Cold and heat exposure 02: Heat after training

Short answer. Heat after training showed a small possible benefit for muscle size and none for strength, pooled from 125 people in studies of fair to poor quality. Both heat and cold reduce soreness, and only cold has a point estimate pointing away from growth. Heat after training is a reasonable thing to do and a poor thing to count on.

What it is

Deliberate heating after a training session. In the studies below that means sauna bathing, or hot water immersion to at least sternum level, at temperatures from around 36°C upward, taken shortly after resistance training. A separate line of research uses local heating applied to one muscle group rather than the whole body.

Heat is sold in the same aisle as cold, often in the same facility, on the opposite logic. Cold calms the tissue down; heat opens it up. Two contradictory stories, one recovery suite, and until recently very little to separate them.

The proposed mechanism

Three chains, all with some support at the cellular level.

Blood flow. Heat dilates vessels and increases muscle blood flow, which is proposed to improve delivery of oxygen, nutrients and amino acids to recovering tissue. Notably, local heating raises muscle blood flow where indirect whole-body heating does not (Heinonen et al., 2011), which is a distinction most popular accounts collapse.

Heat shock proteins. Heat stress increases expression of chaperone proteins involved in protecting and refolding damaged proteins. These have been linked in cell and animal work to both the prevention of atrophy and the support of growth.

Anabolic signaling. Controlled heat stress has been shown in cell models to promote myofibril formation, and heat has been reported to interact with the pathways governing muscle protein synthesis.

Each of these is a reason to expect an effect. None of them predicts a size.

What has actually been tested

The pooled hypertrophy result. A Bayesian meta-analysis pooled six interventions comprising 125 participants, all applying heat therapy shortly after resistance training over blocks of four weeks or longer. It found a small advantage for heat on muscle size: a standardized mean difference (SMD, the gap between groups in standard deviations) of 0.12. The 95% credible interval, the range within which the true effect most probably lies, ran from -0.09 to 0.31, and the posterior probability that the effect was greater than zero was 89.7%. For strength there was little evidence of a difference either way: SMD -0.05, credible interval -0.28 to 0.19, with a 68.7% posterior probability favoring the control condition. The authors rated the included studies as generally fair to poor in methodological quality and asked that the results be read cautiously (Hopper et al., 2026).

Set that beside the equivalent analysis for cold, which found SMD -0.22 with a credible interval of -0.47 to 0.04 (Piñero et al., 2024). Two opposite interventions, two small effects pointing in opposite directions, both with intervals that touch zero. The recovery suite contains a hot tub and a cold tub, and the best current estimate is that each moves twelve-week muscle growth by a fraction neither you nor your training partner would notice.

The whole-body performance picture is not clean. A systematic review comparing whole-body post-exercise heat against passive or placebo recovery found the adaptation evidence limited and conflicting. In one trial of six elite speed skaters, hot water immersion produced larger isometric knee extensor strength gains than passive recovery, while the passive group gained more cross-sectional area in the same muscles and improved more on the skating test (Ahokas et al., 2025). Six participants, three outcomes, two directions.

The same review notes that the evidence for long-term heating enhancing hypertrophy comes largely from sedentary participants and may not transfer to people who already train.

Local heat is a different intervention. Repeated local heat therapy applied to a single muscle group has its own body of work, separate from sauna and full immersion, and it is the setting where the mechanistic case is strongest (Kim et al., 2020). Whether findings from a heating pad on one quadriceps transfer to a person sitting in a sauna is not established, and given the blood flow finding above there is reason to think they might not.

Where heat clearly outperformed cold. In a trial using muscle biopsies after a simulated muscle injury, ten days of hot water immersion reduced perceived pain and circulating damage markers, while cold immersion changed neither (Dablainville et al., 2025). That is an injury model rather than a training model, and the exposure was daily for ten days. It is the most direct head-to-head we have, and it went to heat.

And where neither did anything. In academy rugby players across a competitive season, repeated post-exercise cold or hot water immersion produced no effect on body composition or performance responses (Horgan et al., 2023).

Our reading

The interesting result in this literature is not which tub wins.

Set the two meta-analyses side by side and the effects point in opposite directions. Heat after training, a point estimate of 0.12 favoring heat. Cold after training, -0.22 favoring the control. Both intervals touch zero. Two interventions sold on opposite logic, in the same facility, each moving twelve-week muscle growth by an amount neither you nor anyone watching you would detect.

We would not put much weight on the heat number. Six interventions, 125 people, fair-to-poor study quality, and a strength estimate pointing nowhere. Treating a point estimate of 0.12 with an interval running below zero as a reason to buy a sauna is the move this publication exists to flag.

The asymmetry that does matter is in the cost column, not the benefit column. Both temperatures reduce soreness. Only one of them has a consistent point estimate pointing away from growth. If you want soreness relief and you are training for size, heat is currently the version of that trade with nothing measurable on the debit side, which is a reversal of what most recovery suites assume.

So: heat after training is a reasonable thing to do and a poor thing to count on. We would not restructure a program around 0.12.

Where the goals diverge

Training for size. The point estimate favors heat and the interval includes zero. On current evidence this is a small possible benefit rather than a reliable one, and the analysis behind it pools six interventions in 125 people. It is a reasonable thing to do and a poor thing to count on.

Training for strength. Nothing. The pooled interval sits on zero with the point estimate marginally on the control side.

Recovering between sessions or fixtures. Both heat and cold reduce soreness, and soreness is a poorer signal than it is usually treated as. (→ Inflammation 02: Delayed-onset muscle soreness) If soreness relief is what you want and hypertrophy is the goal you are protecting, heat currently looks like the version of that trade with no cost attached, which is a change from what most facilities assume.

Coming back from an injury. The strongest heat result in this piece comes from an injury model, and it is also the setting where we are least equipped to give useful guidance, because rehabilitation is a clinical decision with a clinician attached.

Interventions on the other side of the thermometer are covered separately. (→ Cold and heat exposure 01: The daily cold plunge; Inflammation 01: Cold water immersion; Cold and heat exposure sub-hub)

The case against

The evidence base is six interventions and 125 people. That is small for a pooled estimate, and the authors' own quality rating of the included studies was fair to poor. A single additional well-run trial could move this interval appreciably.

A posterior probability is not a result. An 89.7% probability that an effect exceeds zero, attached to a point estimate of 0.12 (Hopper et al., 2026), describes a finding that is probably real and definitely tiny. Those two facts get separated constantly in the retelling, and the first one travels while the second does not.

Sedentary and trained participants are not interchangeable. Much of the encouraging heat literature ran in people who were not training, and the review above says so directly.

The whole-body and local literatures have been merged when they should not have been. Local heating raises muscle blood flow; indirect whole-body heating does not. If blood flow is the mechanism, the sauna is not delivering it.

What is not known

Whether hot water immersion and sauna are equivalent is untested. They heat by different routes, reach different depths, and have been pooled together anyway.

The dose is unmapped. Temperature, duration, and how soon after training all vary across studies without anyone having run the comparison that would separate them.

Nobody knows whether heat and cold combined behaves like either one alone. Contrast protocols are widespread and their adaptation effects are largely unexamined. (→ Cold and heat exposure 03: Contrast therapy)

And whether any of this matters at the scale of a training year is unknown. Every trial here ran weeks, not seasons.

Safety

Sauna and hot immersion carry acute risks including dehydration, dizziness on standing, and fainting. Do not use them alone or after drinking alcohol. Anyone pregnant, or with a cardiovascular condition or uncontrolled blood pressure, should speak to a doctor before starting.