Assessing dynamic atrioventricular conduction time to RR-interval coupling in athletes and sedentary subjects

Abstract

Background: Atrioventricular conduction time (AVCT) is influenced by autonomic input and subject to physiological remodeling. Objective: To evaluate beat-by-beat AVCT and RR-interval variability in athletes and healthy sedentary subjects. Methods: Twenty adults, including 10 healthy sedentary (Controls) and 10 elite long-distance runners (Athletes), age, weight and height-adjusted, underwent maximal metabolic equivalent (MET) assessment, and 15-min supine resting ECG recording seven days later. The interval between P-wave and R-wave peaks defined the AVCT. Mean (M) and standard deviation (SD) of consecutive RR-intervals (RR) and coupled AVCT were calculated, as well as regression lines of RR vs. AVCT (RR-AVCT). Concordant AV conduction was defined as positive RR-AVCT slope and discordant otherwise. A multivariate linear regression model was developed to explain MET based on AVCT and RR-interval variability parameters. Significance-level: 5 %. Results: In Athletes, M-RR and SD-RR values were higher than in Controls, whereas M-AVCT and SD-AVCT were not. RR-AVCT slopes were, respectively, 0.038 ± 0.022 and 0.0034 ± 0.017 (p < 0.05). Using a cut-off value of 0.0044 (AUC 0.92 ± 0.07; p < 0.001), RR-AVCT slope showed 100% specificity and 80% sensitivity. In a multivariate model, SD-RR and RR-AVCT slope were independent explanatory variables of MET (F-ratio: 17.2; p < 0.001), showing 100% specificity and 90% sensitivity (AUC 0.99 ± 0.02; p < 0.001). Conclusion: In elite runners, AVCT to RR-interval dynamic coupling shows spontaneous discordant AV conduction, characterized by negative AVCT vs. RR-interval regression line slope. RR-intervals standard deviation and AVCT vs. RR-interval regression line slope are independent explanatory variables of MET.