Abstract:

Objective To investigate the application value of direct solid drug sensitivity test (DST) using sputum specimens from smear-positive pulmonary tuberculosis patients.Methods A total of 389 (from Feb 2010 to Jul 2011) sputum specimens isolated from patients with positive acid-fast tuberculosis were collected. The sputum samples were first used for mycobacterial solid culture and direct DST, and then the solid-culture-positive strains were subjected to conventional proportional susceptibility testing and sodium nitrobenzoate (PNB) medium growth test to differentiate strains. The direct DST was conducted as follows. Based on the specimen traits, 1 to 2 times of 4% NaOH was added to the pre-treatment tube containing sputum specimen, and the screw cap was tightened. The tube was vortexed for 30 seconds until the sputum specimen was liquefied, and then incubated for 15 minutes at room temperature. The pretreated sputum specimens were aspirated with a sterile pipette and evenly inoculated into common acidic Roche medium (control medium for DST), rifampicin (RFP, 40.0μg/ml)-containing acidic Roche medium and isoniazid (INH, 0.2μg/ml)-containing acidic Roche medium (0.1 to 0.15ml per tube). The tubes were incubated at 36℃. The bacterial growth on the culture tube was recorded weekly until the 8th week. On the premise of positive cultures of mycobacteria on the control medium, drug resistance was defined as a percentage of drug resistance of >1% or number of colonies on drug-containing medium >20. The results of direct DST were judged based on the above two drug resistance determination methods. The detection value of DST was analyzed using the proportional method as gold standard.Results Specimens that could not obtain susceptibility test results due to pollution or smear-positive but culture-negative were excluded. As a result, 346 specimens were available for RFP susceptibility test by both the direct method and the proportional method, and 338 specimens were available for INH susceptibility test. When using a resistance percentage of >1% as the definition for drug resistance, the detection rate of RFP resistance by the direct method was 10.4% (36/346), which was lower than that detected by the proportional method (13.9%, 48/346), and the difference was statistically significant (χ ²=24.14, P<0.01). Using the proportional method as the standard, the sensitivity, specificity, and accuracy of the direct solid DST for RFP resistance testing were 75.0% (36/48), 82.9% (247/298), and 81.8% (283/346). No strong consistency between the two methods was shown (Kappa=0.43). For INH resistance, the detection rate by the direct method was 7.1% (24/338), lower than that detected by the proportion method (10.9%, 37/338), and the difference was statistically significant (χ ²=15.29, P<0.01). Taking the proportion method as the standard, the sensitivity, specificity, and accuracy of the direct solid method for INH resistance testing were 64.9% (24/37), 86.0% (259/301), and 83.7% (283/338), respectively. The consistency between the two methods was poor (Kappa=0.39). When using number of colonies on the drug-containing medium more than 20 as the definition for drug resis-tance, the RFP resistance rate tested by the direct method was 10.4% (36/346), which was lower than that of the proportional method. The deference was statistical significance (χ ²=15.87, P<0.01). Using the proportional method as the standard, the sensitivity, specificity, and accuracy of the direct solid DST for RFP resistance testing were 75.0% (36/48), 86.2% (257/298), and 84.7% (293/346), respectively. No strong consistency between the two methods was shown (Kappa=0.49). The INH resistance rate detected by the direct method was 7.1% (24/338), which was lower compared with the proportion method, and the difference was statistically significant (χ ²=9.38, P<0.01). Using the proportional method as the standard, the sensitivity, specificity, and accuracy of the direct solid DST for INH resistance testing were 64.9% (24/37), 88.7% (267/301), and 86.1% (291/338). No strong consistency between the two methods was shown (Kappa=0.43). Conclusion The direct solid DST shows quite different drug resistance results compared with the traditional proportional DST. The clinical diagnosis of drug resistant tuberculosis by direct solid method is not ideal.

Key words: Mycobacterium tuberculosis, Drug resistance, bacterial, Microbial sensitivity tests, Compara-tive study