Lower limit of normal or FEV1/FVC <0.70 in diagnosing COPD: An evidence-based review

Aim: To review the currently available literature comparing the FEV1/FVC <LLN with a fixed value of FEV1/FVC <0.70 in diagnosing airflow obstruction in subjects aged >40 years. Methods: A structured MEDLINE, EMBASE and Cochrane search of English-language literature was conducted. Studies comparing prevalence rates according to the LLN and a fixed value were included. Attention was paid to the choice of the reference test or gold standard used. Results: Eighteen studies met the inclusion criteria. Sixteen studies compared the rates of subjects diagnosed with airflow obstruction by either definition of airflow obstruction without using a non-independent reference standard (level 4 studies). Using a fixed value of FEV1/FVC, an overall higher number of subjects were diagnosed with airflow obstruction that increased with age. Two studies included a follow-up phase comparing risks of either hospitalization or occurrence of respiratory symptoms and mortality (level 2b studies). Adjusted risks of hospitalization (HR 2.6) or mortality (HR 1.3) were significantly larger in subjects with an FEV1/FVC below 0.70 but above the LLN (in-between group) compared to subjects with normal lung function. Conclusion: The prevalence of spirometry-based COPD is greater when using the fixed value of FEV1/FVC in comparison to using the LLN. Based on one longitudinal study the in-between group appears to have a higher risk of hospitalization and mortality; therefore it seems that using the LLN of FEV1/FVC underestimates COPD. In absence of a gold standard of COPD longitudinal research will be necessary to determine which criterion is better and more clinically relevant. (C) 2011 Elsevier Ltd. All rights reserved.