To compare the role of automated real-time nucleic acid amplification technology for rapid and simultaneous detection of mycobacterium tuberculosis and rifampicin resistance (Xpert MTB/RIF) in testing different types of tissue specimens of spinal tuberculosis.

The pus, granulation tissue and caseous necrotic tissue of 223 patients who were diagnosed as having spinal tuberculosis (TB) and received curettage at the Department of Orthopedics of Tianjin Haihe Hospital were collected for bacterial culture and Xpert MTB/RIF assay. Also, the bacterial culture and phenotypic drug-sensitivity test were adopted as the golden standards to calculate the sensitivity and specificity of Xpert MTB/RIF assay.

The positive rate of Xpert MTB/RIF assays on spinal TB patients’ tissue specimens was higher than that of bacterial culture [ (68.61±7.35)% vs. (44.39±6.51)%, P=0.000]. The positive rate of pus was the highest, followed by granulation tissue, and the caseous necrotic tissue was the lowest by both methods with statistically significant difference [bacterial culture: (44.44±7.08)% > (43.86±7.44)% > (30.72±7.31%), P=0.017; Xpert MTB/RIF assay: (73.02±6.33)% > (63.74±7.21)% > (39.87±7.76%), P=0.000]. The positive rate of Xpert MTB/RIF assay was higher than that of bacterial culture in all three types of specimens (pus and granulation tissue: P=0.000, with statistically significant difference; caseous necrotic tissue: P=0.131, without statistically significant difference). The sensitivity of Xpert MTB/RIF assay on mycobacterium tuberculosis was up to (96.97±3.38)% while the sensitivity (97.44%) and specificity (94.74%) of drug-sensitivity test also remained relatively high.

For efficient and accurate diagnosis of TB and drug-resistance test and timely provision of effective treatment, multiple specimens, especially the pus of spinal TB patients, should be collected for Xpert MTB/RIF assays.

Spinal tuberculosis; Molecular diagnostics; Mycobacterium tuberculosis