Evaluation of the predictive validity of two indices for the diagnosis of asthma in a population of preschool children with recurrent wheezing in Bogotá, Colombia: preliminary analysis

Abstract

Introduction: identification of symptomatic infant and young children who will go on to develop persistent asthma symptoms among the whole group of preschool children with recurrent wheezing, would allow for better targeting of secondary prevention actions and therapeutic strategies for those who are most likely to benefit. To aid in the early identification of preschool children who wheeze and are at high risk of developing persistent asthma symptoms, a number of asthma predictive scores have been reported. However, to the best of our knowledge, the applicability and validation of these predictive asthma scores have not been performed in our city. Objective: the aim of the present study was to validate the Asthma Predictive Indice (API) and the Caudri´s predictive asthma score in a population of preschool children with recurrent wheezing living in urban Bogota, Colombia. Methods: we conducted a prospective cohort study in children less than 3 years old with recurrent wheezing who attended our outpatient pediatric pulmonary unit between October 2006 and August 2007. At time of enrollment we collected almost all information necessary to complete the predictive asthma scores proposed by Castro-Rodríguez JA et al (API), and Caudri D et al (PIAMA). To establish if patients had developed active asthma, they were assessed again at 5-6 years old, and parents were asked if the child had wheezing during the previous year, about the frequency of wheezing episodes, and if the child had a doctor diagnosis of asthma during the previous 12 months. Thereafter, we assessed the predictive validity of the scores comparing the predictive asthma scores and the case definition of asthma. Results: of the 43 patients candidates to be included in the study by having completed 5-6 years at the time of this analysis, 41 were included in the analysis of PIAMA score and 31 in the analysis of API score because they had all necessary information to complete the scores. The mean age at enrollment in the study was 27.4 ± 6 months. Of the total of patients, 19 (46.3%) met the definition of active asthma at 5-6 years of life. The loose API score yielded a sensitivity of 0.71 (0.45-0.88) and a specificity of 0.35 (0.17-0.59); the stringent API score yielded a sensitivity of 0.43 (0.21-0.67) and a specificity of 0.88 (0.66-0.97); and the PIAMA score yielded a sensitivity of 0.63 (0.41- 0.81) and a specificity of 0.64 (0.43-0.80) for predicting asthma at school age. Conclusions: when comparing the predictive validity reported in the literature of API and PIAMA scores, with that obtained in a sample of preschool children with recurrent wheezing living in Bogota, we found differences that can be explained by differences in methodology of the studies and a possible inappropriate report of wheezing by parents. These findings suggest that original scores could be used reliably with our patients.