REF-1482 · Gomes et al., 2022 · Journal of Strength and Conditioning Research · Paper · PubMed

Can session RPE steer my lifting?

Perhaps: in 20 trained men over 6 weeks it gave fewer hard sessions than a preset plan; strength 9.68% vs 9.28%, muscle 6.55% vs 9.65%. Too small to show equivalence; rule not in the abstract.

Main caveat: Abstract only, and the rule is missing. The method itself can't be reproduced from it.
limited, mixed evidenceMeasured on recreational to trained athletes

Gomes and colleagues compared an individualised progression based on the session rating of perceived exertion with a conventional predetermined progression. Twenty previously trained young men (n = 20) were randomised to RPE (n = 10) or PP (n = 10). Muscle cross-sectional area and maximum dynamic strength were measured at baseline and after 6 weeks.

The RPE model led to fewer high-intensity sessions, and total volume was not significantly different (45,366.00 ± 10,190.00 vs 47,779.00 ± 5,685.00 kg; p = 0.52). Both groups improved (p < 0.0001): area by 6.55 ± 5.27% vs 9.65 ± 3.63% and strength by 9.68 ± 4.57% vs 9.28 ± 4.01%. The authors encourage trainees to adopt session RPE as a potential tool.

Our reading: a small pilot that shows the method did not obviously fail in six weeks. "Similar gains" in n = 10 per group is not proof of equivalence, the muscle point estimate is about 3 percentage points lower with RPE, and the entry can't say what was done since the rule is missing from the abstract.

Protocol

Design: parallel, randomised, two groups; 6 weeks.
Who: n = 20 previously trained young male subjects: RPE n = 10, PP n = 10.
The rule: the abstract says only "an individualized resistance training progression model based on the session rating of perceived exertion" and "session-to-session physiological response assessments". It does not give the RPE scale, when the rating was taken, the rating that triggers more or less work, what changed (load, reps, sets), or the sets and reps. The predetermined plan is not described either.
Session RPE, how used: as a whole-session rating that steered the next session's workload, per the abstract's wording; details not given.
Outcomes: muscle cross-sectional area (site and method not in the abstract), maximum dynamic strength (exercise not in the abstract), total training volume, number of high-intensity sessions.
Funding and interests: not stated in the abstract record.

Key Results

OutcomeRPE (n = 10)Preset progression (n = 10)Between groups
Muscle area change6.55 ± 5.27%9.65 ± 3.63%"similar"; within-group p < 0.0001
Strength change9.68 ± 4.57%9.28 ± 4.01%"similar"; within-group p < 0.0001
Total training volume45,366.00 ± 10,190.00 kg47,779.00 ± 5,685.00 kgp = 0.52
High-intensity sessionslowerhigherthe number is not in the abstract
Point differences (our arithmetic): muscle area 3.10 percentage points lower with RPE; strength 0.40 higher with RPE; no confidence interval in the abstract.

Mechanisms

Measured: size, strength, volume, count of hard sessions.
Authors' interpretation: progression based on session-to-session physiological response allows fewer high-intensity sessions and "optimization of training stimulus on an individual basis".
Inference: session RPE is a perception, not a physiological response. The abstract's "optimization" isn't shown: no individual responses or optimum were measured in the abstract.

Practical Application

Can session RPE steer my lifting? Perhaps: in 20 trained men over 6 weeks it gave fewer hard sessions than a preset plan; strength 9.68% vs 9.28%, muscle 6.55% vs 9.65%. Too small to show equivalence; rule not in the abstract.
Use it as a hint, not a rule: this paper can support "a rating can inform the next session" as worth trying; it cannot supply thresholds. Any number for a trigger is not from this paper.
Not from this paper: the app logs RPE per exercise, so how to read a jump in logged RPE is for the new progression guide to set, as practice.
Do not read this as autoregulation being proven: REF-1476 says it cannot determine the effect of intrasession autoregulation, and REF-685 mainly favours APRE, a different method.

Limitations

Abstract only, and the rule is missing. The method itself can't be reproduced from it.
Small. Ten men a group; a difference on muscle area of about 3 percentage points, with SDs of 5.27 and 3.63, isn't excluded. The abstract prints no between-group test for the gains, only "similar", and that does not establish equivalence.
The conclusion runs ahead of the data. Encouraging adoption for "optimization" isn't something a null in 20 men can back.
Short and narrow. 6 weeks, young trained men; no women, and no endurance training described.
Measure. How hard "high-intensity" was, and how it was defined, isn't in the abstract.
How this entry was checked and changed
Why this level

n = 20 sits at the floor of the rubric's parallel-RCT 20-39 band, but 10 per group is treated as a small two-arm trial (B design), as the base does for other abstract-only trials; down one for the abstract-only small-study clause, since the progression rule at the centre of the paper is not in the abstract and the conclusion claims more than the design can show