Reliability and Sensitivity of Nocturnal Heart Rate and Heart-Rate Variability in Monitoring Individual Responses to Training Load

Abstract

Purpose: To assess the reliability of nocturnal heart rate (HR) and HR variability (HRV) and to analyze the sensitivity of these markers to maximal endurance exercise. Methods: Recreational runners recorded nocturnal HR and HRV on nights after 2 identical low-intensity training sessions (n = 15) and on nights before and after a 3000-m running test (n = 23). Average HR, the natural logarithm of the root mean square of successive differences (LnRMSSD), and the natural logarithm of the highfrequency power (LnHF) were analyzed from a full night (FULL), a 4-hour (4H) segment starting 30 minutes after going to sleep, and morning value (MOR) based on the endpoint of the linear fit through all 5-minute averages during the night. Differences between the nights were analyzed with a general linear model, and intraclass correlation coefficient (ICC) was used for internight reliability assessments. Results: All indices were similar between the nights followed by low-intensity training sessions. A very high ICC (P .001) was observed in all analysis segments with a range of .97 to .98 for HR, .92 to .97 for LnRMSSD, and .91 to .96 for LnHF. HR increased (P .001), whereas LnRMSSD (P .01) and LnHF (P .05) decreased after the 3000-m test compared with previous night only in 4H and FULL. Increments in HR (P .01) and decrements in LnRMSSD (P .05) were greater in 4H compared with FULL and MOR. Conclusions: Nocturnal HR and HRV indices are highly reliable. Demanding maximal exercise increases HR and decreases HRV most systematically in 4H and FULL segments.