Diagnostic standard for differentiation between bronchial asthma and chronic obstructive pulmonary disease
Liebhart, J.; Dor, A.
Pneumonologia i Alergologia Polska 66(7-8): 373-382
1998
ISSN/ISBN: 0867-7077 PMID: 10354685 Document Number: 493186
The aim of this study was to work out the simplified standard for distinguishing COPD from bronchial asthma. An overall sample of 150 individuals was used for statistical analysis. Sixty one were diagnosed as having COPD and 89 as suffering from asthma. For each patient the modified ATS-DLD-75-C questionnaire was filled out, and laboratory tests results were collected. Laboratory findings included: basic spirometry and flow-volume curve parameters, diurnal PEF variation, tests with bronchodilators (salbutamol, ipratropium bromide and corticosteroids), challenges with histamine and exercise, blood gas analysis, skin tests, chest X-ray, ECG, blood cell count, blood and sputum eosinophilia. Continuous variables were transformed into discrete (dichotomous) ones using commonly accepted threshold values. Then, sensitivity, specificity and accuracy indexes were calculated for each variable and for all possible sets of 2, 3 and 4 variables. Finally, 2 sets of 4 signs and symptoms were selected as the most characteristic of the diseases of interest. Asthma was diagnosed if 3 out of 4 following conditions were present: 1) episodes of shortness of breath and wheezing, 2) smoking index (cigarettes number per day x years of smoking) < or = 200, 3) PC20 < or = 8 mg/ml or delta FEV1 after bronchodilator > or = 15% predicted, 4) diurnal PEF variation > or = 20% predicted. COPD was recognised in the same way on the basis of: 1) productive cough, 2) smoking index > 200, 3) signs of emphysema on the chest X-ray, 4) maximal FEV1 < 80% predicted (after treatment). Diagnosis established using this model was correct in 76% and false (what was very important) in only 2.6% of cases. In the remaining 21% of patients it was uncertain (e.g. both of the diseases confirmed).