Chest
Volume 126, Issue 6, December 2004, Pages 1861-1866
Journal home page for Chest

Clinical Investigations
PULMONARY FUNCTION TRAITS
Validity of the American Thoracic Society and Other Spirometric Algorithms Using FVC and Forced Expiratory Volume at 6 s for Predicting a Reduced Total Lung Capacity

https://doi.org/10.1378/chest.126.6.1861Get rights and content

Objectives:

(1) To compare the performance of three spirometric algorithms developed to predict whether the total lung capacity (TLC) is reduced vs normal or increased, (2) to determine if forced expiratory volume at 6 s (FEV6) can be substituted for FVC in these algorithms, and (3) to determine if ascertainment bias was present in patients referred for the measurement of spirometry and TLC compared to patients referred for spirometry only.

Methods:

We analyzed the results of 219 consenting consecutive patients referred to a New Zealand tertiary hospital respiratory laboratory for spirometry and TLC measurements. Spirometry results from 370 patients referred for spirometry but not lung volumes were used to test for potential ascertainment bias. Spirometry results were analyzed using the lower limit of normal (LLN) values from the third National Health and Nutrition Examination Study reference equations. The equations of Goldman and Becklake, and Crapo were used to classify TLC as normal or abnormal. Receiver operator characteristic curves were used to produce an algorithm using the LLN for FVC and FEV6. The performances of previous algorithms and our own algorithms were analyzed for predicting a reduced lung volume against the “gold standard,” plethysmographic TLC.

Results:

All three algorithms predicted a reduced TLC with an accuracy of approximately 50%. In contrast, all algorithms predicted TLC was either normal or increased with an accuracy of ≥ 99% regardless of the reference set used. The algorithms based on FEV6 performed equally as well as the FVC algorithms. No ascertainment bias was found.

Conclusions:

This study provides evidence that spirometry-based algorithms can accurately predict when TLC is either normal or increased, and can also increase the a priori probability that TLC is reduced to approximately 50%. FEV6 is equivalent to FVC in these predictions.

Section snippets

Materials and Methods

The Canterbury Ethics Committee, New Zealand, approved the study. Consecutive patients referred for testing in the Respiratory Physiology Laboratory at Christchurch Hospital between December 2000 and December 2001 were considered for the study. Patients scheduled for spirometry alone or spirometry and lung volumes were approached, the study was explained, and informed consent was sought. Two hundred nineteen patients referred for spirometry and lung volume measurements signed consent forms and

Results

Subject recruitment is summarized in Table 2. Eligible patients were approached if the laboratory workload allowed the time necessary to get informed consent. Workload issues resulted in only 27% of the spirometry-only patients and 30% of the spirometry and lung volume patients being approached. Of those approached, 67% of the spirometry-only and 61% of the spirometry-only and lung volume groups consented. The effects of workload and lack of consent, the causes of the low percentage of

Discussion

This study confirms the results of Aaron et al3 and Glady et al,4 who found that spirometry patterns could not reliably predict a reduced TLC but could reliably predict a TLC at or above the normal level (NPV = 99 to 100%). The NPV of these patterns was not affected by the reference set used, and substituting FEV6 for FVC led to equivalent outcomes. These findings are consistent with a previous report6 of good performance of FEV6 as a surrogate for FVC in establishing obstructive and

Summary

Clinicians can be confident in spirometry-based algorithms that predict a normal TLC, but restriction cannot be reliably diagnosed with spirometry alone. FEV6 is equivalent to FVC for excluding a low TLC. There was no evidence of ascertainment bias in the selection of patients to get TLC measurements.

ACKNOWLEDGMENT

The authors thank Debbie Murray, Kelly Sallaway, and Fiona McClymont for technical assistance and Janet Embry for editorial assistance.

References (12)

There are more references available in the full text version of this article.

Cited by (44)

  • Lung Function Tests in Clinical Decision-Making

    2012, Archivos de Bronconeumologia
    Citation Excerpt :

    There are no data to document the use of FRC or RV categories in airflow obstruction or TLC in pulmonary restriction in order to classify severity, as done in spirometry. Cases of low TLC and normal VC are exceptional3,6,35–37; therefore these measurements are generally not very useful in subjects with normal VC. Its use has not been demonstrated in the differential diagnosis between emphysema and chronic bronchitis, or between COPD and asthma.36,38

  • Prognostic significance of surrogate measures for forced vital capacity in an elderly population

    2010, Journal of the American Medical Directors Association
    Citation Excerpt :

    Indeed, this measure could not be measured in only 15 of our patients, and in patients not attaining the FVC, who tended to be older and frailer,10 it was significantly associated with mortality. For diagnostic purposes, FEV1/FEV6 is a well-recognized alternative to FEV1/FVC for the diagnosis of bronchial obstruction and FEV6 is an acceptable alternative to FVC for the diagnosis of restriction.11,12,18–20 The proposed index, SFVC, is unlikely to have noticeable diagnostic and classificatory properties because it provides a very approximate estimate of the vital capacity and its repeatability is suboptimal, but further studies should investigate this issue, especially in the elderly.

  • Changes in management of chronic obstructive pulmonary disease (COPD) in primary care: EMMEPOC study

    2010, Respiratory Medicine
    Citation Excerpt :

    Vandevoorde et al.23 calculated the values of FEV1/FEV6 and FEV6 that would best match an FEV1/FVC of 0.70 of predicted values, and they found FEV1/FEV6 and FEV6 suitable for screening purposes in primary care. These data established cutoffs for the detection of obstructive spirometric patterns and were consistent with previous studies stressing the importance of spirometry in primary care, using portable airflow monitoring devices.17,23–25 Despite the fact that the KoKo Peak Pro 6 spirometer used was not validated to the spirometer standards of either the American Thoracic Society or the European Thoracic Society,26 it provides an easily reproducible test and involves measures such as PEF, FEV1% and FEV6%, which are simple to interpret in general practice.

  • When the Heart Is Not to Blame: Managing Lung Disease in Adult Congenital Heart Disease

    2018, Progress in Cardiovascular Diseases
    Citation Excerpt :

    A normal FVC has high negative predictive value in excluding restrictive lung impairment. However, as mentioned above, an abnormally low FVC has modest positive predictive value, and referral for pulmonary function testing with body plethysmography is ideally needed to confirm the diagnosis of restrictive lung disease by demonstrating a reduction in lung volumes.17 DLCO measured during pulmonary function testing may also elucidate the etiology behind the low FVC, by differentiating an intrinsic or parenchymal lung abnormality (low DLCO) from extrinsic restrictive causes (normal DLCO).

View all citing articles on Scopus

This work was performed at the Respiratory Physiology Laboratory, Christchurch Hospital, Christchurch, New Zealand.

View full text