The prognostic value of ECG parameters for 20-year cardiovascular risk and incremental improvement of SCORE2-based prediction modelling: the ATTICA cohort study (2002–2022)

Abstract

We aimed to evaluate the prognostic value of electrocardiographic (ECG) parameters in estimating 20-year cardiovascular disease (CVD) risk and their incremental predictive value when added to the HellenicSCORE II+ model in an apparently healthy adult population. To evaluate whether routine ECG parameters improve long-term cardiovascular risk prediction beyond established SCORE2-based clinical models. This analysis included 1,482 participants (mean age 46±15 years; 49% male) from the population-based prospective ATTICA study who had baseline ECG recordings, comprehensive clinical and biochemical assessment, and complete 20-year follow-up data. During the 20-year follow-up period, 633 of the 1,482 participants (43%) experienced a first fatal or non-fatal CVD event. Out of the 633 CVD cases, 86 were fatal, leading to a 20-year CVD mortality rate of 9.5% in males and 2.4% in females (p<0.001). Almost all the ECG parameters (except for R-R and PR intervals) showed significant associations with 20-year CVD risk. Specifically, a 10 ms prolongation in the P wave duration was associated with a 29% increase in CVD risk (HR 1.029, 95% CI 1.018-1.041), while the same prolongation in the QRS duration was linked to a 32% increase CVD risk (HR 1.032, 95% CI 1.026-1.039). Also, incorporating the QRS duration into the HellenicSCORE II+ model yielded the highest positive NRI, indicating enhanced risk classification (continuous NRI: 0.677, p<0.001). Routine ECG parameters, particularly QRS duration, enhance long-term CVD risk prediction beyond established clinical models. Given the wide availability and low cost of ECG, these findings support further validation and integration of ECG-derived markers into preventive risk assessment strategies.Net reduction against threshold probabilityDecision curves for HellenicSCORE II+