This is an exciting time for pediatric gastroenterology research, with evidence growing rapidly that early environmental experience “programs” the anatomy and functioning of the gastrointestinal (GI) tract, including the intestinal microbiota and the enteric nervous system (ENS), and its connections with the central and autonomic nervous systems. This programming appears to be the result of evolutionary pressures that allow each individual's GI system to conform to its environment. Although usually beneficial, this plasticity may have adverse consequences in which a harmful environment programs the GI tract in an ineffective and ultimately unhealthy way.
For many years, the prolonged periods of unexplained crying that characterize early infancy have been attributed to “colic,” involving GI disturbance and pain (see Background and Scope for details). Although popular, this view of the crying has lacked a precise account of the origins of the GI disturbance, if any, and of the physiological processes that translate GI disorder into infant pain. The concept of environmental programming of GI structure and function potentially offers a way of specifying how colic comes about and how to prevent it. Rapid scientific progress in an area also can give rise, however, to hyperbole and overgeneralization. It is important that the limitations of current evidence are taken into account, particularly where the aim, as it is here, is to inform routine clinical practice as well as advance research.
Together with a critical analysis of up-to-date research, the strengths of this supplement include bringing together researchers from pediatric GI, developmental psychology, and behavioral backgrounds. Any understanding of infant distress, discomfort, or pain requires attention to these psychological processes and to the ways in which they give rise to infant behaviour; it seems fair to say that GI research thus far has been limited in addressing psychological and behavioral factors. Indeed, the position is made more complex by the dependence in much clinical research on parental reports rather than measures of infants. Although parental reports provide the point of contact for clinicians and drive health services costs, they are once removed from measures of infants themselves, resulting in a variety of biases and distortions. Scientific progress is most likely to result from research that combines these different professional backgrounds and sources of methodological expertise.
As Burns and Thapar discuss in the first article, any understanding of how the GI system is programmed needs to be rooted in the knowledge of how growth and developmental processes normally occur. Their review provides an authoritative, up-to-date account of the processes involved. In particular, there is evidence that the anatomy and function of the ENS are modified by experience both before and after birth. Enteric neurons respond to the intestinal environment by changing the expression of receptors and/or neurotransmitters or their associations with other cell types. Dietary factors or high levels of stress can result in ENS adaptations that reset the growth trajectory for the individual involved. Early life stress may increase visceral sensitivity and mucosal permeability, alter the balance in enteric microflora, and increase mucosal mast cell density. Most of the current evidence comes from animal studies, but the findings provide a model for how the early environment may increase visceral sensitivity, and consequently perceived pain, in human infants.
Complementing Burns and Thapar, the 2 articles by Indrio and colleagues list ways in which GI disturbances may result in infant discomfort. As they point out, honing down on any particular GI component is complicated by the complexity and dynamic interplay among the gut microbiota, ENS, and immune, autonomic nervous, and central nervous systems. They posit several potential mechanisms for colic pain, including intestinal distension, disturbed gut motility spasms or gastric emptying, aberrant microbiota colonization, disturbed vagal signaling, and the action of corticotropin-releasing factor in modulating gut motility, inflammation, permeability, and visceral hypersensitivity. In keeping with other studies and reviews (1–5), Indrio et al conclude that gastroesophageal reflux rarely causes crying or colic. This conclusion stands in marked contrast to the evidence that treatment of excessively crying infants involving antireflux medicines is highly prevalent in some countries. They conclude that alterations in the composition of intestinal microflora appear to contribute to several functional disorders in children and adults, including infant colic. They also mention that modification of the intestinal microflora with the use of probiotics represents a promising therapeutic approach to this widespread condition. Other studies, reporting changes in the microbiota composition of colic infants, point to the need to understand the functions of different bacteria (5).
Di Lorenzo's article on early life events develops the previous articles’ proposal that stressful early life experiences (called adverse early life events, or AELs) program later GI and psychological function. The argument is complex, because AELs are not expected to have the same effect in all individuals. Instead, the proposal is that an individual with a predisposing genetic susceptibility who is “primed” by a sensitizing event in early childhood goes on to develop physical and psychological impairments in later life. He draws on supportive human and animal evidence, for example, to suggest that gastric suction at birth is associated with functional intestinal disorders at later ages. Other evidence links early trauma, abuse, and neglect with functional GI disorders and anxiety, depression, and distress in later life. Possible mechanisms include altered stress response, changes in gut microbiota and inflammation, alterations of neuronal structure and function, and modification of the immunological function of the enteric mucosa. Most of the supporting human data are from studies following up older children who have experienced unusually severe trauma, and Di Lorenzo notes the lack of evidence linking AELs with crying and discomfort behaviors during infancy. Nevertheless, in support of this broad approach, several studies have found associations between maternal stress and anxiety during pregnancy and high amounts of infant crying in the weeks after birth (6–8). The implication is that the AEL hypothesis provides a promising basis for research designed to evaluate the GI mechanisms and infant outcomes posited, but is not a basis for routine health service practice at the present time.
The review by Ha-Vinh Leuchter and colleagues focuses on the central nervous system (CNS) rather than the GI tract. In common with the previous articles, it proposes that stress can program system structure and function, drawing on supportive animal evidence, but noting that human findings are inconsistent. The notion of temperament could help us to understand individual differences in emotionality, but the authors acknowledge that the on–off pattern of prolonged crying in early infancy is difficult to reconcile with the idea of a temperamental trait. The concept of regulatory disturbance, wherein the crying is seen as the result of inadequate CNS control of responsiveness to sensory stimulation, leading to prolongation of crying responses, is considered more promising.
The first of Heine's 2 articles focuses on the specific link between cow's-milk allergy (CMA) and infant colic. He attributes the CMA to early programming of the immune system, whereby gut bacteria modulate early immune response and tolerance development. Although methodological concerns remain, there is substantial evidence from studies both of breast-fed and formula-fed infants that CMA can result in prolonged crying during early infancy. An important question is how common this problem is. Although the exact prevalence of CMA in infants with colic is unknown, approximately 2% of infants have early-occurring CMA, ∼50% of whom displayed irritable and colicky behavior following a cow's-milk challenge. As Heine points out, an implication is that most infants displaying prolonged crying in early infancy do not have CMA.
Unfortunately, detection of CMA cases using skin prick testing or measurement of cow's-milk–specific serum immunolobulin E (IgE) antibodies is ineffective because the allergy involves a non-IgE–mediated form of CMA. Instead, diagnosis relies on the demonstration of a reduction in crying after cow's-milk elimination, and crying recurrence after cow's-milk challenge, often together with other symptoms, such as increased vomiting/regurgitation, diarrhea, or eczema. Heine recommends that a time-limited trial of hypoallergenic formula or a maternal elimination diet should be considered in infants who continue to cry persistently beyond 4 months of age or exhibit other clinical features of CMA (persistent diarrhea, vomiting, or eczema). When crying improves within 2 to 4 weeks of commencing the diet, a cow's-milk challenge should be used to confirm the CMA diagnosis; in practice, parents may be unwilling to implement this, however. Nutritional progress and growth need to be carefully monitored and the diet normalized as soon as tolerated, which is usually possible by 9 to 12 months, or 2 to 3 years of age in a minority of cases. Unlike CMA, there is little evidence that lactose intolerance causes prolonged infant crying.
Heine's conclusions are consistent with those of another review (1) and provide practical steps that can benefit parents and infants as part of health services. They, however, also raise questions that require further research. One proviso is that such cases are rare, with an estimated prevalence of <2% compared with approximately 20% for prolonged crying in early infancy (1,9). The method needed to identify these cases, dietary restriction, is challenging to implement, especially in infants who are breast-fed (10), so it will be important to know how many parents are able to comply with the diagnostic and treatment regimens involved. Ideally, simple tests are needed to distinguish these cases and it would be helpful to know how many can be identified by their concomitant diarrhoea, vomiting, or eczema. The need for expert monitoring is clear, but the associated staff training and resourcing may be beyond health services in some places. These questions and the associated cost-effectiveness issues are in urgent need of further research.
Bellaïche et al summarize the evidence on treatments of infant colic as a whole. They agree with Heine that extensively hydrolyzed protein formulas may help formula-fed infants, especially in the presence of other digestive symptoms, whereas restricted maternal diets can help some breast-fed infants. Lactase treatment or soy-based formulas are not recommended. Herbal therapies (notably fennel) appear useful in some cases, but there is a need for more research to establish safety and effectiveness. There is no firm evidence that chiropractic or reflexology interventions are effective. An appropriate medical consultation, providing empathy and reassurance, is considered to be essential for the management of affected infants and parents. They conclude that the use of the probiotic Lactobacillus reuteri could be beneficial to ameliorate symptoms of infant colic, but that there is still insufficient evidence to support the general use of probiotics in all infants with colic or to recommend its use in preventing colic. According to the systematic review of Sung et al, 12 trials have examined the use of probiotics to treat or prevent prolonged crying, 6 finding reductions and 6 not (11). Only 2 of the trials were considered to be free of methodological bias and the studies varied greatly in how infants were selected and fed, the adequacy of the crying measures, when crying was measured, the type and dose of probiotics prescribed, whether they were used preventively or as treatments, and in the robustness of controls. Among the studies evaluated, the most promising seemed to involve the use of L reuteri, which was beneficial in 1 of the prevention and 3 of the management trials. Sung et al also concluded that L reuteri may be effective as a treatment for crying in exclusively breast-fed infants with colic, but the evidence is still insufficient. The implication of this systematic review is that further research with improved methods is needed. Sung and colleagues note that the metaRegister of Controlled Trials lists 5 unpublished studies designed to examine the efficacy of probiotics in the management of breast- and formula-fed infants with colic that are under way. Their findings should help to clarify this complex picture.
Like Ha-Vinh Leuchter et al, St James-Roberts and colleagues focus on CNS development. Their approach does not require environmental programming, either of the CNS or GI tract. Instead, they conclude that prolonged crying in early infancy is a result of normal CNS processes of maturation and development. In support, they cite evidence that infants in general have crying peaks in early infancy at a time when CNS reorganization takes place, as cortical regulation replaces reflex control of behavior. The prolonged and “unsoothable” crying bouts are attributed to a temporary deficit in neurological control of behavior during this transition, so that infants respond strongly or cannot stop crying once it has started. This view has much in common with that of Ha-Vinh Leuchter et al, except that the CNS structure and function of crying babies are considered to be developing normally. Individual differences in how much infants cry are attributed to normal variations in development.
As St James-Roberts and colleagues note, this developmental viewpoint offers a possible explanation for the existence of long and unsoothable crying bouts in early infancy, for the puzzle of why so many apparently healthy babies should cry without a reason, and for the spontaneous resolution of this type of crying with age. It does not, however, explain why the crying clusters in the evenings and lacks direct empirical support from studies linking CNS changes to crying changes. The authors also note that neurodevelopmental processes may account for the nature and age of prolonged crying in babies, but other factors in the infant or the environment may explain the onset of crying periods.
Unlike the other articles discussed above, the implication of this developmental viewpoint is that 1- to 3-month-old infants who cry frequently are healthy, so that they do not need a therapeutic treatment. Instead, the focus is on supporting parents in how to manage the crying and, consequently, on parental vulnerability and its influence on parent–infant interactions and outcomes in the longer term. This echoes the conclusion of Bellaïche et al that a medical consultation providing empathy and reassurance is essential for the management of the affected infants and parents. The emphasis here is on infant crying problems as social and family matters rather than as an infant medical condition.
Di Lorenzo's second article examines other functional GI disorders (FGIDs) that occur in the first 24 months of age. In particular, he reviews the evidence on infantile dyschezia, rumination syndrome, and functional diarrhea, pointing out that 2 other FGIDs, cyclic vomiting syndrome and functional constipation, are more common in older children and do not have distinct characteristics when presenting earlier in life. The Rome symptoms-based criteria have been developed to facilitate diagnosis of FGIDs and avoid extensive, usually fruitless, testing.
Di Lorenzo notes interesting parallels between regurgitation and crying, in that both are common in healthy infants, peak during the early months of age (at 4–6 months in the case of regurgitation), then decline spontaneously. Like crying, regurgitation also is a common source of parental concern and, as a result, is probably overtreated, given that it usually resolves spontaneously and has benign outcomes. Drawing on expert guidelines and more recent evidence, Di Lorenzo provides a concise summary for how and when to evaluate and treat an infant with regurgitation and suspected gastroesophageal reflux. The prevalence of infant rumination is unknown, but it is probably most common in infants with severe neurodevelopmental disorders. Behavioral treatments for rumination usually are effective. Infant dyschezia also is known as grunting baby syndrome based on the behavior of the infant who seems to struggle to have a bowel movement. Unlike infant colic, the distress occurs only during the act of defecation. Treatment involves reassuring parents that the problem will resolve itself and preventing excessive intervention. Functional diarrhea also typically resolves spontaneously with age. Treatment involves parental reassurance and dietary management.
The final article is Heine's review of GI food allergy and intolerance in infants and young children. He distinguishes IgE-mediated from non-IgE-mediated food allergy and both of these from nonimmunological intolerance reactions to foods, such as fat and carbohydrate malabsorption. His review summarizes what is known about the prevalence, causes, and treatments of GI food allergy (including food protein–induced enteropathy, enterocolitis syndrome, and proctocolitis), and food intolerances (including lactose intolerance and malabsorption, fructose malabsorption, celiac disease and nonceliac gluten sensitivity). IgE-mediated allergic reactions to foods (raw egg, peanut, sesame, shellfish, or cow's milk) are common, with an estimated prevalence of approximately 10% in 1 year olds. Symptoms of allergic GI disorders in infancy include an array of disorders, including persistent regurgitation, vomiting, chronic diarrhea, feeding difficulties, unsettled behaviors, and sleep pattern disturbance. Except for an estimated prevalence of approximately 2% for cow's-milk allergy noted above, the prevalence of non-IgE-mediated food allergy—and how many cases involve crying, distress, or discomfort behavior in particular—are unknown. Progress has been made in understanding genetic and other individual vulnerability mechanisms while failure to thrive and other signs of ill health often are present. Diagnosis remains a complex process, typically involving dietary elimination before the precise causation is confirmed.
Taken together, the 10 articles in this review present different theoretical viewpoints about the causes of prolonged infant crying and discomfort, but there is a good deal of consensus about the implications of current evidence for treatments and further research.
Implications of the Evidence for Treatments
1. Prolonged crying is common in the first 3 months of infancy. The prevalence figure depends on the criteria used, but rates of approximately 20% often are reported. Most of these infants are healthy, stop crying without intervention, and have normal long-term development. There is a lack of evidence that prescribed and nonprescribed medications and preparations are safe and effective in reducing crying.
2. It follows that when prolonged crying is the sole presenting symptom, treatments based on medications and preparations are not indicated. Currently, treatments of this sort for the crying are overprescribed (12), as is the case for regurgitation (see Di Lorenzo's second article). In particular, there is no evidence that treatment for gastroesophageal reflux is effective in ameliorating such cases.
3. More research is needed to establish whether colic has any impact on the development of functional and/or psychological problems later in life. In the meantime, the main driver for the use of medicinal treatments is parents’ concern about their baby's health and well-being. Rather than infant treatments, the implication is that health service needs to provide parents with better information and support. Provisional guidelines for this purpose have been put forward (13) and research is needed to evaluate and improve them in routine health service use.
4. A minority of 1- to 3-month-old infants who cry frequently have organic disturbances, including GI disorders. The prevalence is unknown, but is probably <10%, and perhaps <5% of infants (1,13). Many of these infants present clear symptoms of ill health, including failure to thrive, diarrhea, vomiting, or eczema. Health services need the resources to distinguish, prioritize, and provide effective treatments for these cases.
5. There is substantial evidence that cow's-milk allergy causes prolonged crying in approximately 1% to 2% of infants, or 5% to 10% of cases referred to clinicians because of prolonged crying. Detection and treatment of these cases involves dietary elimination and challenge. Nutritional progress and growth need to be monitored carefully by suitably qualified professionals and diet to be normalized as soon as can be tolerated.
6. Prolonged crying after 4 months of age occurs in approximately 5% of infants and is associated with poorer developmental outcomes, whereas symptoms of possible GI disturbance after 4 months of age and severe regurgitation after 6 months of age suggest a more serious illness. Health services need to identify, treat, and monitor these cases.
7. The use of L reuteri supplements to prevent or treat prolonged infant crying offers a potentially exciting development that may change our understanding of the factors involved. The evidence is promising but insufficient and does not provide a basis for routine adoption of probiotic supplements; rather, the findings as a whole point to the need for further research. Additional studies are under way and should provide a clearer understanding within the next 2 years.
Implications of the Evidence for Further Research
1. The hypothetical explanations for how GI disturbance could give rise to prolonged infant crying put forward in this supplement need to be tested in different groups of infants at different ages.
2. Likewise, the hypothesized developmental processes need to be examined by studies that track whether changes in infant psychological and CNS functions are reflected in changes in crying.
3. The 2 schemes proposed by expert clinicians for diagnosing infant colic crying cases (the Rome III and Wessel criteria) are difficult to apply in practice and may be flawed in conceptualizing the infants involved as a distinct group. Research is needed to develop more effective assessment and diagnostic schemes. In particular, methods are needed that measure infant crying and associated behavioral symptoms objectively and that distinguish these from parental concerns and vulnerabilities.
4. The evidence suggests that in addition to prolonged crying, many infants with GI disturbance can be distinguished by growth faltering, failure to thrive, diarrhea, vomiting, or eczema. Research is needed to show what proportion of GI cases are correctly identified, wrongly identified, and overlooked by assessments of these symptoms.
5. Detection of cow's-milk allergy and other GI food intolerances requires dietary interventions that are cumbersome and may be difficult to implement in practice. Evidence about how often these interventions are effective and what factors enhance or hamper their effectiveness in routine health services use is needed.
6. There is a need for research that develops cost-effective tests or protocols that distinguish GI abnormalities.