摘要

复杂冠心病血运重建心脏团队决策实践流程优化研究

复杂冠心病，包括冠状动脉三支病变或左主干病变，是冠心病中最为严重的类型，其治疗决策常涉及多种选择。由于治疗选择的多样性和临床复杂性，欧美指南均推荐复杂冠心病患者需要由多学科心脏团队讨论以确保最佳血运重建决策。

尽管心脏团队在指南中得到明确推荐，但在真实世界中的应用却十分有限，指南推荐的团队决策流程整合至临床实践存在困难。主要原因包括以下几个方面：第一，心脏团队落地实践的资源条件要求高：（1）心脏团队要求多学科专家组成团队召开会议讨论患者的治疗决策，但真实世界中许多中心多学科医生资源匮乏；（2）心脏团队决策造成治疗决策延迟，患者需要下手术台等待决策，可能面临决策后再次进行经皮冠状动脉介入（percutaneous coronary intervention，PCI）治疗，增加患者诊疗负担，增加治疗花费，影响诊疗效率。因此，亟需能高效整合多学科医疗资源、提高决策效率的新流程，促进心脏团队地常规开展。第二，心脏团队实践对于复杂冠心病患者治疗决策和临床预后的改善缺乏高质量的证据。既往研究多为单臂病例系列报道，缺乏真实世界中多学科与单学科决策对比的证据，证据力度有限。心脏团队可能带来的治疗延迟对于临床结局的潜在影响和临床获益是否可以平衡，仍需进一步探索。第三，心脏团队开展流程需进一步优化。前期的心脏团队实践探索表明，心脏团队在决策过程中仍存在优化的空间。许多因素，如医生间差异、团队间差异对于决策的质量和稳定性仍有显著影响，需进一步标准化。亟需对专家决策的具体过程进行深入挖掘，识别心脏团队讨论过程中存在争议的关键人群及其特征，识别促使决策改变的关键因素，作为决策流程优化的潜在靶点，整合至新的实践流程中，在未来的前瞻性研究中加以验证。

针对心脏团队在真实世界无法常规开展，最优实践流程尚不明确这一关键问题，本研究旨在以心脏团队决策在真实世界常规落地实践为目标，开发全新优化决策流程，系统论证优化流程的优势，并探索可进一步优化的干预靶点。本研究总共分为三个阶段：在研究的第一阶段，我们通过一项多中心随机对照试验，验证了一种全新的冠状动脉造影术中实时心脏团队决策模式的可行性和安全性，并具体探讨了实时心脏团队决策模式对于临床效率、决策质量和工作负担的改善潜力。在第一部分的基础上，研究的第二部分对比了多学科心脏团队决策和真实世界单学科决策在决策质量的改善效果，同期探索了两种决策模式对于临床效率和患者预后的影响。第三部分，我们基于多学科心脏团队的讨论语料，聚焦于决策发生改变的讨论过程，探索引起决策发生改变的关键因素，为心脏团队流程的进一步优化提供潜在靶点。

主要的研究内容及结果如下：

复杂冠心病血运重建实时心脏团队决策流程的可行性及效果评价研究

目的

当前指南推荐复杂冠心病患者采用多学科心脏团队获得最佳的治疗决策，但目前的心脏团队资源要求高，临床中无法常规开展。开发实时决策流程，高效整合资源，有望为心脏团队的临床实践提供新的契机。本部分研究旨在评价一种基于通讯技术开发的“实时心脏团队”决策流程的有效性和安全性。

方法

本研究为一项多中心随机对照临床试验，研究纳入2023年8月至2024年5月期间于北京三家大型心脏中心行择期冠状动脉造影并确证为复杂冠心病的患者，按照1：1的比例随机分组至传统心脏团队组（冠状动脉造影术后，根据指南推荐开展线下面对面讨论）和实时心脏团队组（冠状动脉造影术中，基于标准流程开展实时线上讨论）。研究的主要终点为PCI住院时间延长率，主要次要终点为1年随访期间由全因死亡、卒中、心肌梗死、再发心绞痛入院和计划外血运重建组成的复合终点。次要终点包括团队专家的工作负担，团队决策情况，患者实际治疗情况，住院时间和住院花费及不良事件的各个组分等。

结果

本研究共纳入490例复杂冠心病患者，其中传统心脏团队组245例，实时心脏团队组245例。平均年龄为61.2 ± 9.8岁，女性占比为23.1%，三支病变患者占比89.6%，左主干病变患者占比29.8%。实时心脏团队决策组PCI住院时间延长率显著低于传统心脏团队组（15.5% vs. 31.0%，RR=0.49，95% CI: 0.34–0.69，P < 0.001）。1年随访期间，实时组和传统组的MACCE发生率分别为7.0%和9.8%，发生率差值及95% CI为 -2.86% （-2.13% – 7.92%），实时组不劣于传统组（非劣效检验P值 < 0.001）。与传统心脏团队相比，实时心脏团队决策的适宜性，患者实际接受的治疗和决策依从性的差异无统计学显著性，而团队专家的工作负担，患者的治疗等待时间和住院时长的降低具有统计学显著性（P < 0.05）。

结论

对于复杂冠心病患者，实时心脏团队决策可显著提升临床效率，降低PCI住院时间延长率，改善决策适宜性，降低专家工作负担，且临床预后不劣于传统心脏团队决策模式。

第二部分 复杂冠心病血运重建的多学科心脏团队模式与真实世界单学科模式决策效果评价研究

目的

当前指南推荐复杂冠心病患者采用多学科心脏团队获得最佳的治疗决策，但单学科决策仍在临床范围内广泛开展。既往研究多集中于对于心脏团队开展特定环节的探索，缺少与真世界单学科决策的全面对比。本研究基于多学科心脏团队决策临床试验研究对象数据与真实世界单学科决策队列人群研究数据开展，系统评价多学科决策和单学科决策模式对于决策质量和临床预后的影响。

方法

本研究为一项回顾性队列研究，研究基于第一部分“复杂冠心病血运重建实时心脏团队决策模式的可行性及效果评价研究”（多学科团队决策组）和“国人血运重建适宜性评价研究”（单学科决策组）的数据开展。“国人适宜性评价研究”为一项前瞻性多中心队列研究，研究纳入了2016-2017年期间择期行冠状动脉造影的稳定冠心病患者，采集真实世界中患者治疗意愿，介入医生治疗推荐和患者实际治疗策略，临床预后等情况。本研究拟纳入两项研究中经冠状动脉造影证实的复杂冠心病患者，采用倾向性评分匹配（1:1）的方法来均衡多学科团队决策组和单学科决策组患者的基线特征。研究的主要终点为决策适宜率，次要终点包括决策情况，患者实际接受治疗情况，住院时长等效率指标和1年随访期间主要不良心脑血管事件及其组分。

结果

研究共纳入1090例复杂冠心病患者，经过倾向性评分匹配后，多学科心脏团队组和单学科心脏团队组各302例患者入选（平均年龄59.7 ± 9.7，男性占比 78.3%）。两组间的患者基线特征均衡。结果显示，多学科心脏团队组决策适宜性高于单学科决策组（70.2% vs. 48.0%，χ²=30.748，P < 0.001）。两种决策模式下，患者接受的实际治疗差异存在统计学显著性，多学科决策组患者接受冠状动脉旁路移植术（coronary artery bypass grafting，CABG）的比例显著高于单学科决策组。多学科决策组的患者住院时长低于单学科组（3天 [2–7] vs. 5天 [3–12]）。1年随访期间，多学科和单学科决策组的主要心脑血管不良事件发生风险差异无统计学显著性（6.7% vs. 11.6%，HR=0.70，95% CI: 0.40 – 1.23，P = 0.211）。

结论

多学科心脏团队决策可显著提升患者的决策适宜性和治疗的适宜性，提高临床效率，减少住院时长。与单学科决策相比，多学科决策未体现出临床获益上的优势，可能与随访时间较短有关。

第三部分 基于多学科心脏团队讨论语料的复杂冠心病患者血运重建决策改变影响因素探索的定性研究

目的

多学科心脏团队已被血运重建指南推荐用于复杂冠心病血运重建治疗策略的决策。但既往研究多集中于心脏团队作为一种干预方式整体的影响，缺少决策过程的细致理解。此外，对于哪些患者应当纳入心脏团队讨论同样存在争议。本研究基于前期开展的“复杂冠心病心脏团队决策一致性对比研究”中的患者数据及团队讨论语料，聚焦于影响决策发生改变的关键环节，分析归纳关键人群特征和关键影响因素。

方法

本研究为一项定性方法研究。研究基于“复杂冠心病心脏团队决策一致性对比研究”数据开展，该研究为一项随机对照临床试验，于2019年10月至2020年3月纳入480例复杂冠心病患者，为每位患者配备四支心脏团队（包含六名外科医生和六名介入医生），研究中团队内每位专家均先对患者的治疗措施（如PCI或CABG）独立决策；而后，发表决策理由；之后，进行集中讨论；形成最终决策意见。基于上述研究数据和讨论语料，本研究拟分析决策过程中，心脏团队讨论前后专家治疗决策发生改变的人群特征，描述治疗决策改变方向的频数及在不同学科间的分布。进一步聚焦于决策发生改变的讨论语料，通过主题归纳法分析影响心脏团队专家决策向不同方向发生改变的关键因素，总结影响因素相关主题和亚主题。

结果

研究共纳入480例复杂冠心病病例（平均年龄为61.2 ± 9.1岁，男性占比 75.6%），其中355例患者的治疗决策发生改变。与决策未发生改变的患者相比，决策发生改变的患者年龄更高，体重指数更高，合并糖尿病，脑血管疾病，三支病变的比例更高，SYNTAX评分平均值更高，表明治疗决策发生改变的患者临床情况更为复杂。心脏团队会议过程中共产生5760个治疗决策，其中693个治疗决策在团队讨论后发生了改变。在发生改变的治疗决策中，决策结果由PCI更改为CABG，和由CABG更改为PCI的频率最高，且外科医生的决策改变频率高于内科医生。经过主题归纳，影响决策改变的关键因素可总结为四大主题：（1）患者特征；（2）病变特征；（3）手术获益；（4）远期获益。

结论

心脏团队讨论决策过程中，决策发生改变的患者临床病情更加复杂。专家决策发生改变较普遍。影响决策向不同方向发生改变的因素主要包括四大主题：（1）患者特征；（2）病变特征；（3）手术获益；（4）远期获益。未来需开展前瞻性研究验证本研究提出的假设。

Abstract

Optimization of the heart team decision-making process for myocardial revascularization in complex coronary artery disease.

Complex coronary artery disease (CAD), including triple-vessel disease or left main disease, represents the most severe type of CAD, with treatment decisions often involving multiple options. Due to the diversity of treatment options and clinical complexity, both European and American guidelines recommend that patients with complex CAD should be discussed by a multidisciplinary heart team to ensure optimal revascularization decisions.

Although the heart team is explicitly recommended in clinical guidelines, its real-world implementation remains limited. Integrating the guideline-recommended team-based decision-making process into clinical practice presents significant challenges. The reasons are as follows: First, the implementation of a heart team requires substantial resource investment: (1) A heart team consists of multidisciplinary experts who convene meetings to discuss patient treatment decisions. However, in real-world settings, many centers face a shortage of multidisciplinary medical professionals. (2) The heart team decision-making process may lead to delays in treatment, requiring patients to wait between the decision-making and the subsequent intervention. This delay may result in patients undergoing multiple procedures, increasing their healthcare burden, treatment costs, and reducing clinical efficiency. Therefore, there is an urgent need for a novel workflow that efficiently integrates multidisciplinary resources and improves decision-making efficiency, facilitating the routine implementation of heart teams. Second, there is a lack of high-quality evidence demonstrating the benefits of heart team practice in improving treatment decisions and clinical outcomes for patients with complex CAD. Most existing studies were single-arm case series, lacking comparative evidence between multidisciplinary and single-discipline decision-making in real-world settings, limiting the strength of the evidence. The potential impact of treatment delays caused by the heart team on clinical outcomes and whether these delays are outweighed by clinical benefits remain to be further explored. Third, the heart team workflow requires further optimization. Previous exploratory studies have shown that there is still room for improvement in the heart team decision-making process. Several factors, such as inter-physician and inter-team variability, significantly affect decision quality and consistency, necessitating further standardization. It is crucial to investigate the detailed process of expert decision-making, identify key patient groups for whom heart team discussions are most contentious, and determine the critical factors influencing decision changes. These elements could serve as potential targets for optimizing the decision-making process, which should be integrated into an optimized workflow and validated in future prospective studies.

To address the key issues that the heart team cannot be routinely implemented in real-world practice and that the optimal practice workflow remains unclear, this study aimed to develop a novel, optimized decision-making process that facilitated the routine implementation of heart team decision-making, systematically evaluated the advantages of the multidisciplinary heart team, and explored potential intervention targets for further improvement. This study consisted of three phases: In the first phase, we conducted a multicenter randomized controlled trial to evaluate the feasibility and safety of a novel real-time heart team decision-making approach during coronary angiography and assessed its potential to improve clinical efficiency, decision quality, and mitigate physician workload. Building on the first phase, the second phase compared the effectiveness of multidisciplinary heart team decision-making versus real-world single-discipline decision-making in improving decision quality while concurrently assessing the impact of these two approaches on clinical efficiency and patient outcomes. In the third phase, we analyzed heart team discussion transcripts, focusing on discussions that led to changes in decision-making. We explored the key factors driving these changes, identifying potential targets for further optimizing of the heart team workflow.

The main research contents and findings are as follows:

Part I. Safety and Feasibility of a Real-Time Heart Team Decision-Making Approach for Patients with Complex Coronary Artery Disease: A Multicenter Randomized Controlled Trial.

Objective

Current guidelines recommend that patients with complex CAD should be discussed by multidisciplinary heart teams to achieve optimal treatment decisions. However, the high resource demands of traditional heart teams make routine implementation in clinical practice challenging. Advances in communication technology offer the potential to integrate resources efficiently, providing new opportunities for heart team practice. This study aimed to evaluate the feasibility and safety of a “real-time heart team” decision-making process based on communication technology.

Methods

This study was a multicenter randomized controlled clinical trial. Patients diagnosed with complex CAD after elective coronary angiography at three large cardiac centers in Beijing between August 2023 and May 2024 were enrolled and randomized in a 1:1 ratio into either the traditional heart team group (post-angiography, guideline-recommended face-to-face discussion) or the real-time heart team group (intra-procedural online discussion). The primary endpoint was the rate of prolonged percutaneous coronary intervention (PCI) hospital stay. The primary secondary endpoint was a composite outcome, including all-cause mortality, stroke, myocardial infarction, rehospitalization for recurrent angina, and unplanned revascularization during the one-year follow-up. Secondary endpoints included specialist workload, team decision-making process, actual treatment received by patients, length of stay, hospitalization costs, and individual adverse event components.

Results

A total of 490 patients with complex CAD were included, with 245 patients in the traditional heart team group and 245 in the real-time heart team group. The mean age was 61.2 ± 9.8 years, with 23.1% female patients, 89.6% having triple-vessel disease, and 29.8% having left main disease. The prolonged PCI hospitalization rate was significantly lower in the real-time heart team group compared to the traditional heart team group (15.5% vs. 31.0%, RR = 0.49, 95% CI: 0.34 to 0.69,P < 0.001). During the one-year follow-up, the incidence of major adverse cardiovascular and cerebrovascular events (MACCE) was 7.0% in the real-time heart team group and 9.8% in the traditional group, with a risk difference of -2.86% (95% CI: -2.13% to 7.92%). The non-inferiority test demonstrated that the real-time group was non-inferior to the traditional group (P < 0.001). Compared with the traditional heart team, real-time heart team decision-making showed no significant differences in decision appropriateness, actual treatment received by patients, or adherence to decisions. However, specialist workload, patient treatment waiting time, and hospital length of stay were significantly reduced in the real-time group (P < 0.05).

Conclusion

For patients with complex CAD, real-time heart team decision-making significantly improved clinical efficiency, reduced the rate of prolonged PCI hospital stay, optimized decision-making appropriateness, and reduced specialist workload while maintaining non-inferior clinical outcomes compared to the traditional heart team approach.

Part II. Comparative Study of the Effectiveness of Multidisciplinary Heart Team Decision-making Versus Real-world Single-discipline Decision-making for Patients with Complex Coronary Artery Disease.

Objective

Current guidelines recommend that patients with complex CAD should be discussed by a multidisciplinary heart team to achieve optimal treatment decisions. However, single-disciplinary decision-making remains widely practiced in clinical settings. Previous studies have primarily focused on specific aspects of heart team implementation, lacking a comprehensive comparison with real-world single-disciplinary decision-making. This study was based on data from a clinical trial and a real-world single-discipline decision-making cohort study to systematically evaluate the impact of multidisciplinary versus single-discipline decision-making models on decision quality and clinical outcomes.

Methods

This study was a retrospective cohort study, conducted based on data from the “Safety and Feasibility of a Real-Time Heart Team Decision-Making Approach for Patients with Complex Coronary Artery Disease” (multidisciplinary decision-making group) and the "Chinese Revascularization Appropriateness Evaluation Study" (single-discipline decision-making group). The "Chinese Revascularization Appropriateness Evaluation Study" was a prospective, multicenter cohort study that included patients with stable CAD who underwent elective coronary angiography between 2016 and 2017. It collected real-world data on patients' treatment preferences, interventional cardiologists' treatment recommendations, actual treatments received, and clinical outcomes. This study included patients with complex CAD confirmed by angiography from both cohorts. A propensity score matching (1:1) approach was used to balance baseline characteristics between the multidisciplinary heart team group and the single-discipline decision-making group. The primary endpoint was decision appropriateness. Secondary endpoints included decision outcomes, patients' actual treatment, hospital length of stay, efficiency-related metrics, and MACCE and their components during the one-year follow-up.

Results

A total of 1,090 patients with complex CAD were included, with 302 patients in each group after propensity score matching (mean age: 59.7 ± 9.7 years, male: 78.3%). Baseline characteristics were well balanced between the two groups. The results showed that the decision appropriateness rate was significantly higher in the multidisciplinary heart team group than in the single-discipline decision-making group (70.2% vs. 48.0%, χ²=30.748, P < 0.001). The actual treatment received by patients differed significantly between the two decision-making modes, with the multidisciplinary heart team group demonstrating a higher rate of coronary artery bypass grafting (CABG) procedures and improved treatment appropriateness compared to the single-discipline group. Hospital length of stay was significantly shorter in the multidisciplinary heart team group than in the single-discipline group (3days [2-7] vs. 5 days [3-12]). However, during the one-year follow-up, there was no significant difference in the risk of MACCE between the two groups (6.7% vs. 11.6%, HR=0.70; 95% CI: 0.40 to 1.23, P = 0.211).

Conclusion

Multidisciplinary heart team decision-making significantly improved decision appropriateness, treatment appropriateness, and clinical efficiency. However, compared to single-discipline decision-making, multidisciplinary decision-making did not demonstrate a significant advantage in clinical outcomes, possibly due to the relatively short follow-up duration.

Part III. A Qualitative Study on Factors Influencing Revascularization Decision Changes in Patients with Complex Coronary Artery Disease Based on Heart Team Discussion Data.

Objective

Multidisciplinary heart team discussions are recommended by revascularization guidelines for decision-making in complex CAD. However, previous studies have primarily examined the overall impact of heart team as an intervention, with limited understanding of the decision-making process itself. Furthermore, controversy remains regarding which patients should be included in heart team discussions. This study aimed to analyze the key stages influencing decision changes and identify patient characteristics and key factors based on prior discussion transcripts from the "Comparative Study on Heart Team Decision Consistency in Complex Coronary Artery Disease".

Methods

This qualitative study was based on data from the "Comparative Study on Decision Consistency in Heart Teams for Complex Coronary Artery Disease", a randomized controlled trial conducted from October 2019 to March 2020, which enrolled 480 patients. Each patient was evaluated by four heart teams, consisting of six cardiac surgeons and six interventional cardiologists. Experts first made independent treatment decisions (e.g., PCI or CABG), stated their reasoning, participated in team discussions, and then made a second independent decision. Using data and discussion transcripts from the trial, this study quantitatively analyzed the characteristics of patients whose treatment decisions changed after discussion, as well as the frequency and disciplinary distribution of decision shifts. Thematic analysis was then applied to relevant discussion transcripts to identify key factors influencing the direction of decision changes, summarizing the themes and subthemes.

Results

A total of 480 complex CAD cases were included (mean age: 61.2 ± 9.1 years, male proportion: 75.6%) and 355 patients’ treatment decisions were altered. Compared to patients with unchanged decisions, patients with modified decisions were older, had higher body mass index (BMI), and had higher proportions of diabetes, cerebrovascular disease, triple-vessel disease and higher SYNTAX scores, indicating that patients with altered decisions had more complex clinical conditions. A total of 5,760 decisions were made during the heart team meetings, among which 693 decisions were changed after team discussion. Among all decision modifications, changes from PCI to CABG and from CABG to PCI were the most frequent, with surgeons exhibiting a higher frequency of decision changes compared to interventional cardiologists. Through thematic analysis, four key themes influencing decision changes were identified: (1) Patient characteristics; (2) Lesion characteristics; (3) Surgical benefits; (4) Long-term outcomes.

Conclusion

Patients whose decisions were altered during heart team discussions tended to have more complex clinical conditions. Decision modifications among specialists were common, and the primary factors influencing these changes could be categorized into four major themes: (1) patient characteristics, (2) lesion characteristics, (3) surgical benefits, and (4) long-term outcomes. Future prospective studies are needed to validate the hypotheses proposed in this study.