原发性高血压视网膜静脉阻塞风险预警模型构建及验证

doi:10.3969/j.issn.1007-3205.2025.04.014

河北医科大学学报 ›› 2025, Vol. 46 ›› Issue (4): 459-464.doi: 10.3969/j.issn.1007-3205.2025.04.014

原发性高血压视网膜静脉阻塞风险预警模型构建及验证

新疆维吾尔自治区喀什地区第二人民医院眼科，新疆喀什 844000

出版日期:2025-04-25 发布日期:2025-04-17
作者简介:高玮（1988-），女，辽宁新宾人，新疆维吾尔自治区喀什地区第二人民医院副主任医师,医学硕士，从事眼科疾病诊治研究。
基金资助:
喀什地区应用技术研究与开发计划项目（KS2021083）

Construction and validation of a risk early warning model for retinal vein occlusion in patients with primary hypertension

Department of Ophthalmology, the Second People′s Hospital of Kashgar Prefecture, Xinjiang Uygur Autonomous Region, Kashgar 844000, China

Online:2025-04-25 Published:2025-04-17

摘要/Abstract

摘要： 目的构建原发性高血压视网膜静脉阻塞（retinal vein occlusion，RVO）风险预警模型并验证。
方法从2021年1月—2022年12月喀什地区第二人民医院收治的原发性高血压患者1 046例中筛查并发RVO患者32例为RVO组，另在剩余患者中随机选取50例为对照组。RVO患者根据黄斑中心视网膜厚度（central retinal thickness，CRT）是否＞369 μm分为高CRT组（n=10）和低CRT组（n=22），比较RVO组和对照组一般资料，比较高CRT组和低CRT组血浆血小板计数（blood platelet count，BPC）和血小板平均体积（mean platelet volume，MPV）水平，分析原发性高血压并发RVO的影响因素，根据Logistic回归分析结果构建原发性高血压并发RVO的风险预警模型并构建列线图，以原发性高血压并发RVO为阳性样本建立受试者工作特征（receiver operating characteristic，ROC）曲线，并经Hosmer-Lemeshow验证该风险预警模型的拟合度。
结果 RVO组BPC［（271.31±31.42）×109/L］、MPV［（9.52±0.88）fL］水平高于对照组［（256.38±28.45）×109/L、(7.85±0.73）fL,P＜0.05］。高CRT组BPC［（288.31±34.42）×109/L］、MPV［（10.84±0.82）fL］水平高于低CRT组［（262.38±24.45）×109/L、(8.59±0.65）fL,P＜0.05］。多因素Logistic回归分析结果显示，BPC、MPV是原发性高血压并发RVO的影响因素（P＜0.05）。ROC曲线表明，BPC、MPV联合预测的AUC值（AUC=0.928，95%CI：0.876~0.981)明显优于BPC（AUC=0.740，95%CI：0.629~0.851)、MPV（AUC=0.919，95%CI：0.862~0.976)单独预测，且经Hosmer-Lemeshow验证该风险模型的拟合度良好。
结论 BPC、MPV与原发性高血压并发RVO和RVO的严重程度显著相关，且构建的风险模型拟合度良好。

关键词: 原发性高血压, 视网膜静脉闭塞, 血小板计数, 平均血小板体积

Abstract: Objective To construct and validate a risk early warning model for retinal vein occlusion (RVO) in patients with primary hypertension.
Methods From January 2021 to December 2022, 32 patients with concurrent RVO were screened out from 1 046 patients with primary hypertension admitted to the Second People′s Hospital of Kashgar Prefecture, and they were designated as the RVO group. Another 50 patients were randomly selected from the remaining patients as the control group. The RVO patients were divided into a high central retinal thickness (CRT) group (n=10) and a low CRT group (n=22) according to whether the CRT of the macular center was greater than 369 μm. The general data between the RVO group and the control group were compared. The levels of blood platelet count (BPC) and mean platelet volume (MPV) in plasma were compared between the high CRT group and the low CRT group. The influencing factors of RVO complicated with primary hypertension were analyzed. According to the results of Logistic regression analysis, a risk early warning model for RVO complicated with primary hypertension was constructed, and a nomogram was drawn. Using RVO complicated with primary hypertension as the positive sample, a receiver operating characteristic (ROC) curve was established, and the goodness of fit of this risk early warning model was verified by Hosmer-Lemeshow test.
Results The levels of BPC ［(271.31±31.42)×109/L］ and MPV ［(9.52±0.88) fL］ in the RVO group were higher than those in the control group ［(256.38±28.45)×109/L, (7.85±0.73) fL, P＜0.05］. The levels of BPC ［(288.31±34.42)×109/L］ and MPV ［(10.84±0.82) fL］ in the high CRT group were higher than those in the low CRT group ［(262.38±24.45)×109/L, (8.59±0.65) fL, P＜0.05］. The results of multivariate Logistic regression analysis showed that BPC and MPV were the influencing factors of RVO complicated with primary hypertension (P＜0.05). The ROC curve indicated that the area under the curve (AUC) value of the combined prediction of BPC and MPV (AUC=0.928, 95%CI: 0.876-0.981) was significantly better than the individual predictions of BPC (AUC=0.740, 95%CI: 0.629-0.851) and MPV (AUC=0.919, 95%CI: 0.862-0.976). And the goodness of fit of this risk model was verified to be good by Hosmer-Lemeshow test.
Conclusion BPC and MPV are significantly correlated with RVO complicated with primary hypertension and the severity of RVO, and the constructed risk model has a good fit.

Key words: primary hypertension, retinal vein occlusion, blood platelet count, mean platelet volume

高玮, 艾力亚尔·艾尔肯, 迪力夏提·多力坤. 原发性高血压视网膜静脉阻塞风险预警模型构建及验证[J]. 河北医科大学学报, 2025, 46(4): 459-464.

GAO Wei, ELIYAR Erkin, DILSHAT Dolkun. Construction and validation of a risk early warning model for retinal vein occlusion in patients with primary hypertension[J]. Journal of Hebei Medical University, 2025, 46(4): 459-464.

[1]	杨丽妙1, 杨洪乐1, 郝丽颖2, 张曼娜1, 杜欢欢1, 胡蕊1. 显微镜视场数对血涂片细胞计数影响的研究 [J]. 河北医科大学学报, 2024, 45(4): 477-484.
[2]	朱宝莉, 江敏, 姜善雨. 血小板相关参数与早产儿动脉导管未闭的相关性分析 [J]. 河北医科大学学报, 2023, 44(4): 450-453,464.
[3]	刘伟, 王贺, 徐丽丽, 孟丽君, 陶萍, 李建辉. RDW、PLT、HCT与急性非静脉曲张性上消化道出血危险程度的关系及联合预测预后的效能[J]. 河北医科大学学报, 2022, 43(2): 145-149.
[4]	冯月华，赵万建，李军华，曹素芬. 不同病原菌新生儿败血症患儿血糖异常、血小板计数、白细胞计数及C反应蛋白的变化研究[J]. 河北医科大学学报, 2020, 41(4): 422-425.
[5]	彭凌云，丁怀胜，李〓原，肖大宴，刘安康，王〓惠. 氯吡格雷低反应的高危冠心病患者改用替格瑞洛抗血栓治疗效果观察[J]. 河北医科大学学报, 2016, 37(12): 1370-1373,1382.
[6]	梁丽;单明华;徐斌;王杰;杨露. 传统格栅样激光与577nm 黄色微脉冲激光光凝治疗视网膜静脉阻塞并发黄斑水肿的疗效比较[J]. , 2015, 36(3): 356-356.
[7]	高见;郝玉华;闫配;张莉. 循环内皮细胞计数在非缺血型视网膜中央静脉阻塞转变为缺血型患者的表达[J]. , 2013, 34(7): 777-777.
[8]	李波;史敏;宋军;高羽高;王燕. EDTA依赖性假性血小板减少的临床和实验室分析[J]. , 2012, 33(9): 1028-1028.
[9]	单志明;马景学;张一凡. 激光治疗视网膜黄斑分支静脉阻塞[J]. , 2011, 32(8): 905-905.
[10]	刘丽娅;马景学. 视网膜中央静脉阻塞手术治疗的新进展-放射状视神经切开术[J]. , 2007, 28(6): 450-450.
[11]	. 急性脑梗死患者血小板参数的变化及应用[J]. , 2006, 27(4): 285-285.
[12]	. 肾综合征出血热患者血小板参数和功能指标的变化及临床意义[J]. , 2005, 26(6): 705-705.
[13]	苏军华;朱芸;张静. 血细胞分析仪血小板计数影响因素的探讨[J]. , 2003, 24(5): 300-300.
[14]	邱健芬;莫坚. 新生儿缺氧缺血性脑病与血小板参数变化的关系(附96例报告)[J]. , 2003, 24(5): 297-297.

原发性高血压视网膜静脉阻塞风险预警模型构建及验证

Construction and validation of a risk early warning model for retinal vein occlusion in patients with primary hypertension

HTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 14

编辑推荐

Metrics

本文评价