Prediction of new onset atrial fibrillation in acute myocardial infarction using fragmented QRS complex combined with HEART score

Authors:
Open Access
Publisher website
Google Scholar
Abstract
This study explores the predictive value of fragmented QRS (FQRS) wave combined with the HEART score for new-onset atrial fibrillation (NOAF) following emergency percutaneous coronary intervention (PCI) in patients with acute myocardial infarction (AMI). A regression analysis was conducted on clinical data from AMI patients who underwent emergency PCI at the Third Affiliated Hospital of Anhui Medical University between March 2019 and October 2023. A total of 509 AMI patients without a history of atrial fibrillation were included. The presence of FQRS was determined by electrocardiogram, dividing patients into an FQRS group (303 cases) and a non-FQRS group (206 cases). HEART scores were calculated at admission, categorizing patients into low-risk (30 cases), medium-risk (220 cases), and high-risk (259 cases) groups. NOAF occurrence was observed during hospitalization and within 6 months post-discharge. The study compared the incidence of various parameters and NOAF across different groups and analyzed the ROC curves for FQRS, HEART score, and their combined predictive value for NOAF. During the 6-month follow-up, 50 out of 509 AMI patients developed NOAF. The FQRS group showed higher rates of NOAF, elevated VLDL levels, more STEMI cases, and a higher history of alcohol consumption compared to the non-FQRS group. The LVEF value was lower in the FQRS group, with these differences being statistically significant (PPPP=0.014; OR=1.618, 95% CI: 1.148–2.281, P=0.06). The area under the ROC curve for FQRS in predicting NOAF was 0.599 (95% CI: 0.522–0.677), for the HEART score 0.657 (95% CI: 0.584–0.730), and for the combined prediction 0.691 (95% CI: 0.617–0.764). The combined prediction model had a larger ROC curve area than either FQRS or HEART score alone. The FQRS wave and HEART score can predict NOAF in AMI patients following emergency PCI. Combining both improves predictive accuracy.

This publication has 37 references indexed in Scilit: