Study
Temporal dynamics of muscle strength recovery following acute cold-water immersion: a systematic review and meta-analysis.
Yang Zhu, Lele Yang, Tao Liu, Fuya Yao, Qilong Wang, Zheng Yi · PeerJ · 2026
Read the original paper ↗The short version
This systematic review and meta-analysis pooled 22 randomized controlled trials with 375 mostly male participants to ask when, if at all, a single cold-water immersion (CWI) after exercise helps muscles recover. CWI means immersing the body, from navel depth up to the neck, in water of roughly 5 to 20 degrees C for about 6 to 20 minutes. Exercise-induced muscle damage (EIMD) is the loss of force, soreness and raised creatine kinase (CK, a blood marker leaking from stressed muscle cells) that follows hard or unaccustomed exercise, often with delayed onset muscle soreness (DOMS). Earlier reviews mixed time points, so this one separated results at 0, 24, 48 and 72 hours. The included trials enrolled healthy non-athletes and athletes aged about 16 to 49, who exercised with intervals, simulated matches, distance running or heavy eccentric lifting. After exercise, one group sat in cold water and the comparison group rested seated for a similar time. Researchers measured maximal voluntary isometric contraction (MVIC, the hardest possible static muscle squeeze), countermovement jump (CMJ, jump height as a gauge of explosive power), CK, and soreness on a visual analog scale (VAS, a line marked from no pain to worst pain). Maximal strength was unchanged by CWI (g = 0.08, 95% CI -0.11 to 0.26, p = 0.42; a negligible effect) at every time point. Jump performance was worse immediately afterward (g = -0.68, p = 0.01; a medium-sized drop) and showed no difference by 24 to 48 hours (g = 0.38 at 24 h, p = 0.06; g = 0.06 at 48 h, p = 0.68). CK looked lower after CWI (g = -0.43, p < 0.001; a small effect), but the result shrank to non-significance after correcting for likely missing small studies (g = -0.16, 95% CI -0.44 to 0.12; the original benefit was probably overestimated). Soreness was lower (g = -0.58, p = 0.006; a medium effect), though the trials disagreed widely (I² = 87%) and the certainty of evidence was rated very low. The evidence has clear limits: 362 of 375 participants were male, participants could not be blinded to cold water, several trials were tiny, and only one immersion with follow-up of 72 hours or less was studied. The authors suggest the soreness effect is the most plausible benefit and the temporary jump decrement is the main drawback. Repeated use during training and its effect on muscle adaptation were not examined, and they call for larger trials that include women and standardize temperature and duration.
Key findings
- In this review of 22 randomized trials, a single cold-water immersion after exercise did not change maximal muscle strength (MVIC, the hardest possible static muscle squeeze) at any time from immediately after exercise to 72 hours (g = 0.08, 95% CI -0.11 to 0.26; a negligible effect).
- Jump performance was lower immediately after immersion (g = -0.68, a medium-sized drop) but no longer different from rest by 24 to 48 hours, so the effect was temporary.
- Reductions in creatine kinase (CK, a blood marker of muscle damage) and in muscle soreness appeared in the main analysis, but the CK result disappeared after correcting for publication bias (g = -0.16, 95% CI -0.44 to 0.12) and both lost statistical significance when trials were clustered, with very low certainty for soreness.
- Almost all participants were male (362 of 375), participants could not be blinded to cold water, and the review covered only one immersion with follow-up of 72 hours or less, so effects of repeated use on training adaptation were not assessed.
- No adverse reactions were reported in any of the included studies.
Common questions
What healing modality was studied?
Hydrotherapy in the form of acute cold-water immersion: a single bout of immersion in cold water (roughly 5 to 20 degrees C for about 6 to 20 minutes) within one hour after exercise, compared with seated passive recovery.
What condition was addressed?
Recovery from exercise-induced muscle damage in healthy people and athletes, measured as maximal isometric strength, countermovement jump height, creatine kinase (a marker of muscle damage) and delayed-onset muscle soreness at 0, 24, 48 and 72 hours.
What were the main findings?
Maximal strength was unchanged (g = 0.08, 95% CI -0.11 to 0.26). Jump performance was impaired immediately after exercise (g = -0.68, 95% CI -1.16 to -0.20) but not by 24 to 48 hours. Creatine kinase (g = -0.43) and soreness (g = -0.58) were lower in the main analyses, but the CK result became non-significant after trim-and-fill adjustment (g = -0.16) and both were non-significant in cluster-robust analyses.
How strong is the evidence?
Rigorous review (PROSPERO-registered, four databases, PRISMA, GRADE) of 22 RCTs, but certainty ranged from moderate for strength to very low for soreness. All trials were high risk of bias on RoB 2 because immersion cannot be blinded, 362 of 375 participants were male, several trials were very small, and publication bias and heterogeneity affected CK and soreness.
What are the clinical implications?
The authors suggest acute immersion is most relevant when short-term symptom control and next-day recovery readiness are prioritized, and that its timing deserves care when a power-dependent task follows soon after exercise. They note that the findings cannot be extrapolated to women or to repeated use during training, and call for larger sex-balanced trials with standardized dosing.
This summary and its findings were generated by AI from the published paper and may contain imperfections. It describes a study and is not medical advice — talk to a qualified clinician about your own health (terms). The original study is always the source of truth — read it above.