Purpose The physiological consequences of freely chosen cadence during cycling remains poorly understood. We sought to determine the effect of cadence on the respiratory and hemodynamic response to cycling exercise. Methods Eleven cyclists (10 males, 1 female; age, 27 ± 6 yr; VO2max = 60.8 ± 3.7 mL˙kg-1˙min-1) completed four, 6-min constant-load cycling trials at 10% below their previously determined gas exchange threshold (i.e., 63% ± 5% peak power) while pedaling at 60, 90, and 120 rpm, and a freely chosen cadence (94.3 ± 6.9 rpm) in randomized order. Standard cardiorespiratory parameters were measured and an esophageal electrode balloon catheter was used to assess electromyography of the diaphragm (EMGdi) and the work of breathing (Wb). Leg blood flow index (BFI) was determined on four muscles using near-infrared spectroscopy with indocyanine green dye injections. Results Oxygen uptake (VO2) increased as a function of increasing cadence (all pairwise comparisons, P < 0.05). The EMGdi and Wb were significantly greater at 120 rpm compared with all other conditions (all P < 0.01). Vastus medialis and semitendinosus BFI were significantly greater at 120 rpm compared with 60 and 90 rpm (all P < 0.05). Gastrocnemius BFI was higher at 120 rpm compared with all other cadences (all P < 0.01). No difference in BFI was found in the vastus lateralis (P = 0.06). Blood flow index was significantly correlated with the increase in VO2 with increasing cadence in the medial gastrocnemius (P < 0.001) and approached significance in the vastus lateralis (P = 0.09), vastus medialis (P = 0.06), and semitendinosus (P = 0.09). There was no effect of cadence on Borg 0-10 breathing or leg discomfort ratings (P > 0.05). Conclusions High cadence cycling at submaximal exercise intensities is metabolically inefficient and increases EMGdi, Wb, and leg muscle blood flow relative to slower cadences.