Purpose: Cold water immersion (CWI) provides rapid cooling in events of exertional heat stroke. Optimal procedures for CWI in the field are not well established. This meta-analysis aimed to provide structured analysis of the effectiveness of CWI on the cooling rate in healthy adults subjected to exercise-induced hyperthermia.

Methods: An electronic search (December 2014) was conducted using the PubMed and Web of Science. The mean difference of the cooling rate between CWI and passive recovery was calculated. Pooled analyses were based on a random-effects model. Sources of heterogeneity were identified through a mixed-effects model Q statistic. Inferential statistics aggregated the CWI cooling rate for extrapolation.

Results: Nineteen studies qualified for inclusion. Results demonstrate CWI elicited a significant effect: mean difference, 0.03°C·min(-1); 95% confidence interval, 0.03-0.04°C·min(-1). A conservative, observed estimate of the CWI cooling rate was 0.08°C·min(-1) across various conditions. CWI cooled individuals twice as fast as passive recovery. Subgroup analyses revealed that cooling was more effective (Q test P < 0.10) when preimmersion core temperature ≥38.6°C, immersion water temperature ≤10°C, ambient temperature ≥20°C, immersion duration ≤10 min, and using torso plus limbs immersion. There is insufficient evidence of effect using forearms/hands CWI for rapid cooling: mean difference, 0.01°C·min(-1); 95% confidence interval, -0.01°C·min(-1) to 0.04°C·min(-1). A combined data summary, pertaining to 607 subjects from 29 relevant studies, was presented for referencing the weighted cooling rate and recovery time, aiming for practitioners to better plan emergency procedures.

Conclusions: An optimal procedure for yielding high cooling rates is proposed. Using prompt vigorous CWI should be encouraged for treating exercise-induced hyperthermia whenever possible, using cold water temperature (approximately 10°C) and maximizing body surface contact (whole-body immersion).