![]()
Chapter 1
The nature and consequences of feeding problems in infancy
Sheena Reilly, David Skuse and Dieter Wolke
Introduction
The human infant, like other infant mammals, depends completely upon a social relationship for its nutritional requirements and for the maintenance of life and its physical well-being (Lipsett et al., 1985). The earliest relationship between the infant and caregiver may serve a number of different purposes quite apart from the importance of that interaction for the supply of adequate and appropriate nutrition. The relationship may serve as a basis of social dialogue, and the quality of action during feeding may reflect the nature and quality of parent-infant interaction in other settings as well (Wolke, 1994, 1996). The style of the interpersonal relationship during feeding will be influenced by the individual characteristics of both the infant and the caregiver, and if problems develop during the course of that feeding interaction there may be implications for the style and quality of their relationship in other (non-feeding) contexts, as well as for subsequent feeding behaviours, including preferences and aversions (Lindberg, 1994).
The relative neglect of feeding as an issue central to early child development is puzzling. In an editorial in the journal Developmental Medicine and Child Neurology Bax (1989) suggested that the reason why feeding has received little attention in the developmental psychology and paediatric literature is because it is treated so extensively in childcare books that the subject does not seem to be a ‘respectable one’ for the clinician or research worker to investigate. He goes on: ‘A glance at the work of nutritionists, gastroenterologists and even dieticians . . . shows that none of them discusses the problem of actually getting food into the child.’ Bax examined the bibliography of the journal and discovered that in previous years there had been very few publications in the journal regarding infant feeding: one paper in 1986, none in 1987 and two in 1988. Bax issued a challenge to all health professionals involved in paediatrics to pay increased attention to infant feeding.
Recently we have seen the publication of numerous specialist textbooks devoted to the management of feeding problems, the publication of a specialist journal devoted to dysphagia, as well as a plethora of journal articles. Some names stand out as being major contributors to our understanding of infant feeding, such as James Bosma, whose work spans over thirty years of study of the anatomy and physiology of feeding (e.g. Bosma et al., 1967; Bosma, 1977, 1986), Erika Gisel (e.g. Gisel and Patrick, 1988) and Suzanne Evans-Morris, who has made a particular study of the feeding difficulties of infants with cerebral palsy (e.g. Evans-Morris, 1977a, 1977b, 1985).
Early feeding and the success of feeding development in the infant depend upon the organisation of a number of interacting biological and psychological systems in which nutritional requirements, developmental attainments, internal sensations such as hunger and satiety, and cultural pressures impinge upon the child. It is not possible to discuss sensibly the nature and range of feeding problems encountered in infancy without first discussing the normal developmental phases to be observed. The success of the infant's feeding relationship for normal growth and health is dependent upon a number of self-regulatory processes, and also upon a caretaker who is sufficiently responsive to that infant to provide requisite environmental modifications. The infant is born with the capacity to emit a repertoire of signals which indicate needs and which are intended to evoke the appropriate response from the caregiver (Adoph, 1968). A lack of capacity for self-regulation would render the infant's very survival at risk; normal development is contingent upon the infant's ability to signal his or her needs clearly and unambiguously.
Endogenous or constitutional factors comprise a complex repertoire of reflexes and behaviour patterns which are well suited to the task of acquiring food and ingesting it. There are complex schemas of rooting, sucking and swallowing responses which take advantage of the presence of a caregiver who can feed the child responsively, and whose own psychological and physiological resources are normally well coordinated with the infant's physiognomic and behavioural characteristics. In the earliest days the infant is motivated primarily by reflexive and biologically programmed behaviours in order to seek food. In the first few months of life caregivers respond adaptively to their infant's needs and must recognise and respond appropriately to their infant's repertoire of behaviours. Feeding problems are particularly likely to develop when either the infant's endogenous programming is faulty in some way, compromising the clarity of these signals or the coherence of the pre-programmed behaviours, or when the caregiver is for some reason unable accurately to identify or respond to the infant's particular needs (Wolke, 1994). There is a subtle interrelationship between these two major variables which puts a proportion of infants at relatively ‘high risk’. The feeding relationship could thus be seen within the framework of a transactional model (Emde, 1987): constitutional factors (such as the coherence of foetal development, the duration of gestation, and perinatal and postnatal factors that might affect subsequent psychomotor performance) interact with environmental conditions and emerging social, emotional and cognitive abilities that may or may not act in concert with the reflexive developments which are age-appropriate.
The development of ingestive behaviour
Any discussion of feeding problems in infancy should begin with mention of the development of swallowing in utero. Foetal swallowing is thought to contribute to several critical developmental processes including the regulation of amniotic fluid volume and composition, the acquisition and re-circulation of solutes from the foetal environment, and the maturation of the foetal gastrointestinal tract. Controversy surrounds the role of foetal swallowing and voiding in the regulation of amniotic fluid volume. Nevertheless some interesting facts remain. First, the volume of fluid swallowed is strongly correlated with the volume of the fluid within the amniotic cavity. Second, there is marked increase in foetal swallowing in association with increasing gestational age: an 18-week-old foetus swallows 4–11 ml of fluid per day, whereas near-term foetuses (38–40 weeks) have a mean swallowing rate of 198 ml per day (Abramovich et al., 1979). Finally, ultrasound studies have demonstrated that the human foetus can chew, swallow and regurgitate during intrauterine life (Bowie and Clair, 1982), and that the presence of foetal vomiting in utero may be associated with gastrointestinal obstruction.
These studies demonstrate that oral-motor and pharyngeal activity increases throughout gestation (Humphrey, 1964; Hack et al., 1985), indicating that the motor program for sucking and swallowing appears to function well before term thereby providing the foetus with a range of early experiences in utero.
There is growing support from both animal and human studies to suggest that prenatal nutritional deprivation may be associated with increased risk of disease in adulthood (e.g. Barker and Martyn, 1992). The materno-foetal pregnancy environment may be influenced by regulatory mechanisms which control ingestive behaviour; furthermore, foetal swallowing might be modulated by a variety of factors including neuro-behavioural state changes and might be further influenced by hypoxia, hypotension and plasma osmolality changes. Although speculative, there are also suggestions that foetal swallowing might be regulated by the development of appetite sensation, salt appetite and the development of taste. Animal studies have demonstrated that altered osmotic environments modulate not only swallowing activity but also the development of adult sensitivities for thirst and arginine vasopressin secretion and responsiveness.
Therefore, future work with human foetuses will have important implications for foetal and adult dipsogenic regulation. The suggestion that regulatory mechanisms such as the development of appetite, salt and taste sensation are susceptible to influences in the pregnancy environment will alter our concept of the development of ingestive behaviour during the postnatal period. Prenatal constitutional factors may play a far greater role than previously expected.
Learning processes in infancy
Although infants have traditionally been viewed as capable of only reflexive behaviours at birth there is increasing evidence that very young infants, including newborns, can learn the contingencies between their own responses and the consequences they produce. These behavioural changes are not elicited, nor are they a reflection of the general behavioural arousal. In full-term newborns the rapidity of acquisition is specific to the nature of the response and reinforcer and the task parameters. Responses associated with feeding, for example, are learned very rapidly even by newborns (Rovee-Collier, 1987), in particular those associated with aversive experiences. Conditioning procedures have yielded evidence of robust retention of memories after hours, days and even weeks in extremely young infants. The young infant is prepared quite early to take in and learn from a wide range of stimulation, and to associate some experiences with others. Infants of 8 to 12 weeks of age, given the appropriate training conditions, can remember for periods of up to fourteen days without the benefit of a reminder. There is considerable evidence that, by 3 months of age, not only are infants capable of learning the contextual cues, but for periods of days or even weeks after that learning experience those contextual cues can selectively influence retrieval (for review see Rovee-Collier, 1987).
The cry of the human infant is highly effective in eliciting nurturance and bodily contact (Bernal, 1972) and it induces anticipatory milk let-down and increased breast temperature of the lactating mother (Vuorenskoski et al., 1969). In general, newborns are hedonic creatures responding to the incentive-motivational properties of reinforcers with motor changes in behaviour, such as sucking and swallowing, and autonomic changes (Lipsitt et al., 1985). Both the foetus and newborn have functioning chemosensory systems and within a few hours of birth infants are sensitive to subtle changes in gustatory stimulation and have pronounced preferences for sweeter fluids (e.g. Crook, 1979). While taste preferences for sweet foods, salty foods and milk appear to be influenced by a substantial genetic component, exposure of the foetus in utero to flavours in the amniotic fluid and human milk may also contribute to later preferences (Mennella and Beauchamp, 1998). In addition, postnatal maturation of the infant's sensory system is vulnerable to many other experiential influences.
There is normally a close and communicative interaction between a mother and her infant. During feeding the two individuals are responsive to one another and communication between them is warm and relaxed. All the child's sensory receptors become activated, and begin to integrate information from eyes, ears, nose, touch and movement. The development of normal feeding behaviour in infancy is also influenced by the transactional relationship between the infant's and the caregiver's behaviour. The crucial variables include, first, how the nutrition that is offered to the child is presented; second, how the nutrition is accepted by the child; and, third, the caregiver's reaction to how that offering is accepted. The infant's behavioural response during feeding may be an expression of underlying maturation, reflecting the integrity of the autonomic, motor and state subsystems and their transaction with the environment. If problems develop in the feeding relationship they can mirror difficulties elsewhere in the relationship at that time, but can also serve as a pointer to potential problems developing in the future.
Feeding and growth in early infancy
The first six months or so of life is a period of tremendous growth in terms of weight gain; in the first half of the first year after birth the rate at which the child is gaining weight is greater than it will ever be again until puberty, and of course must be sustained by an adequate intake of a diet rich in protein and calories. Relative deficiencies in protein or energy intake at this time are likely to have profound effects upon attained weight for age. The velocity of growth in terms of length is also at a peak shortly after birth, but in this case there is a more gradual diminution over the first two years or so. The rate of growth remains relatively constant until puberty from 6 months, in terms of weight gain, and from 2 years in terms of linear (skeletal) growth. In order to sustain this remarkable rate of growth in infancy the infant needs an adequate supply of calories, energy intake seemingly being the most common limiting factor upon growth, at least within the developed world. The proportion of energy (caloric) intake that is used for growth is higher during the first four to five months after birth than it ever is again; for a period of about five months after birth, in full-term infants, growth accounts for more than 8 per cent of normal energy intake but at no subsequent stage is it more than 5 per cent (Widdowson, 1973). Stunted growth due to chronic early malnourishment may be reversed in later years by a plentiful supply of food; but the longer the early period of under-nutrition the briefer the subsequent period of so-called ‘catch-up growth’ and the smaller the ultimate size of the child (Widdowson, 1973).
There is considerable individual variability in energy needs. Healthy babies who are being breastfed may consume a daily volume of milk at any given age that varies across a two- to threefold range (e.g. Whitehead and Paul, 1981). This variance cannot simply be accounted for by the differences in size of babies or by differences in their velocity of growth. Whitehead (1985) demonstrated that, at least from 2 to 8 months after birth, standard rates of growth can be achieved on energy intakes that are substantially smaller than the current recommended dietary allowances (Department of Health and Social Security, 1979). Towards the end of the first year there is closer correspondence with the WHO/FAO recommendations, possibly because of increased activity levels (FAO/WHO/UNU, 1985).
Historical trends in infant feeding
During the last decade there have been major changes in infant feeding practices. The first seventy years or so of the twentieth century saw a progressive fall in the use of human lactation as the normal way of feeding a young baby (Whitehead et al., 1986). During 1911–1915, 70 per cent of American infants were reported to be breastfed at the time they left hospital (Hirschman and Butler, 1981). Even at 9 months of age the proportion was still as high as 30 per cent. However, by 1946–1950, the corresponding levels had fallen to only 24 per cent and 1 per cent respectively. Quite clearly there had to be a correspondingly high proportion of infants who were consuming cow's milk based formulae, and there is evidence to suggest that solid foods were being given at increasingly early ages. By the 1970s, 80 to 95 per cent of British babies were receiving non-milk solids by 3 months of age (Whitehead et al., 1986). This however fell steadily to 68 per cent in 1990 and 55 per cent in 1995 (Foster et al., 1997), the changes reflecting the recommendations prevailing at the time the two surveys were undertaken.
Babies are said to adjust their volume of intake to the caloric density of the feed, so the daily fluid intake of babies fed over-concentrated feeds is reduced (Whitehead et al., 1986). Despite reduced water intake, caloric intake is, in such circumstances, higher than that of infants fed formulae of lower caloric density. Reduced water intake may have the effect of making infants especially vulnerable to changes in external water balance, and also renders them unable to tolerate long periods without feeding, giving rise to the common complaint that infants fed in this way never seem to be satisfied (Taitz, 1974). Studies on serum osmolality suggested that such babies were in a state of mild hypertonicity. The fault lay not with artificial feeding itself, but with the way in w...
