Editorial Manager(tm) for Journal of Thoracic Imaging Manuscript Draft Manuscript Number: JTI-06-64R2 Title: HIGH-RESOLUTION CT FINDINGS IN PATIENTS WITH PULMONARY TUBERCULOSIS: CORRELATION WITH THE DEGREE OF SMEAR POSITIVITY

Background: There are data about the relationship between morphologic findings on HRCT and the number of acid-fast bacilli (AFB) on sputum smears in patients with pulmonary tuberculosis (PTB). It was also shown that existence of cavities and airspace consolidation might be related to smear positivity in PTB patients. However there is no study suggesting a relationship between AFB on sputum smears and radiological extent of disease based on HRCT findings. Aim: In this study, we investigated a relationship between the degree of smear positivity and radiological extent of disease based on HRCT findings and, the degree of smear positivity and different pulmonary parenchymal changes on HRCTs of the PTB patients. Methods: Sixty-one male patients with PTB (mean age: 22+/-3.2) were included into the study. HRCT images were assessed for patterns, distribution, and profusion of pulmonary abnormalities. Dividing the lungs into three zones, profusion of abnormalities was assessed. A profusion score was given. Patients were divided as smear positive and smear negative and compared for the scores of HRCT findings. Smear positive patients were divided into four groups as per grading of the sputum AFB smear: Group I (sputum1+), Group II (sputum 2+), Group III (sputum 3+) and Group IV (sputum 4+). Correlations were investigated between the degree of smear positivity and the scores of HRCT findings. Results: A significant correlation between radiological extent of the disease based on HRCT and the degree of smear positivity was found (r=0.63, p=0.0001). There were also significant correlations between the degree of smear positivity and the scores of different HRCT findings. Nodule, cavity, and bronchial lesions are the most important contributors of the predictive properties of the total score. There was significant differences for the scores of HRCT findings between smear positive and smear negative patients. Conclusion: Our study suggests that radiological extent of disease based on HRCT findings in patients with PTB correlated with the degree of smear positivity. Different HRCT findings such as nodule, cavitation, ground-glass opacity, consolidation and bronchial lesion are significantly associated with smear-positive PTB. Particularly, nodules, cavities and bronchial lesions might be predictors of smear positivity in patients with PTB. This study also suggests that the thickness of cavity wall and the distance of cavity from central airways might be related to the degree of smear positivity.