脂蛋白(a)联合TyG指数对冠心病合并 2型糖尿病患者首次PCI术后非致命性 心肌梗死的预测价值
Predictive Value of Lipoprotein(a) Combined with TyG Index for Non-Fatal Myocardial Infarction after First PCI in Patients with Coronary Heart Disease Complicated with Type 2 Diabetes
DOI: 10.12677/acm.2026.162571, PDF, HTML, XML,   
作者: 陈 雪:青岛大学附属医院心血管内科,山东 青岛;黄岛区中医医院健康管理科,山东 青岛;李 健*:青岛大学附属医院心血管内科,山东 青岛
关键词: 冠心病2型糖尿病PCI脂蛋白(a)TyG指数预测Coronary Heart Disease Type 2 Diabetes Percutaneous Coronary Intervention Lipoprotein(a) Triglyceride-Glucose Index Prediction
摘要: 目的:探究脂蛋白(a) (lipoprotein(a), Lp(a))联合甘油三酯–葡萄糖指数(triglyceride-glucose index, TyG)对冠心病(coronary heart disease, CHD)合并2型糖尿病(type 2 diabetes mellitus, T2DM)患者首次经皮冠状动脉介入(percutaneous coronary intervention, PCI)术后非致命性心肌梗死的预测价值。方法:回顾性分析2020年1月至2023年1月首次在青岛大学附属医院接受PCI术治疗的586例CHD合并T2DM患者的临床资料,对患者进行术后随访并记录非致命性心肌梗死发生情况。根据术后是否发生非致命性心肌梗死,将患者分为非致命性心梗组(n = 115)和无非致命性心梗组(n = 471)。比较两组患者强效他汀使用率;采用Bootstrap重抽样法(重抽样1000次)校正ROC曲线的AUC及95%置信区间以验证模型稳定性;同时纳入糖化血红蛋白(HbA1c)作为对照指标。采用多因素Cox比例风险回归模型分析术后发生非致命性心肌梗死的危险因素,绘制受试者工作特征(receiver operating characteristic, ROC)曲线并计算曲线下面积(area under the curve, AUC),评估Lp(a)、TyG指数单独及联合检测对术后非致命性心梗的预测效能。结果:非致命性心梗组患者的Lp(a)水平、TyG指数均显著高于无非致命性组(均P < 0.05),而两组患者强效他汀使用率无显著差异(P = 1.000)。多因素Cox回归分析显示,高Lp(a)水平(HR = 3.240, 95% CI: 2.237~4.692, P < 0.05)和高TyG指数(HR = 3.330, 95% CI: 2.297~4.829, P < 0.001)是CHD合并T2DM首次PCI术后发生非致命性心肌梗死的独立危险因素;同时高Lp(a)联合高TyG指数的患者术后发生非致命性心肌梗死的风险最高(HR = 13.380, 95% CI: 6.578~27.213, P < 0.001)。ROC曲线分析显示,Lp(a)、TyG指数及二者联合模型的AUC分别为0.626,0.657,0.727。TyG指数的AUC (0.657)显著高于HbA1c (0.538) (Z = 4.290, P < 0.001)。经Bootstrap 1000次重抽样验证,Lp(a)和TyG指数联合模型平均AUC为0.637 (95% CI: 0.575~0.701),提示模型稳定性良好。结论:Lp(a)、TyG指数与CHD合并T2DM患者首次PCI术后非致命性心肌梗死的发生显著相关,且二者联合检测的预测效能优于单一指标,经Bootstrap验证模型稳定性良好,可作为术后非致命性心肌梗死的有效预测指标。
Abstract: Objective: To explore the predictive value of lipoprotein(a) (Lp(a)) combined with the triglyceride-glucose (TyG) index for non-fatal myocardial infarction after the first percutaneous coronary intervention (PCI) in patients with coronary heart disease (CHD) with type 2 diabetes mellitus (T2DM). Methods: A retrospective analysis was conducted on the clinical data of 586 patients with CHD complicated by T2DM who underwent PCI for the first time at Qingdao University Hospital between January 2020 and January 2023. Patients were followed up postoperatively, and the occurrence of non-fatal myocardial infarction was recorded. Patients were categorized into a non-fatal MI group (n = 115) and a non-fatal MI-free group (n = 471) based on postoperative non-fatal MI occurrence. The use of high-intensity statins was compared between the two groups. Bootstrap resampling (1000 resamples) was employed to correct the AUC and 95% confidence interval of the receiver operating characteristic (ROC) curve to validate model stability. Glycated hemoglobin (HbA1c) was included as a control indicator. Multivariate Cox proportional hazards regression analysis was performed to identify risk factors for postoperative non-fatal myocardial infarction. ROC curves were plotted, and the area under the curve (AUC) was calculated to evaluate the predictive performance of Lp(a) and TyG index, both individually and in combination, for postoperative non-fatal myocardial infarction. Results: Patients in the non-fatal myocardial infarction group exhibited significantly higher Lp(a) levels and TyG indices compared to the non-non-fatal group (both P < 0.05), while there was no significant difference in the use of intensive statins between the two groups (P = 1.000). Multivariate Cox regression analysis revealed that elevated Lp(a) levels (HR = 3.240, 95% CI: 2.237~4.692, P < 0.05) and high TyG index (HR = 3.330, 95% CI: 2.297~4.829, P < 0.001) were independent risk factors for non-fatal myocardial infarction after initial PCI in patients with CHD and T2DM, while patients with both elevated Lp(a) and high TyG index exhibited the highest risk of postoperative non-fatal myocardial infarction (HR = 13.380, 95% CI: 6.578~27.213, P < 0.001). ROC curve analysis revealed AUC values of 0.626, 0.657, and 0.727 for Lp(a), TyG index, and the combined model, respectively. The TyG index AUC (0.657) was significantly higher than those of glycated hemoglobin (HbA1c) (0.538) (Z = 4.290, P < 0.001). Bootstrap resampling (n = 1000) validated the combined Lp(a) and TyG index model with a mean AUC of 0.637 (95% CI: 0.575~0.701), indicating good model stability. Conclusions: Lp(a) and TyG index are significantly correlated with the occurrence of non-fatal myocardial infarction in patients with CHD and T2DM after their first PCI procedure. The predictive performance of their combined detection is superior to that of either index alone. The model’s stability has been verified through Bootstrap analysis, indicating that it can serve as an effective predictor of non-fatal myocardial infarction after the procedure.
文章引用:陈雪, 李健. 脂蛋白(a)联合TyG指数对冠心病合并 2型糖尿病患者首次PCI术后非致命性 心肌梗死的预测价值[J]. 临床医学进展, 2026, 16(2): 1780-1791. https://doi.org/10.12677/acm.2026.162571

1. 引言

冠心病(coronary heart disease, CHD)与2型糖尿病(type 2 diabetes mellitus, T2DM)均是严重危害全球公众健康的重大慢性疾病[1]。CHD以冠状动脉粥样硬化为主要病理基础,是导致心肌梗死、心力衰竭等心血管不良事件的主要原因[1]。T2DM则是一种以胰岛素抵抗(insulin resistance, IR)和/或胰岛素分泌不足为特征的复杂代谢性疾病[2]。二者关系密切,常相互伴发,形成恶性循环[3]。流行病学数据显示,T2DM患者发生CHD的风险较非糖尿病人群高出2~4倍,且其心血管事件发生率和死亡率显著升高[3]。高血糖、胰岛素抵抗、脂代谢紊乱及慢性炎症状态共同构成了T2DM患者加速动脉粥样硬化进程的病理生理基础,使得此类患者冠脉病变往往更为弥漫和严重,临床预后也更差[3] [4]。因此,对CHD合并T2DM这一高危人群进行精准的风险评估与管理,具有极其重要的临床意义。

经皮冠状动脉介入治疗(percutaneous coronary intervention, PCI)是目前重建CHD患者冠状动脉血流、缓解心肌缺血的主要血运重建手段[5]。然而,即使成功实施PCI,部分患者,尤其是合并T2DM者,术后仍面临包括支架内再狭窄、非致命性心肌梗死等在内的主要不良心血管事件风险[5]。因此,探寻有效、简便的生物学指标以早期识别高危患者,对于优化术后管理、改善远期预后至关重要。脂蛋白(a) (Lipoprotein(a), Lp(a))作为一种特殊的脂蛋白,在心血管疾病中扮演着重要的角色[6]。研究显示,高水平的Lp(a)与心血管事件的发生密切相关[7]。此外,TyG指数(triglyceride-glucose index)作为一个新兴的代谢综合征评估指标,能够反映IR状态,近年来越来越多地被应用于心血管疾病的风险评估中[8] [9]。尽管已有研究指出Lp(a)和TyG指数与心血管事件之间的关联,但二者联合应用于PCI术后非致命性心肌梗死的预测价值仍需进一步探讨[10]

基于以上背景,本研究旨在深入探讨Lp(a)联合TyG指数对冠心病合并2型糖尿病患者首次接受PCI治疗后,发生非致命性心肌梗死的预测价值。通过回顾性分析临床资料,比较发生与未发生事件患者的指标差异,并运用多因素回归模型与受试者工作特征曲线分析,旨在明确Lp(a)与TyG指数是否为此类患者术后非致命性心肌梗死的独立危险因素,并进一步评估二者单独及联合应用的预测效能。研究结果有望为临床早期识别高风险个体、实施更积极的二级预防策略提供新的、更具综合性的生物标志物参考依据,从而助力改善CHD合并T2DM患者的长期预后。

2. 对象和方法

2.1. 研究对象与分组

本研究为单中心回顾性队列研究。选取2020年1月至2023年1月青岛大学附属医院首次接受PCI治疗的CHD合并T2DM患者。

研究人群纳入标准:(1) 年龄 ≥ 18岁;(2) 2型糖尿病诊断明确[11];(3) 冠心病诊断明确:冠心病以经冠状动脉造影显示1支至或1支以上的冠状动脉及其主要分支血管狭窄 ≥ 50%为诊断标准[12];(4) 首次进行PCI治疗,临床资料完整。

排除标准:(1) 排除严重肝肾功能衰竭的患者;(2) 排除恶性肿瘤的患者;(3) 排除基线资料缺失脂蛋白(a)的患者;(4) 排除基线资料缺失甘油三酯和空腹血糖的患者;(5) 排除严重血液系统疾病及免疫系统疾病的患者;(6) 排除失访患者。

所有患者均由经验丰富的心脏病介入专家,经过桡动脉或股动脉途径实施造影。造影结果由本院2名以上具有经验丰富的心血管医师分析,并结合冠状动脉造影定量系统,以目测法评估血管病变部分、狭窄程度。对冠状动脉或其分支造影检查显示管腔狭窄 ≥ 70%的病变血管进行介入处理,并对多支血管病变患者采用分次介入的策略,优先处理犯罪血管,在住院期内进行非犯罪血管的处理。所有患者均按介入手术方案植入药物洗脱支架。

本研究经青岛大学附属医院伦理委员会审核并批准(伦理批准号:QYFYWZLL30893),因系回顾性研究,所有分析数据均为匿名,豁免患者知情同意。

2.2. 结局指标

主要结局指标:患者首次PCI术后非致命性心肌梗死的发生情况。诊断参照《全球心肌梗死通用定义(第四版,2018年)》[13]

2.3. 资料采集与随访方法

通过医院电子病历系统收集患者数据资料,具体见表1,Gensini评分计算方式详见补充材料表1

Table 1. Classification and specific content of patient data

1. 患者数据资料分类及具体内容

资料分类

具体内容

人口统计学与 既往史合并症

年龄、性别、冠心病家族史、吸烟史、高血压、降压药使用、高脂血症、降脂药使用(阿托伐他汀 ≥ 40 mg/瑞舒伐他汀 ≥ 20 mg、PCSK9抑制剂)、降糖药使用、心力衰竭、慢性肾脏病(CKD)

人体测量学资料

身高、体重、体重指数(BMI)、收缩压、舒张压、心率

血液标志物资料

血脂全项:空腹血糖、甘油三酯(TG)、总胆固醇(TC)、低密度脂蛋白胆固醇(LDL)、高密度脂蛋白胆固醇(HDL)、载脂蛋白A1 (Apo A1)、载脂蛋白B (Apo B)、Lp(a)

血常规:白细胞计数、中性粒细胞计数、淋巴细胞计数、单核细胞计数、血红蛋白、血小板 计数

心肌酶检测:肌钙蛋白、肌酸激酶同工酶(CK-MB)

肾功能:尿素氮、肌酐、尿酸

肝功能:谷丙转氨酶(ALT)、谷草转氨酶(AST)、总胆红素、白蛋白

6. 其他标志物:糖化血红蛋白(HbA1c)、C反应蛋白(CRP)、纤维蛋白原、D-二聚体

心脏超声数据

左心房内径(LAD)、左心室舒张末期内径(LVDD)、室间隔厚度(IVSTD)、左心室射血分数(LVEF)

冠脉造影资料

左主干病变、三支病变、多支病变、支架植入数量、支架总长度、Gensini评分

出院用药

抗血小板药:阿司匹林、吲哚布芬、氯吡格雷、替格瑞洛

降压药:β受体阻滞剂、钙通道阻滞剂(CCB)、血管紧张素转化酶抑制剂(ACEI)、 血管紧张素Ⅱ受体拮抗剂(ARB)

调脂药:他汀类、前蛋白转化酶枯草溶菌素9抑制剂(PCSK9i)、依折麦布

降糖药:胰岛素、口服降糖药

5. 利尿剂:呋塞米、螺内酯

患者术后通过门急诊记录随访、既往住院资料随访、电话随访进行随访,随访起点为PCI术后出院日,随访终点为2025年1月1日或至非致命性心肌梗死的发生日,中位随访时长为35.58个月(范围:0.47~60.90个月)。

2.4. 统计学方法

数据分析采用SPSS26.0统计软件及R语言(4.5.2版本)。对连续变量进行Shapiro-Wilk正态性检验,连续变量采用中位数(四分位数)表示,两组间比较采用Mann-Whitney U检验。四组间比较采用Kruskal-Wallis H检验。分类变量以频数(%)表示。两组或四组间的差异采用χ2检验和Fisher精确检验。单因素Cox回归分析筛选潜在预后影响因素,并P < 0.05的变量纳入多因素Cox比例风险回归模型,校正混杂因素后分析Lp(a)和TyG指数对非致命性心肌梗死的独立影响及二者交互作用。根据年龄、性别、高血压、高脂血症、心力衰竭、三支病变进行亚组分析,进一步评估Lp(a)、TyG及其联合与非致命性心肌梗死的多因素分层关联。为验证模型稳定性,采用R语言pROC和boot包进行Bootstrap重抽样法(1000次)校正ROC曲线的AUC及95%置信区间。纳入HbA1c作为对照指标,对比TyG指数的预测效能。绘制受试者工作特征(ROC)曲线,评估Lp(a)、TyG指数及二者联合的预测价值,并依据Youden指数确定最佳截断值,计算对应的灵敏度和特异度。所有检测均为双侧检验,P < 0.05被视为具有统计学显著性。

3. 结果

3.1. 纳入人群的基线特征

组间基线特征比较显示:非致命性心肌梗死组空腹血糖、TG、Lp(a)、TyG指数水平均显著高于无非致命性心肌梗死组,而高脂血症、左主干病变、阿司匹林使用占比显著低于无非致命性心肌梗死组,差异均有统计学意义(P < 0.05)。其中Lp(a)、TyG、TG组间差异达极显著水平(均P < 0.001)。具体基线特征比较结果见表2

Table 2. Comparison of baseline characteristics between the non-fatal myocardial infarction group and the non-fatal myocardial infarction-free group

2. 非致命性心梗组与无非致命性心肌梗死组基线特征比较

变量

无非致命性心肌梗死

非致命性心肌梗死

Z/t/χ2

P

N

471

115

高脂血症,n (%)

160 (34.0)

54 (47.0)

6.724

0.010

空腹血糖,mmol/L

7.58 (6.37, 9.88)

8.47 (6.79, 11.24)

−2.723

0.006

TG, mmol/L

1.50 (1.09, 2.27)

1.96 (1.49, 3.07)

−5.297

<0.001

Lp(a), mg/L

173.00 (88.00, 334.00)

308.00 (124.00, 494.20)

−4.198

<0.001

TyG

9.15 (8.70, 9.66)

9.61 (9.04, 10.15)

−5.226

<0.001

左主干病变,n (%)

53 (11.3)

21 (18.3)

4.115

0.043

出院阿司匹林使用,n (%)

385 (81.7)

105 (91.3)

6.171

0.013

Lp(a),脂蛋白(a);TyG,甘油三酯–葡萄糖指数;TG,甘油三酯。

Lp(a)最佳截断值为301 mg/L,而TyG最佳截断值为10.10。根据Lp(a)最佳截断值分为两组,低Lp(a)组(Lp(a) ≤ 301 mg/L, n = 390)和高Lp(a)组(Lp(a) > 301 mg/L, n = 196)。根据TyG最佳截断值分成两组,低TyG组(TyG ≤ 10.10, n = 469)和高TyG组(TyG > 10.10, n = 117)。依据Lp(a)和TyG指数水平将研究对象分为四组(联合分组),组1为低Lp(a)组 + 低TyG组(n = 320)。组2为低Lp(a)组 + 高TyG组(n = 149)。组3为高Lp(a)组 + 低TyG组(n = 70)。组4为高Lp(a)组 + 高TyG组(n = 47)。四组间基线特征比较显示,性别、高脂血症和螺内酯、口服降糖药、胰岛素的使用比例在四组间分布差异均有统计学意义(均P < 0.05)。四组计量资料比较差异均有统计学意义(均P < 0.05),其中Lp(a)和TyG组间差异具有高度统计学意义(均P < 0.001)。四组非致命性心肌梗死发生率比较,差异有统计学意义(P < 0.001),且组4非致命性心肌梗死发生率最高。具体比较结果见表3

Table 3. Comparison of baseline characteristics for non-fatal myocardial infarction in PCI patients in the Lp(a) and TyG combined group

3. Lp(a)、TyG联合分组的PCI术后患者非致命性心肌梗死基线特征比较

变量

组1

组2

组3

组4

F/K/χ2

P

年龄,岁

65.00 (59.00, 70.00)

67.00 (59.00, 72.00)

61.50 (55.00, 68.25)

65.00 (59.00, 70.00)

110.071

0.018

性别,n (%)

17.150

0.001

男性

210 (65.6)

78 (52.3)

39 (55.7)

18 (38.3)

女性

110 (34.4)

71 (47.7)

31 (44.3)

29 (61.7)

高脂血症,n (%)

66 (20.6)

38 (25.5)

65 (92.9)

45 (95.7)

209.622

<0.001

BMI, kg/m2

25.39 (23.67, 27.67)

25.40 (23.41, 27.14)

26.21 (24.56, 29.56)

25.59 (24.03, 27.47)

12.205

0.007

空腹血糖,mmol/L

7.20 (6.32, 9.07)

7.30 (5.77, 8.67)

11.23 (9.35, 13.62)

11.03 (9.65, 13.44)

145.374

<0.001

糖化血红蛋白,%

7.50 (6.80, 8.60)

7.50 (6.70, 8.55)

8.75 (8.08, 9.83)

9.30 (8.10, 10.50)

68.271

<0.001

甘油三酯,mmol/L

1.42 (1.08, 1.88)

1.39 (1.07, 1.80)

3.61 (2.89, 5.42)

3.11 (2.56, 3.72)

231.702

<0.001

总胆固醇,mmol/L

4.31 (3.60, 5.11)

4.47 (3.78, 5.32)

4.71 (4.08, 5.80)

4.82 (4.13, 6.10)

17.588

0.001

载脂蛋白B,g/L

0.82 (0.67, 0.98)

0.88 (0.72, 1.05)

0.92 (0.80, 1.08)

0.94 (0.74, 1.15)

17.043

0.001

Lp(a), mg/L

124.15 (69.25, 193.08)

514.00 (374.15, 745.00)

98.50 (48.25, 152.20)

466.00 (334.00, 655.10)

394.517

<0.001

TyG

9.05 (8.66, 9.42)

9.01 (8.65, 9.32)

10.34 (10.16, 10.63)

10.17 (10.13, 10.33)

281.552

<0.001

CRP, mg/L

0.79 (0.50, 3.98)

1.80 (0.54, 6.22)

1.50 (0.50, 3.81)

2.17 (0.50, 5.96)

14.561

0.002

尿素氮,mmol/L

5.70 (4.70, 6.98)

6.23 (5.21, 7.85)

5.92 (4.90, 7.40)

6.60 (5.22, 8.75)

13.416

0.004

总胆红素,umol/L

15.62 (11.51, 19.40)

13.50 (10.04, 18.45)

11.59 (9.22, 19.00)

11.00 (8.80, 14.70)

28.153

<0.001

血红蛋白,g/L

141.00 (131.00, 153.00)

137.00 (121.50, 145.00)

139.50 (131.00, 149.25)

135.00 (123.00, 143.00)

18.118

<0.001

纤维蛋白原,g/L

3.14 (2.66, 3.60)

3.44 (2.89, 4.07)

3.09 (2.59, 3.54)

3.22 (2.77, 3.89)

16.157

0.001

D二聚体,ng/mL

280.00 (190.00, 390.00)

340.00 (260.00, 430.00)

320.00 (170.00, 370.00)

270.00 (190.00, 400.00)

16.869

0.001

Gensini评分

47.00 (32.00, 69.00)

52.00 (31.50, 79.00)

58.00 (43.50, 83.25)

55.00 (40.00, 94.00)

10.208

0.017

出院用药,n (%)

螺内酯

63 (19.7)

47 (31.5)

16 (22.9)

11 (23.4)

7.990

0.046

口服降糖药

262 (81.9)

120 (80.5)

66 (94.3)

45 (95.7)

12.816

0.005

胰岛素

18 (5.6)

10 (6.7)

3 (4.3)

9 (19.1)

9.721

0.017

非致命性心梗,n (%)

26 (8.1)

41 (27.5)

23 (32.9)

25 (53.2)

74.056

<0.001

组1:低Lp(a)组 + 低TyG组;组2:低Lp(a)组 + 高TyG组;组3:高Lp(a)组 + 低TyG组;组4:高Lp(a)组 + 高TyG组。BMI,体重指数;Lp(a),脂蛋白(a);TyG,甘油三酯–葡萄糖指数;CRP,C反应蛋白。

3.2. 多因素分析Lp(a)、TyG及其联合与非致命性心肌梗死的相关性

Lp(a)、TyG及其联合与非致命性心肌梗死的多因素Cox回归分析结果显示,高Lp(a)水平(HR = 3.226, P < 0.05)和高TyG (HR = 3.241, P < 0.05)是非致命性心肌梗死独立风险因素,即使矫正阿司匹林等大量混杂因素(模型3),关联仍具有统计学意义(P < 0.05)。联合分组分析显示,同时具有高Lp(a)和高TyG的患者发生非致命性心肌梗死风险最高(HR = 8.390, P < 0.05)。Lp(a)与TyG存在显著协同作用,二者均升高时,非致命性心肌梗死风险远高于单一指标升高组。调整更多混杂因素(模型2到模型3),联合分组的HR无下降趋势,提示协同关联具有良好的稳健性。具体结果见表4

联合分组与非致命性心肌梗死的多因素亚组分析结果显示,在所有亚组中,组2、组3、组4发生非致命性心肌梗死风险均显著高于组1 (HR均 > 1,P < 0.05)。组4风险最高,说明Lp(a)和TyG同时升高时,加大风险非致命性心肌梗死风险,并且不受年龄、性别、高血压、高脂血症、心力衰竭、三支病变这些因素影响。具体结果见表5

Table 4. Multivariate Cox regression analysis of Lp(a), TyG, and their combination with non-fatal myocardial infarction

4. Lp(a)、TyG及其联合与非致命性心肌梗死的多因素Cox回归分析

变量

模型1

模型2

模型3

HR

95% CI

P

HR

95% CI

P

HR

95% CI

P

Lp(a)

Lp(a)

1.001

1.001~1.002

<0.001

1.001

1.001~1.002

<0.001

1.001

1.001~1.002

<0.001

低Lp(a)组

参考

参考

参考

高Lp(a)组

3.226

2.228~4.671

<0.001

3.226

2.228~4.671

<0.001

3.240

2.237~4.692

<0.001

TyG

TyG

1.680

1.352~2.088

<0.001

1.680

1.352~2.088

<0.001

1.720

1.388~2.132

<0.001

低TyG组

参考

参考

参考

高TyG组

3.241

2.237~4.696

<0.001

3.242

2.237~4.696

<0.001

3.330

2.297~4.829

<0.001

联合分组

组1

参考

参考

参考

组2

4.124

2.522~6.746

<0.001

4.237

2.590~6.932

<0.001

4.319

2.639~7.070

<0.001

组3

4.640

2.647~8.133

<0.001

7.152

3.571~14.328

<0.001

7.315

3.686~14.516

<0.001

组4

8.390

4.843~14.535

<0.001

13.503

6.590~27.665

<0.001

13.380

6.578~27.213

<0.001

P for trend

<0.001

<0.001

<0.001

组1:低Lp(a)组 + 低TyG组;组2:低Lp(a)组 + 高TyG组;组3:高Lp(a)组 + 低TyG组;组4:高Lp(a)组 + 高TyG组。模型1:未调整;模型2:调整了高脂血症;模型3:调整高脂血症、总胆红素、血红蛋白、左主干病变、三支病变、阿司匹林。Lp(a),脂蛋白(a);TyG,甘油三酯–葡萄糖指数;HR,风险比;CI,置信区间。

Table 5. Multivariate subgroup analysis of Lp(a) and TyG combined stratification with nonfatal myocardial infarction

5. Lp(a)、TyG联合分组与非致命性心肌梗死的多因素亚组分析

亚组

组2 vs组1

组3 vs组1

组4 vs组1

HR

95% CI

P

HR

95% CI

P

HR

95% CI

P

年龄

<65岁

3.974

1.858~8.500

<0.001

10.567

4.141~26.963

<0.001

39.825

13.256~119.646

<0.001

≥65岁

4.369

2.279~8.376

<0.001

5.971

2.570~13.870

<0.001

5.879

2.608~13.250

<0.001

性别

男性

4.547

2.433~8.496

<0.001

10.602

3.970~28.308

<0.001

34.450

11.294~105.087

<0.001

女性

4.346

1.901~9.935

<0.001

7.034

2.822~17.536

<0.001

7.714

3.234~18.402

<0.001

高血压

4.846

2.726~8.614

<0.001

4.460

2.298~8.657

<0.001

5.907

2.937~11.880

<0.001

2.284

0.850~6.135

0.101

4.222

1.431~12.456

0.009

13.087

5.139~33.327

<0.001

高脂血症

20.746

2.654~162.152

0.004

24.296

3.260~181.058

0.002

43.900

5.926~325.192

<0.001

3.467

2.045~5.876

<0.001

6.192

1.868~20.523

0.003

21.605

5.088~91.731

<0.001

心力衰竭

9.814

2.527~38.117

0.001

7.543

0.672~84.648

0.101

331.454

17.222~6378.996

<0.001

3.847

2.258~6.556

<0.001

5.494

3.062~9,859

<0.001

8.237

4.556~14.891

<0.001

三支病变

4.642

2.431~8.863

<0.001

4.634

2.177~9.863

<0.001

6.962

3.357~14.437

<0.001

3.522

1.628~7.618

0.001

4.911

2.120~11.374

<0.001

14.142

6.095~32.816

<0.001

亚组分析调整了高脂血症、总胆红素、血红蛋白、左主干病变、三支病变、阿司匹林。HR,风险比;CI,置信区间。

3.3. Lp(a)、TyG及其联合对非致命性心梗的预测价值

ROC曲线分析显示,Lp(a)、TyG及其联合预测预后的敏感性分别为57.4%、42.6%、77.4%,特异性分别为72.4%、85.4%、62.4%;Lp(a) (AUC = 0.626, P < 0.001)、TyG (AUC = 0.657, P < 0.001)单独对非致命性心肌梗死均有统计学意义的预测价值,但单独预测能力有限。Lp(a)与TyG联合评估时,AUC提升(AUC = 0.727, P < 0.001),达到中等预测效能,且结果稳定,表明Lp(a)、TyG及联合分组对非致命性心肌梗死的发生均具有一定的预测价值,其中联合分组预测价值更高。HbA1c单独对非致命性心肌梗死无统计学意义预测价值。TyG指数的AUC为0.657,HbA1c的AUC为0.538,Z检验显示两者AUC差异有统计学意义(Z = 4.290, P < 0.001)提示TyG指数在预测术后非致命性心梗方面优于HbA1c。具体结果见表6图1

ROC,受试者操作特征曲线;Lp(a),脂蛋白(a);TyG,甘油三酯–葡萄糖指数。

Figure 1. ROC curve evaluating the predictive value of Lp(a), TyG, and their combination for non-fatal myocardial infarction

1. ROC曲线评估Lp(a)、TyG及其联合对非致命性心肌梗死的预测价值

Table 6. ROC curve evaluation of Lp(a), TyG, and their combined predictive value for non-fatal myocardial infarction compared to HbA1c

6. ROC曲线评估Lp(a)、TyG及其联合与HbA1c对非致命性心肌梗死的预测价值

变量

截断值

AUC

95% CI

P

敏感性

特异性

Z值(TyG vs HbA1c)

P值(Z检验)

Lp(a)

301 mg/L

0.626

0.568~0.684

<0.001

0.574

0.724

-

-

TyG

10.10

0.657

0.604~0.710

<0.001

0.426

0.854

4.290

< 0.001

联合分组

-

0.727

0.674~0.780

<0.001

0.774

0.624

-

-

HbA1c

6.45%

0.538

0.482~0.593

0.208

0.939

0.149

-

-

ROC,受试者操作特征曲线;Lp(a),脂蛋白(a);TyG,甘油三酯–葡萄糖指数;AUC,曲线下面积;CI,置信区间。

3.4. 模型稳定性验证

构建Lp(a)联合TyG指数预测模型,其ROC曲线下面积(AUC)为0.638,经Bootstrap 1000次重抽样校正后,模型的平均AUC为0.637 (95% CI: 0.575~0.701),与原始AUC基本一致,提示模型无明显过拟合,稳定性良好。具体见图2

ROC,受试者操作特征曲线;AUC,Sensitivity;敏感度;Specificity,特异度;95% CI,95%置信区间。

Figure 2. Bootstrap-corrected ROC curve of the combined model

2. 联合模型的Bootstrap校正ROC曲线

4. 讨论

CHD合并T2DM患者在成功接受PCI后,仍是主要不良心血管事件(MACE)的高危人群,其中非致命性心肌梗死对患者远期生活质量与预后的负面影响尤为显著[14]。传统风险标志物对该特定人群术后风险的预测能力有限,因此,探寻整合多病理环节的新型、简便生物标志物,对于实现精准的风险分层与早期干预至关重要[15] [16]。本研究通过回顾性队列分析发现,Lp(a)与TyG指数是CHD合并T2DM患者首次PCI术后非致命性心肌梗死发生的独立危险因素,且二者联合检测可进一步提升预测效能。

本研究结果显示,术后发生非致命性心肌梗死的患者,其Lp(a)与TyG指数水平均显著高于未发生事件者(均P < 0.05)。多因素Cox比例风险回归分析进一步证实,Lp(a)、TyG指数及二者联合检测均与CHD合并T2DM患者首次PCI术后非致命性心肌梗死发生显著相关,且高Lp(a)水平(HR = 3.240, 95% CI: 2.237~4.692, P < 0.05)和高TyG指数(HR = 3.330, 95% CI: 2.297~4.829, P < 0.001)均为非致命性心肌梗死发生的独立危险因素,这一发现为临床风险评估提供了新的参考依据。近年来研究证实,Lp(a)是动脉粥样硬化性心血管疾病(ASCVD)的核心风险因素。Lp(a)水平升高与后续急性心肌梗死风险增加显著相关[17] [18]。Lp(a)的致动脉粥样硬化作用主要源于其与LDL相似的结构,易沉积于血管壁,且其载脂蛋白(a) (apo(a))成分所携带的Kringle IV-2重复序列可诱发炎症、促进血栓形成并抑制纤溶,进而加速斑块进展与不稳定化[19]。同时,TyG指数作为反映IR的可靠替代指标,与心血管预后的关联性已被多项研究证实[20] [21]。基于中国T2DM人群的前瞻性研究亦发现,TyG指数升高与MACE发生风险正相关[22]。TyG指数作为IR的简易标志物,可通过代谢紊乱引发血管功能异常,同时激活低度炎症通路加速斑块不稳定与血栓形成[8] [23] [24]。本研究进一步显示,TyG指数的AUC (0.657)显著高于HbA1c (0.538) (Z = 4.290, P < 0.001),提示其在预测术后非致命性心肌梗死方面的效能优于传统血糖指标。二者从不同病理环节参与PCI术后非致命性心肌梗死的发生,这也为联合检测的协同效应提供了理论依据。

本研究进一步证实,Lp(a)与TyG指数联合预测对PCI术后非致命性心肌梗死的预测效能优于单一指标,Lp(a)、TyG指数及其联合预测预后的AUC分别为0.626、0.657、0.727,联合检测时患者非致命性心肌梗死发生风险升高8.390倍(P < 0.05),且ROC曲线下面积显著高于单一指标。此外,经Bootstrap 1000次重抽样验证,Lp(a)与TyG指数联合模型的平均AUC为0.637 (95% CI: 0.575~0.701),提示模型稳定性良好。单一生物标志物仅能反映心血管风险的某一维度,而联合检测可整合动脉粥样硬化与胰岛素抵抗两大病理环节的风险信息,提升预测的全面性。从病理机制来看,二者存在明确的协同作用:Lp(a)主要通过促进动脉粥样硬化和血栓形成增加风险,TyG指数则通过代谢紊乱放大炎症反应[24] [25]。胰岛素抵抗可增强Lp(a)的氧化修饰并加速斑块破裂,而Lp(a)可刺激巨噬细胞释放炎症因子,与TyG指数诱导的游离脂肪酸积累形成炎症叠加效应,这也是联合检测预测效能更优的核心原因[25] [26]

综上所述,本研究表明,Lp(a)与TyG指数可作为评估CHD合并T2DM患者首次PCI术后非致命性心肌梗死风险的实用生物标志物,且联合检测具有更高的预测价值,有助于早期识别高危个体,为临床采取强化降脂(如针对高Lp(a)人群优化LDL-C管理)、改善胰岛素敏感性等个体化二级预防策略提供参考。然而,本研究为单中心回顾性观察性研究,存在一定局限性:首先,未能证实Lp(a)与TyG指数和非致命性心肌梗死之间的因果关系;其次,虽经多因素校正,仍可能存在残余混杂;第三,随访时间有限,结论需进一步验证。未来研究可开展前瞻性临床研究验证本联合预测模型;利用组学技术深入探索Lp(a)与IR交互作用的分子机制;并通过干预性临床试验,评估以调控Lp(a)或改善IR为靶点的治疗策略对患者远期预后的影响。

NOTES

*通讯作者。

参考文献

[1] Virani, S.S., Newby, L.K., Arnold, S.V., Bittner, V., Brewer, L.C., Demeter, S.H., et al. (2023) 2023 AHA/ACC/ACCP/ASPC/NLA/PCNA Guideline for the Management of Patients with Chronic Coronary Disease: A Report of the American Heart Association/American College of Cardiology Joint Committee on Clinical Practice Guidelines. Circulation, 148, e9-e11. [Google Scholar] [CrossRef] [PubMed]
[2] Einarson, T.R., Acs, A., Ludwig, C. and Panton, U.H. (2018) Prevalence of Cardiovascular Disease in Type 2 Diabetes: A Systematic Literature Review of Scientific Evidence from across the World in 2007-2017. Cardiovascular Diabetology, 17, Article No. 83. [Google Scholar] [CrossRef] [PubMed]
[3] Bornfeldt, K.E. and Tabas, I. (2011) Insulin Resistance, Hyperglycemia, and Atherosclerosis. Cell Metabolism, 14, 575-585. [Google Scholar] [CrossRef] [PubMed]
[4] Marx, N., Federici, M., Schütt, K., Müller-Wieland, D., Ajjan, R.A., Antunes, M.J., et al. (2023) 2023 ESC Guidelines for the Management of Cardiovascular Disease in Patients with Diabetes. European Heart Journal, 44, 4043-4140. [Google Scholar] [CrossRef] [PubMed]
[5] Lawton, J.S., Tamis-Holland, J.E., Bangalore, S., Bates, E.R., Beckie, T.M., Bischoff, J.M., et al. (2022) 2021 ACC/AHA/SCAI Guideline for Coronary Artery Revascularization: A Report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. Circulation, 145, e4-e17. [Google Scholar] [CrossRef] [PubMed]
[6] Kronenberg, F., Mora, S., Stroes, E.S.G., Ference, B.A., Arsenault, B.J., Berglund, L., et al. (2022) Lipoprotein(a) in Atherosclerotic Cardiovascular Disease and Aortic Stenosis: A European Atherosclerosis Society Consensus Statement. European Heart Journal, 43, 3925-3946. [Google Scholar] [CrossRef] [PubMed]
[7] Berman, A.N., Biery, D.W., Besser, S.A., Singh, A., Shiyovich, A., Weber, B.N., et al. (2024) Lipoprotein(a) and Major Adverse Cardiovascular Events in Patients with or without Baseline Atherosclerotic Cardiovascular Disease. Journal of the American College of Cardiology, 83, 873-886. [Google Scholar] [CrossRef] [PubMed]
[8] Sánchez‐Íñigo, L., Navarro‐González, D., Fernández‐Montero, A., Pastrana‐Delgado, J. and Martínez, J.A. (2016) The Tyg Index May Predict the Development of Cardiovascular Events. European Journal of Clinical Investigation, 46, 189-197. [Google Scholar] [CrossRef] [PubMed]
[9] da Silva, A., Caldas, A.P.S., Hermsdorff, H.H.M., Bersch-Ferreira, Â.C., Torreglosa, C.R., Weber, B., et al. (2019) Triglyceride-Glucose Index Is Associated with Symptomatic Coronary Artery Disease in Patients in Secondary Care. Cardiovascular Diabetology, 18, Article No. 89. [Google Scholar] [CrossRef] [PubMed]
[10] Luo, T., Chen, S., Cai, J., Liu, Q., Gou, R., Mo, X., et al. (2022) Association between Combined Exposure to Plasma Heavy Metals and Dyslipidemia in a Chinese Population. Lipids in Health and Disease, 21, Article No. 131. [Google Scholar] [CrossRef] [PubMed]
[11] ElSayed, N.A., McCoy, R.G., Aleppo, G., Balapattabi, K., Beverly, E.A., Briggs Early, K., et al. (2024) 2. Diagnosis and Classification of Diabetes: Standards of Care in Diabetes—2025. Diabetes Care, 48, S27-S49. [Google Scholar] [CrossRef] [PubMed]
[12] Fihn, S.D., Gardin, J.M., Abrams, J., Berra, K., Blankenship, J.C., Dallas, A.P., et al. (2012) 2012 ACCF/AHA/ACP/AATS/PCNA/SCAI/STS Guideline for the Diagnosis and Management of Patients with Stable Ischemic Heart Disease: A Report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines, and the American College of Physicians, American Association for Thoracic Surgery, Preventive Cardiovascular Nurses Association, Society for Cardiovascular Angiography and Interventions, and Society of Thoracic Surgeons. Circulation, 126, e354-e471. [Google Scholar] [CrossRef] [PubMed]
[13] Crengle, S. (2020) Comment On: Jaine Et Al Cost-Effectiveness of a Low-Dose Computed Tomography Screening Programme for Lung Cancer in New Zealand. Lung Cancer 2018, 124, 233-240. 2018/10/01. Lung Cancer, 145, 219-220. [Google Scholar] [CrossRef] [PubMed]
[14] Jiang, Z., Ruan, S., Duan, Q., Zhang, W. and Pan, S. (2025) Predictive Value of Lp(a) for MACE in ACS Patients with T2DM Following PCI, with Achieved LDL-C Target: A Cross-Sectional Study. Science Progress, 108, 1-12. [Google Scholar] [CrossRef] [PubMed]
[15] Xu, S., Qin, Z., Yuan, R., Cui, X., Zhang, L., Bai, J., et al. (2022) The Hemoglobin Glycation Index Predicts the Risk of Adverse Cardiovascular Events in Coronary Heart Disease Patients with Type 2 Diabetes Mellitus. Frontiers in Cardiovascular Medicine, 9, Article 992252. [Google Scholar] [CrossRef] [PubMed]
[16] Zhong, M., Tian, X., Sun, Q., Li, L., Lu, Y., Feng, Z., et al. (2023) Correlation of Asprosin and NRG-4 with Type 2 Diabetes Mellitus Complicated with Coronary Heart Disease and the Diagnostic Value. BMC Endocrine Disorders, 23, Articek No. 61. [Google Scholar] [CrossRef] [PubMed]
[17] Nurmohamed, N.S., Moriarty, P.M. and Stroes, E.S. (2022) Considerations for Routinely Testing for High Lipoprotein(a). Current Opinion in Lipidology, 34, 174-179. [Google Scholar] [CrossRef] [PubMed]
[18] Sagris, M., Antonopoulos, A.S., Theofilis, P., Oikonomou, E., Siasos, G., Tsalamandris, S., et al. (2021) Risk Factors Profile of Young and Older Patients with Myocardial Infarction. Cardiovascular Research, 118, 2281-2292. [Google Scholar] [CrossRef] [PubMed]
[19] Bittner, V.A., Szarek, M., Aylward, P.E., Bhatt, D.L., Diaz, R., Edelberg, J.M., et al. (2020) Effect of Alirocumab on Lipoprotein(a) and Cardiovascular Risk after Acute Coronary Syndrome. Journal of the American College of Cardiology, 75, 133-144. [Google Scholar] [CrossRef] [PubMed]
[20] Zeng, J., Li, N., Ma, C., Song, Y., Tian, W., Gao, Y., et al. (2025) Association between Triglyceride-Glucose Index and Postoperative Delirium Following Cardiac Surgery: A Multicenter Retrospective Analysis. Lipids in Health and Disease, 24, Article No. 330. [Google Scholar] [CrossRef
[21] Bulut, M., Celik, F.B., Guvenc, T.S., Yilmaz, Y., Celik, M., Ozyildirim, S., et al. (2024) Usefulness of Triglyceride-Glucose Index and Homeostatic Model Assessment for Predicting Coronary Microvascular Dysfunction. Journal of Clinical Lipidology, 18, e764-e772. [Google Scholar] [CrossRef] [PubMed]
[22] Tai, S., Fu, L., Zhang, N., Zhou, Y., Xing, Z. and Wang, Y. (2022) Impact of Baseline and Trajectory of Triglyceride-Glucose Index on Cardiovascular Outcomes in Patients with Type 2 Diabetes Mellitus. Frontiers in Endocrinology, 13, Article 858209. [Google Scholar] [CrossRef] [PubMed]
[23] Simental-Mendía, L.E., Rodríguez-Morán, M. and Guerrero-Romero, F. (2008) The Product of Fasting Glucose and Triglycerides as Surrogate for Identifying Insulin Resistance in Apparently Healthy Subjects. Metabolic Syndrome and Related Disorders, 6, 299-304. [Google Scholar] [CrossRef] [PubMed]
[24] Vergès, B. (2015) Pathophysiology of Diabetic Dyslipidaemia: Where Are We? Diabetologia, 58, 886-899. [Google Scholar] [CrossRef] [PubMed]
[25] Reyes-Soffer, G., Ginsberg, H.N., Berglund, L., Duell, P.B., Heffron, S.P., Kamstrup, P.R., et al. (2022) Lipoprotein(a): A Genetically Determined, Causal, and Prevalent Risk Factor for Atherosclerotic Cardiovascular Disease: A Scientific Statement from the American Heart Association. Arteriosclerosis, Thrombosis, and Vascular Biology, 42, e48-e60. [Google Scholar] [CrossRef] [PubMed]
[26] van der Valk, F.M., Bekkering, S., Kroon, J., Yeang, C., Van den Bossche, J., van Buul, J.D., et al. (2016) Oxidized Phospholipids on Lipoprotein(a) Elicit Arterial Wall Inflammation and an Inflammatory Monocyte Response in Humans. Circulation, 134, 611-624. [Google Scholar] [CrossRef] [PubMed]