Disclaimer: This is a machine generated PDF of selected content from our databases. This functionality is provided solely for your convenience and is in no way intended to replace original scanned PDF. Neither Cengage Learning nor its licensors make any representations or warranties with respect to the machine generated PDF. The PDF is automatically generated “AS IS” and “AS AVAILABLE” and are not retained in our systems. CENGAGE LEARNING AND ITS LICENSORS SPECIFICALLY DISCLAIM ANY AND ALL EXPRESS OR IMPLIED WARRANTIES, INCLUDING WITHOUT LIMITATION, ANY WARRANTIES FOR AVAILABILITY, ACCURACY, TIMELINESS, COMPLETENESS, NON-INFRINGEMENT, MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Your use of the machine generated PDF is subject to all use restrictions contained in The Cengage Learning Subscription and License Agreement and/or the Gale Academic OneFile Terms and Conditions and by using the machine generated PDF functionality you agree to forgo any and all claims against Cengage Learning or its licensors for your use of the machine generated PDF functionality and any output derived therefrom.
How Accurate Is Spirometry at Predicting Restrictive Pulmonary Impairment?(*) Authors: Shawn D. Aaron, Robert E. Dales and Pierre Cardinal Date: Mar. 1999 From: Chest(Vol. 115, Issue 3) Publisher: Elsevier B.V. Document Type: Article Length: 3,460 words
Full Text: Objective: To determine the accuracy with which spirometric measurements of FVC and expiratory flow rates can diagnose the presence of a restrictive impairment.
Design: The pulmonary function tests of 1,831 consecutive white adult patients who had undergone both spirometry and lung volume measurements on the same visit over a 2-year period were analyzed. The probability of restrictive pulmonary impairment, defined as a reduced total lung capacity (TLC) below the lower limit of the 95% confidence interval, was determined for each of several categoric classifications of the spirometric data, and additionally for each of several interval levels of the FVC and the [FEV.sub.1]/FVC ratio.
Setting: A large clinical laboratory in a university teaching hospital using quality-assured and standardized spirometry and lung volume measurement techniques according to American Thoracic Society standards.
Results: Two hundred twenty-five of 1,831 patients (12.3%) had a restrictive defect. The positive predictive value of spirometry for predicting restriction was relatively low; of 470 patients with a low FVC on spirometry, only 41% had restriction confirmed on lung volume measurements. When the analysis was confined to the 264 patients with a restrictive pattern on spirometry (ie, low FVC and normal or above normal [FEV.sub.1] /FVC ratio), the positive predictive value was 58%. Conversely, spirometry had a very favorable negative predictive value; only 2.4% of patients (32 of 1,361) with a normal vital capacity (VC) on spirometry had a restrictive defect by TLC measurement. The probability of a restrictive defect was directly and linearly related to the degree of reduction of FVC when the FVC was [is less than] 80% of predicted (p = 0.002). Combining the FVC and the [FEV.sub.1]/FVC ratio improved the predictive ability of spirometry; for all values of FVC [is less than] 80% of the predicted amount, the likelihood of restrictive disease increased as the [FEV.sub.1]/FVC ratio increased.
Conclusions: Spirometry is very useful at excluding a restrictive defect. When the VC is within the normal range, the probability of a restrictive defect is [is less than] 3%, and unless restrictive lung disease is suspected a priori, measurement of lung volumes can be avoided. However, spirometry is not able to accurately predict lung restriction; [is less than] 60% of patients with a classical spirometric restrictive pattern had pulmonary restriction confirmed on lung volume measurements. For these patients, measurement of the TLC is needed to confirm a true restrictive defect. (CHEST 1999; 115:869-873)
Key words: diagnosis; lung volumes; respiratory function tests; restrictive lung disease; spirometry; vital capacity
Abbreviations: CI = confidence interval; FRC = functional residual capacity; TLC = total lung capacity; VC = vital capacity
Lung diseases are broadly classified as those leading to airflow obstruction, volume restriction, or a combination of obstructive and restrictive defects. Airflow obstruction can be diagnosed by a low [FEV.sub.1]/FVC ratio on spirometry. Spirometry is also often used as a screening tool to diagnose or rule out restrictive pulmonary impairment; however, its reliability and accuracy at predicting a restrictive process is unknown. Currently, the gold-standard diagnosis of restrictive pulmonary impairment requires measurement of the total lung capacity (TLC) through helium dilution or plethysmograph techniques.
A low FVC can result from true restrictive pulmonary disease, or from airflow obstruction with excessive air trapping. In the absence of airflow obstruction, it has been suggested that a restrictive defect can be diagnosed by spirometry alone. Therefore, a low spirometric FVC together with a normal or high [FEV.sub.1]/FVC ratio has traditionally been classified as a restrictive abnormality. However, studies verifying the accuracy of this interpretation are lacking.
Similarly, there are no studies verifying the accuracy of spirometry at ruling out restrictive impairment. Although it has been suggested that lung volume measurements may be useful in providing evidence of a diagnosis of restrictive lung disease not clearly evident from spirometric testing, this would be true only if spirometry had a poor negative predictive value for ruling out restrictive disease.
It would be convenient to be able to diagnose, or rule out, the possibility of a restrictive defect by spirometry alone, without measurement of TLC, because measurement of lung volumes is time-consuming and expensive. Avoidance of unnecessary lung volume testing can help to achieve a reduction in health care costs. The purpose of this study was to determine the predictive value of spirometry for diagnosing, and for ruling out, a restrictive pulmonary defect in a population of consecutive patients referred to a tertiary care pulmonary function laboratory.
MATERIALS AND METHODS
Test results from all white patients who had undergone both spirometry and lung volume measurements on the same visit from January 1, 1991 to January 1, 1993 were included. Nonwhites, and patients with technically inadequate tests (because of poor patient effort or inability to perform spirometry) were excluded. Pulmonary function tests were ordered at the discretion of the patients’ physicians. If the patient had undergone multiple lung volume measurements over this g-year period, we used only the measurements from their first visit. Lung volumes were always performed after spirometry. For patients who had undergone spirometry before and after administration of a bronchodilator (salbutamol, 200 [micro]g, via metered-dose inhaler with a spacer device), the postbronchodilator spirometry was used as the reference and was compared to the postbronchodilator lung volume measurements.
Spirometric testing was done by experienced pulmonary technologists certified and licensed by The Canadian College of Respiratory Therapists and The Canadian Association of Cardio-Pulmonary Technologists. Spirometry was performed according to American Thoracic Society criteria. Testing was done with the subjects seated, and a maximum forced exhalation was carried out for a minimum of 6 s. After 6 s, the test continued until zero flow was achieved ([is less than or equal to] 50-mL flow extrapolated over 30 s). Testing was repeated until a minimum of three acceptable flow volume loops with [FEV.sub.1] and FVC within 5% were obtained. Helium dilution and plethysmograph techniques were similarly performed according to published guidelines. For helium dilution, patients were asked to perform an inspiratory capacity maneuver every 30 s to speed equilibration, and attempt to open closed lung areas. Mixing time was prolonged until the stability of helium dilution remained constant ([is less than] 0.02% variation) for a minimum of 30 s. For plethysmography, a minimum of two attempts with the FRC reproducible within 5% were made for each patient. Spirometric and helium lung volume measurements were made using a Gould 2450 Complete PF System (Gould; Dayton, OH) or a PK Morgan Complete PF System (PK Morgan; Chatham, England). Spirometric and plethysmographic lung volume measurements were made using a body plethysmograph (2800 AutoBox; Gould). All machines were calibrated with a 3-L volumetric syringe daily.
Lung volume testing was performed by the method of helium dilution in 77% of cases, and by plethysmograph in 22% of patients; 1% of patients had both methods used to determine lung volumes.
Measurements of FVC, [FEV.sub.1], [FEV.sub.1]/FVC, and TLC were expressed as a percentage of predicted to control for the influence of age, gender, and height. We used published predicted equations taken from healthy, nonsmoking white patients. Spirometric reference values used were from Knudson et al for those patients [is less than] 65 years old and from Enright et al for patients [is greater than or equal to] 65. Lung volume reference values used were from Crapo et al. Test values for the FVC, [FEV.sub.1]/FVC ratio, and TLC that fell below the lower limit of the 95% confidence interval (CI) of the predicted value were classified as abnormal.
Spirometry data were classified categorically as being consistent with a normal pattern (normal FVC and [FEV.sub.1]/FVC ratio), an obstructive pattern with a normal FVC (reduced [FEV.sub.1]/FVC ratio, normal FVC), an obstructive pattern with a reduced FVC (reduced [FEV.sub.1]/FVC ratio, reduced FVC), or a restrictive pattern (reduced FVC, normal or above normal [FEV.sub.1]/FVC ratio). The probability of restrictive pulmonary impairment, defined as a reduced TLC below the lower limit of the 95% CI, was determined for each of the categoric classifications of the spirometric data, and additionally for each of several interval levels of the FVC and the [FEV.sub.1]/FVC ratio.
CIs around estimates of probability were calculated using the binomial distribution, except when point estimates for probability were [is less than] 0.05 in which case the Poisson distribution was used. The relationship between individual spirometric variables and the probability of restrictive disease was assessed in a logistic model with [chi square] testing for linear trend, p Values [is less than] 0.05 were accepted as significant.
A total of 1,879 pulmonary function tests were performed on white patients during the 2-year study period. 1,831 of these tests were technically adequate and therefore were included in data analysis.
The mean ([+ or -] SD) age of the patients was 50.8 [+ or -] 17.6 years, and 49.8% of the patients were males. Four hundred seventy patients (25.7%) had an FVC on spirometry that was below the 95% predicted CI. Two hundred twenty-five patients (12.3%) had a restrictive defect confirmed on lung volume testing. Ten percent of those patients who were tested with plethysmography were found to be restricted vs 13.0% of those patients tested using the helium dilution method (p = 0.10).
A simple spirometry classification system is pictured in Table 1. Patients were grouped as having a normal spirometric FVC or a low FVC (FVC below the lower limit of the predicted 95% CI). The sensitivity and specificity of a low FVC for determining true restrictive pulmonary impairment were 86% and 83%, respectively. However, the positive predictive value of spirometry for predicting restriction in this population was relatively low; of 470 patients with a low FVC on spirometry, only 41% had restriction confirmed on lung volume measurements. Conversely, the negative predictive value was quite high; only 2.4% of 1,361 patients with a normal VC on spirometry had a restrictive defect by TLC measurement.
Table 1–FVC vs Lung Volume Testing: Test Characteristics of Spirometry vs the Gold Standard(*) No. of Restricted Patients([dagger]) No. of patients with a low FVC([sections]) 193 No. of patients with a normal FVC 32 Total 225 No. of Nonrestricted Patients ([double dagger]) No. of patients with a low FVC([sections]) 277 No. of patients with a normal FVC
1,329 Total 1,606
(*) Sensitivity = 193/225 = 86%; specificity = 1,329/1,606 = 83%; positive predictive value = 193/470 = 41%; negative predictive value = 1,329/1,361 = 97.6%.
([dagger]) TLC below lower limit of 95% CI.
([double dagger]) Normal TLC.
([sections]) Below lower limit of 95% CI.
Results of the analysis did not differ significantly for patients who had lung volumes measured by helium dilution vs plethysmography. For the 1,410 patients tested by the helium dilution method, the sensitivity and specificity, respectively, of the FVC for determining pulmonary restriction were 86% and 82% vs 82% and 86% for the 421 patients who underwent plethysomography. Positive and negative predictive values were 44% and 97%, respectively, in the patients tested by helium dilution, vs 24% and 99% in the patients tested by plethysmography.
We then reclassified the spirometric data into four physiologic groups (see Table 2). Spirometry data were classified categorically as being consistent with a normal pattern, an obstructive pattern with a normal FVC, an obstructive pattern with a reduced FVC, or a restrictive pattern, and the prevalence of pulmonary restriction among the four spirometric physiologic groups was determined (Table 2). When the definition of a restrictive spirometric pattern was thus narrowed to include only those patients with a low FVC and a normal or above normal [FEV.sub.1]/FVC ratio, the sensitivity of the classification scheme for determining restrictive impairment fell to 68% with a specificity of 93% (Table 3). The positive predictive value was 58%.
Table 2–Prevalence of Pulmonary Restriction Among the Four
Spirometric Physiologic Groups
Spirometric Obstructive with Pattern Normal Normal FVC Total no. of patients 1,101 260 No. of patients with true restriction 30 (2.7%) 2 (0.8%) confirmed on lung volume testing Spirometric Obstructive With Pattern Low FVC Restrictive Total no. of patients 206 264 No. of patients with true restriction 40 (19.4%) 153 (58.0%) confirmed on lung volume testing
Table 3–Restrictive Spirometric Pattern vs Lung Volume Testing: Test Characteristics of Spirometry vs the Gold Standard(*)
No. of Restricted Patients([dagger]) No. of patients with a restrictive 153 spirometric pattern No. of patients with a nonrestrictive 72 spirometric pattern Total 225 No. of Nonrestricted Patients([double dagger]) No. of patients with a restrictive 111 spirometric pattern No. of patients with a nonrestrictive 1,495 spirometric pattern Total 1,606
(*) Sensitivity = 153/225 = 68%; specificity = 1,495/1,606 = 93%; positive predictive value = 153/264 = 58%.
([dagger]) TLC below lower limit of 95% CI.
([double dagger]) Normal TLC.
As depicted in Figure 1, the probability of a restrictive defect was linearly related to the degree of reduction of FVC for all values of FVC [is less than] 80% of predicted (p = 0.002). The probability of restriction was [is less than] 3% if the FVC was [is greater than] 80% of predicted, 27% (95% CI, 23 to 31) if the FVC was between 60 and 80% of predicted, and 50% (95% CI, 43 to 57) if the FVC was [is less than] 60% of predicted.
Combining the FVC and the [FEV.sub.1]/FVC improved the predictive ability of spirometry; for all values of FVC [is less than] 80% of predicted, the likelihood of restrictive disease increased as the [FEV.sub.1]/FVC ratio increased (Fig 2). For example, when the FVC was [is less than] 60% of predicted, the probability of a restrictive defect was 78% (95% CI, 65 to 91) if the [FEV.sub.1]/FVC was [is greater than] 80%, but only 14% (95% CI, 1 to 27) if the [FEV.sub.1]/FVC was [is less than] 40%.
This study has shown that spirometry is very useful at excluding a restrictive defect. However, the classical spirometric pattern of restriction can be misleading, because it represents a true restrictive defect in [is less than] 60% of cases.
In this study, the probability of a restrictive pulmonary defect was only 2.4% for the 1,361 patients with a normal FVC on spirometry. This very low false-negative rate is reassuring, because this rate falls within the expected [Alpha] error of the test, ie, we would have expected as many as 2.5% of normal patients to show a low TLC on lung volume measurements.
The very high negative predictive value of spirometry has enormous cost-saving implications. Lung volumes are, for the most part, only useful for diagnosing pulmonary restriction. Therefore, given the high negative predictive value of spirometry, 1,361 of 1,831 lung volume tests could have been avoided in our patient population. In the province of Ontario, the addition of lung volume measurements to spirometry adds $32.15 (Canadian) to the cost of pulmonary function testing. If we consider that measurement of lung volumes yielded new diagnostic information in only 32 of 1,361 patients with a normal spirometric FVC, then 42.5 tests would have been required to obtain each single test providing new diagnostic information. Thus, for patients with a normal FVC, the cost of one lung volume test providing new diagnostic information was $32.15 x 42.5 = $1,366.38 Canadian. These calculated cost savings might be expected to be even greater in centers in the United States where the charge for performing and interpreting lung volume measurements is on the order of $130.00 US per test.
Conversely, for those patients whose spirometry demonstrated the classical restrictive pattern, only 58% of patients were found to have true restriction when lung volumes were measured. This suggests that measurement of lung volumes is necessary to confirm a restrictive impairment in this subset of patients.
One potential criticism of our study is that lung volume measurements were performed using helium dilution techniques for 77% of patients and by using plethysmography for the remainder. In healthy subjects, helium dilution and plethysmography yield very similar results, and the coefficient of variation of repeated measurements for both methods, on the same subject, is [is less than] 10%. However, in patients with severe airflow obstruction, the true lung volume can be underestimated by the dilution method, unless mixing time is prolonged. To ensure that our method of lung volume measurement did not introduce systematic bias into our data, we repeated analyses for those patients who had lung volumes measured by plethysmography separately from those who had lung volumes measured by helium dilution. Results were similar for both groups of patients. In addition, we calculated the percent variation by which TLC measures of helium dilution differed from plethysmography in a sample of 16 patients from our data set. These 16 patients, all of whom had spirometric evidence of moderate to severe expiratory airflow obstruction (mean [FEV.sub.1], 1.61 [+ or -] 0.76 L; mean [FEV.sub.1]/FVC, 0.55 [+ or -] 0.15), had lung volume measurements done by both methods on the same visit to the pulmonary function laboratory during the study time period. The mean percent variation by which TLC measures of helium dilution differed from plethysmograph in these obstructed patients was only 9.7 [+ or -] 8.1%.
Finally, because helium dilution techniques can only underestimate lung volumes in patients with severe airflow obstruction, the effect, if any, of this potential bias would have been to overdiagnose restrictive impairment in those patients with airflow obstruction. This phenomenon was not seen in our patient population; of 260 patients with obstruction to airflow and a normal FVC, only 0.8% had evidence of restriction on lung volume testing.
The patients were selected for this study if spirometry and lung volumes had been measured on the same visit to the pulmonary function laboratory during the 2-year study period. This may have introduced the possibility of selection bias, ie, only selected patients were sent for both tests. Although selection bias may have been present, the effects of this bias would not have changed the conclusions of this study. Presumably, patients selected by their physicians for lung volume measurements had a greater prior probability of restrictive disease; however, despite this, we found that patients with a normal FVC on spirometry had a very low probability of having restrictive pulmonary impairment on measurement of lung volumes.
Recently, Lefante et al published a study in which they attempted to derive an equation to account for the effects of airflow obstruction on the observed FVC. They found that in patients with an [FEV.sub.1]/FVC ratio [is less than] 0.70, obstruction explained 15 to 17% of the variability in FVC percent predicted. They found that in these obstructed patients, the predicted FVC could be adjusted for the effects of obstruction according to the following equation:
FVC % predicted (adjusted)
= FVC % predicted (observed) + 76
When we used this published equation to adjust the FVC percent predicted for our patients whose spirometries showed an obstructive pattern with a low FVC (n = 206), we found that 103 of 206 patients (50%) had FVC measurements that were within 80% of their adjusted predicted FVC. Of these 103 patients, 9% had restrictive impairment confirmed by lung volumes. This suggests that adjusting the FVC percent predicted for the expected effects of obstruction can further reduce the number of patients whose spirometries show a low VC, although in our population this adjustment produced a 9% false-negative rate.
In summary, this study has served to determine the accuracy with which spirometric measurements of FVC and expiratory flow rates can diagnose the presence of a restrictive impairment. Patients whose FVC fall above the 95% CI of the predicted value are very unlikely to have a restrictive impairment, and in these patients, unless restrictive disease is suspected a priori, measurements of lung volumes can be avoided.
This study has also demonstrated that, contrary to popular opinion, a restrictive pattern on spirometry does not accurately predict true restrictive disease. Even in those patients with a classic spirometric restrictive pattern, the probability of true restrictive disease was [is less than] 60%. For any patient with a low VC, measurement of lung volumes is necessary before the patient is diagnosed as having a restrictive pulmonary impairment.
ACKNOWLEDGMENT: The authors thank Ms. Mary Lunan for her assistance with data gathering.
 Ries A. Measurement of lung volumes. Clin Chest Med 1989; 10:177-186
 American Thoracic Society. Lung function testing: selection of reference values and interpretative strategies. Am Rev Respir Dis 1991; 144:1202-1218
 Crapo R. Pulmonary-function testing. N Engl J Med 1994; 331:25-30
 American Thoracic Society. Standardization of spirometry: 1987 update. Am Rev Respir Dis 1987; 136:1286-1296
 Ferris BG. Epidemiology standardization project. Am Rev Respir Dis 188(Suppl 6); 1978:1-20
 Knudson R, Lebowitz M, Holberg C, et al. Changes in the normal maximal expiratory flow-volume curve with growth and aging.
Am Rev Respir Dis 1983; 127:725-734
 Enright P, Kronmal R, Higgins M, et al. Spirometry reference values for women and men 65 to 85 years of age. Am Rev Respir Dis 1993; 147:125-133
 Crapo R, Morris A, Clayton P, et al. Lung volumes in healthy nonsmoking adults. Bull Europ Physiopath Resp 1982; 18: 419-425