Detection of mycobacteria in clinical specimen using the mycobacteria growth indicator tube (MGIT) and the Lowenstein Jensen medium

The recovery rates of mycobacteria strains isolated from 1200 clinical spec imens using the mycobacteria growth indicator tube (MGIT) system and the co nventional Lowenstein Jensen medium (LJ) were assessed. Of the 87 mycobacte rial isolates recovered, 54 belonged to the M. tuberculosis complex (MTB) a nd 33 to the non-tuberculosis complex (NTM). MGIT recovered 78 (89.65%) myc obacteria isolates (51 MTB (94.44%) and 27 NTM (81.81%) and LJ recovered 70 (80.46%) mycobacteria isolates (49 MTB (90.74%) and 21 NTM (63.63%). Sixty one (70.1%) of the total mycobacteria isolates were recovered with both sy stems (46 (85.2%) MTB and 15 (45.5%) NTM. No significant difference was fou nd between MGIT and LJ (p > 0.05) in both MTB and NTM recoveries. The avera ge detection time for MTB was significantly shorter with MGIT than with LJ, in both the smear-positive specimens (8 vs 30 days: p < 0.0001) and smear- negative specimens (15 vs 30 days: p < 0.001). The average detection time o f NTM was also shorter for MGIT (15 vs 30 days: p < 0.0001). However, the c ontamination rate was higher in MGIT (8.5%) than in LJ (3%). The results su ggest that the use of MGIT contributes to a more rapid and effective diagno sis of mycobacterial infections particulary when combined with the classica l LJ.