Abstract:

Objective: To evaluate the diagnostic value of GeneXpert MTB/RIF Ultra (Xpert Ultra) in testing surgical tissue and pus specimens for tuberculosis. Methods: This prospective research collected surgical tissue and pus specimens from clinically presumptive tuberculosis patients in Xi’an Chest Hospital from January to September 2021. These samples includes 49 surgical tissues and 36 pus from the same lesion of the patients. Xpert Ultra, GeneXpert MTB/RIF (Xpert), BACTEC MGIT 960 liquid culture (MGIT 960 culture), DNA real-time fluorescent quantitative nucleic acid amplification detection (FQ-PCR), RNA real-time fluorescence constant temperature amplification (SAT-RNA), DNA real-time fluorescence constant temperature amplification method (CTM-DNA) were performed on these specimens. The clinical diagnosis result was used as the reference standard to evaluate the diagnostic value of surgical tissues of the above-mentioned various methods; differences of positive rates of the tissues and pus specimens from the same yielded of tuberculosis patients (each 28) by each method were analyzed. Results: The sensitivity of Xpert Ultra (75.0%, 24/32) on tissue samples was significantly higher than that of Xpert (65.6%, 21/32), FQ-PCR (53.1%, 17/32), SAT-RNA (25.0%, 8/24), CTM-DNA (6.3%, 2/30) and MGIT 960 culture (37.5%, 12/32). The differences between Xpert Ultra and other methods were statistically significant (χ²=32.919, P<0.001; χ ²=21.866, P<0.001; χ ²=5.418, P=0.020; χ ²=15.228, P<0.001; χ ²=11.574, P=0.001). Consistency analysis results showed: Xpert Ultra was highly consistent with clinical diagnosis results (Kappa: 0.675), Xpert and FQ-PCR were moderately consistent with clinical results (Kappa: 0.570, 0.444), MGIT 960 culture was mildly consistent (Kappa: 0.294), SAT-RNA and CTM-DNA were very poorly consistent (Kappa: 0.188, 0.044).The areas under the receiver operating characteristic curve were 0.875 for Xpert Ultra, 0.828 for Xpert MTB/RIF, 0.766 for FQ-PCR, 0.688 for culture method, 0.625 for SAT-RNA and 0.531 for CTM-DNA. The positive rates of Xpert Ultra in detecting pus and tissue samples from the same lesion of patients were 78.6% (22/28) and 71.4% (20/28),and the difference was not statistically significant (χ ²=0.382, P=0.537). Conclusion: Xpert Ultra on surgical tissue and pus specimens had a high rapid diagnostic value for tuberculosis, and there was no need to detect different types of specimen from the same lesion of patients at the same time, but considering the availability and quality of sampling, tissue specimens should be preferred.

Key words: Molecular diagnostic techniques, Biopsy, Laboratory techniques and procedures, Evaluation studies