Use of peak flow variability and methacholine responsiveness in predictingchanges from pre-test diagnosis of asthma

Asthma is usually diagnosed clinically. This study investigated how methach oline challenge and peak expiratory pow monitoring influenced change from a pretest clinical diagnosis. Records of 132 patients referred with respiratory symptoms, who subsequentl y had reliable measurements of both airway responsiveness (provocative conc entration of methacholine causing a 20% fall in forced expiratory volume in one second (FEV1 (PC20)) and peak expiratory flow variability (PEFV) were reviewed, Initial and final diagnoses for each patient were classified as: a) definite asthma; b) possible asthma; and c) definitely not asthma. The p redictive value of PEFV and PC20 regarding overall change from pre- to post -test diagnosis, change from initial diagnosis of possible or definitely no t asthma, and change from initial diagnosis of definite asthma, were tested by multiple logistic regression analysis. Odds ratios for PC20 were expres sed per doubling dose, and for PEFV per 5% variability. Clinical diagnosis of definite asthma and definitely not asthma were confir med in 70% and 79% respectively. PC20, but not PEFV, predicted an overall c hange between pre- and post-test diagnosis. Both PC20 and PEFV independentl y predicted change to definite asthma. PEFV and interaction between PC20 an d PEFV predicted a change in those whose initial diagnosis was definite ast hma. Although both measurements showed a significant correlation, there was poor agreement between positive tests. Both peak expiratory now variability and provocative dose of methacholine c ausing a 20% fall in forced expiratory volume in one second influence diagn ostic decision-making in patients with a high pre-test probability of asthm a.