Pooled over 211 heat-acclimation papers, the typical block is 8 ± 4 exposures of 90 ± 36 min and buys a 3.1% time-trial gain
And adding sessions buys a different adaptation from making sessions longer
Key Findings
The protocols actually used, averaged over the literature: 8 ± 4 exposures, 90 ± 36 min per exposure, 39.1 ± 4.8 °C ambient, 2.78 ± 0.83 kPa water-vapour pressure.
Pooled effects of heat acclimation: resting heart rate −5 bpm [−7, −3]; end-exercise heart rate −17 bpm [−19, −14]; resting core temperature −0.19 °C [−0.23, −0.14]; end-exercise core temperature −0.43 °C [−0.48, −0.36]; plasma volume +5.6% [3.8, 7.0]; exercise metabolic rate −87 mL/min [−126, −49]; whole-body sweat rate +163 mL/h [94, 226].
Performance: time-trial +3.1% [1.8, 4.5]; time to exhaustion +49% [35, 61]; incremental exercise time +14% [7, 24]. Inference, not tested: the time-trial number is the one to quote — open-ended time-to-exhaustion tests produce large percentage changes by design.
One extra EXPOSURE buys: haemoglobin mass +1.9 g [0.6, 3.2] and whole-body sweat rate +9 mL/h [1, 17].
full text, checked 2026-09-27: an extra exposure did not further move end-exercise heart rate (0.0 bpm [−0.4, 0.4]) or plasma volume (−0.1% [−0.2, 0.1]) and moved end-exercise core temperature only −0.01 °C [−0.02, −0.00].
One extra 15 MINUTES per exposure buys: end-exercise core temperature −0.04 °C [−0.05, −0.03] and plasma volume +0.4% [0.1, 0.7]. (full text, checked 2026-09-27: end-exercise heart rate −1.1 bpm [−1.6, −0.5].)
Environment matters at the margin: a 5 °C hotter room further lowers end-exercise heart rate by 2 bpm [−3, −1]; 1 kPa more water-vapour pressure adds +37 mL/h [4, 72] of sweat rate. (full text, checked 2026-09-27: a 5 °C hotter room also adds −0.04 °C [−0.06, −0.02] end-exercise core temperature.)
The approach matters less than the exposure. End-exercise heart rate and core temperature fell similarly under controlled hyperthermia (−16 bpm [−18, −14]; −0.43 °C [−0.48, −0.36]) and constant work rate (−17 bpm [−18, −16]; −0.45 °C [−0.49, −0.42]).
full text, checked 2026-09-27: resting heart rate separated more by approach — controlled hyperthermia −8.0 bpm [−9.6, −6.3] vs constant work rate −5.0 [−6.5, −3.6], controlled heart rate −4.1 [−6.3, −1.6], intermittent −5.2 [−7.9, −2.3].
full text, checked 2026-09-27: the authors' framing of kinetics — most adaptations are induced within the first week and generally considered complete after 10-14 days of exposure.
The authors provide an online predictor that estimates HA adaptations from protocol characteristics (University of Canberra).
Practical Application
Pooled over 211 papers, heat acclimation (typically 8 ± 4 exposures of 90 ± 36 min at a mean 39.1 °C) improved time-trial performance by 3.1% [1.8, 4.5], expanded plasma volume by 5.6% and cut end-exercise heart rate by 17 bpm; each extra exposure added 1.9 g haemoglobin mass and 9 mL/h sweat rate, each extra 15 min per exposure added −0.04 °C end-exercise core temperature and +0.4% plasma volume. Plan a real block, not "a few sauna sessions".
Lengthen the session to move core temperature and plasma volume; add sessions to move sweat rate and haemoglobin mass. Which lever to pull depends on which adaptation the race needs.
Tell the athlete to expect about 3%, not a transformation. A time-trial gain of 1.8-4.5% is the honest range; the 49% time-to-exhaustion figure should not be quoted to an athlete as a race gain.
Do not agonise over controlled hyperthermia versus a fixed workload — end-exercise heart rate and core temperature land in the same place. Choose whichever the athlete can actually execute.
Inference, not tested: a humid room is not a worse room — more water-vapour pressure went with more sweat-rate adaptation, and a hotter room with a little more heart-rate adaptation. Use what is available.
Not from this paper; the base's own practice: roughly 8-10 exposures of 60-90 min is the working prescription, built around the literature means above.
Limitations
Meta-regression over heterogeneous protocols, not a trial. Dose-response coefficients are associations across studies, not the result of randomising athletes to more or fewer exposures.
full text, checked 2026-09-27: study quality is a constraint the authors name — 60% of included studies had no control group, 19% did not randomise, 20% did not justify sample size.
full text, checked 2026-09-27: 8.9% female participants — sex-specific dose-response cannot be read from this.
Full text, checked 2026-09-27: linear models only — smooth (non-linear) terms did not converge, so a plateau point cannot be located.
Decay and re-acclimation are not addressed here.
Inference, not tested: the pooled base is trained but not elite, and the highest tiers may adapt less.
How this entry was checked and changed
Why this level
regraded 2026-09-27 from A: meta-regression across 211 mostly small heat-acclimation studies with heterogeneous protocols, many without a control group
Content correction 2026-09-27
Level A -> A- (meta-regression over mostly small, heterogeneous, often uncontrolled studies).
Numbers not in the abstract (participant count/VO2max/% female, per-exposure null effects on HR and plasma volume, per-15-min HR, hotter-room core temperature, resting HR by approach, 10-14-day kinetics, study-quality percentages, model convergence) labelled full text.
Hotter room "−1.8 bpm [−3.0, −0.7]" -> abstract: −2 bpm [−3, −1]; added the abstract's end-exercise HR by approach (−16 vs −17 bpm).
"8-10 exposures of 60-90 min" relabelled house practice; "each extra day" -> each extra exposure; stale REF-002 decay pointers removed; lead now carries the pooled numbers.
Lines labelled "full text; not checked against the abstract" were checked against the open-access full text (PMC12122934): every number in them appears there; relabelled "full text, checked 2026-09-27".
Lines labelled "full text; not checked" were read against the open-access full text (PMC12122934): 1 confirmed and relabelled.