The ratio of the forced vital capacity (FVC) that is not yet expired within the first 3 seconds of a forced exhalation is expressed with the following formula: 1 minus-forced expiratory volume in third seconds (1-FEV3) / FVC (1,2). Originally, Hansen et al. (3) showed that 1-FEV3/FVC may be used for the evaluation of small airways, may be an indicator of the distal expiratory obstruction and was more sensitive than forced expiratory flow (FEF)25-75 % in evaluating small airways (3). In chronic obstructive pulmonary diseases (COPD), although small airways are mainly involved, larger airways are also affected due to a number of factors, including the loss of ciliated epithelial cells, squamous metaplasia, thickening of the basement membrane, mucous gland hypertrophy and hyperplasia (4). All these factors contribute to irreversible obstruction mainly caused by progressive air trapping, which is a prominent feature of COPD. Both the peripheral and proximal airways are also affected not only in COPD but also in asthma. The forced expiratory volume in the first second (FEV1) mainly reflects large airways obstruction, and for FEV1 to become abnormal a significant amount of small airways must be affected (5). Later fractions of forced exhalation those occur after FEV1, such as FEV3 was proposed to be more sensitive to reductions in terminal expiratory flow (1,2,3). For that reason, FEV3, FEV3/FVC ratio and 1-FEV3/FVC were suggested to better assess small airways disease (3). Therefore, both in asthma and COPD, 1-FEV3/FVC may be an indicator of small airways dysfunction and air trapping. In order to detect the presence of air trapping in the lungs, lung volumes should be measured to determine the total lung capacity and the residual volume. However, since these methods are associated with increased medical costs and require sophisticated equipment, they are not widely utilized. However, 1-FEV3/FVC value can be readily calculated by the widely available standard spirometric examination, and thus may help to detect air trapping in patients with obstructive pulmonary disease. In order to test this hypothesis, the present study aimed to investigate the associations of 1-FEV3/FVC in obstructive lung diseases and its relationship with the spirometric measures and lung volumes that assess small airways dysfunction, which reflects hyperinflation and air trapping.

MATERIALS AND METHODS

A retrospective assessment of a total of 1110 participants with at least three acceptable spirometric manoeuvres who underwent body plethysmographic lung volume estimations (ZAN 500 Plethysmography, nSpire, Germany) between 2005 and 2012 at the Pulmonary Function Test Laboratory was carried out. Repeated tests of same person were excluded (according to duplicated name, surnames and identity card numbers). None of the authors have reported a conflict of interest prior to the study. The pulmonologists reviewed all of the pulmonary function tests on a daily basis. The technicians were trained in whole-body plethysmography techniques, and the laboratory supervisor also checked all the steps involved in the test procedures in terms of adherence to the American Thoracic Society and American Thoracic Society/European Respiratory Society guidelines (9-12). The whole-body plethysmography device was calibrated daily according to manufacturer’s guidelines and biological quality control was performed on a monthly basis. Patients younger than 18 years of age were excluded, and only pre-bronchodilator test results were utilized. 1-FEV3/FVC was calculated electronically by whole-body plethysmography for each patient; this can also be calculated by spirometers. 1-FEV3/FVC estimation: After FEV3 and FEV3/FVC measurements were obtained from records of the patients, 1-FEV3/FVC was calculated to show the remaining unexhaled vital capacity ratio in the lung at the end of the 3rd second [(FVC-FEV3)/FVC=1-FEV3/FVC]. There is controversy regarding appropriate criteria to define airflow obstruction by using the fixed threshold of 70% or the lower limits of normal (LLN) for the FEV1/FVC ratio (13). In the present study, firstly, we defined airflow obstruction by using the fixed threshold of 70% for the FEV1/FVC ratio by using pre-bronchodilator spirometry (14,15). Patients were assigned into either the group with FEV1/FVC <70% or the group with FEV1/FVC ≥70%. The two groups were compared in terms of FVC, FEV1, FEV1/FVC, FEF25-75, inspiratory capacity (IC), total lung capacity (TLC), residual volume (RV), RV/TLC, thoracic gas volume at functional residual capacity (FRC-pleth), FEV3, FEV3/FVC, and 1-FEV3/FVC. Secondly, in order to assess whether FEV3/FVC (accordingly, 1-FEV3/FVC) provides additional information on air trapping and hyperinflation to that of FEV1/FVC, we analysed correlations of FEV3/FVC abnormality. We defined FEV3/FVC abnormality by using the redefined LLN criteria for FEV3/FVC (16). Analyses were performed separately, for the whole study population, and the subgroups, including individuals with FEV3/FVC < LLN and FEV3/FVC ≥ LLN. Statistical analyses Statistical analyses were performed using Statistical Package for Social Sciences (SPSS) software version 21.0 (IBM SPSS Statistics for Windows, Armonk, NY: IBM Corp.) Continuous variables were expressed as mean ± standard deviation, whereas categorical variables were shown as the number and percentage of cases. Means and medians were compared using Student’s t-test or Mann-Whitney U-test, depending on the normality distribution of data. A p value <0.05 was considered an indication of statistical significance. In addition, the correlations between variables were tested using Spearman’s correlation analysis. The study protocol was approved by the Ethics Board (Approval No: 83045809/604.01/02-346067).

RESULTS

Of the overall study population, 607 (54.7%) were male, and 503 (45.3%) were female, with a mean age of 52.5±15.6 years, mean FEV3/FVC of 87.05% and 1-FEV3/FVC of 12% (Table 1). Of the total study population, 644 had a FEV1/FVC ratio ≥70%, and 466 had FEV1/FVC <70%. Mean FEV3/FVC was 95.1% in the group with FEV1/FVC ≥70% and 75.9% in the group with FEV1/FVC <70%, while the corresponding 1-FEV3/FVC values in these two groups were 4.9% and 24.1%, respectively (Table 2). The upper 95% confidence limit for 1-FEV3/FVC was 13.7%. Individuals with FEV1/FVC <70% had a significantly higher TLC, RV, FRCpleth, RV/TLC, and a significantly more reduced IC than those with FEV1/FVC ≥70% (Table 2). 1-FEV3/FVC had moderate to strong and significant correlations with RV (r=0.70; p<0.0001), FEF25-75 (r=−0.84; p<0.0001), RV/TLC (r=0.59; p<0.0001), TLC (r=0.47; p<0.0001) and FRCpleth (r=0.61; p<0.0001) in the total study population (Table 3). When analysed separately in the FEV1/FVC ≥70% and FEV1/FVC <70% groups, we observed that 1-FEV3/FVC had significant correlations with RV, RV/TLC, TLC, FRC pleth and FEF25-75 (Table 3). Importantly, 1-FEV3/FVC displayed stronger correlations with RV, RV/LC, TLC, FRCpleth and FEF25-75 in those with FEV1/FVC <70% compared to those with FEV1/FVC ≥70% (Table 3). On the other hand, correlation of 1-FEV3/FVC with IC was weak in the total study population and both FEV1/FVC ≥70% and FEV1/FVC <70% groups (Table 3). In a further analysis, we assessed FEV3/FVC normality by the newly defined FEV3/FVC LLN criteria. A total of 379 (34.1%) of the whole study population were below the LLN for FEV3/FVC. Individuals with FEV3/FVC < LLN had a significantly higher TLC, RV and RV/TLC, and a significantly more reduced IC than those in the group with FEV3/FVC ≥ LLN (Table 4). 1-FEV3/FVC had significant correlations with RV, RV/TLC, TLC, FRCpleth and FEF25-75 in both FEV3/FVC ≥ LLN and FEV3/FVC < LLN groups. 1-FEV3/FVC displayed stronger correlations with RV, RV/TLC, TLC, FRCpleth and FEF25-75 in those with FEV3/FVC < LLN compared to those with FEV3/FVC ≥ LLN (Table 4). More importantly, we observed somewhat higher correlation coefficients for FEV3/FVC with IC, FEF25-75 and the air trapping measures - RV and RV/TLC - in FEV3/FVC < LLN subgroup than the correlations observed in FEV1/FVC <70% subgroup (Table 5). We also observed that FEV1/FVC has a similar or slightly higher level of correlation with TLC, RV, FRCpleth, RV/TLC and FEF25-75 in the total study population and subgroup analyses (Table 3). But when airflow obstruction is defined by FEV3/FVC LLN criterion instead of FEV1/FVC, we observed that 1-FEV3/FVC displays a stronger correlation with TLC (r=0.66, p<0.0001), RV (r=0.67, p<0.0001), RV/TLC (r=0.55, p<0.0001), FRCpleth (r=0.66, p<0.0001), FEF25-75 (r=-0.82, p<0.0001) and even with IC (r=0.30, p<0.0001) (Table 4).

DISCUSSION

In the present study, we report that the fraction of FVC that has not been expired at the end of the first three seconds of the FVC (1-FEV3/FVC), is significantly increased in patients with a FEV1/FVC below 70%. Both groups, including FEV1/FVC <70% and FEV3/FVC < LLN subjects, had significantly increased hyperinflation and air trapping with regard to RV, RV/TLC, TLC compared to FEV1/FVC ≥70% and FEV3/FVC ≥ LLN groups, respectively. We also showed that 1-FEV3/FVC significantly correlates with measures of hyperinflation and air trapping in the whole study population as well as in subgroup analyses, including FEV1/FVC <70% and FEV3/FVC < LLN subjects. Small airways are major contributors to airflow limitation in asthma and COPD (17). Air trapping and premature airway closing are accepted as useful surrogates to assess and quantify small airways obstruction. RV and RV/TLC ratios are useful and widely accepted measures of hyperinflation and air trapping (18). The earliest change associated with airflow obstruction is a reduction in the terminal portion of the spirogram, even though the initial part of the spirogram is barely affected (9). In this context, later fractions of forced exhalation, i.e. those that occur after the first second of exhalation, such as FEV3, were proposed to define reductions in terminal expiratory flow (1,2,3). FEV3 and FEV3/FVC were introduced in the last three decades, first by Crapo et al. (19) in 1981, followed by Miller et al. (20,21) in 1985. Later on, Hansen et al. (16) introduced the concept of 1-FEV3/FVC to identify the increased fraction of the long-time-constant lung units as a measure of late expiratory fraction in their study utilizing data from a smokers and never-smokers population of the Third National Health and Nutrition Examination Survey (22). Our study shows that 1-FEV3/FVC is a promising spirometric parameter that correlates with markers of air trapping and hyperinflation. 1-FEV3/FVC can be easily calculated by using standard spirometry through the measurement of FEV3 at the 3rd second of the forced expiratory manoeuvre. We suggest that 1-FEV3/FVC may be used to assess the presence of hyperinflation and air trapping, especially in settings where the lung volumes cannot be measured. Furthermore, FEV3/FVC LLN criteria define a group with significantly worse spirometric indices (FEV1, FEV3, FEV1/FVC, FEF25-75), and increased RV, RV/TLC, TLC compared to FEV3/FVC ≥ LLN subjects. Previously, FEV3/FVC and 1-FEV3/FVC were reported to be superior to FEF25-75 in the assessment of expiratory airflow limitation, since FEF25-75 can be misleading, with a high rate of false-negative and false-positive results (3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21,22). We observed that FEF25-75 had a high correlation with 1-FEV3/FVC in the total study population as well as in subgroup analyses. Interestingly, we found that FEF25-75 had a higher correlation with RV/TLC and IC than that of 1-FEV3/FVC, whereas 1-FEV3/FVC had a higher correlation with RV, TLC and FRCpleth than that of FEF25-75. But as we did not define normality vs. abnormality according to LLN for FEF25-75, our analysis did not allow a comparison of our results with previous findings. In addition to these results, we also observed that not only FEV3/FVC but also FEV1/FVC was negatively correlated with RV (r=−0.75; p<0.001), RV/TLC (r=−0.63; p<0.001) and TLC (r=0.49; p<0.001). We think this finding is consistent with Hansen’s suggestion that FEV1/FVC and FEV3/FVC are complementary and both ratios are beneficial in the characterization of expiratory airflow obstruction (3). Current studies and our own are still unable to answer the question of which ratio is better, 1-FEV3/FVC or the FEV1/FVC, in diagnosing expiratory airflow obstruction. The potential strengths of this study include the fact that the pulmonary function test laboratory where all of the tests were performed is the most comprehensive and qualified laboratory in the country, accepting referrals for whole-body plethysmography from more than 40 hospitals. For that reason, we believe our analysis reflects a wide range of a patient profile based on reliable measurements. However, its retrospective design with a lack of detailed history of smoking and other exposures, limited us in investigating the effect of smoking on spirometric measures effects of smoking on spirometric measures. In addition, our database does not include the necessary information regarding the medication history of the study participants. This was another limitation of our study. Nevertheless, whether the FEV3/FVC ratio translates into clinically meaningful disease-centred outcomes needs to be evaluated in further observations, together with clinical and radiologic features.

CONCLUSION

1-FEV3/FVC can be easily calculated from routine daily spirometric measurements. 1-FEV3/FVC is a promising marker of air trapping and hyperinflation. We suggest that 1-FEV3/FVC is complementary to FEV1/FVC and recommend clinicians routinely report this measurement and evaluate it together with FEV1/FVC during spirometry.. Conflict of Interest: No conflict of interest was declared by the authors.