The CM challenge was positive in 118 infants at a mean (range) age of 6.7 (2.8-12.7) months giving a cumulative incidence of 1.9 % in the present cohort. In the whey hydrolysate group fewer infants developed CMA than in the CM group (Table 1). In the pasteurized human milk group there was a similar trend. The exclusively breast-fed infants and those exposed to CM were at similar risk.

Table 1. Cumulative Incidence of CMA and ORs for the Risk of CMA according to Feeding Group at Maternity Hospital

Infants with CMA

Table 1. Cumulative Incidence of CMA and ORs for the Risk of CMA according to Feeding Group at Maternity Hospital

Infants with CMA


N (%)

OR (95% CI)

CM formula (1758)

43 (2.4)


Pasteurized human milk (1844)



Whey hydrolysate formula (1715)


0.61 (0.38-1.00)

Own mother's milk (811)


0.85 (0.48-1.51)

During the first 8 weeks of life, 54% of the allergic infants and 55% of the tolerant infants in the randomised groups had been exposed to CM at home compared with only 6% and 20% in the comparison group, respectively. The percentage of allergic infants with a history of obvious parental atopy was almost double that oftolerant infants: 36% vs 19%. In the 4 groups of allergic infants the prevalence was similar: 33%, 38%, 35%, 47% in the CM, pasteurized human milk, whey hydrolysate and comparison groups, respectively. Among the tolerant infants respective figures were: 18%, 19%,20% and 19%. Among the infants given supplementary feedings at hospital, significant risk factors for CMA were exposure to CM at hospital (OR 1.54, 1.04-2.30) and apositive atopic heredity (OR 2.32, 1.53-3.52).

At challenge, 75 (64%) infants showed IgE-mediatedand 43 (36%) non-IgE-mediated response to CM. Of the IgE-positive infants, 68% reacted within 2 hours compared with 16% in the non-IgE-mediated group. The IgE-positive infants had more often urticaria (76% vs 9%), but less frequently atopic dermatitis (28% vs 72%),vomiting (9% vs 30%),diarrhoea (0% vs 23%) and wheezing (1% vs 14%) than those with non-IgE-mediated reaction. During the first 8 weeks at home, the infants with IgE-mediated reaction were exposed to CM less frequently and those who were exposed were given smaller amounts of CM during a shorter period (Table 2). They were breast-fed longer, and greater percentage of them had symptoms suggestive of CMA during exclusive breast-feeding. Of the 50 infants showing symptoms during exclusive breast-feeding, 18 were given CM at hospital. Thus 32 infants were sensitised during exclusive breast-feeding.

Table 2. Infant Formula and Breast-feeding Patterns in the IgE-Positive and IgE-Negative Groups

Table 2. Infant Formula and Breast-feeding Patterns in the IgE-Positive and IgE-Negative Groups

Feeding Pattern



CM formula exposure at maternity hospital



CM formula exposure at home <8 weeks of age



-duration of exposure (wk)

1.3 (0.8-2.0)


- daily dose / infant (mL)

40 (30-70)


- total volume/infant (mL)

400 (210-770)


Symptoms during exclusive breastfeeding



Total breast-feeding (mo)

8.4 (7.8-9.1)

4.5 (3.5-5.7)**

Significant risk factors for the development oflgE-mediated CMA were: exposure to CM at hospital (OR 3.5, 1.2-10.1),breast-feeding during the first 8 weeks at home either exclusively (OR 5.1, 1.6-16.4) or combined with infrequent exposure to small amounts of CM (OR 5.7, 1.5-21.6), and long breast-feeding (OR 3.9, 1.6-9.8). The prevalence of parental atopy was similar in the IgE-positive and IgE-negative groups: 33% and 42%, respectively.

The mean concentrations of IgA and milk-specific IgA were similar in colostrum samples from mothers ofinfants with IgE-mediated and non-IgE-mediated CMA and also from control subjects (Table 3). IFN-y) could be measured only in 27% and IL-6 only in 48% of samples. For both IFN-y) and IL-6, the mean levels of samples with values above the detection level were similar in the study groups and the CIs for these samples were wide (data not shown). The mean concentration ofTGF- pi in colostrum of mothers of infants with IgE-mediated CMA was significantly lower than in samples from mothers of infants with non-IgE-mediated CMA ( t=2.57, p=0.012; Table 3). The level ofTGF- pi in control samples was between the 2 groups ofallergic infants and did not differ from either group.

Table 3. Mean (95% CI) Concentrations ofIgA, Milk-Specific IgA and TGF- P1 in Colostrum Samples of Mothers of Infants with CMA and From Control Mothers [Number of samples]

Mothers of infants with

IgA, S/L

IgA CM antibodies, AU*

TGF-[31, pg/mL**

IgE-mediated CMA

2.4 (2.3-2.4) T691

188 (103-342) T681

589 (413-840) T651

Non-IgE-mediated CMA

2.0(1.5-2.8) T391

114(50-261) T381

1162 (881-1531) T371

Control subjects

2.1 (1.9-2.4) T2071

207 (160-269) T2031

807 (677-963) N261

The concentration ofTGF- P1 in colostrum significantly correlated with the serum antibody levels oflgAto P-lactoglobulin (r=0.204, p=0.04), IgG to a-casein (r=0.237, p=0.02), IgG to CM (r=0.240, p=0.02), diameters of the skin prick test responses with CM (r=-0.228, p=0.02) and SI's to a-casein (r=-0.282, p=0.04) and P-lactoglobulin (r=-0.347, p=0.01) measured from infants with CMA at the time of the challenge. Of the 315 mothers, 19% had obvious atopy. Maternal atopy had no influence on any of the parameters measured from colostrum

0 0

Post a comment