Cow's milk allergy (CMA) is characterized by varied clinical manifestations following cow's milk ingestion. Clinically, patients with CMA usually have symptoms such as vomiting, chronic diarrhea, failure to thrive, anemia, recurrent bronchospasm, and eczema. The underlying mechanism is an abnormal immunologically mediated reaction (1,2). Characterization of such mechanisms may be helpful in further understanding its pathogenesis. Not many years ago, the eosinophil was thought to be a cell that primarily modulated the proinflammatory actions of the mast cell. However, the eosinophil may act as a cytotoxic cell and produce large quantities of lipid mediators, such as LTC4 and PAF. This possibility has changed the view of the potential pathophysiologic role this cell plays in various inflammatory conditions (3). In the early 1980s, doctors recognized that different biologically active mediators are stored in the characteristic granules of eosinophil cells, which play an important role in triggering the inflammatory process. The first described mediators are major basic protein (MBP), eosinophil cationic protein (ECP), eosinophil derived neurotoxin (EDN; known also as eosinophil protein X, EPX) and eosinophil peroxidase (EPO) (4). After stimulation, these proteins are released from the granules of eosinophil cells causing tissue damage. The ECP causes desquamation in the respiratory epithelium (5) or in the gut mucosa (6). Consequently, the submucosal layer will be unprotected against influence from outside allergens, which may lead to further sensitization and the increase of eosinophilic desquamative inflammation. Cilia or brush border will be destroyed. However, the ECP makes a change in the lipid structure of the membrane or produces inadequate transmembrane pores because it inhibits the adenosine triphosphatase (ATPase) activity of the epithelial cell.
The serum ECP level is increased in several allergy-like inflammatory diseases. Previous studies on eosinophil activation are mostly concerned with asthma and allergic rhinitis (7,8). There are only limited data on sECP levels in cow's milk allergy (9,10). Therefore, we thought it would be worthwhile to study its serum level in children. This is why we examined the sECP level of milk-sensitive patients before and during the oral cow's milk challenge test. We also wanted to know whether measurement of serum ECP during oral challenge may serve as a diagnostic tool for cow's milk allergy.
PATIENTS AND METHODS
The study included 35 children with challenge-proven cow's milk allergy. The mean age was 16 months (range, 6–49 months). The detailed clinical data of patients can be seen on Table 1. Children who were considered to have cow's milk allergy had clinical symptoms that ceased when they were no longer given cow's milk. Symptoms reappeared on open challenge. Urticaria, atopic dermatitis, repeated wheezing episodes, failure to thrive, vomiting, diarrhea, and bloody stools were symptoms after cow's milk protein was introduced to the children's diet. The children with cow's milk allergy were on a cow's milk-free diet (soy-based formula or extensively hydrolyzed protein formula). The children were rechallenged at an average of 9 months after the diagnosis to assess whether they could tolerate cow's milk. At that time, all the patients were symptom free. Twenty healthy non-allergic children younger than 5 years, whose first-degree relatives did not have any atopic disease, were included as controls in the study.
The Ethical Committee of Semmelweis University (Budapest, Hungary) gave ethical permission. Parents received detailed information regarding the examination before they signed consent forms.
Before the cow's milk challenge, we determined the baseline level of ECP from a 3-mL blood sample that was drawn from each of the children. The challenge test was begun with 2 mL of cow's milk-based formula with low lactose concentration (Nutrilon low lactose; Nutricia, Zoetermeer, The Netherlands). Increasing doses (5, 10, 20, 50, 100 mL) of this formula were serially administered at 30-minute intervals. The second blood sample was taken 2 hours after the beginning of the challenge, when the patients had consumed 37 mL of formula. The children were thoroughly observed and clinical symptoms (skin, gastrointestinal, airway symptoms, etc.) were registered. The last blood sample was drawn 24 hours after the beginning of the challenge. We determined the ECP level from the serum samples with fluoroimmunoassay (Pharmacia CAP System ECP FEIA; Amersham Pharmacia Biotech Kft, Budapest, Hungary). The levels ranged from 2 μg/L to 200 μg/L. The accuracy of the method was determined by intra-assay testing: from one sample of 10 different patients we made 2 measurements and the variation coefficient was always less than 10% (0, 1–8, 8%). We also determined the absolute eosinophil count from all blood samples.
Statistical analysis was performed by the SAS statistical software package (SAS/STAT Software Release 6.12; SAS Institute, Cary, NC, U.S.A.). Data are presented as median and 95% confidence intervals. Between patients with positive and negative challenge test a two-way (group x time) analysis of variance for repeated measures was performed in which baseline values were included as covariates (ANCOVA). The median nonparametric test was used to compare the values of the whole patient group and controls. A P value <0.05 was considered statistically significant.
After the cow's milk rechallenge test, 10 of 35 children with previously confirmed cow's milk protein allergy had clinical signs, such as skin rash, eczema, diarrhea, poor weight gain, wheezing (positive group). In the remaining 25 children, no reaction was observed at the rechallenge test (negative group). The level of serum immunoglobulin (Ig) E in the positive group (16.4 kU/L) was significantly higher (P = 0.02) than that of negative group (6.3 kU/L).
All individual measurements in the controls and the patients with either positive or negative challenge test results are presented in Figure 1. The median sECP levels were higher in children with positive test results at all time points, but this difference was not statistically significant by ANCOVA analysis. Two hours after the challenge, a slight decrease in the median values of sECP levels was observed in both the positive and negative groups. However, these values were again not significantly different from the baseline values (0 hour). Twenty-four hours after the beginning of the challenge test, the sECP concentrations were similar to the starting values in both positive and negative groups.
If the border between positive and negative sECP values was drawn at 12.5 μg/L, the positive predictive value of this test was 50% and its negative predictive value 85%.
The sECP values of all children with previously confirmed cow's milk allergy were compared with those of the controls. The median value (12.4 μg/L) of the sECP levels (range, 9.8–14.9 μg/L; 95% confidence intervals) of all patients before the cow's milk challenge was significantly higher (P < 0.05) than that of the controls (median, 4.3 μg/L; range 2.6–6.0 μg/L). Two hours after the challenge, the median value (9.4 μg/L) of the sECP levels (range, 7.6–11.3 μg/L) of all patients decreased (P < 0.05) compared with the starting value and was not significantly different from that of the controls. After 24 hours, the median sECP value (11.1 μg/L) of all patients (range, 8.6–13.6 μg/L) again was significantly higher (P < 0.05) compared with that of controls. The absolute eosinophil cell counts varied similarly with the sECP levels. The baseline eosinophil count (0.31 × 106/L) was slightly but not significantly decreased 2 hours after the beginning of challenge (0.28 × 106/L). Twenty-four hours after the challenge, the eosinophil count significantly increased (0.43 × 106/L, P = 0.02) compared with the starting value.
In the present study, eosinophil activation was observed in children with cow's milk allergy. Previous studies have shown that ECP is involved in the pathogenesis of asthma, inflammatory bowel disease, IgE-mediated skin disorders and in several other diseases, such as bronchiolitis, wheezy bronchitis, cystic fibrosis and bronchopulmonary dysplasia (11–15). In previous studies, the sECP level was measured in cow's milk allergic children before the cow's milk challenge test and at 8 hours and 24 hours after the challenge (9,10). Only those children with skin symptoms showed significant elevation in sECP levels, whereas those patients with gastrointestinal symptoms had stable sECP levels. Bengtsson et al. (16) found an increase of ECP in the small intestinal content of patients with milk-related gastrointestinal symptoms diagnosed by double-blind placebo-controlled milk challenge. In other studies (17–19), the fecal ECP level of children with food allergy was elevated after the provocation test without any increase in the serum ECP concentration.
In our present study, the control group had sECP levels that corresponded to the published reference values of nonatopic healthy children (20,21).
In our present study, it is interesting that the sECP level was higher in children with cow's milk allergy than in controls, even when both positive and negative groups were on a milk-free diet before the provocation test. We also observed that those children who had a positive provocation test had a slightly more elevated sECP level when on milk-free diet than those with a negative challenge test. However, there was significant overlap between the values of these two groups, giving a poor positive and negative predictive value. Because all children had good compliance with a strict cow's milk-free diet and were symptom free, this elevation cannot be explained by the allergen exposition. It is more probable that the prolonged persistence of eosinophil activation in allergic children cause the elevation of sECP levels even when the children are on the strict diet. Another surprising observation was that the sECP level was significantly decreased 2 hours after the beginning of milk challenge test compared with baseline level. Because we did not find such an early determination of sECP level after the beginning of challenge test in the literature, we cannot compare our results with other studies. We believe that the eosinophils are shunted to the bowel, where the allergen is located, after the milk challenge test, and the deliberated ECP is excreted mostly to the bowel and not to the blood stream. This hypothesis is supported also by Bengtsson et al. (16) and Kosa et al. (17). This possibility is also supported by our observation that, 2 hours after the cow's milk challenge, the absolute number of eosinophils decreased in the blood. Other studies have shown increase in the sECP level 8 and 24 hours after the beginning of milk challenge test in children with skin symptoms (9,10), but we were not able to show this. One possible explanation for our different findings may be that the majority of our patients with a positive provocation test had gastrointestinal symptoms. In these patients, the activated eosinophils may persist longer in the small intestinal mucosa, causing released ECP from these cells to be metabolized or excreted by the stool before absorption in the blood stream.
The diagnosis of cow's milk allergy is based on the observation of clinical symptoms after withdrawal and challenge of milk (1). However, objective evidence of immunologic activation is also essential. Evidence may include the elevation of sECP level, but it is important to note that, in a considerable percentage of patients with cow's milk allergy, the basic sECP level is in the normal range compared with the control values. This is why the determination of sECP level is not sensitive enough for the diagnosis of food allergy. In those children with cow's milk allergy in whom the sECP level is elevated at the onset of disease, the later decrease of the sECP level may indicate the cessation of this disorder. Therefore, the measurement of sECP level may be helpful in determining the optimal time for the repeat challenge test, when the result will more likely be negative.
This study was supported by the National Scientific Research Fund (OTKA-T31791) and Scientific Council of Health (ETT-297/2000) of Hungary.
1. Savilahti E, Heyman M, MacDonald TT. Diagnostic criteria for food allergy with predominantly intestinal symptoms. J Pediatr Gastroenterol Nutr 1992; 14:108–12.
2. Hill DJ, Ford RPK, Shelton MJ, et al. A study of 100 infants and young children with cow's milk allergy. Clin Rev Allergy 1984; 2:125–29.
3. Wardlaw AJ. Eosinophils in the 1990s: new perspectives on their role in health and disease. Postgrad Med J 1994; 70:536–52.
4. Boyce JA. The pathobiology of eosinophilic inflammation [review]. Allergy Asthma Proc 1997; 18:293–300.
5. Gleich GJ, Flavahan NA, Fujisawa T, et al. The eosinophil as a mediator of damage to respiratory epithelium: a model for bronchial hypersensitivity. J Allergy Clin Immunol 1988; 81:776–81.
6. Levy AM, Gleich GJ, Sandborn WJ, et al. Increased eosinophil granule proteins in gut lavage fluid from patients with inflammatory bowel disease. Mayo Clinic Proceedings 1997; 72:117–23.
7. Bousquet J, Corrigan CJ, Venge P. Peripheral blood markers: evaluation of inflammation in asthma. Eur Respir J 1998; 11(Suppl):42S–48S
8. Meyer P, Persson CG, Andersson M, et al. α2
-macroglobulin and eosinophil cationic protein in the allergic airway mucosa in seasonal allergic rhinitis. Eur Resp J 1999; 13:633–7.
9. Niggemann B, Beyer K, Wahn U. The role of eosinophils and eosinophil cationic protein in monitoring oral challenge tests in children with food-sensitive atopic dermatitis. J Allergy Clin Immunol 1994; 4:963–71.
10. Suomalainen H, Soppi E, Isolauri E. Evidence for eosinophil activation in cow's milk allergy. Pediatr Allergy Immunol 1994; 5:27–31.
11. Gruber W, Eber E, Pfleger A, et al. Serum eosinophil cationic protein and bronchial responsiveness in pediatric and adolescent asthma patients. Chest 1999; 116:301–5.
12. Zweiman B, Atkins P, Allmen von C, et al. Release of eosinophil granule proteins during IgE-mediated allergic skin reactions. J Allergy Clin Immunol 1991; 87:984–92.
13. Talley NJ, Kephart GM, McGovern TW, et al. Deposition of eosinophil granule major basic protein in eosinophil gastroenteritis and celiac diseases. Gastroenterology 1992; 103:137–45.
14. Luck W, Becker M, Niggemann B, et al. In vitro release of eosinophil cationic protein from peripheral eosinophils reflects disease activity in childhood Crohn disease and ulcerative colitis. Eur J Pediatr 1997; 156:921–4.
15. Koller DY, Nilsson M, Enander I, et al. Serum eosinophil cationic protein, eosinophil protein X and eosinophil peroxidase in relation to pulmonary function in cystic fibrosis. Clin Exp Allergy 1998; 28:241–8.
16. Bengtsson U, Knutson TW, Knutson L, et al. Eosinophil cationic protein and histamine after intestinal challenge in patients with cow's milk intolerance. J Allergy Clin Immunol 1997; 100:216–21.
17. Kosa L, Kereki E, Börzsönyi L. Copro-eosinophil cationic protein (ECP) in food allergy. Allergy 1996; 51:963–5.
18. Majamaa H, Laine S, Miettinen A. Eosinophil protein X and eosinophil cationic protein as indicator of intestinal inflammation in infants with atopic eczema and food allergy. Clin Exp Allergy 1999; 29:1502–6.
19. Kapel N, Matarazzo P, Haouchine D, et al. Fecal tumor necrosis factor alpha, eosinophil cationic protein and IgE levels in infants with cow's milk allergy and gastrointestinal manifestations. Clin Chem Lab Med 1999; 37:29–32.
20. Fitch PS, Brown V, Schock BC, et al. Serum eosinophil cationic protein (ECP): reference values in helathy nonatopic children. Allergy 1999; 54:1199–203.
21. Lodrup Carlsen KC, Halvorsen R, Carlsen KH. Serum inflammatory markers and effects of age and tobacco smoke exposure in young non-asthmatic children. Acta Paediatr 1998; 87:559–64.