Summer-winter differences in 24 h variability of heart rate

Objective To examine possible seasonal changes in heart rate variability (H RV). Background Greater than normal mortality from cardiovascular disease (CVD) in the winter has been reported for many countries and might be partly expl ained by considering seasonal changes in CVD risk factors. Depression of HR V is an independent predictor of arrhythmic complications and of cardiac de ath, and it is also among the variables that may be affected by the season of the year. Methods We compared pairs of 24 h HRV data of 120 healthy men who were exam ined once in the summer and once in the winter. Multivariate analyses were performed for each dependent variable (HRV indexes) in separate statistical models with age, resting heart rate, serum level of cholesterol, cigarette smoking, body mass index, sports habits, alcohol consumption, systolic blo od pressure, physical activity at work, years of education, consumption of energy, and season as the independent variables. Results Although there were no seasonal differences in mean R-R interval, a ll indexes of HRV were found to be lower in the summer than they were durin g winter. Differences and 95% confidence intervals were standard deviation (SD) of coupling intervals between normal beats 12 ms, 6-17 ms; SD of 5 min mean R-R intervals 14 ms, 8-20 ms; mean of all 5 min SD of R-R intervals 2 .0 ms, 0.6-2.5 ms; proportion of adjacent R-R intervals differing by > 50 m s 1.5%, 0.6-2.5% and root mean square of the difference between successive normal intervals 3.1 ms, 1.5-4 ms. Multivariate analyses showed that HRV in the winter was less than that in the summer even after adjustment for age, serum level of cholesterol, systolic blood pressure, and body mass index. Conclusions HRV indexes of healthy men vary physiologically by season, with lowest values obtained in the winter. Since low HRV is linked to pathologi c conditions, the significance of seasonal changes for those suffering from CVD and their possible contribution to the greater mortality rates in wint er have to be considered. J Cardiovasc Risk 7:141-146 (C) 2000 Lippincott W illiams & Wilkins.