Comparison of the BACTEC MGIT 960 with Löwenstein-Jensen medium for recovery of mycobacteria from clinical specimens.

Taiwan Provincial Chronic Disease Control Bureau.To evaluate the rate of recovery and the mean time to detection (TTD) of mycobacteria in clinical specimens with two culture systems, the BACTEC MGIT 960 and Löwenstein-Jensen (LJ) medium.We studied 365 specimens, collected from 166 patients. Specimens were processed with standard N-acetyl-L-cysteine (NALC)-NaOH method, then inoculated onto BACTEC MGIT 960 and onto LJ slants.A total of 124 mycobacterial isolates (114 Mycobacterium tuberculosis and 10 non-tuberculous mycobacteria) were detected. The recovery rates were 94% (117/124) with BACTEC MGIT 960 and 75.8% (94/124) with LJ. The rates of contamination for each of the systems were 5.5% with BACTEC MGIT 960 and 4.1% with LJ. The TTDs for mycobacteria were 10.7 days with BACTEC MGIT 960 and 30.6 days with LJ. Excluding the non-tuberculous mycobacteria, the TTDs for M. tuberculosis were 11.1 days with BACTEC MGIT 960 and 30.7 days with LJ. The difference in TTD between smear-positive and smear-negative specimens for either mycobacteria (10.0 vs 12.6 days; P = 0.06) or M. tuberculosis (10.1 vs 12.7 days; P = 0.06) with BACTEC MGIT 960 was not statistically significant.The BACTEC MGIT 960 system can expedite the recovery of mycobacteria in culture. Combined with conventional solid medium, it also increases the overall recovery of mycobacteria in culture.