Abstract
Background
The tuberculin skin test has been used to diagnose latent tuberculosis infection, but is not widely used to diagnose or exclude pulmonary tuberculosis. The objective of this study was to evaluate the diagnostic utility of the tuberculin test in diagnosing and excluding pulmonary tuberculosis, and differentiating pulmonary tuberculosis from nontuberculous mycobacteria (NTM) pulmonary disease, when a sputum acid-fast bacilli (AFB) smear was positive.
Material And Methods
From October 2002 to August 2003, among all the inpatients of the Division of Pulmonary and Critical Care Medicine at Samsung Medical Center, 258 patients with clinical suspicion of pulmonary tuberculosis were enrolled and underwent a tuberculin test.
Results
156 males and 102 females were included, with a mean age of 57.5 years. The final diagnoses included lung cancer in 89 cases (34.5%), pulmonary tuberculosis in 59 cases (22.9%), bacterial pneumonia in 33 cases (12.8%) and NTM pulmonary disease in 24 cases (9.3%). The positive tuberculin test rate was higher in the tuberculosis than non-tuberculosis group; 81.4 (48/59) vs. 42.4% (81/199). (p<0.001). In 208 patients with a negative sputum AFB smear, the result of the tuberculin test was positive in 69.4% (25/36) of the tuberculosis group and in 44.8% (77/172) of the non-tuberculosis group (p=0.007), so a positive result of the tuberculin test could predict pulmonary tuberculosis with 69.4% sensitivity, 55.2% specificity, a 24.5% positive predictive value and a 89.6% negative predictive value. In 50 patients with a positive sputum AFB smear, the positive rates of the tuberculin test were 83.9% (26/31) in tuberculosis group and 21.1% (4/19) in NTM pulmonary disease group (p<0.001), so a positive result of the tuberculin skin test could predict pulmonary tuberculosis with 83.9% sensitivity, 78.9% specificity, a 86.7% positive predictive value and a 75.0% negative predictive value.