Validity of the Actiheart step test for the estimation of maximum oxygen consumption in endurance athletes and healthy controls

Abstract

Submaximal step tests are often used for estimation of maximum oxygen consumption (VO_2max) in humans. The validity of the Actiheart step test for VO_2max estimation was not fully studied yet. Therefore, purpose of the study was to estimate VO_2max using the Actiheart step test and to compare the data with measured VO_2max in endurance trained athletes (ATH) and healthy non-athletes (CON).

68 ATH (54% men, 28.0±5.4 yrs, 20.9±1.7 kg∙m^-2) and 63 CON (46% men, 27.6±5.1 yrs, 22.1±1.7 kg∙m^-2) performed the Actiheart step test and a spiroergometry for assessment of VO_2max. In addition, resting metabolic rate (RMR; indirect calorimetry), maximum heart rate (HR_max; heart rate monitoring system during spiroergometry), and sleeping heart rate (SHR; Actiheart 6-day long term measurement) were determined. Validity of two different Actiheart software entry modes was assessed: (1) AHraw (estimated RMR [Schofield] and HR_max [Tanaka], SHR = 70 bpm) and (2) AHcomplete (measured RMR, HR_max, and SHR). Validity was investigated using linear regression (R² and standard error of the estimate (SEE)) and repeated-measures ANOVA with a Bonferroni post-hoc correction. The level of significance was set to α=0.05.

VO_2max estimated by AHraw was significant related to measured VO_2max in women CON (R²=0.22; p<0.05), whereas when VO_2max was estimated by AHcomplete the relation was significant in women ATH and CON, and in men CON (R²=0.17-0.24; p<0.05). AHraw significantly underestimated VO_2max in the total sample by 8% (51.4±10.2 vs. 55.9±7.6 ml∙kg^-1∙min^-1; p<0.0001), whereas no significant difference between AHcomplete and the criterion method was found (57.0±11.1 vs. 55.9±7.6 ml∙kg^-1∙min^-1; p=0.26).

The Actiheart step test is an acceptable tool for the estimation of VO_2max if an error within 8% can be tolerated. However, accuracy of the VO_2max prediction is much improved when entering measured variables, such as RMR, SHR, and HR_max, into the software.