Background: In patients with cow's milk protein intolerance (CMPI), delayed clinical reactions to cow's milk (CM) ingestion may be misdiagnosed if the clinical symptoms are not "classical" and there is a long time lapse between ingestion of CM and the clinical reaction. The aim was to evaluate the clinical outcome of CMPI in a cohort of CM-intolerant children, with particular attention to the occurrence of clinical manifestations beyond 72 h after CM challenge.
Methods: Eighty-six consecutive patients (44 boys, 42 girls) with new CMPI diagnoses were enrolled; median age at diagnosis was 4 months. Patients were followed up for a mean period of 40 months. In all patients, CMPI diagnosis was made on the observation of symptoms, their disappearance after elimination diet, and their reappearance on double-blind CM challenge. At CMPI diagnosis, immunologic tests to demonstrate IgE-mediated hypersensitivity were performed. After 12 months of CM-free diet, CM tolerance was re-evaluated with a CM challenge continued at home for up to 30 days, according to a double-blind, placebo-controlled method. Patients who did not achieve CM tolerance continued a CM-free diet and subsequently underwent yearly CM challenge.
Results: The percentages of CMPI patients who became CM-tolerant after 1, 2, and 3 years of CM-free diet were 30%, 54.5%, and 70%, respectively. At the end of the follow-up period, 26/86 subjects showed persistent CMPI; these patients had a higher percentage of positivity of total serum IgE (P<0.05), RAST (P<0.01), and cutaneous prick tests for CM antigens (P<0.001) than all the others. At CMPI diagnosis, all patients had a clinical reaction within 72 h from the beginning of the CM challenge; at the subsequent "cure" challenges, we observed patients who first reacted to CM more than 72 h after ingestion. In total, 10 out of 86 patients showed "very delayed reactions"; in these patients, the mean time between the beginning of CM challenge and the onset of a clinical symptom was 13.3 days (range 4-26 days). The number of "very late reactors" increased from the first to the third of the "cure" CM challenges, performed at yearly intervals. The "very delayed" CMPI manifestations in these subjects were constipation (five cases), wheezing (two cases), dermatitis plus constipation (two cases), and dermatitis alone (one case); in 6/10 patients, the symptoms observed at the "cure challenge" were different from those at CMPI onset.
Conclusions: Very delayed clinical reactions to reintroduction of CM in the diet can occur in CMPI patients; thus, accurate follow-up and frequent outpatient observation in patients with a long history of CMPI are probably more useful and safer than prolonged CM challenge.