Adjusted T-SPOT.TB criteria can increase the specificity of diagnosis when differentiating spinal tuberculosis 1 from other spinal infections

15 Background: The ability of T-SPOT.TB to differentiate Mycobacterium tuberculosis infection of the spine from other infections is little known. This study quantified the 16 efficiency, sensitivity, and specificity of the T-SPOT.TB assay to distinguish between spinal tuberculosis (STB) caused by M. tuberculosis and other infections of the spine and 17 evaluated whether diagnostic performance was improved by adjusting the T-SPOT.TB assay criteria. 18 Methods: From January 2010 to May 2020, 147 patients with spinal infections were recruited. Peripheral blood mononuclear cells were collected, and the number of spot- 19 forming cells was observed. Patients’ white blood cell (WBC) counts, C-reactive protein (CRP), erythrocyte sedimentation rate (ESR), procalcitonin (PCT), and TB antibodies 20 were recorded. Specimen/tissue bacteriological culture was the reference standard for sensitivity and specificity. 21 Results: There were 77 (52.4%) participants with confirmed TB and 70 (47.6%) with other infections. The groups were comparable in T-SPOT.TB assay results, age, sex, 22 lesions in the segments, WBC count, CRP, procalcitonin, ESR, and TB antibodies. The sensitivity and specificity of the T-SPOT.TB assay for identifying STB was 88.3% and 23 40.0%, respectively. On the basis of Relative operating characteristic curve (ROC) analysis and the Youden index, when we adjusted the T-SPOT.TB assay’s diagnostic criteria, 24 ESAT-6>12 or CFP-10>19,the sensitivity and specificity of the T-SPOT.TB assay for identifying STB was 83.1% and 64.3%, respectively. 25 Conclusion: The T-SPOT.TB assay has great sensitivity to distinguish STB from other spinal infections; however, the specificity is extremely low. Specificity can be 26 significantly improved while sensitivity is guaranteed by adjusting the diagnostic criteria. 27


29
Tuberculosis (TB) is a communicable disease caused by infection with the bacterium Mycobacterium tuberculosis, and TB is a major cause of ill health worldwide. Most 30 TB cases in 2018 occurred in Southeast Asia (44%), including China (9%) [1]. Extrapulmonary TB accounts for 10% of cases, of which half involve the musculoskeletal system.

31
The spine is the most common musculoskeletal site involved in extrapulmonary TB (1 to 2% of cases) [2]. Early treatment can reduce the incidence of physical disability and 32 injury in STB, but diagnosis is extremely challenging. There are scarce bacteria in the articular effusion of lesion sites, and specimens are not easy to obtain, which reduces the 33 positive rate of puncture fluid or joint surgical specimens [3].

34
The T-SPOT.TB is an assay of interferon (IFN)-γ release from M. tuberculosis-specific effector T-cells stimulated by the Mycobacterium-specific antigens ESAT-6 (early-35 secreted antigenic target 6) and CFP-10 (culture filtrate protein 10). Thus, the antigens ESAT-6 and CFP-10 have been successfully utilized to determine the presence of M. 36 5 examination suggested spinal infection; no severe underlying disease or human immunodeficiency virus infection was present; and the patient was followed up for at least three 54 months.

55
The confirmation of STB was based on the identification of M. tuberculosis in tissues or specimens by culture or by PCR in addition to clinical, radiographic, or other 56 supporting evidence and medical history suggestive of TB.

57
Probable (rather than confirmed) STB was considered when the results of M. tuberculosis culture were unclear, but pathological examination indicated TB infection, and 58 clinical, radiographic, and other supporting tests and medical history suggested TB.

59
The diagnosis of other infection was made when the results of culture or PCR in tissues or specimens indicated infection other than M. tuberculosis, anti-TB treatment 60 prior to surgery was ineffective, and the pulmonary and bacteriological culture of sputum was negative.

61
Finally, 147 patients were apportioned to three groups as follows. The confirmed STB group comprised 35 patients for whom the culture or PCR results indicated M.

85
Among the diagnostic groups (confirmed STB, probable STB, and other infection), CRP and ESR were statistically significant (P=0.03, P=0.01, respectively), but there 86 were no significant differences in age, sex, lesions in the segments, WBC count, PCT ( Table 2 and Table 3). Based on the chi-squared test, TB antibodies was significant 87 differences (P = 0.003, Table 3), but no differences were shown in CRP, ESR, PCT, TB antibodies, WBC count, sex by logistic regression (Table 4).

113
In the present study, the 147 patients were apportioned to 3 groups according to the results of culture or PCR: confirmed STB, probable STB, and other infection, with 35, 114 42, and 70 cases, respectively. In the diagnosis of spinal infection, the T-SPOT.TB test results were compared with traditional serological test results (e.g., ESR, CRP). The 115 traditional serological parameters include WBC count, ESR, CRP, PCT, and TB antibodies. However, WBC count, ESR, CRP, PCT and TB antibodies had no significant 116 specificity for the differential diagnosis of spinal infections, while the results of the T-SPOT.TB tests were statistically significant (P < 0.01) for the differential diagnosis of nevertheless, the specificity was 40%, which was much lower than that in other reports. A false-positive rate of 60% makes it difficult to use this assay to determine the cause   (Table 4). The result shows that specificity increased to more than 60% with almost constant sensitivity guaranteed by adjusting the diagnostic criteria of the T-SPOT.TB 126 assay. Some studies have suggested that the diagnostic efficiency of ESAT-6 and CFP-10 may not be exactly the same. CFP-10 was responsible for significantly more positive