QRS micro-fragmentation as a mortality predictor in the emergency department: a dual-center retrospective cohort study

Background Emergency departments (EDs) are increasingly challenged by high patient volumes. QRS microfragmentation (QRSµf) is a novel ECG-based marker of abnormal ventricular depolarization and has been shown to be a strong predictor of all-cause mortality in both patients with heart disease and the general population. Purpose To evaluate whether QRSµf is associated with short-term mortality in patients presenting at the ED, thereby assessing its potential as a readily available risk stratification tool to facilitate patient care and clinical decision-making in emergency care. Methods In this large, dual-center retrospective cohort study, all patients presenting to the ED who underwent ECG recording as part of routine clinical care were included. QRSµf was calculated according to established technologies and dichotomized at the previously established cut-off value of 3.5%. The primary endpoint was death from any cause, with a follow-up period of 1 year. Multivariable analyses were adjusted for patient age, gender, heart rate, QRS duration, QTc and R-T angle. Analyses were performed in a derivation center 1 (inclusion period September ’19 until June ’22), and subsequently in a validation center 2 (inclusion period January ’18 until December ’22). Results A total of 10,915 patients were included in center 1 (median age of 69 years [54 – 80], 48% female), and 47,566 patients in center 2 (median age of 66 years [50 – 79], 49% female). QRSµf was above the cut-off point of 3.5% for 1,053 patients (10%) in center 1, and for 8,265 patients (17%) in center 2. During a median follow-up of 1 year, mortality occurred in 1,355 patients (12%) in center 1 and 5,908 patients (12%) in center 2. Median QRSµf was significantly higher in patients who died during follow-up compared to survivors (2.0% [1.3 – 3.0] vs 1.6%[1.2 – 2.3] in center 1 (p<0.0001), and 2.4% [1.8 – 3.5] vs 2.2% [1.6 – 3.0] in center 2 (p<0.0001)). In a multivariable Cox regression QRS>3.5% was significantly associated with a higher incidence of mortality with a hazard ratio (HR) of 1.43 (95%CI 1.21 – 1.69, p<0.0001) for center 1, and a HR of 1.24 (95%CI 1.16 – 1.32, p<0.0001) for center 2 (Table 1). Conclusion QRSµf is independently associated with 1-year mortality in patients presenting at the ED for any reason. These findings suggest QRSµf may hold potential as a risk stratification tool in emergency care.