Effect of the Changing the Lower Limits of Normal and the Interpretative Strategies for Lung Function Tests.
10.4046/trd.2006.61.2.129
- Author:
Seung Won RA
1
;
Ji Seon OH
;
Sang Bum HONG
;
Tae Sun SHIM
;
Chae Man LIM
;
Youn Suck KOH
;
Sang Do LEE
;
Woo Sung KIM
;
Dong Soon KIM
;
Won Dong KIM
;
Yeon Mok OH
Author Information
1. Division of Pulmonary and Critical Care Medicine, Department of Internal Medicine and the Clinical Research Center for Chronic Obstructive Airway Diseases, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Korea. ymoh55@amc.seoul.kr
- Publication Type:Original Article
- Keywords:
Lung function test;
Interpretation;
Lower limits of normal;
Algorithm
- MeSH:
Chungcheongnam-do;
Consensus;
Humans;
Korea;
Lung*;
Pulmonary Disease, Chronic Obstructive;
Respiratory Function Tests*;
Spirometry
- From:Tuberculosis and Respiratory Diseases
2006;61(2):129-136
- CountryRepublic of Korea
- Language:Korean
-
Abstract:
BACKGROUND: To interpret lung function tests, it is necessary to determine the lower limits of normal (LLN) and to derive a consensus on the interpretative algorithm. '0.7 of LLN for the FEV1/FVC' was suggested by the COPD International Guideline (GOLD) for defining obstructive disease. A consensus on a new interpretative algorithm was recently achieved by ATS/ERS in 2005. We evaluated the accuracy of '0.7 of LLN for the FEV1/FVC' for diagnosing obstructive diseases, and we also determined the effect of the new algorithm on diagnosing ventilatory defects. METHODS: We obtained the age, gender, height, weight, FEV1, FVC, and FEV1/FVC from 7362 subjects who underwent spirometry in 2005 at the Asan Medical Center, Korea. For diagnosing obstructive diseases, the accuracy of '0.7 of LLN for the FEV1/FVC' was evaluated in reference to the 5th percentile of the LLN. By applying the new algorithm, we determined how many more subjects should have lung volumes testing performed. Evaluation of 1611 patients who had lung volumes testing performed as well as spirometry during the period showed how many more subjects were diagnosed with obstructive diseases according to the new algorithm. RESULTS: 1) The sensitivity of '0.7 of LLN for the FEV1/FVC' for diagnosing obstructive diseases increased according to age, but the specificity was decreased according to age; the positive predictive value decreased, but the negative predictive value increased. 2) By applying the new algorithm, 34.5% (2540/7362) more subjects should have lung volumes testing performed. 3) By applying the new algorithm, 13% (205/1611) more subjects were diagnosed with obstructive diseases; these subjects corresponded to 30% (205/681) of the subjects who had been diagnosed with restrictive diseases by the old interpretative algorithm. CONCLUSION: The sensitivity and specificity of '0.7 of LLN for the FEV1/FVC' for diagnosing obstructive diseases changes according to age. By applying the new interpretative algorithm, it was shown that more subjects should have lung volumes testing performed, and there was a higher probability of being diagnosed with obstructive diseases.
