In recent years the proliferation of new instrumental techniques such as ultrasonography, pediatric endoscopy, and 24-hour computed pH-metry has produced a remarkable increase in interest and knowledge in upper gastrointestinal tract diseases in the pediatric age group (1).
Cow's milk allergy (CMA) and gastroesophageal reflux (GER) are among the most common gastroenterologic disturbances in the first year of life. These two entities have many common features (i.e., clinical symptoms, epidemiology, and natural history) that appear to be closely associated, sometimes overlapping, and even interdependent (2). In clinical practice, the distinction between GER and CMA is important, because it determines a different therapeutic approach. For this reason, as pediatric gastroenterologists, we have been particularly attracted by the relationship existing between GER and CMA.
Our research (3–5) originated from the observation that in infants less than 1 year of age, failures of conventional GER therapy were sometimes corrected by avoiding cow's milk. This suggested that GER could be a CMA manifestation.
The results of our investigations allowed us to identify common features between GER and CMA, determine the frequency of the GER–CMA association during the first year of life, verify the possible dependence of GER on CMA, and identify the clinical, laboratory, and instrumental differences between GER and CMA-induced GER.
CHARACTERISTICS OF GASTROESOPHAGEAL REFLUX AND COW'S MILK ALLERGY
Cow's milk allergy is a condition that affects 0.3% to 7.5% of infants (6,7). It is often a familial disorder (8). Several immunologic mechanisms are involved in the pathogenesis of CMA. Although it was once widely believed that immunoglobulin (Ig)E was always involved in the pathogenesis of CMA (the condition was otherwise known as cow's milk intolerance), today we know that a large majority of cases of CMA in the first year of life are caused by immunologic mechanisms other than IgE-mediated reaction (9–12).
On the basis of clinical features and underlying pathogenetic mechanisms, Hill et al. (10) have identified three groups of infants with CMA. The first group is composed of immediate reactors who manifest allergy a few minutes after the ingestion of cow's milk, mainly as cutaneous eruptions or anaphylactic shock. In this group, an IgE-mediated mechanism is demonstrable. The second group is composed of intermediate reactors who react to cow's milk within hours of ingestion of moderate quantities. In these cases the symptoms are mainly gastrointestinal (i.e., vomiting, regurgitation, and diarrhea) and IgE is not involved as a pathogenetic mechanism. The third group (late reactors) consists of patients who show gastrointestinal, respiratory, or cutaneous symptoms several hours or days after the ingestion of cow's milk. In these patients an IgE-mediated mechanism is seldom demonstrated, whereas an elevated T-cell reactivity is often revealed in vitro.
In our experience it seems that most patients affected with CMA in the first year of life belong to the second group, and this is probably the reason they show no positivity in laboratory tests intended to detect reaginic activity (paper radioimmunosorbent test [PRIST], radioallergosorbent test [RAST], and skin-prick test [SPT]). High titers of circulating IgG antibodies against β-lactoglobulin are often present in their sera, but the clinical significance of this phenomenon is still debated.
IgE-mediated reactions are more frequent in patients who do not recover in the first 2 years of life but in whom the disease persists until childhood. The natural history of CMA is somewhat controversial. Some researchers assert that symptoms tend to regress within 3 years (13–15), but it is also well known that there are forms that last longer and that patients may remain symptomatic beyond 4 years of age (16–18).
Cow's milk allergy affects the gastrointestinal tract, particularly the small intestine. Some investigators affirm that an enteropathy is always present in an infant with CMA (19), whereas others point out that not all patients show small bowel mucosal damage (20).
The clinical appearance (Table 1) is represented by typical symptoms such as vomiting, chronic diarrhea, failure to thrive, anemia, recurrent bronchospasm, and cutaneous eczema, as well as by unusual symptoms only recently associated with CMA such as infantile colic, chronic constipation, apnea, and sleeping disorders (21–25). Some of these symptoms can represent the initial symptoms of GER.
As mentioned before, laboratory tests are not useful in diagnosis of CMA. A correct diagnosis is based on the disappearance of symptoms when the patient is given a cow's milk–free diet and on their reappearance on cow's milk challenge. The gold standard for CMA diagnosis is still the double-blind, placebo-controlled food challenge, which must be performed in a hospital (26,27). This procedure represents the only way to achieve a real diagnosis of CMA and to avoid the risk of underestimating or overestimating the disease in the first year of life.
Gastroesophageal reflux is not an all-or-none phenomenon. It is a physiological event that affects almost all infants in the first year of life and has to be differentiated from a pathologic condition in accordance with the quantity of reflux (28). Physiological GER is clinically characterized by episodes of regurgitation and vomiting in an otherwise healthy and well-thriving infant, whereas pathologic GER, also called GER disease (GERD), is characterized by symptoms and complications such as esophagitis, with consequent hematemesis and/or iron deficiency anemia, failure to thrive, apnea, obstructive airway disease, aspiration pneumonia, and other effects (29,30).
The prevalence of GER in young infants, although difficult to ascertain, because most infants are not brought to medical attention (31), can be estimated between 1% and 10%(32). As for the natural history of GER, it must be emphasized that reflux disappears spontaneously within the first 18 months of life in 80% of cases, with or without conservative therapy. It can persist until 24 months in 15% of infants, and in 5% it does not resolve and transforms into GERD (33).
As mentioned before, the clinical features of GER (Table 2) vary widely, from common symptoms such as vomiting and regurgitation to unusual ones such as respiratory diseases and apnea (34). Vomiting and regurgitation can be considered the most common symptoms of GER in the first year of life, but such symptoms can be attributed to numerous other disorders in this age group: CMA, renal disorders, metabolic disease, brain tumors, and disorders of the central nervous system (35).
The diagnosis of GER and the distinction between GER and GERD is achieved by clinical evaluation and by complementary investigations such as 24-hour computed pH-metry and endoscopy. Twenty-four-hour pH-metry is the gold standard for the diagnosis of GER. In the evaluation of the tracings produced by pH-monitoring, several parameters and criteria can be used. Among these, the percentage of total reflux time is one of the most reliable (36). A complementary way to evaluate the results of this investigation is to observe the patterns of the pH-monitoring tracing. The pattern is important because it provides more detailed information on the type of GER (37). In our experience, it also helps to identify infants with CMA.
Endoscopy is mandatory for the diagnosis of esophagitis. It has been demonstrated that some cases of esophagitis misdiagnosed with 24-hour pH-metry were correctly diagnosed only after endoscopic examination (38). In histologic examination the typical lesion attributable to esophagitis is the infiltration of neutrophils and particularly of eosinophils (39,40). The real significance of the presence of eosinophils in the esophageal biopsy specimen is still debated.
The appropriate timing and means of performing pH-metry, endoscopy, and other investigations have been recently outlined by the ESPGAN Working Group on Gastroesophageal Reflux in a position paper concerning propositions for the diagnosis and treatment of GERD in children (34). In this paper the committee proposes that infants with uncomplicated GER can be safely treated before beginning investigations such as pH-metry and endoscopy. These investigations are otherwise mandatory if symptoms persist or if the child has an unusual manifestation of GER. The Working Group also suggests another important consideration concerning the possibility of secondary GER. It is in fact stressed that GER can be dependent on different conditions such as infections (i.e., urinary tract infections and gastroenteritis), CMA, metabolic disorders, and other diseases. In this case, GER represents a symptom more than a single disease entity.
Comparing the aforementioned characteristics of CMA and GER in the first year of life, it was noted that the two diseases share common characteristics (Table 3), not only epidemiologically (both affect a large percentage of infants in their first year of life and have a similar incidence), but also clinically (symptoms at onset are often overlapping) and, as far as natural history is concerned (in a large majority of cases both regress within the second year of life). Furthermore, GER can be secondary to CMA.
These considerations are particularly relevant in differential diagnosis. For example, when treating an infant with recurrent vomiting in whom GER or CMA is suspected, the first step is to determine whether only one of the two entities is responsible for the clinical picture, whether they coexist, or whether GER is dependent on CMA. In this last situation, only the elimination of cow's milk and its derivatives from the diet leads to resolution of the disease. Our research has progressed from the definition of the frequency of GER and CMA association in the first year of life.
Frequency of Association of Gastroesophageal Reflux and Cow's Milk Allergy During the First Year of Life
Before our investigation, the association between GER and CMA in the first year of life had been described in only one study in the literature (32), but no prospective studies existed to evaluate the frequency of this association. For this reason, we undertook a 3-year prospective study in a large number of patients, breast-fed or formula-fed, who were observed in our department for symptoms compatible with the diagnosis of GER. We performed a complete diagnostic work-up, including 24-hour pH-metry and upper gastrointestinal endoscopy. In this way, 204 patients were selected as having GER. The diagnosis was based on the finding of a total reflux time greater than 5.2% at pH-metry and of esophagitis observed during endoscopic examination.
At the moment of hospital admission the clinical history of these patients was carefully recorded, and laboratory examinations were performed to investigate concomitant CMA (Table 4). All patients with clinical signs suggesting associated CMA or with at least one positive test result began an exclusion diet with a formula containing hydrolyzed proteins. After at least 3 months, the patients underwent a double-blind cow's milk challenge to confirm the diagnosis of CMA. In addition, 24 hours before and 24 hours after the challenge, all these patients underwent intestinal biopsy at the ligament of Treitz. In accordance with the criteria outlined by ESPGAN, only patients who had clear clinical reactions after cow's milk challenge, accompanied by alterations in immunologic test results, which confirmed hypersensitivity to cow's milk protein, were considered to have CMA. In this way, 85 of 204 patients were diagnosed with CMA.
The analysis of our data revealed some important facts: First, there is a high, and perhaps unexpected, frequency of the GER–CMA association during the first year of life. As many as 41.8% of the patients with GER we observed consecutively were shown to have concomitant CMA. Although we are unable to make an exact calculation of CMA in the general population of our region, the data reported in the literature indicate a range between 3% and 5%. The very high prevalence of CMA in patients with GER cannot therefore be attributed to the frequency of this disease in the general population.
Second, the association between GER and CMA was sometimes apparent simply by recording clinical history carefully. In slightly more than 25% of the cases of infants affected with GER, there were concomitant symptoms of allergy. However, these symptoms were not overtly manifest, and the parents had consulted our department because of the presence of other symptoms, including some of those we considered as selection criteria for our study. Therefore, it is possible (but it is not applicable in 75% of cases) that a precise clinical history may be sufficient to diagnose CMA and GER.
Third, in cases with no clinical signs or symptoms of CMA, the suspicion of CMA arises on the basis of positive laboratory test results including, above all, IgG anti-β-lactoglobulin assay (Table 5) and skin prick test. An IgG anti-β-lactoglobulin assay was very sensitive although nonspecific in screening for suspected CMA. In one of our previous studies (41) we described that IgG anti-β-lactoglobulin serum values, obtained using an immunoenzymatic microplate method (Betalactotest; Eurospital, Trieste, Italy), were distributed differently in a group of patients with CMA (50th percentile: 101%, 90th percentile: 147%) when compared with healthy control subjects (50th percentile: 17%, 90th percentile: 48%) and patients affected with gastroenterologic disturbances different from CMA (50th percentile: 15%, 90th percentile: 56%).
Skin prick tests were performed by using full-cream cow's milk, lactalbumin, and β-lactoglobulin (Lofarma Diagnostic) and considering as positive any weal diameter that exceeded that of the control and was more than one fourth the size of the histamine weal.
Fourth, we noticed that, in regard to the pH-monitoring characteristics, none of the parameters commonly used in interpreting the results was useful in distinguishing between subjects with GER and subjects with GER and CMA. On the contrary, there was a significant difference in Jolley scores at the time of the diagnosis in the subjects of the two categories. The score was much higher in patients with GER and CMA than in those with GER only. Most probably, this is not a spurious result, but rather the expression of a specific characteristic that we observed in most pH-monitoring tracings in patients with GER and CMA. In fact, we noted that these infants almost always had normal esophageal pH during feeding and immediately after feeding; however, after this period, there was an evident, constant, and progressive decrease in esophageal pH, and a record of prolonged periods of pH less than 4 were recorded. Therefore, the Jolley score (42), excluding the two postprandial hours in its calculation, fully confirmed our impression concerning the characteristic of reflux in patients with CMA.
Are Gastroesophageal Reflux and Cow's Milk Allergy Associated, or Does Cow's Milk Induce Reflux?
After describing the strong association between GER and CMA, we were interested in checking the possible dependence of GER on CMA. For this reason, we selected 96 consecutive infants fed milk formulas containing cow's milk protein who had been referred to our clinic for observation and showed signs or symptoms compatible with a diagnosis of GER and/or CMA. All the patients underwent 24-hour esophageal pH monitoring, and those with gastroenterologic symptoms also underwent endoscopy of the upper gastrointestinal tract. All patients were then given an elimination diet without cow's milk and its derivatives, by using a milk formula based on hydrolyzed casein. Those patients with improved symptoms after receiving a cow's milk protein–free diet underwent a double-blind milk challenge 6 to 8 weeks later. An intestinal biopsy of the ligament of Treitz was performed both before and 24 hours after the challenge in these patients.
Patients were considered to have GER when they showed endoscopic evidence of esophagitis or showed a clear link between the observation of the clinical symptom (apnea or episode of bronchospasm, fits of crying) and an episode of gastroesophageal reflux recorded during the 24-hours pH monitoring. Conversely, patients were considered to have CMA if symptoms improved during the cow's milk protein–free diet and if a cow's milk protein challenge was positive. On the basis of this study design (Fig. 1), after endoscopy and cow's milk challenges, we were able to classify our patients into four groups: 33 with primary GER not associated with CMA, 14 with GER associated with CMA, 25 with CMA without GER, and 24 without either GER or CMA. The high frequency of GER associated with CMA was confirmed again; CMA was recorded in 14 (30%) of 47 patients with GER.
Furthermore, because the cow's milk protein–free diet alone determines a significant improvement in symptoms in patients with GER and CMA, we formulated the hypothesis that the association of these two diseases is not fortuitous, but rather causative, so that GER and esophagitis may be directly dependent on CMA.
Clinical, Laboratory, and Instrumental Differences Between GER and CMA-Induced GER
In both our studies, the clinical symptoms of patients with GER and CMA did not statistically differ from those observed in patients with GER. Only the immunologic data and analyses of the pH-metry tracing appeared to be more useful in making a correct diagnosis. It must be stressed, however, that patients with GER and CMA more commonly manifest the concomitant typical CMA symptoms, such as diarrhea, dermatitis, or rhinitis.
Of all the immunologic tests, concentrations of IgG anti-β-lactoglobulin proved to be most useful (in our latter study it was elevated in 13/14 cases of GER and CMA). The usefulness of determination of IgG anti-β-lactoglobulin serum levels in the diagnosis of CMA is still debated. Some investigators consider this test to be a diagnostic tool (43–45), whereas others find it not generally helpful (46–49). In our experience (41), the distribution of IgG anti-β-lactoglobulin serum levels in CMA patients is significantly different if compared with healthy subjects or with subjects affected with other gastroenterologic disturbances. We therefore consider the determination of IgG anti-β-lactoglobulin to be a reliable test in screening patients with suspected CMA who are candidates for a cow's milk–free diet and, subsequently, a cow's milk challenge.
Conversely, it is not advisable to use the intestinal biopsy after cow's milk challenge as a routine test to support the hypersensitivity reaction. In fact, in our latter study, of 39 patients with a clear and severe clinical reaction to the cow's milk protein challenge, we recorded damage to the intestinal mucosa in only 31 (the sensitivity of intestinal biopsy in the diagnosis of CMA in our study was therefore 79%).
Twenty-four-hour esophageal pH monitoring was also highly useful in distinguishing between forms of primary GER and GER secondary to CMA. In fact, all the patients who had esophageal pH gradually decreasing between one meal and the next, giving rise to the characteristic phasic pattern, proved to have CMA.
In our latter study, as shown in Table 6, this typical pattern was not observed in any of the study patients without CMA. Furthermore, the sensitivity of the analysis of the tracing in the diagnosis of GER and CMA was high: 12 of 14 patients with secondary GER showed the characteristic phasic pattern, which was also discovered in patients with CMA but without GER. In our study population, 24 of 25 patients with CMA, but without esophagitis, had a phasic pattern. In addition, once again, none of the classic pH-metric indices were useful in distinguishing between the cases of primary GER and those with GER secondary to CMA.
The profile of the phasic pattern is shown in Figure 2, where it is compared with a fraction of tracing with a completely random distribution of reflux periods.
In Figure 3 a 3-hour fraction of the same two tracings is analyzed for the number of recordings of each pH value during each 30-minute fraction between one meal and the next. It can be seen that after analyzing the phasic tracing and inserting the data into the double-entry analytical table, the most frequent value recorded in each 30-minute fraction is distributed along the main diagonal of the table. In the cases in which the tracing was not phasic, there was a completely random distribution of these parameters.
FINAL COMMENTS AND INDICATIONS FOR CLINICAL PRACTICE
Even though our work suggests a role for CMA in the phenomena underlying GER and an association in the first year of life, many aspects still remain unclear. In the past 2 years, our research in this field has led us to the conclusion that the relationship between GER and CMA is not always causative, and not all infants affected with GER and CMA have milk-induced reflux. Sometimes the two entities simply coexist. This is proved by the existence of a small population of infants in whom GER symptoms do not subside completely after a cow's milk–free diet. For these infants a pharmacologic anti-GER intervention is often required.
This feature probably reflects the great variety of mechanisms involved in the pathogenesis of GER that differently prevails in subjects with GER associated with CMA (in whom GER has to be considered as primary) compared with those with CMA-induced GER.
We believe that the physiopathology of antireflux mechanisms in subjects with primary GER differs from the one working in patients with GER secondary to CMA. In the first case, the transitory and inappropriate relaxation of the lower esophageal sphincter can be considered the main pathogenetic mechanism (50). As shown by the pH-metric phasic pattern, in patients with GER secondary to CMA, there seems to be a persistent functional incapacity of the lower esophageal sphincter (probably due to allergic inflammation). As a result, the intraesophageal electrode records the progressive postprandial decrease in gastric pH. Further investigations are necessary to clarify the pathogenesis of GER secondary to CMA, especially using instrumental techniques such as manometry, ultrasonographic determination of gastric emptying, and electrogastrography. Some efforts in these directions have recently been undertaken by other researchers (51,52). In some cases, it has been demonstrated that the ingestion of cow's milk in allergic patients causes a mucosal inflammatory response with consequent alterations in normal gastric myoelectrical activity, resulting in symptoms of foregut dysmotility.
For this reason, in cases of milk-induced reflux, the cow's milk–free diet resolves the functional incapacity of the lower esophageal sphincter. Nevertheless, if primary GER coexists with CMA and if there are transitory and inappropriate relaxations of the lower esophageal sphincter, recovery is achieved only after pharmacologic therapy.
Our experience is limited to the resolution of symptoms for cases of milk-induced reflux. Therefore, to fully confirm the relationship, it is advisable to perform a second instrumental work-up, including pH-metry and endoscopy, during the cow's milk–free diet, to demonstrate the normalization of the pH-metry tracing and the disappearance of infiltration.
Indeed, there is still considerable debate on the histologic findings in esophageal biopsies and especially on the role of eosinophils. It is well known that eosinophils in the esophageal mucosa represent a marker for GERD-induced peptic esophagitis (39,40). Kelly et al. (53) have recently reported 10 pediatric patients with upper gastrointestinal symptoms attributed to GERD and esophageal histology characterized by eosinophilic infiltration whose symptoms were not improved after conventional antireflux medical therapies but in whom almost all symptoms resolved after 2 to 6 weeks of cow's milk–free diet. The positive response to the cow's milk challenge, confirming the diagnosis of CMA, seems to attribute the presence of eosinophilic infiltrate to allergic inflammation, rather than to peptic esophagitis. Some investigators have suggested differentiating allergic from peptic esophagitis on the basis of the presence of mast cells (54), whereas others have cured eosinophilic esophagitis with corticosteroids (55).
Although further studies are required to fully explain these phenomena to confirm the high frequency of association of GER and CMA found in our population of infants less than 1 year of age, we believe that many useful indications for clinical practice may emerge from our studies. The approach to the infant with suspected GER and CMA has to be different, with regard for various clinical situations. Our suggestions may be summarized as follows:
1. According to the recommendations of the ESPGAN Working Group on Gastroesophageal Reflux, well-thriving infants who show repetitive regurgitation and none of the complications related to GER or other signs of CMA can be safely treated with milk-thickening agents, antacids, and even prokinetics without performing any complementary investigation.
2. During the first 12 months after birth, when facing an infant with signs of complicated GER, unusual symptoms, or failure of GER therapy, it is advisable to perform a complete work-up for diagnosing both GER and CMA, given the very high prevalence of GER associated with CMA. The suggested work-up should include recording an accurate clinical history, laboratory examinations, pH-metry, and endoscopy.
3. The recording of an accurate clinical history is very important because, although the symptoms of GER associated with CMA are the same as those observed in primary GER, up to 25% of patients have mild typical CMA-associated manifestations (diarrhea, dermatitis or rhinitis).
4. Among laboratory tests, the determination of IgG anti-β-lactoglobulin serum levels is useful in supporting the diagnosis of CMA.
5. The evidence of a typical phasic pattern is a sensitive and specific instrumental index for diagnosing CMA and identifying an association between GER and CMA.
6. Intestinal permeability studies, proposed (38) as reliable markers of intestinal damage and thus considered useful in diagnosing CMA, are not easily performed in all hospitals. Furthermore, they are somewhat aspecific for CMA diagnosis and of limited sensitivity in those cases without enteropathy. In this regard, we question the need for routine use of intestinal biopsy after cow's milk protein challenge to support the hypersensitivity reaction.
7. If grade 3 esophagitis is present and a CMA-induced GER has been demonstrated, the cow's milk–free diet cannot be used as the exclusive therapeutic approach, and a conventional pharmacologic treatment must be associated with the diet.
In conclusion, our studies appear to imply that in just under half of the cases of GER in infants less than 1 year of age there is an association with CMA. The frequency of this association should prompt physicians to screen for possible concomitant CMA in all infants with GER in this age group. Further studies are necessary to clarify the pathogenetic mechanisms underlying these two pathologic conditions and determining their association.
1. Thomson M. Disorders of the oesophagus and stomach in infants. Baillieres Clin Gastroenterol 1997; 11:547–71.
2. Forget P, Arends JW. Cow's milk protein allergy and gastroesophageal reflux. Eur J Pediatr 1993; 152:704–11.
3. Iacono G, Carroccio A, Cavataio F, et al. Gastroesophageal reflux and cow's milk allergy in infants: A prospective study. J Allergy Clin Immunol 1996; 97:822–7.
4. Cavataio F, Iacono G, Montalto G, Soresi M, Tumminello M, Carroccio A. Clinical and pH-metric characteristics of gastroesophageal reflux secondary to cow's milk protein allergy. Arch Dis Child 1996; 75:51–6.
5. Cavataio F, Iacono G, Montalto G, et al. Gastroesophageal reflux associated with cow's milk allergy in infants: Which diagnostic examinations are useful? Am J Gastroenterol 1996; 91:1215–20.
6. Collins–Williams C. Cow's milk allergy in infants and children. Int Arch Allergy 1962; 20:38–41.
7. Gerrard JW, MacKenzie J, Goluboff N, Garson JZ, Maningas CS. Cow's milk allergy: Prevalence and manifestations in an unselected series of newborns. Acta Pediatr Scand 1973; 234:(Suppl)1–21.
8. Deamer WC, Gerrard JW, Speer F. Cow's milk allergy: A critical review. J Fam Pract 1979; 9:223–32.
9. Bahna SL, Gandi F. Milk hypersensitivity: I. Pathogenesis and symptomatology. Ann Allergy 1983; 50:218–23.
10. Hill DJ, Firer MA, Shelton MJ, Hosking CS. Manifestations of milk allergy in infancy: Clinical and immunological findings. J Pediatr 1986; 109:270–6.
11. Høst A. Cow's milk protein allergy and intolerance in infancy. Some clinical, epidemiological and immunological aspects. Pediatr Allergy Immunol 1994; 5(suppl 5):1–36.
12. Høst A, Jacobsen HP, Halken S, Holmenlund D. The natural history of cow's milk protein allergy/intolerance. Eur J Clin Nutr 1995; 49(suppl 1):S13–8.
13. Kuitunen P, Visakorpi JK, Savilhati E, Pelkonen P. Malabsorption syndrome with cow's milk intolerance. Arch Dis Child 1975; 50:351–6.
14. Kjellman NI, Bjorkstein B, Hattevig G, Falth–Magnusson K. Natural history of food allergy. Ann Allergy 1988; 61:83–7.
15. Bock SA. Prospective appraisal of complaints of adverse reactions in children during the first 3 years of life. Pediatrics 1987; 79:683–7.
16. Schrander JJP, Oudsen S, Forget PP, Kuijten RH. Follow-up study of cow's milk protein intolerant infants. Eur J Pediatr 1992; 151:783–5.
17. Bishop JM, Hill DJ, Hosking CS. Natural history of cow milk allergy: Clinical outcome. J Pediatr 1990; 116:862–7.
18. Iacono G, Cavataio F, Montalto G, Soresi M, Notarbartolo A, Carroccio A. Persistent cow's milk protein intolerance in infants: The changing faces of the same disease. Clin Exp Allergy 1998; 28:817–23.
19. Jalonen T. Identical intestinal permeability changes in children with different clinical manifestations of cow's milk allergy. J Allergy Clin Immunol 1991; 88:737–42.
20. Hill DJ, Davidson GP, Cameron DJ, Barnes GL. The spectrum of cow's milk allergy in childhood: Clinical, gastroenterological and immunological studies. Acta Paediatr Scand 1979; 68:847–52.
21. Walker–Smith JA. Diagnostic criteria for gastrointestinal food allergy in childhood. Clin Exp Allergy 1995; 25(suppl 1):20–3.
22. 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.
23. Iacono G, Cavataio F, Montalto G, et al. Intolerance of cow's milk and chronic constipation in children. N Engl J Med 1998; 339:1100–4.
24. Camfield P, Camfield C, Bagnell P, Rees E. Infant apnea syndrome: A prospective evaluation of etiologies. Clin Pediatr (Phila) 1982; 21:684–7.
25. Kahn A, François G, Sottiaux M, et al. Sleep characteristics in milk-intolerant infants. Sleep 1988; 11:291–7.
26. Goldman AS, Anderson DW Jr Sellers WA, Saperstein S, Kniker WT, Halpern SR. Milk allergy: I. Oral challenge with milk and isolated milk proteins in allergic children. Pediatrics 1963; 32:425–43.
27. Bock SA, Sampson HA, Atkins FM, et al. Double-blind, placebo-controlled food challenge (DBPCFC) as an office procedure: A manual. J Allergy Clin Immunol 1988; 82:986–97.
28. Boyle JT. Gastroesophageal reflux in the pediatric patient. Gastroenterol Clin North Am 1989; 18:315–38.
29. Orenstein S. Gastroesophageal reflux. Curr Probl Pediatr 1991; 21:193–241.
30. Sondheimer JM. Gastroesophageal reflux: Update on pathogenesis and diagnosis. Pediatr Clin North Am 1988; 35:103–16.
31. Milla PJ. Reflux vomiting. Arch Dis Child 1990; 65:996–9.
32. Forget P, Arends JW. Cow's milk protein allergy and gastro-oesophageal reflux. Eur J Pediatr 1985; 144:298–300.
33. Cucchiara S, Emilano M, Calabrese F. Il. Reflusso gastroesofageo. Scient Nutr Tod 1995; 2:17–21.
34. Working Group on Gastro-oesophageal Reflux Disease of the European Society of Pediatric Gastroenterology and Nutrition. A proposition for the diagnosis and treatment of gastro-oesophageal reflux disease in children: A report from a working group on gastro-oesophageal reflux disease. Eur J Pediatr
35. Orenstein S. Dysphagia and vomiting. In: Wyllie, Hyams, eds. Pediatric gastrointestinal disease.
Philadelphia: W.B. Saunders, 1993:135–50.
36. Vandenplas Y, Goyvaerts H, Helven R, Sacre L. Gastroesophageal reflux, as measured by 24-hour pH-monitoring, in 509 healthy infants screened for risk of sudden infant death syndrome. Pediatrics 1991; 88:834–40.
37. Jolley SG, Dale GJ, Herbst JJ, Matlak ME. The significance of gastroesophageal reflux patterns in children. J Pediatr Surg 1981; 16:859–65.
38. Staiano A, Troncone R, Simeone D, et al. Differentiation of cow's milk intolerance and gastroesophageal reflux. Arch Dis Child 1995; 73:439–42.
39. Winter HS, Madera JL, Stafford RJ, Grand RJ, Qunilan JE, Goldman H. Intraepithelial eosinophils: A new diagnostic criterion for reflux esophagitis. Gastroenterology 1982; 83:818–23.
40. Hyams JS, Ricci A Jr Leichtner AM. Clinical and laboratory correlates of esophagitis in young children. J Pediatr Gastroenterol Nutr 1988; 7:881–4.
41. Iacono G, Carroccio A, Cavataio F, et al. IgG anti-betalactoglobulin (betalactotest): Its usefulness in the diagnosis of cow's milk allergy. Ital J Gastroenterol 1995; 27:355–60.
42. Jolley SG, Johnson DJ, Herbst JJ, Pena A, Garnier R. An assessment of gastro-esophageal reflux in children by extended pH-monitoring of the distal esophagus. Surgery 1978; 84:16–24.
43. Burgin–Wolff A, Signer E, Friess HM, Berger R, Birbaumer A, Just M. The diagnostic significance of antibodies to various cow's milk allergy (fluorescence immunosorbent test). Eur J Pediatr 1980; 133:17–24.
44. Fallström SP, Ahlstedt S, Carlsson B, Lönnerdal B, Hanson LA. Serum antibodies against native, processed and digested cow's milk proteins in children with cow's milk protein intolerance. Clin Allergy 1986; 16:417–23.
45. Taylor CJ, Hendrickse RG, McGaw J, MacFarlane SBJ. Detection of cow's milk protein intolerance by an enzyme-linked immunosorbent assay. Acta Paediatr Scand 1988; 77:49–54.
46. Udall JN. Serum antibodies to exogenous proteins: The significance? J Pediatr Gastroenterol Nutr 1989; 8:145–6.
47. Burks AW, Williams LW, Casteel HB, Fiedorek SC, Connaughton CA. Antibody response to milk proteins in patients with milk-protein intolerance documented by challenge. J Allergy Clin Immunol 1991; 85:197–202.
48. James JM, Sampson HA. Immunological changes associated with the development of tolerance in children with cow milk allergy. J Pediatr 1992; 121:371–7.
49. Keller KM, Burgin–Wolff A, Lippold R, With S, Lentze MJ. The diagnostic significance of IgG cow's milk protein antibodies re-evalued. Eur J Pediatr 1996; 155:331–7.
50. Kawahara H, Dent J, Davidson G. Mechanisms responsible for gastroesophageal reflux in children. Gastroenterology 1997; 13:399–408.
51. Knafelz D, Smith VV, St. Louis D, Kafrista P, Lindley KJ, Milla PJ. Electrogastrographic abnormalities in patients with allergen induced gastric mucosal inflammation (abstract). Ital J Gastroenterol Hepatol 1998; 30(suppl 1):A34.
52. Ravelli AM, Pillan MN, Ugazio AG. Effects of cow's milk protein allergy on gastric antral electrical activity and gastric emptying (abstract). Ital J Gastroenterol Hepatol 1998; 30(suppl 1):A33.
53. Kelly K, Lazenby AJ, Rowe P, Yardley JH, Perman JA, Sampson HA. Eosinophilic esophagitis attributed to gastroesophageal reflux: Improvement with an amino acid-based formula. Gastroenterology 1995; 109:1503–12.
54. Justinich CJ, Kalafus D, Esposito P, et al. Mucosal mast cells distinguish allergic from gastroesophageal reflux induced esophagitis. J Pediatr Gastroenterol Nutr 1996; 23:342.
55. Faubion WA Jr Perrault J, Burgart LJ, Zein NN, Clawson JL, Freese DK. Treatment of eosinophilic esophagitis with inhaled corticosteroids. J Pediatr Gastroenterol Nutr 1998; 27:90–3.