摘要

第一部分 左心室质量指数预测肥厚型梗阻性心肌病手术患者的临床结局：心脏核磁共振研究

目的：左心室质量指数（Left ventricular mass index,LVMI）是反应左心室肥厚程度的重要指标。既往研究显示，在肥厚型心肌病（Hypertrophic cardiomyopathy,HCM）患者中，与左心室壁最大厚度（Maximal wall thickness of left ventricle, MWT）相比，LVMI在预测HCM患者不良预后上敏感度更高。然而，在肥厚型梗阻性心肌病（obstructive HCM,oHCM）手术患者中，LVMI对患者中远期预后的预测价值尚不明确。本研究旨在探讨在oHCM手术患者中基线LVMI与临床预后的关系。

方法：回顾性连续纳入2011年6月至2016年7月期间在本中心接受室间隔心肌切除术的oHCM患者490例。所有患者术前均接受心脏核磁共振（Cardiac magnetic resonance, CMR）检查，89例患者在术后1年完成了CMR复查。本研究的主要终点为全因死亡（包括心脏移植，心血管疾病相关性死亡以及心源性猝死），次要终点为心力衰竭、新发或复发性房颤以及复合终点事件（包含全因死亡、心力衰竭、缺血性卒中，永久性起搏器植入、新发或复发性房颤）。通过最大选择秩统计量法确定LVMI对全因死亡事件的截断值，将患者分为高截断值组（82.35g/m²）和低截断值组（<82.35 g/m²）以比较两组患者的基线差异。

结果：在本研究中，患者的平均年龄为47.35±11.63岁，其中195例（39.8%）为女性。与低截断值组相比，高截断值组患者年龄更小、女性比例更低、左房内径更大、MWT，钆延迟强化阳性（LGE+）比例及钆延迟后增强容积百分比（LGE%）更高。在中位随访82.4个月（四分位间距：64.0至101.8个月）期间，共发生21例全因死亡、44例心力衰竭、40例新发或复发性房颤及124例复合终点事件（包括21例死亡、44例心力衰竭、40例新发或复发性房颤、8例缺血性卒中及11例起搏器植入）。采用最大选择秩统计量法确定LVMI对全因死亡的最佳截断值为82.35 g/m²。Kaplan-Meier曲线显示，LVMI≥82.35 g/m²组的全因死亡累积发生率显著高于LVMI<82.35 g/m²组（p=0.004）。Cox多因素分析表明，左房内径及基线LVMI[LVMI连续值：HR（Hazard ratio，风险比） 1.012，95%CI（confident interval, 置信区间） 1.005-1.019，p<0.001；LVMI截断值：HR 9.92，95%CI 1.317-74.75，p=0.026）]为全因死亡的独立预测因子。在进一步校正MWT和LGE%后，LVMI仍与全因死亡独立相关（HR 1.009，95%CI 1.0001-1.018，p=0.047）。同时，LVMI在预测术后1年、3年和5年生存率的曲线下面积（AUC）分别为0.815（敏感性：100%，特异性：68.0%，截断值108.51 g/m²）、0.689（敏感性：100%，特异性：44.7%，截断值84.32 g/m²）和0.679（敏感性：100%，特异性：41.8%，截断值82.74 g/m²）。同样地，采用最大选择秩统计量法确定LVMI对各次要终点的截断值。Kaplan-Meier曲线显示，高LVMI组的新发或复发性房颤（LVMI>70.08 g/m²）、心力衰竭（LVMI>77.01 g/m²）及复合终点（LVMI>76.77 g/m²）累积发生率均显著高于低LVMI组。 Cox多因素分析表明，LVMI截断值与新发或复发性房颤（HR 5.937，95%CI 1.381-25.521，p=0.017）及复合终点（HR 1.639，95%CI 1.069-2.512，p=0.023）独立相关。

结论：在接受室间隔心肌切除术的oHCM患者中，基线LVMI是独立于MWT以及LGE%外远期全因死亡事件的重要预测因子，对1-,3-,5年生存率预测的敏感度为100%，同时也是新发或复发性房颤及复合终点事件的独立危险因素。

冠状动脉肌桥增加肥厚型梗阻性心肌病患者术前房颤风险：一项单中心大样本队列研究

目的：冠状动脉肌桥（Myocardial Bridging, MB）是一种先天性发育异常，其特征为心外膜冠状动脉在心肌内走行，导致心肌内节段收缩期受压及血流动力学紊乱。既往研究表明，MB与多种形式的心律失常有关。然而，在肥厚型梗阻性心肌病（Obstructive Hypertrophic Cardiomyopathy, oHCM）患者中，MB与房颤（Atrial Fibrillation, AF）的相关性仍存争议。此外，室间隔心肌切除术（Septal myecotomy thrapy,SRT）是缓解oHCM相关梗阻的金标准，可同期处理并存的心脏瓣膜病变或冠状动脉疾病。本研究旨在探讨oHCM患者中MB与AF的相关性，并评估SRT同期处理MB的安全性及对预后的影响。

方法：连续纳入2015年1月至2019年12月期间在阜外医院接受SRT的oHCM患者968例，其中144例合并MB。所有患者均在术前接受冠状动脉造影。当患者合并MB时，记录MB的位置、长度及收缩期最大压缩程度。通过单变量及逐步多变量Logistic回归分析识别AF相关因素。在MB亚组中，采用最小绝对收缩与选择算子（Least Absolute Shrinkage and Selection Operator, LASSO）回归筛选预测变量并纳入最终多变量模型，明确MB位置、长度、压缩程度等与AF的关系。使用Kaplan-Meier曲线拟合对数秩检验比较组间无事件生存率。

结果：与非MB组相比，MB组患者年龄更小（平均52.3岁 vs. 58.6岁，p<0.001），AF发生率更高（26.4% vs. 11.5%, p<0.001）。多变量Logistic回归显示MB是AF的独立危险因子[OR（比值比）3.73，95%CI[置信区间] 2.32–5.98，p<0.001]。亚组分析表明，左前降支近段MB（OR 4.22，95%CI 1.40–12.68，p=0.010）、收缩期压缩程度（每增加1%：OR 1.04，95%CI 1.04–1.08，p=0.013）及MB长度（每增加1毫米：OR 1.06，95%CI 1.004–1.11，p=0.034）与AF风险显著相关。在144例MB患者中，100例接受外科干预（心肌桥松解术或冠状动脉旁路移植术）。与非MB组相比，MB组患者的灌注时间（108.5±40.7分钟 vs. 96.7±37.3分钟，p<0.001）及主动脉阻断时间（75.2±28.3分钟 vs. 64.7±26.9分钟，p<0.001）更长。非MB组术后机械通气时长较MB组的更长（17小时[13-19小时] vs. 16小时[12-18小时]，p=0.068。非MB组中有2例患者术中死亡，死因分别为室间隔破裂及低心排血量综合征。MB组1例患者于出院10天后发生心源性猝死。中位随访3.25年（四分位距：2.03–4.61）期间，全队列中观察到18例死亡、36例新发AF及118例复合终点事件。Kaplan-Meier分析显示，MB组与非MB组在全因死亡率（p=0.86）、新发AF（p=0.91）及复合终点（p=0.56）无显著差异。

结论：MB及其解剖学特征（位置、长度及收缩期压缩程度）与oHCM患者AF风险显著相关。SRT同期处理MB不显著增加手术风险，建议在对MB进行评估后可考虑实施。

第三部分 室上性异位活动预测肥厚型梗阻性心肌病患者术后房颤、新发房颤及临床结局

目的：室上性异位活动（Supraventricular Ectopic Activity, SVEA）是房颤（Atrial Fibrillation, AF）的潜在触发因素之一，常与不良预后有关。在肥厚型梗阻性心肌病（Obstructive Hypertrophic Cardiomyopathy, oHCM）患者中，AF是常见的心律失常，术后房颤（Postoperative Atrial Fibrillation, POAF）以及新发 AF是oHCM手术患者中远期心力衰竭（Heart Failure,HF）、缺血性卒中及心血管死亡的重要危险因素。因此，明确SVEA的类型和频率对POAF、新发AF以及临床心血管不良事件的预测作用，对提高oHCM患者围手术期管理能力及中远期心血管不良事件的预防具有重要的临床意义。本研究旨在纳入术前24小时动态心电图（Holter）参数以探究SVEA的频率和类型对oHCM患者的POAF、新发 AF和临床结局的预测价值。

方法：本研究采用回顾性队列研究设计，连续纳入961例接受外科治疗的oHCM患者。所有患者术前均完成24小时Holter监测。本研究的主要事件包括POAF、新发 AF以及复合终点事件（包括全因死亡事件、新发AF、HF）。在研究中，为探究SVEA频率和类型对结局的预测，我们将患者根据房性早搏（Premature atrial contraction, PACs）负荷三分位数（Tertile）进行分组，比较不同程度PACs负荷下患者的基线特征。采用多因素logistic回归分析SVEA与POAF的关联，应用Cox比例风险模型评估SVEA对远期预后的预测价值。通过受试者工作特征曲线下面积（area under the curve, AUC）评估预测效能，并采用净重分类改善指数（net reclassification improvement, NRI）和综合判别改善指数（integrated discrimination improvement, IDI）量化预测模型的增量价值。

结果：在本研究中POAF发生率为20.7%，且与PACs负荷呈正相关（Tertile1、Tertile2、Tertile3组POAF的发生率分别为14.7%、18.7%、29%，趋势性p<0.001）。多因素分析显示，室上性心动过速（SVT）是POAF的独立预测因子[OR （比值比）2.15, 95%CI（置信区间） 1.32-3.49）]，包含SVT的预测模型AUC为0.710（95%CI 0.670-0.750）。中位随访2.9年期间，观察到全因死亡12例、新发AF 60例及复合终点事件139例。 Spearman相关分析表明，PACs负荷与新发AF（r=0.32, P<0.001）及复合终点事件（r=0.28, P=0.002）呈正相关。Kaplan-Meier生存分析显示，PACs>200次/天患者的new-onset AF及复合终点事件的累积发生率显著高于PACs<200次/天的患者。Cox多因素分析表明，PACs >200次/天是新发AF （HR（风险比）3.13, 95%CI 1.74-5.62, P<0.001）及复合终点事件（HR 2.00, 95%CI 1.30-3.06, P=0.002）的独立危险因素。将PACs>200次/天纳入多变量模型后，预测新发AF [NRI=0.264（P=0.008）、IDI=0.033（P=0.010）]和复合终点事件[NRI=0.233（P=0.014）、IDI=0.014（P=0.024）]的效能显著提升。

结论：在本研究中，oHCM患者POAF的发生率为20.7%，且POAF的发生率随着PACs负荷的增加而升高。此外，PACs负荷与新发AF 和复合终点事件的发生率呈正相关。在PACs >200次/天的患者中，新发AF和复合终点事件的风险显著增加。将PACs >200次/天纳入多变量模型后，模型对新发AF和复合终点事件的预测效能显著提高，表明PACs >200次/天可能是预测oHCM患者新发AF 和不良预后的最佳阈值。

Abstract

Part 1 Prognostic value of left ventricular mass index in obstructive hypertrophic cardiomyopathy undergoing septal myectomy:insights from a Cardiac Magnetic Resonance study

Objectives: The left ventricular mass index (LVMI) is a critical indicator reflecting the degree of left ventricular hypertrophy. Previous studies have shown that compared to maximal left ventricular wall thickness(MWT) , LVMI serves as a more sensitive predictor of adverse outcomes in hypertrophic cardiomyopathy(HCM) patients. However, the prognostic value of LVMI for mid- to long-term outcomes in patients with obstructive HCM (oHCM) following myectomy remains unclear. This study aimed to investigate the relationship between LVMI and clinical outcomes in oHCM patients after surgery.

Methods: We retrospectively enrolled 490 oHCM patients who underwent myectomy at Fuwai Hospital between June 2011 and July 2016. All patients underwent preoperative cardiac magnetic resonance (CMR) imaging, with 89 patients completing CMR re-examination one year postoperatively. The primary endpoint of this study was all-cause mortality (including heart transplantation, cardiovascular disease-related death, and sudden cardiac death), and the secondary endpoints were heart failure, new or recurrent atrial fibrillation (AF), and composite endpoint events (including all-cause mortality, heart failure, ischemic stroke, permanent pacemaker implantation, new or recurrent AF). The Maximally Selected Rank Statistics method was used to determine the cutoff value of LVMI for all-cause mortality, categorizing patients into high cutoff (≥82.35 g/m²) and low cutoff (<82.35 g/m²) groups for baseline comparisons.

Results: The mean age of the study population was 47.35 ± 11.63 years, with 195 patients (39.8%) being female. Compared to patients in the low cutoff group,those in the high cutoff group were much younger, less likely to be female, had a greater left atrial diameter, higher MWT, higher proportion of positive late gadolinium enhancement (LGE+), and higher percentage of LGE%. During an mean follow-up of 82.4 months (interquartile range: 64.0 to 101.8 months), there were 21 all-cause deaths, 44 heart failures, and 124 composite endpoint events (including 21 deaths, 44 heart failures, 40 new or recurrent AF, 8 ischemic strokes, and 11 pacemaker implantation).  The maximum selection rank statistic method determined the optimal cutoff value of LVMI for all-cause mortality to be 82.35 g/m². Kaplan-Meier curves showed that the cumulative incidence of all-cause mortality in the LVMI≥82.35 g/m² group was significantly higher than in the LVMI<82.35 g/m² group (p=0.004). Cox multivariable analysis indicated that left atrial diameter and LVMI [LVMI continuous value: HR (Hazard ratio) 1.012, 95%CI (confidence interval) 1.005-1.019, p<0.001; LVMI cutoff value: HR 9.92, 95%CI 1.317-74.75, p=0.026) were independent predictors of all-cause mortality. After further adjustment for MWT and LGE%, LVMI remained independently associated with all-cause mortality (HR 1.009, 95%CI 1.0001-1.018, p=0.047). Additionally, the area under the curve (AUC) for LVMI in predicting 1-year, 3-year, and 5-year survival rates were 0.815 (sensitivity: 100%, specificity: 68.0%), 0.689 (sensitivity: 100%, specificity: 44.7%), and 0.679 (sensitivity: 100%, specificity: 41.8%), respectively. Furthermore, Kaplan-Meier curves showed that the cumulative incidence of new or recurrent AF (LVMI>70.08 g/m²), heart failure (LVMI>77.01 g/m²), and composite endpoints (LVMI>76.77 g/m²) were significantly higher in the high LVMI group compared to their counterparts. Cox multivariable analysis showed that LVMI cutoff values were independently associated with new or recurrent AF (HR 5.937, 95%CI 1.381-25.521, p=0.017) and composite endpoints (HR 1.639, 95%CI 1.069-2.512, p=0.023).

Conclusion: In oHCM patients undergoing myectomy, baseline LVMI is an important predictor of all-cause mortality independent of MWT and LGE%, with a sensitivity of 100% for predicting 1-, 3-, and 5-year survival rates, and is also an independent risk factor for new or recurrent AF and composite endpoints.

Part 2 Myocardial bridging is associated with increased risk of preoperative atrial fibrillation in adult patients with obstructive hypertrophic cardiomyopathy：A Large single-center cohort study

Aims: Myocardial bridging (MB), characterized by an epicardial coronary artery traversing intramurally through the myocardium, induces systolic compression of the tunneled segment and hemodynamic disturbances.Previous studies have shown that MB is associated with various forms of arrhythmias.  However, the association between MB and atrial fibrillation (AF) in obstructive hypertrophic cardiomyopathy (oHCM) remains controversial. Moreover, septal myectomy remains the gold standard for alleviating oHCM-related obstruction, with concurrent surgical management of coexisting valvular or coronary pathology when clinically indicated. This study aimed to investigate the relationship between MB and AF in oHCM and evaluate the safety of concurrent MB intervention during septal myectomy.

Methods:We enrolled 968 consecutive oHCM patients undergoing septal myectomy, including 144 with angiographically confirmed MB. All patients underwent coronary angiography to assess MB characteristics, including location, length, and maximal systolic compression degree. Univariable and stepwise multivariable logistic regression analyses were performed to identify factors associated with AF. In the MB subgroup analysis, least absolute shrinkage and selection operator (LASSO) regression was employed to select predictive variables for inclusion in the final multivariable model. Kaplan-Meier curves with log-rank testing were utilized to compare event-free survival between groups.

Results: MB patients were significantly younger (mean age 52.3 vs. 58.6 years, p<0.001) and exhibited higher AF prevalence (26.4% vs. 11.5%, p<0.001) than non-MB patients. Multivariable logistic regression showed that MB was an independent risk factor for AF [OR (odds ratio) 3.73, 95%CI (confidence interval) 2.32–5.98, p<0.001]. Subgroup analysis indicated that proximal left anterior descending MB (OR 4.22, 95%CI 1.40–12.68, p=0.010), systolic compression degree (per 1% increase: OR 1.04, 95%CI 1.04–1.08, p=0.013), and MB length (per 1 mm increase: OR 1.06, 95%CI 1.004–1.11, p=0.034) were significantly associated with AF risk. Among the 144 MB patients, 100 underwent surgical intervention (unroofing or coronary artery bypass grafting). Compared to the non-MB group, the MB group had longer perfusion time (108.5±40.7 minutes vs. 96.7±37.3 minutes, p<0.001) and aortic cross-clamp time (75.2±28.3 minutes vs. 64.7±26.9 minutes, p<0.001). The non-MB group had a longer postoperative mechanical ventilation duration than the MB group (17 hours [13-19 hours] vs. 16 hours [12-18 hours], p=0.068). There were 2 operative deaths in the non-MB group due to septal rupture and low cardiac output syndrome, respectively. One MB patient experienced sudden cardiac death 10 days after discharge. During a median follow-up of 3.25 years (interquartile range: 2.03–4.61), 18 deaths, 36 new-onset AF cases, and 118 composite endpoint events were observed in the entire cohort. Kaplan-Meier analysis showed no significant differences between the MB and non-MB groups in all-cause mortality (p=0.86), new-onset AF (p=0.91), or composite endpoints (p=0.56).

Conclusion: MB and its anatomical characteristics (location, length, and systolic compression degree) are significantly associated with AF risk in oHCM patients. Concurrent management of MB during septal myectomy does not significantly increase surgical risk, and it is recommended to consider intervention after evaluating MB.

Part 3  Supraventricular ectopic activity predicts postoperative atrial fibrillation, new-onset atrial fibrillation, and worse survival in patients with obstructive hypertrophic cardiomyopathy

Objectives: Supraventricular ectopic activity (SVEA) is a known trigger for atrial fibrillation (AF) and is linked to adverse clinical outcomes. In patients with obstructive hypertrophic cardiomyopathy (oHCM), AF is a prevalent arrhythmia, with both postoperative atrial fibrillation (POAF) and new-onset AF serving as critical risk factors for long-term complications such as heart failure, stroke, and cardiovascular mortality following myectomy. Understanding the predictive role of SVEA type and frequency in POAF, new-onset AF, and adverse cardiovascular events holds significant clinical value for enhancing perioperative management and mitigating long-term risks in oHCM patients. This study leverages preoperative 24-hour Holter monitoring to assess the predictive utility of SVEA frequency and type for POAF, new-onset AF, and adverse outcomes in this patient population.

Methods: This retrospective study included 961 consecutive oHCM patients who underwent surgical intervention at our institution. Preoperative 24-hour Holter monitoring was performed for all participants. The primary endpoints comprised POAF, new-onset AF, and a composite of all-cause mortality, new-onset AF, and heart failure. Patients were stratified into tertiles based on premature atrial contraction (PACs) burden to assess the predictive value of SVEA frequency and type on clinical outcomes, with baseline characteristics compared across these groups. Multivariate logistic regression was used to analyze the association between SVEA and POAF, and Cox proportional hazards models were applied to evaluate the predictive value of SVEA for long-term outcomes. The predictive performance was assessed using the area under the receiver operating characteristic curve (AUC), and the net reclassification improvement (NRI) and integrated discrimination improvement (IDI) were used to quantify the incremental value of the predictive models.

Results: The incidence of POAF in this study was 20.7%, and it was positively correlated with PACs burden (POAF incidence in Tertile 1, Tertile 2, and Tertile 3 groups was 14.7%, 18.7%, and 29%, respectively, with p<0.001). Multivariate analysis showed that Supraventricular tachycardia (SVT) was an independent predictor of POAF [OR (odds ratio) 2.15, 95%CI (confidence interval) 1.32-3.49], and the AUC of the predictive model including SVT was 0.710 (95%CI 0.670-0.750). During a median follow-up of 2.9 years, 12 all-cause deaths, 60 new-onset AF cases, and 139 composite endpoint events were observed. Spearman correlation analysis indicated that PACs burden was positively correlated with new-onset AF (r=0.32, P<0.001) and composite endpoint events (r=0.28, P=0.002). Kaplan-Meier survival analysis showed that the cumulative incidence of new-onset AF and composite endpoint events was significantly higher in patients with PACs >200 beats/day compared to those with PACs <200 beats/day. Cox multivariate analysis demonstrated that PACs >200 beats/day was an independent risk factor for new-onset AF [HR (hazard ratio) 3.13, 95%CI 1.74-5.62, P<0.001] and composite endpoint events (HR 2.00, 95%CI 1.30-3.06, P=0.002). After incorporating PACs >200 beats/day into the multivariable model, the predictive performance for new-onset AF [NRI=0.264 (P=0.008), IDI=0.033 (P=0.010)] and composite endpoint events [NRI=0.233 (P=0.014), IDI=0.014 (P=0.024)] significantly improved.

Conclusion: In this study, POAF occurred in 20.7% of oHCM patients, with incidence rising alongside increased PACs burden. Additionally, PACs burden was positively correlated with the incidence of new-onset AF and composite endpoint events. Patients exhibiting PACs > 200 beats/day faced a substantially elevated risk of these outcomes. Incorporating this threshold into multivariate analysis significantly enhanced predictive accuracy for new-onset AF and adverse events, indicating that PACs >200 beats/day may serve as the optimal benchmark for risk stratification in oHCM patients.