基于临床检验指标的脑梗死合并高血压发生风险的预测模型构建

吴璇; 韦文婕; 陈琳; 朱明要; 韩玮玮; 黄晴; 方家敏

doi:10.16766/j.cnki.issn.1674-4152.004541

基于临床检验指标的脑梗死合并高血压发生风险的预测模型构建

doi: 10.16766/j.cnki.issn.1674-4152.004541

1.
蚌埠医科大学第二附属医院检验科，安徽蚌埠 233080
2.
蚌埠医科大学公共卫生学院，安徽蚌埠 233030

基金项目:

安徽省高校自然科学研究重点项目 2023AH051915

企业委托横向项目 2026HX0301012

详细信息

通讯作者:
方家敏，E-mail：fjm5250@163.com

中图分类号: R743.3 R544.1
计量
- 文章访问数: 9
- HTML全文浏览量: 3
- PDF下载量: 0
- 被引次数: 0
出版历程
- 收稿日期: 2025-12-30
- 网络出版日期: 2026-11-10

Construction of a risk prediction model for cerebral infarction complicated with hypertension based on clinical laboratory indicators

1.
Department of Clinical Laboratory, the Second Affiliated Hospital of Bengbu Medical University, Bengbu, Anhui 233080, China

摘要

摘要: 目的通过临床检验指标，建立基于Lasso-logistic回归的脑梗死合并高血压发生风险的预测模型，并评估该模型的诊断价值。方法纳入2022年1月—2024年12月于蚌埠医科大学第二附属医院收治的127例脑梗死患者，按照是否合并既往高血压病史分为脑梗死组（CI，72例）与脑梗死合并高血压组（CIHT，55例）。比较2组患者临床资料和检验指标，通过多因素logistic回归筛选脑梗死合并高血压病发生的危险因素，并建立Nomogram模型，绘制ROC曲线分析其预测价值。结果 CI组患者美国国立卫生研究院卒中量表（NIHSS）评分、BMI、CRP、D-二聚体和中性粒细胞与淋巴细胞计数比值（NLR）明显低于CIHT组（P < 0.05）；多因素logistic回归模型分析结果提示，BMI、CRP、D-二聚体、NLR与NIHSS评分均为脑梗死合并高血压病的影响因素；基于以上5个独立影响因素构建Nomogram模型，并绘制ROC曲线，其AUC为0.885，灵敏度为77.8%，特异度为88.9%，约登指数为0.667，表明该模型区分性能表现良好；校准曲线分析显示，平均绝对误差（MAE）控制在0.021，且校准曲线贴合理想曲线，表示模型具有较好的校准性能。结论脑梗死合并高血压病发生的影响因素包括BMI、CRP、D-二聚体、NLR和NIHSS评分，以此构建的Nomogram预测模型校准度良好，且具有临床实用性。
- 脑梗死合并高血压 /
- 检验指标 /
- Nomogram模型 /
- 风险预测
Abstract: Objective To establish a prediction model for evaluating the occurrence risk of cerebral infarction combined with hypertension based on Lasso-logistic regression through clinical indicators, and to evaluate the diagnostic value of the model. Methods A total of 127 confirmed patients of cerebral infarction admitted to the Second Affiliated Hospital of Bengbu Medical University from January 2022 to December 2024 were included. Based on the presence or absence of a history of hypertension, patients were assigned to two groups: the cerebral infarction group (CI) and the cerebral infarction complicated with hypertension group (CIHT). Clinical data and laboratory indicators were compared, and multivariate logistic regression was utilized to analyze the risk factors associated with cerebral infarction accompanied by hypertension, and the Nomogram model was established. The ROC curve was drawn to analyze its predictive value. Results The National Institute of Health stroke scale (NIHSS) scores, BMI, CRP, D-dimer, and neutrophil-lymphocyte ratio (NLR) in the CI group were significantly lower than those in the CIHT group (P < 0.05). The results of the multivariate logistic regression revealed that BMI, CRP, D-dimer, NLR, and NIHSS were all risk factors for cerebral infarction complicated with hypertension. Based on these five independent influencing factors, the Nomogram model was constructed, and the ROC curve was drawn. The AUC was 0.885, the sensitivity was 77.8%, the specificity was 88.9%, and the Youden's index was 0.667, reflecting the excellent discriminatory ability of the established model. The calibration curve results showed that the mean absolute error (MAE) was 0.021, and the calibration curve was highly consistent with the ideal curve, demonstrating a satisfactory calibration of the model. Conclusion BMI, CRP, D-dimer, NLR, and NIHSS scores were identified as risk factors for cerebral infarction complicated with hypertension. The constructed Nomogram prediction model has a good calibration and clinical practicality.
- Cerebral infarction complicated with hypertension /
- Clinical laboratory indicators /
- Nomogram /
- Risk prediction

HTML全文

图 1 Lasso回归的特征变量选择

注：A为十倍交叉验证图；B为收缩系数图。

Figure 1. Feature variable selection in Lasso regression

下载: 全尺寸图片幻灯片

图 2 脑梗死合并高血压发生风险因素的Nomogram模型构建

Figure 2. Construction of Nomogram for predicting the risk of cerebral infarction complicated with hypertension

下载: 全尺寸图片幻灯片

图 3 Nomogram列线图预测模型的校准曲线

Figure 3. Calibration curve of the nomogram prediction model

下载: 全尺寸图片幻灯片

图 4 Nomogram列线图预测模型预测脑梗死合并高血压发生风险的ROC曲线

Figure 4. ROC curve of the Nomogram prediction model for predicting the risk of hypertension in patients with cerebral infarction

下载: 全尺寸图片幻灯片

表 1 2组脑梗死患者临床资料与检验指标比较

Table 1. Comparison of clinical data and test indicators between the two groups of patients with cerebral infarction

组别	例数	性别(男/女，例)	年龄(x±s，岁)	BMI (x±s)	CRP [M(P₂₅, P₇₅)，mg/L]	D-二聚体(x±s，mg/L)	NLR (x±s)	NIHSS评分[M(P₂₅, P₇₅)，分]	糖尿病[例(%)]
CI组	72	27/45	63.53±12.72	22.25±3.45	6.40(3.20, 8.50)	1.38±0.49	2.85±1.09	5.00(3.00, 7.00)	19(26.4)
CIHT组	55	22/33	67.67±11.15	24.91±2.71	10.50(4.50, 14.10)	2.66±0.73	3.56±1.57	11.00(6.00, 15.00)	17(30.9)
统计量		0.082^a	1.923^b	4.692^b	-3.700^c	11.653^b	2.991^b	-6.212^c	0.312^a
P值		0.771	0.063	0.014	0.001	0.001	0.003	0.001	0.581
注：^a为χ²值，^b为t值，^c为Z值。

下载: 导出CSV

表 2 脑梗死合并高血压发生影响因素的logistic回归分析

Table 2. Logistic regression analysis of influencing factors for cerebral infarction combined with hypertension

变量	B	SE	Waldχ²	P值	OR(95% CI)
BMI	0.214	0.076	2.712	0.001	1.239(1.058~1.442)
CRP	0.132	0.063	2.164	0.036	1.141(1.009~1.276)
D-二聚体	0.556	0.268	2.038	0.042	1.743(1.018~2.978)
NLR	0.513	0.213	2.431	0.022	1.670(1.113~2.528)
NIHSS	0.121	0.052	6.597	0.003	1.129(1.032~1.243)

下载: 导出CSV

参考文献(23)

[1]	CHEN L Y, LIU F, TIAN X, et al. Impact of cerebral microbleeds on cognitive functions and its risk factors in acute cerebral infarction patients[J]. Neurol Res, 2023, 45(6): 564-571. doi: 10.1080/01616412.2022.2164456
[2]	江应露, 张苗苗, 张涛, 等. 平肝熄风化痰通络方对老年脑梗死急性期合并高血压患者的临床疗效[J]. 中成药, 2025, 47(10): 3274-3278.
[3]	BAI X, FU Z, SUN Z, et al. Thrombectomy using the EmboTrap clot-retrieving device for the treatment of acute ischemic stroke: a glimpse of clinical evidence[J]. Am J Neuroradiol, 2022, 43(12): 1736-1742. doi: 10.3174/ajnr.A7708
[4]	ZHENG D L, LI X M, FU Y P. Risk factors of acute cerebral infarction in patients with primary hypertension[J]. Ir J Med Sci, 2023, 192(5): 2441-2445. doi: 10.1007/s11845-022-03206-4
[5]	黄晴, 李孟磊, 胡富勇. 急性缺血性脑卒中病人rt-PA静脉溶栓后预后不良的列线图风险预测模型构建与评估[J]. 蚌埠医科大学学报, 2024, 49(6): 731-735. Huang Q, Li M L, Hu F Y. Construction and evaluation of a nomogram risk prediction model for poor prognosis in patients with acute ischemic stroke after intravenous rt-PA thrombolysis[J]. Journal of Bengbu Medical University, 2024, 49(6): 731-735.
[6]	ZHENG H D, HU Y H, YE K, et al. Development and validation of a nomogram for preoperative prediction of tumor deposits in colorectal cancer[J]. World J Gastroenterol, 2023, 29(39): 5483-5493. doi: 10.3748/wjg.v29.i39.5483
[7]	许静, 张蕊, 王化彬, 等. 超早产儿早期死亡风险列线图预测模型的建立[J]. 中华围产医学杂志, 2024, 27(5): 394-401. Xu J, Zhang R, Wang H B, et al. Establishment of a nomogram prediction model for early mortality risk in extremely preterm infants[J]. Chinese Journal of Perinatal Medicine, 2024, 27(5): 394-401.
[8]	中国卒中学会, 中国卒中学会神经介入分会, 中华预防医学会卒中预防与控制专业委员会介入学组. 急性缺血性卒中血管内治疗中国指南2023[J]. 中国卒中杂志, 2023, 18(6): 684-711. Chinese Stroke Society, Chinese Stroke Society Neurointervention Branch, Interventional Group of Stroke Prevention and Control Committee, Chinese Preventive Medicine Association. Chinese guidelines for intravascular diagnosis and treatment of acute ischemic stroke 2023[J]. Chinese Journal of Stroke, 2023, 18(6): 684-711.
[9]	SHEN Z Y, BAO N N, TANG M, et al. Tenecteplase vs. alteplase for intravenous thrombolytic therapy of acute ischemic stroke: a systematic review and meta-analysis[J]. Neurol Ther, 2023, 12(5): 1553-1572. doi: 10.1007/s40120-023-00530-4
[10]	迪丽达热·艾尔肯, 西合热扎提·阿布力米提, 田睿思. 急性缺血性脑卒中在院内发生的风险因素及预测价值分析[J]. 中华全科医学, 2025, 23(10): 1643-1646. Aierken D, Abulimiti X, Tian R S. Analysis of influencing factors and predictive value of in-hospital acute ischemic stroke[J]. Chinese Journal of General Practice, 2025, 23(10): 1643-1646.
[11]	金哲宇, 刘政, 吕建斌. NCCT-CTA-CTP一站式检查定量参数对急性脑梗死溶栓患者的预后评估价值[J]. 现代实用医学, 2022, 34(8): 1031-1034. Jin Z Y, Liu Z, Lyu J B. The prognostic assessment value of quantitative parameters from NCCT-CTA-CTP one-stop examination for patients with acute cerebral infarction undergoing thrombolysis[J]. Modern Practical Medicine, 2022, 34(8): 1031-1034.
[12]	郝井兰, 张成刚, 孙京华, 等. 高同型半胱氨酸水平对急性脑梗死患者神经功能、认知功能及预后的影响[J]. 西部医学, 2025, 37(6): 866-870. Hao J L, Zhang C G, Sun J H, et al. The impact of high homocysteine levels on neurological function, cognitive function, and prognosis in patients with acute cerebral infarction[J]. Medical Journal of West China, 2025, 37(6): 866-870.
[13]	房硕, 董兰真, 刘洋洋, 等. HO-1、sLOX-1水平与急性缺血性脑卒中患者神经功能、认知功能的关系[J]. 西部医学, 2025, 37(9): 1386-1389, 1395. Fang S, Dong L Z, Liu Y Y, et al. The relationship between HO-1, sLOX-1 levels and neurological and cognitive functions in patients with acute ischemic stroke[J]. Medical Journal of West China, 2025, 37(9): 1386-1389, 1395.
[14]	GAO Y, LIU K D, FANG S K. Trend analysis of stroke subtypes mortality attributable to high body-mass index in China from 1990 to 2019[J]. BMC Public Health, 2024, 24(1): 2155-2160. doi: 10.1186/s12889-024-19615-2
[15]	旷南岳, 李辉, 林娟, 等. 新疆生产建设兵团60699例成年居民身体质量指数与高血压、糖尿病和血脂异常患病率的相关分析[J]. 安徽医药, 2023, 27(12): 2425-2428. Kuang N Y, Li H, Lin J, et al. Correlation analysis of body mass index with the prevalence of hypertension, diabetes, and dyslipidemia in 60 699 adult residents of the Xinjiang Production and Construction Corps[J]. Anhui Medical and Pharmaceutical Journal, 2023, 27(12): 2425-2428.
[16]	杨光, 陈奎. 急性缺血性脑卒中患者血清D-二聚体、中性粒细胞与淋巴细胞比值、C反应蛋白/白蛋白与静脉溶栓预后的相关性分析[J]. 临床内科杂志, 2024, 41(8): 544-546. Yang G, Chen K. Correlation analysis of serum D-dimer, neutrophil-to-lymphocyte ratio, C-reactive protein/albumin ratio, and intravenous thrombolysis prognosis in patients with acute ischemic stroke[J]. Journal of Clinical Internal Medicine, 2024, 41(8): 544-546.
[17]	王刚, 仲婷婷, 孙炎, 等. 免疫及炎性因子与急性缺血性卒中后神经损伤的相关性分析[J]. 临床急诊杂志, 2022, 23(2): 91-94. Wang G, Zhong T T, Sun Y, et al. Correlation analysis between immune and inflammatory factors and neurological injury after acute ischemic stroke[J]. Journal of Clinical Emergency Medicine, 2022, 23(2): 91-94.
[18]	TATLISULUOGLU D, TEZCAN B, MUNGAN I, et al. Predicting postoperative ischemic stroke problems in patients following coronary bypass surgery using neutrophil-lymphocyteratio, platelet-lymphocyte ratio, and red blood cell distribution width values[J]. Kardiochir Torakochirurgia Pol, 2022, 19(2): 90-95.
[19]	王学成, 许晓红, 王中新. C反应蛋白/清蛋白比值与急性缺血性脑卒中严重程度相关性分析[J]. 国际检验医学杂志, 2023, 44(6): 749-751. Wang X C, Xu X H, Wang Z X. Correlation analysis between C-reactive protein/albumin ratio and severity of acute ischemic stroke[J]. International Journal of Laboratory Medicine, 2023, 44(6): 749-751.
[20]	张琳翊, 王丽琨, 伍国锋, 等. 中性粒细胞与淋巴细胞比值联合血肿量对脑干出血患者预后的预测价值[J]. 贵州医科大学学报, 2025, 50(12): 1806-1812. Zhang L Y, Wang L K, Wu G F, et al. The prognostic predictive value of the neutrophil to lymphocyte ratio combined with hematoma volume in patients with brainstem hemorrhage[J]. Journal of GuiZhou Medical University, 2025(12): 1806-1812.
[21]	LATTANZI S, NORATA D, BROGGI S, et al. Neutrophil-to-lymphocyte ratio predicts early neurological deterioration after endovascular treatment in patients with ischemic stroke[J]. Life(Basel), 2022, 12(9): 1415. DOI: 10.3390/life12091415.
[22]	朱晓雯, 徐嘉蔚, 王玉飞, 等. 出血性和缺血性脑卒中患者凝血、炎症指标水平变化及诊断价值分析[J]. 国际检验医学杂志, 2021, 42(19): 2317-2320. Zhu X W, Xu J W, Wang Y F, et al. Analysis of changes in coagulation and inflammatory marker levels and diagnostic value in patients with hemorrhagic and ischemic stroke[J]. International Journal of Laboratory Medicine, 2021, 42(19): 2317-2320.
[23]	马慧, 杜和宝, 苏增锋, 等. 急性脑梗死患者血浆D-二聚体水平对病情及预后的影响[J]. 神经损伤与功能重建, 2022, 17(9): 547-549. Ma H, Du H B, Su Z F, et al. The impact of plasma D-dimer levels on the condition and prognosis of patients with acute cerebral infarction[J]. Neural Injury and Functional Reconstruction, 2022, 17(9): 547-549.