Effects of cold water immersion vs body cryotherapy on delayed onset muscle soreness and jump performance following acute strenuous exercise in physically active participants: A systematic review and meta-analysis

Abstract

Background:

Delayed onset muscle soreness (DOMS) and reduced athletic performance are common consequences of high-intensity exercise. Cold-water immersion (CWI) and body cryotherapy (BC) are widely used recovery methods, but their relative effectiveness remains unclear. To systematically compare the effects of CWI and BC on alleviating DOMS and restoring jump performance, with implications for optimizing post-exercise recovery and reducing muscle damage.

Methods:

This systematic review and meta-analysis included 13 randomized controlled trials with a total of 214 participants aged 20–47.2 years. Databases searched included PubMed, Excerpta Medica Database (EMBASE), Cochrane Central Register of Controlled Trials, Web of Science, covering the period from January 1, 2002 to December 10, 2024. The primary outcomes were the alleviation of DOMS and the recovery of jump performance. Data were analyzed using Review Manager 5.3, with standardized mean difference or mean difference used as effect size metrics.

Results:

The analysis showed that CWI was significantly more effective than BC in alleviating DOMS, particularly within 24 hours (mean difference = 1.07; 95% confidence interval [CI]: 0.70–1.43; P < .00001). However, this difference was not statistically significant at 48 hours (mean difference = 0.47; 95% CI: −0.04 to 0.98; P = .07). In terms of jump performance recovery, BC showed a small and fragile effect on 24-hour jump performance (standardized mean difference = 0.52; 95% CI: 0.03–1.02; P = .04), while no significant differences were observed between CWI and BC at 48 hours (standardized mean difference = 0.26; 95% CI: −0.82 to 1.33; P = .64). The analysis of DOMS showed low heterogeneity, whereas the jump performance recovery analysis exhibited high heterogeneity.

Conclusion:

CWI appears more effective than BC for short-term DOMS relief. BC may offer a modest, short-term advantage in jump performance recovery, though this finding is limited by heterogeneity and a marginal effect size. Both methods support post-exercise recovery by accelerating the restoration of muscle function and reducing soreness, which can benefit athletes and physically active individuals globally.

1. Introduction

Recovery after exercise is a critical topic in sports science. Athletes, especially after high-intensity exercise, often face muscle damage and functional fatigue, such as delayed onset muscle soreness (DOMS) and reduced athletic performance (e.g., diminished jump ability).^[1,2] These issues directly affect athletes’ training outcomes and competitive performance. DOMS typically arises after high-intensity or unaccustomed exercise and manifests as muscle soreness, swelling, and limited range of motion.^[3–5] Similarly, the decline in jump performance is often a hallmark of muscle fatigue and damage.^[6,7] To mitigate these adverse effects, various recovery methods have been proposed, with cryotherapy – including body cryotherapy (BC) and cold-water immersion (CWI) – being common and widely used approaches.^[8,9]

CWI involves submerging the body in cold water, leveraging water’s thermal conductivity to lower body temperature, thereby reducing inflammation, relieving pain, and accelerating recovery.^[10] Studies have shown that CWI can significantly alleviate DOMS and improve athletic performance, particularly within 48 hours post-exercise.^[11] On the other hand, BC exposes the body to extremely low air temperatures (typically ranging from −110 to −140°C) to reduce muscle temperature and inflammation,^[12] thereby speeding up recovery.^[13] In recent years, BC has gained wide adoption in the sports field and has demonstrated potential in reducing muscle damage markers and enhancing muscle functional recovery.^[14]

The mechanism of BC primarily involves vasoconstriction induced by low temperatures, which reduces blood flow and the migration of inflammatory factors, thus alleviating muscle pain and swelling.^[15] Although the application methods of BC and CWI differ, their shared goal is to trigger physiological responses by lowering body temperature, such as reducing muscle soreness, accelerating recovery, and decreasing the accumulation of metabolic byproducts.^[16,17] whole-body cryotherapy (WBC) generates a stronger temperature gradient compared to CWI, but the poor thermal conductivity of air limits its cooling effect on deeper tissues. In contrast, CWI effectively cools the skin and muscle tissues through the marginally greater thermal conductivity of water, making it more effective in reducing superficial muscle inflammation and soreness. Despite their differences in mechanisms, both have demonstrated positive effects on promoting recovery, reducing DOMS, and restoring athletic performance. The 24- and 48-hour post-exercise time points were selected because they represent the typical peak and early resolution phases of exercise-induced muscle damage and functional impairment.^[18] These windows are highly relevant to athletic training cycles, where transient recovery within 1 to 2 days is often essential for maintaining performance in subsequent sessions or competitions.^[19] Jump performance (e.g., countermovement jump, SJ) is a well-validated,^[20] functional measure of neuromuscular readiness and lower-body power, frequently used in both research and applied sports settings to assess recovery status.^[21]

Currently, extensive research has explored the application of BC in exercise recovery, particularly the effects of CWI and WBC on alleviating DOMS and improving jump performance.^[22,23] However, existing findings are inconsistent. Some studies highlight the significant efficacy of WBC in reducing muscle damage and improving functional recovery,^[24] while others suggest that CWI yields better recovery outcomes.^[25] Although previous studies have indicated that both WBC and CWI improve DOMS, most studies are limited by small sample sizes, heterogeneous study designs, and a lack of systematic comparative analyses. Therefore, conducting a systematic meta-analysis based on existing data to compare the efficacy of these 2 methods in alleviating DOMS and restoring jump performance is essential. Physiologically, CWI relies on water’s 25-fold higher thermal conductivity compared to air to transiently cool superficial muscle compartments, thereby reducing nerve conduction velocity and the release of inflammatory mediators.^[26,27] BC, in contrast, generates a −110 to −140°C skin temperature gradient that evokes strong cutaneous vasoconstriction and a subsequent “rebound” hyperemia; yet its limited penetration depth (<2 mm) may spare deeper muscle layers.^[13,28] These mechanistic differences predict that CWI should outperform BC in DOMS alleviation (greater cooling dose), whereas BC might transiently restore neuromuscular output via faster cutaneous nociceptor inhibition. However, empirical findings are conflicting: an RCT favor CWI,^[29] whereas 2 report marginally greater efficacy of BC. Conversely, chronic post-exercise cooling has been shown to attenuate mitochondrial biogenesis and long-term performance gains,^[30] underscoring the need to restrict BC/CWI to short-term recovery only. The unresolved contradiction provides the theoretical impetus for the present systematic comparison. This analysis aims to synthesize existing evidence to inform the selection of recovery modalities for athletes and active individuals. Therefore, this systematic review and meta-analysis aim to directly compare the efficacy of CWI and BC in alleviating DOMS and restoring jump performance at 24 and 48 hours post-exercise. We hypothesize that CWI will be more effective for reducing DOMS in the first 24 hours, while BC will be marginally greater for the immediate recovery of jump performance.

2. Materials and methods

2.1. Literature search

The literature search was conducted independently by 2 researchers. Search terms included: CWI, cold-water therapy, cryotherapy, whole-BC, partial-BC, DOMS, and jump performance (e.g., countermovement jump or SJ). The databases searched included PubMed, Excerpta Medica Database (EMBASE), Cochrane Central Register of Controlled Trials, and Web of Science. The search period was from January 1, 2002 to June 15, 2024, with the final search conducted on December 10, 2024. The search period commenced in 2002 because this year marks the emergence of the first robust, modern RCTs investigating CWI and BC as structured recovery modalities in athletic populations. For example, the PubMed search strategy used the following query:(#1 Cold-water immersion [Title/Abstract] OR #2 Cold-water therapy [Title/Abstract] OR #3 Cryotherapy [Title/Abstract] OR #4 Whole-body cryotherapy [Title/Abstract] OR #5 Partial-body cryotherapy [Title/Abstract]) AND (#6 Delayed onset muscle soreness [Title/Abstract] OR #7 Jump performance [Title/Abstract]) AND (#8 Randomized controlled trials [Title/Abstract]). The literature screening and data extraction were conducted independently by 2 researchers (S.L. and Y.M.) using a double-blind approach. In cases of disagreement, a third researcher resolved the issue through discussion.

Articles retrieved from each database were imported into the EndNote X9 reference management software for initial screening and inclusion based on the PRISMA guidelines.

2.2. Selection criteria

2.2.1. Inclusion criteria

Healthy adults (≥18 years) engaging in exercise designed to induce muscle soreness or fatigue. BC, involving exposure to very cold air (≤−100°C) for 2 to 4 minutes. CWI, involving immersion in cold water (≤15°C) for 5 to 15 minutes. The primary outcomes were muscle soreness (DOMS), assessed using a visual analog scale or a numeric rating scale, and jump performance (e.g., countermovement jump height). Outcomes were restricted to the first 48 hours post-intervention to align with the acute recovery phase most critical for athletic readiness and training continuity. Randomized controlled trials (RCTs) and language: studies written in English.

2.2.2. Exclusion criteria

Combined interventions that may interfere with the effects of the primary intervention. Training programs involving long-term interventions. Studies lacking complete data or data presented only in formats such as conference abstracts or short papers, which cannot be effectively analyzed. Non-RCTs, review articles, conference papers, study protocols, or correspondence articles.

2.3. Literature data extraction

The literature screening and data extraction were conducted independently by 2 researchers using a double-blind approach. Initially, duplicate articles, non-RCT studies, review articles, conference papers, study protocols, and correspondence articles were excluded. By reading the abstracts, studies meeting the inclusion criteria were identified, and studies not meeting the criteria were excluded. Finally, the 2 researchers reviewed the full texts of the remaining studies to confirm eligibility. In cases of disagreement, a third researcher resolved the issue through discussion. During the data extraction process, researchers extracted and recorded relevant metrics from the included studies, including: author and year of publication, country of the study, participant characteristics, interventions, outcome measures (see Table 1 for details).

Table 1.

Basic characteristics of included studies.

Authors and year	Country	Subjects	Age	Subject type	Interventions	Outcomes
Hohenauer et al, 2020	Switzerland	PBC: 10 CWI: 10 CON: 8	BC: 22.4 ± 3.0 CWI: 21.9 ± 2.0 CON: 23.3 ± 2.6	Healthy amateur female trainees, no resistance training history	BC: Exposure to −60°C for 30 s, −135°C for 2 min CWI: Immersion in 10°C water for 10 min CON: Resting for 10 min	DOMS, VJP
Guilhem et al, 2013	Thailand	CRYO: 16 CONT: 16	BC: 21.31 ± 1.03 CON: 21.31 ± 1.03	Healthy males, no musculoskeletal or nervous system disorders	BC: −30°C air pulse cryotherapy, 4 times/d, 5 min each, for 5 d CONT: Resting, no cryotherapy	DOMS
Hohenauer et al, 2018	Switzerland	CWI: 9 PBC: 10	CWI: 26.0 ± 4.3 BC: 25.8 ± 4.5	Healthy males, regular moderate-intensity endurance sports	CWI: Immersion in 10°C water for 10 min BC: Exposure to −60°C for 30 s, −135°C for 2 min	DOMS, VJP
Haq et al, 2022	UK	WBC1: 9 WBC4: 8 CWI: 8 CON: 8	BC1: 35.3 ± 14.9 BC4: 27.8 ± 5.9 CWI: 38.1 ± 12.5 CON: 47.2 ± 12.0	Healthy males, regular sports participants	BC1: −120°C cryotherapy for 3 min, 1 h later BC4: −120°C cryotherapy for 3 min, 4 h later CWI: Immersion in 15°C water for 10 min CON: Resting	DOMS
Lima et al, 2017	Brazil	BC: 19	BC: 21.6 ± 2.0	Healthy females, no systematic strength training or upper-limb activity history	BC: 20 min/session, 2 times/d, for 4 d, 6–7 h intervals CON: Other arm without cryotherapy	DOMS
Ferreira et al, 2014	Brazil	PBC: 13 CON: 13	BC: 20.2 ± 2.5 CON: 20.3 ± 2.2	Healthy males, regular moderate-intensity physical activity	BC: −110°C cryotherapy for 3 min CON: Exposure to 21°C for 3 min	DOMS
Wilson et al, 2017	UK	WBC: 10 CWI: 11 Placebo: 10	BC: 37.7 ± 8.9 CWI: 41.3 ± 7.6 Placebo: 40.6 ± 7.2	Trained male endurance runners (marathon)	BC: −85°C cryotherapy for 3 min, rest 15 min, then 4 min CWI: Immersion in 8°C water for 10 min Placebo: Simulated cherry juice intake (no antioxidants)	DOMS
Rose et al, 2014	Australia	WBC: 5 CWI: 4 CON: 4	BC: 26 CWI: 26 CON: 26	Healthy males, moderate resistance trainers	BC: Exposure to −140°C for 3 min CWI: Immersion in 11°C water for 3 min CON: Resting at room temperature for 3 min	DOMS
Hausswirth et al, 2011	France	WBC: 9 CON: 9	BC: 31.3 ± 4.1 CON: 30.3 ± 5.1	High-level endurance runners	BC: 3 whole-body cryotherapy sessions, 3 min each, at −110°C CON: Passive rest	DOMS
Costello et al, 2011	France	WBC: 18 CON: 18	BC: 20.8 ± 1.2 CON: 20.8 ± 1.2	Healthy university students	BC: Whole-body cryotherapy (−110°C), 3 min each CON: Room temperature (15°C), 3 min each	DOMS
Fonda et al, 2013	Slovenia	WBC: 11	BC: 26.9 ± 3.8	Healthy males, regular moderate-intensity physical activity	WBC: −140 to −195°C, 3 min/d, for 6 d	DOMS, SJ, CMJ
Oakley et al, 2013	USA	BC: 23 CON: 10	BC: 24.0 ± 4.0	Healthy individuals (including trainees and non-trainees)	BC: Ice application to the right leg, 20 min/session, 3 times/d, for 72 h CON: No intervention	DOMS
Abaïdia et al, 2016	France	CWI: 10 WBC: 10	23.4 ± 4	Physically active males, no recent leg injuries	CWI: Immersion in 10°C water for 10 min BC: −110°C whole-body cryotherapy for 3 min, with a 2-wk interval between experiments	DOMS, CMJ

BC body cryotherapy, CMJ = counter-movement jump, CWI = cold-water immersion, DOMS = delayed onset muscle soreness, SJ = squat jump, VJP = vertical jump height, WBC = whole-body cryotherapy

2.4. Literature quality assessment

The risk of bias in the included studies was independently assessed by 2 researchers using the Cochrane Handbook for Systematic Reviews of Interventions (5.1.0) tool. Due to the nature of cryotherapy interventions, participant blinding was not feasible. However, outcome assessors were blinded where possible. Six bias categories were evaluated using the Cochrane Risk of Bias tool: random sequence generation, allocation concealment, blinding of outcome assessors, incomplete outcome data, selective reporting, and other biases.

2.5. Data analysis and processing

Meta-analysis was conducted using Review Manager 5.3, employing standardized mean difference (SMD) as the effect size due to variations in scoring scales across studies.^[31] Effect sizes were interpreted based on Cohen’s criteria, with SMD < 0.2 indicating a small effect, 0.2 SMD < 0.5 a small effect, 0.5 SMD < 0.8 a moderate effect, and SMD ≥ 0.8 a large effect.^[32] Heterogeneity was assessed using I², applying a fixed-effect model for I² < 50% and a random-effect model for I² > 50%.^[33] Sensitivity analysis was conducted to identify heterogeneity sources, and publication bias was evaluated using a funnel plot.^[34] The intervention protocols for CWI and BC varied across studies. CWI temperatures ranged from 8 to 15°C, with durations of 3 to 10 minutes. BC temperatures ranged from −30 to −195°C with exposures of 2 to 5 min/session. The timing of interventions post-exercise and frequency of application are detailed in Table 1.

3. Results

3.1. Literature search results

A total of 1763 articles were identified through database searches and other sources. After removing duplicates, 561 articles remained. Based on predefined inclusion and exclusion criteria, 173 articles were selected after an initial screening of titles and abstracts that were clearly irrelevant. Further exclusions were made for review articles, non-RCTs, non-human studies, incomplete data, poor-quality studies, and inconsistent outcome measures, leaving 36 articles after full-text review. Finally, 13 articles were included in the meta-analysis. The study selection process is shown in Figure 1.

Figure 1.

Flow diagram of study selection. WOS = Web of Science.

3.2. Basic information on the inclusion of the literature

Thirteen RCTs were included in the meta-analysis based on strict inclusion and exclusion criteria. The participants, aged 20 to 47.2 years, included 214 individuals in total (see Table 2 for details). The subjects comprised both healthy individuals and athletes. Interventions included BC, CWI, and passive recovery or no intervention as controls. BC: temperatures ranged from −110 to −195°C, for 3 minutes, conducted daily or on alternate days. CWI: water temperatures ranged from 8 to 15°C, lasting 3 to 10 minutes. LocalBC: Included ice packs or air pulses at temperatures of −30 to −60°C, lasting 5 to 20 minutes. Control groups: Included passive recovery or exposure to room temperature. The duration of each intervention ranged from 3 to 20 minutes, with frequencies of 1 to 4 sessions/d, over an intervention period of 1 to 6 days. Outcome measures included DOMS and athletic performance metrics, such as vertical jump height (VJP), countermovement jump (CMJ), and squat jump (SJ).

Table 2.

Literature quality evaluation form of Cochrane.

	A	B	C	D	E	F	G	Total scores
Wilson et al[35]	1	0	0	0	1	1	1	4
Hohenauer et al[36]	1	1	0	0	0	1	1	4
Guilhem et al[37]	1	1	0	0	1	1	1	5
Hohenauer et al[38]	1	1	0	0	1	1	1	5
Haq et al[39]	1	1	0	0	1	1	1	5
Lima et al[40]	1	0	0	0	0	1	1	3
Ferreira et al[41]	1	1	0	0	0	1	1	4
Rose et al[14]	1	0	0	0	0	1	1	3
Hausswirth et al[42]	1	1	0	0	0	1	1	4
Costello et al[43]	1	0	0	0	0	1	1	3
Fonda and Sarabon[44]	1	1	0	0	0	1	1	4
Oakley et al[45]	1	1	0	0	0	1	1	4
Abaïdia et al[46]	1	0	0	0	1	1	1	4

No selective reporting or other potential sources of bias were identified in any of the studies (low risk). Among the included studies, 3 were classified as high-quality, while the remaining 10 were of moderate quality. Detailed data are presented in this table, with the results of the bias assessment illustrated in Figures 4 and 5.

3.3. Literature quality evaluation results

The quality evaluation of the included literature revealed that all 13 studies were RCTs (Table 2). For random sequence generation, 8 studies employed methods such as digital tools, lottery, or block randomization (low risk), while the remaining 5 studies did not specify the randomization process (unclear risk). None of the studies reported allocation concealment (unclear risk), nor did they provide a clear description of blinding of assessors (unclear risk). Regarding outcome data completeness, 5 studies provided complete data (low risk), whereas 8 studies had incomplete data (high risk).

4. Results

4.1. The effectiveness of BC in alleviating DOMS

Data from 20 included studies. The results indicated an overall effect size of mean difference (MD) = −0.82 (95% confidence interval [CI]: −1.03 to −0.60; Z = 7.45; P < .00001), demonstrating a significant benefit of the experimental group in reducing DOMS (Fig. 2). Subgroup analysis revealed an effect size of MD = −0.74 (95% CI: −1.04 to −0.44; Z = 4.81; P < .00001, I² = 16%) within 24 hours and MD = −0.90 (95% CI: −1.20 to −0.59; Z = 5.74; P < .00001, I² = 19%) within 48 hours. These findings indicate that the experimental group outperformed the control group at both time points, with a more pronounced effect at 48 hours. The low heterogeneity (I² < 20%) suggests strong consistency across the studies.

Figure 2.

Body cryotherapy versus control group on delayed onset muscle soreness: meta-analysis. BC = body cryotherapy, CI = confidence interval, SD = standard deviation.

4.2. Comparison of BC and CWI in reducing DOMS

Data from 11 studies were included to compare the effects of BC and CWI. The results showed an effect size of MD = 0.86 (95% CI: 0.57–1.16; Z = 5.70; P < .00001), indicating that CWI (n = 98) was more effective than BC (n = 99) in alleviating DOMS (Fig. 3). When analyzed across all time points, CWI demonstrated a slight advantage over BC in reducing DOMS, and the results were robust. The heterogeneity test revealed no heterogeneity among the studies (I² = 0%, P = .61), and a fixed-effect model was used for analysis.

Figure 3.

Cold water immersion versus cryotherapy on delayed onset muscle soreness: meta-analysis. BC = body cryotherapy, CI = confidence interval, CWI = cold-water immersion, SD = standard deviation.

Subgroup analysis by post-intervention time showed that within 24 hours, the effect size was MD = 1.07 (95% CI: 0.70–1.43; Z = 5.72; P < .00001), indicating that CWI was significantly more effective than BC in reducing DOMS (as higher values indicate greater pain reduction). The results were statistically significant, with no heterogeneity (I² = 0%), and study outcomes were consistent. Within 48 hours, the effect size was MD = 0.47 (95% CI: −0.04 to 0.98; Z = 1.81; P = .07), showing no statistically significant difference, with low heterogeneity (I² = 4%) and consistent results across studies.

4.3. Comparison of BC and CWI in restoring jump performance

A total of 7 studies were included for analysis. The results showed an effect size of SMD = 0.39 (95% CI: −0.18 to 0.96; Z = 1.35; P = .18), indicating no significant difference between the BC and CWI groups in improving jump height (Fig. 4). Heterogeneity testing revealed substantial heterogeneity among the studies (I² = 72%, P = .0008), and a random-effects model was applied for analysis. Given the high heterogeneity (I² = 72%), we conducted sensitivity analysis by sequentially removing each study. The exclusion of Wilson et al reduced I² to 58%, suggesting it contributed notably to heterogeneity.^[35]

Figure 4.

Cold water immersion versus body cryotherapy on jump: meta-analysis. BC = body cryotherapy, CI = confidence interval, CWI = cold-water immersion, SD = standard deviation.

Subgroup analysis based on post-intervention time showed that within 24 hours, the effect size was SMD = 0.52 (95% CI: 0.03–1.02; Z = 2.06; P = .04), suggesting that BC was significantly more effective than CWI in restoring jump performance, with statistically significant results and low heterogeneity (I² = 28%). However, within 48 hours, the effect size was SMD = 0.26 (95% CI: −0.82 to 1.33; Z = 0.47; P = .64), indicating no significant advantage of BC over CWI in jump performance recovery at this time point. The heterogeneity was high (I² = 83%), suggesting reduced result stability. CWI is more effective than for BC at 24 hours was statistically significant, the modest SMD (0.52) and the lower confidence interval limit near 0 (0.03) suggest that the clinical relevance of this difference may be limited.

4.4. Publication bias test

The funnel plots from the meta-analyses comparing the effects of BC and CWI on DOMS and jump performance showed that most scatter points were symmetrically distributed and shifted upwards, indicating no significant publication bias among the included studies. These findings suggest that the results of the analyses are reliable (Figs. 5 and 6).

Figure 5.

Funnel plots of pain degree.

Figure 6.

Funnel plots of Qswestry disability.

5. Discussion

This study compared the effects of BC and CWI on DOMS and jump performance recovery. A total of 11 RCTs involving 197 participants were included, examining 2 distinct interventions and a control group. The temperature range for BC was −30 to −195°C, while CWI therapy utilized temperatures between 5 and 15°C. The findings indicated that CWI was significantly more effective than BC in alleviating DOMS, while BC demonstrated marginally greater performance in restoring jump capacity within the first 24 hours.

5.1. CWI vs BC in alleviating DOMS

This study found that CWI outperformed BC in reducing DOMS. Within 24 hours post-intervention, CWI showed a significantly greater effect (MD = 1.07; 95% CI: 0.70–1.43; P < .00001). This aligns with findings from Hohenauer et al, which demonstrated that CWI significantly alleviates muscle soreness in the short term.^[38] CWI facilitates recovery by promoting alternating vasoconstriction and vasodilation, enhancing the clearance of metabolic waste and reducing localized inflammation. The improved blood and lymphatic flow contributes to the removal of waste products and the reduction of inflammation. Guilhem et al noted that CWI aids in lactic acid clearance, a key factor in DOMS, potentially alleviating exercise-induced muscle soreness more effectively.^[37]

In contrast, BC operates via extreme localized cooling (typically below −30°C), which suppresses blood flow and metabolic activity to reduce inflammation and swelling.^[40] BC typically uses gaseous or localized cooling techniques that induce short-term vasoconstriction and reduce local blood flow, mitigating the spread of inflammation.^[39] However, this blood flow suppression may limit BC’s ability to promote waste clearance and circulation improvement in the short term, making it less effective than CWI in these aspects.

At 48 hours post-intervention, the difference between CWI and BC diminished and was no longer significant (MD = 0.47; 95% CI: −0.04 to 0.98; P = .07). This suggests that the short-term effects of CWI are more pronounced, whereas BC may lack sustained efficacy over a longer duration. Similar findings were reported by Hohenauer et al, who noted that the initial advantages of CWI over BC in DOMS recovery diminish over time.^[36] These results reinforce the idea that CWI excels in early recovery phases, likely due to its ability to enhance circulation and reduce muscle tension.

Moreover, Ferreira-Junior et al found that BC could produce sustained repair effects on deeper muscle tissues at 48 hours post-intervention.^[41] This indicates that BC may hold an advantage for deeper muscle recovery, particularly for prolonged recovery periods, by effectively controlling deep muscle inflammation and swelling. In contrast, the marginally greater short-term benefits of CWI are attributed to its strong impact on localized blood flow, which helps alleviate soreness in superficial muscles.

The mechanisms underlying CWI and BC fundamentally differ. CWI’s effects are primarily driven by temperature-induced vascular changes, promoting metabolic waste clearance and reducing inflammatory responses.^[47] It directly impacts the muscle area through immersion in water at controlled temperatures for specific durations. BC, on the other hand, focuses on transient temperature reduction to suppress blood flow and metabolic activity, thereby mitigating localized inflammation.^[42] This distinction explains why CWI is more advantageous in the initial recovery phase, particularly in the removal of lactic acid and other metabolic byproducts. BC, while effective at reducing swelling and inflammation through transient cooling, may not directly promote circulation or waste clearance as effectively, resulting in less sustained impact over longer recovery periods compared to CWI.

5.2. CWI vs BC in restoring jump performance

Our study found that BC significantly outperformed CWI in restoring jump performance within 24 hours. The BC group achieved an effect size of SMD = 0.52 (95% CI: 0.03–1.02; P = .04), indicating faster recovery of jump ability. This aligns with findings from Costello et al, who reported that BC significantly improved jump performance in the short term, particularly following high-intensity exercise.^[43] BC’s ability to reduce neuromuscular damage through cold air exposure-induced vasoconstriction and metabolic slowing appears to facilitate accelerated neuromuscular function recovery.^[44]

The short-term advantage of BC may be closely tied to its unique mechanisms. Exposure to extremely low temperatures transiently induces localized vasoconstriction, reducing inflammation and swelling, and mitigating neuromuscular damage.^[45] In the early stages of recovery, this mechanism is particularly effective in preventing muscle function decline and restoring physical performance. Similar findings were reported by Abaïdia et al, who observed improvements in jump ability within 24 hours post-BC.^[46] However, by 48 hours post-intervention, the differences between BC and CWI were no longer significant (SMD = 0.26; 95% CI: −0.82 to 1.33; P = .64). This suggests that while BC provides transient benefits, these effects diminish over time, converging with the effects of CWI. This is consistent with Lima et al, who found that CWI’s recovery effects became more pronounced over time, whereas BC’s benefits were less sustained.^[40] The enhanced recovery observed with CWI at 48 hours may be attributed to its prolonged mechanisms of action, including buoyancy, temperature regulation, and improvements in local circulation, which help eliminate metabolic waste, reduce muscle soreness, and promote long-term muscle recovery. This shift in effectiveness could also be explained by differences in intervention methods. BC’s extreme low temperatures and short exposure times may produce transient but limited effects, while CWI offers more gradual and sustained recovery benefits. Hohenauer et al also noted that CWI facilitates long-term recovery by enhancing blood circulation and reducing local temperature, with effects becoming evident after 48 hours.^[38] This indicates that CWI’s benefits in promoting blood flow and clearing metabolic waste manifest more strongly during extended recovery periods, particularly in alleviating muscle soreness and enhancing muscle recovery. The high heterogeneity in jump performance outcomes (I² = 72%) underscores the need for cautious interpretation. Our sensitivity and subgroup analyses suggest that differences in training level and jump test protocols may contribute to variability. Future studies should standardize jump assessments and stratify participants by training status to minimize heterogeneity.

Another factor influencing these findings could be differences in participant training levels and recovery demands. Our study primarily involved healthy recreational athletes, whereas studies like those by Wilson et al and Haq et al focused on elite athletes. Athletes at different training levels may respond differently to BC interventions.^[35,39] For example, elite athletes may be more sensitive to BC’s transient recovery effects due to their heightened neuromuscular fatigue following intense training and competition, coupled with a greater urgency for recovery.

5.3. Research limitations and future prospects

This study has several limitations that should be noted. First, the relatively small sample size and the predominant focus on recreational athletes restrict the generalizability and external validity of the findings, particularly for elite or professional athletic populations. Second, there is notable heterogeneity in the intervention protocols across the included studies, such as differences in temperature, duration, and frequency of BC and CWI, which may have impacted the consistency of the results. Third, this study primarily focuses on short-term recovery effects within 48 hours post-intervention, with limited exploration of the long-term efficacy and sustainability of these recovery modalities. Additionally, the lack of investigation into the integration of other recovery strategies, such as compression garments, nutritional interventions, or active recovery, represents a missed opportunity to assess the comprehensive and synergistic effects of combined approaches.

Future research should address these limitations by expanding the sample population to include a broader range of participants, encompassing individuals with diverse athletic levels, ages, and physiological characteristics, thereby improving the generalizability of the findings. Standardizing intervention protocols, particularly concerning temperature, exposure duration, and frequency, is essential to reduce methodological heterogeneity and enhance comparability between studies. Longitudinal studies should extend beyond the immediate recovery period to examine the long-term effects of BC and CWI on sustained recovery and athletic performance. Furthermore, mechanistic studies are needed to explore the specific physiological processes underlying these interventions, such as their roles in inflammation regulation, metabolic waste clearance, and neuromuscular restoration. By addressing these gaps, future research can provide more robust, evidence-based guidance for the effective and tailored application of these recovery modalities in both recreational and elite sports settings.

6. Conclusion

CWI is more effective than BC in reducing DOMS within 24 hours. BC may provide a short-term advantage in jump performance recovery, although this result should be interpreted with caution due to modest effect sizes and heterogeneity. By 48 hours, the effects of both methods converge. Future research should optimize parameters and explore long-term effects for practical applications.

Acknowledgments

We extend our gratitude to Tsinghua University for providing the essential academic environment that greatly supported our research. Additionally, we sincerely thank the research institutions and students who contributed their assistance.

Author contributions

Conceptualization: Shuang Liu, Yifei Ma.

Data curation: Shuang Liu.

Formal analysis: Shuang Liu.

Writing – original draft: Shuang Liu.

Writing – review & editing: Shuang Liu.