REF-989 · Rosenblat et al., 2023 · Sports Medicine, Vol. · Paper · PubMed

The Additional Effect of Training Above the Maximal Metabolic Steady State on VO2peak, Wpeak and Time-Trial Performance in Endurance-Trained Athletes: A Systematic Review, Meta-analysis, and Reality Check

A-fairly strong evidenceMeasured on trained to highly trained athletesWritten before October 2026, being re-checked

A systematic review and meta-analysis asking what training above the maximal metabolic steady state (MMSS) adds to VO2peak, peak power (Wpeak) and time-trial (TT) performance of endurance-trained athletes.

Eligible studies were prospective, used trained participants, and compared an intervention group doing training both above and below MMSS with a comparator doing only below-MMSS training.

Fourteen studies lasting 2 to 12 weeks met the criteria, covering 171 recreational and 128 competitive endurance athletes, with mean ages from 15 to 43 years and mean VO2peak 38-68 mL·kg-1·min-1. Adding above-MMSS work gave a further 2.5 mL·kg-1·min-1 of VO2peak gain (95% CI 1.4-3.6, p < 0.01, with I2 = 0%).

Detecting that effect (SMD 0.44) with sufficient power would need at least 81 participants per group. Wpeak and TT performance showed no intensity-specific effect, partly because the samples were smaller.

Participants

Fourteen studies; 171 recreational and 128 competitive endurance athletes. Mean age across studies 15 to 43 years; mean VO2peak across studies 38 to 68 mL·kg-1·min-1. Sex and sport are not given in the abstract.

Protocol

Sources: Medline and SPORTDiscus, from inception until February 23, 2023.
Inclusion: published in academic journals, in English, prospective, trained participants, an intervention group with training above and below MMSS, a comparator group training only below MMSS, and results for VO2peak, Wpeak or TT performance.
Duration of included studies: 2 to 12 weeks.
Comparison: between groups (above + below MMSS vs below MMSS only), pooled across studies.
"Reality check": a sample-size calculation for the pooled VO2peak effect.

Key Results

VO2peak: adding above-MMSS training gave a bigger gain, by 2.5 mL·kg-1·min-1 more (95% CI 1.4-3.6, p < 0.01, with I2 = 0%).
Adequate power to detect a significant VO2peak effect of this size (SMD 0.44) would need at least 81 participants per group.
For Wpeak and TT performance, no effect specific to intensity emerged, partly because fewer participants were available.
Authors' conclusion: in endurance-trained athletes, one mesocycle containing some above-MMSS training can raise VO2peak more than a mesocycle kept entirely below MMSS; the evidence is not enough to say that Wpeak or TT performance adapt alongside it.

Practical Application

"I am already trained — does adding work above threshold raise my VO2max, and will I be faster for it?" In a meta-analysis of 14 controlled studies of 2 to 12 weeks (171 recreational and 128 competitive endurance athletes; between-group comparison), adding training above the maximal metabolic steady state gave a 2.5 mL·kg-1·min-1 greater VO2peak gain (95% CI 1.4-3.6; I2 = 0%) than training only below it — but no intensity-specific effect on peak power or time-trial performance was found. The stimulus-to-VO2peak link is supported in trained athletes; the performance payoff inside one block is not demonstrated.
"Not demonstrated" is partly a power problem: the authors state the Wpeak and TT analyses had a smaller sample. It is an absence of evidence within 2 to 12 weeks, not proof of no effect.
The "reality check": detecting the VO2peak effect (SMD 0.44) needs 81 per group — larger than a typical training study. Inference, not tested here: a single athlete's response to one block is far noisier than this pooled estimate, so one flat block does not show the stimulus failed.
The comparator trained only below MMSS; the review does not compare formats of above-MMSS work, nor say how much of it is needed. Interval shape is covered by SYNTH-33.

Limitations

Fourteen studies, 2 to 12 weeks, small groups (171 recreational and 128 competitive athletes across all arms).
Wpeak and TT analyses rest on fewer participants than VO2peak; a null result there is inconclusive.
Mixed recreational and competitive athletes with a wide VO2peak range (38 to 68); no subgroup result is given in the abstract.
How MMSS was determined in each study, and the type and amount of above-MMSS training, are not given in the abstract.
English-language, two databases.
Written from the abstract and record metadata only (full text not read).
How this entry was checked and changed
Why this level

meta-analysis of 14 short studies (2 to 12 weeks) with small groups; the authors' own power calculation asks for 81 per group; the Wpeak and time-trial comparisons are underpowered