[Contribution of serum total immunoglobulin E measurement in the diagnosis of respiratory allergic diseases]

ABSTRACT

Serum total IgE measurement is one of the first allergic inflammation marker test. It has been currently used for the diagnosis of allergy. However, when it has been widely used in clinical practice, we discover its limit in the diagnosis of allergic disease. We carried a retrospective study about 1111 consecutive patients, seen in the Allergology Unit of the Home Ministry Hospital with suspicion of atopic disease. The aim of our study is to evaluate the contribution of serum total IgE measurement in the diagnosis of respiratory allergic diseases. Our population was divided into three groups. Group A 771 allergic symptomatic subjects [371 children with mean age of 7.8 +/- 3.4 years and 467 adults with mean age of 30.1 +/- 6.6 years]. Group B 117 controls [70 children with mean age of 6.6 +/- 3 years, and 47 adults with mean age of 30.7 +/- 6 years]. 221 patients with doubtful diagnosis of allergy were excluded [group C]. If we consider the cut-off I of 150 UI [for adults], 75% of allergic subjects have elevated serum total IgE, 62% of adults and 77.6% of children. When doubling the cut-off [cut-off II], 49% of allergic subjects have serum total IgE over 300 UI/ml, 36% of adults and 57% of children. In the control group 35.8% of patients have total IgE over the cut-off value I and 17% over the cut-off value II. The variation of serum total IgE was related in allergic subjects to the type and the number of symptoms. The levels of total IgE were higher in asthma than in rhinitis and more elevated if the patient has several symptoms. With cut-off I we found a quite good sensibility of the test of 75% [62% in adults and 77.6% in children] and a poor specificity of 64% [68% in adults and 61% in children], with a negative predictive value of 28% and a positive predictive value of 93%. When doubling the cut-off, the sensibility falls to 49% and the specificity was improved at 82%. The serum total IgE measurement was in agreement with prick-tests in 80% of cases and with Phadiatop in 56% of cases if we consider cut-off I. Our results showed a large overlapping of serum total IgE in both allergic and control populations. The measurement of serum total IgE is not helpful for the diagnosis of allergic respiratory diseases because it has a quite good sensibility, a poor specificity and a poor negative predictive value. The test was more sensible and less specific in children