Construction and validation of a prediction model for UGIB in patients receiving dual antiplatelet therapy after PCI
10.3969/j.issn.1009-0126.2025.05.012
- VernacularTitle:经皮冠状动脉介入治疗后双联抗血小板治疗发生上消化道出血的模型构建与验证
- Author:
Jingyi YANG
1
;
Bing PAN
1
;
Shengxiang FENG
1
;
Jiexin MING
1
;
Hongwei YU
1
Author Information
1. 121000 锦州医科大学研究生培养基地(锦州市中心医院)心血管内科
- Publication Type:Journal Article
- Keywords:
percutaneous coronary intervention;
hemorrhage;
risk factors;
logistic models
- From:
Chinese Journal of Geriatric Heart Brain and Vessel Diseases
2025;27(5):596-600
- CountryChina
- Language:Chinese
-
Abstract:
Objective To explore the risk factors for UGIB after dual antiplatelet therapy(aspirin+P2Y12 receptor antagonist)in elderly patients with acute coronary syndrome(ACS)undergoing PCI,and then construct a nomogram model.Methods A total of 1590 elderly patients diagnosed with ACS and then undergoing PCI in Jinzhou Central Hospital from January 2019 to June 2022 were retrospectively recruited,and randomly divided into a testing group(n=1114)and a valida-tion group(n=476)in a ratio of 7∶3.Univariate and multivariate logistic regression analyses were applied to identify the independent risk factors for UGIB events,and then a prediction model was constructed and verified for its diagnostic performance.The predictive value of our prediction model for UGIB events was compared with that of PRECISE-DAPT scoring system.Results There were no statistical differences in baseline data between the testing group and the validation group(P＞0.05).The incidence of UGIB was 3.9％(62/1590)in the patients.Multivariate logistic regression analysis indicated that age(OR=1.064,95％CI:1.032-1.097,P=0.000),alcohol(OR=2.433,95％CI:1.220-4.823,P=0.011),heart failure(OR=3.734,95％CI:1.882-7.404,P=0.000),gastrointestinal ulcer/bleeding(OR=3.030,95％CI:1.391-6.618,P=0.005),and Cr(OR=1.017,95％CI:1.014-1.040,P=0.000)were independent risk factors for UGIB in these patients after dual antiplatelet therapy.The constructed nomogram model based on the risk factors obtained an AUC value of 0.806(95％CI:0.739-0.872)in the testing set and 0.838(95％CI:0.737-0.945)in the validation set.The AUC value of the PRECISE-DAPT scoring system in predicting UGIB was 0.674(95％CI:0.583-0.766),which was significantly lower than our mod-el(P＜0.05),indicating the nonogram showing good discriminability.Calibration curve analysis and H-L goodness of fit test revealed that our model had good consistency and was well fit(vali-dation set:P=0.846,testing set:P=0.326).Decision curve analysis displayed that our model showed good potential clinical benefits.Conclusion Age,history of gastrointestinal ulcer/bleed-ing,alcohol,heart failure and Cr are independent risk factors for UGIB in these post-PCI patients after dual antiplatelet therapy.The prediction model constructed with these factors has good dis-criminability,calibration and fitting,shows sound clinical application,and can be served as an ef-fective prediction tool for UGIB events in the patients.
