Abstract:

Objective To evaluate the value of mycobacterial biphasic L-J medium in the diagnosis of tuberculosis, and to explore its feasibility in China. Methods We conducted a multicenter trial to consecutively enroll all patients with suspected TB at 3 prefecture-level TB clinics in 3 provinces of China from October 2014 to October 2015. Of the 2320 suspicious patients enrolled, 15 cases had either one or two of the two cultures (mycobacterial modified L-J medium and mycobacterial biphasic L-J medium) with contamination or missing results, and their comparison of the culture results of the two media is not included in the analysis. 3 sputum specimens collected per patient (immediate sputum, night sputum, morning sputum) for Ziehl-Neelson stain smear microscopy (“smear”), and 2 sputum specimens with good traits were selected simultaneously for modified L-J medium and biphasic L-J medium culture. The positive detection rate of the two media was compared by paired χ² test using SPSS 24.0 software, and P<0.05 was considered statistically significant; The detection efficiency of the biphasic L-J medium was evaluated by the culture results of modified L-J medium and the clinical diagnosis as the standard; The difference between the detection time of modified L-J medium and biphasic L-J medium was statistically analyzed by Wilcoxon rank sum test. P<0.05 was considered statistically significant. Results The positive detection rates of modified L-J medium and biphasic L-J medium were 19.65% (453/2305) and 22.00% (507/2305), respectively, and the difference was statistically significant (χ²=22.091, P=0.000); Based on the culture results of modified L-J medium, the sensitivity of biphasic L-J medium was 91.39% (414/453), and the specificity was 94.98% (1759/1852), and the agreement rate was 94.27% (2173/2305);Based on clinical diagnosis, the area under the ROC curve of modified L-J medium and biphasic L-J medium were 0.694 (95%CI: 0.672-0.715), 0.707 (95%CI: 0.685-0.728), respectively; The median (quartile)(M (Q₁, Q₃)) of days when modified L-J medium reported a positive result was 28 (21, 34)d, while the M (Q₁, Q₃) of days when biphasic L-J medium reported a positive result was 18 (14, 24)d. The difference was statistically significant (Z=15.114, P=0.000). Conclusion Biphasic L-J medium has a higher positive detection rate than modified L-J medium. This method has a high diagnostic value, and the time for reporting positive results is significantly shortened. It is a reference method that can be used as a clinical diagnosis for tuberculosis, and has certain application value in grassroots laboratories in China.

Key words: Mycobacterium tuberculosis, Culture media, Bacteriological techniques, Diagnosis, Comparative study