Construction of Risk Prediction Model for Cerebral Hemorrhage in Middle-Aged and Elderly Patients with Hypertension

TIAN Yanshou; YANG Jiaojiao; ZHOU Lihua; WANG Bingchong

doi:10.3969/j.issn.1004-437X.2024.21.012

Abstract：Objective To construct a risk prediction model for cerebral hemorrhage in middle-aged and elderly patients with hypertension,and explore its clinica implications.Methods A total of 228 newly diagnosed middle-aged and elderly patients with hypertensive intracerebral hemorrhage admitted to Luohe Central Hospital(the First People's Hospital of Luohe)from May 2023 to March 2024 were included in the occurrence group.A 1∶1 ratio was used to select 228 middle-aged and elderly patients with primary hypertension who did not have intracerebral hemorrhage during the same period for outpatient review and included in the non-occurrence group.General information and possible factors of intracerebral hemorrhage occurrence were collected for both groups of patients,and logistic regression analysis was used to analyze the influencing factors of intracerebral hemorrhage in middle-aged and elderly patients with hypertension.A nomogram prediction model was constructed and evaluated and validated.Results There were significant differences in gender,age ≥60 years,body mass index(BMI),smoking history,alcohol consumption history,diabetes,coronary heart disease,hyperlipidemia,hypertension classification,H-type hypertension,use of antiplatelet drugs,disease duration,health promotion lifestyle profile-Ⅱ(HPLP-Ⅱ)score,emotion dysregulation scale-short(EDS-S)score,and Pittsburgh sleep quality index(PSQI)score between the two groups of patients(P＜0.05).Logistic regression analysis showed that diabetes,coronary heart disease,hyperlipidemia,hypertension classification,H-type hypertension,use of antiplatelet drugs,EDS-S score,and PSQI score were risk factors for cerebral hemorrhage in elderly hypertensive patients,while HPLP-Ⅱ score was a protective factor(P＜0.05).Based on the results of logistic regression analysis,a nomogram risk prediction model was developed to evaluate and validate the model.The evaluation and validation results showed that the model had an area under the curve(AUC)of 0.827 for predicting cerebral hemorrhage in middle-aged and elderly hypertensive patients,with a calibration degree of 0.874.Conclusion Diabetes,coronary heart disease,hyperlipidemia,hypertension classification,H-type hypertension,application of antiplatelet drugs,EDS-S score,PSQI score,and HPLP-Ⅱ score are all independently associated with cerebral hemorrhage in elderly hypertensive patients.Based on these indicators,a nomogram risk prediction model can be constructed to have good predictive value for cerebral hemorrhage and can be used as an effective model for clinical prediction of hypertensive cerebral hemorrhage,and guide the follow-up clinical work.