A Machine Learning Model for the Prediction of No-Reflow Phenomenon in Acute Myocardial Infarction Using the CALLY Index

Background: Acute myocardial infarction (AMI) constitutes a major health problem with high mortality rates worldwide. In patients with ST-segment elevation myocardial infarction (STEMI), no-reflow phenomenon is a condition that adversely affects response to therapy. Previous studies have demonstrated that the CALLY index, calculated using C-reactive protein (CRP), albumin, and lymphocytes, is a reliable indicator of mortality in patients with non-cardiac diseases. The objective of this study is to investigate the potential utility of the CALLY index in detecting no-reflow patients and to determine the predictability of this phenomenon using machine learning (ML) methods.

Methods: This study included 1785 STEMI patients admitted to the clinic between January 2020 and June 2024 who underwent percutaneous coronary intervention (PCI). Patients were in no-reflow status, and other clinical data were analyzed. The CALLY index was calculated using data on patients' inflammatory status. The Extreme Gradient Boosting (XGBoost) ML algorithm was used for no-reflow prediction.

Results: No-reflow was detected in a proportion of patients participating in this study. The model obtained with the XGBoost algorithm showed high accuracy rates in predicting no-reflow status. The role of the CALLY index in predicting no-reflow status was clearly demonstrated.

Conclusions: The CALLY index has emerged as a valuable tool for predicting no-reflow status in STEMI patients. This study demonstrates how machine learning methods can be effective in clinical applications and paves the way for innovative approaches for the management of no-reflow phenomenon. Future research needs to confirm and extend these findings with larger sample sizes.