SALT LAKE CITY — The use of an age-adjusted lower limit of normal in the FEV1/FVC ratio to define chronic obstructive pulmonary disease in the elderly results in a large problem of underdiagnosis, Dr. David M. Mannino said at the annual meeting of the American College of Chest Physicians.
The Global Initiative on Obstructive Lung Disease (GOLD) criteria provide a better approach to defining abnormal lung function in the elderly. The GOLD standard relies on a fixed ratio of the postbronchodilator forced expiratory volume in 1 second to forced vital capacity (FEV1/FVC) of less than 0.70 as the threshold regardless of age, explained Dr. Mannino of the University of Kentucky, Lexington.
How best to define COPD in the elderly has recently become a source of controversy among pulmonologists. The GOLD criteria, developed under the auspices of the National Heart, Lung, and Blood Institute and World Health Organization, have been highly influential since their publication in 2001.
But GOLD critics have noted that the FEV1/FVC ratio declines with age, based on population norms. For example, the age at which the ratio's lower limit of normal (LLN) slips below the 0.70 threshold is 52 years in white women, 41 years in white men, 54 years in black women, and 48 years in black men. This observation gave rise to concern that reliance on the GOLD criteria might result in overdiagnosis of mild COPD in a large fraction of elderly individuals whose lung function actually fell within the bounds of normal for their age.
In response to this argument, the latest American Thoracic Society/European Respiratory Society guidelines for the interpretation of spirometry, published in 2005, recommend adoption of the age-adjusted LLN to classify obstruction on spirometry. But there is a problem with this recommendation: It's based on analysis of cross-sectional data. And cross-sectional data do not provide any information about longitudinal outcomes—which is what really matters, Dr. Mannino said.
To elevate the level of the debate by introducing outcomes data, he and his coinvestigators turned to the National Institutes of Health-sponsored prospective epidemiologic Cardiovascular Health Study.
Dr. Mannino reported on 4,965 Cardiovascular Health Study participants aged 65 years and older who underwent baseline spirometry and up to 11 years of follow-up. Twelve percent were current smokers, and 42% were former smokers; 95% were white, and 57% were women.
The population of particular interest in this analysis was the 1,134 subjects whose baseline FEV1/FVC was less than 0.70 but above the LLN. The outcomes that were measured in this analysis were death or COPD-related hospitalization during follow-up.
Death occurred in 32.6% of the 4,965 subjects during follow-up, and 18.8% had one or more COPD-related hospitalizations. The subgroup whose FEV1/FVC fell between 0.70 and the LLN had an adjusted highly significant 30% increased risk of mortality and a 2.6-fold increased risk of COPD-related hospitalization during follow-up, compared with asymptomatic subjects with normal lung function.
“I think it's important to realize that if these people were characterized using the lower limit of normal, they would all be counted as normal—and they're clearly not normal,” Dr. Mannino noted, adding that it's likely that these sorts of patients would benefit from intervention.
“We need to be very cautious in changing guidelines based on cross-sectional data,” he continued. “Particularly in the area of lung function, cross-sectional and longitudinal data tend to be very different. You have to be alive for us to look at your lung function decline. And sick people tend not to show up for follow-up examinations. So cross-sectional data are always biased.”
With the lower limit of normal, they would all be counted as normal—and they're clearly not normal. DR. MANNINO