Force-velocity test on a stationary cycle ergometer: methodological recommendations

Force-velocity tests performed on stationary cycle ergometers are widely used to assess the torque- and power-generating capacities of the lower limbs. The aim of this study was to identify how testing and modeling procedures influence the assessment of individual torque-cadence and power-cadence relationships. Seventeen males completed 62 +/- 16 pedal cycles from six 6-s all-out efforts interspersed with 5 min of rest. True measures of maximal power for a particular cadence were obtained for 24 +/- 3 pedal cycles, while power was only 94 +/- 3% of the true maximum in 19 +/- 5 pedal cycles. Pedal cycles showing maximal levels of power also displayed higher levels of electromyography (EMG: 89 +/- 7 vs . 87 +/- 7%) and coactivation (34 +/- 11 vs . 31 +/- 10 arbitrary units), as well as lower variability in crank torque and EMG profiles. Compared with the linear and second-order polynomial models that are traditionally used, a better goodness of fit was obtained when the torque-cadence and power-cadence relationships were predicted using second- and third-order polynomials, respectively. The later modeling procedures also revealed an asymmetry in the power-cadence relationship in most participants (i.e., 15 out of 17) and provided a better estimation of maximal cadence [C-max: 214 +/- 20 revolutions/min (rpm)] from the x-intercept of power-cadence relationships (C-0: 214 +/- 14 rpm). Therefore, we recommend predicting the individual shapes of torque- and power-cadence relationships using second- and third-order polynomial regressions after having selected pedal cycles during which true measures of cadence-specific maximal power were recorded. NEW & NOTEWORTHY This study is the first to demonstrate that suboptimal activation of the lower limb muscles accompanied reductions in cadence-specific levels of torque and power produced during a force-velocity test performed on a stationary cycle ergometer. This research is also the first to show that, in most noncyclist participants, torque-cadence relationships are not linear, whereas power-cadence relationships display asymmetric shapes, with power production decreasing rapidly when cadence increases beyond 180 revolutions/min.