A Comparison of Physiological Response to Incremental Testing on Stationary and Dynamic Rowing Ergometers.

Abstract

Purpose: The purpose of the current study was to compare responses to graded exercise testing (GXT) on 2 popular commercial rowing ergometers. Methods: A cohort of 23 subelite male rowers (age 20 [2] y, height 1.88 [0.06] m, body mass 82.0 [8.8] kg) performed a GXT on both stationary (Concept2 [C2]) and dynamic (RowPerfect3 [RP3]) rowing ergometers. Physiological responses including oxygen consumption (VO2), heart rate (HR), blood lactate concentration (BLa), stroke rate (SR), and minute ventilation (VE) were recorded. BLa data were plotted graphically and anaerobic threshold was identified using the Dmax method. Workload, HR, and VO2 at Dmax were interpolated. Physiological responses at maximal exercise and at Dmax were compared, along with response across a discrete range of submaximal workloads. Results: At maximal exercise, no significant differences in HR, VO2, or BLa were observed (P > .05); however, VEpeak was significantly higher during RP3 tests (T = 2.943, P < .05). No significant differences in HR, VO2, or BLa at Dmax were observed (P > .05). When comparing across submaximal workloads, HR was significantly higher with the RP3 at 2 distinct workloads (210 and 240 W; P < .05), while SR was higher during RP3 testing at all workloads (F = 56.7, P < .05). When SR was fixed as a covariate, the effect of ergometer on HR response was not significant. A significant workload by ergometer interaction effect was observed for SR with higher data recorded on the RP3 (F = 3.48, P < .01). Levels of agreement for GXT-derived measures of anaerobic threshold (Dmax) were deemed unacceptable. Conclusions: These results indicate that while some differences in HR and VE response were observed between ergometers, these differences were a result of SR alterations between ergometer type. While no differences in response at Dmax were observed, the poor levels of agreement between ergometers suggests that prescription of GXT-derived threshold for training should ideally be specific to the rowing ergometer upon which the test was performed.