Accuracy of Timika X-ray scoring system to predict the treatment outcomes among tuberculosis patients in India

Background Timika scoring system is a radiographic grading tool, widely employed for grading the severity of tuberculosis (TB). We evaluated the predictive accuracy of this tool for adverse treatment outcomes among TB patients in Indian setting. Methods We undertook a longitudinal analysis of cohort data under the RePORT-India consortium. Cohort having participants with active pulmonary TB were included. CXRs were independently scored by chest physicians. Timika scoring system had a total score of 140. The predictive nature of the tool was assessed using the ROC analysis. Results Around 364 laboratory confirmed TB patients were enrolled. The mean (SD) of overall Timika score was 62.3 (24.9). Sputum conversion was achieved among 218/260 (83.8%) patients available at end of intensive phase. AUC for Timika score was 0.53 (95% CI: 0.43–0.63) and for percent lung affected, was 0.56 (95% CI: 0.46–0.65). Unfavorable treatment outcome was observed among 67/287 (23.3%) at the end of continuation phase. AUC for percent lung affected was 0.62 (95% CI: 0.54–0.70) and for Timika score was 0.59 (95% CI: 0.51–0.67). Conclusion Both Timika scoring system and percent lung affected had poor predictive accuracy, highlighting the inability of a single CXR scoring system to predict the treatment outcome in Indian setting.