第一部分 甘油三酯与C反应蛋白水平联合对普通人群不良心血管预后的预测价值

【目的】探索普通人群中甘油三酯（triglyceride，TG）联合C反应蛋白（C-reactive protein，CRP）水平对于心血管不良预后的价值。

【方法】本研究纳入的研究对象来自于美国国家健康和营养调查（National Health and Nutrition Examination Survey，NHANES）。采集的研究对象基线特征指标主要包括：吸烟史、饮酒史、体力活动、自我报告的疾病史、收缩压、体重指数（body mass index，BMI）、TG、CRP、高密度脂蛋白胆固醇（high-density lipoprotein cholesterol，HDL-C）、低密度脂蛋白胆固醇（low-density lipoprotein cholesterol，LDL-C）、总胆固醇（total cholesterol，TC）、糖化血红蛋白（glycohemoglobin，HbA1c）、估算肾小球滤过率（estimated glomerular filtration rate，eGFR）。主要观察终点为全因死亡，次要终点为心血管死亡。本研究将CRP＞0.3mg/dL定义为高CRP，TG＞2.3mmol/L定义为高TG。根据TG和CRP将研究人群分为四组：组Ⅰ（低CRP低TG）、组Ⅱ（低CRP高TG）、组Ⅲ（高CRP低TG）、组Ⅳ（高CRP高TG）。

【结果】本研究最终纳入NHANES（2001-2010）共11293例受访者，中位随访时间为159.0月。将TG和CRP作为连续变量纳入到使用限制性立方样条（restricted cubic spline，RCS）的Cox单因素及多因素回归分析，结果显示：在单因素分析中，TG和CRP水平均与全因死亡及心血管死亡风险具有非线性相关性；在多因素分析中，TG与全因死亡及心血管死亡风险无独立相关性，CRP与全因死亡及心血管死亡均独立相关，与全因死亡仍保持非线性关系。分组后单因素Cox回归分析提示，从组Ⅰ至组Ⅳ的全因死亡及心血管死亡风险逐渐显著增加，后三组相对于组Ⅰ的HR分别为1.456（95% CI：1.181-1.795）、1.580（95% CI：1.434-1.741）、2.063（95% CI：1.690-2.518）。但在调整了基本人口学特征[年龄、性别、种族、贫困收入比值（poverty income ratio，PIR）、教育程度，即模型1]，以及在模型1的基础上进一步调整其他多因素[心血管疾病（cardiovascular disease，CVD）、心力衰竭、卒中、BMI分组、吸烟状态、体力活动、LDL-C、HbA1c，即模型2]后，组Ⅱ相对组Ⅰ的全因死亡和心血管死亡风险升高均无统计学意义，调整后两个模型的HR分别为1.105（95% CI：0.898-1.360）、1.020（95% CI：0.744-1.398）；组Ⅲ和组Ⅳ与组Ⅰ相比全因死亡及心血管死亡风险均逐渐升高，组Ⅲ调整两个模型后的HR分别为1.306（95% CI：1.177-1.448）、1.277（95% CI：1.110-1.469），组Ⅳ的HR分别为1.652（95% CI：1.362-2.003）、1.532（95% CI：1.178-1.991）。TG和CRP联合分组的加入显著提高了传统预测模型对全因死亡（NRI：8.6%, P=0.030, IDI 0.2%, P<0.001）及心血管死亡（NRI：12.2%，P<0.001，IDI 0.5%，P<0.001）的预测效能。

【结论】普通人群CRP水平升高时，高TG水平与不良预后包括全因死亡及心血管死亡风险独立相关，高TG合并高CRP水平显著增加普通人群不良预后风险，TG联合CRP对普通人群不良预后具有重要预测价值。将TG联合CRP分组纳入传统模型可显著提高其对全因死亡及心血管死亡的预测效能。

第二部分 甘油三酯与C反应蛋白水平对冠心病患者预后的预测价值

【目的】在冠心病（coronary artery disease，CAD）患者队列中，通过探索甘油三酯（triglyceride，TG）与超敏C反应蛋白（high-sensitivity C-reactive protein，hsCRP）水平与不良心血管事件的相关性，为CAD二级预防提供思路和线索。

【方法】纳入2016年1月至2019年11月就诊于中国医学科学院阜外医院、诊断为冠状动脉粥样硬化性心脏病且年龄≥18岁的住院患者。排除标准：（1）既往恶性肿瘤病史；（2）严重肝肾功能不全（转氨酶水平＞10倍正常上限，血肌酐水平＞442μmol/L）；（3）住院时处于急性感染期；（4）自身免疫病活动期或使用糖皮质激素药物治疗；（5）严重心功能不全（左室射血分数＜30%）。采集患者的人口学数据、病史、药物治疗情况及空腹血TG、hsCRP及其他血生化指标。研究的主要终点为主要不良心血管事件（major adverse cardiovascular events，MACE），即由全因死亡、非致命性心肌梗死、非致命性卒中及计划外血运重建组成的复合终点，次要终点为全因死亡。

【结果】最终纳入分析1604例患者，中位随访时间为4.88年，共发生354例MACE，包括67例全因死亡、36例非致命性心肌梗死、33例非致命性卒中和218例计划外血运重建。初步统计学分析提示hsCRP与全因死亡风险独立相关（HR=1.083，95% CI：1.022-1.147）。hsCRP水平升高显示出使MACE风险增加的趋势（HR=1.026，95% CI：0.993-1.060），但未达到统计学意义（P=0.123)；TG与MACE和全因死亡均未显示出独立相关性。基于Cox回归的RCS分析提示TG、hsCRP均与MACE呈非线性相关，求得TG的潜在切点为3.81mmol/L。以TG≥3.9mmol/L作为界值分组，并与hsCRP水平纳入到多因素回归模型中，结果提示TG≥3.9mmol/L与MACE风险升高显著相关（HR=1.624，95% CI：1.022-2.582），hsCRP水平显示出了使MACE风险升高的趋势（HR=1.025，95% CI：0.992-1.059），但无统计学意义。使用倾向性匹配分析针对性别（男性）、年龄、目前吸烟、收缩压、BMI及经皮冠状动脉介入治疗（percutaneous coronary intervention， PCI）史进行1：2匹配（TG≥3.9mmol/L vs. TG＜3.9mmol/L），匹配后的队列生存结果提示TG≥3.9mmol/L与MACE风险升高相关（P=0.022），与全因死亡风险无显著相关性（P=0.4）。

【结论】hsCRP与CAD患者全因死亡风险独立相关；当TG≥3.9mmol/L时，无论hsCRP水平高低，TG水平升高均与CAD患者主要不良心血管事件风险独立相关，因此在接受有效他汀治疗后TG≥3.9mmol/L时，可能需要积极应用降甘油三酯药物进一步减少CAD患者的残余风险。

第三部分 冠状动脉斑块负荷指标与甘油三酯和高敏C反应蛋白的相关性及其对冠心病患者经皮冠状动脉介入治疗术后预后的预测价值

【目的】在接受经皮冠状动脉介入（percutaneous coronary intervention，PCI）治疗的冠心病患者中，探索术前冠状动脉计算机断层血管成像（coronary computed tomography angiography，CCTA）所得到的斑块负荷指标与甘油三酯（triglyceride，TG）与超敏C反应蛋白（high-sensitivity C-reactive protein，hsCRP）水平的相关性，并探讨术前CCTA得到的斑块负荷指标与PCI术后不良心血管结局的关系。

【方法】本研究为回顾性队列研究，纳入了2010年1月1日至2022年11月30日期间于中国医学科学院阜外医院行首次PCI治疗且术前接受CCTA检查的冠心病患者。收集患者人口学数据、病史、药物治疗及TG、hsCRP及其他生化检查等信息，CCTA 影像由64层以上计算机断层成像（computed tomography，CT）扫描，使用Deep Blue工作站分析患者CCTA影像数据。斑块识别和测量由工作站根据Hounsfield单位（Hounsfield unit，HU）值自动进行，分为低衰减斑块 （<30 HU）、纤维斑块（30-150 HU）、钙化斑块（>350 HU）和管腔（150-350 HU）并测量各组成部分体积。分别计算总斑块体积、总斑块体积比、总低衰减斑块体积、总低衰减斑块占比、总纤维斑块体积、总纤维斑块占比、总钙化斑块体积和总钙化斑块占比。主要观察终点为主要不良心血管事件（major adverse cardiovascular events，MACE）, 即包含全因死亡、非致命性心肌梗死、卒中和非计划血运重建的复合终点。

【结果】最终纳入分析患者230例。TG升高（≥1.7mmol/L）与总斑块体积、总斑块体积比、总低衰减斑块体积、总纤维斑块体积升高显著相关，与总低衰减斑块占比、总纤维斑块占比和总钙化斑块体积及占比无显著相关性；同样，hsCRP升高（≥2mg/L）与总斑块体积、总斑块体积比、总低衰减斑块体积、总纤维斑块体积升高显著相关，与总低衰减斑块占比、总纤维斑块占比和总钙化斑块体积及占比无显著相关性。

患者平均随访时间为4.8年，共发生67例MACE，包括6例全因死亡、9例非致命性心肌梗死、6例非致命性卒中和46例计划外血运重建。单因素分析结果显示总斑块体积、总斑块体积比、总低衰减斑块体积、总纤维斑块体积与MACE风险显著相关（P<0.05）；总低衰减斑块占比、总纤维斑块占比与MACE风险无显著相关性；总钙化斑块体积及占比与MACE风险均无显著相关性。在多因素Cox分析中，通过多模型调整协变量后，总斑块体积、总斑块体积比、总低衰减斑块体积、总纤维斑块体积与MACE风险独立相关。与第1三分位区间相比，总斑块体积和总纤维斑块体积的第2三分位区间与MACE风险无显著相关性；与第1三分位相比，总斑块体积和总纤维斑块体积的第3三分位区间的风险比（hazard ratio，HR）分别为为2.06（95% CI：1.03-4.15）和2.23（95% CI：1.11- 4.46)。与第1三分位区间相比，总斑块体积比、总低衰减斑块体积第2三分位区间与MACE风险的HR分别为2.34（95% CI：1.17-4.67）和2.33（95% CI：1.12-4.81）。总低衰减斑块占比、总纤维斑块占比与MACE风险无显著相关性；总钙化斑块体积和占比与MACE风险也无显著相关性。使用RCS分析指出，部分冠状动脉斑块负荷相关指标，包括总斑块体积、总低衰减斑块体积、总纤维斑块体积与MACE呈非线性关系。

【结论】冠心病患者TG和hsCRP水平与PCI术前CCTA获得的斑块负荷指标，包括总斑块体积、总斑块体积比、总低衰减斑块体积及总纤维斑块体积，均具有相关性；上述指标与PCI术后不良预后风险独立相关。PCI术前CCTA可以评估血运重建术后患者的残余风险，对于总斑块负荷高的患者，需要进一步强化危险因素管理包括降低TG和改善炎症状态。

Part Ⅰ Association of Triglyceride and C-Reactive Protein Levels with Adverse Cardiovascular Outcomes in the General Population

【Objective】 To investigate the associations between triglyceride (TG) and C-reactive protein (CRP) levels and adverse cardiovascular outcomes in the general population.

【Methods】 The participants of this study were selected from the National Health and Nutrition Examination Survey (NHANES). The baseline characteristics of the included participants mainly included smoking history, alcohol consumption, physical activity, self-reported medical history, systolic blood pressure, body mass index (BMI), TG, CRP, high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), total cholesterol (TC), glycosylated hemoglobin (HbA1c), and estimated glomerular filtration rate (eGFR). The endpoints of the study were cardiovascular death and all-cause death. In this study, CRP > 0.3 mg/dL was defined as high CRP, and TG > 2.3 mmol/L was defined as high TG. According to TG and CRP levels, the study participants were divided into four groups: Group Ⅰ (low CRP and low TG), Group Ⅱ (low CRP and high TG), Group Ⅲ (high CRP and low TG), and Group Ⅳ (high CRP and high TG).

【Results】 A total of 11293 NHANES participants (2001-2010) were included in this study, and the median follow-up time was 159.0 (29.0-193.0) months. TG and CRP were included as continuous variables in Cox univariate and multifactorial regression analyses using restricted cubic spline (RCS). Univariate analysis revealed that both TG and CRP levels were nonlinearly correlated with the risk of all-cause death and cardiovascular death. According to the multivariate analysis, TG was not independently associated with the risk of all-cause death or cardiovascular death, and CRP was independently associated with all-cause death and cardiovascular death, and the relationship with all-cause death remained nonlinear. Univariate Cox regression analysis after grouping suggested that the risk of all-cause death and cardiovascular death gradually increased from group Ⅰ to group Ⅳ. And comparing to Group Ⅰ, the hazard ratios (HRs) of the three groups were 1.456 (95% CI: 1.181-1.795), 1.580 (95% CI: 1.434-1.741) and 2.063 (95% CI: 1.690-2.518), respectively. However, after adjusting for basic demographic characteristics (age, sex, race, poverty income ratio, education level; Model 1) and other factors (cardiovascular disease, heart failure status, stroke status, BMI, smoking status, physical activity status, LDL-C level, HbA1c level; Model 2), there was no statistically significant increase in the risk of all-cause death or cardiovascular death in Group II compared with Group Ⅰ. The HRs after adjusting the two models were 1.105 (95% CI: 0.898-1.360) and 1.020 (95% CI: 0.744-1.398), respectively. Compared with those in group Ⅰ, the risks of all-cause death and cardiovascular death in groups III and IV gradually increased. The HRs after adjusting the two models in group III were 1.306 (95% CI: 1.177-1.448) and 1.277 (95% CI: 1.110-1.469), respectively. And the HRs in group IV were 1.652 (95% CI: 1.362-2.003) and 1.532 (95% CI: 1.178-1.991), respectively. The addition of TG-CRP group significantly improved the predictive performance of the traditional model for all-cause death (NRI: 8.6%, P=0.030; IDI: 0.2%, P<0.001) and cardiovascular death (NRI: 12.2%, P<0.001; IDI: 0.5%, P<0.001).

【Conclusion】 In the general population, TG-CRP group was significantly associated with adverse outcomes, including cardiovascular death and all-cause death, and elevated TG levels were independently associated with the risk of adverse outcomes only when combined with elevated CRP levels. The inclusion of TG-CRP group in the traditional model can significantly improve its predictive power.

Part II: Effect of TG and CRP Levels on the Prognosis of Patients with Coronary Artery Disease

【Objective】 To investigate the association between triglyceride (TG) and high-sensitivity C-reactive protein (hsCRP) levels and the risk of adverse cardiovascular events in a cohort of patients with coronary artery disease (CAD) to provide clues for the secondary prevention of CAD.

【Methods】 Inpatients aged ≥18 years who were diagnosed with CAD between January 2016 and November 2019 were included. The exclusion criteria were as follows： (1) history of malignant tumor; (2) severe hepatic or renal dysfunction (aminotransferase level > 10 times the upper limit of normal, serum creatinine level > 442 μmol/L); (3) acute infection at the time of hospitalization; (4) active autoimmune disease or treatment with glucocorticoid drugs; and (5) severe cardiac insufficiency (left ventricular ejection fraction < 30%). Demographic data, medical history, drug treatment, and biochemical indicators were collected. The primary endpoint of the study was major adverse cardiovascular events (MACEs), a composite endpoint consisting of all-cause death, nonfatal myocardial infarction, nonfatal stroke, and unplanned revascularization. The secondary endpoint was all-cause death.

【Results】 During a median follow-up of 4.88 years, 354 MACEs occurred, including 67 all-cause deaths, 36 nonfatal myocardial infarctions, 33 strokes, and 218 unplanned revascularizations. Preliminary statistical analysis suggested that hsCRP was independently associated with the risk of all-cause death [hazard ratio (HR) =1.083, 95% CI: 1.022-1.147]. The level of hsCRP increased the risk of MACEs (HR=1.026, 95% CI: 0.993-1.060) but did not reach statistical significance (P=0.123). TG was not independently associated with MACEs or all-cause death. Restricted cubic spline (RCS) analysis indicated that TG and hsCRP were nonlinearly correlated with MACE, and the potential target point of TG was 3.81 mmol/L. TG concentration ≥3.9 mmol/L and hsCRP level were included in the multivariate regression model, and the results suggested that TG concentration ≥3.9 mmol/L was significantly associated with an increased risk of MACEs (HR=1.624, 95% CI: 1.022-2.582). hsCRP tended to increase the risk of MACEs (HR=1.025, 95% CI: 0.92-1.059), but the difference was not statistically significant. Propensity score matching was used to match sex (male), age, current smoking status, systolic blood pressure, BMI, and PCI history at a 1:2 ratio (TG ≥ 3.9 mmol/L vs. TG < 3.9 mmol/L). The survival analysis results of the matched cohort suggested that TG ≥ 3.9 mmol/L was associated with an increased risk of MACEs (P=0.022) but was not significantly associated with the risk of all-cause death (P=0.4).

【Conclusion】hsCRP was independently associated with the risk of all-cause death in CAD patients. When TG ≥3.9 mmol/L, regardless of the level of hsCRP, an increase in TG was independently associated with the risk of major adverse cardiovascular events in CAD patients. Therefore, when TG ≥3.9 mmol/L after receiving effective statin treatment, it may be necessary to actively apply TG-lowering drugs to further reduce the residual risk in CAD patients.

Part Ⅲ Coronary Plaque Burden Variables Correlated with Levels of Triglycerides and High-Sensitivity C-Reactive Protein and Associated with Prognosis after Percutaneous Coronary Intervention.

【Objective】This study aimed to investigate the correlation of levels of triglyceride (TG) and high-sensitivity C-reactive protein (hsCRP) with coronary plaque burden variables derived from coronary computed tomography angiography (CCTA) before patients underwent their first percutaneous coronary intervention (PCI) procedure, and the association of coronary plaque burden variables and major adverse cardiovascular events (MACEs) after PCI.

【Methods】 This retrospective cohort study included patients aged 18-90 years with CAD who received CCTA before their first PCI between January 1, 2010, and November 30, 2022. CCTA images were obtained via 64-layer computed tomography (CT) scans and above. Deep Blue workstation was used to retrospectively analyze the CCTA image data of the patients. Plaque identification and measurement were performed automatically by the workstation according to the Hounsfield unit (HU) value, and plaques were divided into low-attenuation plaques (<30 HU), fibrous plaques (30-150 HU), calcified plaques (>350 HU) and lumen (150-350 HU), and the volume of each component was measured. The total plaque volume, total percent atheroma volume, total low-attenuation plaque volume, total low-attenuation plaque fraction, total fibrous plaque volume, total fibrous plaque fraction, total calcified plaque volume and total calcified plaque fraction were calculated. The primary endpoint was major adverse cardiovascular events (MACEs), a composite endpoint that included all-cause death, nonfatal myocardial infarction, stroke, and unplanned revascularization.

【Results】 A total of 230 patients were included in the analysis. An increase in TG (TG≥1.7 mmol/L) was significantly correlated with an increase in total plaque volume, the total percent atheroma volume, total low-attenuation plaque volume, and total fibrous plaque volume but was not significantly correlated with the fraction of total low-attenuation plaque, the total fibrous plaque fraction or the total calcified plaque volume or fraction. Similarly, an increase in hsCRP (≥2 mg/L) was significantly correlated with an increase in total plaque volume, total percent atheroma volume, total low-attenuation plaque volume, and total fibrous plaque volume but was not significantly correlated with the fraction of total low-attenuation plaque, total fibrous plaque fraction, or total calcified plaque volume or fraction.

During a median follow-up of 4.8 years, 67 MACEs occurred, including 6 all-cause deaths, 9 nonfatal myocardial infarctions, 6 nonfatal strokes, and 46 unplanned revascularizations. Univariate analysis revealed that total plaque volume, total percent atheroma volume, total low-attenuation plaque volume, and total fibrous plaque volume were significantly correlated with the risk of MACEs (P< 0.05). The fractions of total low-attenuation plaque and total fibrous plaque were not significantly correlated with the risk of MACEs. The total calcified plaque volume and fractions were not significantly associated with MACEs. According to multivariate Cox analysis, total plaque volume, total percent atheroma volume, total low-attenuation plaque volume, and total fibrous plaque volume were independently associated with MACEs after adjusting for covariates in multiple models. Compared with the first tertiles, the second tertiles of total plaque volume and total fibrous plaque volume were not significantly associated with MACEs, and the hazard ratios (HRs) of their third tertiles were 2.06 (95% CI: 1.03-4.15) and 2.23 (95% CI: 1.11-4.46), respectively. Compared with those of the first tertiles, the HRs of the second tertiles of total percent atheroma volume and total low attenuation plaque volume were 2.34 (95% CI: 1.17-4.67) and 2.33 (95% CI: 1.12-4.81), respectively. The fractions of total low-attenuation plaque and total fibrous plaque were not significantly associated with MACEs. The total calcified plaque volume and fraction were not significantly associated with MACEs. Restricted cubic spline (RCS) analysis revealed that total plaque volume, total low-attenuation plaque volume, and total fibrous plaque volume exhibited a nonlinear relationship with MACE.

【Conclusion】Selected pre-PCI CCTA-derived variables, including total percent atheroma volume, total plaque volume, total low-attenuation plaque volume, and total fibrous plaque volume, were significantly correlated with TG and hsCRP levels. These coronary plaque burden variables were also independently associated with the risk of poor prognosis after PCI, suggesting that CCTA before PCI reveals the residual risk after revascularization. For patients with a high total plaque burden, it is necessary to further intensify risk factor management, including lowering triglyceride level and ameliorating the inflammatory status.