Development and Validation of a Long-Term Mortality Prediction Model for Acute Coronary Syndrome Patients Based on Cardiopulmonary Exercise Testing

Background. Acute coronary syndrome (ACS) is a major global health burden with a high risk of adverse outcomes. Existing predictive models (e.g., GRACE) primarily rely on static indicators and focus on short-term prognosis, limiting their ability to comprehensively assess patient status and predict long-term mortality. To address the need for improved long-term risk prediction, this study developed and validated a long-term mortality prediction model for ACS patients based on cardiopulmonary exercise testing (CPET) and other clinical indicators. Methods . This study included ACS patients who were treated at Tongji Hospital in Shanghai from January 1, 2007, to December 31, 2018, according to the inclusion criteria. Demographic data, medical histories, CPET indicators, laboratory indicators, and other baseline data of all included patients were collected, and their mortality was followed up. All data sets were randomly divided into derivation and validation cohorts in a ratio of 7/3. Least absolute shrinkage and selection operator regression and Cox multivariate analysis were used to identify independent risk factors affecting ACS prognosis, and a risk prediction model was established using nomograms. Results . A total of 299 patients were included in this cohort (211 in the derivation cohort and 88 in the validation cohort), with an average age of 57.00 years, including 280 males (93.6%). The median follow-up time was 3821 days, and 46 cases (15.4%) reached the study endpoint. The derivation cohort identified four independent predictive factors: age, blood urea nitrogen (BUN), ejection fraction (EF), and heart rate reserve (HRR), and a Nomogram scoring model was constructed based on these factors. The C indexes values of the derivation and the validation cohorts were 0.83 (0.76, 0.89) and 0.72 (0.56, 0.88), respectively. Calibration curves indicated good consistency between model predictions and actual observations. Conclusions . A model established based on four CPET indicators—age, BUN, EF, and HRR—can effectively predict the long-term all-cause mortality risk of ACS, providing a new tool for the long-term management of ACS. Trial registration. Registry: Chinese Clinical Trial Registry; TRN: ChiCTR2100052199; Registration date: October 22, 2021.