Are incremental exercise relationships between rating of perceived exertion and oxygen uptake or heart rate reserve valid during steady-state exercises?

Background Rating of perceived exertion (RPE) is considered a valid method for prescribing prolonged aerobic steady-state exercise (SSE) intensity due to its association with physiological indicators of exercise intensity, such as oxygen uptake (V̇O 2 ) or heart rate (HR). However, these associations between psychological and physiological indicators of exercise intensity were found during graded exercise tests (GXT) but are currently used to prescribe SSE intensity even though the transferability and validity of the relationships found during GXT to SSE were not investigated. The present study aims to verify whether (a) RPE-HR or RPE-V̇O 2 relations found during GXTs are valid during SSEs, and (b) the duration and intensity of SSE affect these relations. Methods Eight healthy and physically active males (age 22.6 ± 1.2 years) were enrolled. On the first visit, pre-exercise (during 20 min standing) and maximal (during a GXT) HR and V̇O 2 values were measured. Then, on separate days, participants performed 4 SSEs on the treadmill by running at 60% and 80% of the HR reserve (HRR) for 15 and 45 min (random order). Individual linear regressions between GXTs’ RPE (dependent variable) and HRR and V̇O 2 reserve (V̇O 2 R) values (computed as the difference between maximal and pre-exercise values) were used to predict the RPE associated with %HRR (RPE HRR ) and %V̇O 2 R (RPE V̇O2R ) during the SSEs. For each relation (RPE-%HRR and RPE-%V̇O 2 R), a three-way factorial repeated measures ANOVA ( α = 0.05) was used to assess if RPE (dependent variable) was affected by exercise modality ( i.e. , RPE recorded during SSE RPE SSE or GXT-predicted), duration ( i.e. , 15 or 45 min), and intensity ( i.e. , 60% or 80% of HRR). Results The differences between RPE SSE and GXT-predicted RPE, which were assessed by evaluating the effect of modality and its interactions with SSE intensity and duration , showed no significant differences between RPE SSE and RPE HRR . However, when RPE SSE was compared with RPE V̇O2R , although modality or its interactions with intensity were not significant, there was a significant ( p = 0.020) interaction effect of modality and duration yielding a dissociation between changes of RPE SSE and RPE V̇O2R over time. Indeed, RPE SSE did not change significantly ( p = 0.054) from SSE of 15 min (12.1 ± 2.0) to SSE of 45 min (13.5 ± 2.1), with a mean change of 1.4 ± 1.8, whereas RPE V̇O2R decreased significantly ( p = 0.022) from SSE of 15 min (13.7 ± 3.2) to SSE of 45 min (12.4 ± 2.8), with a mean change of −1.3 ± 1.5. Conclusion The transferability of the individual relationships between RPE and physiological parameters found during GXT to SSE should not be assumed as shown by the results of this study. Therefore, future studies modelling how the exercise prescription method used (e.g., RPE, HR, or V̇O 2 ) and SSE characteristics ( e.g. , exercise intensity, duration, or modality) affect the relationships between RPE and physiological parameters are warranted.