Xpert MTB/RIF联合T-SPOT.TB检测对脊柱结核的早期诊断价值

The Early Diagnosis Value of Combined Xpert MTB/RIF and T-SPOT.TB Assays for Spinal Tuberculosis

摘要: 目的：通过统计学分析，探究利福平耐药实时荧光定量核酸扩增技术(Xpert MTB/RIF)联合结核感染Ｔ细胞斑点试验(T-SPOT.TB)对脊柱结核早期诊断的临床价值。方法：通过样本获取，回顾性收集2021年12月至2024年9月于青岛市胸科医院有疑似脊柱结核临床表现的患者65例，经确诊分为脊柱结核组(32例)和非脊柱结核组组(33例)。所有研究对象进行脓液的抗酸杆菌涂片检测、病理组织Xpert MTB/RIF和外周血T-SPOT.TB检测比较各单一检测方法和联合检测的诊断阳性率、阴性率和准确率，采用灵敏度、特异度、阳性预测值、阴性预测值和受试者工作特征曲线对各检测方法的诊断价值进行评价。结果：T-SPOT.TB、Xpert MTB/RIF、T-SPOT.TB与Xpert MTB/RIF串并联检测的阳性率分别为93.75%、81.25%、78.13%和96.88%，远高于涂片抗酸染色的阳性率43.75%。并联检测的阳性率高于串联检测及各单一检测，差异有统计学意义(P < 0.05)。T-SPOT.TB与Xpert MTB/RIF并联检测的假阴性率最低，为1.53%；Xpert MTB/RIF检测和T-SPOT.TB与Xpert MTB/RIF串联检测均无假阳性。不同检测方法对脊柱结核及非脊柱结核的诊断正确率均有统计学差异(P > 0.05)。T-SPOT.TB、Xpert MTB/RIF、T-SPOT.TB与Xpert MTB/RIF串并联检测的准确率分别为89.23%、90.77%、89.33%和90.77%，高于涂片抗酸染色的准确率63.08%，差异有统计学意义(P < 0.05)。T-SPOT.TB与Xpert MTB/RIF并联检测的特异度为96.55%，阳性预测值为96.88%，均高于串联检测和单一检测。T-SPOT.TB、Xpert MTB/RIF、T-SPOT.TB与Xpert MTB/RIF串并联检测与涂片抗酸染色相比，曲线下面积有统计学差异(P < 0.05)。并联检测的曲线下面积最大，为0.91 (0.84, 0.98)，Xpert MTB/RIF检测的曲线下面积其次，为0.90 (0.84, 0.97)，且两者差异无统计学意义(P > 0.05)。结论：T-SPOT.TB、Xpert MTB/RIF及其串并联检测用于脊柱结核的诊断相比涂片抗酸染色诊断效能更优，采用两者并联检测的效能最高，但若考虑成本效益，可选择Xpert MTB/RIF单一检测。

Abstract: Objective: To investigate the clinical value of combining Xpert MTB/RIF (rifampicin-resistant real-time fluorescence quantitative nucleic acid amplification detection technology) and T-SPOT.TB (Tuberculosis Infection T-Cell Spot Test) in the early diagnosis of spinal tuberculosis (STB) through statistical analysis. Methods: A retrospective collection was conducted from December 2021 to September 2024 at Qingdao Chest Hospital, involving 65 patients with clinical manifestations suggestive of spinal tuberculosis. After confirmation, the patients were divided into a spinal tuberculosis group (STB, 32 cases) and a non-spinal tuberculosis group (non-STB, 33 cases). All study subjects underwent acid-fast bacillus smear testing of pus, Xpert MTB/RIF testing of pathological tissues, and T-SPOT.TB testing of peripheral blood. The diagnostic positivity rate, negativity rate, and accuracy rate of each single detection method and combined detection were compared. The diagnostic value of each detection method was evaluated using sensitivity, specificity, positive predictive value, negative predictive value, and receiver operating characteristic (ROC) curve analysis. Results: The positive rates of T-SPOT.TB, Xpert MTB/RIF, T-SPOT.TB and Xpert MTB/RIF in series and parallel were 93.75%, 81.25%, 78.13% and 96.88%, respectively, which were significantly higher than that of acid-fast staining of sputum (43.75%). The positive rate of parallel detection was higher than that of series detection and each single detection, with statistically significant differences (P < 0.05). The false negative rate of T-SPOT.TB and Xpert MTB/RIF parallel detection was the lowest, at 1.53%. There were no false positives in either Xpert MTB/RIF detection or T-SPOT.TB and Xpert MTB/RIF tandem detection. There were statistically significant differences in the diagnostic accuracy rates of different detection methods for spinal tuberculosis and non-spinal tuberculosis (P > 0.05). The accuracy rates of T-SPOT.TB, Xpert MTB/RIF, T-SPOT.TB and Xpert MTB/RIF in series and parallel were 89.23%, 90.77%, 89.33% and 90.77%, respectively, which were higher than that of acid-fast staining of sputum (63.08%), with statistically significant differences (P < 0.05). The specificity of parallel detection of T-SPOT.TB and Xpert MTB/RIF was 96.55%, and the positive predictive value was 96.88%, both of which were higher than those of series detection and single detection. The areas under the curve of T-SPOT.TB, Xpert MTB/RIF, T-SPOT.TB and Xpert MTB/RIF in series and parallel were statistically different from that of acid-fast staining of sputum (P < 0.05). The area under the curve of parallel detection was the largest, which was 0.91 (0.84, 0.98), followed by that of Xpert MTB/RIF detection, which was 0.90 (0.84, 0.97), and there was no statistically significant difference between the two (P > 0.05). Conclusion: T-SPOT.TB, Xpert MTB/RIF and their series and parallel detections have better diagnostic efficacy for spinal tuberculosis than acid-fast staining of sputum. The efficacy of parallel detection of the two is the highest, but if cost-effectiveness is considered, Xpert MTB/RIF single detection can be selected.