ОПТИМИЗАЦИЯ РАННЕЙ НОЗОЛОГИЧЕСКОЙ ВЕРИФИКАЦИИ ОДЫШКИ ЛЕГОЧНОГО ИЛИ СЕРДЕЧНОГО ГЕНЕЗА

The aim of this study was early differentiation of breathlessness due to pulmonary or cardiac disease. The study was conducted in three stages. Firstly, we examined 359 patients aged 18 to 90 years with confirmed diagnosis of asthma ( n = 134, or 37.3 %), COPD (n = 114, or 31.8 %), and chronic heart failure (CHF) not related to lung disease ( n = 111, or 30.9 %). There were 154 males (42.9 %) and 205 females (57.1 %). We compared frequencies of P.M.Simon's "language of dyspnea" descriptors (1990) used by patients and excluded characteristics used equally often by patients with asthma, COPD or CHF. This allowed development a modified vocabulary of dyspnea (MVD) which consisted of 7 phrases and 6 clusters. Secondly, we developed a priori probability model to calculate probability of each diagnosis (PPD) using data from MVD and modified Bayesian classifier. Finally, we validated the PPD model with estimation of its sensitivity and specificity for asthma, COPD and CHF by comparing of a priori and a posteriori diagnoses. This analysis included a new cohort of 123 patients (75 (61.0 %) males, 48 (39.0 %) females) aged 18 to 89 years with breathlessness as the leading symptom. All the patients fulfilled MVD before the diagnosis to be confirmed. Subsequent standard clinical examination revealed asthma in 47 (38.2 %), COPD in 46 (37.4 %), CHF in 30 (24.4 %) respondents.