• James E Hansen, Xing Guo Sun, David Adame, and Karlman Wasserman.
• Division of Respiratory and Critical Care Physiology and Medicine, Department of Medicine, Los Angeles Biomedical Research Institute at Harbor-UCLA Medical Center, Torrance, CA 90502, USA. jhansen@labiomed.org
• Respir Med. 2008 Dec 1; 102 (12): 1777-83.

BackgroundExtensively used current guidelines of the American Thoracic Society/European Respiratory Society (ATS/ERS) define a positive aerosolized bronchodilator (BD) response as: "...an increase in FEV(1) and/or FVC > or = 12% of control and > or = 200 mL." We hypothesized that BD responsiveness was better assessed using a statistical approach, linked to each individual's spirometric measurements, rather than the variability of others.DesignWe retrospectively analyzed 1-year's pre- and post-BD spirometric tests from our hospital's clinical laboratory. Using measurements of forced expiratory volume in 1-s (FEV(1)), forced expiratory volume in 3-s (FEV(3)), and forced vital capacity (FVC) from each of three satisfactory forced pre-BD and three satisfactory forced post-BD spirometric maneuvers, we classified each of 313 consecutive patient studies as responders or non-responders in two ways. First, we used ATS/ERS guideline criteria based on population variability. Second, we used unpaired, single-tailed t-tests at P<0.05 for FEV(1), FEV(3), and FVC, considering the variability of and difference between each individual's pre- and post-BD maneuvers.Results135 studies were both ATS/ERS and t-test non-responders, three were ATS/ERS responders and t-test non-responders, 86 were ATS/ERS and t-test responders, and 89 were ATS/ERS non-responders and t-test responders. The latter 89 included many patients with either low baseline FEV(1) (< 1.50 L) who could not reach the 200 mL increase criterion or high baseline FEV(1) (> 3.00 L) who could not reach the 12% increase criterion.ConclusionsWe believe individual t-tests may categorize patient's BD responsiveness better than ATS/ERS guideline criteria which are based on population responses and require both fixed volume and percentage changes. Its usefulness by others remains to be shown.

24 keywords...

hide…