REF-387 · Hill, Howatson, van Someren et al., 2014 / Brown et al., 2017 · British Journal of Sports Medicine · Paper
Compression Garments and Recovery from Exercise-Induced Muscle Damage: two meta-analyses (Hill 2014; Brown 2017)
Sample (pooled)
Hill 2014: 12 studies, outcomes measured at baseline and at 24, 48 or 72 h post-exercise
Brown 2017: 23 peer-reviewed studies on healthy participants; change scores after resistance, running, or non-load-bearing endurance ("metabolic") exercise; moderators: time (0-2, 2-8, 24, >24 h), pressure (<15 vs ≥15 mmHg), training status
Pooled participant number is not given in either abstract
Key Findings
Hill 2014 (damaging exercise): moderate effects of compression on DOMS (Hedges' g = 0.403, 95% CI 0.236-0.569), muscle strength (g = 0.462, 0.221-0.703), muscle power (g = 0.487, 0.267-0.707) and CK (g = 0.439, 0.171-0.706), all p < 0.001 — compression garments are effective in enhancing recovery from muscle damage
Brown 2017 (all exercise types): small, very likely overall benefit, ES = 0.38 (95% CI 0.25-0.51), p < 0.001 — not influenced by garment pressure (p = 0.06) or training status (p = 0.64)
Strength recovery benefited most: ES = 0.62 (0.39-0.84); large at 2-8 h (ES = 1.14) and >24 h (ES = 1.03)
By exercise type: largest after resistance exercise, ES = 0.49 (0.37-0.61), and at >24 h after it, ES = 1.33 (0.80-1.85); after metabolic exercise the effect was significant (p = 0.01), with a large benefit only for cycling performance at 24 h post-exercise (ES = 1.05, 0.25-1.85)
Authors' recommendation: compression garments for strength recovery from 2-8 h and >24 h, most effective after resistance exercise, and to enhance next-day cycling performance
Mechanism
Not from this paper: the usual explanation is that external compression enhances venous return and limits oedema, with a sensory component reducing perceived soreness — mechanisms were not tested in either meta-analysis. Brown 2017 found no pressure dependence (p = 0.06), which argues against a simple hydrostatic dose-response.
Practical Application
Two meta-analyses: after damaging exercise, compression garments moderately improved soreness, strength, power and CK recovery (Hill 2014, 12 studies, Hedges' g 0.403-0.487); across 23 studies of all exercise types the overall benefit was small (ES 0.38) but strength recovery gained most (ES 0.62), the largest effects came after resistance exercise and beyond 24 h, and next-day cycling performance improved (ES 1.05 at 24 h).Wear them after the hard session and into the next day — the effect shows up hours to a day later.
Post-race / post-hard-session: put compression on afterwards and keep it on into the next day — the strongest windows in Brown 2017 were 2-8 h and >24 h
Garment pressure did not change the effect (p = 0.06) and neither did training status (p = 0.64) — a comfortable fit is fine; do not chase a mmHg number
Not from this paper; the base's own practice: compression during travel and in congested blocks or multi-stage events — low cost, low downside
Inference, not tested: unlike cold-water immersion (REF-011), no meta-analysis shows compression blunting training adaptation, but neither meta-analysis measured adaptation — "does not block adaptation" is untested, not proven
Limitations
Heterogeneous garment pressures and protocols; Brown 2017 notes the pressure and training-status questions are limited by paucity of reported data
Mostly resistance and running protocols; cycling evidence is one 24 h time point
Subjective outcomes (DOMS) dominate Hill 2014; Brown 2017 is performance-based
How this entry was checked and changed
Content correction 2026-09-27
"23 studies (Hill 2014) ... n≈1,100 pooled" -> Hill 2014 pooled 12 studies, Brown 2017 23; no pooled n in either abstract. Corrected.
"MVC recovery SMD ≈ +0.30" -> Hill: g = 0.462; Brown: ES = 0.62. Corrected. "Swelling/limb girth reduced", "perceived tightness consistently reduced", "continuously 24-48 h post-exercise", "~20-30 mmHg", "12-24 h post", "15-40 mmHg range" -> not in either abstract; removed.
"Does NOT block adaptation (unlike CWI) — safe for regular use" -> untested in these meta-analyses; relabelled as inference.
Sport "cycling, running, swimming" -> strength, running, cycling (the exercise modes Brown 2017 pooled).