Keywords: BFR training, kaatsu, Blood flow restriction training, physical therapy, muscle strength
Why is this study important?
Blood flow restriction (BFR) training has been widely adopted by rehabilitation providers since 2017, when a meta-analysis in the British Journal of Sports Medicine first showed it could produce greater strength gains than standard low-load exercise in clinical musculoskeletal populations (Hughes et al., 2017). That evidence, however, rested on trials with methodological limitations, populations that didn’t always reflect typical clinical caseloads, and no formal analysis of pain, disability, or adverse events. This paper (Ferlito et al., 2026) addresses those gaps directly because it draws on larger, higher-quality randomized-controlled trials and expands the outcome measures to include pain and disability alongside strength, while also systematically tracking adverse responses. The result is a much sturdier evidence base that lets clinicians justify BFR in a plan of care not just on strength gains, but on the functional and safety outcomes that matter most in day-to-day practice.
How did the study go about this?
Forty-five randomized controlled trials were included. Each one had to enroll patients from clinical musculoskeletal populations, use low-load BFR exercise (bodyweight up to 50% 1RM), and compare it against either low-load exercise alone or high-load exercise (70% 1RM or higher) without BFR, over at least 8 sessions. We required trials to use clearly defined sets, reps, and quantifiable loading so relative intensities could be compared cleanly across groups. This meant excluding studies using resistance bands, a popular approach in practice, but one that’s harder to standardize against a percentage-based loading scheme. Any BFR trial meeting these criteria was included regardless of what pressure prescription it used, since part of our goal was understanding whether pressure approach itself changes outcomes. From each trial we pulled every reported measure of strength, pain, disability, and adverse events.
To judge how trustworthy each trial’s results were, we scored them with the Risk of Bias 2 and PEDro tools, which flag design issues like inadequate blinding or selective reporting, and then rated our overall confidence in the pooled evidence using GRADE (very low, low, moderate, high certainty). Studies with more methodological flags received lower overall GRADE scores whereas groups of studies that had less issues received higher GRADE scores. Because studies measured outcomes in different ways, we converted results to a common scale wherever possible, and because some trials contributed multiple results (say, both strength and pain data), we used statistical methods built to handle that lack of independence rather than treating every number as a separate, unrelated data point. We also tested for publication bias, checking whether negative or null findings might be missing from the published record in a way that would make BFR appear more effective than it really is and subgroups to find whether the ways in which BFR was applied impacted the observed outcomes. Together, this gives readers a rigorous, transparent basis for judging exactly how BFR performs against low- and high-load exercise.
What did the study find?
For our primary outcome, muscle strength, low-certainty evidence (drawn from 22 studies) showed that low-load BFR exercise was superior to low-load exercise alone. Contrary to the 2017 meta-analysis (Hughes et al., 2017), we found no meaningful difference between low-load BFR and high-load exercise, a finding supported by moderate-certainty evidence from 18 studies.
For our secondary outcomes, low-load BFR consistently outperformed low-load exercise alone on both pain and disability, with no signal of increased adverse events, though the adverse-event data remain too sparse for confident comparisons. Against high-load exercise, BFR produced similar disability outcomes and a slight edge in pain relief, but these differences were small and unlikely to be clinically meaningful. As with the other comparisons, low-load BFR did not appear to increase adverse events relative to high-load exercise, though again, the underlying data are thin.
We also investigated factors that could help explain how the observed benefits might differ depending on how BFR is applied by examining whether outcomes varied by body region (upper vs. lower limb), pressure prescription method (individualized vs. fixed), strength testing approach (specific vs. non-specific), load retesting during the program, and intervention duration. None of these factors significantly altered BFR’s effectiveness compared with either low-load or high-load exercise, suggesting the benefits we observed are consistent across common variations in how BFR is applied in clinical practice.
Overall, certainty in these findings ranged from very low to moderate. While this represents real progress beyond the 2017 review, the field still needs more consistent, standardized trials before these results can be considered settled science.
What are the key take-home points?
For patients who are load- or pain-compromised and unable to tolerate high loads (Rolnick et al., 2021), the current evidence suggests that low-load exercise paired with BFR is a suitable alternative that replicates many of the benefits of high-load exercise, including muscle strength, pain, and disability outcomes, with similar rates of adverse events (though evidence here remains especially sparse).
Conversely, applying BFR to the low loads typically used in clinical musculoskeletal rehabilitation appears to offer superior effects on muscle strength, pain, and disability compared with low-load exercise alone, although the strength of these recommendations varied, likely due to differences in populations, outcome measures, and BFR methodologies across the included trials.
The biggest take-home point, however, is this: contrary to the pioneering 2017 meta-analysis (Hughes et al., 2017), a more rigorous analytical approach applied to a larger sample of clinical musculoskeletal populations shows no clear difference in muscle strength between low-load BFR and high-load exercise over a training period. This challenges the prevailing paradigm that BFR is merely a “bridge” toward higher-load training, one assumed to offer comparable hypertrophy but inferior strength gains.
Taken together, these findings show that despite the wide variety of ways BFR is applied in practice, low-load BFR resistance exercise produces superior effects on many of the outcomes rehabilitation providers track most closely, compared with conventional low-load training, while matching the gold standard of high-load resistance exercise.
Authors Nicholas Rolnick DPT, MS, CSCS
GenAI Statement: The initial draft and preparation of this blog post were completed entirely by the author without the use of generative AI. Generative AI tools (Claude Ver. 1.4) were utilized strictly during the post-drafting phase to optimize narrative flow, refine structural editing, and enhance readability for a non-scientific audience.
References
Ferlito, J. V., Rolnick, N., Kamis, O., de Queiros, V., Lopez, P., & Hughes, L. (2026). Do blood flow restriction exercises offer additional benefits when compared to conventional exercises in musculoskeletal rehabilitation? A systematic review and meta-analysis. British Journal of Sports. https://doi.org/10.1136/bjsports-2025-111452
Hughes, L., Paton, B., Rosenblatt, B., Gissane, C., & Patterson, S. D. (2017). Blood flow restriction training in clinical musculoskeletal rehabilitation: a systematic review and meta-analysis. British Journal of Sports Medicine, 51(13), 1003–1011.
Rolnick, N., Kimbrell, K., Cerqueira, M. S., Weatherford, B., & Brandner, C. (2021). Perceived Barriers to Blood Flow Restriction Training. Frontiers in Rehabilitation Sciences, 2, 14.