VALIDATION OF HEART RATE MONITOR-BASED PREDICTIONS OF OXYGEN UPTAKE AND ENERGY EXPENDITURE

Montgomery, PG, Green, DJ, Etxebarria, N, Pyne, DB, Saunders, PU, and Minahan, CL. Validation of heart rate monitor-based predictions of oxygen uptake and energy expenditure. J Strength Cond Res 23(5): 1489-1495, 2009-To validate (V) over dotO(2) and energy expenditure predictions by the Suunto heart rate (HR) system against a first principle gas analysis system, well-trained male (n = 10, age 29.8 +/- 4.3 years, (V) over dotO(2) 65.9 +/- 9.7 ml.kg(-1).min(-1)) and female (n = 7, 25.6 +/- 3.6 years, 57.0 +/- 4.2 ml.kg(-1).min(-1)) runners completed a 2-stage incremental running test to establish submaximal and maximal oxygen uptake values. Metabolic cart values were used as the criterion measure of (V) over dotO(2) and energy expenditure (kJ) and compared with the predicted values from the Suunto software. The 3 levels of software analysis for the Suunto system were basic personal information (BI), BI + measured maximal HR (BIhr), and BIhr + measured (V) over dotO(2) (BIhr + v). Comparisons were analyzed using linear regression to determine the standard error of the estimate (SEE). Eight subjects repeated the trial within 7 days to determine reliability (typical error [TE]). The SEEs for oxygen consumption via BI, BIhr, and BIhr + v were 2.6, 2.8, and 2.6 ml.kg(-1).min(-1), respectively, with corresponding percent coefficient of variation (%CV) of 6.0, 6.5, and 6.0. The bias compared with the criterion (V) over dotO(2) decreased from -6.3 for BI, -2.5 for BIhr, to -0.9% for BIhr + v. The SEE of energy expenditure improved from BI (6.74 kJ) to BIhr (6.56) and BIhr + v (6.14) with corresponding %CV of 13.6, 12.2, and 12.7. The TE values for (V) over dotO(2) were similar to 0.60 ml.kg(-1).min(-1) and similar to 2 kJ for energy expenditure. The %CV for (V) over doto(2) and energy expenditure was similar to 1 to 4%. Although reliable, basic HR-based estimations of (V) over dotO(2) and energy expenditure from the Suunto system underestimated (V) over dotO(2) and energy expenditure by similar to 6 and 13%, respectively. However, estimation can be improved when maximal HR and (V) over dotO(2) values are added to the software analysis.