Stretching flops for sore muscles; massage and contrast therapy hold up

Delayed-onset muscle soreness, the ache that arrives after intense or unaccustomed exercise, reduces muscle strength and brings pain and inflammation, which is why an entire recovery industry exists to treat it. Systematic reviews of physiotherapy treatments for soreness have been published since the 1990s, but they frequently contradict one another, leaving athletes and clinicians guessing about what genuinely helps. An umbrella review, which pools and grades the reviews themselves, is a way to rise above that noise and ask which treatments hold up across the whole evidence base at once.

The authors searched six databases from 1998 to February 2024 for systematic reviews of randomised controlled trials of any post-exercise treatment used by physiotherapists to reduce soreness in healthy adults. Twenty-nine systematic reviews qualified, covering 863 unique randomised trials and 24 distinct treatments. Each review's methodological quality was assessed with the AMSTAR-2 tool, overlap between reviews was accounted for, and an evidence map graded every treatment by effect size and strength of evidence at each follow-up window from immediately post-exercise out to 96 hours.

Several therapies showed significant pain reduction at multiple time points: contrast therapy performed well immediately after exercise, massage registered at 24 and 48 hours, and compression, kinesiotaping and cryostimulation showed effects at 48 hours, with cooling, phototherapy and heat therapy also appearing at various windows. Effect sizes ranged from small for cooling therapy (0.36) to large for heat therapy (1.82). Evidence for stretching, exercises and electrical stimulation was weak. The sobering backdrop is that 17 of the 29 reviews were of critically low methodological quality and only two were rated high.

This is a map of a large but messy literature rather than a final verdict. It confirms that hands-on and temperature-based therapies have the most consistent support for easing soreness, while undercutting the enduring popularity of stretching as a recovery tool. The dominance of low-quality reviews, plus the heterogeneity and uncertainty the authors flag, limits how far any single treatment's grade can be trusted. What is needed next is higher-quality reviews and trials with consistent outcome measures, so the rankings can firm up rather than shuffle with each new synthesis.

If you chase recovery after brutal sessions, the broad direction here is useful: massage, contrast therapy, cooling, compression, phototherapy and kinesiotaping all carry some evidence of easing soreness at different points in the days after training. Stretching, often the first thing people reach for, has weak support in this analysis, as do light exercise and electrical stimulation. Given the quality problems in the underlying reviews, treat this as a guide to where the evidence leans rather than a guarantee that any particular therapy will work for you.