第一部分：非ST段抬高型急性冠脉综合征的残存风险评估：冠状动脉CT血管造影斑块量化分析

背景：即使是在有创血流储备分数的指导下进行最优的治疗策略选择，非ST段抬高型急性冠状动脉综合征（non–ST-segment elevation acute coronary syndrome，NSTE-ACS）患者在未来发生主要心血管不良事件（major adverse cardiovascular events，MACEs）的风险仍是慢性冠状动脉综合征患者的两倍。对于NSTE-ACS患者，超过一半的MACEs和未血运重建的病变（残存病变）相关。有创腔内成像技术相关研究发现，如果冠状动脉斑块脂质含量高、斑块负荷重，则未来引发MACEs的风险将明显增加。冠状动脉CT血管造影（coronary CT angiography，CCTA）作为一种无创检查技术，也可以量化冠状动脉斑块成分，而且相较于腔内成像技术操作简便，更适合筛查高危的未血运重建斑块。对于NSTE-ACS患者，关于CCTA斑块量化分析对未血运重建斑块相关的MACEs的预后评估价值尚缺少有力的证据。

目的：本前瞻性队列研究致力于证实NSTE-ACS患者富含脂质成分的未血运重建斑块更容易引发MACEs，CCTA可用来界定出这些高危的未血运重建的斑块。

材料和方法：本前瞻性多中心队列研究在国内五家医院从2017年11月到2019年1月纳入诊断为NSTE-ACS的患者。参与者在诊断为NSTE-ACS后立刻进行CCTA检查，随后在一天内进行有创冠状动脉造影检查。接诊医师盲于CCTA检查结果，根据现行指南决定最佳诊疗策略。本研究中，斑块成分的测量用以下CT值临界值：脂质成分(<30 HU)，纤维成分(30–350 HU)，钙化成分(≥350 HU)。未血运重建斑块定义为在CCTA图像中可肉眼识别且此次入院没有接受血运重建的粥样硬化斑块。测量每个未血运重建斑块的脂质、纤维、钙化成分的斑块容积和负荷。在参与者水平，斑块容积定义为所有未血运重建斑块容积之和，斑块负荷定义为未血运重建斑块容积之和除以所有相应冠状动脉区域的血管容积之和。对于没有未血运重建斑块的参与者，斑块容积和负荷定义为0。本研究终点定义为和未血运重建斑块相关的复合MACEs，包括心源性死亡、心肌梗死、血运重建、因不稳定心绞痛再次入院。Cox比例风险模型用于验证脂质容积和脂质负荷是未血运重建斑块相关的复合MACEs的独立预测因子。

结果：共有342名参与者的768个未血运重建斑块纳入分析。通过中位随访时间4.0年的随访， 82名参与者发生未血运重建斑块相关的MACEs，4年未血运重建斑块相关的MACEs率为23.9%（95% 可信区间: 19.1%-28.5%）。在斑块水平，90个未血运重建斑块被裁定为引发MACEs的斑块。相较于未发生未血运重建斑块相关MACEs的参与者，发生未血运重建斑块相关MACEs的参与者有更高的平均脂质容积（37.6 mm³ 比 7.3 mm³ , P < 0.001）和平均脂质负荷（12.8% 比3.8%，P < 0.001）。在斑块水平，引发未血运重建斑块相关MACEs的斑块也有更高的平均脂质容积（18.3 mm³ 比2.7 mm³ , P < 0.001）和平均脂质负荷（17.7%比2.8%, P < 0.001）。在参与者水平，脂质核心负荷是未血运重建斑块相关MACEs的独立预测因子（风险比, 1.06； 95%可信区间: 1.02-1.09; P = 0.001），最佳临界值为2.8%。在斑块水平，脂质负荷是未血运重建斑块相关MACEs的独立预测因子（风险比, 1.03； 95% 可信区间: 1.01-1.05; P = 0.01），最佳临界值为7.2%；脂质容积也是未血运重建斑块相关MACEs的独立预测因子（风险比, 1.02； 95% 可信区间: 1.00-1.03; P = 0.02），最佳临界值为10.1 mm³。

结论：对于NSTE-ACS患者，作为CCTA图像中反应斑块脂质成分含量的参数，脂质负荷增加有助于确定容易发生未血运重建斑块相关MACEs的患者和斑块。脂质容积增加也可以用于筛查易损斑快。

第二部分：非ST段抬高型急性冠脉综合征患者的有创冠状动脉造影的看门人：冠状动脉CT血管造影

背景：超过1/3的非ST段抬高型急性冠脉综合征（non-ST-segment elevation acute coronary syndrome，NSTE-ACS）患者没有阻塞性冠状动脉疾病，这些患者承受了来自本可以避免的有创冠状动脉造影（invasive coronary angiography，ICA）的风险。另外， 32.5%的NSTE-ACS患者伴有冠状动脉慢性完全闭塞性病变（chronic total occlusion, CTO），冠状动脉CT血管造影（coronary CT angiography ，CCTA）可以判断冠状动脉狭窄，也可以为经皮冠状动脉介入术（percutaneous coronary intervention ，PCI）提供术前信息，判断CTO病变的PCI成功率。

目的：本文的研究目的是调查CCTA在NSTE-ACS患者中筛查出冠状动脉节段水平管腔狭窄＜70% 的鲁棒性，并调查CCTA对伴有CTO的NSTE-ACS患者能否提供额外的信息以减少PCI手术失败率。

材料和方法：本研究是一项对前瞻性多中心队列的事后分析研究，研究群体为从2017年12月到2019年1月在国内多中心纳入诊断为NSTE-ACS的患者组成的前瞻性队列。在诊断为NSTE-ACS后，参与者立即进行CCTA检查，CCTA的检查结果盲于诊疗医师。ICA检查在CCTA检查后的1天内进行。根据欧洲心脏病学会指南，参与者被分为高危组和低危组。将钙化积分≥400的参与者归为高钙化组，钙化积分＜400参与者归为低钙化组。节段水平冠状动脉狭窄的评估均使用国际心血管CT协会提出的冠状动脉18节段模型进行评估。将每个冠状动脉节段狭窄阳性定义为直径狭窄 ≥ 70%，以ICA为标准判断CCTA的准确性。根据冠状动脉节段水平的狭窄程度数据，整合血管水平和患者水平的冠状动脉狭窄程度。使用灵敏度、特异度、阳性预测值（positive predictive value ，PPV）、阴性预测值（negative predictive value，NPV）来评估CCTA的诊断能力。CCTA诊断狭窄的NPV代表了CCTA能够筛查出没有明显冠状动脉狭窄的能力。CTO病变以远的冠状动脉节段不再进行冠状动脉狭窄程度分析，但用于可识别性分析。CTO病变以远的冠状动脉节段只要有对比剂填充既认为该节段可识别。CCTA和ICA判断CTO以远可识别节段均使用上述标准。

结果：本队列共有347名 NSTE-ACS 参与者的资料可用于本事后分析研究。其中，高危组196人，低危组151人。18名参与者因有PCI手术史，未计算钙化积分。剩余329名参与者中，48名归为高钙化组，281名归为低钙化组。在参与者水平，ICA中出现冠状动脉狭窄＜70%的频率为25.9%。CCTA的NPV为91.3%。7名假阴性NSTE-ACS参与者均为单支病变。高危组和低危组之间，高钙化组和低钙化组之间，CCTA的诊断能力参数均没有显著差异。共有5054 个冠状动脉节段用于节段水平CCTA诊断狭窄能力的判断，前降支近段和中段最容易发生冠状动脉狭窄，狭窄发病率均为32.3%。对冠状动脉18节段模型中所有节段，CCTA判断冠状动脉狭窄的NPV非常稳健（范围：94.7% - 100.0%）。CCTA的特异度也很稳健，在冠状动脉18节段模型所有节段中的范围为94.0% - 100.0%。灵敏度范围为47.6% - 100.0%，PPV范围为58.3% - 100.0%。在纳入的347名NSTE-ACS参与者中，87（25.1%）名参与者具有CTO病变，共有93个CTO病变。CCTA较ICA识别了更多CTO病变以远的冠状动脉节段 (71.0%比48.8%)。92.0%的CTO患者接受PCI治疗，但31.3%的患者PCI未成功。诊疗医师查看CCTA图像后，在患者第二次入院PCI治疗时，80% PCI失败的患者成功进行了PCI治疗。

结论：在我们的队列中约有四分之一的NSTE-ACS患者的冠状动脉狭窄程度 <70%，另外约有四分之一的NSTE-ACS患者伴有CTO病变。CCTA 可以稳健的界定出冠状动脉狭窄程度< 70% 的NSTE-ACS患者，使其免于ICA检查，还可以为伴有CTO病变的患者提供额外更精准的信息。因此，约有一半NSTE-ACS患者将从ICA前的CCTA检查获益。

第三部分：冠状动脉CT血管造影斑块量化分析对非ST段抬高型急性冠脉综合征和慢性冠脉综合征的鉴别能力

背景：慢性冠状动脉综合征（chronic coronary syndrome，CCS）出现冠状动脉斑块急性破裂或侵蚀的时候，可以引发非ST段抬高型急性冠状动脉综合征（non-ST-segment elevation acute coronary syndrome，NSTE-ACS）。明确NSTE-ACS患者罪犯斑块和CCS患者斑块的差异，对了解NSTE-ACS发生机制、对NSTE-ACS患者精准化治疗都具有重要意义。冠状动脉CT血管造影（coronary CT angiography，CCTA）是可以量化冠状动脉斑块成分的无创检查技术，其量化斑块的准确性也在很多以腔内成像技术为标准的研究中被证实。

目的：本文致力于比较NSTE-ACS患者罪犯病变和CCS患者狭窄最重病变之间的斑块差异，以期用于NSTE-ACS和CCS的鉴别诊断。

材料和方法：本研究是一项回顾性病例对照研究，实验组来自从2017年12月到2019年1月在国内多中心组建的NSTE-ACS患者前瞻性队列。该队列患者在诊断为NSTE-ACS后均立即进行CCTA检查。因为慢性闭塞性病变中含有大量陈旧性血栓成分，目前CCTA分析软件尚不能将冠状动脉粥样硬化斑块和陈旧性血栓鉴别开，实验组排除队列中伴有慢性闭塞性病变的参与者。对照组纳入进行CCTA检查的CCS患者，回顾性筛选2023年9月到2024年2月在一所三甲医院进行门诊CCTA检查的CCS患者。根据患者年龄和性别，1:1匹配实验组（NSTE-ACS组）和对照组（CCS组）。实验组的患者选取引发NSTE-ACS的罪犯病变进行测量，对照组的患者选取狭窄最重的病变进行测量。斑块成分的测量用以下CT值临界值：脂质成分(<30 HU)，纤维成分(30–350 HU)，钙化成分(≥350 HU)。测量每个冠状动脉斑块的脂质、纤维、钙化成分的斑块容积和斑块负荷，总斑块容积和斑块负荷，冠状动脉斑块区域管腔狭窄，斑块长度，冠状动脉开口到斑块的距离。记录冠状动脉斑块所在的血管位置。比较两组之间斑块参数的差异性。使用斑块区域管腔狭窄作为判断NSTE-ACS的基线模型（模型1）。将多变量Logistic回归分析中有意义的CCTA冠状动脉斑块参数添加到基线模型，组成新模型（模型2）。比较两个模型判断NSTE-ACS的性能。

结果：实验组共纳入258名NSTE-ACS患者。根据年龄和性别，经过1:1倾向性评分匹配后，实验组和对照组完全匹配，实验组和对照组的平均年龄均为58.8±11.1岁，男性均占比71.7%（185/258）。NSTE-ACS患者罪犯病变的狭窄程度高于CCS患者狭窄最重病变（71.9% 比 37.4%）。在斑块成分和含量的比较中，NSTE-ACS组罪犯病变的脂质容积（9.8 mm³ 比 0.4 mm³）和脂质负荷（5.0% 比 0.3%）、纤维容积（50.2 mm³ 比 19.9 mm³）和纤维负荷（36.0% 比 15.0%）、总斑块容积（84.2 mm³ 比 37.8 mm³）和总斑块负荷（52.9% 比 30.9%）均高于CCS组的最重狭窄病变，钙化容积（0.5 mm³ 比 3.5 mm³）和钙化负荷（0.3% 比 2.8%）均低于CCS组的最重狭窄病变。在多变量Logistic回归分析中，纤维负荷、钙化负荷、狭窄程度均为NSTE-ACS的独立判断因子。纤维负荷、钙化负荷、狭窄程度的最佳临界值分别为23.3%、16.8%、44.6%。基线模型1的受试者特征曲线下面积（area under the curve, AUC）为0.699（0.654 – 0.754）。管腔狭窄、纤维负荷、钙化负荷组成的模型2的AUC为0.782 （0.741 - 0.822），两个模型的AUC之差为0.083，P ＜0.001。

结论：CCTA图像的斑块成分量化信息可用于诊断NSTE-ACS。纤维负荷、钙化负荷、狭窄程度均为NSTE-ACS患者罪犯病变的独立判断因子，这些判断因子组成的模型提高了基线模型对NSTE-ACS患者罪犯病变的诊断能力。

Part 1：Residual risk in non–ST-segment elevation acute coronary syndrome: quantitative plaque analysis at coronary CT angiography

Backgrounds: Even under the guidance of invasive fractional flow reserve, the incidence of major adverse cardiovascular events (MACEs) in the future in patients with non-ST-segment elevation acute coronary syndrome (NSTE-ACS) is still approximately twice that of patients with chronic coronary syndrome. For NSTE-ACS patients, over half of MACEs are associated with lesions without revascularization. Researches on invasive intravascular imaging technique have found that if the lipid content of plaques and the plaque burden is high, the risk of triggering MACEs in the future will significantly increase. As a non-invasive technique, coronary CT angiography (CCTA) can also quantify the constituents of coronary artery plaques, and it is easier to operate than invasive intravascular imaging technique. There is still a lack of evidence regarding the prognostic value of plaque constituents quantification in CCTA for nonrevascularized plaques in NSTE-ACS patients.

Objective: The prospective cohort study aims to confirm that nonrevascularized plaques with high lipid core constituents in NSTE-ACS patients are more likely to trigger MACEs, and CCTA can be used to define these high-risk nonrevascularized plaques.

Materials and Methods: The prospective multicenter cohort study included patients diagnosed with NSTE-ACS in five hospitals in China from November 2017 to January 2019. Participants immediately underwent CCTA examination after being diagnosed as NSTE-ACS, followed by invasive coronary angiography examination within one day. Cardiologists were blinded to the CCTA examination results and decided the optimal treatment strategy based on current guidelines. In this study, the measurement of plaque constituents used the following CT threshold values: lipid constituents (<30 HU), fibrous constituents (30-350 HU), and calcified constituents (≥ 350 HU). Nonrevascularized plaque was defined as atherosclerotic plaque that can be identified by radiologists in CCTA images and did not receive revascularization during hospitalization. The plaque volume and burden of lipid, fibrous, and calcified constituents were measured for all nonrevascularized plaques. At the participant level, plaque volume was defined as the sum of all nonrevascularized plaque volumes in the patient, and plaque burden was defined as the sum of nonrevascularized plaque volumes divided by the sum of vessel volumes of corresponding coronary artery regions. For participants without nonrevascularized plaques, plaque volume and burden were defined as 0. The endpoint of the study was defined as composite MACEs associated with nonrevascularized plaques, including cardiac death, myocardial infarction, revascularization, and readmission due to unstable angina. The Cox proportional hazards regression model was used to validate that lipid volume and lipid burden are independent predictors of composite MACEs associated with nonrevascularized plaques.

Results: A total of 768 nonrevascularized plaques from 342 participants were included in the analysis. Through a median follow-up time of 4.0 years, 82 participants experienced MACEs related to nonrevascularized plaques, with a 4-year MACEs rate of 23.9% (95% confidence interval: 19.1-28.5). At the plaque level, 90 nonrevascularized plaques which triggered MACEs were identified. Compared to participants who did not experience nonevascularized plaque-related MACEs, participants with nonevascularized plaque-related MACEs had higher lipid volume (37.6 mm³ vs. 7.3 mm³, P < 0.001) and lipid burden (12.8% vs. 3.8%, P < 0.001). At the plaque level, nonevascularized plaques that triggered MACEs also had higher lipid volume (18.3 mm³ vs. 2.7 mm³, P < 0.001) and lipid burden (17.7% vs. 2.8%, P < 0.001). At the participant level, lipid burden was an independent predictor of nonevascularized plaque-related MACEs (hazard ratio, 1.06; 95% confidence interval: 1.02-1.09; P = 0.001), with an optimal threshold value of 2.8%. At the plaque level, lipid burden was an independent predictor of nonevascularized plaque-related MACEs (hazard ratio, 1.03; 95% confidence interval: 1.01-1.05; P = 0.01), with an optimal threshold value of 7.2%. Lipid volume was also an independent predictor of nonevascularized plaque-related MACEs (hazard ratio, 1.02; 95% confidence interval: 1.00-1.03; P = 0.02), with an optimal threshold value of 10.1 mm³.

Conclusion: As a variable associated with lipid-rich plaque in coronary CT angiography, increased lipid core burden can help to identify NSTE-ACS patients and plaques at high risk for subsequent nonrevascularized plaque-related MACEs. Vulnerable plaques could also be predicted by high lipid core volume.

Part 2: Gatekeeper for invasive coronary angiography in patients with non-ST-segment elevation acute coronary syndrome: coronary CT angiography

Backgrounds: More than one-third of non-ST-segment elevation acute coronary syndrome (NSTE-ACS) patients do not have obstructive coronary artery disease, and these patients are at risk from invasive coronary angiography (ICA) that could have been avoided. In addition, 32.5% of NSTE-ACS patients have coronary artery chronic total occlusion (CTO). Coronary CT angiography (CCTA) could identify coronary stenosis, provide preoperative information for percutaneous coronary intervention (PCI), and predict the success rate of PCI.

Objectives: The purposes of this study were to investigate the robustness of CCTA in screening coronary artery luminal stenosis<70% in segment level in NSTE-ACS patients, and to investigate whether CCTA can provide additional information to reduce the failure rate of PCI surgery in NSTE-ACS patients with CTO.

Materials and methods: This study was a post hoc analysis of a prospective multicenter cohort consisting of patients diagnosed with NSTE-ACS from December 2017 to January 2019 in multiple centers in China. After being diagnosed with NSTE-ACS, participants immediately underwent CCTA examination, and the results of CCTA examination were blinded to the cardiologists. ICA examination was conducted within 1 day after CCTA examination. According to the guidelines of the European Heart Association, participants were divided into high-risk and low-risk groups. Participants with a calcification score ≥ 400 were classified as the high calcification group, while those with a calcification score<400 were classified as the low calcification group. The evaluation of segmental coronary artery stenosis was performed using the 18-segment coronary artery model proposed by the Society of Cardiovascular Computed Tomography. Each positive coronary artery segment stenosis is defined as a diameter stenosis of ≥ 70%, and the accuracy of CCTA for diagnosing coronary artery stenosis is evaluated using ICA as the standard. The degree of coronary artery stenosis at the vascular and participant levels were recorded based on the data of the degree of coronary artery stenosis at the segmental level. Use sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) to evaluate the diagnostic ability of CCTA. The NPV of CCTA for diagnosing coronary artery stenosis represents the ability of CCTA to screen for non-significant coronary artery stenosis. The coronary artery segments further from the CTO lesion will no longer be analyzed for the degree of coronary artery stenosis, but will be used for recognizability analysis. The coronary artery segment further away from the CTO lesion was considered identifiable as long as there was contrast agent filling. Both CCTA and ICA used the above criteria to determine identifiable segments further from CTO.

Results: A total of 347 NSTE-ACS participants in this cohort available for this post hoc analysis study. Among them, there were 196 participants in the high-risk group and 151 participants in the low-risk group. We did not calculate calcification scores of 18 participants due to a history of PCI. Among the remaining 329 participants, 48 were classified as high calcification group and 281 were classified as low calcification group. At the participant level, the frequency of coronary artery stenosis<70% in ICA was 25.9%. The NPV of CCTA is 91.3%. All of seven false negative NSTE-ACS participants had single vessel significant lesions. There was no significant difference in the diagnostic ability parameters of CCTA between the high-risk group and the low-risk group, as well as between the high calcification group and the low calcification group. A total of 5054 coronary segments were used to judge the ability of CCTA to diagnose stenosis at the segment level. The proximal and middle segments of the left anterior descending artery were most prone to have coronary artery stenosis, and the stenosis incidence rate were both 32.3% for the proximal and middle segments of the left anterior descending artery. In the 18-segment coronary artery model, the NPV of CCTA for coronary artery stenosis was very robust (range: 94.7% -100.0%). The specificity of CCTA was also very robust, ranging from 94.0% to 100.0% in all segments of the 18-segment coronary artery model. The sensitivity range was 47.6% -100.0%, and the PPV range was 58.3% -100.0%. Among the 347 NSTE-ACS participants, 87 (25.1%) participants had CTO lesions. A total of 93 CTO lesions were identified. CCTA identified more coronary artery segments further away from CTO lesions than ICA (71.0% vs. 48.8%). 92.0% of CTO participants received PCI treatment, but 31.3% of them had unsuccessful PCI. After reviewing CCTA images by cardiologists, 80% of participants who had unsuccessful PCI during index hospitalization successfully underwent PCI treatment during their second admission.

Conclusion: About a quarter of NSTE-ACS patients in our cohort had coronary artery stenosis<70%, and approximately a quarter of NSTE-ACS patients had CTO lesions. CCTA can robustly identify NSTE-ACS patients with coronary artery stenosis degree<70% who can avoid ICA examination. CCTA can provide additional and more accurate information for patients with CTO lesions. Therefore, approximately half of NSTE-ACS patients will benefit from upfront CCTA before ICA.

Part:3：The discriminative ability of CCTA plaque measurement for non-ST-segment elevation acute coronary syndrome and chronic coronary syndrome

Backgrounds: When acute rupture or erosion of coronary artery plaques occurs in patients with chronic coronary syndrome (CCS), it can lead to non-ST-segment elevation acute coronary syndrome (NSTE-ACS). In order to better understand the mechanism of NSTE-ACS occurrence and precise treatment of NSTE-ACS, it will be of great significance by clarifying the differences between culprit plaques in NSTE-ACS patients and plaques with most severe stenosis in CCS patients. As a non-invasive examination technique, coronary CT angiography (CCTA) can quantify the composition of coronary artery plaques, and its accuracy in quantifying plaques has been confirmed in many studies with intravascular imaging as reference.

Objective: The study aims to distinguish NSTE-ACS from CCS through comparing the plaque differences between the culprit lesion in NSTE-ACS patients and the most severe stenosis plaque in CCS patients in CCTA imaging.

Materials and methods: This study was a retrospective case control study, with case group from a prospective NSTE-ACS cohort established in multiple centers in China from December 2017 to January 2019. All patients in this cohort underwent CCTA examination immediately after being diagnosed with NSTE-ACS. Because coronary chronic total occlusion contains chronic thrombus components, and the current CCTA plaque analysis software cannot distinguish coronary atherosclerotic plaque from thrombus, the NSTE-ACS group excluded participants with chronic total occlusion. The control pair included CCS patients who underwent outpatient CCTA examination in a tertiary hospital from September 2023 to February 2024. Patients with NSTE-ACS and patients with CCS were propensity matched 1:1 for age and gender. The culprit lesion in NSTE-ACS patients and the most severe stenosis lesion in control pair were used for measurement. The measurement of plaque components used the following CT threshold: lipid constituents (<30 HU), fibrous constituents (30-350 HU), and calcified constituents (≥ 350 HU). The plaque volume and plaque burden of lipid, fibrous, and calcified constituents of each coronary artery plaque, as well as the total plaque volume and plaque burden were measured. The luminal stenosis, plaque length, distance from the coronary artery ostium to the plaque were also measured. The location of the coronary artery plaque was recorded. Compare the differences in plaque parameters between NSTE-ACS group and CCS group. Using luminal stenosis as the baseline model for assessing NSTE-ACS (Model 1). Add CCTA coronary artery plaque parameters with statistical significance from multivariate logistic regression analysis to the baseline model to form a new model (Model 2). Compare the performance for diagnosing NSTE-ACS of the two models.

Results: A total of 258 NSTE-ACS patients were included in the study. Based on age and gender, after a 1:1 propensity score matching, the NSTE-ACS group and CCS group were completely matched. The average age of both groups was 58.8 ± 11.1 years, with males accounting for 71.7% (185/258). The degree of stenosis of culprit lesions in NSTE-ACS patients is higher than that of the most severe stenosis lesions in CCS patients (71.9% vs. 37.4%). In the comparison of plaque composition and content, the lipid volume (9.8 mm³ vs. 0.4 mm³) and burden (5.0% vs. 0.3%), fibrous volume (50.2 mm³ vs. 19.9 mm³) and burden (36.0% vs. 15.0%), total plaque volume (84.2 mm³ vs. 37.8 mm³) and burden (52.9% vs. 30.9%) of the culprit lesions in the NSTE-ACS group were higher than the most severe stenosis lesions in the CCS group. The calcification volume (0.5 mm³ vs. 3.5 mm³) and burden (0.3% vs. 2.8%) was lower in the NSTE-ACS group. In multivariate logistic regression analysis, fibrous burden, calcified burden, and luminal stenosis were independent predictive factors for NSTE-ACS. The optimal threshold values for fibrous burden, calcified burden, and luminal stenosis were 23.3%, 16.8%, and 44.6%, respectively. The AUC (area under the curve) of model 1 was 0.699 (0.654-0.754). The AUC of Model 2 composed of stenosis, fibrous burden, calcified burden was 0.782 (0.741-0.822), and the difference in AUC between the two models was 0.083, P<0.001. 

Conclusion: The quantitative information of plaque constituents in CCTA images can be used to diagnose NSTE-ACS. Fibrous burden, calcified burden, and stenosis are independent predictor factors for culprit lesions in NSTE-ACS patients. The model composed of these factors improves the diagnostic ability of the baseline model for culprit lesions in NSTE-ACS patients.