| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Guest Access | Sign In via User Name/Password
|
|
|
|||||||
Guest Access | Sign In via User Name/Password |
|||||||||
Pittsburgh, PA
Dr. Townsend is principal/consultant, M. C. Townsend Associates, LLC, and adjunct assistant professor, University of Pittsburgh Graduate School of Public Health, Departments of Epidemiology and Environmental and Occupational Health.
Correspondence to: Mary C. Townsend, DrPH, M.C. Townsend Associates, 289 Park Entrance Dr, Pittsburgh, PA 15228; e-mail: mary.townsend4{at}verizon.net
Although the pulmonary community has generally agreed over time on what constitutes serious airways obstruction, there has been disagreement over the past 40 years on where the line should be drawn to separate normal from abnormal. Traditional clinical practice has relied on fixed cut-off points, such as 80% of the predicted value, to distinguish normal from abnormal results.1 But since the mid-1960s, clinicians and researchers have noted that the distributions of the primary spirometric measurements, FEV1 and FVC, are homoscedastic with age, so that normal healthy pulmonary function measures are likely to remain in a roughly fixed position relative to their predicted values as the subject ages.23456 For the past 15 years, the American Thoracic Society (ATS) official spirometry interpretation statements have discouraged the use of fixed spirometry cut-off points to define abnormality because normal individuals will drift below those fixed points with age, even if the subjects are not symptomatic or exposed to known respiratory hazards.
Many widely used spirometric indexes, such as the observed FEV1/FVC ratio and the FEV1 percentage of predicted, decline with age, although the impact of this behavior on the interpretation of results is not always appreciated. The FEV1/FVC ratio decreases with age because the FEV1 declines more rapidly with age than the FVC in normal subjects. And when an individual remains in the same position relative to his predicted FEV1 over time, his FEV1 percentage of predicted will decrease with age because of the shrinking denominator of the percentage of predicted index. For example, a young man’s FEV1 of 3.3 L (83% of his predicted 4 L) will decrease to 2.3 L (77% of his predicted 3 L) if he remains 0.7 L below his predicted value as he ages. Thus, the percentage of predicted index, which is intended to remove the effect of age on lung function, does so only partially.
To address these issues, the 1991 ATS Official Statement on Selection of Reference Values and Interpretative Strategies7 stated the following: (1) "in adults, it is not acceptable to use a fixed FEV1/FVC ratio as a lower limit of normal"; (2) "the use of 80% of predicted for a lower limit of normal for adult pulmonary function parameters is not recommended. This criterion [80% of predicted] works only for average persons and for a limited number of parameters"; and (3) "normal ranges should be based on calculated fifth percentiles." The "average persons" for whom 80% of predicted works were often white males, close to 6 feet tall and in their early 30s; and in this group, there is little difference between 80% of predicted and the fifth percentile lower limit of normal (LLN) that was endorsed by the ATS. However, for individuals with smaller predicted values, because they are shorter, older, nonwhite, and/or female, important differences between a fixed percentage of predicted cut-off and the fifth percentile LLN can be observed.89 Since the fifth percentile LLN declines with age as the distribution shifts downward in an aging population, the problems with using fixed cut-offs to define abnormality are avoided. The 2005 ATS-European Respiratory Society (ERS) Statement on Interpretative Strategies for Lung Function Tests10 reemphasized the importance of using fifth percentile LLNs and avoiding fixed cut-offs to designate abnormal pulmonary function.
However, between the 1991 and 2005 ATS official statements on interpretation of lung function, the fixed cut-off definitions of abnormality were once again put forward. In 2001, the World Health Organization and the National Heart, Lung, and Blood Institute introduced the Global Initiative for Chronic Obstructive Lung Disease (GOLD) to reduce the severity and prevalence of COPD, which has become the fourth-leading cause of death in the United States.11 GOLD targeted primary care practitioners and espoused the earlier fixed cut-off definitions of abnormality, ie, an observed FEV1/FVC ratio < 70% and an FEV1 percentage of predicted < 80%. These definitions were also adopted in 2004 by the ATS/ERS position paper on Standards for the Diagnosis and Treatment of Patients with COPD.12 Reverting back to fixed cut-off definitions of what constitutes abnormal test results will have the greatest impact when spirometry is used to screen groups who may be at risk for COPD but who show no symptomatic evidence of disease. If the fixed cut-off criteria of GOLD and the ATS/ERS position paper are used to draw the line between normal and abnormal, the risk of false conclusions seriously increases, since "at risk" individuals tend to be older, and have lower percentages of predicted FEV1 and lower FEV1/FVC ratios due to their aging alone, with or without COPD.
The article by Hansen et al13 in this issue of CHEST (see page 349) illustrates how the proportion of false-positive results increases dramatically with decade of age among never-smokers in the third National Health and Nutrition Examination Survey when airways obstruction is defined as an observed FEV1/FVC percentage < 70%, compared with the 5% who would be called abnormal when the fifth percentile LLN is used. The problem of increasing false-positive results with age certainly also plagues the evaluation of smokers, but positive findings cannot necessarily be treated as false in these subjects since they have active smoking exposure. The ad hoc "Falling Ratio Working Group" has recently comprehensively described the problems of false-positive results and competing definitions of abnormality, as well as provided fifth percentile LLNs for many widely used reference value prediction equations. Their comments and software are available on Dr. Philip Quanjer’s Web page at http://www.spirxpert.com/controversies/controversy.html. The reader is advised to take advantage of this resource.
Finally, supporters of using fixed cut-off points to define abnormality often comment on the "simplicity" of using a fixed FEV1/FVC ratio and FEV1 percentage of predicted to determine abnormal results. However, if more of the currently available spirometers would display the fifth percentile LLN (which they can easily do), interpretation incorporating the LLN would be greatly simplified. It is unfortunate that a number of spirometer manufacturers claim that the fifth percentile LLN is used in their interpretation algorithms (following the 2005 ATS-ERS recommendations), but they fail to include the LLN in their spirometry printouts. Primary care health professionals will be able to easily understand and use the LLN to evaluate their patients when the LLN is included as a standard feature in the printout of test results. When LLNs are printed out, primary care health professionals can avoid using the "simple" but flawed definitions of abnormality put forward by GOLD, and reduce the probability of false-positive results when screening for COPD.
Footnotes
The author has no conflicts of interest to disclose.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
