BACKGROUND: COPD is diagnosed by using FEV 1 /FVC, which has limitations as a diagnostic test. We assessed the validity of several measures derived from the expiratory phase of the flow-volume curve obtained from spirometry to diagnose COPD: the slopes that correspond to the volume expired after the 50% and 75% of the FVC, the slope formed between the peak expiratory flow (PEF) and the FVC, and the area under the expiratory flow/volume curve. METHODS: We conducted a cross-sectional diagnostic test study in 765 consecutive subjects referred for spirometry because of respiratory symptoms. We compared the reproducibility and accuracy of the proposed measures against post-bronchodilator FEV 1 /FVC < 0.70. We also evaluated the proportion of respiratory symptoms for the FEV 1 /FVC, FEV 1 per FEV in the first 6 s (FEV 6 ), and the PEF slope. RESULTS: The subjects had a mean age of 65.8 y, 57% were women, and 35% had COPD. The testretest intraclass correlation coefficient values were 0.89, 0.85, and 0.83 for FEV 1 /FVC, FEV 1 /FEV 6 , and the PEF slope, respectively. The area under the curve values were 0.93 (expiratory flow/volume), 0.96 (potential expiratory flow/volume), 0.97 (potential expiratory flow/volume at 75% of FVC), and 0.82 (potential expiratory flow/volume at 50% of FVC). The area under the receiver operating characteristic curve was 0.99 for FEV 1 /FEV 6 , 0.99 for the slope at 50% of the FVC, and 0.98 for the PEF slope. CONCLUSIONS: The FEV 1 /FEV 6 , PEF slope, and 50% FVC slopes had similar diagnostic performances compared with FEV 1 /FVC.

No abstract