Physiological and perceptual responses to sprint interval exercise using arm versus leg cycling ergometry

Increases in power output and maximal oxygen consumption ( $$ max) occur in response to sprint interval exercise (SIE), but common use of "all-out" intensities presents a barrier for many adults. Furthermore, lower-body SIE is not feasible for all adults. We compared physiological and perceptual responses to supramaximal, but "non-all-out" SIE between leg and arm cycling exercise. Twenty-four active adults (mean ​± ​SD age: [25 ​± ​7] y; cycling $$ max: [39 ​± ​7] mL·kg^-1·min^-1) performed incremental exercise using leg (LCE) and arm cycle ergometry (ACE) to determine $$ max and maximal work capacity (Wmax). Subsequently, they performed four 20 ​s bouts of SIE at 130% Wmax on the LCE or ACE at cadence ​= ​120-130 ​rev/min, with 2 ​min recovery between intervals. Gas exchange data, heart rate (HR), blood lactate concentration (BLa), rating of perceived exertion (RPE), and affective valence were acquired. Data showed significantly lower (p ​< ​0.001) absolute mean ([1.24 ​± ​0.31] L·min^-1 vs. [1.59 ​± ​0.34] L·min^-1; d ​= ​1.08) and peak $$ ([1.79 ​± ​0.48] L·min^-1 vs. [2.10 ​± ​0.44] L·min^-1; d ​= ​0.70) with ACE versus LCE. However, ACE elicited significantly higher (p ​< ​0.001) relative mean ([62% ​± ​9%] $$ max vs. [57% ​± ​7%] $$ max, d ​= ​0.63) and peak $$ ([88% ​± ​10%] $$ max vs. [75% ​± ​10%] $$ max, d ​= ​1.33). Post-exercise BLa was significantly higher ([7.0 ​± ​1.7] mM vs. [5.7 ​± ​1.5] mM, p ​= ​0.024, d ​= ​0.83) for LCE versus ACE. There was no significant effect of modality on RPE or affective valence (p ​> ​0.42), and lowest affective valence recorded (2.0 ​± ​1.8) was considered "good to fairly good". Data show that non "all-out" ACE elicits lower absolute but higher relative HR and $$ compared to LCE. Less aversive perceptual responses could make this non-all-out modality feasible for inactive adults.