Prevalence of Supranormal Pulmonary Function Test Values Between a Military and Nonmilitary Cohort

Discussion

Minimal information has been previously published on the frequency or significance of supranormal PFTs. Understandably, if elevated values are found during the performance of spirometry in a symptomatic patient, this finding is not typically indicative of a disease process and reassures the clinician that an obstructive or restrictive process is not present. Based on current population studies and spirometry reference values (eg, NHANES III), the derivation of expected normal values should be closer to 100% of predicted because of normal population distributions. However, knowing the baseline spirometry value of any given patient can be important in determining the presence or absence of pulmonary disease, as it can identify individual changes in pulmonary function over time. Especially in a younger, more fit population, does a normal spirometry in the presence of symptoms represent active pulmonary disease? This study has begun to answer several questions regarding supranormal PFT values. First, supranormal values were a common finding across all age ranges in our population and did not favor a specific age group where more comprehensive reference values may be lacking. Second, differentiating between AD personnel and their non-active counterparts did not demonstrate an increased frequency of supranormal PFT values. On the basis of this study, we have reservations about automatically defining the AD military population as having supranormal values on PFTs.

The joint American Thoracic Society/European Respiratory Society recommendations on pulmonary function testing and interpretation were published in 2005 and made several important changes.⁹ The primary change included incorporation of variances in the published reference equations from a higher percentage of minority populations.⁸ In addition, interpretation was changed to incorporate 95% CIs to define the lower limit of normal. There are multiple factors that can influence the performance of spirometry, and they are related primarily to patient effort and ability to consistently reproduce the forced expiratory maneuver. Incorrect height and age may drastically alter the predicted value, and thus incorrectly, the percent of predicted may be in the normal or supranormal range. Although both the upper and lower limits of normal for most reference equations are established, generally there is no clinical use of the upper limit of normal in the interpretation of spirometry. It is presumed that a patient being evaluated for pulmonary disease with normal values for FEV₁, FVC, and FEV₁/FVC would not have significant disease. In many instances, the evaluation for the presence of pulmonary abnormalities may end with a normal spirometry. In those evaluations in which there is concern for occupational disease, serial PFTs are used to detect significant clinical change. However, as the discussion in the American Thoracic Society guidelines adequately points out, there is a vast difference in the evaluation of healthy subjects and patients with disease or symptoms. Applying the same concept of “normal” based solely on PFT findings would be inappropriate, and the interpretation very much depends on the clinical question being asked.¹⁰

Early investigations suggested an increase in lung function during adolescence, especially noted in those individuals with high levels of activity.^11,12 One particular group with supranormal values includes competitive swimmers due to their particular type of exercise. An Australian study of 8 swimmers compared with 8 runners showed higher FEV₁ and FVC in the swimmer group, which were related primarily to an increase in chest wall width.¹³ This confirmed earlier findings in several other small studies of swimmers.^14,15

Given the hypothesis that highly athletic individuals have supranormal PFT values, the published literature on PFTs in competitive athletes should provide ample information. Six published articles were identified that measured resting spirometry values in long-distance runners (both marathon and triathlon participants).^16–21 These studies described primarily the effect of running on PFT values and provided baseline spirometry values in small groups of highly trained athletes. Table 4 provides the demographic information along with actual and predicted values for FEV₁ and FVC. From this group of 67 athletes with a mean age of 27 years, the mean FVC was 102.7%, and FEV₁ was 101.6%. None of the mean values from these 6 studies reached the 110% cutoff as defined in our study, although this group was younger (27 vs 31 y) than our AD population.

Because our study is retrospective in nature, there are several limitations to the data provided. Because the spirometers in our laboratory do not identify values outside the 95th percentile for the upper limit of normal, we had to establish a cutoff for review based on the percent of predicted, which showed a modest correlation with 90% and 95% CI. Other factors such as the indication for spirometry and smoking history were not listed in the PFT report, and we were unable to distinguish between symptomatic and asymptomatic individuals. Interpretation of spirometry is limited without other pertinent clinical information such as symptoms or underlying lung disease. Additional potential biases include the comparison of AD military with dependents and retirees. We assumed an overall improved fitness for the AD group but did not have any direct measurements. Also, although the repeatability of the values within a given test is required by PFT standards, we were unable to determine in our cohort if the supranormal values could be repeated over time. Finally, there are limited large population studies with PFT reference values with which we can compare our data.

Another consideration that has been discussed is the role of spirometry screening for all military service members. Spirometry is currently not used as a screening tool in asymptomatic individuals prior to enlistment or commissioning into the United States Armed Forces. Current guidelines do not recommend “screening” for common lung diseases in the general population and limit surveillance to those individuals with potential occupational exposures. There are a multitude of confounding factors surrounding this issue, including cost, manpower, quality control, and timing of screening spirometry, that make such a proposition difficult to accomplish. The main reasons not to perform spirometry screening include: (1) the limited use of a screening test in an asymptomatic population, (2) the very small likelihood of developing future respiratory disease in this population, and (3) the burden of evaluating abnormal baseline tests that may affect military careers.