全身化疗联合支气管镜下局部灌注异烟肼治疗支气管结核的疗效与安全性研究

A Study on the Efficacy and Safety of Systemic Chemotherapy Combined with Bronchoscopic Local Instillation of Isoniazid in the Treatment of Bronchial Tuberculosis

摘要: 目的：评估全身化疗联合支气管镜下局部灌注异烟肼(INH)治疗支气管结核的临床疗效与安全性。方法：采用回顾性队列研究设计，纳入30例支气管结核患者。观察组(n = 17)在FDC-2HREZ/4HR标准化疗基础上加用支气管镜下局部灌注INH 0.2 g/次，每周1次，连续1~2个月；对照组(n = 13)仅接受标准化疗。主要终点为总有效率；症状缓解、痰菌转阴因缺失数据比例高，降级为探索性终点。缺失数据采用完整病例分析，并以“未记录归为失败”、“未记录排除”、“多重插补”进行敏感性分析。基线不均衡采用多因素logistic回归及倾向性评分校正。结果：观察组总有效率为76.5%，对照组为53.8%，差异无统计学意义(P = 0.362; OR = 2.79, 95% CI 0.58~13.32; RR = 1.42, 95% CI 0.81~2.51)。探索性终点中，观察组症状缓解率为66.7%，对照组为27.8% (P = 0.045; OR = 5.20, 95% CI 1.25~21.56; RR = 2.40, 95% CI 1.06~5.41)，但对照组72.2%未记录，结果受缺失数据限制，仅供参考。观察组基线痰涂片阳性比例显著高于对照组(66.7% vs 22.2%, P = 0.025; OR = 7.00, 95% CI 1.59~30.82; RR = 3.00, 95% CI 1.19~7.56)。探索性分析显示，观察组随访痰菌转阴率高于对照组(55.6% vs 5.6%, P = 0.005; OR = 21.25, 95% CI 2.30~196.10; RR = 10.00, 95% CI 1.42~70.20)，但对照组88.9%未记录，结论需谨慎。安全性方面，观察组气道狭窄/肺不张发生率为14.3%，对照组为0% (P = 0.389)；任意不良事件发生率分别为27.8%和5.6% (P = 0.177; OR = 6.54, 95% CI 0.68~62.99; RR = 5.00, 95% CI 0.65~38.59)，差异均无统计学意义。结论：支气管镜下局部灌注INH对支气管结核总有效率未见显著提高；症状缓解率和痰菌转阴率的探索性分析显示潜在获益，但受样本量小、基线不均衡及缺失数据限制，结论需谨慎解读。未来应开展多中心、大样本随机对照试验，并结合药代动力学研究进一步验证。

Abstract: Objective: To evaluate the clinical efficacy and safety of systemic chemotherapy combined with bronchoscopic local instillation of isoniazid (INH) in treating bronchial tuberculosis. Methods: A retrospective cohort study design was used, enrolling 30 patients with bronchial tuberculosis. The observation group (n = 17) received standard FDC-2HREZ/4HR chemotherapy plus bronchoscopic local instillation of INH at 0.2 g per session, once weekly for 1~2 months; the control group (n = 13) received standard chemotherapy alone. The primary endpoint was overall response rate. Due to high missing data rates, symptom resolution and sputum conversion were downgraded as exploratory endpoints. Missing data were analyzed using complete-case analysis, with sensitivity analyses performed under assumptions of “unrecorded as failure”, “unrecorded excluded”, and multiple imputation. Baseline imbalances were adjusted using multivariable logistic regression and propensity score adjustment. Results: The overall response rate was 76.5% in the observation group versus 53.8% in the control group, with no statistically significant difference (P = 0.362; OR = 2.79, 95% CI 0.58~13.32; RR = 1.42, 95% CI 0.81~2.51). For exploratory endpoints, symptom resolution rates were 66.7% in the observation group versus 27.8% in the control group (P = 0.045; OR = 5.20, 95% CI 1.25~21.56; RR = 2.40, 95% CI 1.06~5.41), although 72.2% of the control group had unrecorded data, limiting interpretation. The baseline proportion of positive sputum smears was significantly higher in the observation group than in the control group (66.7% vs 22.2%, P = 0.025; OR = 7.00, 95% CI 1.59~30.82; RR = 3.00, 95% CI 1.19~7.56). Exploratory analysis showed a higher sputum conversion rate during follow-up in the observation group compared to the control group (55.6% vs 5.6%, P = 0.005; OR = 21.25, 95% CI 2.30~196.10; RR = 10.00, 95% CI 1.42~70.20), but 88.9% of the control group had unrecorded data, so conclusions should be interpreted cautiously. Regarding safety, the incidence of airway stenosis/atelectasis was 14.3% in the observation group versus 0% in the control group (P = 0.389); the rates of any adverse events were 27.8% and 5.6%, respectively (P = 0.177; OR = 6.54, 95% CI 0.68~62.99; RR = 5.00, 95% CI 0.65~38.59), with no significant differences. Conclusion: Bronchoscopic local instillation of INH did not significantly improve the overall response rate in bronchial tuberculosis. Exploratory analyses suggested potential benefits in symptom resolution and sputum conversion, but these findings are limited by small sample size, baseline imbalance, and missing data, requiring cautious interpretation. Future multicenter, large-scale randomized controlled trials combined with pharmacokinetic studies are warranted to further validate these results.