Towards a dynamic model to estimate evolving risk of major bleeding after percutaneous coronary intervention

Nathan C Hurley, Nihar Desai, Sanket S Dhruva, Rohan Khera, Wade Schulz, Chenxi Huang, Jeptha Curtis, Frederick Masoudi, John Rumsfeld, Sahand Negahban, Harlan M Krumholz and Bobak J Mortazavi

PLOS Digital Health, 2025, vol. 4, issue 6, 1-15

Abstract: While static risk models may identify key driving risk factors, the dynamic nature of risk requires up-to-date risk information to guide treatment decision making. Bleeding is a complication of percutaneous coronary intervention (PCI), and existing risk models produce only a single risk estimate anchored at a single point in time, despite the dynamic nature of this risk. Using data available from the National Cardiovascular Data Registry (NCDR) CathPCI, we trained 6 different tree-based machine learning models to estimate the risk of bleeding at key decision points: 1) choice of access site, 2) prescription of medication before PCI, and 3) choice of closure device. We began with 3,423,170 PCIs performed between July 2009 through April 2015. We included only index PCIs and removed anyone who had missing data regarding bleeding events or underwent coronary artery bypass grafting during the index admission. We included 2,868,808 PCIs; 2,314,446 (80.7%) before 2014 for training and 554,362 (19.3%) remaining for validation. This study considered all data available from the Registry prior to patient discharge: patient characteristics, coronary anatomy and lesion characterization, laboratory data, past medical history, anti-coagulation, stent type, and closure method categories. The primary outcome was any in-hospital bleeding event within 72 hours after the start of the PCI procedure. Discrimination improved from an area under the receiver operating characteristic curve (AUROC) of 0.812 using only presentation variables to 0.845 using all variables. Among 123,712 patients classified as low risk by the initial model, 14,441 were reclassified as moderate risk (1.4% experienced bleeds), while 723 were reclassified as high risk (12.5% experienced bleeds). Static risk prediction models have more predictive error than those that update risk prediction with newly available data, which provides up-to-date risk prediction for individualized care throughout a hospitalization.Author summary: Clinical risk models used for treatment decision making are often static models used with fixed input at a fixed point of time. Risk of adverse events, however, is dynamic, changing throughout admissions because of treatment decision making. This work looks at the risk of major bleeding for patients undergoing percutaneous coronary intervention, showing the changes in patient risk estimation throughout the course of treatment. By identifying the changes in risk of bleeding at different points in time, we demonstrate the need for more dynamic evaluation of risk estimates, providing potential changes in treatment decision making throughout admissions, accounting for prior treatment decisions made. The models demonstrate an improvement in discrimination in predicting risk of major bleeding and demonstrates a reclassification of a subset of patients, particularly demonstrating the need for re-evaluating bleeding risk (and thus treatment with bleeding avoidance therapies) at various stages of patient admission before discharge. Models that update risk prediction with newly available data, which provides up-to-date risk prediction, enable individualized care throughout a hospitalization.

Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
https://journals.plos.org/digitalhealth/article?id=10.1371/journal.pdig.0000906 (text/html)
https://journals.plos.org/digitalhealth/article/fi ... 00906&type=printable (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:plo:pdig00:0000906

DOI: 10.1371/journal.pdig.0000906

Access Statistics for this article

More articles in PLOS Digital Health from Public Library of Science
Bibliographic data for series maintained by digitalhealth ().