REF-1585 · Kirby et al., 2021 · European Journal of Applied Physiology · Paper · PubMed · Free full text
In trained middle-distance runners who chose their own group (12 sauna, 8 control places), about 28 min in a 101-108 °C dry sauna after mostly easy runs, three times a week for 3 weeks, lowered heart rate (11 bpm more than control), rectal and skin temperature in a fixed-pace heat run and raised VO2max (+0.27 L/min) and speed at 4 mmol/L lactate (+0.6 km/h) more than training alone; 7 weeks added only 0.1 °C, and 19 of 37 starters dropped out injured
Main caveat: Self-allocation: runners chose their group, so motivation or other differences cannot be excluded; the authors argue both groups were equally invested.
Kirby and colleagues tested whether a hot dry sauna after normal training works as heat acclimation and helps performance in cool conditions. Runners chose either normal training or normal training plus about 28 min in a 101-108 °C sauna straight after runs, three times a week.
After 3 weeks the sauna group, compared with control, had lower peak rectal temperature, skin temperature and heart rate in a 30-min run at a fixed pace in 40 °C heat, more active forearm sweat glands, and larger gains in VO2max, ramp time to exhaustion and speed at 4 mmol/L lactate, with training volume not differing significantly between groups.
Six sauna participants who carried on to 7 weeks lowered peak rectal temperature by only 0.1 °C more. The authors call the method an effective, practical way to acclimate and an ergogenic aid; they also report a 51% dropout, mostly injuries from training in both groups.
Participants
51 runners recruited; of 37 who began, 19 did not finish because of injury; 20 group places were analysed at 3 weeks (sauna 12: 9 women, 3 men; control 8: 4 women, 4 men). Two women took part in both groups in turn, about 5 weeks apart. Our arithmetic: 37 − 19 = 18 people, which matches 20 places only with those two counted twice.
Sauna vs control at baseline: 19 ± 1 vs 20 ± 2 y, 169 ± 7 vs 172 ± 6 cm, VO2max 59.3 vs 58.4 ± 7.2 mL/kg/min as printed (all p > 0.05).
Women: eumenorrhoeic 7, oligomenorrhoeic 1, amenorrhoeic 1, hormonal contraception 3; 5 of the 8 normally menstruating women did every test in the same cycle phase, and the 2 pill users were tested in the active-pill phase.
All took 65 mg ferrous sulphate daily from 2 weeks before the first test to the end.
Protocol
Sauna: 101-108 °C (at about 1.8 m), 5-10% humidity, entered within about 5 min of finishing an outdoor run, usually an easy or long run; asked to stay 30 min (achieved 28 ± 2 min), sitting upright, drinking as wanted; 3 ± 1 sessions a week, 9 ± 1 sessions before the cool-condition tests and 10 ± 1 before the heat test; 290 ± 48 min in total; peak heart rate in the sauna 127 ± 10 bpm.
Heat test: 30 min at 9 km/h, 2% gradient, 40 °C and 40% humidity, no drinking.
Cool-condition tests (about 18 °C): stepwise lactate test to speed at 4 mmol/L, then a ramp VO2max test with the same protocol repeated to measure time to exhaustion.
Training: 6 ± 1 (sauna) vs 7 ± 2 (control) sessions a week, 53.3 ± 23.3 vs 54.5 ± 28.2 km a week; type, frequency, distance and session RPE did not differ significantly.
Analysis: ANCOVA on the 3-week change with baseline as covariate (sex added for VO2max and lactate speed); a sensitivity analysis without the second-arm data of the two women is in the supplement.
Funding: University of Birmingham; no competing interests declared.
Key Results
| Outcome, before → 3 weeks | Sauna | Control | Difference in change (95% CL), p |
|---|---|---|---|
| Peak rectal temperature, heat test (°C) | 38.6 ± 0.5 → 38.3 ± 0.3 | 38.6 ± 0.4 → 38.5 ± 0.3 | 0.2 [0.0, 0.4] larger fall, p = 0.04 |
| Peak skin temperature (°C) | 36.4 ± 1.3 → 35.7 ± 0.8 | 36.3 ± 1.4 → 36.5 ± 1.2 | 0.8 [0.1, 1.5] larger fall, p = 0.03 |
| Peak heart rate, heat test (bpm) | 163 ± 24 → 152 ± 19 | 157 ± 16 → 158 ± 21 | 11 [0, 22] larger fall, p < 0.05 |
| Forearm active sweat glands (per cm²) | 51 ± 24 → 70 ± 20 | 45 ± 20 → 44 ± 24 | p < 0.01 |
| VO2max (L/min) | 3.42 ± 0.81 → 3.65 ± 0.75 | 3.72 ± 0.76 → 3.73 ± 0.55 | +0.27 [0.05, 0.49], p = 0.02 |
| Ramp time to exhaustion (s) | 426 ± 57 → 474 ± 59 | 439 ± 62 → 433 ± 79 | +54 [17, 90], p < 0.01 |
| Speed at 4 mmol/L lactate (km/h) | 15.6 ± 1.7 → 16.2 ± 1.7 | 16.8 ± 1.5 → 16.8 ± 1.4 | +0.6 [0.1, 1.0], p = 0.01 |
Rise in rectal temperature during the heat test did not differ significantly (p = 0.14); sweat loss fell in the control group but not in the sauna group (difference 0.2 kg, p = 0.02); thermal sensation and comfort fell more with sauna.
Running economy (p = 0.73), respiratory exchange ratio (p = 0.14), resting heart rate (p = 0.90), body mass (p = 0.43), EPO (p = 0.82) and VEGF (p = 0.09, detectable in 10 of 20 samples) did not differ significantly between groups.
7 weeks (6 sauna participants, no control): peak rectal temperature 0.3 °C below baseline at 3 weeks and 0.4 °C at 7 weeks, a further 0.1 °C (p = 0.03); no other physiological heat-test variable changed further (authors), but peak RPE fell from 12 ± 3 to 10 ± 2 by 7 weeks, lower than at 3 weeks too (p = 0.03).
Practical Application
Sauna after runs: in 20 group places of trained club runners (mostly women) who picked their own group, about 28 min in a 101-108 °C sauna after mostly easy runs three times a week for 3 weeks brought heat-acclimation signs and larger lab gains in VO2max and lactate speed than training alone, while training did not differ significantly between groups — so in this group the sauna did not cost fitness. The gains come from a small, non-randomised study with heavy dropout; across 10 small trials the pooled effect of post-exercise heat on performance is trivial and uncertain (REF-334b).
Timing as tested: within about 5 min of finishing, typically after easy or long runs; a sauna after hard intervals was not the case studied.
Dose as tested: about 10 sessions of about 28 min over 3 weeks, drinking freely; 7 weeks added little (0.1 °C).
The paper does not mention adverse events from the sauna; the dropouts were mainly injuries from normal training, in both groups.
Everyone took 65 mg ferrous sulphate a day, which the authors suggest may have helped the oxygen-carrying response. Inference, not tested here: a runner whose iron status is unknown or low may respond differently.
Practice, not from this paper: SYNTH-04 lists a 15-30 min sauna straight after training as the base's own practice for heat preparation; this study's dose sits inside that window (our comparison), and where SYNTH-04 notes that sauna rests only on the low-certainty pool of REF-334b, this is a single direct sauna trial.
Limitations
Self-allocation: runners chose their group, so motivation or other differences cannot be excluded; the authors argue both groups were equally invested.
51% dropout, mostly injury; two women counted in both groups without modelling the repeated measure.
The heat test was at a fixed pace; the authors say the effect on self-paced performance in heat or cool conditions is unknown. Speed at a fixed 4 mmol/L is one lactate marker.
The 7-week arm had 6 people and no control.
EPO and VEGF were often undetectable; blood volume and haemoglobin mass were not measured.
How this entry was checked and changed
Why this level
12 and 8 (10 per arm) would be B+ for a randomised trial, but allocation was self-chosen; held at B- because 19 of 37 starters were lost to injury and two women filled a place in each group without the repeated measure being modelled. Outcomes are laboratory capacity measures (VO2max, lactate speed, ramp time to exhaustion) and heat strain at a fixed pace, not a race or time trial