研究目的     

      评价γ -干扰素释放试验（Interferon Gamma Release Assay， IGRA） 在风湿免疫病人群中诊断成人活动性结核病（Active Tuberculosis， ATB）的诊断效能，并从卫生经济学角度评价 IGRA 在风湿免疫病人群中筛查潜伏性结核感染（ Latent Tuberculosis Infection， LTBI）的价值， 为卫生决策者选择经济有效的筛查方案提供科学依据。

研究方法

第一部分： γ -干扰素释放试验与结核菌素皮肤试验用于诊断风湿免疫病人群活动性结核病准确性的系统综述与 Meta 分析

      在 PubMed、 Embase、 Cochrane Library、 中国知网和万方医学网中检索 TSPOT.TB、 QFT-GIT 和结核菌素释放试验（Tuberculin Skin Test， TST） 在风湿免疫病人群中诊断 ATB 的诊断准确性研究。由 2 名研究员独立完成文献筛查、数据提取、偏倚风险评估，并用 Stata 16.0 软件进行 Meta 分析，采用 GRADE 分级系统对系统综述进行证据质量评价。

第二部分：γ -干扰素释放试验用于风湿免疫病人群结核感染筛查的卫生经济学评价

       根据结核病的自然史与筛查诊断的临床路径，利用 TreeAge Pro2022 软件构建决策树-马尔可夫模型，基于本课题组的前期研究，采用文献的二次研究、检索已出版的专著或已公开发表的文献和专家咨询等方式获取模型所需率值参数、成本参数和贴现率等信息后代入模型。比较各筛查策略的成本、效用和增量成本效果比（Incremental Cost-Effectiveness Ratio， ICER），并进行单因素敏感性分析和概率密度敏感性分析。

研究结果

第一部分： γ -干扰素释放试验与结核菌素皮肤试验用于诊断风湿免疫病人群活动性结核病准确性的系统综述与 Meta 分析

      共纳入 15 篇 T-SPOT.TB、 QFT-GIT 和 TST 在风湿免疫病人群中诊断 ATB 的诊断准确性研究， 其中涉及 T-SPOT.TB 诊断准确性的研究有 12 篇，涉及 QFT-GIT 诊断准确性研究的有 1 篇， 涉及 TST 诊断准确性的研究有 11 篇。Meta 分析结果显示，在风湿免疫病人群中， T-SPOT.TB 诊断 ATB 的合并灵敏度（Sensitivity， SEN）、特 异度（Specificity， SPE）和合并曲线下面积（Area Under Curves， AUC） 分别为 0.82 （95%CI： 0.73， 0.88）、 0.84（95%CI： 0.71， 0.92） 和 0.88（95%CI： 0.83， 0.92）； TST 诊断 ATB 的 SEN、 SPE 和 AUC 分别为 0.42（95%CI： 0.32， 0.53）、 0.84（95%CI： 0.77， 0.89） 和 0.69（95%CI： 0.63， 0.75）。 由于纳入的涉及 QFT-GIT 的研究数量不足， 因此未进行 Meta 分析。 T-SPOT.TB 与 TST 在风湿免疫病人中诊断 ATB 的合并SEN 和 SPE 的证据质量均被评估为低级证据。

第二部分：γ -干扰素释放试验用于风湿免疫病人群结核感染筛查的卫生经济学评价

       成本参数研究发现， TST、 T-SPOT.TB、 QFT-GIT 的平均筛查成本分别为 73.6元/次， 524.2 元/次和 544.5 元/次， ATB 平均诊断费用为 835 元/次， ATB 平均治疗费用和预防性抗结核治疗费用分别为 2738.59 元/次和 1555.15 元/次。 以无筛查方案相比， TST、 T-SPOT.TB、 QFT-GIT、 TST/QFT-GIT 和 TST/T-SPOT.TB 的 ICER 分别为 3973.237 元/质量调整生命年（Quality Adjusted Life Years， QALY）、 8955.593 元/QALY、 9513.658 元/QALY、 10915.929 元/QALY 和 10401.950 元/QALY；以 TST作为对照时， T-SPOT.TB、 QFT-GIT、 TST/QFT-GIT 和 TST/T-SPOT.TB 的 ICER 分别为 31881.153 元/QALY、 40329.112 元/QALY、 44333.829 元/QALY 和 33477.392元/QALY；以上一个优势策略为对照， TST 相对于无筛查策略的 ICER 为 3973.237元/QALY， T-SPOT.TB 相较于 TST 筛查方案的 ICER 为 31881.153 元/QALY， TST/TSPOT.TB 相较于 T-SPOT.TB 筛查方案的 ICER 为 39702.781 元/QALY。 QFT-GIT 筛查方案与 TST/QFT-GIT 筛查方案的成本高于 T-SPOT.TB 方案，但获得的效用值低于 T-SPOT.TB 方案，属于绝对劣势方案。单因素敏感性分析显示对模型影响最大的5 个参数分别为 LTBI 健康效用值、 治愈结核健康效用值、 ATB 的病死率、 非结核分枝杆菌（Mycobacterium Tuberculosis， MTB） 感染者的 LTBI 发病率和 LTBI 的ATB 发病率。概率敏感性分析显示，随着意愿支付阈值的提高，优势筛查策略由 TST单独筛查策略逐渐转换为 T-SPOT.TB 单独筛查策略，最后转换为 TST/T-SPOT.TB联合筛查策略。

研究结论

（1） 在风湿免疫病人群中采用 T-SPOT.TB 诊断 ATB 的灵敏度高于 TST， 特异度差异不显著；

（2）在风湿免疫病人群中， QFT-GIT 诊断 ATB 的诊断准确性研究和报道 IGRA不确定结果发生率的文献较少， 还需要进一步的研究来加以验证；

（3）对风湿免疫病人群采用 TST、 IGRA 或 TST 联合 IGRA 筛查结核感染并开展 LTBI 预防性抗结核治疗具有成本效用价值；

（4）在风湿免疫病人群中开展 LTBI 筛查时可优选 TST 或 T-SPOT.TB 单独筛查策略，经济允许的情况下，可优选 TST/T-SPOT.TB 联合筛查方案。

Objective

     To evaluate the diagnostic efficiency of interferon-gamma release assay (IGRA) for active tuberculosis (ATB) in patients with rheumatic diseases, and to evaluate the value of IGRA for screening latent tuberculosis infection (LTBI) in this population from a health economics perspective, and to provide scientific evidence for health decision-makers to choose screening strategies.

Methods

Part I: The accuracy of interferon-gamma release assays and tuberculin skin tests for the diagnosis of active tuberculosis in rheumatic diseases patients: systematic review and meta-analysis

      We searched the PubMed, Embase, Cochrane Library, CNKI, and WanFang Data databases for studies that assessed the diagnostic accuracy of T-SPOT.TB, QFT-GIT or TST in diagnosing ATB in rheumatic disease patients. Two researchers independently conducted literature screening, data extraction, and bias assessment. Meta-analysis was performed using Stata 16.0 software, and the quality of evidence was evaluated using the GRADE grading system for systematic reviews.

Part II: Health Economics Evaluation of Interferon Gamma Release Assays for Screening Tuberculosis Infection in Patients with Rheumatic Diseases

       A decision tree-Markov model was constructed using TreeAge Pro 2022 software based on the natural history of tuberculosis and the clinical pathway of screening and diagnosis. Model parameters, including rate parameters, cost parameters, and discount rates, were obtained through secondary research of relevant literature, retrieval of published textbooks or papers, and expert consultations, building upon previous research conducted by our team. Cost, utility, and incremental cost-effectiveness ratio (ICER) of various screening strategies were compared, and deterministic sensitivity analysis and probability density sensitivity analysis were conducted.

Results

Part I: The accuracy of interferon-gamma release assays and tuberculin skin tests for the diagnosis of active tuberculosis in rheumatic diseases patients: systematic review and meta-analysis

       A total of 15 studies that evaluated the diagnostic accuracy of TST, T-SPOT.TB, and QFT-GIT in diagnosing ATB in rheumatic disease patients were included in the metaanalysis. Among them, there were 12 studies assessing the diagnostic accuracy of TSPOT.TB, one study on QFT-GIT, and 11 studies on TST. The meta-analysis results showed that in rheumatic disease patients, the combined sensitivity (SEN), specificity (SPE), and area under the curve (AUC) of T-SPOT.TB for diagnosing ATB were 0.82 (95%CI: 0.73, 0.88), 0.84 (95%CI: 0.71, 0.92), and 0.88 (95%CI: 0.88, 0.92), respectively. The SEN, SPE, and AUC of TST for diagnosing ATB were 0.42 (95%CI: 0.32, 0.53), 0.84 (95%CI: 0.77, 0.89), and 0.69 (95%CI: 0.63, 0.75), respectively. Meta-analysis for QFTGIT was not performed due to the insufficient number of studies available. The quality of evidence for the combined SEN and SPE of T-SPOT.TB and TST in diagnosing ATB in rheumatic disease patients was assessed as low-level evidence.

Part II: Health Economics Evaluation of Interferon Gamma Release Assays for Screening Tuberculosis Infection in Patients with Rheumatic Diseases

      Cost parameter study found that the average screening costs for TST, T-SPOT.TB, and QFT-GIT were 73.6 Yuan/time, 524.2 Yuan/time, and 544.5 Yuan/time, respectively. The average diagnosis cost for ATB was 835 Yuan/time, and the average treatment cost and preventive anti-tuberculosis treatment cost for ATB were 2738.59 Yuan/time and 1555.15 Yuan/time, respectively. Compared to the no screening strategy, the ICER for TST, T-SPOT.TB, QFT-GIT, TST/QFT-GIT, and TST/T-SPOT.TB were 3973.237 Yuan/Quality Adjusted Life Years (QALY), 8955.593 Yuan/QALY, 9513.658 Yuan/QALY, 10915.929 Yuan/QALY, and 10401.950 Yuan/QALY, respectively. When TST was used as the reference strategy, the ICER for T-SPOT.TB, QFT-GIT, TST/QFT-GIT, and TST/TSPOT.TB were 31881.153 Yuan/QALY, 40329.112 Yuan/QALY, 44333.829 Yuan/QALY, and 33477.392 Yuan/QALY, respectively. Compared to the previous dominant strategy, the ICER for TST relative to the no screening strategy was 3973.237 Yuan/QALY, while the ICER for T-SPOT.TB relative to the TST strategy was 31881.153 Yuan/QALY, and the ICER for TST/T-SPOT.TB relative to the T-SPOT.TB strategy was 39702.781 Yuan/QALY. The costs of QFT-GIT and TST/QFT-GIT screening strategies were higher than the TSPOT.TB strategy, while the utility values obtained were lower, indicating that they were dominated strategies. deterministic sensitivity analysis identified the top 5 parameters that had the greatest impact on the model were the health utility value of LTBI, the health utility value of cured tuberculosis, mortality rate of ATB after treatment, the incidence rate of LTBI in non-MTB infected patients, and the incidence rate of ATB in LTBI. Probabilistic sensitivity analysis showed that as the willingness-to-pay threshold increased, the dominant screening strategy shifted from TST to T-SPOT.TB and ultimately to TST/TSPOT.TB strategy.

Conclusion

(1) In rheumatic disease patients, the sensitivity of T-SPOT.TB for diagnosing ATB is higher than that of TST, while the specificity difference is not significant.

(2) There are relatively few studies on the diagnostic accuracy of QFT-GIT in diagnosing ATB and the occurrence of indeterminate results with IGRA in rheumatic disease patients, and further research is needed to validate these findings.

(3) Screening for tuberculosis infection using TST, IGRA, or TST combined with IGRA, and providing LTBI preventive anti-tuberculosis treatment to patients with rheumatic diseases is cost-effective.

(4) When screening for LTBI in rheumatic diseases patients, TST or T-SPOT.TB alone can be preferred, and TST/T-SPOT.TB screening can be preferred if economically feasible.