【摘要】 背景 膜周围型室间隔缺损（peri-membranous ventricular septal defect,pmVSD）是最常见的先天性心脏病之一。随着介入封堵技术的发展，经导管治疗已成为pmVSD的重要替代方案，但术后心律失常仍是限制其推广的主要并发症，尤其是传导阻滞相关的心律失常。交界性心律失常（junctional arrhythmia,JA）是术后常见的过渡性表现，其在术后心脏传导系统恢复中的意义尚未明确。目的 探讨pmVSD介入封堵术后早期发生JA的特征及其对术后束支传导阻滞持续存在的预测价值。设计单中心巢式病例对照研究。方法 以2011年1月至2022年1月在四川大学华西第二医院儿童心脏中心成功接受室间隔缺损介入封堵手术且术后完成至少1年心电图随访的pmVSD患儿为队列人群，观察术后1年内束支传导阻滞和JA的发生情况及转归。将出现束支传导阻滞的患儿根据术后1年时的心电图结果分为持续性束支传导阻滞组与可恢复性束支传导阻滞组，分析术后JA等暴露因素对束支传导阻滞恢复的影响和预测价值。主要结局指标术后1年束支传导阻滞是否持续存在。结果 2 622例接受室间隔缺损介入封堵手术的pmVSD患儿中，术后1周内出现束支传导阻滞357例，JA 233例，同时合并束支传导阻滞和JA 129例。110例（30.8%）在术后1年时仍存在持续性束支阻滞。合并JA的多束支阻滞（multiple branch bundle blocks,MBBB）和右束支阻滞（right branch bundle block,RBBB）患儿术后1年时束支阻滞持续存在的比例分别低于单纯性MBBB和RBBB患儿，P值分别为0.001和0.025。（1）可恢复性MBBB组（n=31）与持续性MBBB组（n=16）比较，合并JA比例（83.3%vs. 16.7%，χ~2=11.153,P=0.001）和QRS间期[（80.56±14.31）ms vs.(94.88±17.71)ms,t=-4.179,P=0.038）差异有统计学意义，多因素分析显示，JA[OR=0.200(95%CI:0.043～0.932),P=0.040]和QRS间期[OR=1.102(95%CI:0.102～1.190),P=0.014]为独立影响因素。JA预测可恢复性MBBB的ROC曲线下面积为0.747(95%CI:0.590～0.904)，敏感度为68.8%(95%CI:41.3%～81.9%)，特异度为80.6%(95%CI:62.5%～92.5%);QRS间期预测持续性MBBB的ROC曲线下面积为0.793(95%CI:0.631～0.956)，敏感度为75.0%(95%CI:47.6%～92.7%)，最佳截断值为94.44 ms，特异度为93.6%(95%CI:78.6%～99.2%)。（2）可恢复性RBBB组（n=76）与持续性RBBB组（n=45）比较，合并JA比例（22.37%vs.6.67%，χ~2=5.051,P=0.025）、RV5+SV1[(2.38±0.99)mV vs.(2.95±1.30) mV,t=-2.123,P=0.037]和RV1[(0.60±0.32)mV vs.(1.14±0.91)mV,t=-2.395,P=0.028）差异有统计学意义；多因素分析显示，RV1为独立风险因素[OR=70.045(95%CI:12.820～382.705),P<0.001]；基于单因素筛选变量的列线图评分=（1-JA）×3.5+(RV5+SV1-0.5)/4.5×10+RV1/3.5×100，持续性RBBB组的评分为（41.43±16.78）分，显著高于可恢复性RBBB组的（24.15±7.81）分；列线图ROC曲线下面积为0.861(95%CI:0.778～0.944)，最佳截断值为36.28，敏感度为80.0%(95%CI:65.4%～90.4%)，特异度为93.4%(95%CI:85.3%～97.8%)。结论 pmVSD介入封堵术后早期JA可作为部分束支阻滞类型良性转归的预测信号。更多还原

【Abstract】 Background Perimembranous ventricular septal defect(pmVSD) is one of the most common congenital heart diseases. With advances in device technology, transcatheter device closure has become an important alternative to surgery; however, postoperative arrhythmias,especially conduction block-related events,remain a major limitation to its broader adoption. Junctional arrhythmias(JA) are frequently observed as transient postoperative phenomena, yet their role in the recovery of the cardiac conduction system is not well defined. Objective To characterize early postoperative JA after transcatheter device closure of pmVSD and to evaluate its predictive value for the persistence of bundle branch block(BBB) at one year. Design Single-center nested case-control study. Methods We identified a cohort of children who underwent successful transcatheter device closure of pmVSD at the Pediatric Cardiology Center of West China Second University Hospital, Sichuan University, between January 2011 and January 2022, and who completed at least one year of ECG follow-up. We recorded the occurrence and course of BBB and JA within the first postoperative year. Patients who developed postoperative BBB were stratified by their ECG status at one year into persistent BBB and recovered BBB groups. We analyzed the impact and predictive value of postoperative JA and ECG parameters on BBB recovery. Main outcome measures Persistence of BBB at one year after the procedure. Results Among 2,622 children undergoing transcatheter pmVSD closure, BBB occurred within one week postoperatively in 357 children, JA in 233 children, and both in 129 children. At one year. A total of 110 patients(30.8%) had persistent BBB. The proportions of patients with persistent BBB at one year were lower in those with JA coexisting with multi-branch bundle block(MBBB) or right bundle branch block(RBBB) than in those with isolated MBBB or RBBB(P=0.001 and P=0.025, respectively). MBBB subgroup: Compared with the persistent MBBB group(n=16), the recovered MBBB group(n=31) had a higher incidence of JA(83.3% vs. 16.7%, χ~2=11.153, P=0.001) and a shorter QRS duration(80.56 ± 14.31 ms vs. 94.88 ± 17.71 ms, t=-4.179, P=0.038). In multivariable analysis, JA(OR=0.200, 95% CI: 0.043-0.932; P=0.040) and QRS duration(OR=1.102, 95% CI:1.102-1.190; P=0.014) were independent factors. The ROC AUC for JA predicting recoverable MBBB was 0.747(95% CI:0.590-0.904), with a sensitivity of 68.8%(95% CI:41.3%-81.9%) and a specificity of 80.6%(95% CI:62.5%-92.5%). For QRS duration predicting persistent MBBB, the AUC was 0.793(95% CI:0.631-0.956), with a sensitivity of 75.0%(95% CI:47.6%-92.7%), an optimal cut-off of 94.44 ms, and a specificity of 93.6%(95% CI:78.6%-99.2%). RBBB subgroup: Relative to the persistent RBBB group(n=45), the recovered RBBB group(n=76) more often had JA(22.37% vs. 6.67%, χ~2=5.051, P=0.025) and lower voltage indices: RV5+SV1(2.38 ± 0.99 vs. 2.95 ± 1.30 mV, t=-2.123, P=0.037) and RV1(0.60 ± 0.32 vs. 1.14 ± 0.91 mV, t=-2.395, P=0.028). In multivariable analysis, RV1 was an independent risk factor for persistence(OR=70.045; 95% CI:12.820-382.705; P<0.001). A nomogram based on variables selected by univariable screening yielded the score: score =(1-JA) × 3.5 +(RV5+SV1-0.5)/4.5 × 10 +(RV1)/3.5 × 100. Scores were higher in the persistent RBBB group than in the recovered group(41.43 ± 16.78 vs. 24.15 ± 7.81). The nomogram’s ROC AUC was 0.861(95% CI:0.778-0.944); the optimal cut-off was 36.28, with a sensitivity of 80.0%(95% CI:65.4%-90.4%) and a specificity of 93.4%(95% CI:85.3%-97.8%). Conclusion Early postoperative JA after transcatheter pmVSD closure appears to be a favorable indicator for the benign course of certain BBB phenotypes, supporting its utility in prognostic assessment.更多还原