作者：沙吉旦·阿不都热衣木，宋星明，哈力沙·艾尔肯江，迪丽胡玛尔·阿布来提，帕丽达·阿布来提

单位：

830054新疆维吾尔自治区乌鲁木齐市，新疆医科大学第一附属医院综合内三科

Units：

Third Department of General Internal Medicine, the First Affiliated Hospital of Xinjiang Medical University, Urumqi 830054,China

关键词：

高血压；高同型半胱氨酸血症；H型高血压；影响因素分析；预测模型

Keywords：

Hypertension; Hyperhomocysteinemia; H-type hypertension; Root cause analysis; Prediction model

CLC：

R 544.1

DOI：

10.12114/j.issn.1008-5971.2023.00.172

Funds：

自治区科技支疆项目（2020E02131）

摘要：

　目的　分析H型高血压的影响因素，并构建其预测模型。方法　收集2020—2021年新疆医科大学第一附属医院综合内三科收治的高血压患者520例，根据血清同型半胱氨酸（Hcy）将其分为单纯高血压组（Hcy＜15μmol/L，n=381）和H型高血压组（Hcy≥15 μmol/L，n=139）。比较两组临床资料和实验室检查指标，高血压患者血清Hcy影响因素分析采用单因素、多因素线性回归分析，H型高血压的影响因素分析采用多因素Logistic回归分析。构建H型高血压的预测模型，并绘制ROC曲线以评估该预测模型对H型高血压的预测价值。结果　H型高血压组男性占比、脂蛋白a〔Lp（a）〕、血肌酐（Scr）、血尿酸（BUA）、胱抑素C（CysC）、白细胞计数、中性粒细胞计数高于单纯高血压组，年龄大于单纯高血压组，载脂蛋白A（ApoA）低于单纯高血压组（P＜0.05）。单因素线性回归分析结果显示，性别、年龄、冠心病史、ApoA、Scr、BUA、CysC、白细胞计数、中性粒细胞计数、血小板计数可能是高血压患者血清Hcy的影响因素（P＜0.05）。多因素线性回归分析结果显示，性别、年龄、BUA、CysC、中性粒细胞计数、血小板计数是高血压患者血清Hcy的影响因素（P＜0.05）。多因素Logistic回归分析结果显示，男性、年龄增长及BUA、CysC、中性粒细胞计数升高是H型高血压的危险因素（P＜0.05）。根据多因素Logistic回归分析结果，构建H型高血压的预测模型如下：Y=0.627×男性+0.027×年龄+0.003×BUA+0.854×CysC+0.031×中性粒细胞计数-6.830。ROC曲线分析结果显示，该预测模型预测H型高血压的AUC为0.706〔95%CI（0.656，0.760）〕，最佳截断值为0.273，灵敏度为63.3%，特异度为70.3%，约登指数为0.336。结论　性别、年龄、BUA、CysC、中性粒细胞计数、血小板计数是高血压患者血清Hcy的影响因素；男性、年龄增长及BUA、CysC、中性粒细胞计数升高是H型高血压的危险因素，而基于上述危险因素构建的预测模型对H型高血压具有一定预测价值。

Abstract：

　Objective To analyze the influencing factors of H-type hypertension, and construct its prediction model.Methods A total of 520 hypertensive patients admitted to the Third Department of General Internal Medicine of the FirstAffiliated Hospital of Xinjiang Medical University from 2020 to 2021 were collected. They were divided into simple hypertensiongroup [homocysteine (Hcy) < 15 μmol/L, n=381] and H-type hypertension group (Hcy ≥ 15 μmol/L, n=139) according to serumHcy. The clinical data and laboratory examination indexes were compared between the two groups. The influencing factors ofserum Hcy in hypertensive patients were analyzed by univariate and multivariate linear regression analysis, and the influencingfactors of H-type hypertension were analyzed by multivariate Logistic regression analysis. According to the results of multivariateLogistic regression analysis, the prediction model of H-type hypertension was constructed, and the ROC curve was drawn toevaluate the predictive value of the prediction model for H-type hypertension. Results The proportion of males, lipoprotein a[Lp (a) ] , serum creatinine (Scr) , blood uric acid (BUA) , cystatin C (CysC) , white blood cell count and neutrophil count in H-typehypertension group were higher than those in simple hypertension group, the age was older than that in simple hypertension group,and apolipoprotein A (ApoA) was lower than that in simple hypertension group (P < 0.05) . Univariate linear regression analysisresults showed that gender, age, history of coronary heart disease, ApoA, Scr, BUA, CysC, white blood cell count, neutrophil count and platelet count may be the influencing factors of serum Hcy in patients with hypertension (P < 0.05) . Multivariate linearregression analysis results showed that gender, age, BUA, CysC, neutrophil count and platelet count were the influencing factorsof serum Hcy in patients with hypertension (P < 0.05) . Multivariate Logistic regression analysis showed that male, and increasedage, BUA, CysC, neutrophil count were risk factors for H-type hypertension (P < 0.05) . According to the results of multivariateLogistic regression analysis, the prediction model of H-type hypertension was constructed and as follows : Y=0.627×male+0.027×age+0.003×BUA+0.854×CysC+0.031×neutrophil count-6.830. ROC curve analysis showed that the AUC of the predictionmodel predicting for H-type hypertension was 0.706 [95%CI (0.656, 0.760) ] , the optimal cutoff value was 0.273, the sensitivitywas 63.3%, the specificity was 70.3%, and the Youden index was 0.336. Conclusion Gender, age, BUA, CysC, neutrophil countand platelet count were the influencing factors of serum Hcy in patients with hypertension. Male, and increased age, BUA, CysC,neutrophil count are risk factors for H-type hypertension, and the prediction model constructed based on the above risk factors hascertain predictive value for H-type hypertension.

ReferenceList：