A high-resolution computed tomography-based scoring system to differentiate the most infectious active pulmonary tuberculosis from community-acquired pneumonia in elderly and non-elderly patients

Objective

The objective of this study was to use high-resolution computed tomography (HRCT) imaging to predict the presence of smear-positive active pulmonary tuberculosis (PTB) in elderly (at least 65 years of age) and non-elderly patients (18–65 years of age).

Methods

Patients with active pulmonary infections seen from November 2010 through December 2011 received HRCT chest imaging, sputum smears for acid-fast bacilli and sputum cultures for Mycobacterium tuberculosis. Smear-positive PTB was defined as at least one positive sputum smear and a positive culture for M. tuberculosis. Multivariate logistic regression analyses were performed to determine the HRCT predictors of smear-positive active PTB, and a prediction score was developed on the basis of receiver operating characteristic curve analysis.

Results

Of 1,255 patients included, 139 were diagnosed with smear-positive active PTB. According to ROC curve analysis, the sensitivity, specificity, positive predictive value, negative predictive value, false positive rates and false negative rates were 98.6 %, 95.8 %, 78.5 %, 99.8 %, 4.2 % and 1.4 %, respectively, for diagnosing smear-positive active PTB in elderly patients, and 100.0 %, 96.9 %, 76.5 %, 100.0 %, 3.1 % and 0.0 %, respectively, for non-elderly patients.

Conclusions

HRCT can assist in the early diagnosis of the most infectious active PTB, thereby preventing transmission and minimizing unnecessary immediate respiratory isolation.

Key Points

? HRCT can assist in the early diagnosis of the infectious active PTB ? HRCT imaging is useful to predict the presence of smear-positive active PTB ? Predictions from the HRCT imaging are valid even before sputum smear or culture results