中国全科医学 ›› 2023, Vol. 26 ›› Issue (14): 1716-1725.DOI: 10.12114/j.issn.1007-9572.2022.0755

所属专题: 心血管最新文章合辑 全科质控专项研究

• 论著 • 上一篇    下一篇

心脏代谢指数对体质量正常人群代谢异常表型的预测价值研究

陈一佳1, 戚圣香1, 杜金玲1,2, 王琛琛1, 周海茸1, 叶青1, 秦真真1, 苏健3, 武鸣3, 洪忻1,2,*()   

  1. 1.210003 江苏省南京市疾病预防控制中心慢性非传染病防制科
    2.211112 江苏省南京市,南京医科大学公共卫生学院流行病与卫生统计学系
    3.210009 江苏省南京市,江苏省疾病预防控制中心慢性非传染病防制所
  • 收稿日期:2022-06-20 修回日期:2022-12-12 出版日期:2023-05-15 发布日期:2023-01-05
  • 通讯作者: 洪忻

  • 作者贡献:陈一佳负责文章的构思与设计、论文撰写与修订;戚圣香、杜金玲负责形式分析;王琛琛、周海茸负责文献/资料收集;叶青、秦真真负责文献/资料整理;苏健、武鸣负责对文章审查和编辑写作;洪忻负责文章的质量控制与审校,并对文章整体负责、监督管理。
  • 基金资助:
    江苏省第五届"333工程"(BRA2020090); 江苏省领军人才与创新团队计划(K201105); 江苏省卫健委2020年度医学研究项目(M2020085); 南京市医学科技发展项目(ZKX18049,YKK17199)

Predictive Value of Cardiometabolic Index for Metabolically Obese Phenotype in Normal Weight Population

CHEN Yijia1, QI Shengxiang1, DU Jinling1,2, WANG Chenchen1, ZHOU Hairong1, YE Qing1, QIN Zhenzhen1, SU Jian3, WU Ming3, HONG Xin1,2,*()   

  1. 1. Department of Non-communicable Disease Prevention, Nanjing Municipal Center for Disease Control and Prevention, Nanjing 210003, China
    2. Department of Epidemiology and Health Statistics, School of Public Health, Nanjing Medical University, Nanjing 211112, China
    3. Institute for Non-communicable Disease Control, Jiangsu Provincial Center for Disease Control and Prevention, Nanjing 210009, China
  • Received:2022-06-20 Revised:2022-12-12 Published:2023-05-15 Online:2023-01-05
  • Contact: HONG Xin

摘要: 背景 心脏代谢指数(CMI)是一种衡量血脂的简易指标,与糖尿病、脑卒中密切相关。体质量正常代谢异常(MONW)表型个体糖尿病、心脑血管疾病发病和死亡风险更高,正确识别MONW表型个体对代谢相关疾病的预防和控制至关重要。但CMI对MONW表型预测价值如何,相关研究较少。 目的 探讨CMI与MONW表型的关系,并评估CMI对MONW表型的预测价值。 方法 采用多阶段分层整群抽样方法调查南京市≥18岁的常住居民,调查时间为2017-01-01至2018-06-30。收集患者基本资料,采用多因素稳健Poisson回归模型评价CMI对MONW表型的RR值及其95%CI。绘制受试者工作特征(ROC)曲线评估相关指标对MONW表型的预测能力,采用DeLong检验比较各指标间的ROC曲线下面积(AUC),并进一步探讨不同性别、年龄分层CMI预测MONW表型的价值。 结果 共纳入30 408例研究对象,其中男13 213例,女17 195例;体质量正常代谢正常(MHNW)表型23 691例,MONW表型6 717例。MHNW表型和MONW表型研究对象年龄、受教育程度、职业、吸烟、饮酒、体力活动、静态行为时间、高红肉摄入、疾病史、用药史、身高、腰围(WC)、体质指数(BMI)、总胆固醇(TC)、三酰甘油(TG)、高密度脂蛋白胆固醇(HDL-C)、低密度脂蛋白胆固醇(LDL-C)、收缩压(SBP)、舒张压(DBP)、空腹血糖(FPG)、TG/HDL-C、腰高比(WHtR)和CMI比较,差异有统计学意义(P<0.05);男性MHNW表型研究对象和MONW表型年龄、受教育程度、职业、吸烟、饮酒、静态行为时间、高红肉摄入、疾病史、用药史、身高、WC、BMI、TC、TG、HDL-C、LDL-C、SBP、DBP、FPG、TG/HDL-C、WHtR和CMI比较,差异有统计学意义(P<0.05);女性MHNW表型研究对象和MONW表型年龄、受教育程度、职业、吸烟、饮酒、体力活动、静态行为时间、疾病史、用药史、身高、WC、BMI、TC、TG、HDL-C、LDL-C、SBP、DBP、FPG、TG/HDL-C、WHtR和CMI比较,差异有统计学意义(P<0.05)。所有研究对象Q1~Q4组例数分别为7 739、7 940、7 904、6 825例,CMI范围分别为≤0.253、0.254~0.382、0.383~0.539、≥0.540。男性研究对象Q1~Q4组例数分别为2 697、3 410、3 661、3 445例,CMI范围分别为≤0.281、0.282~0.407、0.408~0.569、≥0.570。女性研究对象Q1~Q4组例数分别为5 042、4 530、4 243、3 380例,CMI范围分别为≤0.235、0.236~0.361、0.362~0.516、≥0.517。校正各项混杂因素后,CMI四分位数分组是所有研究对象、男性研究对象和女性研究对象代谢表型的影响因素(P<0.05)。多因素稳健Poisson回归模型分析显示CMI每增加1个SD,总人群、男性和女性MONW表型发生的风险分别增加68%、55%、81%。男性研究对象中CMI对MONW表型的预测能力高于WHtR(Z=18.97,P<0.001)、TG/HDL-C(Z=12.53,P<0.001)、WC(Z=23.85,P<0.001)和BMI(Z=24.13,P<0.001);女性研究对象中CMI对MONW表型的预测能力高于WHtR(Z=27.38,P<0.001)、TG/HDL-C(Z=15.27,P<0.001)、WC(Z=30.83,P<0.001)和BMI(Z=30.84,P<0.001)。女性研究对象CMI预测MONW表型的AUC大于男性(Z=-6.10,P<0.001)。在男性研究对象中,18~34岁CMI预测MONW表型的AUC为0.835〔95%CI(0.818,0.852)〕,高于35~44岁(Z=1.55,P=0.04)、45~54岁(Z=6.92,P<0.001)、55~64岁(Z=4.95,P<0.001)、≥65岁(Z=7.92,P<0.001);在女性研究对象中,18~34岁CMI预测MONW表型的AUC为0.832〔95%CI(0.817,0.847)〕,高于35~44岁(Z=1.95,P=0.03)、45~54岁(Z=2.56,P=0.02)、55~64岁(Z=3.79,P<0.001)、≥65岁(Z=5.71,P<0.001)。 结论 CMI与MONW表型的患病风险呈正相关,且CMI具有较强的预测效能,可作为识别体质量正常人群中MONW表型的有效工具,尤其适用于18~34岁人群。

关键词: 代谢疾病, 血脂异常, 心脏代谢指数, 受试者工作特征曲线, 预测价值

Abstract:

Background

Cardiometabolic index (CMI) is a simple index to measure blood lipid, which is closely related to diabetes and stroke. Metabolically obese normal weight (MONW) individuals have higher risks of morbidity and mortality of diabetes and cardiovascular and cerebrovascular diseases. Correctly identifying individuals with MONW phenotype is essential for the prevention and control of metabolism-related diseases. However, there are few studies on the predictive value of CMI for MONW phenotype.

Objective

To investigate the association between CMI and MONW phenotype, and to evaluate the predictive value of CMI for MONW phenotype.

Methods

The multistage stratified cluster sampling method was used to select permanent residents aged ≥18 years as subjects from Nanjing. The investigation time was from January 1, 2017 to June 30, 2018. The basic data of subjects were collected and multivariate robust Poisson regression model was used to evaluate the RR value with 95%CI of CMI for MONW phenotype. The receiver operating characteristic (ROC) curve was used to evaluate the predictive ability of CMI, waist-to-height ratio (WHtR) , triglyceride/ high-density lipoprotein cholesterol (TG/HDL-C) ratio, waist circumference (WC) and body mass index (BMI) for MONW phenotype. DeLong test was used to compare the area under the ROC curve (AUC) of the above-mentioned five indicators, and to further explore the value of CMI in predicting MONW phenotype in different gender and age groups.

Results

A total of 30 408 people were included, including 13 213 males and 17 195 females, 23 691 cases of MHNW and 6 717 cases of MONW. There was statistically significant difference in age, education, occupation, current smoking, alcohol consumption, physical activity, duration of static behavior, high red meat intake, disease history, medication history, height, waist circumference (WC) , body mass index (BMI) , total cholesterol (TC) , triglyceride (TG) , high density lipoprotein cholesterol (HDL-C) , low density lipoprotein cholesterol (LDL-C) , systolic blood pressure (SBP) , diastolic blood pressure (DBP) , fasting blood glucose (FPG) , TG/HDL-C, waist-height ratio (WHtR) and CMI (P<0.05) . There was statistically significant difference in age, education, occupation, current smoking, alcohol consumption, duration of static behavior, high red meat intake, history of disease, medication, height, WC, BMI, TC, TG, HDL-C, LDL-C, SBP, DBP, FPG, TG/HDL-C, WHtR, and CMI of male MHNW and NONW phenotypes (P<0.05) . There was statistically significant difference in age, education, occupation, current smoking, alcohol consumption, physical activity, duration of static behavior, history of disease, medication, height, WC, BMI, TC, TG, HDL-C, LDL-C, SBP, DBP, FPG, TG/HDL-C, WHtR, and CMI of the female MHNW and NOWN phenotype subjects (P<0.05) . The number of Q1 to Q4 groups was 7 739, 7 940, 7 904, 6 825, and the CMI range was ≤0.253, 0.254 to 0.382, 0.383 to 0.539, and ≥0.540, respectively. Male subjects in Q1 to Q4 were 2 697, 3 410, 3 661, 3 445, and the CMI range was ≤0.281, 0.282 to 0.407, 0.408 to 0.569, and ≥0.570, respectively. 5 042, 4 530, 4 243 and 3 380 female subjects in Q1 to Q4 group were studied, and the CMI ranges were ≤0.235, 0.236-0.361, 0.362-0.516 and ≥0.517, respectively. After adjusting for confounding factors, the CMI quartile grouping was the factor affecting metabolic phenotype in all subjects, male subjects, and female subjects (P<0.05) . Multivariate robust Poisson regression model analysis showed that the risk of MONW phenotype in the general population, male and female increased by 68%, 55% and 81% with each additional SD of CMI. In male subjects, CMI predicted MONW phenotype better than WHtR (Z=18.97, P<0.001) , TG/HDL-C (Z=12.53, P<0.001) , WC (Z=23.85, P<0.001) and BMI (Z=24.13, P<0.001) . The predictive power of CMI for MONW phenotype in female subjects was higher than that of WHtR (Z=27.38, P<0.001) , TG/HDL-C (Z=15.27, P<0.001) , WC (Z=30.83, P<0.001) and BMI (Z=30.84, P<0.001) . The AUC value of CMI predicted MONW phenotype in female subjects was higher than that in male subjects (Z=-6.10, P<0.001) , and the difference was statistically significant. In male subjects, the AUC predicted by CMI from 18 to 34 years old was 0.835〔95%CI (0.818, 0.852) 〕, higher than that of 35 to 44 years old (Z=1.55, P=0.04) , 45 to 54 years old (Z=6.92, P<0.001) , 55 to 64 years old (Z=4.95, P<0.001) , ≥65 years old (Z=7.92, P<0.001) ; In female subjects, the AUC predicted by CMI from 18 to 34 years old was 0.832〔95%CI (0.817, 0.847) 〕, which was higher than that of 35 to 44 years old (Z=1.95, P=0.03) , 45 to 54 years old (Z=2.56, P=0.02) , 55 to 64 years old (Z=3.79, P<0.001) , ≥65 years old (Z=5.71, P<0.001) .

Conclusion

CMI was positively associated with the risk of the MONW phenotype, which has strong predictive power and can be used as an effective tool to identify MONW phenotype in the general population, especially in 18-34 years-old people.

Key words: Metabolic diseases, Dyslipidemias, Cardiometabolic index, Receiver operating characteristic curve, Predictive value