Abstract
538
Background: More than 50% of patients who completed a standard course of anti-tuberculous therapy (ATT) for pulmonary tuberculosis and declared cured based current standard of care has residual metabolic activity (RMA) in the lungs on FDG-PET/CT. Dormant or slow-growing bacilli, which show poorer response to ATT, are one the causes of RMA. Patients with RMA on end-of-treatment (EOT) FDG-PET/CT are at risk of disease relapse. This study aimed to evaluate patient and imaging predictors of RMA in patients who completed a standard course of ATT for pulmonary tuberculosis.
Methods: Patients who completed a standard course of ATT and declared cured based on clinical evaluation, negative sputum microscopy for acid and alcohol-fast bacilli, and negative mycobacterial culture were prospectively recruited to undergo FDG-PET/CT scan within four weeks of completing treatment. Images were assessed for the presence of RMA on PET and lung changes due to tuberculosis on CT. We evaluated several patient factors and CT lung changes for their ability to predict RMA on EOT FDG-PET.
Results: Seventy-five patients were included with a mean age of 36.09 years ± 10.49. There were 47 males and 50 participants were HIV-infected with a median CD4 count of 255 cells/µL (Interquartile range, IQR=147 - 448). Among the HIV-infected, 35 patients had undetected HIV viremia. Out of 75 patients included, 41 (54.67%) had RMA on their EOT FDG-PET/CT. Among patient factors tested, age above 45 years (Odds ratio, OR [95% CI] = 0.23 [0.06 - 0.91], p=0.028); CD4 count ≥ 200 cells/µL (OR [95% CI] = 0.14 [0.04 - 0.53], p=0.002); and undetectable HIV viremia (OR [95% CI] = 0.17 [0.40 - 0.70]) were significant predictors of RMA on EOT FDG-PET/CT. CT findings predictive of RMA were cavity (OR [95% CI] = 3.76 [1.38 - 10.25], p=0.008); pulmonary nodules in tree-in-bud pattern (OR [95% CI] = 30.857 [6.44 - 147.97], p<0.001); lung consolidation (OR [95% CI] = 9.28 [1.11 - 77.51]); and bronchiectasis (OR [95% CI] = 2.95 [1.14 - 7.63]). Patient factors (including gender, smoking, body mass index, hemoglobin level and C-reactive protein) and CT changes (including lung fibrosis, pleural effusion, and parenchyma calcification) were not significant predictors of RMA. Conclusion: RMA on EOT FDG-PET/CT is a known risk factor for tuberculosis relapse. Factors predictive of RMA in patients who completed a standard course of ATT are older age, virologically and immunologically controlled HIV infection, and CT features such as lung cavity, nodules, consolidation, and bronchiectasis.