Clinical Prediction Model for Target Lesion Revascularization Within One Year After Drug-Eluting Stent Implantation: Nationwide Analysis Within the Netherlands Heart Registration.

PURPOSE: This study aimed to develop and validate a clinical prediction model for target lesion revascularization (TLR) at one-year follow-up in patients undergoing percutaneous coronary intervention (PCI) with implantation of a drug-eluting stent (DES).

PATIENTS AND

METHODS

Using real-world data from the Netherlands Heart Registration, we included patients treated with at least one DES between 2019 and October 2022. The study cohort comprised 96,758 patients from 29 centers, while a sub-cohort of 11,789 patients from 11 centers had additional procedural data. Multivariable logistic regression with backward stepwise selection was used for model development. Final coefficients were pooled across 20 imputed datasets, and separate models were built for both cohorts.

RESULTS

The study cohort model included age, sex, interaction between age and sex, diabetes mellitus, renal insufficiency, multivessel disease, out-of-hospital cardiac arrest (OHCA), cardiogenic shock, previous myocardial infarction, previous coronary intervention, access site, number of treated vessels, PCI in the left main, and PCI in arterial or venous grafts. The optimism-corrected area under the curve (AUC) was 0.645 (95% confidence interval (CI): 0.633-0.657). The sub-cohort model incorporated total stent length in the left anterior descending artery, arterial or venous grafts, and left main, along with renal insufficiency, previous coronary intervention, and OHCA, yielding an AUC of 0.662 (95% CI: 0.631-0.693).

CONCLUSION

The developed models provide a basis for predicting one-year TLR using routinely collected data. Despite strong calibration, their modest discrimination suggests a need for more detailed variables and advanced modelling approaches to improve performance.