作者：谢姚屹，黄妍菊，樊敏，岑花桃，黄志林，李燕丽，卢苗值，吴志武

单位：

530003广西壮族自治区南宁市第三人民医院神经科二区

单位（英文）：

Neurology Division 2, the Third People's Hospital of Nanning, Nanning 530003, China

关键词：

高血压；脑梗死；早期神经功能恶化；影响因素分析；列线图

关键词（英文）：

Hypertension; Brain infarction; Early neurological deterioration; Root cause analysis; Nomograms

中图分类号：

R 7544.1 R 743.33

DOI：

10.12114/j.issn.1008-5971.2024.00.079

基金项目：

广西壮族自治区卫生健康委员会自筹经费科研课题（Z-A20221198）

摘要：

目的　探讨H型高血压合并脑梗死患者发生早期神经功能恶化（END）的影响因素，并构建其风险 预测列线图模型。方法　选取2022年7月—2023年3月南宁市第三人民医院收治的150例H型高血压合并脑梗死患者为 建模集，另选取2023年4—12月南宁市第三人民医院收治的100例H型高血压合并脑梗死患者为验证集。收集患者的 临床资料，以脑梗死发病24～72 h内美国国立卫生研究院卒中量表（NIHSS）评分较入院时增加≥2分或死亡判定为 发生END。采用多因素Logistic回归分析探讨H型高血压合并脑梗死患者发生END的影响因素；采用Rstudio软件建立 H型高血压合并脑梗死患者发生END的风险预测列线图模型；进行Hosmer-Lemeshow拟合优度检验，绘制校准曲线、 ROC曲线及决策曲线。结果　建模集150例H型高血压合并脑梗死患者发生END 39例（26.00%）。未发生END者与 发生END者年龄、有吸烟史者占比和入院时NIHSS评分、同型半胱氨酸（Hcy）、中性粒细胞计数、超敏C反应蛋白 （hs-CRP）、TG、总胆红素（TBiL）、D-二聚体及亚甲基四氢叶酸还原酶（MTHFR）C677T基因型分布比较，差异 有统计学意义（P＜0.05）。多因素Logistic回归分析结果显示，年龄和入院时NIHSS评分、Hcy、中性粒细胞计数、 hs-CRP、D-二聚体及MTHFR C677T基因型分布为H型高血压合并脑梗死患者发生END的独立影响因素（P＜ 0.05 ）。 Hosmer-Lemeshow拟合优度检验及校准曲线分析结果显示，建模集与验证集中该列线图模型拟合较好（χ 2 = 3.828 ， P=0.872；χ2 =4.220，P=0.838）。ROC曲线分析结果显示，该列线图模型预测建模集与验证集H型高血压合并脑梗 死患者发生END的AUC分别为0.830〔95%CI（0.764～0.896）〕、0.784〔95%CI（0.678～0.890）〕。决策曲线分析 结果显示，在建模集中，当阈值概率＞0.05时，H型高血压合并脑梗死患者的净获益率＞0；在验证集中，当阈值概 率＞0.10时，H型高血压合并脑梗死患者的净获益率＞0。结论　年龄和入院时NIHSS评分、Hcy、中性粒细胞计数、 hs-CRP、D-二聚体及MTHFR C677T基因型分布为H型高血压合并脑梗死患者发生END的独立影响因素，基于上述因素 构建的列线图模型对H型高血压合并脑梗死患者发生END具有较好的区分能力和校准度及良好的临床净获益。

英文摘要：

Objective To explore the influencing factors of early neurological deterioration (END) in patients with H-type hypertension and cerebral infarction, and to construct the nomogram model for predicting its risk. Methods A total of 150 patients with H-type hypertension and cerebral infarction admitted to the Third People's Hospital of Nanning from July 2022 to March 2023 were selected as the modeling set, and 100 patients with H-type hypertension and cerebral infarction admitted to the Third People's Hospital of Nanning from April to December 2023 were selected as the validation set. The clinical data of the patients were collected, and the National Institutes of Health Stroke Scale (NIHSS) score increased by ≥ 2 points or death within 24-72 hours of cerebral infarction was determined as the occurrence of END. Multivariate Logistic regression analysis was used to explore the influencing factors of END in patients with H-type hypertension and cerebral infarction. The nomogram model for predicting the risk of END in patients with H-type hypertension and cerebral infarction was constructed by using the Rstudio software. Hosmer-Lemeshow goodness of fit test was conducted and calibration curve, ROC curve and decision curve were drawn. Results In the modeling set, among 150 patients with H-type hypertension and cerebral infarction, 39 cases (26.00%) experienced END. There were significant differences in age, proportion of patients with drinking history, and NIHSS score at admission, homocysteine (Hcy) , neutrophil count, hypersensitive C-reactive protein (hs-CRP) , TG, total bilirubin (TBiL) and D-dimer and methylene tetrahydrofolate reductase (MTHFR) C677T genotype distribution between the patients without END and patients with END (P < 0.05) . Multivariate Logistic regression analysis showed that age, and NIHSS score at admission, Hcy, neutrophil count, hs-CRP and D-dimer and MTHFR C677T genotype distribution were the independent influencing factors of END in patients with H-type hypertension and cerebral infarction (P < 0.05) . The results of Hosmer-Lemeshow goodness of fit test and calibration curve analysis showed that the nomogram model fitted well in modeling set ( χ 2 =3.828, P=0.872) and validation set (χ 2 =4.220, P=0.838) . The results of ROC curve analysis showed that the AUC of the nomogram model for predicting END in patients with H-type hypertension and cerebral infarction in modeling set and validation set was 0.830 [95%CI (0.764-0.896) ] , 0.784 [95%CI (0.678-0.890) ] , respectively. The results of decision curve analysis showed that when the high risk threshold was > 0.05 in the modeling set, the net benefit rate of patients with H-type hypertension and cerebral infarction was > 0; when the high risk threshold was > 0.10 in the validation set, the net benefit rate of patients with H-type hypertension and cerebral infarction was > 0. Conclusion Age, and NIHSS score at admission, Hcy, neutrophil count, hs-CRP, D-dimer and MTHFR C677T genotype distribution are the independent influencing factors of END in patients with H-type hypertension and cerebral infarction. The nomogram model constructed based on the above influencing factors has a high degree of discrimination and calibration for END in patients with H-type hypertension and cerebral infarction, and have good clinical benefits.

参考文献：