Abstract
PURPOSE: Heart rate (HR)-derived variables (HR reserve [HR peak - HR rest ], chronotropic indices [attenuated HR response to exercise], HR recovery [attenuated HR response to exercise recovery], and peak HR index [HR peak /HR rest ]) together with peak oxygen uptake (VO 2peak ) are potential prognostic variables for cardiovascular and all-cause mortality. However, heart rate index (HRI) has not been established as a surrogate for VO 2peak , whether measured (Meas) or estimated (Est), during cycle ergometry (CE) and/or treadmill testing (TT). METHODS: HR-derived prognostic variables to assess cardiovascular outcomes were identified from 150 studies obtained from MEDLINE and Google Scholar searches. The Meas CE/TT-VO 2peak was reported in 81 studies (21 773 participants) and Est CE/TT-VO 2peak in 69 studies (331 435 participants). Using the HRI equation with metabolic equivalent (METs) = 6 × HRI - 5 (where HRI = peak HR/resting HR), HRI-VO 2peak was calculated from HR data reported in the 150 studies. The HRI-VO 2peak was then compared against group mean data for both Meas CE/TT-VO 2peak and Est CE/TT-VO 2peak . RESULTS: The difference between Meas CE/TT-VO 2peak and HRI-VO 2peak was 1% (7.15 ± 3.25 METs vs 7.08 ± 3.02 METs [ P = .833], respectively). By comparison, the difference between Est CE/TT-VO 2peak and HRI-VO 2peak was 25.6% (8.94 ± 2.36 METs vs 7.12 ± 2.27 METs [ P < .001], respectively). Moreover, HRI equation estimation of VO 2peak showed greater overprediction for TT, 26.6%, than for CE, 11.9%. CONCLUSIONS: The Meas-VO 2peak and HRI-VO 2peak agreed closely. When compared with HRI-VO 2peak , Est-VO 2peak from currently used exercise-based equations shows significant overprediction. Use of HRI and/or Fitness Registry and the Importance of Exercise National Database (FRIEND) registry equations warrant consideration for more accurately estimating VO 2peak .