Dissociating external power from intramuscular exercise intensity during intermittent bilateral knee‐extension in humans

KEY POINTS: Continuous high‐intensity constant‐power exercise is unsustainable, with maximal oxygen uptake ([Formula: see text]) and the limit of tolerance attained after only a few minutes. Performing the same power intermittently reduces the O(2) cost of exercise and increases tolerance. The extent to which this dissociation is reflected in the intramuscular bioenergetics is unknown. We used pulmonary gas exchange and (31)P magnetic resonance spectroscopy to measure whole‐body [Formula: see text] , quadriceps phosphate metabolism and pH during continuous and intermittent exercise of different work:recovery durations. Shortening the work:recovery durations (16:32 s vs. 32:64 s vs. 64:128 s vs. continuous) at a work rate estimated to require 110% peak aerobic power reduced [Formula: see text] , muscle phosphocreatine breakdown and muscle acidification, eliminated the glycolytic‐associated contribution to ATP synthesis, and increased exercise tolerance. Exercise intensity (i.e. magnitude of intramuscular metabolic perturbations) can be dissociated from the external power using intermittent exercise with short work:recovery durations. ABSTRACT: Compared with work‐matched high‐intensity continuous exercise, intermittent exercise dissociates pulmonary oxygen uptake ([Formula: see text]) from the accumulated work. The extent to which this reflects differences in O(2) storage fluctuations and/or contributions from oxidative and substrate‐level bioenergetics is unknown. Using pulmonary gas‐exchange and intramuscular (31)P magnetic resonance spectroscopy, we tested the hypotheses that, at the same power: ATP synthesis rates are similar, whereas peak [Formula: see text] amplitude is lower in intermittent vs. continuous exercise. Thus, we expected that: intermittent exercise relies less upon anaerobic glycolysis for ATP provision than continuous exercise; shorter intervals would require relatively greater fluctuations in intramuscular bioenergetics than in [Formula: see text] compared to longer intervals. Six men performed bilateral knee‐extensor exercise (estimated to require 110% peak aerobic power) continuously and with three different intermittent work:recovery durations (16:32, 32:64 and 64:128 s). Target work duration (576 s) was achieved in all intermittent protocols; greater than continuous (252 ± 174 s; P < 0.05). Mean ATP turnover rate was not different between protocols (∼43 mm min(−1) on average). However, the intramuscular phosphocreatine (PCr) component of ATP generation was greatest (∼30 mm min(−1)), and oxidative (∼10 mm min(−1)) and anaerobic glycolytic (∼1 mm min(−1)) components were lowest for 16:32 and 32:64 s intermittent protocols, compared to 64:128 s (18 ± 6, 21 ± 10 and 10 ± 4 mm min(−1), respectively) and continuous protocols (8 ± 6, 20 ± 9 and 16 ± 14 mm min(−1), respectively). As intermittent work duration increased towards continuous exercise, ATP production relied proportionally more upon anaerobic glycolysis and oxidative phosphorylation, and less upon PCr breakdown. However, performing the same high‐intensity power intermittently vs. continuously reduced the amplitude of fluctuations in [Formula: see text] and intramuscular metabolism, dissociating exercise intensity from the power output and work done.