Sensitivity and Specificity of Early Diagnostic Tests of Lung Function in Smokers

• DOI: 10.1378/chest.79.1.6
• Publication Date: 2007-11-07T18:25:26Z
• Authors: James A. Dosman; David J. Cotton; Brian L. Graham; Debra L. Hall; Robert Li; Frank Froh; G. Dudley Barnett
• Keywords: Adult; Male; Nitrogen; Respiration; Smoking; Middle Aged; Respiratory Function Tests; 03 medical and health sciences; 0302 clinical medicine; Breath Tests; Spirometry; Humans; Female; Lung Diseases, Obstructive; Lung Volume Measurements; Pulmonary Ventilation; Lung

What are the relative sensitivities and specificities of the "early" tests of lung dysfunction? We describe the findings from a study of virtually the entire population of a rural pollution-free community. Using abnormal spirometry as a marker of obstructive disease, we evaluated the two tests obtained from the single-breath nitrogen curve, closing volume (CV/VC) and the slope of the alveolar plateau (delta N2/L), as well as combinations of the two tests. While CV/VC is highly specific (92.3 percent in male and 94.0 percent in female subjects), it lacks sensitivity (36.8 percent in male and 13.3 percent in female subjects) and is abnormal in only 10.0 and 6.5 percent of male and female smokers, respectively, a percentage not dissimilar from the percentage with abnormal spirometry. However, delta N2/L, abnormal in 24.1 percent of male smokers and 28.8 percent of female smokers, rates reasonably well with regard to both sensitivity (63.2 percent in male and 66.7 percent in female subjects) and specificity (79.3 percent in male and 74.0 percent in female subjects). The group of smokers with abnormal delta N2/L did include fair numbers with abnormal spirometry (20.7 percent in male and 16.1 percent in female subjects). A combination of the two tests (abnormal in either delta N2/L and/or CV/VC) has good sensitivity (68.4 and 80.0 percent for male and female subjects, respectively) and specificity (74.3 and 69.0 percent for male and female subjects, respectively).