| Study | Design & population | Finding & comparison |
|---|---|---|
| 2025 · Solomon TPJ ↗ | Systematic review and meta-analysis Healthy adults in post-exercise heating trials Participants: 199. Sauna or hot-water immersion | Pooled performance effects were uncertain, and evidence certainty was low to very low despite changes in some physiological measures. Comparison: Same exercise training without post-exercise heating Small underlying trials |
| 2021 · Kirby NV ↗ | controlled endurance-training intervention trained middle-distance runners Participants: 20. post-exercise sauna · 3 weeks; subset continued to 7 weeks | Added sauna improved several tested exercise-capacity markers; extended exposure produced limited additional change. Comparison: normal training without added sauna Small trained sample; no randomization stated in abstract; surrogate performance markers. |
| 2023 · Ashworth E ↗ | partially randomized crossover heat-acclimation experiment military-context participants Participants: 25. sauna versus hot-water immersion · 5 days | Both modalities produced modest heat adaptations; the performance improvement was not statistically significant. Comparison: sauna versus water immersion; 15 participants completed both Only a subset crossed over; small sample; specific loaded walking task. |
Original publication: The effect of post-exercise heat exposure (passive heat acclimation) on endurance exercise performance: a systematic review and meta-analysis.
What did the review compare?
The review asked whether passive heating after exercise improves endurance performance beyond the same training without the added heat. That comparison is more specific than asking whether people improve during a period containing exercise and sauna. A matched underlying training program helps distinguish the added intervention from the training shared by both groups.
The included heating approaches were sauna and hot-water immersion. The pooled question therefore does not concern one universal sauna technology or cabin setting. A reader evaluating a product claim should identify whether the relevant result actually isolates that exposure. Shared classification as passive heating is not enough to give every modality, schedule or population an identical performance effect.
Who contributed to the evidence base?
The underlying studies enrolled healthy young adults. Across the registered participant totals, men substantially outnumbered women. This affects how readily a pooled athletic conclusion can be applied to groups that were less represented, including older adults or people with medical conditions.
Pooling expands the evidence view, but it does not turn selected small experiments into a representative population sample. The studies also differed in exercise tasks, heating methods and assessment environments. Those differences are important when interpreting the average result. A review can organize heterogeneous evidence while still leaving uncertainty about which particular exposure, if any, produces a meaningful improvement for a defined athlete and task.
Why did the review distinguish performance from physiology?
The review examined actual performance outcomes alongside aerobic-capacity, heat-response and other physiological measures. Those categories answer different questions. A change in exercise temperature, heart rate or another marker may help explain a possible adaptation without establishing better completion of a performance task.
This distinction is essential to interpreting the synthesis. Some physiological measures changed, yet the authors considered the performance effect uncertain. Combining the categories into one broad positive conclusion would erase that finding. A performance claim should therefore start with the relevant task estimate, then use marker results as context rather than substitutes. The review is valuable because it keeps the practical outcome question explicit.
What did the synthesis establish about endurance performance?
It did not establish a dependable endurance-performance benefit from post-exercise heating. The pooled estimates were uncertain, and the authors rated evidence certainty low to very low across outcomes. That conclusion should remain visible even if selected individual studies or physiological results were favorable.
Uncertainty does not prove that every possible program is ineffective. It means the present synthesis cannot support a firm general conclusion about improvement. A reader should not replace that conclusion with either a guaranteed benefit or an absolute statement that research can never find one. The useful next question is which future trials could reduce the uncertainty by matching task, exposure, population and comparison more closely.
How do confidence and prediction intervals differ?
A confidence interval expresses uncertainty around the modeled pooled estimate. A prediction interval addresses the range of effects that might be expected across settings under the model’s assumptions. They describe different aspects of uncertainty and should not be relabeled as interchangeable statistics.
That distinction matters when considering a new athlete or program. Even a more precisely estimated average would not guarantee the same effect in every future setting. Here, the source also reported limited predictive certainty. This page avoids adding numerical interval limits absent from the registered result facts. The central interpretation is that the review’s uncertainty concerns both the estimated average and the transfer of that estimate to another study context.
| Evidence item | What it addresses | What it cannot supply alone |
|---|---|---|
| Pooled performance estimate | Average modeled intervention effect | A guaranteed gain for a new athlete |
| Confidence interval | Uncertainty around that pooled estimate | The full range of future setting effects |
| Prediction interval | Variation expected across settings under assumptions | Individual clinical suitability |
| Physiological marker change | A measured response or possible mechanism | Proof of faster task performance |
Why does a meta-analysis still receive low certainty?
The authors identified small samples, high risk of bias, imprecision and other limitations affecting confidence in the findings. Combining studies statistically cannot remove those weaknesses automatically. The review’s design label therefore should not be mistaken for a guarantee that its intervention effect is established.
The reported certainty assessment was conducted by the review authors. This editorial summary does not claim to have independently repeated their full risk-of-bias or grading process. It records the conclusion and its stated basis, while preserving the abstract-only appraisal boundary. A full assessment would inspect eligibility, study selection, data extraction, statistical choices and the underlying papers before endorsing stronger claims about the pooled result.
How can favorable and null trials coexist?
The runner study found selected capacity-marker and heat-test improvements beyond normal training. The military comparison reported modest physiological adaptation but no statistically clear performance improvement in its loaded task. The hematological crossover trial found no clear added benefit over endurance training in its assessed variables.
These results can coexist because they differ in participants, tasks, exposure and measured outcomes. A review should not be interpreted as a vote among favorable titles. Nor should a single positive marker result overrule a performance-focused synthesis without addressing those design differences. The sources help define which questions remain open, rather than supplying interchangeable estimates of one universal sauna effect.
Can the review identify the best sauna schedule?
Not from the uncertain, mixed evidence summarized here. The review combined sauna and hot-water interventions with differing schedules. It does not establish a universally effective temperature, exposure duration or frequency for improving a particular endurance event.
A practical schedule assembled from separate favorable studies would be a new intervention. Borrowing a session length from one source and a weekly frequency from another does not mean the combination was tested. The relevant evidence needs to keep the preceding exercise, heating method, comparison and performance assessment together. This is particularly important when a tool offers a precise-looking plan: precision in displaying assumptions is different from validation of a performance-enhancing program.
What are the review and disclosure boundaries here?
The indexed abstract and verified corpus abstraction were read for this source summary. The paper has a free full-text link, but this appraisal has not repeated its searches, checked every included study or completed a full-text methodological assessment. It also does not claim that the review captures every study available after its publication.
Funding and conflict-of-interest details are unreported in the abstract-level abstraction. Those gaps are not assumed absent. The historical safety review supplies separate context about contraindications and alcohol-related concerns; it cannot turn pooled athletic findings into personal clearance. The defensible conclusion is uncertain endurance benefit, with low-certainty evidence and no optimal routine established by this summary.
Explore the evidence with a tool.
Questions, considered
Questions about this study
Did the 2025 post-exercise heating meta-analysis establish better endurance performance?
No. It judged the performance benefit uncertain, with imprecise pooled effects and low or very low certainty across outcomes. Some physiological measures changed, but those changes did not provide firm evidence of a meaningful performance improvement. Better-controlled trials were needed before drawing a confident practical conclusion.
Follow the sourceDid the post-exercise performance review examine sauna alone?
No. It combined eligible studies using sauna or hot-water immersion after exercise. The comparison groups completed the same training without the heating intervention. This design addresses supplementary passive heating in general; its pooled result should not be presented as a precise estimate for one particular sauna technology.
Follow the sourceWhy did the endurance review distinguish confidence and prediction intervals?
A confidence interval describes uncertainty around the pooled average, whereas a prediction interval considers how a future comparable study might differ. The review found poor precision and limited predictive certainty. That combination cautions against promising that an individual athlete will reproduce an apparently favorable average effect.
Follow the sourceFollow the evidence
Sources & references
Research summaries in this edition are based on indexed abstracts unless otherwise noted. A source link is not a claim of medical review.
- The effect of post-exercise heat exposure (passive heat acclimation) on endurance exercise performance: a systematic review and meta-analysis.
Solomon TPJ, Laye MJ
Volume 17, issue 1, pages 4 · PMID 39762944
Open-access flag: yes.Abstract-level editorial review - Intermittent post-exercise sauna bathing improves markers of exercise capacity in hot and temperate conditions in trained middle-distance runners.
Kirby NV, Lucas SJE, Armstrong OJ, Weaver SR, Lucas RAI
Volume 121, issue 2, pages 621-635 · PMID 33211153
Open-access flag: yes.Abstract-level editorial review - Post-exercise, passive heat acclimation with sauna or hot-water immersion provide comparable adaptations to performance in the heat in a military context.
Ashworth E, Cotter J, Kilding A
Volume 66, issue 1, pages 49-60 · PMID 35332846
Open-access flag: not recorded.Abstract-level editorial review - Hematological Adaptations to Post-Exercise Sauna Bathing With No Fluid Intake: A Randomized Cross-Over Study.
Sitkowski D, Cisoń T, Szygula Z, Surała O, Starczewski M, Sadowska D, Malczewska-Lenczowska J
Volume 93, issue 4, pages 795-803 · PMID 34727008
Open-access flag: not recorded.Abstract-level editorial review - Benefits and risks of sauna bathing.
Hannuksela ML, Ellahham S
Volume 110, issue 2, pages 118-126 · PMID 11165553
Open-access flag: not recorded.Abstract-level editorial review