REF-1721 · Helmer et al., 2026 · European Journal of Applied Physiology, · Paper · PubMed

Ten aerobic sessions on an underwater bike in 35 °C water lengthened a 30-min time trial in hot air by 1.1 km, against 0.5 km after the same training in 25 °C water, in a randomised trial of 24 recreational athletes (12 per group); core and skin temperature, sweat rate, plasma volume, heart rate and VO2max did not change differently between groups

Main caveat: One trial of 24 (12 per group), mostly men (22 M / 2 F); read from the abstract only.
limited evidenceMeasured on recreational to trained athletesRead from the abstract only

Helmer and colleagues asked whether exercising in hot water, rather than in hot air or by sitting in a hot bath, prepares athletes for performance in the heat. Twenty-four recreational athletes (22 men, 2 women) were randomised to 10 aerobic sessions on an underwater bike in 35 °C water or in 25 °C water.

Two to three days before and after the training block they rode a 30-min time trial in hot air (38.4 ± 1.4 °C, 45.3 ± 6.9% humidity) and had a cardiac echography, blood analysis and a VO2max test in temperate conditions. Both groups covered more distance after training, the hot-water group more (group effect p = 0.024): +1.1 km against +0.5 km.

Power output tended to improve more in the hot-water group (+27 W vs +13 W), but this was not significant (p = 0.096). No physiological measure taken during the hot time trial (core and skin temperature, sweat rate) or in the temperate tests (plasma volume, heart rate, VO2max) changed differently between groups.

The authors conclude that cycling in hot water improves performance in hot conditions without changing physiological parameters measured in temperate conditions.

Participants

n = 24 recreational athletes (22 men, 2 women), randomised 12 to each group.
Age, sport, training volume, fitness and heat-acclimation history: not in the abstract.

Protocol

Intervention: 10 aerobic training sessions on an underwater bike in 35 °C water (hot water acclimation, HWA).
Control: the same 10 sessions in 25 °C water (CON).
Before and 2-3 days after: 30-min time trial at 38.4 ± 1.4 °C and 45.3 ± 6.9% RH, with core temperature, skin temperature and sweat rate measured; cardiac echography, blood analysis and a cardiopulmonary VO2max test in temperate conditions.
Not in the abstract: session duration and intensity, the number of sessions per week and the length of the block, immersion depth, how core temperature was measured, whether assessors were blinded, dropouts, and confidence intervals for the power result.

Key Results

Outcome (before → after)HWA, 35 °C waterCON, 25 °C waterBetween groups
30-min time-trial distance in the heat+1.1 km (bracketed interval +0.7 to +1.4, type not stated in the abstract; p < 0.001)+0.5 km (+0.1 to +0.8; p = 0.009)group effect p = 0.024
Time-trial power output+27 W+13 Wp = 0.096, not significant
Core and skin temperature, sweat rate in the hot time trialno difference in change between groupsnot significant (values not in the abstract)
Plasma volume, heart rate, VO2max in temperate testsno difference in change between groupsnot significant (values not in the abstract)

Practical Application

Training on an underwater bike in 35 °C water improved a 30-min time trial in the heat by about twice as much as the same training in 25 °C water (+1.1 vs +0.5 km), in one small trial of recreational athletes. It is a heat-preparation route for someone with access to an underwater bike; the abstract gives no session dose to copy.
Where it meets [SYNTH-04]: the guide's access table offers passive hot-water immersion at 40 °C or more for athletes who cannot train in the heat; the guide lists this trial beside it as the one test of an active in-water route, at a lower water temperature (35 °C) with a performance gain in the heat, which agrees with the guide's point that training in the heat beats the same training in thermoneutral conditions for heat performance (REF-1667).It sits against the guide in two places, on thin evidence: the guide's thermoregulatory row says heat training lowers exercise core temperature and heart rate and raises sweat rate more than the same training in thermoneutral conditions (REF-1667), and the guide judges a block by heart rate at a fixed submaximal intensity (practice); here core and skin temperature and sweat rate in the hot time trial, and plasma volume, heart rate and VO2max in temperate tests, were not shown to change differently between groups, while the time trial did.With 12 per group and no values in the abstract this is no detected difference, not evidence of no adaptation, and heart rate at a fixed load in the heat, the guide's actual check, is not reported. The guide's dose, access table and plasma-volume and haemoglobin-mass routes are not tested here.
Where it meets [SYNTH-06]: the guide calls heat the cheaper experiment because it needs no travel; this route needs a heated pool with an underwater bike, so it is no-travel only where such a pool exists (inference, not tested here).
Inference, not tested here: because no adaptation marker separated the groups, this trial gives no sign that heart rate or sweat rate would show whether this route worked; the hot-weather time trial was the only measure reported to differ between groups (whether the markers changed within each group is not in the abstract).
Not from this paper: whether it beats land-based heat training or a hot bath after training, whether it helps performance in cool conditions (only a temperate VO2max test was run, with no between-group difference), and how many sessions or minutes are needed.

Limitations

One trial of 24 (12 per group), mostly men (22 M / 2 F); read from the abstract only.
The power-output gain the authors describe as larger in the hot-water group was not significant (p = 0.096).
No physiological mechanism was shown: none of the measured markers changed differently between groups, so why performance improved is not established.
Both groups trained in water; the control was 25 °C water exercise, not land training, so the comparison isolates water temperature, not the aquatic mode.
Not in the abstract: session dose and schedule, the athletes' fitness and sports, blinding, dropouts, how the time trial was performed, the values of the physiological measures, and funding or conflicts.
How this entry was checked and changed
Why this level

a parallel trial of 12 per arm (the rubric's B+ band) held one step lower because only the abstract was read and it gives no session dose, blinding or dropouts; the main outcome is performance, reported with an interval, so no further step