Variability of cardiopulmonary exercise testing in patients with atrial fibrillation and determination of exercise responders to high-intensity interval training and moderate-to-vigorous intensity continuous training.

Abstract

Disabling atrial fibrillation (AF)-related symptoms and different testing settings may influence day-to-day cardiopulmonary exercise testing (CPET) measurements, which can affect exercise prescription for high-intensity interval training (HIIT) and moderate-to-vigorous intensity continuous training (M-VICT) and their outcomes. This study examined the reliability of CPET in patients with AF and assessed the proportion of participants achieving minimal detectable changes (MDC) in peak oxygen consumption (V̇O2peak) following HIIT and M-VICT. Participants were randomized into HIIT or M-VICT after completing two baseline CPETs: one with cardiac stress technologists (CPETdiag) and the other with a research team of exercise specialists (CPETresearch). Additional CPET was completed following 12 weeks of twice-weekly training. Reliability of CPETdiag and CPETresearch was assessed by intraclass correlation coefficient (ICC) and dependent t-tests. The MDC score was calculated for V̇O2peak using a reliable change index. The proportion of participants achieving MDC was compared between HIIT and M-VICT using chi-square analysis. Eighteen participants (69±7 years, 33% females) completed two baseline CPETs. ICC was significant for all measured variables. However, peak power output (POpeak: 124±40 vs. 148±40 watts, p<0.001) and HR (HRpeak: 136±22 vs. 148±30 bpm, p=0.023) were significantly greater in CPETresearch than CPETdiag. Few participants achieved MDC in V̇O2peak (5.6 mL/kg/min) with no difference between HIIT (0%) and M-VICT (10.0%, p=0.244). POpeak and HRpeak differed significantly in patients with AF when CPETs were repeated under different settings. Caution must be practiced when prescribing exercise intensity based on these measures as under-prescription may increase the number of exercise non-responders.