Hill, David J.; Hosking, Clifford S.
Fussing and crying, especially in the evening, are normal developmental phenomena in infants in the first 3 months of life (1). Unexplained paroxysms of irritability, fussing, or crying that persist for more than 3 hours a day, for more than 3 days a week (2) are said to represent a separate clinical condition termed colic (3). During such episodes the legs may be drawn up to the abdomen, and the infant may become flushed. Abdominal distension and increased passage of flatus are often noted. However, the term colic implies a mechanism responsible for the distress displayed by these infants. Such a mechanism has never been demonstrated. For this reason many recent workers have referred to these infants as exhibiting crying, fussing, or distressed behavior.
In the following discussion the term colic will be used interchangeably with each of these terms, although it is to be emphasized that use of this term in no way implies the cause of the infant's distress.
Table 1 shows the varying incidences of colic reported from different countries (4). Most infants in these studies had severe colic, although no criteria have yet defined mild, moderate, or severe colic.
Broadly, pediatricians managing infantile colic take one of two approaches: 1) They assume that colic is an extreme form of normal crying behavior. The abdominal distension and excess flatus are attributed to air swallowing. Colic is managed with a behavioral approach. 2) They assume that colic is an entity distinct from normal crying behavior and results from an adverse reaction to foods.
This review will discuss:
1. Evidence that infants with colic differ significantly in their pattern of distressed behavior from noncolicky infants.
2. The incidence of colic in children with known food allergy.
3. The incidence of food allergy in children with colic.
4. The outcome of trials focusing on the dietary management of colic.
5. The outcome of studies involving behavior modification in colic.
6. The outcome of a preliminary trial of modification of maternal diet in very young infants with colic.
7. Findings of a systematic review of different treatments for colic.
Some of the data presented in this paper have been published elsewhere (5).
EVIDENCE THAT INFANTS WITH COLIC DIFFER SIGNIFICANTLY IN THEIR PATTERN OF DISTRESSED BEHAVIOR FROM NONCOLICKY INFANTS
Brazelton (1) used parental recording on crying charts to document the natural history of distressed behavior in infancy. Figure 1 summarizes the pattern of distressed or fussing behavior in a group of 80 noncolicky infants studied in the first 12 weeks of life. Brazelton showed how distressed behavior frequently deteriorated until the children were approximately 6 weeks of age and then gradually improved. Both he and Wessel et al. (2) described the predominance of fussing behavior during the late afternoon and evening, although Wessel noted that distress was not confined to this period in one third of infants. As will be discussed, our studies suggest that this pattern of distress may be influenced by whether the infant is breast or formula fed.
Brazelton contrasted the pattern of distressed behavior in noncolicky normal infants with that of a group of colicky infants. He found that the colicky infant's distress was more marked, peaked later than the normal age of 6 weeks, and persisted to the 12-week follow-up (1).
The distress diaries of Wessel et al. and Brazelton were not validated against objective methods for measuring crying; however, Barr et al. (6) developed a 24-hour crying chart validated against objective measurements of infant distress. This group compared information recorded by parents on specifically designed crying data sheets with data voice-activated audio-tape recordings (VAR) of negative vocalizations (i.e., crying) during a 24-hour period. Their study of 10 infants showed a moderate correlation between VAR “negative” vocalization clusters and (1) the frequency of diary-recorded episodes of crying or fussing behavior (R = +0.64;P = 0.03) and (2) the duration of crying (R = +0.67;P = 0.02).
Elimination of one outlier considerably strengthened the correlation between the VAR objective data and episodes of crying-fussing behavior (R = +0.85) and duration of crying (R = +0.90). Using these validated crying charts, Hunziker and Barr (7) confirmed the natural history of distressed behavior previously described by Brazelton (1).
Recent studies from our center have compared the pattern and duration of distressed behavior in a group of 30 colicky and noncolicky infants (8). Figure 2 shows the higher levels of distressed behavior in the colicky infants compared with that in the noncolicky infants. The findings were confirmed in a separate group of 90 colicky infants. The evaluation of distressed behavior on an hour-by-hour basis confirmed the predominance of nocturnal symptoms, but similar to Hide and Guyer (9), we found colic and distressed behavior frequently occurred in colicky infants during other periods (Fig. 3). Table 2 shows the duration of distress broken down into 6-hour periods, highlighting the increased distress in colicky compared with noncolicky infants over these four 6-hour periods.
COMPARISON OF TIMINGS OF DISTRESS BETWEEN BREAST-FED AND BOTTLE-FED COLICKY INFANTS
In our colicky patients, the levels of distress at different times during 24 hours were compared between the formula-fed and breast-fed infants. Although the median total distress levels were similar during 24 hours (300 minutes vs. 325 minutes), formula-fed infants showed more distress before 12 noon than breast-fed infants (90 vs. 55 min;P = 0.05), whereas breast-fed infants were more distressed than bottle-fed infants after noon (263 minutes vs. 150 minutes;P = 0.03). When results were analyzed on an hourly basis, formula-fed infants showed significantly more distress at 9:00 A.M. (P = 0.01). In Figure 4, levels of distress behavior are shown during 6-hour periods. The bottle-fed colicky infants showed no significant difference in median distress times during these four 6-hour periods (20 minutes, 75 minutes, 60 minutes, 105 minutes), whereas, breast-fed infants showed more distress in the two 6-hour periods after 12 noon (10 minutes, 30 minutes, 95 minutes, 120 minutes;Fig. 4).
Relevant to these observations Axelsson et al. (10) noted that β-lactoglobulin appeared in the breast milk 4 hours after cow's milk ingestion, but the highest concentrations of β-lactoglobulin in breast milk were found 8 to 12 hours after ingestion. Paganelli et al. (11) demonstrated cow's milk–complexed antigen rapidly appeared in the serum within an hour of ingestion. Thus, if milk protein intolerance contributes to distress behavior, formula feeding with high-dose antigen exposure is more likely to elicit a rapid-onset distress response than is prolonged low-dose antigen exposure through breast milk. These observations are consistent with the different timing of distress behavior in our breast-fed and bottle-fed infants with colic.
Together these data suggest that the distressed behavior of colicky infants may not simply represent an exaggeration of the normal distressed pattern of noncolicky infants. Hunziker and Barr (7) reported that “supplemental carrying” reduced distress behavior in noncolicky infants but this was not found in colicky infants (12), results that are consistent with this hypothesis.
INCIDENCE OF COLIC IN CHILDREN WITH KNOWN FOOD ALLERGY
Studies from our group have used cow's milk allergy (CMA) as a model of food allergy. Description of the effects of cow's milk challenge in young children with suspected CMA demonstrate the diverse manifestations of this disorder (13). Associated markers of immune hypersensitivity mechanisms have also been reported (14–17). In a sequential group of 100 patients with challenge-proven CMA (median age, 16.2 months) 44% of infants displayed irritable and colicky behavior during the cow's milk challenge procedure (13).
A clustering algorithm (K-means) identified three groups of patients with common clinical features (13). Predominant clinical findings in each of the three groups of patients along with relevant immunologic markers are shown in Tables 3 and 4. Children who had immediate reactions (group 1), reacted to small volumes of cow's milk within 1 hour of beginning the cow's milk challenge procedure. In contrast, the children in group 2 (intermediate reactors) usually tolerated 60 to 200 ml of cow's milk before symptoms of vomiting and diarrhea developed over several hours. The third group of patients usually tolerated near-normal volumes of cow's milk for 24 to 72 hours before symptoms of CMA developed.
From these data, each of the groups showed statistically significant differences in incidence of challenge-induced skin eruptions, vomiting, diarrhea, and respiratory symptoms. By contrast, the incidence of distressed behavior (i.e., colic elicited by cow's milk challenge) was the same in the three groups. As shown in Table 3, each of the clinical groups exhibited different immunologic features. Thus, colic is not associated with any one immunologic marker of CMA, in that the children with immunoglobulin (Ig)E immediate-type hypersensitivity (group 1), non-IgE–associated enteropathy (group 2), or delayed onset T-cell associated disease (group 3) all showed features of colic.
One other point to emerge from these studies was that children with hypersensitivity to cow's milk frequently demonstrated similar reactions to other foods, including eggs, peanuts, nuts, wheat, soy formula, and extensively hydrolyzed casein and whey hydrolysate preparations (18,19).
Thus, children with true CMA often show colic and distressed behavior as one manifestation of the disease. It remains less clear how many children with colic, putatively due to CMA, fail to display other manifestations of CMA.
INCIDENCE OF FOOD ALLERGY IN CHILDREN WITH COLIC
Iacono et al. (20) put 70 cow's milk formula–fed infants with severe colic on a soy milk formula. After 1 week, the parents of 50 of the infants reported improvement of colic that relapsed within 24 hours after cow's milk was reintroduced into the infants' diets. Within 3 weeks, 8 of 50 infants showed effects of soy allergy, and at the age of 9 months, 18 of 50 patients had other symptoms of CMA at challenge. Only 1 of 20 of the patients in whom colic did not respond initially to soy milk showed other features of CMA (P < 0.02).
Lothe et al. (21) noted a similar phenomenon. Of 43 infants with colic who responded to exclusion of cow's milk, 18% showed other features of CMA by the age of 6 months and 13% retained these features to at least 12 months of age.
Although the severity of colic at initial examination in these infants may have overshadowed other minor features of CMA, it is more likely that colic represented one early manifestation of true food protein hypersensitivity.
Sleep disturbance is a major feature in infants with colic. Kahn et al. (22) identified 15 (median age, 13 months) in whom sleep disturbance resolved within 5 weeks of beginning a cow's milk–free diet. In a subsequent double-blind, placebo-controlled challenge, this sleep disturbance recurred within 4 days of the reintroduction of cow's milk. The effect of milk exclusion and reintroduction was monitored by the use of polysomnographs to document arousal and sleep disturbance patterns and was evaluated by measurement of skin water evaporation during non–rapid-eye-movement sleep. Gastroesophageal reflux and other cardiorespiratory dysrhythmias as a source of the distressed sleep pattern were excluded. Kahn et al. did not determine how many children had symptoms other than sleep disturbance and distressed behavior after reintroduction of cow's milk, but more than half showed features of eczema, vomiting, and diarrhea before receiving the cow's milk–free diet.
TRIALS OF DIETARY MANAGEMENT OF COLIC
After a preliminary study (23), Jakobsson et al. (24) noted that one third of breast-fed colicky infants had remission and then relapse of colic when mothers excluded and subsequently reintroduced cow's milk. Evans et al. (25), however, were unable to confirm these findings but associated increased colic with increased variety of foods in the maternal diet. Maternal ingestion of cow's milk, egg, chocolate, fruit, and nuts was implicated in producing symptoms. Lothe et al. (21) reported that 11 of 60 colicky infants receiving cow's milk formula responded to soy formula; another 32 improved after administration of a casein hydrolysate formula. These preliminary studies have been criticized because some of the populations were selected (25), the numbers were too small to identify a real effect (26), the protocol was not truly double-blind (22), and objective data of crying or fussing behavior were not obtained (26).
Other investigations of the role of diet in colic have addressed some of these shortcomings. Lothe et al. (27) implemented a 5-day cow's milk–free diet using casein hydrolysate. A marked diminution of distress symptoms occurred in 24 of 27 colicky infants. In these infants, the total crying time decreased from 5.6 hours to 0.7 hours (P < 0.001). The 24 responding infants then entered into a randomized, double-blind crossover trial of whey protein formula. The active or placebo challenge was conducted on a single day with an intervening 3-day washout period. Crying and disturbed sleep were more prevalent in the whey-feeding phase (whey, 3.2 ± 2.4 hours vs. placebo 1.0 ± 1.6 hours); (P < 0.01). Of the 24 infants challenged, 18 (two thirds of the original study population) demonstrated increased distress on challenge with milk protein.
In a crossover design of 17 colicky infants, Forsyth (28) ensured the smell of formula would not lead to unblinding of the study. Casein hydrolysate, or casein hydrolysate and cow's milk formula were fed alternately for four periods, each of which lasted 4 days. Significant decreases in distressed behavior were noted in the first two formula-change periods only. During the four formula challenge periods, only two (11.8%) of the infants showed a reproducible effect of formula change on colic behavior. Previously, Lothe et al. (21) had suggested approximately 25% of infants with colic responded to dietary change.
The conclusions of Forsyth's study have been questioned because of the small number of patients studied and the accuracy of diary recording for 16 days. In addition, critics have noted that the volume of milk protein administered in challenge was only half that normally provided to infants of this age. Furthermore, the effect of returning these infants to normal volumes of cow's milk formula upon completion of the study was not recorded. Forsyth concluded that diet was likely to be only one factor in the causation of colic (28). He particularly drew attention to the feelings of helplessness, frustration, and decreased confidence in parenting ability that parents of colicky patients experience. Results of these studies are summarized in Table 5.
STUDIES OF BEHAVIOR MODIFICATION IN COLIC
In a series of studies, Taubman (29) demonstrated the effectiveness of parental counseling in the management of distressed infants. In the first phase of a study of 21 colicky infants, he found that parental counseling reduced distressed behavior to an extent similar to that of introduction of a cow's milk–free diet. In the second phase of his study, however, patients who had received only dietary treatment in the first phase returned to their prestudy diets while their parents received counseling. The distressed behavior of these infants with diet-responsive colic further decreased with parental counseling. Taubman concluded that in many infants who have colic the crying results from parental misinterpretation of infant cries and is not caused primarily by milk protein allergy. The relatively small number of patients in this study and the difficulty of blinding counseling procedures bring the validity and generalizability of Taubman's findings into question. Nevertheless, they emphasize the importance of parental counseling in the management of infantile colic.
Hunziker and Barr (7) also implied that distressed behavior in infancy reflects parental misinterpretation of normal crying behavior. After their observation that normal infants showed less crying behavior when regularly nurtured by supplemental carrying, Barr et al. (12) studied the effect of supplemental carrying on 66 colicky infants. Overall, the treatment group carried their infants 6.1 hours per day (2.2 hours more than the control group) throughout the intervention period. At 6 weeks of age, when a significant effect of treatment was expected, none could be shown. Hunziker and Barr concluded that this difference in response to carrying may be due to an underlying pathologic process such as protein hypersensitivity or irritable bowel.
Studies by Rautava et al. (30) suggest an important role for maternal distress during pregnancy and childbirth and unsatisfactory sexual relationships, but not for sociodemographic factors, in the cause of colic in Finnish infants. Wolke et al. (31) examined the effect of different behavior strategies in 92 mother and colicky infant pairs. After 3 months, all of the infants' distress had improved. Infants whose mothers received advice on behavior modification had a 51% reduction in distress, however, compared with infants of those mothers received empathic support (37%) and infants in the control group (35%). Of note, all but 3 of the 92 infants in this study were ingesting hypoallergenic formula during the study period. Thus, although behavior modification clearly reduced distress in this cohort of colicky infants, the role of hypoallergenic formula in influencing the outcome of the study was not examined. The results of these studies are summarized in Table 6.
The Melbourne Colic Study
The role of diet was examined in an investigation of 115 colicky infants who reported distress for more than 3 hours per day for more than 3 days per week for more than 3 weeks (32). Infants were referred from community-based pediatric facilities and were studied during a 1-week period. All mothers of breast-fed infants were provided an artificial color-free, preservative-free, additive-free diet program. In addition, those assigned the active low-allergen diet excluded not only cow's milk but also other common food allergens including egg, wheat, peanut, nuts, fish, and shellfish. Formula-fed infants were assigned a casein hydrolysate preparation (low-allergen diet) or cow's milk–based formula. In the double-blind, placebo-controlled study, the response to diet was assessed by comparing the level of distress behavior at the outset and at the end of 1-week diet treatment. Parents recorded distress levels on the previously validated infant distress charts described earlier.
If successful outcome was defined as a reduction in distress of 25% or more, after adjusting for age and feeding mode, infants with an active diet had a significantly higher rate of improvement than those with the control diet (odds ratio [OR] = 2.32; 95% confidence interval [CI] 1.07–5.0;P = 0.03). In addition, the results were assessed by comparing the distress ratio on day 8 to that on day 1 for infants assigned the active diet, compared with those assigned the control diet. Distress was reduced by 39% (95% CI [26–50]) in infants receiving the active diet compared with 16% (95% CI, [0–30]) for those consuming the control diet. After adjusting for age and feeding mode, these differences proved statistically significant (P = 0.012). This finding translates into a median reduction of 117 minutes in distress scores for infants consuming the active diet, compared with 46 minutes for those consuming the control diet.
Those results were achieved by studying a cohort of infants whose contact was restricted to the study period. Subjects entered the study irrespective of previous use of anticolic medications, dietary manipulation, or presence of psychosocial factors likely to contribute to colic. The findings have general applicability, because infants were referred by community-based pediatricians, family practitioners, and child health practitioners. They had distress times comparable to infants in other investigations of diet and behavior intervention strategies in colic.
The apparent benefits of diet in this study must be tempered by the following limitations of the study, however. First, a high withdrawal rate was noted with 36 infants not completing the full study, though no difference was observed whether active or control diets were assigned to patients, nor did socioeconomic factors differ between the withdrawals and those who completed the study. Second, in determining sample size, based on the data available, the incidence of spontaneous improvement in colic was underestimated by as much as 30% in the different groups. This meant that only one group of patients (breast-fed infants older than 6 weeks) included a sufficient number of patients to adequately test the effect of diet in different subgroups.
One striking feature that emerged from the study was the difference in the effect of the two diet programs on breast-fed infants less than 6 weeks of age. In this age group, only infants receiving the active diet decreased their distress, whereas those with the control diet increased distress during the study period. The breast-fed infants younger than 6 weeks of age receiving the active diet decreased distress by 24% (73 minutes per 24 hours), whereas those consuming the control diet increased distress by 34% (67 minutes per 24 hours). This difference in the response to diet was statistically significant (P = 0.006). When these data were examined according to whether the infants reduced distress by 25% or more, 9 of 12 infants assigned the active diet reduced distress according to this parameter, whereas only 1 of 10 assigned the control diet achieved this reduction (P = 0.015).
The findings of this study suggest that diet is one factor that contributes to colic in infants, but it was unclear whether this effect was global or whether only a subpopulation of distressed infants are responsive. The results suggest that, in particular, breast-fed infants younger than 6 weeks of age may benefit from this intervention; in older infants, the incidence of spontaneous remission of colic is such that a large number of patients must be studied to detect an effect of diet change. These findings suggest that some infants have colic because of transient protein intolerance that improves, beginning at approximately 6 weeks of age.
In our study, the degree of dietary restriction on mothers with breast-feeding infants was severe. If this approach is to be followed for more than 1 week, nutritional support for such mothers is essential.
DIET MODIFICATION IN VERY YOUNG COLICKY INFANTS
The Melbourne Colic Study identified a group of very young infants, less than 6 weeks of age, whose distress improved markedly if mothers consumed a low-allergen diet. In a subsequent open pilot study, mothers of 17 breast-fed colicky infants less than 6 weeks of age were given a low-allergen diet for 2 weeks. The low-allergen diet was that used in The Melbourne Colic Study, but, to ensure optimal nutrition, it was supplemented by an amino acid–based preparation, Elemental O28 (SHS, Liverpool, UK). This ensured that daily energy intake exceeded 2600 calories, whereas the previous low-allergen diet used in The Melbourne Colic Study provided only 1900 calories per day and was inadequate for prolonged use.
Of the 17 patients, 9 showed reduced distress behavior by more than 25% during the study period; 7 of these infants demonstrated relapse of distress behavior within 48 hours after the maternal diet was normalized. These results are summarized in Table 7. The significance of these findings and their clinical application must now be tested prospectively.
SYSTEMATIC REVIEW OF DIFFERENT TREATMENTS IN COLIC
A recent systematic review (33) of more than 300 research articles on the treatment of infantile colic confirmed that hypoallergenic milk formula but not antacids alone or low lactose formula were effective in the treatment of colic. However, dicyclomine, an anticholinergic agent, was found in this review to be the most effective treatment for colic. This drug cannot be recommended because of the potential serious side effects it may have in an essentially self-limiting condition. Nonetheless, the effectiveness of a drug that alters gut motility may give a clue to the mechanism of distress in some infants with colic. These data, combined with the observation that infants who are destined for the development of colic show increased levels of the gut hormone motilin, which modulates gastrointestinal motility, in blood at birth as well as in infants with colic (34) should stimulate new studies on the role of gut motility in the cause of distress behavior in colicky infants.
Colic is a clinical entity that can be defined and quantified. No general consensus has emerged about its cause, which is likely to be multifactorial. The two major schools of thought attribute colic to either a disturbance in the parent–child interaction (behavioral) or a food protein hypersensitivity reaction (allergy) with perhaps either or both these effects manifesting in an infant predisposed to gut motility disturbance. The following observations may help to reconcile the apparent differences between these conflicting views.
First, Barr et al. (12) observed that colicky infants younger than 6 weeks respond poorly to a form of behavior modification shown to benefit normal crying infants (6). Second, Forsyth (28) noted that the response to diet change in formula-fed infants with colic is transient. Our group's study (32) demonstrated that breast-fed infants less than 6 weeks of age respond particularly well to a hypoallergenic diet. Finally, Wolke et al. (30) noted that older infants with excessive crying who were ingesting hypoallergenic formula diets responded to behavior modification.
Thus, we hypothesize that in colicky infants, transient food protein intolerance associated with disturbance in gut motility in the first weeks of life (34) leads to distress that may then persist, due to behavior patterns and secondary disturbance in parent-infant interactions.
Question: The association with colic and possibly CMA and diet is intriguing. Undeniably, a group of patients with colic improve while consuming a hydrolized formula such as Nutramigen. Is it possible that the mechanism is physiologic rather than immunologic? Does the size of the particle and the ease of absorption through the gastrointestinal tract determine whether there is an immunologic reaction and ergo a diagnosis of CMA?
Dr. Hill: With regard to immunologic markers associated with colic, I am not aware of any systematic studies that have examined the possible basis of food protein intolerance in these infants.
Question: If we assume that there is an improvement in some cases of colic when the infant is fed a hydrolized formula, the mechanism by which that occurs is not necessarily immunologic.
Dr. Hill: There are two studies that highlight the importance of intolerance to milk protein, compared with some other component of cow's milk formula or breast milk, as the cause of colic. The first, by Jakobsson (23), showed that, in a double-blind placebo-controlled trial, when whey protein in gelatin capsules was reintroduced to the mother's diet after exclusion of cow's milk from the maternal diet, the patient experienced a relapse of colic. The other study in bottle-fed infants by Lothe et al. (27) showed remission of symptoms on exclusion of cow's milk from infants' diets, and relapse when lactose-free whey protein in capsules was reintroduced. Sequential analysis of both these studies showed a high correlation between infantile colic and exposure to milk protein.
Question: How about those patients who improve with hydrolyzed formula? What is their subsequent clinical course? Do they have other manifestations of CMA?
Dr. Hill: Iacono et al. (20) and Lothe et al. (21) showed that 20% to 40% of infants with colic who respond to a milk-exclusion diet show other features of cow's milk protein intolerance when challenged 3 to 6 months after initial diagnosis.
Question: Did they look at predictors of allergy in the infants, such as family history of allergic disease or poor IgE, in the patients who did not respond to the change in diet?
Dr. Hill: No.
Question: How about the effect of adding counseling to diet changes. Has that been looked at?
Dr. Hill: Yes. Taubman's (29) data compared the effect of counseling and change in diet in 20 infants (median age, 6 weeks) who were studied for 9 days. A significant decline in distress levels was noted in the 10 infants whose parents received counseling and in the 10 breast-feeding infants whose mothers were placed on a cow's milk–free diet or bottle-fed infants who received casein hydrolyzed formula. The number of patients and the design of the study make it unclear whether these effects of diet change and counseling may be additive.
Question: This suggests that the effect of counseling might be additive.
Dr. Hill: Taubman's (29) data and the data of Wolke et al. (31) confirm the importance of parental counseling in the management of colic.
Question: Are there features that distinguish the child who is going to be responsive to counseling versus the one who is not going to respond? What is the role of diet?
Dr. Hill: In the Melbourne Colic Study (32) the subgroup of infants who did best were those breast-fed infants less than 6 weeks of age whose mothers received a low-allergen diet. There is good evidence that much distress behavior in infants starts remitting after the age of 6 weeks. Together, these findings suggest that much colic during this period may be due to a dietary protein intolerance which may be a transient phenomenon. Perhaps the continuing distress behavior, which responds to parental counseling, reflects a secondary disturbance in family dynamics consequent upon distress in infancy. These observations should be tested prospectively.
Question: What is your interpretation of the efficacy of low-lactose formulas for infants? What do you think of the trenchant lactose intolerance that was originally alluded to in some reports?
Dr. Hill: Data from a Cochrane evidence-based medicine review of treatments of infantile colic found a low-lactose diet was not of benefit in the management of infants with colic behavior (33).
Comment: We have looked at a few children who are clearly IgE-mediated cow's milk sensitive, who respond on challenge to cow's milk, who drink lactose-free formula, and who have positive skin test responses to cow's milk formula and negative skin test responses to lactose-free formula. These children also respond when fed the enzyme in cow's milk, lactase. Apparently, a single company in New Zealand provides this lactose-free formula. They use a process that involves heating and some denaturation that does something to the protein as well as taking the lactose out. Clearly, this lactose-free formula is not just lactose-free.
Question: I almost never hear low socioeconomic status families reporting colic. Is this an observation you have made? Also, do you think there is a population of children who never see a specialist and who are put through “formula roulette” by the general practitioner or pediatrician who tries rotating the infant through a number of formulas until the level of the infant's irritability is acceptable to the family? Do you think there are a large number of infants who are probably responding adversely to their formulae but who never see a physician?
Dr. Hill: These questions can be addressed by conducting community-based studies. Participants in such research programs are probably going to be of middle-class socioeconomic status. One can expect that other factors, particularly disturbance in family dynamics, are greater in low socioeconomic status families; one can only guess at the impact of this disturbance on colic behavior in infants. Prospective studies should further define a group of infants who are likely to be most responsive to dietary intervention measures. Once those clinical profiles have been defined, the practice of formula roulette should be minimized.
Question: Are the patients in the Melbourne study involved in a long-term follow-up study as well? What would such a study show?
Dr. Hill: The Melbourne Colic Study was only of 1 week's duration. We have not had the opportunity to observe these patients long-term. I have referred to the long-term studies of Iacono et al. (20) and Lothe et al. (27) showing that 20% to 40% of infants with colic who respond to dietary exclusion programs subsequently show features of true cow's milk protein intolerance 3 to 6 months after initial diagnosis when challenged with cow's milk protein.
The authors thank Penny Millar for her assistance.
1. Brazelton TB. Crying in infancy. Pediatrics 1960; 26:579–88.
2. Wessel MA, Cobb SC, Jackson EB, et al. Paroxysmal fussing in infancy: sometimes called “colic.”Pediatrics 1954;14:421–4, 434.
3. Illingworth RS. Three months of colic. Arch Dis Child 1954; 29:165–74.
4. Lothe L. Studies on infantile colic. Thesis; University of Lund, Malmo, 1989.
5. Hill DJ, Hosking CS. Infantile colic and food hypersensitivity. In: Metcalfe DD, Sampson HA, Simon RA, eds. Food Allergy: Adverse reactions to foods and food additives. 2nd ed. Boston: Blackwell Science, 1997: 259–68.
6. Barr RG, Kramer MS, Boisjoly C, et al. Parental diary of infants cry and fuss behaviour. Arch Dis Child 1988; 63:380–7.
7. Hunziker VA, Barr RG. Increased carrying reduces infant crying: A randomized control trial. Pediatrics 1986; 5:641–7.
8. Hill DJ, Menahem S, Hudson I, et al. Charting infant distress: An aid to defining colic. J Pediatr 1992; 121:755–8.
9. Hide DW, Guyer BM. The prevalence of infant colic. Arch Dis Child 1982; 57:559–60.
10. Axelsson I, Jakobsson I, Lindberg T, Benediktsson B. Bovine beta-lactoglobulin in human milk. Acta Paediatr Scand 1986: 75:702–7.
11. Paganelli R, Atherton DJ, Levinsky RJ. Differences between normal and milk allergic subjects in their immune responses after milk ingestion. Arch Dis Child 1983; 58:201–6.
12. Barr RG, McMullan SJ, Speiss H, et al. Carrying as colic “therapy”: A randomized control trial. Pediatrics 1986; 87:623–30.
13. Hill DJ, Firer MA, Shelton MJ, et al. Manifestations of milk allergy in infancy: Clinical and immunological findings. J Pediatr 1986; 109:270–6.
14. Firer MA, Hosking CS, Hill DJ. Humoral immune response to cow's milk in children with cow's milk allergy. Int Arch Allergy Appl Immunol 1987: 84;173–7.
15. Hill DJ, Ball G, Hosking CS. Clinical manifestations of cow's milk allergy in childhood: I. Association with in vitro cellular immune responses. Clin Allergy 1988; 18:469–79.
16. Hill DJ, Duke AM, Hosking CS, et al. Clinical manifestations of cow's milk allergy in childhood: II. The diagnostic value of skin tests and RAST. Clin Allergy 1988; 5:481–90.
17. Hill DJ, Ball G, Hosking CS, Wood PR. Gamma-interferon production in cow milk allergy. Allergy 1993; 48:75–80.
18. Bishop JM, Hill DJ, Hosking CS. Natural history of cow milk allergy: Clinical outcome. J Pediatr 1990; 116:862–7.
19. Hill DJ, Cameron DJS, Francis DEM, Gonzalez A, Hosking CS. Challenge confirmation of late onset reactions to extensively hydrolysed formula in infants with multiple food protein intolerance. J Allergy Clin Immunol 1995; 96:386–94.
20. Iacono G, Carroccio A, Montalto G, et al. Severe infantile colic and food intolerance: A long term prospective study. J Pediatr Gastroenterol Nutr 1991; 12:332–5.
21. Lothe L, Lindberg T, Jakobsson I, et al. Cow's milk formula as a cause of infantile colic: A double-blind study. Pediatrics 1982; 70:7–10.
22. Kahn A, Mozin MJ, Rebuffat E, et al. Milk intolerance in children with persistent sleeplessness: A prospective double-blind crossover evaluation. Pediatrics 1989; 84:595–603.
23. Jakobsson I, Lindberg T. Cow's milk proteins cause infantile colic in breast-fed infants. Lancet 1978; 312:437–439.
24. Jakobsson I, Lindberg T. Cow's milk proteins cause infantile colic in breast-fed infants: A double-blind crossover study. Pediatrics 1983; 71:268–71.
25. Evans RW, Fergusson DM, Allardyce RA, et al. Maternal diet and infantile colic in breast-fed infants. Lancet 1981; 315:1340–1342.
26. Sampson HA. Infantile colic acid and food allergy: fact or fiction. J Pediatr 1989: 115:583–4.
27. Lothe L, Lindberg T. Cow's milk whey protein elicits symptoms of infantile colic in colicky formula-fed infants: A double-blind crossover study. Pediatrics 1989; 83:262–6.
28. Forsyth BW. Colic and the effect of change in formulas: A double-blind multiple-crossover study. J Pediatr 1989; 115:521–6.
29. Taubman B. Parental counselling compared with elimination of cow's milk or soy milk protein for the treatment of infantile colic: A randomised trial. Pediatrics 1988; 81:756–61.
30. Rautava P, Helenius H, Lehtonen L. Psycho-social predisposing factors for infantile colic. BMJ 1993: 307:600–4.
31. Wolke D, Gray P, Meyer R. Excess infant crying: A control study of mothers helping mothers. Pediatrics 1994; 94:322–32.
32. Hill DJ, Hudson IL, Sheffield LJ, Shelton MJ, Hosking CS. A low allergen diet is a significant intervention in infantile colic: Results of a community-based study. J Allergy Clin Immunol 1995; 96:886–92.
33. Lucassen PL, Assendelft WJ, Gwbels JW, van Eijk JT, van Geldrop WJ, Knuistingh Neven A. Effectiveness of treatment for infantile colic: Systematic review. BMJ 1998; 316:1563–9.
34. Lothe L, Ivarsson SA, Ekman R, Lindberg T. Motilin and infantile colic. Acta Paediatr Scand 1990;79:4:410–6.
35. Boulton TJ, Rowley MP. Nutritional studies during early childhood: III. Incidental observations and temperament, habits, and experiences of ill-health. Aust Paediatr J 1979; 15:87–90.
36. Thomas DB. Aetiological associations in infantile colic: An hypothesis. Aust Paediatr J 1986; 17:292–5.
37. Carey WB. “Colic”: Primary excessive crying as an infant: Environmental information. Pediatr Clin North Am 1984; 31:993–1005.
Colic; Infants; Food allergy
© 2000 Lippincott Williams & Wilkins, Inc.