Zdorovʹe Rebenka (Sep 2019)

Prediction of the severe course of bronchial asthma in children

  • V.A. Klymenko,
  • O.S. Kozhyna

DOI
https://doi.org/10.22141/2224-0551.14.5.2019.177406
Journal volume & issue
Vol. 14, no. 5
pp. 307 – 311

Abstract

Read online

Background. Bronchial asthma (BA) is a chronic he­terogeneous respiratory disease in children, 339 million people in the world suffer from it. Severe course of BA is characterized by difficulties in disease control and remains the cause of children’s di­sability and mortality. The study was purposed to improve healthcare delivery to patients suffering from BA by means of therapy individualization taking into account prognosis of severe course of disease. Creation of mathematical model to predict the severe BA course in children was the study objective. Materials and methods. The study included 70 patients aged 6 to 17 years with BA diagnosis and 20 apparently healthy children. One hundred and forty-two clinical and paraclinical parameters (personal data, complaints, case and life history, laboratory and instrumental results, including clinical blood and urine test, coprogram, spirography, immunolo­gical indicators and total immunoglobulin E as well as allergy tests, etc.) were analyzed. Both quantitative and qualitative characters were selected. Each qualitative character was marked as 1, if a patient had it, and 0, is the character was absent. Mathematical model for severe BA course prognosis was developed using logistic regression with step-by-step inclusion of regressors to analyze the functions and select significant criteria. Results. There were 10 most significant factors affecting the prognosis: atopic dermatitis, allergic rhinitis, blood eosinophilia, CD8 absolute number and CD25 relative number in the blood serum, total IgE, sensitization to allergens of cat hair, rabbit hair, sheep wool, house dust. The model efficiency was tested in 40 adolescents suffering from BA, including 20 patients with severe disease course and 20 patients with intermittent BA. The model specificity (0.85), sensitivity (0.90), positive predictive value (0.86) and negative predictive value (0.11) were defined. Conclusions. A mathematical model has been developed to predict the development of a severe BA course, which takes into account child’s past medical history, indicators of clinical, immunological blood test and sensitization data.

Keywords