第一部分：金属裸支架、第一代和第二代药物洗脱支架用于急性心肌梗死患者的有效性与安全性比较研究

研究背景：冠状动脉粥样硬化性心脏病（简称冠心病）是常见的心血管疾病，严重危害人类的健康。与单纯药物治疗相比，经皮冠状动脉介入术（PCI）显著改善了冠心病患者的治疗效果及预后。经过多年的技术开发与临床应用，冠状动脉支架先后经历了不同的发展阶段。与最初的金属裸支架（BMS）相比，近年来开发的药物洗脱支架（DES）进一步提高了经皮冠状动脉介入术（PCI）的有效性与安全性。急性心肌梗死（AMI）是冠心病发展过程中的严重事件，其病理学基础通常为不稳定粥样硬化斑块发生破裂，导致激活的血小板聚集于病变部位形成急性血栓、并阻塞冠状动脉内血流，导致心脏组织因缺血缺氧而发生急性坏死，AMI严重影响患者的预后，需要积极防治。冠状动脉内置入BMS和DES是治疗AMI的重要手段，但是对于我国患有急性ST段抬高型心肌梗死（STEMI）、并接受PCI术的患者而言，BMS与第一代和第二代药物洗脱支架（G1-DES和G2-DES）的有效性和安全性尚缺乏较大规模的研究报告。

研究目的：本研究目的：在明确诊断为STEMI、并接受急诊PCI术的患者当中，比较BMS、G1-DES和G2-DES的有效性与安全性差异，为冠状动脉支架的优化临床应用提供真实世界的证据。

研究方法：2004年4月至2014年5月之间于中国医学科学院阜外医院明确诊断为STEMI，并接受急诊PCI术的1,623例连续患者。排除标准包括：（1）因支架内再狭窄、或支架内血栓形成而导致的STEMI，共排除95例患者，（2）置入除BMS、G1-DES或G2-DES以外类型的支架、或同时置入不同种类型支架的患者，共排除305例患者。最终纳入1,223例患者数据。根据患者置入支架类型将其分为3组，分别为BMS组、G1-DES组和G2-DES组。随访时间为1年，随访调查的终点包括有效性终点和安全性终点。有效性终点为主要不良心血管事件（MACE），包括由靶血管相关心肌梗死（TV-MI）、靶血管重建（TVR）、靶病变重建（TLR）和心源性死亡组成的联合终点，以及上述独立终点事件；安全性终点为支架内血栓，本研究所报告的支架内血栓根据ARC定义为确定的（definite）和可能性大的（probable）支架内血栓。经过1年随访后，应用卡方检验、Kaplan-Meier法比较了靶血管相关心肌梗死（TV-MI）、靶血管重建（TVR）、靶病变重建（TLR）、心源性死亡和支架内血栓的发生率，并应用多因素Cox比例风险回归模型、以及Logistic回归模型比较了上述不良事件在不同支架类型之间的发生风险。

研究结果：在确诊为STEMI并接受急诊PCI术的1,223名连续患者当中，1年随访结果表明与BMS组相比，G1-DES和G2-DES组的MACE事件发生率显著降低（分别为20.4%、4.2%和4.1%，P < 0.001；BMS与G1-DES相比，P < 0.001；BMS与G2-DES相比，P < 0.001），而在G1-DES和G2-DES组之间，MACE事件发生率无显著差异（P = 0.969）。在MACE事件当中，TV-MI发生率在分别为2.6%（BMS）、0.3%（G1-DES）和1.0%（G2-DES），三组间比较结果具有显著性差异（P = 0.017），而事后分析表明G1-DES组的TV-MI事件低于BMS组（P = 0.009），在BMS组和G2-DES组之间（P = 0.070）、以及在G1-DES组和G2-DES组之间（P = 0.216）均无显著差异。此外与BMS组相比，G1-DES和G2-DES组的TVR、TLR和心源性死亡率显著降低（TVR分别为10.4%、2.5%和1.8%；TLR分别为7.8%、1.1%和1.2%；心源性死亡率分别为9.2%、1.7%和1.8%；P值均为< 0.001），而上述事件发生率在G1-DES和G2-DES组之间无显著差异（P值分别为0.458、> 0.999和0.908）。1年内血栓支架发生率在三组间未见差别（P = 0.354）。Kaplan-Meier曲线显示在G1-DES组和G2-DES组当中，1年内无TLR生存率和无心源性死亡生存率明显高于BMS组（P值均为< 0.001），而G1-DES组和G2-DES组之间则无此差异（P值分别为0.927和0.700）。多因素Cox比例风险回归分析显示与BMS相比，G1-DES和G2-DES是降低TVR（G1-DES，HR 0.24，95% CI [0.11-0.50]，P < 0.001；G2-DES，HR = 0.40，95% CI [0.27- 0.58]，P < 0.001）和TLR（G1-DES，HR = 0.15，95% CI [0.05-0.43]，P < 0.001；G2-DES，HR = 0.39，95% CI [0.25-0.61]，P < 0.001）的保护性因素，而TVR和TLR的发生风险在上述两组之间无显著差异。Logistic回归分析显示，支架类型不是导致支架内血栓的危险因素。

研究结论：（1）在确诊为急性ST段抬高型心肌梗死、并接受急诊PCI治疗的患者当中，BMS组MACE事件发生率明显高于G1-DES和G2-DES组，而在G1-DES和G2-DES组之间MACE事件发生率无显著差异；（2）与金属裸支架相比，置入G1-DES或G2-DES能够在1年之内降低TVR和TLR的发生率和发生风险，而在G1-DES和G2-DES之间尚无明显差别；（3）在置入上述三种类型支架后1年内，支架内血栓的发生率和发生风险无明显差异；此外，TV-MI和心源性死亡风险在三组间无显著差异。

关键词：急性ST段抬高型心肌梗死，金属裸支架，药物洗脱支架，靶血管重建，靶病变重建。

第二部分：急性冠状动脉综合征患者置入第一代和第二代药物洗脱支架的有效性与安性比较研究

研究背景：急性冠状动脉综合征（ACS）包括急性ST段抬高型心肌梗死（STEMI）、非ST段抬高型心肌梗死（NSTEMI）和不稳定型心绞痛，其共同的病理学机制为不稳定斑块破裂或侵蚀引起不同程度的血栓形成，从而导致冠状动脉内血流急性减少。ACS通常表现为突发性，且具有潜在致命性，故应受到重视并积极防治。既往研究报道与稳定型冠心病相比，ACS是增加支架内血栓的独立危险因素，因此在针对ACS介入治疗过程中需要谨慎选择冠状动脉支架类型。针对G1-DES和G2-DES用于ACS患者介入治疗的有效性和安全性比较结果尚有争议，而目前尚未见文献报道G1-DES和G2-DES用于中国ACS患者的作用比较研究。

研究目的：本研究拟在确诊为ACS并接受冠状动脉支架置入术的患者中，明确G1-DES和G2-DES的有效性和安全性差异，从而为我国ACS患者介入治疗支架类型选择提供一定理论依据。

研究方法：本研究前瞻性筛选纳入2013年1月至2013年12月中国医学科学院阜外医院确诊为急性冠状动脉综合征，并接受PCI治疗的连续患者。患者排除标准包括（1）仅接受经皮冠状动脉腔内成形术（PTCA）治疗，而未接受支架置入的患者；（2）G1-DES或G2-DES以外类型的支架、或同时置入不同种类型支架的患者；（3）诊断为急性心肌梗死或不稳定型心绞痛以外的非ACS患者。最终纳入3，929例患者数据，包括急性ST段抬高型心肌梗死心肌梗死862例，非ST段抬高型心肌梗死284例和不稳定型心绞痛2,783例。根据患者置入支架的类型，将其分为G1-DES组和G2-DES组。随访时间为2年，随访终点事件同上文所述，本研究所报告的支架内血栓包括确定的（definite）、可能性大的（probable）和可能的（possible）支架内血栓。采用倾向性评分匹配法，以1:1比例匹配了两组患者，匹配后患者数目为2，284例，随后应用卡方检验、Kaplan-Meier法和多因素Cox比例风险回归模型在两组间分析并比较了终点事件的发生率和危险因素。

研究结果：在接受PCI治疗的ACS患者当中，由倾向性评分匹配法筛选了2,248例患者，G1-DES组和G2-DES组各包含1,124例患者。2年随访结果显示，两组间MACE事件发生率无显著差异（分别为4.8%和3.9%，P = 0.302）。在G1-DES和G2-DES两组之间TV-MI发生率无显著差异（分别为0.7%和0.7%，P > 0.999），而TVR（分别为5.5%和3.5%，P = 0.019）、TLR（分别为4.1%和2.5%，P = 0.033）和心源性死亡率（分别为0.5%和1.4%，P = 0.032）则明显不同。在G1-DES和G2-DES之间，急性、亚急性、晚期、极晚期和2年累计支架内血栓发生率均无显著差异（急性支架内血栓：0.0%和0.2%，P = 0.500；亚急性支架内血栓：0.0%和0.1%，P > 0.999；晚期支架内血栓：0.2%和0.2%，P = 0.790；2年累计支架内血栓：0.7%和0.5%，P = 0.826）。经Kaplan-Meier分析发现，在G2-DES组TLR生存率显著增高（P = 0.038），而无心源性死亡生存率显著降低（P = 0.039）。多因素Cox比例风险回归分析结果显示，与G1-DES相比，G2-DES是能够降低TLR的保护性因素（HR = 0.592，95% CI [0.370-0.947]，P = 0.029），而TV-MI、心源性死亡和累计支架内血栓的发生风险在两组间无明显差异（TV-MI：HR = 1.121，95% CI [0.380-3.303]，P = 0.836；心源性死亡：HR = 2.469，95% CI [0.883-7.057]，P = 0.084；累计支架血栓：HR = 1.197，95% CI [0.471-3.041]，P = 0.706）。

研究结论：（1）在确诊为急性冠状动脉综合征并接受PCI治疗的患者当中，G1-DES和G2-DES之间MACE事件发生无无显著性差异；（2）与G1-DES相比，G2-DES是能够降低靶病变重建的保护性因素；（3）在G1-DES和G2-DES之间，靶病变相关心肌梗死、心源性死亡和累计支架内血栓的发生风险无显著差异。

关键词：急性冠状动脉综合征，药物洗脱支架，靶血管重建，靶病变重建，支架内血栓。

Part 1: Comparison of effectiveness and safety among bare-metal stent, and first and second generation drug-eluting stent in Chinese patients diagnosed with ST-segment elevated acute myocardial infarction

Background: Coronary artery disease (CAD) is a quite common cardiovascular disorder that threatens public health worldwide. Percutaneous coronary intervention (PCI) technique remarkably improved outcomes in patients with CAD. Coronary stents experienced continuous development in these years, and the emerging drug-eluting stents (DES) further increased the effectiveness and safety of PCI as compared with bare-metal stents (BMS). Acute myocardial infarction (AMI) is the end-stage malignant event of CAD. Pathologically the unstable atherosclerotic plaque breaks and promotes platelets to aggregate around the lesion site, forming an acute thrombus and thoroughly blocking the blood flow in the coronary artery. Thus the cardiac tissue necrosis occurs due to acute ischemia, which severely worsens prognosis in patients and is supposed to be prevented. Implantation with BMS or DES in the culprit coronary artery has been a standardized approach to treat ACS. Nevertheless it has not been fully elucidated whether BMS is different from DES, or whether G1-DES is significantly different from G2-DES in respect to effectiveness and safety in Chinese patients diagnosed as STEMI.

Object: The study aims to demonstrate and compare the effectiveness and safety among BMS, G1-DES and G2-DES in patients diagnosed as STEMI and receiving PCI treatment, and provide real-world evidence to refine the practice of coronary artery stent implantation.

Method: A consecutive group of patients who were diagnosed as STEMI and received emergency PCI between April 2004 and May 2014 in Fuwai Hospital, National Center for Cardiovascular Diseases, Chinese Academy of Medical Sciences, Beijing, China were retrospectively enrolled in the study. The exclusion criteria included (1) STEMI due to intra-stent restenosis or stent thrombosis (95 cases) and (2) patients who received stent implantation other than BMS, G1-DES or G2-DES, or received more than 1 type of stents (305 cases). Finally 1,223 cases remained in the study. Patients were divided into 3 groups according to the type of implanted stent, namely BMS group, G1-DES group and G2-DES group. The endpoints included effective and safety outcomes. The effective endpoint is defined as MACE (major adverse cardiac events) including a composition of target vessel-related myocardial infarction (TV-MI), target vessel revascularization (TVR), target lesion revascularization (TLR) and cardiogenic death, as well as these individual events. The safety endpoint id defined as definite and probable stent thrombosis based on the Academic Research Consortium criteria. The follow up lasted for 1 year. The occurrence of TV-MI, TVR, TLR, cardiogenic death and stent thrombosis were compared by Chi-square test and Kaplan-Meier method. Besides multivariate Cox proportional-hazard regression analysis and Logistic regression methods were applied to compare the hazard/odd ratio of the above mentioned adverse events among the 3 groups.

Results: In the 1,223 consecutive patients diagnosed as STEMI and receiving emergency PCI treatment, the 1-year-follow-up data revealed that the occurrence of MACE was higher in the BMS group as compared with DES-1 and DES-2 groups (BMS 20.4%, G1-DES 4.2% and G2-DES 4.1%, P < 0.001; BMS vs. G1-DES, P < 0.001; BMS vs G2-DES, P < 0.001), while it was not different between G1-DES and G2-DES groups (P = 0.969). The occurrence of TV-MI in BMS, G1-DES and G2-DES groups were 2.6%, 0.3% and 1.0% respectively, and differed among the 3 groups (P = 0.017). By post-hoc Chi-square analysis we found that TV-MI was reduced in the G1-DES group compared to BMS group (P = 0.017), while it was similar between BMS and G1-DES (P = 0.070) and between G1-DES and G2-DES groups (P =0.216). Furthermore compared BMS both G1-DES and G2-DES reduced the occurrence of TVR (10.4%, 2.5% and 1.8%), TLR (7.8%, 1.1% and 1.2%) and cardiogenic death (9.2%, 1,7% and1.8%) (all P values < 0.001) while the occurrence rate of these events were not different between the G1-DES and G2-DES group (P values were 0.458,> 0.999 and 0.908). The stent thrombosis rates were not differed among the 3 groups within 1 year (P = 0.354). It was demonstrated by Kaplan-Meier analysis that the TLR- and cardiogenic free survival rates were both higher in the DES groups as compare to that in the BMS group (all P values < 0.001), and did not show statistical difference between G1- and G2-DES groups. By further analysis with multivariate Cox proportional hazard regression model, it was elucidated that both G1- and G2-DES were predictive of reduced TVR (G1-DES, HR 0.24, 95% CI [0.11-0.50], P < 0.001; G2-DES, HR = 0.40, 95% CI [0.27- 0.58], P < 0.001) and TLR (G1-DES, HR = 0.15, 95% CI [0.05-0.43], P < 0.001; G2-DES, HR = 0.39, 95% CI [0.25-0.61], P < 0.001), while the hazard ratios of TV-MI, TLR or cardiogenic death were not different between G1- and G2-DES groups. In addition the Logistic regression analysis revealed that stent type was not predictive of stent thrombosis.

Conclusion: (1) In patients diagnosed as STEMI and receiving PCI treatment, MACE occurred more frequently in the BMS group, and did not differ in occurrence between G1-DES and G2-DES groups. (2) G1-DES and G2-DES were protective factors of TVR and TLR within 1 year, while the risk of these events were not different between the 2 groups. (3) The occurrence and risk of TV-MI thrombosis were similar among BMS, G1-DES and G2-DES groups, and the risk of TV-MI or cardiogenic death were not different among the 3 groups.

Key words: Acute ST-segment elevated myocardial infarction (STEMI), bare-metal stent (BMS), drug-eluting stent (DES), target vessel revascularization (TVR) and target lesion revascularization (TLR).

Part 2: Comparison of effectiveness and safety between first and second generation drug-eluting stent in Chinese patients diagnosed with acute coronary syndrome

Background: Acute coronary syndrome (ACS) consists of ST-segment elevated acute myocardial infarction, non-ST-segment elevated acute myocardial infarction and unstable angina pectoris. The common pathological mechanism underlying ACS is complete or incomplete coronary thrombosis caused by ruptured or eroded atherosclerotic plaque, leading to an acute decrease in coronary blood flow. ACS usually presents with sudden onset and potential lethality, and therefore should be cautiously prevented. According to previous reports ACS increases hazard ratio of stent thrombosis after PCI treatment as compared with stable coronary disorder. Thus due attention should be pay to the selection of stent type during the PCI treatment in patients with ACS. Yet there are controversies on the effectiveness and safety issues between G1-DES and G2-DES in the therapy of ACS, and it has not been reported whether G2-DES is superior to G1-DES in the setting of ACS in Chinese patients.

Objective: The current study aims to compare the effectiveness and safety between G1-DES and G2-DES in Chinese patients diagnosed as ACS and receiving PCI treatment, and to provided theoretical evidence of appropriate selection of stent type in the PCI treatment in ACS patients.

Methods: A consecutive group of patients who received PCI treatment between January 2013 to December 2013 in Fuwai Hospital, National Center for Cardiovascular Diseases, Chinese Academy of Medical Sciences, Beijing, China were retrospectively enrolled in the study, and those diagnosed as ACS were screened and identified. The exclusion criteria included (1) patients who received percutaneous transluminal coronary angioplasty (PTCA) without stent implantation, (2) patients who had stent implantation other than G1-DES or G2-DES or had more than 1 type of stent implanted, and (3) patients who were not diagnosed as ACS. A total of 3,929 patients were enrolled, including 862 cases with STEMI, 284 cases with NSTEMI and 2,783 cases with unstable angina pectoris. All patients were divided in to G1-DES and G2-DES group according to the implanted stent type. The follow up lasted for 2 years, and the outcomes are described in Part 1 in this paper. Stent thrombosis were defined as definite, probable or possible thrombosis according to the Academic Research Consortium criteria. A propensity score matching method was applied to match the patients in a 1:1 manner. The Chi-square test, Kaplan-Meier analysis and multivariate Cox proportional hazard regression analysis were applied to compare the occurrence rate and hazard ratio of outcome events.

Results: From all the patients who receiving PCI treatment due to ACS, 2,248 cases were selected by propensity score matching. After 2 years follow up, there was no significant difference on the occurrence of MACE between G1-DES and G2-DES groups (4.8% vs. 3.9%, P = 0.302). The occurrence of TV-MI did not differ between the 2 groups (0.7 vs. 0.7, P > 0.999). Meanwhile statistical differences on TVR (5.5% vs. 3.5%, P = 0.019), TLR (4.1% vs. 2.5%, P = 0.033) and cardiogenic death (0.5% vs. 1.4%, P = 0.033) were detected. we found that the TLR-free survival rate increased and cardiogenic death-free survival rate decreased in the G2-DES group (P = 0.038 and 0.039). The multivariate Cox proportional hazard regression analysis identified that G2-DES is predictive of reduced TLR as compared with G1-DES (HR = 0.592, 95% CI [0.370-0.947], P = 0.029), while the risks of TV-MI, cardiogenic death or cumulative stent thrombosis were similar between G1- and G2-DES groups (TV-MI, HR = 1.121, 95% CI [0.380-3.303], P = 0.836; cardiogenic death, HR = 2.469, 95% CI [0.883-7.057], P = 0.084; cumulative stent thrombosis, HR = 1.197, 95% CI [0.471-3.041], P = 0.706).

Conclusion: (1) In patients receiving PCI due to ACS, the occurrence of MACE was not significantly different between G1-DES and G2-DES groups. (2) G2-DES was identified as a protective factor of TLR as compared to G1-DES. (3) The occurrence and risk of TV-MI, cardiogenic death or cumulative stent thrombosis were not statistically different between G1-DES and G2-DES groups.

Key words: Acute coronary syndrome, drug-eluting stent, target vessel revascularization, target lesion revascularization and stent thrombosis.