第一部分 肥厚型心肌病的新型预后评估因素研究

（一）右心室受累在肥厚型心肌病患者预后评估中的价值

中文摘要

背景和目的：在临床实践过程中，合并右心室受累的肥厚型心肌病（hypertrophic cardiomyopathy，HCM）患者并不少见。然而以往关于HCM临床表现和预后的研究主要集中在左心室，比如左心室肥厚和左心室流出道梗阻等等，而右心室系统在HCM的诊疗和研究中往往被忽略。本研究的目的是评估HCM患者合并右心室受累对预后的影响以及右心室受累与肌小节基因突变的关系。

方法：本研究前瞻性地纳入了945例接受了右心室影像学评估的HCM患者，在排除肺动脉高压的患者之后，最终有893例HCM患者纳入了分析。在本研究中，超声心动图和/或心脏磁共振用于评价HCM患者是否存在右心室肥厚、右心室梗阻和右心室钆对比剂延迟强化。合并上述三种结构性右心室异常中任意一种的患者即被认为存在右心室受累。本研究的中位随访时间为3.0年，采用单因素和多因素Cox回归的方法评估右心室受累与HCM患者预后的关系，主要研究终点是心血管死亡，次要研究终点是全因死亡和心力衰竭相关死亡。669例患者接受了基因检测，采用单因素和多因素logistics回归的方法评估右心室受累与肌小节基因突变的关系。

结果：893例患者的平均年龄是47.2 ± 14.2岁，其中590例（66.1%）为男性。在所有893例HCM患者中，共114例（12.8%）被发现合并有右心室受累。与不合并右心室受累的患者相比，合并有右心室受累的HCM患者更年轻、左心室受累更严重、存在HCM家族史的比例也更高。预后分析的结果显示，合并右心室受累是HCM患者不良预后的独立危险因素[心血管死亡：校正风险比（hazard ratio，HR） 4.191，95%置信区间（confidence interval，CI） 1.706-10.298, P =0.002；全因死亡：校正HR 3.013，95% CI 1.290-7.039，P =0.011；心力衰竭相关死亡：校正HR 14.142，95% CI 2.337-85.559，P =0.004]，随后的敏感性分析也报告了同样的结果。基因分析的结果显示，相比不合并右心室受累的HCM患者，合并右心室受累的患者被检出肌小节基因型阳性的比例更高（合并右心室受累相比不合并右心室受累：57.0% versus 31.0%，P <0.001）。其中，肌球蛋白结合蛋白C3（myosin binding protein C3，MYBPC3）基因的致病（pathologic，P）/可能致病（likely pathologic，LP）的突变与右心室受累的关系最为密切。logistics回归的结果显示肌小节基因突变阳性[校正比值比（odd ratio，OR） 1.993，95% CI，1.170-3.396，P = 0.011]和MYBPC3的P/LP突变（校正OR 2.007，95% CI，1.064-3.783，P =0.031）与HCM合并右心室受累独立相关。

结论：合并右心室受累是HCM患者发生心血管死亡、全因死亡和心力衰竭相关死亡的独立危险因素。肌小节基因突变，特别是MYBPC3的P/LP突变，与这一表型的出现有密切的关系。

（二）心脏指数在肥厚型心肌病患者预后评估中的价值

中文摘要

背景和目的：肥厚型心肌病（hypertrophic cardiomyopathy，HCM）患者因为预后异质性大，因此需要准确的危险分层，以识别高危患者。心功能障碍是HCM的重要临床特征之一，但是目前针对HCM的危险分层策略中对心功能的考虑尚不充分，仅2020年美国心脏学会发布的指南中纳入了左心室射血分数这一心功能的指标。左心室射血分数是代表心脏收缩功能的指标，HCM患者中左心室射血分数常表现为正常或升高。而HCM主要表现为舒张功能障碍，使用左心室射血分数对HCM患者进行危险分层显然是不够的。因此在HCM的危险分层策略中，急需一个更为合适的指标来评价HCM患者的心功能情况。心脏指数反映了心脏的泵功能，有望成为比左心室射血分数更合适的HCM心功能评价指标。本研究主要探讨了HCM患者出现心脏指数降低的临床意义。

方法：本研究共前瞻性地纳入了927例接受过心脏磁共振心脏指数检查的HCM患者。本研究的中位随访时间为4.3年。主要研究终点是心血管死亡，次要研究终点是心脏性猝死（sudden cardiac death，SCD）和全因死亡。在预后分析时，使用surv_cutpoint的方法寻找心脏指数的最佳截断值，采用单因素和多因素Cox回归的方法评估心脏指数降低与HCM患者预后的关系。为了探究心脏指数对SCD预测的增量价值，我们将心脏指数降低和左心室射血分数降低分别加入2014年欧洲心脏学会提出的HCM Risk-SCD模型，在原有模型基础上构建新模型，并用C-统计量判断模型预测准确性，使用综合判别改善指数和净重新分类指数来评价模型优化的程度。

结果：心脏指数降低定义为心脏指数小于2.42 L/min/m²。在4.3年的随访期间，51（5.5%）例患者出现了全因死亡，40例（4.3%）患者出现了心血管死亡。在出现心血管死亡的HCM患者中，23（20.0%）为SCD。经过严格校正后的多因素Cox回归的结果表明，心脏指数降低是HCM患者心血管死亡[校正风险比（hazard ratio，HR） 2.976，95%置信区间（confidence interval，CI） 1.35-6.529，P =0.007]，SCD（校正HR 6.385，95% CI 2.086-19.544，P =0.001）和全因死亡（校正HR 2.428，95% CI 1.235-4.774，P =0.010）风险升高的独立危险因素。在模型优化研究中，通过在HCM Risk-SCD模型中加入心脏指数降低这一指标，原模型的C-统计量从原来的0.691（95% CI 0.584-0.798，P =0.002）增加到了0.762（95% CI 0.663-0.860，P <0.001），综合判别改善指数为0.021 (95% CI 0.004-0.038，P =0.018)，净重新分类指数0.560（95% CI 0.152-0.968，P =0.007)。而在原模型中加入左心室射血分数减低这一指标并不能改善原模型的效能。与左心室射血分数降低相比，心脏指数降低对所有终点的预测准确性也更高。

结论：心脏指数降低是HCM患者不良预后（包括心血管死亡、SCD和全因死亡）的强独立预测因素。将心脏指数降低加入HCM Risk-SCD模型能够有效改善模型的效能，增加预测准确性。加入左心室射血分数降低则不能改善模型。在预测各个终点时，心脏指数降低都比左心室射血降低具有更好的预测准确性。

（三）心肌损伤生物标志物在肥厚型心肌病患者预后评估中的价值

中文摘要

背景和目的：血清肌钙蛋白和肌酸激酶同工酶MB型（creatine kinase，MB，CK-MB）是易于检测和可靠的亚临床心肌损伤的特异性生物标志物。在心肌被破坏后，肌钙蛋白和CK-MB可以从心肌细胞被释放入血，是临床判断急性心肌损伤，包括急性心肌梗死和心肌炎的经典指标。不同于急性心肌损伤，肥厚型心肌病（hypertrophic cardiomyopathy，HCM）患者在其自然病程中存在持续而慢性的心肌损伤，最终导致心肌纤维化和心功能障碍。在临床实践中可以发现肌钙蛋白和CK-MB的升高在不合并急性心肌损伤的HCM患者中并不少见，但是其临床意义仍然缺乏充分的证据。本研究着重探讨了心肌肌钙蛋白I（cardiac troponin I，cTnI）和CK-MB在HCM患者预后预测中的作用，以期明确血清心肌损伤生物标志物的检测对HCM患者危险分层的意义。

方法：本研究纳入了1999年至2019年在中国医学科学院阜外医院连续入组的1045例HCM患者，他们均有基线cTnI和CK-MB的数据。本研究的排除标准是患者接受了经皮冠状动脉介入治疗或冠状动脉旁路移植术，或有肾功能衰竭。研究终点包括全因死亡、心血管死亡和心脏性猝死（sudden cardiac death，SCD）。血清cTnI和CK-MB作为连续变量和分类变量进行分别分析。在预后分析方面，本研究采用单因素和多因素Cox比例风险回归模型评估cTnI和CK-MB的水平与各结局之间的关系。本研究还找到了cTnI和CK-MB在HCM患者中的最佳截断值，并将“cTnI和CK-MB共同升高”加入2014 HCM Risk-SCD模型，并用C-统计量评估模型的预测准确性，使用综合判别改善指数和净重新分类指数来评价模型优化的程度。726例患者接受了基因检测，用以评价cTnI和CK-MB与肌小节基因突变的关系。

结果：本研究最终分析了970例随访成功的患者（平均年龄49.3岁；36.4%为女性)。本研究的中位随访时间为4.3年，随访期间87例患者出现结局。多因素Cox比例风险模型的结果显示，HCM患者血清cTnI水平和CK-MB水平的升高（每升高1 IU/L）与全因死亡[cTnI每升高0.05 ng/mL：校正风险比（hazard ratio，HR） 1.038，95%置信区间（confidence interval，CI） 1.024-1.052，P <0.001；CK-MB每升高1 IU/L：校正HR 1.021，95% CI 1.007-1.035，P =0.004]，心血管死亡（cTnI每升高0.05 ng/mL：校正HR 1.040，95% CI 1.026-1.055，P <0.001；CK-MB每升高1 IU/L：校正HR 1.025，95% CI 1.007-1.043，P =0.006）和SCD（cTnI每升高0.05 ng/mL：校正HR 1.045，95% CI 1.028-1.063，P <0.001；CK-MB每升高1 IU/L：校正HR 1.032，95% CI 1.013-1.052，P =0.001）风险升高独立相关。此外我们进一步发现，血清cTnI和CK-MB同时升高的HCM患者的预后比cTnI或CK-MB单独升高的HCM患者，以及cTnI和CK-MB均未升高的HCM患者更差。血清cTnI和CK-MB同时升高的HCM患者相较cTnI或CK-MB单独升高的HCM患者，以及cTnI和CK-MB均未升高的HCM患者也表现出更严重的心肌纤维化。更重要的是，将“cTnI和CK-MB同时升高”加入2014 HCM Risk-SCD模型可以显著提高模型对于SCD的预测能力，以及重分类的能力（综合判别改善指数 0.064，95% CI 0.029-0.100，P <0.001；净重新分类指数 0.652，95% CI 0.326-0.977，P <0.001)。基因检测的结果也提示，血清cTnI和CK-MB同时升高的HCM患者肌小节基因阳性的检出率更高。

结论：综合评价亚临床心肌损伤生物标志物，cTnI和CK-MB，对HCM患者的不良结局预测具有相当大的临床价值。临床上进行常规的cTnI和CK-MB的检测可能有助于识别高危SCD患者并指导植入式心律转复除颤器的植入，提高HCM一级预防的质量。

Part I Novel Prognostic Predictors in Patients with Hypertrophic Cardiomyopathy

I-1: Prognostic Significance of Structural Right Ventricular Involvement in Patients with Hypertrophic Cardiomyopathy

ABSTRACT

Background and aims: Right ventricular involvement is not uncommon in patients with hypertrophic cardiomyopathy (HCM) in clinical practice. However, previous studies on the clinical manifestations and prognosis of HCM mainly focused on the left ventricle, such as left ventricular hypertrophy and left ventricular outflow tract obstruction. The involvement of right ventricle is often overlooked. This study aimed to evaluate the prognostic impact of right ventricular involvement in patients with HCM, and the associations between right ventricular involvement and sarcomeric gene mutations.

Methods: A total of 945 patients with HCM who underwent right ventricular assessment by echocardiography and/or cardiac magnetic resonance were prospectively enrolled in this study. The exclusion criterion in this study was the presence of pulmonary hypertension. After exclusion, 893 patients with HCM were finally included in the analysis. In this study, the presence of right ventricular hypertrophy, right ventricular obstruction, and right ventricular late gadolinium enhancement were evaluated in patients, and those with any one of the above three structural right ventricular abnormalities were identified as having right ventricular involvement. The median follow-up time was 3.0 years. Univariable and multivariable Cox regression models were used to evaluate the associations between right ventricular involvement and the outcomes of patients with HCM. The primary endpoint of this study was cardiovascular death; the secondary endpoints were all-cause death and heart failure-related death. A total of 669 patients underwent genetic testing. Univariable and multivariate logistic regression models were used to evaluate the associations between right ventricular involvement and sarcomeric gene mutations in patients with HCM.

Results: The mean age of the participants was 47.2 ± 14.2 years, and 590 (66.1%) of them were male. Of the 893 patients, 114 (12.8%) were identified as having right ventricular involvement. Compared with patients without right ventricular involvement, those with right ventricular involvement were younger, had more severe left ventricular manifestations, and had a higher prevalence of family history of HCM. The survival analysis showed that right ventricular involvement was independently associated with an increased risk of cardiovascular death (adjusted hazard ratio [HR] 4.191, 95% confidence interval [CI] 1.706-10.298, P =0.002), All-cause death (adjusted HR 3.013, 95% CI 1.290-7.039, P =0.011), and heart failure-related death (adjusted HR 14.142, 95% CI 2.337-85.559, P =0.004). These results were then confirmed by a sensitivity analysis. The genetic analysis showed that patients with right ventricular involvement were more likely to carry sarcomeric gene mutations than those without right ventricular involvement (57.0% versus 31.0%, P <0.001). The pathologic (P) and likely pathologic (LP) mutations of myosin binding protein C3 (MYBPC3) were most closely related to right ventricular involvement. The results of logistics regression showed that sarcomeric gene mutations (adjusted odd ratio [OR] 1.993, 95% CI 1.170-3.396, P =0.011) and MYBPC3 P/LP mutations (adjusted OR 2.007, 95% CI 1.064-3.783, P =0.031) were independently associated with the presence of right ventricular involvement in patients with HCM.

Conclusion: Right ventricular involvement is an independent risk factor for cardiovascular death, all-cause death, and heart failure-related death in patients with HCM. Sarcomeric gene mutations, especially the P/LP mutations of MYBPC3, have significant contributions to the right ventricular involvement in patients with HCM.

I-2: Prognostic Significance of Cardiac Index in Patients with Hypertrophic Cardiomyopathy

ABSTRACT

Background and aims: Hypertrophic cardiomyopathy (HCM) has strong heterogeneity in patient outcomes. Therefore, accurate risk stratification is urgently needed to identify patients at high risk. Cardiac dysfunction is one of the most important clinical features of HCM. However, only the 2020 American Heart Associate/American College of Cardiology guidelines include a cardiac function indicator, left ventricular ejection fraction, in their recommendations for evaluating the risk in patients with HCM. The left ventricular ejection fraction represents the systolic function of the heart, which is usually normal or mildly elevated in patients with HCM. Since HCM is typically characterized as having diastolic dysfunction, there is a frank deficiency using left ventricular ejection fraction in risk stratification of HCM. Therefore, a more appropriate indicator to evaluate the cardiac function is urgently needed in patients with HCM. The cardiac index reflects the pumping function of the heart and is expected to be a more suitable index than the left ventricular ejection fraction to evaluate cardiac function in patients with HCM. The aim of this study was to investigate the clinical significance of cardiac index in patients with HCM.

Methods: A total of 927 patients with HCM and measurement of cardiac index by cardiac magnetic resonance were prospectively enrolled in this study. The median follow-up was 4.3 years. The primary endpoint was cardiovascular death. The secondary endpoints were sudden cardiac death (SCD) and all-cause death. The surv_cutpoint function was used to find the optimal cut-off value of the cardiac index. The univariable and multivariate Cox regression models were used to evaluate the associations between reduced cardiac index and outcomes in patients with HCM. To investigate the incremental value of cardiac index in predicting SCD, the reduced cardiac index was added to the 2014 European Society of Cardiology HCM Risk-SCD model, and C-statistics, integrated discrimination improvement, and net reclassification improvement were used to evaluate the performance of the model. The prediction value of the cardiac index was then compared with that of the left ventricular ejection fraction.

Results: Reduced cardiac index was defined as a cardiac index of less than 2.42 L/min/m². During 4.3 years of follow-up, 51 patients (5.5%) experienced all-cause death and 40 (4.3%) died from cardiovascular causes. Among patients with HCM who experienced cardiovascular death, 23 (20.0%) had SCD. Multivariable Cox regression showed that reduced cardiac index was an independent risk factor for cardiovascular death (adjusted hazard ratio [HR] 2.976, 95% confidence interval [CI] 1.35-6.529, P <0.05). P =0.007), SCD (adjusted HR 6.385, 95% CI 2.086-19.544, P =0.001) and all-cause death (adjusted HR 2.428, 95% CI 1.235-4.774, P =0.010) in patients with HCM. In the model improvement analysis, addition of reduced cardiac index into the 2014 HCM Risk-SCD model increased the C-statistic of the original model from 0.691 (95% CI 0.584-0.798, P =0.002) to 0.762 (95% CI 0.663-0.860, P =0.002). The integrated discrimination improvement was 0.021 (95% CI 0.004-0.038, P =0.018), and the net reclassification improvement was 0.560 (95% CI 0.152-0.968, P =0.007). Addition of reduced left ventricular ejection fraction to the original model did not improve the performance of the original model. The reduced cardiac index also had greater predictive accuracy for all endpoints compared with reduced left ventricular ejection fraction.

Conclusion: Reduced cardiac index is a strong independent risk factor for poor prognosis in patients with HCM, including cardiovascular death, SCD and all-cause death. Adding reduced cardiac index into the 2014 HCM Risk-SCD model can effectively improve the model performance; while adding reduced left ventricular ejection fraction failed to achieve such improvement. Reduced cardiac index had better predictive accuracy than reduced left ventricular ejection when predicting all endpoints of the study.

I-3: Prognostic Significance of Myocardial Injury Biomarkers in Patients with Hypertrophic Cardiomyopathy

ABSTRACT

Background and aims: Serum troponins and creatine kinase MB (CK-MB) are readily detectable and reliable specific biomarkers of subclinical myocardial injury. Troponins and CK-MB can be released into the blood from cardiomyocytes due to myocardial damage. Troponins and CK-MB are classical indicators of acute myocardial injury, including acute myocardial infarction and myocarditis. Different from acute myocardial injury, patients with hypertrophic cardiomyopathy (HCM) are in non-acute myocardial injury in their natural courses, which eventually leads to myocardial fibrosis and cardiac dysfunction. The elevation of troponin and CK-MB is not uncommon in patients with HCM without acute myocardial injury, but its clinical significance still lacks sufficient evidence. This study investigated on the role of cardiac troponin I (cTnI) and CK-MB in patients with HCM and tried to figure out the use of myocardial injury biomarkers in risk stratification of patients with HCM.

Methods: A total of 1045 consecutive HCM patients with baseline cTnI and CK-MB data were enrolled in this study between 1999 and 2019 at Fuwai Hospital. Patients were excluded if they underwent percutaneous coronary intervention or coronary artery bypass grafting or if they had renal failure. The endpoints of this study were all-cause death, cardiovascular death, and sudden cardiac death (SCD). Serum cTnI and CK-MB levels were analyzed for both continuous and categorical variables. For survival analysis, univariate and multivariate Cox proportional hazards regression models were used to evaluate the associations between cTnI and CK-MB levels and outcomes. The optimal cut-off values for cTnI and CK-MB are identified in this study, and the indicator "both cTnI and CK-MB elevated" was added to the 2014 HCM Risk-SCD model for model improvement analysis. The predictive accuracy of the model was evaluated by C-statistic, and the model improvement was evaluated by the integrated discrimination improvement, and the net reclassification improvement. A total of 726 patients underwent genetic testing. The associations between cTnI, CK-MB and sarcomeric gene mutations were evaluated.

Results: A total of 970 patients (mean age 49.3 years; 36.4% female) were finally analyzed. During the median follow-up time of 4.3 years, 87 patients experienced mortality. The results of multivariate Cox proportional hazards models showed that higher levels of cTnI and CK-MB were associated with an increased risk of all-cause death (per 0.05 ng/mL increase in cTnI: adjusted hazard ratio [HR] 1.038, 95% confidence interval [CI] 1.024-1.052, P <0.001; per 1 IU/L increase in CK-MB: adjusted HR 1.021, 95% CI 1.007-1.035, P =0.004), cardiovascular death (per 0.05 ng/mL increase in cTnI: adjusted HR 1.040, 95% CI 1.026-1.055, P <0.001; per 1 IU/L increase in CK-MB: adjusted HR 1.025, 95% CI 1.007-1.043, P =0.006) and SCD (per 0.05 ng/mL increase in cTnI: adjusted HR 1.045, 95% CI 1.028-1.063, P <0.001; per 1 IU/L increase in CK-MB: adjusted HR 1.032, 95% CI 1.013-1.052, P =0.001). We further found that patients with HCM with both cTnI and CK-MB elevated had worse prognoses than those with only cTnI or CK-MB elevated, and those with neither cTnI nor CK-MB elevated. Patients with HCM with both cTnI and CK-MB elevated also show more severe myocardial fibrosis than those with only cTnI or CK-MB elevated, and those with neither cTnI nor CK-MB elevated. More importantly, adding "both cTnI and CK-MB elevated" into the 2014 HCM Risk-SCD model significantly improved the model performance for predicting SCD (integrated discrimination improvement 0.064, 95% CI 0.029-0.100, P <0.001; net reclassification improvement 0.652, 95% CI 0.326-0.977, P <0.001). The results of genetic testing also suggested that the yield rate of sarcomeric gene mutations was higher in patients with HCM with elevated levels of cTnI and CK-MB.

Conclusion: Comprehensive evaluation of the biomarkers of subclinical myocardial injury, cTnI and CK-MB, has considerable clinical value in predicting the adverse outcome in patients with HCM. Routine clinical assessments of cTnI and CK-MB may help to identify patients at a high risk of SCD and guide the implantation of an implantable cardioverter defibrillator, which improves the quality of primary prevention of HCM.