Chronic obstructive pulmonary disease (COPD) was always believed to be a disease of progressive accelerated decline in lung function, even after quitting smoking. However, as it turns out, that time-honored (and still-taught) paradigm, based on some unsure assumptions and weak primary data, is not true. More recent investigations incorporating robust data sets from large populations with longer follow-up are showing that in fact, progression of lung function decline in people with COPD is highly variable and unpredictable. One study, for example, found that only 18% of people with COPD experienced a statistically significant decline in lung function over several years. Another showed that almost half of the studied COPD patients had normal or nearly normal lung function decline, compared to historical norms.
This begs the question, is lung function decline in COPD predictable? M. Bradley Drummond, Donald Tashkin, Robert Wise et al in the June 15, 2012 AJRCCM analyzed a mountain of data from the Lung Health Study to conclude that, yes: the lower lung function (FEV1/FVC) at the time of diagnosis of COPD in smokers, the greater the risk for rapid decline in lung function and early mortality.
What They Did
Authors analyzed 5,885 participants of the Lung Health Study (a randomized trial on smoking cessation), all of whom were smokers with an FEV1/FVC ratio less than 0.70 at the time of enrollment, recruited from the community and without a COPD diagnosis previously. 96% were white and two thirds were men. These were largely young or middle-aged people; average age was 49 years old. They all had repeated spirometric assessments and were followed for 12 years.
What They Found
Lower FEV1/FVC ratio at the time of enrollment was independently associated with increased risk for rapid lung function decline and early mortality. This effect persisted after controlling for sex, age, tobacco exposure history, and ongoing smoking status during the study. The decline in lung function relationship was linear and robust, while the risk of death in the stratified FEV1/FVC cohorts had highly overlapping confidence intervals (unsurprising, given the relative infrequency of death in comparison with the far larger body of spirometry data on each patient).
Significantly, excessive lung function decline was only observed when FEV1/FVC was below 0.65 at the time of enrollment.
What It Means
Authors argue that consideration should be given to revising the GOLD guidelines, with their threshold of an FEV1/FVC ratio of 0.70 for diagnosis of COPD, pointing out that this is never been validated as predictive of mortality, lung function decline, or other outcomes. Rather, in active smokers, the threshold should be redefined as 0.65, they suggest. They make the case that their data argues for renewed attention to advocating screening spirometry: presumably, smokers in the highest risk cohorts could be counseled more aggressively and effectively to quit smoking. Screening spirometry has in fact never been shown to be helpful by “scaring straight” active smokers. It’s at least possible, though, that if younger smokers are screened with spirometry and presented with evidence that they are at the highest possible risk for disability and death by continued smoking, it could help them quit.
Drummond MB et al. Spirometric Predictors of Lung Function Decline and Mortality in Early Chronic Obstructive Pulmonary Disease. AJRCCM 2012;185:1301-1306.