Background and Objectives: CB‑NAAT is a semi‑quantitative nested real‑time PCR which detects both MTB and RIF resistance directly from clinical specimens. This study was conducted to measure the diagnostic yield of CBNAAT in diagnosis of tuberculous meningitis (TBM). Methods: The present study was conducted among 62 children suffering from TBM in the department of Paediatrics at Subharti Medical Collegefrom November 2019 to July 2020. Complete physical examination was done including level of consciousness, signs of meningeal irritation (neck stiffness, Kernig's sign, Brudzinski's sign), cranial nerve involvement, etc.Following a lumbar puncture with standard and sterile method, about 10 ml CSF was obtained, transported to the laboratory within 1 h, and divided into four tubes: 1 (1-2 ml) for CSF cells, 2 protein and glucose, 3 bacterial smear and culture and 4 (8-10 ml) for TB PCR testing.Data so collected was tabulated in an excel sheet and analysed using SPSS 24.00 for window (SPSS inc, Chicago, USA). Results: Out of the 62 patients of TBM, 9 patients (14.5 %) had a positive CBNAAT in CSF while only one patient (1.61 %) had positive result for Ziehl -Neelsen staining for acid fast bacilli.In our study, sensitivity, positive predictive value and diagnostic efficacy CBNAAT was 47.30%, 98.20% and 47.04% respectively. Conclusion: It can be concluded from the results that CBNAAT is an efficient technique for detection of tuberculous meningitis in CSF samples. Its simplicity, speed and automation, and detection of resistance at the same time makes this technique a very attractive tool for the rapid diagnosis of TB meningitis, especially in suspected cases.