Analysis of risk factors and development of prediction model for long term prognosis in patients with ischemic heart failure after percutaneous coronary intervention

Lifang Su¹Xianghua Fu¹Yunfa Jiang¹Yanbo Wang¹Boyan Tian¹Yang Fu¹Qing Wang¹Wei Zhi¹Yi Li¹Zhengkun Guan²Xinshun Gu^1*

• ¹Second Hospital of Hebei Medical University, Shijiazhuang, China
• ²Fourth Hospital of Hebei Medical University, Shijiazhuang, Hebei Province, China

Background: This study aimed to investigate the factors influencing the long-term prognosis of patients with ischemic heart failure (IHF) after percutaneous coronary intervention (PCI) , and to develop and validate a nomogram prediction model based on these key factors. Methods: In this single-center and retrospective study, consecutive patients diagnosed with IHF who underwent PCI at the main campus of the Second Hospital of Hebei Medical University between January 2019 and September 2023 were included. A nomogram prediction model was developed based on key factors identified by Cox regression and least absolute shrinkage and selection operator (LASSO) regression. Additionally, the patients treated at the branch campus of the Second Hospital of Hebei Medical University during the same period for external validation. Results: Age, New York Heart Association (NYHA) classification III or IV, residual diseased coronary arteries≥2, left ventricular ejection fraction (LVEF), left ventricular end-diastolic dimension (LVEDD), and the application of angiotensin receptor-neprilysin inhibitor (ARNI) during follow-up as factors significantly associated with MACE. The nomogram prediction model based on these six factors had an area under the curve (AUC) of 0.764 (95% CI: 0.680-0.847) for the 5-year receiver operating characteristic (ROC) analysis, and the model's concordance index (C-index) was 0.713, indicating good discriminative ability at the 5-year. Calibration curve and decision curve analysis demonstrated the model's consistency and clinical utility. The external validation of the model yielded an AUC of 0.707, and the C-index was 0.691. Multivariate Cox regression showed that NYHA classification III or IV, residual diseased coronary arteries ≥ 2, and LVEDD were independent risk factors for MACE, while the use of ARNI during follow-up was an independent protective factor. Conclusions: The nomogram prediction model, incorporating age, NYHA classification III or IV, residual diseased coronary arteries ≥2, LVEF, LVEDD, and the use of ARNI during follow-up, demonstrates strong predictive value for long-term MACE in patients with ICM after PCI. NYHA classification III or IV, residual diseased coronary arteries ≥2, and LVEDD were identified as independent risk factors for MACE, while the use of ARNI during follow-up was found to be a protective factor.