Difficulty in Initiating and Maintaining Sleep Associated with COW'S Milk Allergy in Infants

Sleep 10(2):116-121, Raven Press, New York 1987, Association of Professional Sleep Societies Difficulty in Initiating and Maintaining Sleep Associated with COW'S Milk Allergy in Infants A. Kahn, E. Rebuffat, D. Blum, G. Casimir, 1. Duchateau, M. 1. Mozin, and *R. lost Pediatric Sleep Laboratory and Department of Immunology, Free University of Brussels, Belgium, and *Nestle Research Department, Switzerland Summary: To confirm that sleeplessness in infants can be related to an undiagnosed allergy to cow's milk proteins, 71 infants were studied. Group I consisted of 20 infants referred for chronic insomnia that had appeared in the early days of life. Group II was made up of 31 infants admitted for skin or digestive symptoms attributed to cow's milk intolerance; 13 of these infants were shown to sleep as poorly as the infants of group I. Group III consisted of 20 infants with no history of sleep disturbance or milk allergy. The three groups of infants were comparable for sex and age. Laboratory tests revealed immunologic reactions to milk in all the infants in groups I and II. The sleep of the insomniac infants (group I, and the 13 "poor sleepers" in group II) became normal after cow's milk was eliminated from the diet. Insomnia reappeared when the infants in group I were challenged with milk. We conclude that infants with clinically evident milk allergy may suffer from sleeplessness and that when no evident cause for a chronic insomnia can be found in an infant the possibility of milk allergy should be given serious consideration. Key Words: Insomnia Sleep-Milk-protein allergy. Persistent settling and waking difficulties, associated with disturbing behavior, restlessness and intense crying, are encountered in up to 20% of infants aged <1 year (1). The symptoms have been attributed to a variety of causes, including inappropriate parental behavior (2), adverse environmental conditions (3), chronic physical discomfort, upper airway obstructions (4), gastroesophageal reflux (5), or the delayed effects of neonatal asphyxia (6). Food allergies are occasionally reported as a cause of insomnia (5), although there is little supporting evidence. In a preliminary study we reported that the sleep of eight infants with chronic insomnia was normalized after these infants were fed a milk-free diet for a previously undiagnosed milk protein allergy (7). In the present study, we report a further group of 20 patients with severe sleeplessness attributed to milk protein. We further compare their sleep characteristics with those of patients referred for skin and digestive symptoms attributed to milk allergy and to those of infants with no symptom of allergy and no insomnia. Accepted for publication December 1986. Address correspondence and reprint requests to Dr. A. Kahn at Pediatric Sleep Laboratory, Hopital Universitaire des Enfants, CHU Brugmann, Place Van Gehuchten, 1020 Bruxelles, Belgium. 116

INSOMNIA AND ALLERGY IN INFANTS 1I7 PATIENTS AND METHODS From July 1984 to July 1986, 71 infants were selected from an outpatient clinic and from the general pediatric ward. The ethical committee approved our protocol and informed parental consent was obtained in each case. The infants were divided into three main groups according to the main complaints on presentation. Patients Group I: Insomnia. From 120 infants referred by their pediatricians for chrom~ waking and crying during sleep hours, 20 were selected because their sleep had been severely disturbed, almost since birth, and an allergy to cow's milk protein was suspected. On the first visit, an interview failed to reveal any usual cause for the infant's poor sleep. The parents were asked to keep a log for 7 days describing the child's sleep schedule. On the second visit, a standard medical and psychologic protocol was followed to rule out the most frequently reported causes for chronic insomnia in infants (1-5). It was completed by an all-night polygraphic recording, which excluded further causes of arousals, such as obstructive apneas (4) or esophageal reflux (5). Esophageal acidity was studied with a ph probe positioned 3 cm above the cardia and with continuous recording of the changes in esophageal ph. The recordings were normal for each infant, and confirmed the frequent awakenings and short total sleep time reported by the parents. When atopy was suspected because of family history of allergy and occasional mild episodes of diarrhea or skin rashes, these infants were given series of allergy tests. Cow's milk was removed from the diet for 4 weeks by feeding the infants exclusively with a new routine hypoallergenic infant formula based on trypsinic degradation of cow's milk proteins (L.H.A. Formula, Nestle). Follow-up interviews were conducted with the help of our dietitian to evaluate the child's progress. During the treatment period, the parents were asked to keep a log describing the child's sleep schedule. After 4 weeks, a second polygraphic recording was performed to confirm the reported increase in total sleep time and the decreased number of arousals. Milk was then reintroduced into the diet under close medical surveillance. The diagnosis of milk allergy was confirmed if sleeplessness and agitated behavior reappeared within 4 days (7,8). Cow's milk was again excluded from the diet, and improvement ensued within 5 days. Every infant in this group responded according to the protocol. Group II: Milk allergy. Thirty-one infants were referred with skin and digestive manifestations attributed to cow's milk allergy. After the introduction of cow's milk into their diet, the infants developed severe skin irritation, eczema (n = 15), repeated airway infections, vomiting, or diarrhea (n = 28). Referring pediatricians attributed the symptoms to cow's milk allergy, and the same allergy tests were performed as on group I. The diagnosis was confirmed when the symptoms disappeared after the infants were fed the milk-free diet (Alfare, Nestle). Although none of the parents spontaneously complained about their infant's sleep, they were asked to fill in a sleep questionnaire and to keep a sleep log for 4 weeks after exclusion of milk from the diet. According to this information the group was further subdivided into two subgroups. Eighteen infants with normal sleep behavior formed the "good sleepers" subgroup; 13 infants with frequent awakenings and short sleep time formed the "poor sleepers" subgroup. Group III: Control infants. Twenty normal infants without previous history of sleep disturbance or allergy-related symptoms were selected from children hospitalized for minor interventions or examinations. With the parents' consent, blood was drawn from

118 A. KAHN ET AL. the infants to test for allergy and parents were asked to fill in a sleep questionnaire. No further investigation or regimen was instituted for these infants. Allergy Tests Assay kits for IgE (Prist Phadebas) were used to determine serum IgE levels. In vitro radioallergosorbent tests (RAST, Pharmacia Fine Chemicals) were conducted to identify specific IgE against (3-lactoglobulin. Antibodies against (3-lactoglobulin were measured. Statistical Analysis Statistical analysis was performed using analysis of variance (ANOVA), chi-square, and Wilcoxon signed-rank tests, with a level of significance of 0.05. RESULTS The main characteristics of the 71 infants are shown in Table 1. The groups were comparable for age and sex of the infants. In the 20 infants referred for chronic insomnia, poor sleep had been noticed since birth. Six of the infants had been sleepless while being breastfed. Three infants had been treated with sedative medications, phenothiazine syrups, without improvement in their sleep behavior. Two were being fed with soya milk because of previous eczema that had cleared before their referral. In group II, no family or personal characteristics, including the duration and apparent severity of their skin and digestive symptoms, differentiated the two subgroups of good and poor sleepers. The reported sleep duration significantly differentiated the three groups. The 20 infants in group I and the 18 poor sleepers in group II slept significantly less than the remaining 33 infants. No difference in total sleep time was seen between the infants in group I and the poor sleepers in group II or between the good sleepers in group II and the infants in group III. All infants in group I and the 18 poor sleepers in group II had 2-9 arousals per night..the good sleepers in group II and the infants in group III had no arousal during the night or awakened only occasionally. The tests for milk protein allergy were positive for the 38 infants in groups I and II but for none of the infants in group III (Table 2). Milk allergy was confirmed clinically by the favorable evolution during the exclusion diet for the infants in groups I and II and by the sleeplessness that followed the oral milk challenge in the infants in group I. Within 4 weeks of the milk-free diet, the sleep characteristics of all infants in group I and of the poor sleepers in group II became normal. Their sleep duration during the TABLE 1. Characteristics of the infants studied Sleep duration (h) Main complaint Mean age Sex Group n on admission (weeks) (M/F) Night Day I 20 Insomnia 25.5 ± 10.1 8/12 5.1 ± 1.0 1.0 ± 0.5 II 31 Milk allergy 31.5 ± 20.0 15/16 18 Poor sleepers 5.9 ± 2.0 1.5 ± 1.0 13 Good sleepers 9.5 ± 1.5 3.0 ± 1.5 III 20 Controls 30.5 ± 16.5 12/8 9.9 ± 1.5 4.5 ± 2.4 F test NS NS om 0.01 NS, not significant. Seventy-one infants were studied. Data are presented as absolute values, means, and SD.

INSOMNIA AND ALLERGY IN INFANTS 119 TABLE 2. Results of laboratory tests for allergy [3-Lactoglobulin High High antibody IgE eosinophil titers titers counts Group n (>100 V) (>10 Vlml) (>400) I 20 20/20 2120 3/20 II 18 18/18 2118 1118 13 13/13 3/13 1113 III 20 0/20 0120 1120 Seventy-one infants were studied. Numbers represent number of infants with results above normal values. Normal values are shown in parentheses for each test. day, as during the night, could not be differentiated from that of the good sleepers in group II and of the control infants in group III. DISCUSSION The infants with chronic sleeplessness had all the characteristics of the insomniac patients suffering from milk allergy reported previously (7), and their sleep was severely disturbed, despite the absence of any usual cause of repeated awakenings described in infants (1-6). Allergy to cow's milk was suspected from history and laboratory tests (8) and it was confirmed by the improvement in symptoms after milk was avoided, the recurrence of symptoms after an oral challenge with milk, and a new improvement of sleep after a second trial of milk elimination (8). The prevalence of this cause of sleep disturbance in children is not known. Although the prevalence of milk allergy is as high as 3% (8), only a fraction of the pediatric population referred for poor sleep could suffer from it. Because most insomniac infants will respond successfully to changes in family life and sleep routine (2), only the most persistent and severe cases should be considered potential candidates for the diagnosis of milk allergy. When cow's milk is eliminated from the diet of the infants with poor sleep, their sleep time becomes similar to that of normal control children of the same age. Still, three caveats must be mentioned about possible limitations of the dietary treatment of TABLE 3. Effects of diet on sleep duration Before After 3 weeks Group: Characteristic diet of diet p I: Insomnia (n = 20) Duration of daytime sleep 1 ± 0.5 3.0 ± 1.5 0.01 Duration of nighttime sleep 5.1 ± 1.2 9.9 ± 2.0 0.01 II: Milk allergy, poor sleepers (n = 18) Duration of daytime sleep 1.5 ± 1.0 4.5 ± 2.4 0.01 Duration of nighttime sleep 5.9 ± 2.0 9.8 ± 1.0 0.01 II: Milk allergy, good sleepers (n = 13) Duration of daytime sleep 3.0 ± 1.5 3.9 ± 1.6 NS Duration of nighttime sleep 9.5 ± 1.5 9.4 ± 1.2 NS NS, not significant. Influence of the milk-free diet on sleep duration, during the day (from 0701 to 1900 h), and during the night (from 1901 to 0700 h). Numbers represent means and SD. Statistical analysis was made with Wilcoxon's signed-rank test.

120 A. KAHN ET AL. sleep. First, the exclusion of all cow's milk protein can be a tedious procedure, as many food preparations contain f3-lactoglobulin. The accidental reintroduction of only small quantities of cow's milk protein into the infant's diet can cause the return of poor sleep. This can occur despite breastfeeding, as f3-lactoglobulin can enter the mother's milk if she drinks cow's milk (8). Second, soya milk may not be the best choice for replacement of cow's milk because up to 5% of allergic infants can also suffer from soya protein intolerance (8). In such case, preference should be given to a hypoallergenic hydrolized infant diet. Third, the diet should be maintained for at least 3-4 weeks until sleep is normalized. This implies that the families must be closely supported during the long treatment period. Considering the favorable evolution of the patient's sleep once the diet is correctly modified, the administration of sedative drugs to these infants should be avoided. Such medications do not improve the sleep quality of these infants (7) and can be dangerous (9). We cannot yet explain the relation between milk allergy and sleeplessness. The poor sleep could be related to physical discomfort, as well as to changes in metabolism of the central nervous system, for instance, through the local release of histamine. Neither do we know what the natural evolution of the sleep characteristics of these infants would have been had the diet not been modified. The clinical manifestations of cow's milk intolerance usually improve after the third year of life, perhaps through a change in intestinal permeability to ingested proteins (8). Although the sleep disturbances could thus diminish with age, one cannot exclude the possibility that some infants not treated will eventually develop chronic insomnia. In the group of 31 infants referred by their pediatricians because of digestive, skin, or respiratory symptoms attributed to milk allergy, 18 slept as poorly as the patients referred for chronic insomnia. This information was revealed only when the parents were specifically questioned about sleep behavior of their child. The poor sleep of the allergic children was overlooked by their pediatricians, who were probably more concerned about the classical symptoms of food allergy. Both these and the sleep disturbances were corrected by exclusion of milk from.the diet. We are unable to explain why 13 of the infants referred for atopy slept normally whereas 18 were poor sleepers. No significant difference was seen between the two groups of infants, either in their individual histories, clinically, or in laboratory tests. An immediate type of allergic reaction does not seem to account for the development of poor sleep, as indicated by the absence of elevated immunoglobulin E levels in the blood of most infants. Further studies will be needed to clarify this question. We conclude that infants with clinically evident milk allergy may suffer from sleeplessness, although this sleep disturbance can be overlooked. Furthermore, when no evident cause for a chronic insomnia can be found in an infant, the possibility of milk allergy should be given serious consideration. Acknowledgment: We thank Professor H. L. Vis for encouragement and Dr. I. Ingenbleek, Ph.D: (Nestle) for constant help. This work was supported by the Fondation Nationale de la Recherche Scientifique grant No. 3,4543,83. REFERENCES 1. Bax Meo. Sleep disturbance in the young child. Br Med J 1980;5:1177-9. 2. Ferber R, eds. Solve your child's sleep problems. New York: Simon and Schuster, 1985.

INSOMNIA AND ALLERGY IN INFANTS 121 3. Moore T, Ucko LE. Night waking in early infancy: part I. Arch Dis Child 1957;32:333-42. 4. Guilleminault C, Anders TF. Sleep disorders in children. Adv Pediatr 1976;22:151-75. 5. Association of Sleep Disorders Centers and the Association for the Psychophysiological Study of Sleep. Diagnostic classification of sleep and arousal disorders; Roffwarg HP (chairman). Sleep 1979;2:21-57. 6. Jones NB, Ferreira MCR, Brown MF. The association between perinatal factors and later night waking. Dev Med Child NeuroI1978;20:427-34. 7. Kahn A, Mozin MJ, Casimir G, Montauk L, Blum D. Insomnia and cow's milk allergy in infants. Pediatrics 1985;76:880-4. 8. Bahna SL, eds. Allergies to milk. Orlando, FL: Grune & Stratton, 1980. 9. Kahn S, Blum D. Phenothiazines and sudden infant death syndrome. Pediatrics 1982;70:75-8. Sleep, Vol. /0, No.2, 1987