Cow’s milk protein allergy (CMPA) is the most common food allergy in infants, with incidence rates estimated between 2% and 3% in developed countries (1-4). The majority of infants with CMPA develop symptoms before one month of age. CMPA debuting after 12 months of age is extremely rare (2, 3). Allergic reactions to cow’s milk proteins (CMP) are most often IgE-mediated, causing cutaneous, gastrointestinal and respiratory symptoms. Food allergy manifestations may vary from clearly IgE-mediated to mixed reactions dominated by eosinophilic granulocytes as effector cells or clearly non-IgE mediated reactions. A differentiation between IgE-mediated and non-IgE mediated reactions to CMP cannot be made based on clinical symptoms alone (3). Studies have shown that infants with IgE sentitization to CMP have an increased risk of persisting CMPA, and are more likely to develop persisting allergy to other foods and inhalant allergies before three years of age (1, 3, 4). Food allergy (FA) and atopic dermatitis (AD) dominate in early childhood, while allergic asthma (AA) and rhinoconjunctivitis (RC) are more common later. This progression from FA and AD to AA and subsequently to RC is known as the atopic march (5).

Longitudinal prospective birth cohort studies provide a desciption of the natural course of allergic diseases and sensitization. There are very few longitudinal studies on the natural history of CMPA and the long term development of atopic diseases in children diagnosed with CMPA. The studies conducted on the natural course of CMPA report varying recovery rates, most likely due to the differences in methods and design.

From the 1-year birth cohort in this study, the clinical course of CMPA regarding remission rate during the first 15 years of life has previously been reported in detail (1, 6). Clinical, epidemiological and immunological aspects of CMPA in the same study population have been reviewed and evaluated in relation to other studies on this subject (3). A 26 year follow-up on the natural course of sensitization and allergic diseases in a random sample of 276 children (16%) from the same birth cohort has previously been reported (7). Further-more, other studies from the same birth cohort have been published (8-12).

The aim of this study was to investigate 1) the natural course of CMPA in a 1-year birth cohort of Danish children from birth until 15 and 26 years of age and 2) compare the development of atopic diseases in a group of children with CMPA (group A) with the development of atopic diseases in a random sample of 276 children from the same birth cohort (group B).

A cohort of 1,749 newborns, born during 1985 at the Odense University Hospital, was followed up prospectively for the development of CMPA during their first year of life. The infants were referred to the paediatric clinic at the hospital when one or more of the following symptoms occurred: Recurrent wheezing, rhinitis, atopic eczema, urticaria, erythema/exanthema, vomiting and/or diarrhoea not caused by coincidental infections or other demonstrable causes; failure to thrive and infantile colic not disappearing after advice on feeding technique. This part of the study has previously been described in detail (1, 8).

The diagnosis of CMPA was established by the following, generally accepted, criteria (1, 3, 6, 13, 14): a definite disappearance of symptoms after each of two dietary eliminations of cow’s milk and cow’s milk products; recurrence of identical symptoms after one challenge; and exclusion of lactose intolerance and coincidental infection.

Details on elimination and challenge procedures have previously been described in full (1, 3, 15). Once a diagnosis of CMPA was confirmed, the milk-free diet was continued until a new milk challenge had shown development of tolerance to cow’s milk protein. All infants with confirmed CMPA were rechallenged at 12 months of age and the infants with non-confirmed CMPA (negative milk challenge) were also reviewed at 12 months of age. In the event of continued clinical sensitivity to cow’s milk protein, rechallenges were performed at 18, 24, and 36 months of age. Infants with CMPA during the first year of life were investigated every 6 months until 3 years of age. Thereafter, re-challenge was performed every 12 months until 15 years of age, to assess continued clinical sensitivity. The one remaining subject with CMPA from 15 to 26 years of age, was not rechallenged further, due to the severity of symptoms persisting until 26 years of age.

SPT (ALK Soluprick®, standard panel) (16) and specific sIgE testing (CAP RAST at 18 months, 3, 5, and 10 years and UniCAP at 15 and 26 years; Pharmacia Diagnostics, Uppsala, Sweden) (17) were carried out against inhalant allergens and food allergens in all children with CMPA at 18 months, 3, 5, 10, 15, and 26 years of age. Furthermore, lung function was measured by spirometry at 10, 15 and 26 years of age.

Based on the results of SPT and specific sIgE testing to cow’s milk, the reactions were immuno-logically classified as A) IgE-mediated CMPA when SPT was ≥ 2+ and/or RAST ≥ class 2 or B) non-IgE-mediated CMPA when SPT <2+ and RAST < class 2. Method for classification of SPT and RAST has been described in detail by Høst et. al. (1).

From the same birth cohort, a population based sample of 276 (16%) was randomly selected for prospective follow-up of atopic diseases (6, 7).

At 0, 6 and 12 months the parents filled in a questionnaire regarding atopic heredity, environ-mental factors, respiratory tract infections, wheezing, eczema, adverse reaction to foods and other possible signs of allergic disease. At ages 18 months, 5, 10, 15, and 26 years, all participants underwent physical examination by a doctor at Hans Christian Andersen Children′s Hospital, Odense University Hospital, and were interviewed regarding the same factors as in the before-mentioned questionnaire.

Specific sIgE testing was also carried out in children at 18 months, 5, 10, 15, and 26 years of age (CAP RAST at 18 months, 5 and 10 years and UniCAP at 15 and 26 years; Pharmacia Diagnostics, Uppsala, Sweden). Lung function was measured by spirometry at 10, 15, and 26 years of age. (6, 7)

The study population of children with CMPA (group A) was compared to the population based sample (group B) regarding the current prevalence of atopic diseases; AD, AA and RC.