Mechanical efficiency during sub-maximal cycling is underestimated because negative muscular power is ignored

The in vivo mechanical efficiency of muscles has often been estimated during sub-maximal cycling. In this approach, it has implicitly been assumed that the average amount of positive mechanical muscle power equals the average mechanical power output, i.e., that no power is dissipated by muscles. Here, we investigated the validity of this assumption using an optimal control musculoskeletal model. We identified optimal muscle stimulation patterns for 4 cadences (60, 80, 100 and 120RPM) and 5 levels of average mechanical power output (50, 100, 150, 200 and 250W). We found that the amount of negative mechanical muscular power was substantial, with the average across all conditions being -84,6W (56,4%). The amount of negative mechanical muscular power was found to increase with increasing cadence and was independent of the average mechanical power output. To investigate the effect of negative muscular power on in vivo estimates of the muscular efficiency, we used our simulation results to correct gross efficiencies measured during sub-maximal cycling. The resulting increases in the gross efficiency were substantial, with the average increasing from 16.9% to 27.5%. These results suggest that current estimates of the muscular efficiency during sub-maximal cycling underestimate the true muscular efficiency.