the child must be sleepy, waking up, or even hungry. In other words, at the neonatal stage yawning could be an important tool for directing the efforts of the caregiver onto ‘the right track’ (Fig. 7.9). Up to 2 years of age, and before they become fully verbal, children need a lot of empathic understanding on the part of those caring for them. Furthermore, sleep and hunger are two basic and vital states. A caregiver can miss many other nuances, but not these vital needs.
Fig. 7.9 Fetal yawn. Fetus at 22 weeks’ gestation (the same as in Fig. 7.6 c, d). The mouth, seen as a black hole, is wide open and the yawn lasted 6 s. Interestingly, this rather non human image (the fetus seen from below looks almost like a rat) had an ‘infec tious’ impact on the mother, who started yawning too
86
Yawning, unlike crying and screaming, elicits empathy and sympathy, not irritation, anxiety or exasperation, thus simplifying the task. Furthermore, in children yawning would not need to be bidirectional. When babies and children are tired, and they communicate this by yawning, the caregiver usually takes the initiative to put them to bed. At such moments they may like to be lulled to sleep, or be read a repetitive story, but clearly are not willing to engage in complex social interactions. On the other hand, though, babies (and young children) are not keen to recognize fatigue or boredom in their caregiver’s face when she or he yawns while they may be needing or demanding some action or interaction. As the saying goes, ‘Children’s needs come first’. Mother or father may be tired, but still have to prepare some food or be asked to participate in some story-telling or preparatory ritual for sleep. If this were the case, fetuses yawning in utero would be displaying a preparatory and anticipatory function, helping them to have their vital needs better understood and more easily met by those caring for them after birth.
References 1. Physiognomy: Webster’s timeline history, 1226 2007. Icon Group, San Diego 2. Khune L (1917) The science of facial expressions. New edn 2008. Health Research, Pomeroy, Washington 3. Duchenne de Boulogne GB (1862) The mechanism of human facial expression. Cuthbertson RA (ed and trans) (1990) Cam bridge University Press, Cambridge 4. Ekman P (2006) Darwin and facial expressions: a century of research in review. Malor Books, Cambridge, Massachusetts 5. Ekman P, Friesen WV (2003) Unmasking the face, 10th edn. Malor Books, Cambridge, Massachusetts 6. Fridlund AJ (1994) Human facial expressions: an evolutionary view. Academic Press, San Diego 7. Izard CE (1971) The face of emotion. Appleton, New York 8. Russell JA (1994) Is there universal recognition of emotion from facial expression? Psychol Bull 15:102 141 9. Hinde RA (1985) Was ‘the expression of emotions’ a misleading phrase? Animal Behav 33:985 992 10. Kagan J (1978) On emotions and its development: a working paper. In: Lewis M, Rosenblum LA (eds) The development of affect. Plenum Press, New York 11. Van Hoof JARAM (1976) The comparison of facial expressions
7 Facial Expressions in man and higher primates. In: von Cranach M (ed) Methods of inference from animal to human behaviour. Aldine Press, Chicago 12. Meltzoff AN, Moore MK (1977) Imitation of facial and man ual gestures by neonates. Science 198:75 78 13. Piaget J (1951) Play, dreams and imitation in childhood. Nor ton, New York 14. Spelke E (1976) Infants’ intermodal perception of events. Cogn Psychol 8:553 560 15. Degenaar M, Lokhorts GJ (2005) Molyneux problem. In: Stan ford Encyclopedia of Philosophy. http:/plato.stanford.edu/en tries/molyneux problem/. Accessed 12 Feb 2010 16. John Locke (1690) An essay concerning human understand ing, 2nd edn. Clarendon Press, Oxford 17. Gopnik A, Meltzoff AN, Kull P (1999) The scientist in the crib. William Morrow, London 18. Schmuckler MA, Jewell DT (2007) Infants’ visual proprio ceptive intermodal perception with imperfect contingency information. Dev Psychobiol 49:387 398 19. Russel JA, Fernandez Dol JM (1998) The psychology of fa cial expressions. Cambridge University Press, Cambridge 20. Rizzolati G, Fadiga L, Fogassi L, Gallese V (2002) From mirror neurons to imitation: facts and speculations. In: Meltzoff AN, Prinz W (eds) The imitative mind: development, evolution, and brain bases. Cambridge University Press, Cambridge 21. Rizzolati G, Sinigaglia C (2008) Mirrors in the brain: how our minds share actions, emotions, and experience. Oxford University Press, Oxford 22. Stern DN (1985) The interpersonal world of the infant. Basic Books, New York 23. Bates E, Bretherton I, Snyder LS (1991) From first words to grammar: individual differences and dissociable mechanisms. Cambridge University Press, Cambridge 24. The Royal College of Obstetricians and Gynaecologists (1997) Fetal awareness: report of a working party. RCOG Press, London 25. Hepper PG (1992) Fetal psychology. An embryonic science. In: Nijhuis JG (ed) Fetal behavior: developmental and peri natal aspects. Oxford University Press, Oxford 26. Provine RR (1989) Contagious yawning and infant imitation. Bull Psychon Soc 27:125 126 27. Anderson JR, Meno P (2003) Psychological influences on yawning in children. Curr Psychol Lett Behav Brain Cogn 11, vol 2, pp 1 7 28. Schmuckler MA, Jewell DT (2007) Infant’s visual proprio ceptive intermodal perception with imperfect contingency information. Dev Psychobiol 4:387 398 29. Jacoboni M (2008) Mirroring people: the new science of how we connect with others. Farrar, Straus and Giroux, New York 30. Trevarthen C, Aitken KJ, Papoudi D, Robarts JZ (1996) Chil dren with autism: diagnosis and interventions to meet their needs. Jessica Kingsley, London
Rest-Activity Cycles, Clusters and the Ontogeny of Sleep
8
With the assistance of Luisa Bocconi, Chiara Boschetto, Florinda Ceriani, Alessandra Kustermann and Cinzia Zoppini
Keywords REM • Electro-oculogram • EEG • Children • Polysomnographic recordings • Active sleep • Quiet sleep • Fetal heart rate • Central nervous system • Wakefulness
Sleep is a very widespread, almost ubiquitous phenomenon. Mammals, birds, most reptiles, amphibians, and fishes sleep. Sleep has different characteristics throughout the animal kingdom [1, 2], but common ground can be found in the near-total inactivity, the decreased capacity to react to environmental stimuli, and the cyclic quality of sleep. Even insects and molluscs display prolonged sleep-like periods of inactivity. Given its widespread and peculiar nature, sleep has always fascinated, puzzled, and disquieted mankind. In antiquity sleep was considered a double-faceted phenomenon. On the one hand, it was regarded as a sweet merging with the realm of oblivion from which one reemerged to a new life, restored and reborn each new day. On the other hand, it was viewed as a plunge into a condition akin to death, the darkness and unconsciousness of the eternal sleep. The Greeks captured well this dual nature in their mythology as Hypnos, the god of sleep and the son of Night and Darkness, was the twin brother of Thanatos, the god of death. Dreams were not considered to be creations of those who dreamt them, but emanations visiting the dreamer from above, direct revelations radiating from the gods. As such, dreams were considered important for their predictive, prophetic qualities. All dreamt, but only a few could be dream interpreters, and only particular dreams were enlightening. Down the centuries sleep was largely disregarded for its restorative qualities, and considered almost solely as a container of enlightening dreams [3]. In 1900 Freud threw a whole new light on dreams, which he considered products of the individual psyche. A. Piontelli, Development of Normal Fetal Movements. © Springer Verlag Italia 2010
Dreams no longer came from the gods, but were psychological phenomena pertaining to the single individual. Dreams contained residues of our daily lives, and vestiges of our past, but especially unconscious, repressed thoughts. According to Freud, repressed wishes were mainly sexual in their content, and their meaning, being abhorrent to the waking mind, had to be uncovered and interpreted. Dreams were thus the ‘royal road’ to the unconscious, the main avenue to unveiling our repressed wishes. Sleep was almost entirely considered by Freud as the ‘guardian’ of dreams [4]. Sleep research devoid of dream interpretation properly began with the pioneering work of Aserinsky and Kleitman in Chicago. Interestingly, it was while observing sleeping infants that Aserinsky noticed periods of sleep characterized by bursts of eye movements. Following these preliminary observations the research was extended to adults. In 1955 two different states of sleep in the adult, one with eye movements (rapid eye movement or REM sleep) and one without (Non-REM or N-REM sleep) were described [5]. Alongside eye electrodes (the electro-oculogram, or EOG), William C. Dement, a student in the Kleitman laboratory, placed electrodes on the scalp of his subjects (the electroencephalogram, or EEG), which included infants, and discovered that during REM periods the brain displayed an intense activity similar to wakefulness, hence the name ‘paradoxical sleep’ [6]. Until then sleep had been regarded essentially as state of quiescence of the central nervous system. Other towering figures of early sleep research during
88
the 1950s and the 1960s were Giuseppe Moruzzi in Italy and William C. Magoun in the USA, who observed the effects of lesions and stimulation at various levels in the brainstem, and Michel Jouvet in Marseille, who studied sleep predominantly in animal preparations [7, 8]. Following the golden era of the 1950s and the 1960s sleep research became widespread and pursued by many investigators. However, despite innumerable researches and many hypotheses, the full function of dreams still eludes our knowledge.
8.1 Sleep in Children For a long while children, especially young, pre-verbal children, were somewhat ignored by mainstream research. It is only fairly recently that so-called behavioural states in the neonate have been fully recognized and described using mixed criteria both by direct observation of neonates and by analysis of polysomnographic recordings, the joint recording of the EEG, EOG, muscular activity (electromyography, or EMG), electrocardiogram (or ECG), and various other parameters such as oxygen saturation and air flow [9]. The term ‘behavioural state’, signifying various sleeping and waking states, is currently used to indicate ‘clearly distinguishable and relatively stable functioning conditions of the organism periodically and predictably recurring over time’ [10, 11]. These functioning conditions comprise the presence or lack of bodily motions and ocular movements, the type of breathing, and specific characteristics of the EEG and of the EMG recorded at the level of the chin. Two basic types of sleep have been distinguished in the neonate. Some confusion has been created by the various names they have been given. One is active sleep, also called state 1, irregular, paradoxical, or REM sleep. The other is quiet sleep, also called state 2, or regular, or N-REM sleep. For reasons of clarity and simplicity only the terms active and quiet sleep will be used in this book. This terminology is also better suited to describing fetal phenomena pertaining to the first 25 weeks of pregnancy. During active sleep cardiac and breathing rhythms are irregular, breathing is paradoxical, apnoeas are possible, rapid eye movements are present, and so is intense and turbulent bodily motion, muscular tone is elevated, and the infant displays a variety of facial expressions. During quiet sleep all activities which come under the
8 Rest-Activity Cycles, Clusters and the Ontogeny of Sleep
control of the neuro-vegetative system are much more regular. Hence cardiac and breathing rhythms are regular, less accelerated and non-paradoxical. No rapid eye movements can be noted, muscular tone ‘falls’ and the overall motility is greatly reduced, being almost exclusively represented by startles and short, occasional local movements. At birth, active sleep is largely prevalent, but soon quiet sleep becomes increasingly predominant [12]. The concept of behavioural states is currently widely used and taken into account during the assessment of individual infants. Neonatologists wait until the infant is in a favourable state, neither fretting restlessly nor sound asleep, before examining it. Increasingly, parents who ask for a consultation about their children’s sleeping problems are helped to distinguish various states and to no longer regard some of these as distressing conditions, but as a predictable chain of naturally occurring events.
8.2 Behavioural States in Premature Infants and Mature Fetuses Behavioural states have been thoroughly investigated in the premature infant. Severely premature infants are continuously asleep and their eyes are kept almost constantly closed. However, they basically alternate between active and quiet sleep. During quiet sleep they are immobile and respiration, although generally assisted, is regular. Active sleep is characterized by more or less pronounced unrest, presence of startles, intermittent eye movements and irregular breathing with frequent apnoeas [13]. At 24 weeks the EEG is mostly silent, with some brief (<30 s) and irregular signs of activity called ‘brushes’. From 27 weeks signs of EEG activity become more prolonged, especially during phases of unrest. ‘Brushes’ may be recorded in different areas of the brain, and can be asymmetric. This kind of tracing, commonly called discontinuous, persists until 30 weeks. It is only from 32 weeks that the EEG becomes increasingly continuous, and active and quiet sleep become well differentiated. At 36 weeks the first brief episodes of quiet wakefulness begin to appear [14]. Ultrasonography has made it possible to identify in the human fetus well-defined phases of sleep comparable to the stages of neonatal sleep starting at 34 36 weeks [15, 16].
8.3 Early Fetal Functioning: Rest-Activity Cycles and Clusters of Activities
8.3 Early Fetal Functioning: Rest–Activity Cycles and Clusters of Activities It should be clear that when we talk about active and quiet sleep in the early fetus, we are only using an analogy to designated states which may contain precursors of neonatal sleep, but which also could differ from it, even profoundly. As described, active sleep in the full-term fetus and in the neonate is accompanied by the surfacing of gross somatic activity and by fetal heart rate (FHR) accelerations and eye movements. Quiet sleep, by contrast, is characterized by the near absence of movement (only startles and brief body motions), absence of eye movements, and a stable FHR. Before birth, active sleep is largely predominant. Parameters used to distinguish these states cannot be applied to earlier stages of pregnancy. Eye movements begin to be noticed from 16 18 weeks, but even then are extremely sporadic, slow and difficult to visualize. If the fetus is turned in an unfavourable position its eyes, and accordingly its eye movements, cannot be observed. Rapid eye movements have been detected in the fetal sheep at around 7 months’ pregnancy [17]. Respiration is discontinuous, and mainly irregular. As described, regular breathing movements surface occasionally during periods of rest. Furthermore, up to 20 22 weeks fetal motions are not coupled with FHR accelerations. When the fetus moves, its heart beat does not increase in frequency. This dissociation, amongst other things, reflects the still scant contribution of the parasympathetic nervous system and the predominance of the vagal component to fetal functioning [18, 19]. The non-coupling of FHR with movements, due to late vagal innervation, may be an important energy-saving method. Early fetuses move a lot. If their movements were linked with an increased cardiac output, this would result in an increase in metabolic rate and a decrease in growth rate. Prior to 34 36 weeks’ gestation authors do not talk about behavioural states, and even less so about sleep, but about rest activity cycles. Fetuses cycle rapidly between periods of turbulent motion and periods of absolute or relative rest: we are talking about a few minutes and seconds, rather than the hours which characterize waking and sleep in life after birth. These rapidly alternating cycles are generally considered disorganized. Activity cycles are assumed to comprise any kind of motion and rest cycles are regarded as totally devoid of these.
89
However, well-organized phases of sleep cannot be postulated to begin to operate without a previous preparatory history or even a simple past. Analysis of the temporal sequence of various phenomena within rest activity cycles shows that these are not as chaotic as usually thought, but are structured differently, and their organization changes rapidly with increasing gestational age. Again, we are talking here about a few weeks rather than months or years as in life after birth. Up to 20 weeks most fetal activities function in an either/or way. General movements are exclude breathing and swallowing movements and vice versa, and none of these activities take place during cycles of rest [20]. Fetuses are either in an aggregate in which they perform general movements (cluster A) or in another in which they ‘breathe’ and swallow (cluster B), or in a cycle of rest. The aggregates are denominated ‘clusters’ (Fig. 8.1). When one cluster of activities is present the other is not. These activities are basically competitive, and become coordinated only gradually during the second half of pregnancy. By 25 weeks, competition still largely prevails over cooperation and coordination. Local motions, startles, twitches and hiccups, on the other hand, are shared with all clusters. In addition, breathing and swallowing tend to surface before the beginning of periods of rest. Viewed in a sequence, a cluster of general movements is followed by one of breathing and swallowing motion, which in turn is followed by a cycle of rest. Finally, the circle is closed and general movements surface again. Once a cluster is over a shift to another phase occurs. Startles are not clustered phenomena; however, by increasingly concentrating at both ends of cycles of rest they could be viewed as phenomena heralding or triggering a shift to another phase. Hiccups, on the other hand, despite producing a jerk of the fetal body which can be quite strong, do not hasten or herald any change when surfacing within periods of rest. One could say that hiccups do not act as a perturbator, while startles can have such a function towards the end of phase shifts. Additionally, within the same cluster some activities can be performed synchronously whilst other activities are only performed separately. Competition initially prevails over cooperation within clusters too. Local movements, startles and twitches are ‘compatible’ with all other activities. Hiccups are compatible with general movements, local movements, startles and twitches,
90
8 Rest-Activity Cycles, Clusters and the Ontogeny of Sleep
General Movements Localized Motions Rest Synchr Syn chrono onous us Mot Motion ionss
St rtles Sta tl Hiccup ps Twitch Twi tches tch es
b
Fetall Breathing thing M Movements Swallowingg Reggula ular Feta etal Brea reathi thing ng
a
c
Fig. 8.1 Clusters. 15 weeks’ gestation. A cluster was considered such when composed of at least two successive episodes of events belonging to the same aggregate. The mean number of the sum of episodes belonging to cluster A and B was calculated and plotted against the mean number of tail events marking the transition from one cluster to the next (A to B and B to A). Behavioural events within the clusters substantially outnumbered those that marked the transition between the clusters. These data strongly suggest that the temporal distribution of such behavioural phenomena is far from random. a See Fig. 2.4 legend for an explanation of the chart design. At this stage clusters are still separate. As can be seen, the cluster in which general movements (re ) emerge does not contain any breathing (blue) or swallowing (light pink) k motions and vice versa. The cycle of rest (from 21 to 29.31 minutes) does not include any of the above mentioned activities. However, during the rest phase (the absence of any coloured bar indicates absolute rest), startles ( (yellow ), localized movements (green), hiccups (grey) and twitches (shocking pink) surface. b Clustering of behavioural phenomena. The number of times each single behavioural episode within cluster A (general movements present) and within cluster B (fetal breathing movements and swallowing motions present) is directly followed by an episode belonging to the same (◆) or the other cluster (■). Each point represents the mean ± standard deviation (plotted as a vertical bar over each point) of the above mentioned events calculated for 30 cases over 1 h observation time per week. c Comparison of the total number of behavioural phenomena belonging to cluster A (▲) and to cluster B (■) during gestation. Each point represents the mean ± standard deviation (plotted as a vertical bar over each point) of the above mentioned events calculated for 30 cases over 1 h observation time per week. (Panel b and c from [20])
but not with breathing or swallowing. Breathing movements and swallowing, although surfacing within the same cluster, are not performed synergistically nor in alternation [20].
Towards 20 weeks important changes start to occur. Clusters and ‘incompatibilities’ begin to dissolve and cooperation between some activities starts to emerge, albeit very briefly. General movements and breathing
8.3 Early Fetal Functioning: Rest-Activity Cycles and Clusters of Activities
movements can be synchronous, even if for a few seconds only. Swallowing, however, remains incompatible with general movements. Swallowing and breathing also begin to be performed adjacently and occasionally in alternation, heralding the fine coupling of these activities after birth. Hiccups continue to be incompatible with both breathing and swallowing. Although fetuses largely continue to operate in the previous non-cooperative mode, by occasionally coupling various activities whose synergism is essential for survival after birth, they also begin to be able to function, albeit briefly and sporadically, in a postnatal mode. Additionally, from 16 18 weeks occasional and isolated episodes of shallow and regular breathing similar to those which will characterize breathing during quiet sleep begin to surface within periods of rest. Another fact may be of interest. From the start, cycles of rest are characterized by decreased muscular tone as judged initially by adhesion of the fetal body to the uterine surface. Up to 13 weeks, during cycles of rest fetuses lie in the supine position with their spine and head adhering to the placenta or the uterine wall. Arms and legs, on the other hand, float freely in the amniotic fluid. Later, when the whole fetal body can no longer lie ‘flat’ and in contact with the uterine surface, a fall of the neck and spine tone with extreme forward ‘drop’ and curvature, and occasionally a fall of chin tone accompanied by mouth opening, can be noted. Arms and legs, having become ‘heavier’, are generally semi-flexed and not in contact with the uterine surface (Fig. 8.2). As soon as fetuses enter a phase when general movements are performed, fetal nuchal and axial tone rise, heralding or merging with the change. The distance between the fetal body and the uterine wall shows an increase ranging between 4 and 6 mm. Operators carrying out a nuchal translucency test, usually performed between 11 and 13 weeks gestation to evaluate the probability of Down syndrome (which can be evidenced by increased nuchal thickness), are well aware of the variation in nuchal tone, although they may not interpret it as such: they simply find it annoyingly impossible to measure nuchal thickness during a full-blown spell of fetal generalized movement, including the pauses between fragmented burst. Accurate measurements can only be taken while the fetus is at rest. Cycles of rest become moderately more prolonged as gestation progresses. As pregnancy progresses, nuchal and spine tone can be considerably raised during a spell of activity, and
91
a
b
c
Fig. 8.2 Resting tone. a Fetus at 14 weeks’ gestation. The whole body adheres to the uterine surface, save for the legs, which can be seen floating in the amniotic fluid. The chin touches the chest, indicating a drop in neck muscular tone. b Fetus at 16 weeks’ gestation. The uterine cavity is less even and more restricted and no longer allows for complete adherence of the entire fetal body to the uterine wall. The head and thorax adhere to the wall. The lower (lumbosacral) part of the body and the legs and arms all float in the amniotic fluid. The chin touches the chest, indicating a drop in neck muscular tone. c Fetus at 22 weeks’ gestation. The fetus can no longer lie in the complete supine position. The spine and the neck are completely bent forwards, and the head, unsupported by the neck, is lying on the uterine wall. The chin touches the chest and the lower jaw is open, further indicating a drop of chin tone. The legs are completely folded, also indicative of reduced muscular tone
92
fetuses can be seen ‘sitting up’ unsupported (Fig. 8.3). This observation would suggest that variations of muscular tone are displayed from very early on in gestation, with an increase in tone during cycles of activity when general movements surface.
8 Rest-Activity Cycles, Clusters and the Ontogeny of Sleep
ing to the reticular formation of the brainstem [23]. These in turn control spinal motoneurons [24]. The initial dependence of general movements on startles for commencement could be explained by this connection.
8.3.1 Ontogeny of Sleep and its Possible Precursors As mentioned previously, until quite recently the ontogeny of sleep could only be studied through direct observation of the premature and with the aid of polygraphy (the simultaneous recording of various physiologic activities) [9]. The clinician dealing with severely premature infants is well aware of most of the phenomena of physiological fetal functioning, but clearly regards these as solely pathological and due to immaturity. For instance, the severely premature infant cannot feed autonomously. Sucking requires the smooth coordination of various activities such as rooting, opening of the mouth, grasping and holding the nipple, squeezing it, and finally swallowing. The premature infant is hardly capable of performing this complex sequence [21]. Breathing also has to be assisted, being frequently interrupted by prolonged apnoeas. Behavioural manifestations of the fetus living within its natural environment and their variations can give us important indications about physiological functioning and development. They can also provide some indications about the neural substrate presiding over these same manifestations. Nevertheless, when talking about the human fetus one is clearly linking later phenomena and later physiological functioning to much earlier phenomena. This being so, the considerations expressed in this section are necessarily of a speculative nature. Some of the building blocks of later phenomena may be there, but considerable reshuffling and further events will be added as the fetus continues to grow. Studies on the ontogenesis of the brain have mainly focused on the formation of the cortex and the diencephalon. However, the vital structures of the brainstem, specifically the bulb and the pons, have been found to organize first [22]. From 10 to 25 weeks’ gestation the central nervous system undergoes innumerable transformations. From 10 weeks, fetuses show signs of brainstem functioning. Startles are known to originate within neurons belong-
a
b
c
Fig. 8.3 Active tone. a Fetus at 17 weeks’ gestation. b Fetus at 19 weeks’ gestation. c Fetus at 21 weeks’ gestation. All fetuses keep their head raised sustained by their neck. The head and cervical spine are not supported by any portion of the uterine surface. In the fetus in c the lumbar column is also raised. All fetuses point their feet against the wall, also indicating increased muscular tone
8.3 Early Fetal Functioning: Rest-Activity Cycles and Clusters of Activities
When dependence becomes no longer functional, inhibitory mechanisms or biochemical or other transformations could account for the change that de-couples startles from general movements. Between 10 and 12 weeks, fetuses show other signs of brainstem functioning. ‘Breathing’ and other nonspinal activities such as swallowing begin to emerge. Brainstem reticular structures also control sleep. Breathing in particular is inextricably linked with sleep [25, 26] (Fig. 8.4). Activity cycles show some analogies with active sleep of the mature fetus [15, 16]. Initially general movements surface when breathing movements and swallowing are not present. However, towards midpregnancy these mutually incompatible clusters begin to merge. As in active sleep, gross somatic activity is coupled with irregular, paradoxical breathing. Rest cycles, on the other hand, show some analogies with quiet sleep [27]. During rest cycles the fetus displays no movements, save short and occasional local ones, and some startles also emerge. Regular, superficial breathing movements emerge in isolated bouts at around 16 weeks. Before then, breathing movements are only present during activity cycles. The vital structures of the brainstem, specifically the bulb and the pons, which are known to organize first, in turn keep the forebrain and cortical structures under synaptogenic and maturational control [28, 29]. Inhibitory subcortical structures are not mature during
Cerebral hemisphere
93
the first half of pregnancy. Inhibitory circuits start functioning towards the end of pregnancy and beyond. The lack of motor feedback inhibition during activity cycles appears to be functional to this stage by allowing general movements to be performed. Intense bodily activity accompanies cycles of activity and will continue to accompany active sleep during the first few months after birth. This activity could be viewed as an important sensorimotor stimulation to the brain. Fetal motions are not stereotyped, and fetuses constantly adapt to the changing requirements of the external and internal environment (Fig. 8.5). General movements could thus be viewed as a component of prenatal ‘learning’ and neural development. The first episodes of wakefulness do not appear in the fetus before 36 weeks [15, 16]. Wakefulness is a late addition, which presumably arises in order to meet the different requirements of the physical and social postnatal environment. Around 3 4 months the neonate is awake for long enough intervals of time to consistently interact with its environment and starts learning from it. Interestingly, at about the same time general movements are no longer observed during active sleep [30]. Localized movements, on the other hand, with their goal-directed, nondiffuse nature, may become the kind of movements to be found during waking. Nine tumultuous months separate the early fetus from the neonate, but some building blocks of various
Diencephalon Brain stem
Infidbular stem Cranial nerves
Reticular core
Spinal cord
Fig. 8.4 The fetal brain at 15 weeks’ gestation. Ascending and descending projections originating in the brain stem are represented with arrows. The magnified detail panel shows a re ciprocal circuit between neurons con trolling general movements (GM) and those involved in ‘respiration’ (BR), which could explain the clustering observed. From [20]
94
8 Rest-Activity Cycles, Clusters and the Ontogeny of Sleep
trajectories start to be laid down near the beginning of fetal life. Knowledge of the remote ontogeny can help elucidate later phenomena. As the great Aristotle declared, ‘He who sees things grow from the beginning will have the best view of them’ [31].
References
a
b
c
Fig. 8.5 Adapting motions to changing conditions. Fetus at 20 weeks’ gestation. The intrauterine space is becoming progressively restricted. The legs have considerably elongated, and their muscular development and ossification (bone production, seen in white) have progressed. The fetus has become skilled at adapting its movements to its changed internal and external conditions. The legs are folded (a), stretched almost in parallel (b), and then stretched in alternation (c) in order to initiate a general movement provoking a complete turning round (not seen here)
1. McGinty D, Szymusiak R (1994) Neurobiology of sleep. In: Saunders NA, Sullivan CE (eds) Sleep and breathing. Lung biology in health and disease series, vol 71, 2nd edn. Marcel Dekker, New York 2. Lacrampe C (2003) Sleep and rest in animals. Firefly Books, Richmond Hill, Ontario 3. Schutzman S (1976) The history of sleep. Gallimaufry, San Francisco 4. Freud S (1900) The interpretation of dreams. Standard edn vols 4 and 5. Hogarth Press, London 5. Aserinsky E, Kleitman N (1953) Regularly occurring periods of eye motility, and concomitant phenomena, during sleep. Science 118:273 274 6. Dement W, Kleitman N (1957) The relation of eye movements during sleep to dream activity: an objective method for the study of dreaming. J Exp Psychol 55:543 553 7. Moruzzi G, Magoun HW (1949) Brain stem reticular forma tion and activation of the EEG. Electroencephalogr Clin Neu rophysiol 1:455 473 8. Jouvet M, Michel F (1960) Release of the “paradoxal phase” of sleep by stimulation of the brain stem in the intact and chronic mesencephalic cat. CR Soc Biol 154:422 425 9. Kelmanson IA (2006) Polysomnography as the objective method of sleep study. In: Kemalson IA (author) Sleep and breathing in infants and young children. Nova Science Pub lishers, New York 10. Wolff PH (1966) The causes, controls and organization of behavior in the neonate. Psychol Iss 5:1 105 11. Prechtl HFR (1974) The behavioural states of the newborn infant. Brain Res 76:1304 1311 12. Salzarulo P (2003) Il primo sonno: sviluppo dei ritmi sonno veglia nel bambino. Bollati Boringhieri, Torino 13. Curzi Dascalova L, Giganti F, Salzarulo P (2008) Neuro physiological basis and behavior of early sleep development. In: Marcus CL, Carroll JL, Donnelly DF, Loughlin GM (eds) Sleep in children: developmental changes in sleep patterns, 2nd edn. Informa Healthcare, London 14. Kelmanson IA (2006) Sleep and breathing in infants and young children, chap ‘Sleep ontogeny’. Nova Science Pub lishers, New York 15. Nijhuis JG, Prechtl HFR, Martin CB (1982) Are there behav ioural states in the human fetus? Early Hum Dev 6:117 124 16. Arduini D, Rizzo G, Giorlandino C et al (1986) The devel opment of fetal behavioural states: a longitudinal study. Prenat Diagn 6:117 124 17. Schwab K, Groh T, Schwab M, Witte H (2009) Nonlinear analysis and modelling of cortical activation and deactivation patterns in the immature fetal electrocorticogram. Chaos 19:511 517
References 18. Van Leeuwen P, Lange S, Bettermann H et al (2009) Fetal heart rate variability and complexity in the course of pregnancy. Early Hum Dev 54:259 269 19. Porter FL (2001) Vagal Tone. In: Singer LT, Zeskind PS (eds) Biobehavioral assessment of the infant. Guilford Press, New York 20. Piontelli A (2006) On the onset of human fetal behavior. In: M Mancia (ed) Psychoanalysis and neuroscience. Springer Verlag, Milano 21. Mizuno K, Ueda A (2003) The maturation and coordination of sucking, swallowing, and respiration in preterm infants. J Pediatr 142:36 40 22. Spreafico R (1994) Some developmental aspects of telen cephalic structures involved in motor control. In: Fedrizzi E, Avanzini G, Crenna P (eds) Motor development in children. John Libbey, London 23. Alajuanine T, Gastaut H (1955) La syncinésie sursaut et l’épilépsie sursaut à déclanchement sensoriel or sensitif in opiné. I. Les faits anatomocliniques (15 observations). Rev Neurol 93:29 41 24. Llinàs RR, Terzuolo CA (1964) Mechanisms of supraspinal ac tions upon spinal cord activities. J Neurophysiol 27:579 591
95 25. Jouvet M (1962) Recherches sur les structures nerveuses et les mécanismes responsables des différentes phases du sommeil physiologique. Arch Ital Biol 100:125 206 26. Lydic R, Baghdoyan AH (2002) Neurochemical evidence for cholinergic modulation of sleep and breathing. In: Carley D, Radulovacki M (eds) Sleep related breathing disorders. Marcel Dekker, New York 27. Dreyfus Brisac C (1968) Sleep ontogenesis in early human pre maturity from 24 to 27 weeks of conceptual age. Dev Psychobiol 1:162 169 28. Rakic P (1988) Specification of cerebral cortical areas. Science 241:170 176 29. Marin Padilla M (1995) Prenatal development of fibrous (white matter), protoplasmic (gray matter), and layer I astrocytes in the human cerebral cortex: a Golgi study. J Comp Neurol 357:554 572 30. Foulkes D (1999) Children’s dreaming and the development of consciousness. Harvard University Press, Cambridge, Massa chusetts 31. Aristotle. Historia animalium, books 1 10. In: Gotthelf A, Balme DM (eds) Cambridge classical texts and commentaries. Cambridge University Press, Cambridge, UK
Twin Fetuses and Twin Myths
9
With the assistance of Luisa Bocconi, Chiara Boschetto, Elena Caravelli, Alessandra Kustermann, Umberto Nicolini, Sarah Salmona, Beatrice Tassis, Laura Villa and Cinzia Zoppini
Keywords Intrapair stimulation • Monochorionic • Dichorionic • Zygosity • Monozygotic • Behavioural individuality • Myths • Communication • Maternal emotions • Bereavement
Since time immemorial twin births have been perceived as an extraordinary, often disquieting phenomenon. Many legends and myths flourished around their origins. Twins were thought to be the result of adultery, considered as the incarnation of evil spirits and ancestors, or, as in the myth of Castor and Pollux, one was regarded as having godly origins and the other not. The intrauterine activities of twins have also been subject to much speculation, frequently transforming twins into highly socially interactive partners. Twins have been imagined to fight or even to have intercourse in utero. Several legends, updated and modernized, continue to this day and still have a bearing on the lives of many twins. In some areas of the world these legends can even cost the life of one or both twins [1]. Only in 1875 did Sir Francis Galton’s work ‘The history of twins as a criterion of the relative power of nature and nurture’ bring twins into the realm of inquisitive science [2]. Since then innumerable studies have centred on twins as an ‘experiment in nature’ in the attempt to resolve the nature/nurture controversy: trying to tease out the relative strength of the genetic inherited makeup and the various environmental components in determining different characteristics of the individual. Due to fertility treatments, births of twins are currently on the increase and consequently the focus on twin fetuses has increased too. Because of the many complications potentially associated with twin gestations, research, articles, and treatises on twin pregnancies are plentiful. However, fetal movements in twins pregnancies have A. Piontelli, Development of Normal Fetal Movements. © Springer Verlag Italia 2010
been investigated little in general and least of all with regard to the first half of gestation. Researchers have concentrated mainly on the last trimester of pregnancy and found a significant degree in coincidence in fetal activity, suggesting that evoked fetal movements are operative in the later stages of twin gestations.
9.1 Beginnings of Intrapair Stimulation and its Relevance for Our Knowledge of the Sensory Capacities of All Fetuses Twins can be used as a perfect ‘experiment in nature’ to demonstrate the functioning in utero of two sensory modalities which would otherwise be impossible to verify in the fetus within its natural environment. Intrapair stimulation, with responses to touch and pressure originating from the other twin, proves per se that tactile and proprioceptive sensitivity that is, touch and muscular/bodily sense are operative in utero [3, 4]. The simple fact that a reaction is produced in a twin when it is pushed by its co-twin adds to our knowledge of the sensory capacities of all fetuses. These sensory modalities had otherwise only been investigated by Minkowski [5], Hooker [6] and Humphrey [7] in spontaneously or surgically aborted pre-agonal fetuses that they stimulated in various ways. In utero, movements can be considered as evoked when an otherwise immobile twin fetus is dislodged by some movement of the co-twin and responds to this
98
displacement by starting a movement of its own within 1 s or less. Whilst fetal movements start becoming evident by 7 weeks, intrapair stimulation before 10 or 11 weeks’ gestation can be considered a fairly exceptional event [3]. Twins are usually implanted too distantly from each other, the space within their amniotic sacs is too big, the membranes dividing the sacs are too thick and their movements too weak for them effectively to reach their co-twin earlier. Ideally one would have to observe monoamniotic twins twins sharing both the same placenta and the same amniotic sac- as soon as fetal movements begin, but they are extremely rare, representing only 1% of all monozygotic twin pregnancies, which in turn have a rate of occurrence calculated to be 1 per 250 live births [8, 9]. Nevertheless, intrapair stimulation can occasionally be noted as early as 9 weeks. Whether its beginning in the human fetus coincides with the onset of fetal movements still remains an open question. Consequently, whether proprioceptive and tactile sensitivity are already operative by then also remains to be determined. Before 11 weeks the motions of one twin can elicit a counter-reactive motion in the co-twin only if the twins share the same placenta and are hence called monochorionic (from the Greek words mono meaning one and chorion meaning the placenta) (Fig. 9.1). Evoked responses, at this initial stage, represent an average of 2% of the overall movements. At 11 weeks, save possibly in exceptional cases, intrapair stimulation continues to occur only in monochorionic twins. The sharing of the same placenta (favouring closer proximity), and the particular thinness and reduced number of the membranes dividing the amniotic sacs connected with this type of pregnancy, are likely to enhance the possibility of mutual stimulation. Evoked movements come to represent 6% of the overall activity of these pairs. At 12 weeks all monochorionic pairs and 60% of dichorionic ones respond to mutual stimulation. Evoked activity represents 10% of the overall activity of monochorionic pairs and 6% of the dichorionic ones. From 13 weeks onwards intrapair stimulation begins to be noted in all dichorionic pregnancies. However, evoked responses are still slightly more frequent in monochorionic pairs (13% versus 10% in dichorionic pairs). This delay can be accounted for by the fact that the separate placentas of dichorionic twins can be implanted in different and quite distant sites. More importantly, per-
9 Twin Fetuses and Twin Myths
a
b
Fig. 9.1 Dichorionic and monochorionic twins (10 weeks’ gesta tion). Early in pregnancy twins are distinguished according to the type of placentation, or chorionicity (from the Greek word chorion meaning placenta). Dichorionic twins have two separate placentas whilst monochorionic ones share one. Early ascertainment of chori onicity type guides the subsequent clinical conduct, monochorionic twins being ‘at risk’. All dizygotic twins and roughly 30% of monozygotic twins have separate placentas. The remaining 70% of monozygotic twins are monochorionic. a Dichorionic twins. The separate placentas can be seen at the opposite poles of the gestational sacs. A thick, full set of membranes (arrow) separates the two sacs. The arrow also indicates the so called ‘lambda sign’, another sign of dichorionic placentae. At this stage the gestational sacs of dichorionic twins are too distant, the twin fetuses too small, and the membranes dividing them too thick to allow mutual stim ulation. As pregnancy progresses, increasing fetal growth and in creased proximity will allow it. b Monochorionic twins. The shared placenta can be seen on the left hand side of the gestational sac. A very thin single layer of membranes (arrow) divides the two sacs. Even at this very early stage intrapair stimulation is possible. These monozygotic twins already show marked growth discordance, twin A being almost double the size of twin B
haps, the membranes dividing the amniotic sacs are considerably thicker, and multi-layered, rendering early contact of sufficient strength fairly improbable (Fig. 9.2). With advancing gestational age twins grow rapidly
9.2 Features of Rest Cycles Revealed by Twins
Fig. 9.2 Dichorionic twins, 11 weeks’ gestation: 4D image. These twins have separate but adjacent or so called fused placentas, seen in the lower part of the sacs. However, the membranes, whose thickness is enhanced by the 4D image, impede contact. Both twins are moving their limbs, but the movements of each are unlikely to be sensed by the co twin
in size, with a corresponding increase in the vigour of their movements. This results in a greater likelihood of contact between them. Soon contact becomes inevitable. From 14 to 25 weeks, evoked movements are invariably and increasingly present in all twin fetuses. General movements are the prevalent form of evoked reaction until 13 weeks, although from 11 weeks rare discrete responses are occasionally noted. At 11 12 weeks these are represented almost exclusively by hand face contacts and by isolated leg movements. After 13 weeks evoked general movements decline steadily in line with the decline of spontaneous general movements. On the other hand, discrete evoked responses increase and become more varied from 13 weeks on. Isolated arm movements, leg movements, head anteflexion, head retroflexion, head rotation and stretches are all present. By 16 weeks generalized and localized responses are equally represented. By 20 weeks localized movements predominate over generalized ones. Interestingly, from 22 to 25 weeks, whenever touched on their face, twins predominantly (80%) react with head rotations towards the side of the stimulus, possibly intimating the start of the rooting reflex. The rooting reflex, one of the so-called primitive reflexes, is present at birth and helps neonate orienting towards the breast. When stroked on the chin or mouth the newborn moves its head automatically towards the side where it has been touched, making searching motions and opening its mouth until it latches onto the nipple.
99
Although prevailing modes of evoked responses are observed at different gestational ages, no stereotyped patterns of evoked motility can be noted at any gestational age (Fig. 9.3). By 15 weeks intrapair stimulation is already a constant and, up to 25 weeks, an increasing feature of all twin gestations. From the occasional contacts that could be noticed at 10 weeks, reactive movements come to represent almost one-third of all movements by 20 25 weeks [3, 4]. Thus, intrapair stimulation appears to be an important, consistent and unique determinant in the intrauterine behaviour of all twins from late first trimester/early midgestation onwards. In addition to the countless intrauterine environmental factors affecting any fetus, the twin fetus is also influenced by the motions of its co-twin. Mutual influence extends well beyond behavioural aspects and can affect fundamental features of the intrauterine life of twins. Twins are usually born earlier and smaller than singletons, and can suffer from several, even fatal complications due to the presence of the co-twin which do not affect singletons. However, twins would stand apart from singletons even if the existence of intrapair stimulation was the only distinguishing factor. Whether the increased stimulation in some way affects the neural or even the future social maturation of twin fetuses remains an open question, and one that would merit further investigation [10].
9.2 Features of Rest Cycles Revealed by Twins A fetus emerging from a cycle of rest usually first shows some isolated sign of activity signalling that it is coming out of this phase. A slight movement of one leg, hand or arm, a slight stretch of its head or spine, or a twitch heralds the change. Even more frequently and regularly, startles anticipate and precipitate the change. If some kind of stimulation from one twin reaches the other when motor signs begin to indicate that it is entering an activity cycle, the change of phase is accelerated. In other words, intrapair stimulation can be regarded as a potential environmental perturber precipitating the entry into an active cycle when a fetus is near a point of change. On the other hand, intrapair stimulation has no effect when it occurs at a time of stability. From 10 weeks, during rest cycles, periods can be noted when twin fetuses appear to be impervious to even the strongest intrapair
100
9 Twin Fetuses and Twin Myths
Fig. 9.3 Monozygotic (Mz) and dizygotic (Dz) twins: evoked movements. These ‘actograms’ show a detailed analysis of all movements, including subdivisions of localized movements. Evoked movements display intrapair differentiation at all gestational ages under consideration. Each twin, whether monozygotic or not, has its own fairly distinct ‘style’ of reaction
stimulation. The body of the stimulated twin is passively displaced within the amniotic fluid often quite far from its original position, but shows no response and falls back unperturbed into an inactive state [3] (Fig. 9.4). Given the random nature of intrapair stimulation, these periods cannot be demarcated and quantified. However, the relatively high frequency of stimuli originating from the other co-twin during an activity phase would make them appear not to be chance events. Periods of sensorimotor inhibition appear to be present from very early in gestation within cycles of rest. Cycles of rest are characterized by refractoriness at mid-cycle. Only during periods
of ‘instability’, towards the beginning or the end of cycles of rest, does intrapair stimulation become increasingly effective as a ‘perturber’ precipitating the shift to another phase [11, 12]. As with startles, intrapair stimulation can act as a ‘perturber’.
9.3 Similarities and Differences Until fairly recently the investigation of aspects of twin behaviour in utero more properly linked to traditional twin research was limited by the fact that zygosity could
9.3 Similarities and Differences
101
Twin A (15 Weeks g a.)
General Movementts Locali lizedd M Moti tions Rest Syn y chronous Motions
Startles St t tl Hiccup cup u s Twitches
Fetall Breathing thingg M Movements Swallowingg Reggular Feta etal Brea e thingg ea
Twin B (15 Weeks g.a.)
General Movementts Locali lizedd M Moti tions Rest Syn y chronous Motions
Startles Hiccup Hi ps Twitches
Fetal Breathing g Movements Swall llowing i g Reg gular Fetal Breathingg
Fig. 9.4 Intrapair stimulation: a ‘perturber’ during periods of instability. See Fig. 2.4 legend for an explanation of the chart design. The twins are in similar, but not coincident (or ‘in phase’) cycles. In twin A, elicited localized movements at minutes 4 (0 10 s and 38 45 s), 24 (1 5 s), 26 (17 28 s) and 27 (48 55 s) do not cause a change of cycle or cluster. The fetus is in a cycle of rest and continues to be unperturbed. However, in the same twin an elicited localized movement at minute 4 (55 60 s) precipitates a change: the twin enters a cycle where general movements are displayed. This elicited movement acts as a perturber at a point of instability, thus hastening a change whilst the other elicited movements did not. The same can be seen for twin B: elicited movements at minutes 3 (24 32 s), 23 (0 10 s), 24 (14 20 s) and 30 (34 42 s) do not act at points of instability and thus do not cause a shift to another cluster, whereas a movement at minute 10 (0 8 s) causes a perturbation in this case a short general movement
102
only be determined in sets of twins of opposite sex, and hence clearly dizygotic. Diagnostic refinements now allow zygosity to be determined reliably in utero, even when the placenta is monochorionic [13]. Prenatal determination of zygosity has allowed twin research to start in utero. The possibility of diagnosing monochorionic placentas with a good degree of accuracy early in pregnancy, apart from its clinical relevance, has also made it possible to compare the behaviour of two populations of twins: monozygotic monochorionic and opposite-sex dizygotic in utero. Roughly 70% of all monozygotic twins share the same placenta and are hence monochorionic, and all dizygotic twins have separate placentas and are hence dichorionic. Since 30% of all monozygotic twins have separate placentas, when confronted with same-sex twins, one has to wait until birth to determine if these are monozygotic or not [9]. Although ideally one would eventually like to be able to evaluate the behaviour of all types of twins in order to remove the possible variable represented by gender-linked behavioural differences, prenatal uncertainty regarding zygosity would mean observing an enormous population of twin pregnancies. Furthermore, since the majority of monozygotic twins share the same placenta, these are the most representative subgroup of the entire population of monozygotic twins. It may well be, however, that some behavioural characteristics of monozygotic twins found by twin studies may be related specifically to this type of twinning. Intrapair similarities in levels of spontaneous activity are initially greater in monozygotic than in dizygotic twins. Monozygotic twins have similar, but not identical initial activity levels, and their mean differences increase considerably with gestational age. Dizygotic twins show much higher differences from the start and the differences decrease only slightly over time. However, by 20 25 weeks monozygotic twins have reached the same degree of diversity as dizygotic twins, indicating clear behavioural autonomy [14] (Fig. 9.5). Furthermore, when one analyses individual movements, the larger behavioural differences initially found in the activity levels of dizygotic twins are no longer noted. Examined in detail, both monozygotic and dizygotic twins merge together behaviourally and become indistinguishable. Each fetus, regardless of its zygosity, has its own fairly distinctive way of acting and reacting. As in life after birth, monozygotic twins may be considered behaviourally alike at a macroscopic level of analysis, but can never be considered identical upon detailed examination. Furthermore, without the interference
9 Twin Fetuses and Twin Myths
Fig. 9.5 Monozygotic and dizygotic twins: activity levels. Intrapair developmental trends. The overall activity level of each fetus during each observation was determined by adding the duration of all spontaneous movements, expressed as a percentage of the time of observation. Activity levels in monozygotic twins initially show the greater intrapair similarities. However, by 20 22 weeks’ gesta tion, monozygotic twins have reached almost the same degree of intrapair diversity as dizygotic ones
of the often strikingly similar postnatal appearance of monozygotic twins, dissimilarities in behaviour can be observed even more clearly during prenatal life. All fetuses can only perform certain activities and functions at given gestational ages. However, not even monozygotic twins carry them out in exactly the same way (Fig. 9.6). Complications may further increase the diversity in the level and type of motions of all twins. However, monochorionic-monozygotic twins, being more prone to various prenatal complications, are more likely to be affected. For instance, marked discrepancies in amniotic fluid volume can develop. Due to scarcity of amniotic fluid, one twin can become constrained in its movements, whilst those of the other may be enhanced by an abundance of liquid. More severe complications can have more dramatic effects. By the time fetal motions become observable, the fetus is already 7 weeks old. Although it may sound like a joke to an adult, a day, and even more so a week, is an enormous amount of time by embryological standards. By the time twins begin to move, a lot could have already happened to set them on behaviourally distinct paths. Embryology and molecular genetics tell us that monozygotic twins are never identical from the start [15]. All this throws a new, less deterministic light on the lives of socalled identical twins which allows space for diversity within apparent homogeneity and identity (Fig. 9.7).
9.3 Similarities and Differences
103
Fig. 9.6 Monozygotic and dizygotic twins: spontaneous movements. On detailed analysis, differences between monozygotic and dizygotic twins are no longer noted. These ‘actograms’ show clear intrapair differences in all twins, regardless of their zygosity
Fig. 9.7 Monozygotic twins at birth: impact of the intrauterine environment. The intrauterine environment is considered as being ‘neutral’ or a non variable in many twin studies. However, the remarkable difference in size between these twins indicates a profound impact of the prenatal environment on them. Twin 1 weighed 2500 g and twin 2 1400 g at birth. Differences in weight and height were seen to continue at subsequent follow ups. At 8 years twin 2 was still much shorter and lighter than his co twin
104
9 Twin Fetuses and Twin Myths
9.4 Behavioural Individuality
9.5 Universal Myths
Twin fetuses provide a unique opportunity to observe a fetus and compare its behaviour and that of its cotwin simultaneously. The co-twin can become a ‘control’. As we have said, all twin fetuses are behaviourally quite independent, showing differing levels of activity, preferential positions and fairly distinctive favoured movements. Another means of investigating behavioural differences and autonomy in twin fetuses is to determine whether their rest activity cycles coincide in time. From the beginning, variance and autonomy control these factors. Only at 10 11 weeks do rest activity cycles appear to be slightly more concomitant in monozygotic twins. However, at 10 11 weeks most activities have only just begun to emerge sporadically. Later cycles of rest and the cyclic emergence of different clusters of activities are neither more nor less coincident than those of dizygotic twins. In all twins different clusters, including cycles of rest, are sometimes in phase and sometimes not. In the early stages of pregnancy cycles alternate so rapidly that coincidence or disparity are more or less equally likely to occur. With advancing gestation, as clusters become more defined and prolonged, coincidence becomes less pronounced [16]. The irregular coincidence of various clusters and cycles in twin fetuses tells us that an internal autonomous ‘clock’ regulates each twin. Monozygotic twins therefore have an added element of ‘autonomy’ that distinguishes one from the other. This is particularly relevant in the case of those twins who share the same placenta. Monochorionic twins have vascular communications through their shared placenta. These communications can involve often serious complications, and should one twin die for any reason, the other is also likely to die or to be seriously damaged by the death of its co-twin [17]. This aspect clearly links the majority of monozygotic twins in terms of life and death during their intrauterine existence. However, paradoxically, this same condition of extreme mutual dependence also demonstrates further behavioural autonomy. In theory, substances could be exchanged through vascular connections, triggering more coincident behavioural cycles. The fact that rest activity cycles are no more in phase in monochorionic twins than in dizygotic ones disproves this and further emphasizes the prenatal behavioural autonomy and individuality.
Many myths encumber fetal life, and twins are doubly exposed to a host of myths. Some myths regarding all fetuses will be discussed first. The image of the fetus which most readily comes to the public’s mind is that of early first trimester. Visualization on an ultrasound scan of the fetal body in its entirety is only possible up to 16 18 weeks. Beyond this gestational age, only parts of the fetal body can be observed at any one time. A third-trimester fetus merely visible in segments is clearly less glamorous and less prone to being used for illustrations in non-specialist publications. Early fetuses move a lot and this lively activity is also more appealing to a non-specialist audience, which would find relatively long stretches of immobility tedious to watch. Growth and development are extremely concentrated and accelerated in fetal life. We often talk about days and weeks rather than months and years as in postnatal life. What is appropriate to ascribe to a certain gestational age may not be equally pertinent to an earlier or later phase. Despite the rapidity of all developmental changes, most people refer to ‘the fetus’ in general, quite independent of gestational age. By referring to ‘the fetus’, they generally mean the familiar early, active first-trimester/midpregnancy one. The lively activity of the early fetus is all too often interpreted as representing wakefulness, which is then considered identical to intentionality and consciousness. In fact, an early active fetus is not awake at all. Brief spells of waking have been noted only towards the end of pregnancy [18]. In addition, although wakefulness is a precondition for consciousness, the two do not necessarily coincide [19]. Malformed fetuses born without the cerebrum, cerebellum and flat bones of the skull called anencephali can be intermittently awake and exhibit several behavioural patterns of healthy newborns [20]. Ultrasonography has uncovered what was once withheld from view by nature during the 9 months of pregnancy, opening up a visual dimension which used only to start after birth. This premature unveiling has, amongst other things, triggered the habit of applying the phenomenon of ‘meaning attribution’ to the physiologically concealed motions of the fetus. Meaning attribution is a fundamental mechanism in postnatal life which helps us to care for our newborns and to deal with the complex nuances of social life in general. However, attributing the meaning associated with neonatal or even adult manifes-
9.6 Twins: Open to Mutual Communication
tations to the behaviour of the fetus, especially of the early fetus, is neither functional nor appropriate.
9.6 Twins: Open to Mutual Communication The ‘mystic’ union so often attributed to twins in postnatal life is frequently believed to date back to prenatal life. Intrapair stimulation, which is indeed limited to twins, is taken to mean various types of communication. Twin fetuses are considered as particularly interactive companions, relating to each other in all sorts of complex and sophisticated ways. Even kissing and fighting have been ‘scientifically’ described to occur in utero [21]. Apart from any other consideration, it is difficult to imagine how elaborate behavioural and emotional patterns such as kissing could be carried out across the membranes which separate the two different amniotic sacs in which 99% of all twin pregnancies are contained. These activities, like all complex emotional and social patterns, clearly only belong to later stages of postnatal life. Intrapair stimulation, which is a feature of twin pregnancies, seems to be considered relevant solely to support the belief that twins are highly communicative partners. As a consequence, all fetuses are regarded as potentially open to communication. However, although fetuses are rapidly and progressively preparing to enter into a social world, this preparation is geared towards the different stimuli and more ‘mature’ human beings that they will encounter in the extrauterine environment. Fetuses are simply not equipped for complex social interactions with other fetuses. Human communication is foreign and nonfunctional to the usually solitary 9 months of gestation. The severely premature infant is in many ways simply a fetus removed from its natural environment. Though certainly not numb to various forms of stimulation, including human contact, it clearly does not enjoy an intense and lively ‘social’ life. Most of the time, premature infants lie inert and hypotonic, crushed by gravity, dormant and barely responsive to all the commotion surrounding them. Even more so, pre-viable twin fetuses incapable of living autonomously outside the uterus can hardly be assumed to be fit for communication with each other at all.
9.7 Maternal Emotions and their Impact on the Twin Fetus The impact of maternal emotions on the fetus is a ‘hot’
105
topic amongst prenatal psychologists, who concentrate especially on ‘negative’ emotions such as anxiety and stress. The research on the impact of maternal emotions on the fetus is fraught with enormous and, for the moment, unresolved methodological problems. To mention but a few: our knowledge of the physical substrata of emotions is still primitive, our understanding of placental crossing is also limited, and our comprehension of the mechanisms regulating fetal behaviour is virtually nil. Yet twins have been thought of as a possible ‘ideal model’ to evaluate the impact of maternal emotions on the fetus [21]. According to this model, since both twins are equally and systematically affected by the same maternal emotion or by the same by-product of an emotion, this proves per se that behavioural alterations are caused by the emotion. The underlying assumption is that twins live in and share exactly the same intrauterine environment. Even assuming that maternal ‘stressors’, of whatever origin and kind, can reach the fetus, twin fetuses would only receive the same quality of such substances, never the same quantity. For instance, pregnancy-induced hypertension is a frequent complication in twin pregnancies [22]. However, this condition is not proven to have any causal relation to maternal anxiety. The opposite may well be true. Furthermore, pregnancy-induced hypertension does not impinge on both twins in the same way. Twin fetuses have different umbilical cords, an unequal distribution of placental mass; they float in different amounts of amniotic fluid and have differing blood flows. Their environment, including potential maternal ‘stressors’, can hardly be regarded as ‘equal’. Uterine contractility is particularly elevated in twin gestations. Uterine contractions have also been postulated to be the result of maternal anxiety and stress. According to this hypothesis, uterine contractions, by applying pressure on the amniotic sac, would produce increased tension in the amniotic fluid. This pressure would be perceived as disruptive by the fetus, which would then start to move more. However, even assuming that this ‘increased pressure’ hypothesis were true, the vast majority of twins are contained in separate amniotic sacs and the amniotic fluid is unequally distributed between the two. Pressure on sacs which are different from each other may well impact differently on each twin. Basically, all these hypotheses confound the lively fetal movements with anxiety-driven hyperactive states after birth [4]. Twin gestations, owing to the various
106
complications and many discomforts associated with them, are anxiety-provoking pregnancies. Already overanxious prospective mothers of twins should be spared the additional burden of having to try and suppress their emotions, which may often be impossible anyway, and whose harmful effects are in any case in doubt.
9.8 Bereavement in the Twin Fetus Anecdotal evidence from children and adults reporting a sense of loss which they can only explain in terms of a wish for a reunion with a dead co-twin fetus is accumulating. As a consequence, therapy groups for ‘prenatal mourners’ are proliferating. However, it is impossible to ascertain how much of this longing and loss is derived from real reminiscences of sensations felt in utero and how much is the result of later constructions belonging to postnatal life alone. The event of fetal loss, and especially of a late fetal loss, always has a deep impact on parents. It is not hard to imagine how this could reverberate in only too many ways on the surviving twin. The ‘missing’ twin fetus can easily become the vehicle for projection of all that has been missed in life and the inevitable dissatisfactions connected with this [23]. Only when proximity between the twins has become continuous and intrapair stimulation a consistent component of the intrauterine environment due to increasing crowding could one postulate that a surviving twin might feel some kind of ‘loss’. Again, one should not forget that the majority of twins live in separate environments. What a twin fetus might ‘miss’ would clearly not be a whole, distinct person, but just the stimulation, however strong, arising from the other twin fetus or perhaps some kind of ‘animal’ warmth deriving from the proximity of another living being. Though some kind of loss may be experienced by the survivor, for the time being, we cannot assess whether the death of a twin in utero remains forever embedded in the unconscious of a surviving twin [24 25].
References 1. Piontelli A (2008) Twins in the world. McMillan/Palgrave, New York 2. Galton F (1875) The history of twins as a criterion of the relative powers of nature and nurture. J Anthropol Inst Great Britain Ireland 5:391 406
9 Twin Fetuses and Twin Myths 3. Piontelli A, Bocconi L, Kustermann A et al (1997) Patterns of evoked behaviour in twin pregnancies during the first 22 weeks of gestation. HFR Prechtl (ed) Special issue. Early Hum Dev 50:39 46 4. Piontelli A (2002) Twins: from fetus to child. Routledge, London 5. Minkowski M (1922) Über frühzeitige Bewegungen, Reflex und muskuläre Reaktionen beim menschlichen Fötus and ihre Beziehungen zum fötalen Nerven und Muskelsystem. Schweiz Med Jahrbuch 52:721 724 and 751 755 6. Hooker D (1952) The prenatal origin of behaviour. University of Kansas Press, Lawrence, Kansas 7. Humphrey T (1978) Function of the nervous system during prenatal life. In: Uwe Stave (ed) Perinatal physiology. 2nd edn. Plenum Medical, New York 8. Moore KL, Persaud TVN (2007) The developing human: clinically oriented embryology. Saunders, Philadelphia 9. Baldwin VJ (1994) Pathology of multiple pregnancy. Springer, New York Berlin Heidelberg 10. Hepper PG (1992) Fetal psychology. An embryonic science. In: Nijhuis JG (ed) Fetal behaviour: developmental and perinatal aspects. Oxford University Press, Oxford 11. Glass L, Mckey MC (1988) From clocks to chaos. The rhythms of life. Princeton University Press, Princeton 12. Kelso JAS, Sholz JP, Schoner G (1986) Non equilibrium phase transitions in coordinated biological motion: critical fluctuations. Phys Lett 118:279 284 13. Benacerraf BR (1998) Ultrasound of fetal syndromes. Churchill Livingstone, Philadelphia 14. Piontelli A, Bocconi, L, Boschetto C et al (1999) Differences and similarities in the intra uterine behaviour of monozygotic and dizygotic twins. Twin Res 2:264 273 15. Bomsel Helmreich O, Al Mufti W (1991) Zigosité et déterminisme des grossesses gémellaires et multiples. In: Papiernick Berkhauer E, Pons JC (eds) Les grossesses multiples. Pons, Paris 16. Piontelli A (1995) Non shared intra uterine environmental factors: pre and post natal observations of twins. Invited Address, Fifteenth Annual Spring Meeting of the American Psychological Association, Santa Monica, 25 30 April 17. Fox H, Sebire N (2007) Pathology of the placenta. Saunders, Philadelphia 18. Parkes MJ (1991) Sleep and wakefulness: do they occur in utero? In: Hanson MA (ed) Fetal and neonatal brainstem. Cam bridge University Press, Cambridge 19. LeDoux J (1966) The emotional brain. Simon and Schuster, New York 20. Peiper A (1963) Cerebral functions in infancy and childhood. Consultants Bureau, New York 21. Arabin B, Gembruch U, van Eyck J (1995) Intrauterine be haviour. In: Keith LG (ed) Multiple pregnancy: epidemiology, gestation, and perinatal outcome. Parthenon, New York 22. Kochenour NK (1992) Obstetric management in multiple gestation. In: Fanaroff AA, Martin RJ (eds) Neonatal perinatal medicine. Mosby Year Book, St Louis, Missouri 23. Piontelli A (2000) Is there something wrong? The impact of technology in pregnancy. In: Raphael Leff J (ed) Perinatal loss and breakdown. IPA Publications, London 24. Piontelli A (1999) Twins in utero. In: Sandbank A (ed) Twin and triplet psychology. Routledge, London 25. Piontelli A (1992) From fetus to child. Routledge, London
Conclusions: Movement is Life
10
Keywords Hiccups • Breathing • Feeding • Glycaemia • Sleep • Apnoeas • Lung formation • Swallowing • Sucking • Startles • Generalized movements • Locomotion • Touch • Proprioception • Twins • Body schema • Vestibular system • Somatosensory homunculus • Perception • Facial expressions • Mirror neurons
This work has endeavoured to provide a basic description of how fetal movements arise and evolve. In describing this it has also tried to explain in what ways our prenatal movements may begin to shape us all. According to obstetrical standards, the first sign of life is the heartbeat, present in the embryo by 4 5 weeks. If the heartbeat is absent it means that pregnancy has come to a halt. If the heartbeat is present, at all subsequent checks it will always be the first movement the obstetrician will look for. However, if other fetal movements are detected first, the heart can wait. Movement means the fetus is alive. In this book I have tried to explain how fetal movements are the building blocks that lie at the foundation of many essential phenomena of our existence, including some higher functions of our brains (Fig. 10.1).
10.1 Fetal Movements: Varied and Varying Functions Some prenatal movements, such as hiccups, have an function that is obscure, or perhaps they have no function at all. Some authors postulate that hiccups may be just a remnant of our amphibian past. Others view it as a grounding for fetal breathing movements. However, the meaning of hiccups is still unclear. Hiccups are one of the first movements observed in the fetus and a phenomenon that continues throughout our existence. A. Piontelli, Development of Normal Fetal Movements. © Springer Verlag Italia 2010
Fig. 10.1 Behavioural phenomena: developmental trends. All be havioural phenomena discussed in this book, apart from facial expressions, clusters and the behaviour of twins, are recapitulated in this figure in order to give the reader an overall view of these phenomena and their developmental trends. GMs, General Move ments; St, Startles; FBM, Fetal Breathing Movements; Hic, Hic cups; Sw, Swallowing; LM, Localized Movements; Tw, Twitches; Cl, Clonuses
Given its high rate of recurrence during the early stages of pregnancy, its physiological function could be found in the initial stages of fetal life. Some other motions have an apparently clear and
108
exclusively preparatory role geared for postnatal life. We all have to be equipped for breathing before we take our first breath, or for feeding before we latch onto the breast. However, matters are not as simple as that. Although breathing has a long preparatory history during the 9 months preceding parturition, prenatal breathing is profoundly different from postnatal aerial respiration. Fetuses do not take in oxygen when they inhale or let out carbon dioxide when they exhale. These vital functions are carried out through the placenta and the umbilical cord. Breathing is the only movements influenced by glycaemia, increasing after a good meal and suppressed by hypoglycaemia. It is also one of the first movements to disappear when oxygen is required in ‘higher-ranking’, easily damaged regions such as the brain and the adrenals. Thus, unlike in postnatal life, fetal breathing is less immediately vital than other functions. Suspension of breathing within a reasonable, not yet precisely quantified span lasting possibly several hours rather than a few minutes, does not mean cerebral demise and death. Unlike during postnatal life, fetal breathing is not continuous but occurs in bouts. Except perhaps occasionally during the last stages of pregnancy, breathing movements can become continuous only when fetuses are in a terminal condition, and as such are taken as an ominous sign. Breathing bouts are separated by apnoeic intervals of various duration. Especially in the preterm infant, respiration tends to be very irregular, with episodes of periodic breathing defined as ‘pauses in respiratory movements that last for up to 20 seconds alternating with breathing’ [1]. Periodic breathing is also observed in neonates and in the adult at high altitude or suffering from serious conditions such as severe congestive heart failure. Periodic breathing and also more prolonged apnoeas are the norm in the fetus. In the premature and the neonate, severe apnoeas require assisted ventilation, and could even underlie the sudden infant death syndrome, or cot death. Sleep-related apnoea is increasingly being recognized and taken into account as a major health problem not only in adults, but also in children and adolescents [2]. Fetuses suspend their main and most frequent form of respiration, paradoxical breathing, when at rest and instead occasionally display sporadic and short bouts of shallow and regular respiration interspersed by long apnoeic intervals. Some cases of sleep-related apnoea may well have their remote origin in a proneness to cease breathing when in deep phases of sleep and rest. On the other hand, breathing movements fulfil other
10 Conclusions: Movement is Life
important functions at various stages during pregnancy, as has been revealed by studies of animals as well as impaired fetuses. For instance, by squeezing the thorax, they contribute to lung formation and maturation in various ways ranging from the secretion of growth factors to the enlargement and development of alveolar structures or the expulsion of limited amounts of surfactant in the amniotic fluid. Lung development in the fetus is seriously impaired by consistent diminution or cessation of breathing movements of whatever origin. Swallowing is another function apparently clearly geared towards postnatal life. Neonates need to be able to suck and swallow, and to coordinate these ingestive activities with breathing. Proper sucking is barely present during the first half of pregnancy. Although fetuses increasingly begin to alternate some swallowing bouts with breathing movements, the fine-tuned sequences necessary for milk intake are not yet reached by 25 weeks. Severely premature infants do not latch onto or suck from the breast, and if fed orally tend to stop breathing, regurgitate and easily choke. The most severely premature infants have to be tube-fed. Swallowing movements in the fetus, just like breathing ones, are not geared solely towards postnatal nutrition, but have other vital additional prenatal functions. For instance, swallowing contributes to the regulation of the amniotic fluid. Additionally, some nutrients and growth factors are found in the amniotic fluid. By swallowing and ingesting them, fetuses contribute both to nutrition and to gastroenteric development. Some other movements, such as startles, seem to have an exclusively transitory role and one restricted to early fetal life. For a brief period in time startles set general movements in motion. However, startles continue to fire even beyond this stage and are never linked to general movements in a one-to-one manner. Startles are found to accumulate at both ends of cycles of rest, precipitating or heralding a change, but they can also emerge more sparsely at apparently any time. We do not know why this is so. Startles may have additional unknown functions, or they may just be vestigial relics which continue to emerge aimlessly and ‘obtusely’ beyond their time. Furthermore, we do not know whether those startles which occasionally accompany the initial stages of falling asleep even in the adult are linked in some way with prenatal startles. Even more disquietingly, we do not know whether some forms of epilepsy especially so-called startle epilepsy could be related to prenatal startles in any way. Startles set general move-
10.2 Shaping a sense of our Boundaries
ments in motion, and general movements with their burst-like global quality are reminiscent of generalized seizures. However, generalized movements are differently performed, gracefully and smoothly. Furthermore, they are performed in a watery medium which helps to give them a fluent quality absent from every movement performed on earth. However, could it be that later in life some disorderly startled movements trigger some equally disordered generalized movements with all the characteristics of proper generalized seizures? We have no answer to these and other questions, but knowing how and when some prenatal motions such as breathing and swallowing movements or startles emerge may help to pose questions which could be useful for the comprehension of postnatal phenomena whether in the premature, the neonate at term or even in the adult. Various explanations have been given for other motions more properly related to movement as we designate it in life after birth, and specifically to general and localized movements. According to some authors, generalized movements, amongst their many other functions, prevent adhesion of the fetal skin to the uterine surface [3]. Adhesion would cause tearing and ulceration of the extremely fragile fetal skin, similarly to the ‘decubitus plagues’ seen in bed-ridden, non-mobilized patients after birth. The fetal skin is initially very thin, consisting of only two layers. Between 12 and 20 weeks the fetal epithelium increases from two to five layers. However, the stratum corneum (horny layer), the keratinized part of the epidermis, is not yet formed. Keratinization begins after 20 weeks and spreads fairly rapidly. By 24 weeks keratinized cells cover almost all areas of the body and the fetal skin nearly corresponds to the skin of the neonate [4]. Roughly speaking, the keratin layer the outer layer of the skin forms an impervious barrier between the outer surface of the body and the environment, protecting it from damage, infection and water loss. Despite these changes at 25 weeks the fetal skin is still very vulnerable. Doctors caring for severely premature infants are well aware of this vulnerability. Even misplaced, carelessly removed, too adhesive, or infrequently changed plasters can leave lifelong scars. The non-adhesion explanation certainly justifies one element of general movements, but clearly not all its facets. Again, animal studies and impaired fetuses have shown that movements are important for muscle differentiation and elongation, and for the prevention of
109
muscular atrophy deriving from disuse. Additionally, early fetal movements play a role in shaping the skeletal system, and in the development of the nervous system, including the delineation of sleep [3]. Even the shape of the pharynx, with cascade effects on future functions including phonation, would be affected by fetal head movements, especially head retroflexions [5]. Other preparatory functions, such as locomotion, are fulfilled as well. ‘Motor learning’, commonly called ‘muscle memory’, is a well-known phenomenon and one that occurs even in lower organisms [6, 7]. Motor learning provokes changes and refinements through time, thus equipping fetuses with the competence to perform increasingly refined movements, be it grasping or starting to use their hands as tools or exercise their legs to perform alternate stepping eventually resulting, months later, in the capacity to walk in the erect stance. In other words, fetal movements perform many and varied crucial roles.
10.2 Shaping a Sense of our Boundaries By the time we are born we all have a me/not me discernment and at least a sense of our boundaries. Touch and proprioception are the main shapers of this sense. Newborn twins can beautifully illustrate the me/not me distinction. Infant twins are often placed near each other for several reasons ranging from comfort or convenience to pure delight and fun. Many twins touch their co-twin, or suck their nose or chin. Whoever is watching them usually exclaims in a thrilled voice, ‘he believes it’s his own nose’ or ‘she thinks she is sucking her own chin.’ However, if one records these events on tape and then plays it back in slow motion or frame by frame, an expression of surprise or puzzlement can be noted which may last only a few seconds, but nevertheless is there. Both the perpetrator and the receiver seem to distinguish that the action was not self-contained: that it arose from or reached something beyond its own body, something extraneous to its self. Albeit not consciously, twins realize that they touched or were touched (or sucked) by an external entity. Touch is unique amongst other senses in having a dual aspect: when anyone touches, he or she is also simultaneously being touched. Touch is both an active movement and passive receptivity. Even by touching ourselves we both touch and are touched by whatever part of our body we have contacted [8]. The twin who
110
touches the other twin senses that it has touched an extraneous element which has also touched it (the first twin). Aristotle considered touch the most indispensable of all the senses and the basis for the others, and yet he viewed touch as hierarchically less ‘noble’ and more ‘animal’ compared to the nobleness and primacy of vision. Centuries later we still live in a world permeated and infatuated with vision. Infants are not born blind, although fetuses do not rely on vision, but on ‘proximal’ senses such as touch and proprioception. Besides other considerations, in the narrow space of the uterus vision with its correlated sense of distance would be unnecessary anyway.
10.3 Building a Body Schema and a Proto-Sense of Self In addition to a sense of our boundaries, largely built through touch, we are also born with a body schema. Since the definition of body schema is complex and hotly debated, the interested reader should refer to specialist publications for further clarification [9, 10]. The term ‘body schema’ and some of its correlated properties will be used here in a simple way, meaning an unconscious body awareness which includes the changing position of the body in space and all its postures. Our body schema is not formed once and for all, but is constantly refined by movements, changing bodily proportions, and sensory inputs. However, the overall body schema, together with an (albeit rudimentary) sense of constancy in the doer, although unconsciously, is also maintained through space and time. Without this constancy we would probably be at a loss. The phenomenon known as the phantom limb, though still far from clarified, probably illustrates this permanence: over 95% of amputees feel sensations where the arm or leg used to be, as if it were still there. After birth the body schema becomes interpersonal, as the me/not me distinction involves other people as well. Whether in the late stages of pregnancy twins may acquire an interpersonal notion of their co-twin is doubtful. Nevertheless, the question is debated and needs definitive proof one way or the other. A body schema lies at the foundation of a proper sense of self which will evolve fully after birth. However, our sense of self has its remote roots in our intrauterine past and is shaped by our movements from
10 Conclusions: Movement is Life
the early prenatal stages. Movement, and the registration of that movement in a developing proprioceptive system a system which registers its own self-movement contributes to the way we come to be conscious of ourselves, to communicate with others and to live in the postnatal world [11]. Apart from the heartbeat, the first fetal movements start at around 7 weeks. We do not know whether movements and sensitivity begin at the same time. The testing carried out by Hooker on pre-agonal aborted fetuses seemed to indicate that they do. Nevertheless, by 9 weeks, as twin fetuses again illustrate, all fetuses are open to tactile and proprioceptive sensitivity. From then on movement and sensing start to be inextricably linked, and even the simplest movement provides fetuses with a sensory feedback. The division made here between localized movements as providers of tactile sensitivity and generalized movements of proprioception is rather arbitrary. Localized movements invariably provide proprioceptive feedback and generalized movements provide tactile feedback. Each type of movement can be considered as the main source of the sensitivity associated here with it. Generalized movements start before localized ones. Localized movements, with their goal-directedness, are used principally for touching sensitive areas of the body and increasingly for ‘exploratory’ contacts with various components of the intrauterine environment. Although they anticipate the voluntary movements to be performed after birth, they do not provide a thorough sense of the body in space. Turbulent general movements, on the other hand, with their often complete turning round (especially during the early stages of pregnancy), endow fetuses with a complete map of their bodies and with a sense of their bodies in space. During their tossing and turning the entire fetal body is touched, stimulated and moved. In addition, proprioception provides a constant adjustment of bodily spatial orientation. Presumably a global sense of our body in space begins to be built through generalized movements. The vestibular system in the inner ear, by supplying continuous information of the movements and orientation of our head in space, should also participate in the feedback information fetuses receive. When and whether the vestibular system starts functioning in utero is still open to debate. The onset of the vestibular function has been investigated especially in fetal rats [12]. In animals the onset of the vestibular function has been shown to occur before birth. Its supporting structures
10.4 Forming the Cortical Homunculus and its Curious Layout?
are said to develop ‘early’ during pregnancy, although how early is not specified. In life after birth the vestibular apparatus functions in close connection with vision, and in concert with proprioception, forming a control system that allows us to maintain an erect stance as well as a broad range of stable and unstable postures. Those unfortunate individuals whose vestibular apparatus for whatever reason does not function adequately, have to keep their eyes open and maintain a very rigid relationship between the head and the body in order to orient themselves in space. Unless they do so they may fall down, experiencing vertigo, nausea and a strong sense of bodily unsteadiness. Vestibular stimulation in utero is provided especially by the pregnant mother’s body movements, which were found to be critical for the development of the vestibular system in animals. It may well be that, additionally and apart from maternal movements, generalized movements with frequent head motions and accelerations may also be critical for vestibular development. Neonates like to be rocked and cradled, all motions reminiscent of the movements experienced in utero, whether originated by their mother’s movements or by their own. Local movements can be differentiated from general movements. From the start local movements are directed towards particularly sensitive areas of the body abundantly innervated by sensory fibres. The face is the main target, and the face is touched with the hands. Less sensitive areas such as the thorax and the stomach are hardly ever touched. The earliest contacts start with light scratching motions in the region separating the face from the cranium, where trigeminal innervation ends. Occasionally the hand may also scratch the cranium at the border with the insensitive area of the fontanelles. All these are barely sensitive or non-sensitive areas bordering with very sensitive ones. Could it be that fetuses sense the demarcation and from then on mostly direct their ‘explorations’ towards areas that they can sense? Soon the barely sensitive areas are almost no longer touched, and the hands are pressed against the face. Increasingly, and predominantly, the lips are touched. The nose is generally neglected and, as stated, only later on are the hands also pressed against the eyes. Hand face contacts continue throughout pregnancy, and well beyond the first 25 weeks. Another important point of contact is the soles of the feet. Fetal feet are disproportionately big and the soles are another sensitive area. Feet are very frequently
111
pushed against the uterine wall and touch it even when at rest, thus providing a lot of tactile feedback. Around 16 20 weeks, when the most sensitive areas of the body have been actively and repeatedly touched, the fetus begins to make contact with components of the intrauterine environment as well. Such ‘explorations’ are made with the hands, and increasingly also with the tongue. At around 22 25 weeks fetuses start licking their environment. Neonates do just the same, using their tongue for ‘oral exploration’ in order to explore objects that are handed to them or placed near them, as in the in utero case of a co-twin’s face and nose. Oral exploration is prominent in the first few postnatal months, and though it becomes subordinate to visual exploration by the middle of the first year remains a common behaviour throughout childhood [13]. Could it originate from the first licking movements in utero?
10.4 Forming the Cortical Homunculus and its Curious Layout? The so-called sensory cortical homunculus within the primary parietal somatosensory cortex is remote from the young fetus which is just beginning to form a cortex. The sensory cortical homunculus is a physical representation of the sense and motor information received from the rest of the body, a mapping of body surfaces in the brain. The resulting image is a disfigured human (homunculus means small man in Latin) with huge hands, lips and face in comparison to the rest of the body which is relatively undersized. In the distorted ‘homunculus’ the hands, face, mouth and lips are all over-represented. The tongue is also over-represented, and the area corresponding to the feet is quite large (Fig. 10.2). These are the very parts of the body that the fetus begins to touch or stimulate with touch during its goal-directed, localized movements. Looking at the somatosensory homunculus, we can note other interesting facts. While the hand and the face are over-represented, the cranium is not. The homunculus almost looks like an anencephalus, without the cranial vault. Its face terminates where trigeminal innervation stops and where fetal touching or scratching also stops. The eyes and the nose are also small compared to the huge area of the lips. As has been mentioned, fetuses barely touch their eye-sockets during the first half of pregnancy. The nose, too, is touched
112
10 Conclusions: Movement is Life
the phenomenon of the phantom limb. Although a good deal of neurological development is genetically coded and organizes itself in predesigned circuits, at least some of that development requires sensory and motor input the very input provided by prenatal movements [15, 16]. Prenatal movements may be important building blocks in the development and the shaping of our somatosensory cortex, and possibly of other cortical areas as well.
10.5 Building on Expressive Repertoire
Fig. 10.2 The sensory cortical homunculus. The cortical ho munculus was described by Penfield and Broadly in 1937 [21] and Penfield and Rasmussen in 1950 [22]. The ‘homunculus’, located within the primary parietal somatosensory cortex, is a physical representation of the sensory and motor information re ceived from the rest of the body a mapping of body surfaces in the brain. The resulting image is a disfigured human with huge hands, lips and face in comparison to the rest of the body, which is relatively undersized. The homunculus represents the relative proportions of the sources of fetal sensations
just fleetingly when fetuses make contact with the rest of their face. Other authors have wondered why the somatosensory homunculus has hands next to face and genitals next to feet. They hypothesized that this curious layout may reflect the classical ‘fetal position’ in utero [14]. Fetuses, especially beyond the first half of pregnancy, become more and more cramped for space and increasingly take up the fetal position, resulting in their crossed feet being in contact with their genitals. Of course the over-represented areas are sensitive areas in life after birth as well. However, we do not touch our faces as much as fetuses do. When we scratch our heads to express perplexity, we do not do so only within the limits of trigeminal innervation. It would not be unreasonable to postulate that prenatal sensory feedback may play a role in the so-called ‘wiring up’ of the brain, including our somatosensory map. This does not mean that our somatosensory homunculi may not also be hard-wired. Those unfortunate people born without one or even all limbs still report
Besides the above-discussed possible functions of localized and general movements, another aspect of fetal life which has been largely neglected is facial expressions. Only fairly recently have 3D and, especially, 4D ultrasonography given us the possibility of studying them. Infants have been shown to be capable of imitation from birth [17]. Fetuses do not use vision while in utero and yet they do imitate some facial expression in the adult soon after birth. Relevant to fetuses and to neonates is what is called cross-modal integration the integration of information from different sensory modalities. In order to imitate a facial gesture such as tongue protrusion or mouth opening, for example, it is necessary for the infant to translate its experience from one sensory modality, vision, into another sensory modality, proprioception. At birth the infant possesses an already organized innate body schema, and this organization, built through movement and through movement feedback, informs perception. Furthermore, perception is intermodal from the start: experience in one sense ‘educates’ other sensory modalities [18, 19]. As reported in this book, fetuses start practising facial gestures and expressions close to mid-pregnancy. From then on fetuses display facial expressions that we would classify as smiling, or crying, or even silent vocalizations, and they certainly stick out their tongues. One could say that facial expressions which have long been practised in utero are transferred to the visual modality at birth. Infants ‘recognize’ with sight what they have long rehearsed in utero. Is this response pure imitation or some form of recognition? The practising of facial expression and silent vocalizations in utero may be relevant for another implication. Through facial expressions we can read the inten-
References
tions and states of the other. If neonates were born without an expressive repertoire, their caregivers would be at loss in trying to understand and respond to their needs. The pre-verbal child needs to be able to produce facial expressions in order to express its requirements. If an infant cries, the mother takes action and picks it up, cradles, feeds or puts it down to sleep according to various characteristics associated with the cry. The infant smile is a powerful tool to captivate the attention of the caregiver and elicit a sympathetic response driving him or her to interact with the infant. The same can be said with infant vocalizations which stir the mother to engage in a ‘proto-conversation’. Tactile and kinetic modes can be added to the proto-conversation [20]. In this dialogue, mothers talk in ‘baby talk’ or ‘motherese’, more properly called ‘child-directed speech’. The intonation of baby talk is different from that of usual adult speech, being high pitched, with short and simple words, repetitions and many glissando variations. Baby talk contributes to mental development and to the development of speech. Although fetuses are clearly a long way away from speech, when engaged in displaying facial expressions they may be paving the way for future language as well as for other future social interactions. In contrast to many other movements, facial expressions practised in utero may have an entirely anticipatory role, and a primary one for the start of intersubjectivity the entry in the postnatal social world.
References 1. Tucker Blackburn S (2007) Maternal, fetal, and neonatal phys iology, 3rd edn. Saunders, St Louis, Missouri 2. Stores G, Wiggs L (2001) Sleep disturbance: a serious, wide spread, yet neglected problem in disorders of development. In: Stores G, Wiggs L (eds) Sleep disturbance in children and adolescents with disorders of development: its significance and management. MacKeith Press, London 3. Brown JK, Omar T, O’Regan M (1997) Brain development and the development of tone and movement. In: Connolly KJ, Forssberg H (eds) Neurophysiology of motor development. MacKeith Press, London
113 4. Coulombe PA, Omary MB (2002) ‘Hard’ and ‘soft’ principles defining the structure, function and regulation of keratin fil aments. Curr Opin Cell Biol 14:110 122 5. Jeffrey N (2005) Cranial base angulation and growth of the human fetal pharynx. Anat Rec A 28:491 499 6. Li CS, Padoa Schioppa C, Bizzi E (2001) Neuronal correlates of motor performance and motor learning in the primary motor cortex of monkeys adapting to an external force field. Neuron 30:593 607 7. Kandel ER (2009) The biology of memory: a forty year per spective. J Neurosci 29:12748 12756 8. Paterson M (2007) The senses of touch. Berg Publishers, Oxford 9. De Preester H, Knockaert V (2005) Body Image and Body Schema. John Benjamin Publishing Co, Amsterdam. 10. Pfeifer R (2006) How the Body Shapes the Way We Think: A New View of Intelligence. MIT Press (Bradford Books), Cambridge, MA 11. Gallagher S (2006) How the body shapes the mind. Oxford University Press, Oxford 12. Ronca AE , Fritzsch B, Alberts JR, Bruce LL (2000) Effects of microgravity on vestibular development and function in rats: genetics and environment. Korean J Biol Sci 4:215 221 13. Jones SS (2006) Exploration or imitation? The effect of music on 4 week old infants’ tongue protrusions. Infant Behav Dev 29:126 130 14. Farah MJ (1998) Why does the somatosensory homunculus have hands next to face and feet next to genitals? A hypothesis. Neural Comput 10:1983 1985 15. Vanderhaegen P, Lu Q, Prakash N et al (2000) A mapping label required for normal scale of body representation in the cortex. Nature Neurosci 3:358 365 16. Strobel G (2000) Genes or environment: what shapes the sensory homunculus? Focus. News from Harvard Medical, Dental, and Public Health School. April 7:1 5 17. Meltzoff AN, Borton RW (1979) Intermodal matching by human neonates. Nature 282:403 404 18. Schweinle A, Wilcox T (2007) Intermodal perception and physical reasoning in young infants. Infant Behav Dev 27:246 265 19. Schmuckler Mark A, Jewell DT (2006) Infants’ visual pro prioceptive intermodal perception with imperfect contingency information. Dev Psychobiol 49:387 398 20. Field T (2003) Touch. MIT Press, Cambridge, Massachusetts 21. Penfield, Boldrey E (1937) Somatic motor and sensory rep resentation in the cerebral cortex of man as studied by electrical stimulation. Brain: 60: 389 443. 22. Penfield W, Rasmussen T (1950) The cerebral cortex of man. A clinical study of localization of function. Macmillan, New York
Glossary
A Acini Small sac-like dilations which in the lungs are at the end of the anatomic terminal units consisting of alveolar ducts and alveoli stemming from a terminal bronchiole. Actin A protein which in combination with myosin is responsible for muscle contraction. Active sleep Considered to be the equivalent of REM sleep in the neonate. During active sleep, sucking motions, twitches, smiles, frowns, irregular breathing and gross limb movements (the opposite of the typical REM sleep paralysis seen at later ages) are noticed. Adaptive function Adaptation is a process whereby an o ganism becomes better suited to its present habitat. Both the habitat of the fetus and the fetus itself undergo rapid changes. Therefore it is possible for a function to be adaptive to a particular stage, but not to a later or earlier one. Alveoli Small outpouchings along the walls of the alveolar sacs and alveolar ducts; through these walls gas exchange takes place between alveolar gas and the pulmonary capillary bed. Anencephaly A lethal malformation characterized by absence of the cranial vault and of most or all the cerebral hemispheres of the brain. Anticipatory functions Activities required for postnatal development, and prerequisites for postnatal functions. Aphasia Inability to use or understand (spoken or written) language due to a brain lesion. Apnoea Absence of breathing (respiration). Apnoeic intervals Used in this book to indicate pauses (apnoea) between two bouts of breathing. In the fetus apnoeic intervals can last from several seconds to many minutes. Arterial chemoreceptor Sense organ (such as the carotid body, the aortic bodies, or the glomus jugulare) which is sensitive to chemical changes in the blood stream, especially reduced oxygen content, and reflexively increases both respiration and blood pressure.
116
Glossary
Ascending reticular formation Diffuse system of ascending fibres arising in the reticular formation that is responsible for the control of states of consciousness. Athetoid Used in this book to indicate slow, sinuous movements of the fingers reminiscent of the motions of octopus tentacles. Autonomic nervous system (or involuntary nervous system) A part of the nervous system that regulates key involuntary functions of the body, including the activity of the heart muscle; of the smooth muscles, including the muscles of the intestinal tract; and of the glands. The autonomic nervous system has two divisions: the sympathetic nervous system, which accelerates the heart rate, constricts blood vessels, and raises blood pressure, and the parasympathetic nervous system, which slows the heart rate, increases intestinal and gland activity, and relaxes sphincter muscles. Axial skeleton The skeleton of the head and trunk. Axon The efferent process of a nerve cell. Each nerve cell has a long fibre (the axon), through which it conducts impulses away from its body and passes them to other target cells or to effector organs (i.e. organs that produce an effect such as contraction or secretion) in response to nerve stimulation. B Behavioural states One can speak of behavioural states in the fetus when particular conditions of several variables recur in specific, fixed combinations and these combinations are fixed in time. If one of the criterion variables changes to a new state, the other variables follow suit, either simultaneously or nearly so. Following these criteria, four distinct patterns can be recognized in the fetus from around 36 to 40 weeks: State 1F: Quiescence, which can be regularly interrupted by brief, gross body movements, mostly startles. No eye movements. Fetal heart rate stable with isolated accelerations related to fetal movements. State 2F: Frequent and periodic gross body movements. Eye movements continuous and rapid. FHR variable, with frequent accelerations associated with movements. State 3F: No gross body movements. Continuous eye movements. Fetal heart rate stable and with no accelerations. State 4F: V Vigorous, continuous bodily activity. Eye movements present. Fetal heart rate unstable, with large and long-lasting accelerations, often fused into sustained tachycardia. Blood flows Used in this book to mean placental blood flows. Blood flows between mother and fetus through the placenta, supplying oxygen and nutrients to the fetus and carrying away fetal waste products. Body image An internal representation in the conscious experience of visual, tactile and motor information of corporeal origin. Implies a conscious awareness of one’s body. Body scheme or schema Unconscious awareness of one’s body and the position of its parts or non-conscious performance of the body.
117
Bradyarrhythmia Any disturbance in the heart rhythm. Bradycardia Slowness of the heartbeat. In the fetus and the premature, heart rate less than 120 beats per minute. Generally associated with hypoxia. Brainstem (also called encephalic trunk) The stem-like part of the base of the brain that is connected to the spinal cord. Most cranial nerves originate in the brainstem. Besides being a relay station between the brain and the spinal cord, the brainstem controls vital bodily functions such as breathing, swallowing, heart rate, blood pressure, sleep/wakefulness and consciousness. The brainstem consists of the midbrain, pons and medulla oblongata. Broca’s area Region of the brain (generally in the left frontal brain) responsible for the motor or executive aspects of speech. C Canalicular phase The phase of prenatal lung development, lasting in different parts of the lungs from the 16th 17th gestational week to the 26th week or later. Basic structures of the gas-exchanging parts of the lungs develop and become vascular, and primordial alveoli called terminal saccules begin to form, enabling respiration to begin. The canalicular phase is followed by the terminal saccular phase, which starts from the 26th week or later in different parts of the lung and lasting until near birth. Capillary bed The minute vessels that connect the arterioles and venules (minute arterial and venous branches proximal to the capillaries), forming a network in nearly all parts of the body. Their walls act as semipermeable membranes for the exchange of various substances, including fluids, between the blood and tissue fluid. Central pattern generators Generally small, autonomous neural networks endogenously producing patterned outputs. Central pattern generators underlie the production of most rhythmic motor patterns. Cerebellum The portion of the brain located in the back of the head (the so-called posterior cranial fossa) between the cerebrum and the brainstem. It is principally involved in the control and coordination of voluntary and involuntary movements as well as balance. Chromosomal or genetic testing Done to determine whether the fetus has certain hereditary or spontaneous genetic disorders. Ultrasonography and certain blood tests, together with maternal age, can give an indication of increased risk for some genetic defects. However, certainty can only be reached with invasive tests: basically, chorionic villus sampling and amniocentesis. Clonuses A series of rapid, rhythmic, alternating contractions and relaxations of one or more muscles. Club foot A common malformation of the foot (or feet), evident at birth, in which the foot is turned in sharply.
118
Glossary
Clusters Used in this book to denote temporal areas (clusters) within which some activities but not others emerge. Colour flow Doppler imaging Colour flow Doppler uses standard ultrasonographic methods to produce a picture of a blood vessel. A computer converts the Doppler effects into colours that are overlaid on the image of the blood vessel and that represent the speed and direction of blood flow through the vessel. Colour flow Doppler is used to check the health of a fetus. Blood flows can be checked in the umbilical cord, across the placenta, or in the heart and brain of the fetus. This test can show whether the fetus is getting enough oxygen and nutrients. Conceptional age Time elapsed between the day of conception and the day of delivery. In pregnancies resulting from assisted reproductive technologies, the date of fertilization or implantation can be exactly known. However, even in these pregnancies, gestational age is usually conventionally applied by adding 2 weeks to the conceptional age. Conditioning A form of learning in which a stimulus initially incapable of evoking a certain response acquires the ability to do so by repeated pairing with another stimulus that does elicit the response. Corpus callosum The great transverse arched band of nervous tissue connecting the two cerebral hemispheres, thus allowing communication between the right and left sides of the brain. Corrected age (frequently called adjusted age) The term used to describe children up to 3 years of age who were born prematurely. Corrected age is the age of the child as it would have been had the child been born on the expected delivery date. It is estimated by subtracting from the actual age the difference between 40 weeks and the number of weeks of gestation at which the child was born; e.g. for a child born at 28 weeks’ gestation, the corrected age will be actual age since birth less 12 weeks (40 28 = 12). Cranial nerves The nerves of the brain, which emerge from or enter the skull (the cranium), as opposed to the spinal nerves, which emerge from the vertebral column. There are 12 cranial nerves, each of which is accorded a roman numeral and a name. Four are purely motor, four are mixed (with motor and sensory functions), and two (the olfactory and the optic) are not properly cranial nerves as they are composed of fibres belonging to the central nervous system. Very simply listed, they are: I Olfactory (special sense: smell). II Optic (special sense: sight). III Oculomotor (motor: movement of the eyeball, lens and papillary sphincter). IV Trochlear (motor: superior oblique muscle of the eye). V Trigeminal (mixed sensory: ophthalmic, maxillary and mandible areas; motor: controls muscles of mastication). VI Abducens (motor: moves the eye outwards). VII Facial (mixed motor: controls muscles of the face, scalp, ear; controls salivary glands; sensory: receives taste sensation from the anterior two-thirds of the tongue). VIII Acoustic (special sense: hearing and equilibrium). IX Glossopharyngeal (sensations from the pharynx and the posterior one-third of the tongue). X Vagus (mixed sensory and motor to thoracic and abdominal viscera). XI Accessory (motor: permits movements of head and shoulders). XII Hypoglossal (motor: controls muscles of the tongue). Cross-modal integration Integration of information from different sensory modalities.
119
Crown–rump length Measurement of the length of human embryos and fetuses from the top of the head (crown) to the bottom of the buttocks (rump). D Deafferentation Elimination or interruption of the afferent nerve impulses (carrying impulses toward the central nervous system). Decerebrate posture Body posture that involves the arms and legs being held straight out, the toes being pointed downward, and the head and neck being arched backwards. Decerebrate preparation Experimental animal that has had cerebral function interrupted. Dendrite The short, arm-like process of a neuron which carries the impulse to the cell body. It is usually branched like a tree. Desynchronized EEG Fast and low-voltage electroencephalographic (EEG) activity resembling that of wakefulness. Dextrocardia Presence of the heart in the right hemithorax. Dichorionic twins Twins with two separate placentas. All dizygotic (or so-called fraternal) and roughly 30% of all monozygotic (or so-called identical) twins are dichorionic. Diencephalon The part of the brain between the telencephalon and the mesencephalon, including the thalami and most of the third ventricle. Dynamical systems Used in this book as a concept to help understand fetal rest activity cycles. During cycles of rest at mid-cycle, fetuses are impervious to various forms of stimulation or perturbation (be it a startle or a kick from a co-twin). These impervious time regions can be considered periodic fixed points. However, cycles of rest progressively shift again towards a cycle of activity. When nearing the phase-shift, fetuses become increasingly open to perturbations that can hasten or precipitate the change. Similarly, at the end or tail of a cycle of activity fetuses again become progressively impervious to perturbations and are ‘attracted’ towards fixed points. E Ectopic implantation Embedding of the embryo in the wrong place. Ectopic pregnancy A pregnancy taking place outside the uterus as a result of an ectopic implantation. The main risk is rupture at the site of implantation with consequent potentially life-threatening bleeding. Electric coupling Two or more circuits are said to be electrically coupled if energy can transfer electrically from one to the other.
120
Glossary
Embryonic phase of lung development The earliest period of lung development, lasting from the 3rd to the 6th week after conception. During this period a primordial trachea and larynx, a respiratory diverticulum (lung bud), and finally bronchial buds are formed. Endogenous Developing or originating within the organism, or arising from causes within the organism. Epidermal ridges Minute corrugations of the skin. They compose a sculpturing, termed dermatoglyphics, which characterizes the surface of the palms of the hands and the soles of the feet. These areas lack hair and sebaceous glands (which produce especially fat), but sweat glands (as well as nerve endings) are numerous. Epiglottis The lid-like cartilaginous structure overhanging the entrance to the larynx and serving to prevent food from entering the larynx and trachea during swallowing. Epiphenomenon A secondary phenomenon that is a by-product of another phenomenon. Epithelium The cellular layer that covers internal and external organs of the body, including the skin, blood vessels, body cavities and glands. The epithelium is composed of contiguous cells with a minimum of intercellular substance. Ethogram A list of all the different kinds of behaviours an animal species can exhibit. Eupnoeic breathing Normal or easy respiration. Extreme prematurity Term applied to any infant born before 28 completed weeks of gestation. F Fast twitch muscle fibres Muscle fibres suited for short-duration, intense activity. Widely prevalent at birth. W Fetal biophysical profile Ultrasonographic evaluation of fetal state before or near parturition based on five variables: fetal heart rate, breathing movements, general movements, muscle tone and amniotic fluid volume. G Gastro-oesophageal reflux Reflux (regurgitation) of the stomach contents into the oesophagus. If repeated it can cause oesophagitis, an inflammation of the oesophagus that can result in damage to the mucosa, with erosion and ulceration. Stricture, scarring and occasionally perforation may occur in severe cases. General movements Periodic bursts of whole-body activity. Gestalt Used in this book as the visual recognition of an overall pattern of behaviour.
121
Gestational age The time measured from the first day of a woman’s last menstrual cycle to the current date (including the date of the delivery). The first day of the last menstrual period occurs approximately 2 weeks before ovulation and 3 weeks before implantation of the blastocyst . Most women know when their last period began but not when ovulation occurred. As long as the menstrual dates are remembered accurately, the method of estimating the date of delivery is accurate, and it is widely used. Inaccuracies are generally minor: in the order of 4 6 days. Gestational age is expressed as completed weeks, and is usually 2 weeks shorter than conceptional age. Glossopharyngeal nerve (cranial nerve IX) A mixed cranial nerve closely linked with the vagus nerve. Its motor fibres innervate muscles of the soft palate and pharynx; sensory fibres supply chiefly the posterior third of the tongue and its taste buds, and the pharynx. Its parasympathetic fibres innervate principally the parotid gland. H Habituation Extinction or decrease of a conditioned reflex over time by repetition of the conditioned stimulus. Handedness The property of using one hand more than the other. Haptic Refers to the sense of touch in all its forms. Hard palate The first section of the bony part of the mouth, located in front of the soft palate. Hedonic Used in this book to mean pleasure/displeasure-giving, or having a pleasurable/non-pleasurable colouring. Homeostasis The maintenance of steady states in the organism by coordinated physiological processes or feedback mechanisms. Hyperexplexia (also called startle disease, hyperekplexia) Congenital condition of exaggerated startle reactions. Hyperreflexia Exaggeration of reflexes. Hypertonus Increased muscular tonus. Hypophysis (or pituitary gland) A small gland located in the middle of the base of the brain. It consists of two parts, the anterior pituitary (adenohypophysis) and the posterior pituitary (neurohypophysis), and is connected and functionally related to the hypothalamus. These two organs work together and regulate almost all metabolic processes in the body by means of various hormones and other chemical messengers. Hypothalamus Region of the diencephalon forming the floor of the third ventricle and including neighbouring associated nuclei. The hypothalamus secretes substances that influence the function of the pituitary and other glands and is involved in the control of body temperature, hunger, thirst and other processes that regulate body equilibrium.
122
Glossary
Hypotonic urine Up to near parturition fetuses normally produce hypotonic, i.e. diluted or less concentrated urine. Hypoxia Oxygen deficiency. I Intracranial hypertension Increased intracranial pressure. In the fetus this is generally due to an accumulation of cerebrospinal fluid, less frequently to an accumulation of blood from any kind of haemorrhaging, or to brain tumours. Intrathoracic (or pleural) pressure The pressure between the visceral pleura (the portion of the pleura investing the lungs and lining their fissures, completely separating the different lobes) and the parietal pleura (the portion of the pleura lining the walls of the thoracic cavity) in the pleural cavity. K Keratin layer (or stratum corneum or horny layer) The upper layer of the skin, characterized by the presence of a protein called keratin. Keratin is also found in hair and nails. Kinaesthesia (o movement sense) Refers to the capacity to locate the body and its relative segments in space by sensing its motion, weight and position. Kyphosis Curvature of the spine, the convexity of the curve being posterior. L Limb buds Lateral swellings of the embryonic trunk which will develop into limbs. Limbic system A term loosely applied to a group of brain structures associated with olfaction and believed to be connected with certain aspects of emotion and behaviour. Lordosis Inward curvature of the spine. M Meconium The pasty greenish mass consisting of various substances such as mucus, desquamated epithelial cells, lanugo hairs and vernix caseosa that collects in the intestine of the fetus and forms the first faecal discharge of the newborn. Meconium aspiration syndrome Fetal distress (especially during labour) can trigger intestinal contractions and a relaxation of the anal sphincter with consequent ‘staining’ of the amniotic fluid. If the infant inhales the stained fluid, it may obstruct the airways, impeding the start of breathing, or by being an irritant it can cause chemical pneumonia. Medulla oblongata The medulla oblongata, a portion of the brainstem, functions
123
primarily as a relay station for the crossing of motor tracts between the spinal cord and the brain. It also contains respiratory, vasomotor and cardiac centres, as well as many mechanisms for controlling reflex activities such as coughing, swallowing and vomiting. Mesencephalic locomotor region A region of the mesencephalon defined functionally, rather than anatomically, and thought to play a role in many aspects of motor control. However, it is currently apparent that locomotion is driven by a distributed system rather than a specific centre in the brain. Mesencephalon (or midbrain) The part of the brain developed from the middle of the three primary vesicles of the embryonic neural tube. The mesencephalon contains structures important in the visual and auditory pathways (the trochlear and oculomotor nerves arise in the mesencephalon), as well as other structures thought to control locomotor activity. Mirror neuron system Mirror neurons are a class of neurons that discharge both when individuals perform a given motor act and when they observe others performing that same motor act. The human mirror system is involved in understanding the actions of others and the intentions behind them, and it underlies mechanisms of observational learning. Monochorionic twins Twins sharing the same placenta. Morphology or anomaly scan Ultrasoound scan usually performed at 18 20 weeks to assess fetal structural (morphological) organ development. The morphology scan gives indications not about any possible genetic defects, but about the anatomical structures visible in the scan. Muscular atrophy Decrease in muscle mass. Myoclonus Abrupt, shock-like contractions of muscles, irregular in amplitude and rhythm, and generally asymmetrical in distribution. Myofilaments Ultramicroscopic thread-like structures contained in striated muscles. Myosin A protein which in combination with actin is responsible for muscle contraction. Myotube (or myotubule) A developing muscle cell or fibre with a tubular appearance and with a centrally, rather than peripherally located nucleus. N Neonatal period Generally defined as the first 28 days after birth. Neuron Any of the conducting cells of the nervous system. A typical neuron consists of a complete nerve cell, including the cell body, one long process (the axon), and several short radiating processes (dendrites).
124
Glossary
Non-stress test The non-stress test is a way of externally monitoring the fetus. It measures the fetal heart rate accelerations in conjunction with normal fetal movements and any uterine activity (cardiotocography). N-REM sleep Non-REM sleep the dreamless phase of sleep. During N-REM sleep the brain waves on the EEG recording are typically slow and of high voltage, the breathing and heart rate are slow and regular, the blood pressure is low, and the sleeper is relatively still. N-REM sleep is divided into four stages of increasing depth eventually leading to REM sleep. Nuchal translucency test The nuchal scan is primarily (but not exclusively) a screening for Down syndrome and can be performed from 11 weeks and 3 days to 13 weeks and 6 days. The nuchal fold has been found to be thicker in fetuses with Down syndrome, because there is a greater collection of fluid between two layers of the skin of the fetus’s neck. Once the nuchal fold has been measured, this measurement is fed into a computer along with the woman’s age (as the risk of Down syndrome increases with maternal age) and a combined blood test measuring the concentrations of specific substances. The computer calculates the risk that a birth defect might be present. If the test is ‘screen positive’ the woman will be informed that she has a higher chance of having a baby with Down syndrome. This does not mean that the baby will necessarily have Down syndrome, nor in the case of a ‘screen negative’ (risk <1:250) result can the test give a 100% reassurance that the baby will not have it. However, the woman may give further consideration to undergoing genetic testing, namely chorionic villus sampling and amniocentesis. Currently many centres add an evaluation of the fetal nasal bone to the evaluation of nuchal thickness. O Oligohydramnios Reduced amount of amniotic fluid. Ontogenetic adaptations Transient capacities and functions suitable for survival at the stage in which they emerge, r but which may be unnecessary for or even incompatible with the requirements of later stages. Ontogeny (also called ontogenesis) The origin and development of an individual organism from embryo to adult. Oropharynx The oral pharynx, situated below the level of the lower border of the soft palate and above the larynx, as distinguished from the nasopharynx and laryngeal pharynx. Ossification (also known as osteogenesis) The process of bone formation or of a bony substance by conversion of fibrous tissue or of cartilage into bone. P Pacemaker Used in this book to mean a group of cells that has the ability to initiate periodic activity which does not depend on any external stimuli.
125
Palmar grasp reflex One of the so-called primitive or infantile reflexes. Refers to the fact that when an object is placed in the infant’s hand and touches its palm, the fingers close and grasp it. Parasympathetic nervous system Broadly speaking the part of the autonomic nervous system that slows down the heart rate, increases intestinal and glandular activity, and relaxes sphincter muscles (muscles that surround normal bodily openings and when contracted close them). Perceiving As opposed to simple sensing. A more complex operation involving the interpretation of sensations to give them meaning. Periodic breathing Breathing characterized by regularly recurring periods of apnoea, pauses in respiratory movements that last up to 20 s, alternating with breathing. Phantom limb The perception of sensations, often including pain, in an arm or leg long after the limb has been amputated. Pharyngeal plexus A network of nerves connecting with the pharynges. Phrenic nerve A mixed (sensorimotor) nerve innervating primarily the diaphragm, but also the pleura, pericardium, peritoneum and sympathetic plexuses. Phylogenesis The sequence of events involved in the evolutionary development of a species. Plexus A network of interrelating nerves. Polyhydramnios Increased amount of amniotic fluid. Polyneural innervation Innervation supplied by several nerves. Pons The portion of the brainstem containing the respiratory centre and located between the midbrain and the medulla oblongata. Popliteal angle The angle formed between the leg and the thigh. In order to measure it the physician flexes the infant’s thigh on the abdomen and then extends the knee to its limit. Pregnancy dating The so-called dating scan, normally performed at between 8 and 13 weeks’ gestation, has the objective of determining the fetal ultrasonographic age and the fetus’s due date. The crown rump length (the length of the fetus from the top of its head to its sacral bone, or, more simply, its bottom) is measured. Ultrasonographic age evaluated early in pregnancy is much more accurate than the simple unchecked calculation of gestational age. Pressure phosphenes A subjective sensation of light caused by pressure on the eyeballs.
126
Glossary
Proprioception Implies a feedback mechanism: when the body moves or is moved, information about this movement is returned to the central nervous system, which makes continuous adjustments in the movement and balance. Proprioceptors Sensory receptors located in the muscles, tendons and joints. Proto-self A basic level of awareness to be found even in lower animals. This is merely a short- term collection of neural patterns of activity which represent the current state of the organism. Pseudoglandular stage (or phase) The phase of prenatal lung development lasting from about the 6th to the 16th week, during which time repeated branching of the bronchi and bronchioles takes place, forming primordial conductive airways resembling exocrine glands (glands discharging their secretion through a duct on an internal or external surface of the body). Q Quickening The first feeling on the part of the pregnant woman of fetal movements. Normally occurs between the 4th and 5th month of pregnancy. Quiet sleep Considered analogous to N-REM sleep in the neonate; characterized by minimal large or small muscle movements and rhythmic breathing cycles. R Reflex action (also called Reflex) In its simplest form an automatic, involuntary and unlearned reaction to a stimulus. Reflex arc The neural pathway traversed by an impulse during a reflex action. The impulse travels to a nerve centre over afferent fibres (sensory fibres leading to it) and the response travels outward over efferent fibres (motor pathways carrying the impulse away) to an effector target, be it a muscle, gland or organ, capable of responding to a stimulus, especially a nerve impulse. Reflexogenic Producing a reflex action or resulting from a reflex action. REM sleep (rapid eye movement sleep, also called paradoxical sleep) Sleep characterized by low muscle tone and a rapid, low-voltage EEG, indicating that the activity of the brain is similar to that during waking hence the name ‘paradoxical sleep’. However, the body is paralysed due to atonia (or fall of muscular tone), This is another reason for the name paradoxical. The relative amount of REM sleep varies considerably with age. A newborn baby spends more than 80% of total sleep time in active/REM sleep. Rest–activity cycles Prior to 34 36 weeks’ gestation fetuses are commonly regarded as alternating rapidly between epochs of turbulent motion (activity cycles) and epochs of absolute rest (rest cycles).
127
Reticular formation A diffuse network of nerve fibres and cells in various areas of the brainstem and the diencephalon. The reticular formation is thought to be a complex, highly integrated mechanism that exerts both inhibition and facilitation on almost every type of activity of the central nervous system. Rooting reflex One of the so-called primitive or infantile reflexes present at birth, helping the neonate in orienting itself towards the breast. When stroked on the chin or mouth the neonate moves its head automatically towards the side where it has been touched, making searching motions and opening its mouth until it latches to the nipple. S Sense of body Used in this book to refer to the beginnings of me/not-me sensations and of a sense of bodily demarcation. Sensing As opposed to the more complex faculty of perceiving. The simple receiving and reacting to a sensory feedback. Sensitization A form of aversive conditioning in which the frequency of an undesirable behaviour is lessened by associating the behaviour with unpleasant stimuli. Sensory cortical homunculus A disfigured map of the human body that shows the relative amount of cerebral cortex surface area given over to processing the different sensory inputs of the human nervous system. Sensory input or sensory feedback Conveying of information pertaining to the senses. Sleep myoclonus (or hypnic jerks) Non-pathological myoclonic jerks of the limbs (and in the neonate also of the body) occurring when falling asleep or when asleep. Particularly frequent in the neonate, but also present in the adult. Slow twitch muscle fibres Muscle fibres suited to endurance activity that develop gradually after birth. Soft palate The muscular, non-bony part of the roof of the mouth located behind the hard palate. Somatosensory Pertaining to bodily sensations. Somatosensory cortex Area of the parietal lobe of the cortex receiving sensations from the skin and deep tissues, and the seat of the sensory cortical homunculus. Somatosensory demarcation Used in this book to refer to a proto-sense of one’s own bodily boundaries. Spinal cord The major column of nerve tissue that lies within the vertebral canal and from which spinal nerves emerge. The spinal cord and the brain constitute the central nervous system. During prenatal life the spinal cord initially acts alone, sending
128
Glossary
motor impulses and receiving sensory ones from the periphery. As brain development progresses, the spinal cord becomes increasingly connected with the brain, transmitting and receiving impulses from it. However, the developmental steps leading to this connection and its timing are not yet fully clarified. Spinal motor neurons Efferent neurons with a motor function, conveying motor impulses. Their cell body is located in the anterior grey column of the spinal cord. Their axons pass by way of a peripheral nerve to skeletal muscles. Startle Fetal startles are spontaneous shock-like jerks of the entire fetal body lasting about 1 s each. Startle epilepsy A form of epilepsy characterized by seizures triggered by unexpected sensory stimuli. Startle reflex or Moro reflex Reflex elicited by altering the equilibrium or the plane between the child’s head and trunk, or by a loud noise. The head of the infant is lifted and then allowed to fall backward, or the infant is pulled up by both hands from a lying position and then let go. The normal infant abducts (spreads out) and extends arms and fingers, and then brings them back together as if in an embrace and clenches its fists. Normally seen up to 3 4 months of age. Stepping reflex A reflex response of the newborn and young infant, characterized by alternating stepping movements with both legs as in walking, elicited when the infant is held upright so that both soles touch a flat surface while the infant is moved forward accompanying any step taken. Surfactant A substance secreted by the cells of the alveoli (the tiny air sacs in the lungs) which serves to reduce the surface tension of pulmonary fluids, thus contributing to the elasticity and expansion of lung tissue. Sympathetic nervous system A part of the autonomic nervous system broadly speaking presiding over acceleration of heart rate, constriction of blood vessels and raising of blood pressure. Synapse Region of communication between two neurons, where the impulse passes from the axon of one neuron to a dendrite or the cell of another. T Tachypnoea Abnormally fast breathing. Taxonomy The practice and science of classification. Used in this book to refer to classification of behavioural events. Tonic seizure Seizure characterized by continuous tension or contraction. T Tonic–clonic seizure Seizure characterized by rapid involuntary alternating muscular contractions (tonic phase) and relaxations (clonic phase).
129
Thoracic sympathetic chain A series of nerves leaving the spinal cord between the 6th and 12th thoracic vertebrae (T6 and T12) and innervating intercostal muscles, and by the pharyngeal plexus, a network of nerves connected with the pharynx. The efferent pathways pass mainly through the vagus and the phrenic nerve. Trigeminus or trigeminal nerve (cranial nerve V) A mixed nerve. The chief nerve of sensation for the face, which also controls most of the muscles used for chewing. Twitches Brief and small involuntary contractions of a muscle or a group of muscles. U Ulnar opposition The capacity of the small and ring fingers to rotate across the palm of the hand to meet the thumb. Ultrasonography, ultrasound Ultrasonography is an imaging technique employing ultrasound waves acoustic waves that, as the name implies, are of a frequency beyond the range of human hearing. A sound wave is generated by a transducer enclosed within the probe and made to travel into the body, focusing at a desired depth. Materials on the transducer, and a water- based gel placed between the probe and the patient’s skin, enable the sound waves to be transmitted efficiently into the body. Different tissues have different densities and the sound is reflected back as an echo to the transducer anywhere there are density changes in the body. The sound wave then returns to the transducer, and the transducer turns the vibrations caused by the sound waves into electrical pulses which are then processed and transformed into images. In obstetrics the image is a two-dimensional (2D) real-time (or B-mode) representation of the targeted section of the body. A continuous (real-time) picture of the moving fetus can be depicted on a monitor. Ultrasonography, 3D and 4D In 3D and 4D ultrasonography, unlike 2D, the sound waves are sent not straight into the body, but at different angles. The returning echoes are then processed by a computer program, resulting in a reconstructed 3D image of the fetal outer surface or of its internal organs. No movement is shown. Fourdimensional ultrasonography captures the reconstructed 3D images speedily and animates them, showing fetal movement. V Vagus nerve (or cranial nerve X) The vagus is a mixed nerve originating in the medulla oblongata, a part of the brainstem, and wandering all the way down from the brainstem to the colon hence its name, from the Latin vagare meaning to wander. The vagus supplies nerve fibres to the pharynx, larynx, trachea, lungs, heart, oesophagus and the intestinal tract as far as the transverse portion of the colon. The vagus nerve also brings sensory information back to the brain from the ear, tongue, pharynx and larynx. Vallecular space In the suckling neonate, a roughly triangular space at the far back of the tongue where milk accumulates in the initial phase of sucking. It is delimited by the back of the tongue, the soft palate and the front of the epiglottis.
130
Glossary
Vestibular feedback The perception of balance, head position, acceleration and deceleration obtained from the semicircular canals in the inner ear. W Wernicke’s area Region of the brain (generally the left posterior gyrus) responsible for the capacity to understand rather than to utter language. Wharton’s jelly The mucoid connective tissue that constitutes the matrix of the umbilical cord. Z Zygosity The genetic state of the zygote (the fertilized ovum) with reference to identity (monozygosity) or non-identity (dizygosity or heterozygosity) of two or more genes.
Subject Index
A Active sleep 26, 44, 87-89, 93, 117 Albertus Magnus 49 Anderson and Meno 85 Anencephali 29, 35, 104 anger 77, 78 apnoea 19, 29, 36, 40, 43, 110, 117, 127 apnoeas 42, 43, 88, 92, 109, 110 arm movements 5, 61, 62, 99 arterial chemoreceptors 37, 39 Aserinsky and Kleitman 87 Asim Kurjak 4, 54 athetosis 59, 62 autism 77, 85
B Bailey 30 behavioural individuality 97, 103 bereavement 97, 106 bimanual prehension 58, 61, 64 body schema 69, 71, 109, 112,114 brainstem 7, 16, 17, 19, 30, 36, 39, 42, 45, 54, 62, 88, 92, 93, 119, 125, 127, 131 breathing 4, 5, 22, 23, 26, 29, 30-33, 36, 37, 39-45, 49-51, 54, 60, 61, 64, 80, 88-93, 109, 110, 117, 119, 122, 124-126, 128, 130
C Carl Wernicke 56 central nervous system 5, 7, 8, 16, 17, 20, 30, 40, 62, 69, 75, 87, 92, 120, 121, 128-130 central pattern generator 7, 16, 17, 19, 21, 25, 29, 30, 36, 42, 54, 119 cerebral palsy 3, 27, 59, 73, 74 children 19, 26, 27, 35, 54, 74, 78, 85-88, 106, 110, 120
clonuses 5, 19, 24, 25, 27, 109, 119 clusters 5, 16, 19, 23, 25, 42, 59, 61, 87, 89, 90, 93, 103, 104, 109, 120 communication 54, 61, 82, 85, 97, 104, 105, 120, 130 corticalization 54, 59, 61 cranial nerves 30, 54, 59, 65, 119, 120 cross-modal integration 77, 80, 82, 114, 121 crown rump length 19, 20, 121, 128
D Darwin 26, 36, 61, 77, 78 Dawes 37, 39 Democritus 49 dichorionic 97-99, 102, 121 disgust 77, 78, 80 dreaming 19, 26 dreams 87, 88 Duchenne de Boulogne 77
E EEG 26, 87, 88, 121, 126, 128 Ekman and Friesen 78 Elizabeth Bates 82 EOG 87, 88 Epicurus 49 epilepsy 19, 27, 110, 130 Ernst Haeckel 33
F facial expressions 4, 5, 36, 51, 59, 77-80, 82, 83, 85, 77-79, 81, 88, 109, 114, 115 fear 19, 20, 23, 24, 26, 77, 78, 85, 78 feeding 30, 42, 43, 49-51, 109, 110 fetal thumb sucking 5, 49
132
FHR 89, 118 foot positioning 59, 61 Frank Manning 40 Freud 27, 87
G Galen 29, 77 gasping 5, 29, 36, 37 general movements 5, 7-10, 12, 13, 16-18, 20-23, 25, 27, 32, 42-44, 46, 50, 51, 54, 59-61, 69, 89-93, 99, 101, 109-114, 122, 123 generalized motions 14, 43 Giuseppe Moruzzi 88 glycaemia 109, 110
H hand motions handedness 5, 49, 54, 56, 123 hand-shaping 59, 61 happiness 77, 78 Henrique Rigatto 39 Hepper 4, 85 Hildegard of Bingen 49 Hippocrates 29, 49, 77 Hooker 2, 7, 97, 112 Hubel and Wiesel hyperexplexia 19, 123 hyperplasia 59 hypertrophy 59 hypotonic urine 49, 52, 124 hypoxia 29, 36, 37, 42, 43, 52, 119, 124
I intrapair stimulation 97-101, 105, 106
J Jean Piaget 69, 85 Jost and Pollicard 44
K kyphotic 7, 8
Subject Index
L Legallois 36 locomotion 71, 109, 111, 125 Lombroso 77 lordosis 7-9, 124 lung formation 109, 110 lung liquid 39, 44, 54
M maternal emotions 97, 105 meconium 29, 36, 37, 44, 124 Michel Jouvet 88 Miller 51 mirror neurons 77, 82, 85, 109, 125 Molyneux 82 monochorionic 97, 98, 102, 104, 125 monozygotic 97, 98, 100, 102-104, 121 Moro reflex 19, 20, 23, 130 Moruzzi and Magoun 35 motor patterns 7, 8, 17, 25, 119 myotubes 7, 13, 59 myths 1, 3, 5, 97, 104
N nuchal tone 7, 91
O oligohydramnios 49, 52, 74, 126
P Paillard 69, 71 Paul Broca 56 perception 59, 74, 75, 82, 85, 109, 114, 127, 132 phrenic nerve 29, 30, 33, 45, 127, 131 physiognomy 77 Piaget 69, 80, 85 Piero Salzarulo 35 polyhydramnios 49, 54, 127 polysomnographic recordings 87, 88 popliteal angle 7, 13, 15, 73, 127 Prechtl IX, 2, 7, 27, 43 primitive stepping 7, 14
133
proprioception 59, 74, 75, 82, 109, 111-114, 128 Provine 35, 85 pulmonary hypoplasia 39, 44 Q quiet sleep 19, 87-89, 91, 93, 128
R REM sleep 19, 26, 87, 88, 117, 126, 128, 129 Royal College of Obstetricians and Gynaecologists 84
S Shortt 31 SIDS 29, 36, 43 Sir Francis Galton 97 Sir Joseph Barcroft and Don Barron 39 sneeze 29 somatosensory homunculus 109, 113, 114 startles 4, 5, 16, 19-25, 27, 31, 32, 40-43, 46, 51, 80, 88-90, 92, 93, 99-101, 109-111, 130 sucking 5, 22, 23, 42-44, 49, 50, 54-56, 92, 109-111, 117, 132 supine kicking 7, 14 swallowing 5, 22, 23, 30-33, 42-44, 46, 49-54, 61, 81, 89-93, 101, 109-111, 119, 122, 125 synchronous diaphragmatic flutter 29
T tachipnoeic Thomas Lumsden 36 tongue protrusion 51, 77, 80-82, 114 twitches 5, 19, 22, 24-27, 32, 43, 46, 51, 89, 90, 101, 109, 117, 131
U ultrasounds 5
V Van Hoof 79 vestibular system 109, 112, 113 vocalizations 77-80, 82, 114, 115 volar pads 7, 11 von Ahlfeld 39
W wakefulness 8, 24, 26, 30, 35, 36, 39, 83, 87, 88, 93, 104, 119, 121 William C. Dement 87 William C. Magoun 88
Z zygosity 97, 102, 103, 132