Effects of Blood Flow Restriction Therapy in Individuals With Knee Osteoarthritis: Overview of Systematic Reviews

Abstract

Background:

Knee osteoarthritis (KOA) is frequently accompanied by quadriceps weakness. Resistance training is recommended, but high-intensity resistance training (HIRT) may be poorly tolerated in some patients, whereas low-intensity resistance training (LIRT) may yield smaller strength gains. Low-intensity blood flow restriction training (LIBFR) has been proposed as a low-load strategy to stimulate muscle adaptations; however, the evidence base has been summarized in multiple systematic reviews with variable conclusions.

Objective:

To synthesize and appraise the evidence on the efficacy and safety of LIBFR in individuals with KOA through an overview of systematic reviews.

Methods:

We searched MEDLINE, EMBASE, Cochrane Database, Web of Science, CINAHL, SPORTDiscus, PEDro, Epistemonikos, and gray literature (ProQuest, Global ETD) for systematic reviews of randomized clinical trials comparing LIBFR with LIRT and/or HIRT. Reviews were eligible if they evaluated supervised exercise-based LIBFR delivered ⩾2 to 3 sessions/week for ⩾4 weeks. Two reviewers independently screened, extracted data, and assessed methodological quality (AMSTAR 2). Certainty of evidence was appraised using Grading of Recommendations Assessment, Development, and Evaluation (GRADE) based on KOA-specific quantitative syntheses, applying a hierarchical approach in the presence of overlap across reviews.

Results:

Eighteen systematic reviews (57 randomized trials; 5656 participants) were included. Compared with LIRT, LIBFR was associated with greater improvements in quadriceps strength (standardized mean difference [SMD] = 0.75; 95% confidence interval [CI] = [0.40 to 1.10]) and muscle hypertrophy (SMD = 0.81; 95% CI = [0.10 to 1.52]), although certainty of evidence was very low. Compared with HIRT, LIBFR was associated with fewer adverse events (risk ratio [RR] = 0.26; 95% CI = [0.09 to 0.72]).Across comparisons, effects on pain and functional outcomes were generally similar, and between-group differences were not consistently observed.

Conclusions:

The LIBFR may improve quadriceps strength and hypertrophy relative to LIRT and may have a more favorable adverse-event profile than HIRT, but confidence in these estimates is limited by low-/very low-certainty evidence and variability in protocols and reporting. Well-designed, longer-term randomized trials with standardized LIBFR parameters and robust safety reporting are needed to clarify patient-important outcomes.

Keywords

osteoarthritis knee blood flow restriction kaatsu resistance training systematic review meta-analysis

Background

Osteoarthritis (OA) is the most prevalent form of arthritis worldwide, affecting approximately 302 million people and imposing a major public health and economic burden.¹ Knee osteoarthritis (KOA) is among the leading causes of disability in older adults.¹ Osteoarthritis is a complex, chronic, degenerative joint disease that involves not only cartilage but also multiple joint tissues, including subchondral bone, synovium, ligaments, and periarticular musculature.^1,2

Clinically, KOA is commonly characterized by pain and stiffness (often after inactivity), reduced range of motion, effusion, and muscle weakness, contributing to reduced function, participation, and quality of life (QoL).^3,4 Knee osteoarthritis is also associated with impaired balance/proprioception and increased fall risk.⁵ Risk factors include non-modifiable factors (age and female sex) and modifiable factors such as obesity, low physical activity, persistent joint overload, muscle imbalances, and reduced lower-limb muscle mass and strength.^6-8

Current management is multimodal and typically combines pharmacological and non-pharmacological strategies.⁹ In advanced cases, total knee arthroplasty may be considered when conservative care fails to provide adequate symptom relief or functional improvement.¹⁰ Due to limited long-term benefits of medication and the need to address functional limitations, guidelines increasingly emphasize non-pharmacological approaches such as education, self-management, weight loss (when appropriate), and structured exercise.^9,11,12 Progressive resistance training is strongly recommended because quadriceps weakness is a consistent clinical finding and is associated with knee pain and disability in OA.^13-15 Accordingly, resistance exercise is frequently endorsed as a first-line non-pharmacological therapy.¹⁶

High-intensity resistance training (HIRT; ~60-85% of 1-repetition maximum [1RM]) is traditionally recommended to optimize hypertrophy and strength gains in healthy populations.¹⁷ However, in KOA, higher external loads may be poorly tolerated due to pain, symptom exacerbation, or concerns regarding mechanical stress.¹⁸ Low-intensity resistance training (LIRT) is often more tolerable but typically produces smaller hypertrophy/strength gains than HIRT.^19,20 Low-intensity blood flow restriction training (LIBFR) has emerged as a potential strategy to elicit strength and hypertrophy adaptations with low external loads.¹⁸ Several systematic reviews have examined LIBFR in KOA, but conclusions remain inconsistent and/or uncertain regarding effectiveness and safety.^4,18,21-24 Therefore, an overview of systematic reviews (review of reviews) may help consolidate the evidence, clarify consistencies and discrepancies across reviews, and provide a more accessible synthesis for clinicians and stakeholders.^25,26 Overviews can serve as a pragmatic entry point to the literature by integrating findings from multiple systematic reviews into a single synthesis and supporting evidence-informed decision-making.^25,27

Objective

This overview of systematic reviews evaluates the evidence on LIBFR for KOA, comparing it to standard low-intensity and high-intensity protocols. The goal is to clarify whether LIBFR offers unique benefits or poses distinct risks regarding pain, muscle strength, hypertrophy, physical function, self-reported function, QoL, and adverse events. It also examines the methodological quality of the included reviews and assesses the overall certainty of the evidence using established frameworks.

Methods

Registration Protocol

This overview of systematic reviews adhered to recommendations provided in the Cochrane Handbook for Systematic Reviews of Interventions.²⁸ The report was prepared following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines (Supplemental Appendix 1).²⁹ The study protocol was registered in the International Prospective Register of Systematic Reviews (PROSPERO): CRD42022367209. Adherence to these standards ensured transparency, reproducibility, and methodological rigor.

Inclusion Criteria for Systematic Reviews

Types of Systematic Reviews

We searched for both Cochrane and non-Cochrane systematic reviews that evaluated randomized controlled trials (RCTs) and quasi-randomized trials (quasi-RCTs). The Quasi-RCTs were considered because, in certain contexts, the allocation process—while not strictly randomized—may still provide meaningful comparative data relevant to our research questions, in accordance with guidance from the Cochrane Handbook. Eligible systematic reviews had to present a clearly defined search strategy, explicit inclusion and exclusion criteria, and a synthesis of included studies (quantitative or qualitative). Reviews were required to provide a critical appraisal of the included studies, including risk of bias assessment. In addition, reviews that reported the certainty of evidence, for instance, by using the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) approach, were considered more informative. Narrative reviews without systematic methods, diagnostic/prognostic reviews, economic evaluations, or other non-systematic review types were excluded. If a systematic review included non-randomized studies but allowed separate extraction of RCT data, it remained eligible.

Types of Participants

This overview included systematic reviews of individuals over 40 with KOA diagnosed by clinical criteria or imaging. The Knee Osteoarthritis Outcome Score (KOOS) and Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) helped characterize symptom severity but were not the sole diagnostic methods.^30-32 Participants underwent LIBFR. Reviews covering other populations (eg, postoperative patients, athletes) were excluded unless KOA data were reported separately, ensuring greater homogeneity. This approach accounted for differing rehabilitation protocols in postsurgical or athletic groups compared with standard KOA management, thus maintaining a more consistent clinical profile within the included populations.

Types of Interventions

We considered systematic reviews evaluating LIRT (20-40% 1RM) with blood flow restriction (BFR) vs conventional low- or high-intensity exercise in KOA. Both open and closed kinetic chain exercises were included at least twice weekly for at least 4 weeks. The BFR typically involves using pneumatic cuffs on the proximal leg to partially occlude venous return, enhancing the training stimulus at lower loads.

Types of Outcomes Measured

Primary outcomes were pain, muscle strength, hypertrophy, and physical function. Pain used validated scales (visual analog scale [VAS], WOMAC Pain, KOOS Pain); strength used isometric or isokinetic dynamometry; hypertrophy used magnetic resonance imaging (MRI) or ultrasound; and function used performance-based tests (Time Up and Go [TUG], gait speed, 6-Minute Walk) plus self-reported scales (WOMAC Function, KOOS Function). Follow-ups (4-12 weeks) may limit long-term inferences, especially for hypertrophy or QoL. Secondary outcomes were intervention formats, adverse events, and QoL. Although QoL (Short Form-36 Health Survey [SF-36], 5-level EQ-5D version [EQ-5D]) is critical, it was secondary to streamline data synthesis. Results were grouped by intervention type and follow-up duration, noting short follow-ups might not reveal full improvements or harms.

Exclusion Criteria

We excluded (1) non-systematic reviews (eg, narrative reviews, expert opinions) or reviews without a reproducible search strategy; (2) reviews not primarily synthesizing RCTs/quasi-RCTs; (3) reviews in which the target population was not KOA or where KOA data could not be isolated; (4) interventions not matching LBFR (eg, BFR applied without exercise, aerobic-only BFR, or non-rehabilitative modalities); (5) interventions delivered <2 sessions/week and/or <4 weeks, because these doses are less likely to capture training-related adaptations and would increase clinical and statistical heterogeneity; (6) reviews without patient-relevant outcomes of interest (pain, strength, hypertrophy, or function); and (7) conference abstracts, protocols, and duplicate publications.

Data Sources and Research Strategy

Under the guidance of an experienced librarian (FRAS), 2 independent reviewers (FAM, GJA) conducted a comprehensive search of the following databases:

PubMed (via MEDLINE OvidSP) (searched May 15, 2023).

EMBASE (OvidSP) (searched May 17, 2023).

Cochrane Database of Systematic Reviews (Cochrane Library) (searched May 18, 2023).

Web of Science Core Collection (SCI-EXPANDED) (searched May 24, 2023).

CINAHL & Sports Discuss (via EBSCO host) (searched July 11, 2023).

PEDro (searched June 8, 2023).

Epistemonikos (searched July 19, 2023).

Gray literature via ProQuest: Brazilian Digital Library of Theses and Dissertations (BDTD), Global ETD Search (Networked Digital Library of Theses and Dissertations) (searched July 19, 2023).

Although we explored theses and dissertations, other potentially relevant sources of gray literature—such as trial registries or preprint repositories—were not systematically searched, which may limit the identification of recent or unpublished studies. We employed a search strategy based on the PICOS framework (Population, Intervention, Comparator, Outcome, Study design), using both MeSH terms and non-indexed keywords to capture a broader range of literature. Boolean operators (AND, OR, NOT) were applied to refine the search (see Supplemental Appendices 2 and 3). No date or language restrictions were imposed. Additionally, we performed backward citation searching of reference lists in eligible reviews and consulted experts in the field to identify any overlooked sources.

Data Collection and Analysis

Search results were exported to the Rayyan platform (http://rayyan.qcri.org) for deduplication and screening. Two independent reviewers (FAM, GJA), blinded to each other’s decisions, screened titles and abstracts. Disagreements were resolved by consensus or by involving a third reviewer (WRM). The full texts of potentially eligible reviews were then obtained for further assessment based on the predefined inclusion/exclusion criteria. The PRISMA flowchart was used to document each stage of the selection process.

Data Extraction and Management

We used a standardized data extraction form to ensure consistency and extracted the following from each included review:

Review Features: Objectives, inclusion criteria, participant details, interventions, comparators, outcomes, databases searched, dates, and language restrictions.

Primary Study Characteristics: Study design, sample sizes, participant demographics, intervention details, outcome measures, and bias assessments.

Results: Effect estimates (mean or standardized mean differences [SMDs], risk ratio [RRs]), confidence intervals (CIs), heterogeneity (I²), and any GRADE ratings.

Additional Details: Protocol registration, funding sources, conflicts of interest, and relevant methodological data.

Two reviewers (FAM, GJA) performed data extraction independently, resolving discrepancies by discussion or a third reviewer (WRM). No vital information was missing, so no authors were contacted.

Because overlapping primary studies can affect overviews, we mapped all trials from each review and calculated the Corrected Covered Area (CCA).³³ The degree of overlap was noted in tables, described in section “Results,” and considered in section “Discussion” to inform conclusions. This step aimed to minimize bias from duplicated data across multiple reviews.

Assessment of Methodological Quality of Included Reviews

We used the “A Measurement Tool to Assess Systematic Reviews” (AMSTAR 2)³⁴ to evaluate the methodological quality and risk of bias of the included reviews. Two authors (FAM, GJA) independently scored each review on AMSTAR 2’s 16 items. Discrepancies were resolved by consensus or via a third reviewer (WRM). Following AMSTAR 2 guidelines, we did not generate an overall score but instead classified the confidence in each review’s findings as high, moderate, low, or critically low.

Data Synthesis

We prespecified that systematic reviews including mixed clinical populations would be retained for descriptive mapping of LIBFR implementation (eg, variation in cuff width, pressure prescription, load, and frequency), provided they included a KOA component. However, quantitative synthesis and GRADE were restricted to effect estimates derived from KOA participants only, using reviews that conducted KOA-specific meta-analyses or reported KOA subgroup data that were clearly separable from other conditions.

We employed a narrative synthesis, with summary tables and figures to present findings. When effect estimates were available, we reported means, SMDs, and 95% CIs. Heterogeneity was assessed using the I² statistic; high heterogeneity prompted subgroup analyses and potential outlier exclusion to explore variability.³³ We documented evidence quality with GRADE, relying on reviews’ assessments or performing our own. Because multiple reviews addressed similar topics with overlapping primary studies, we prioritized Cochrane reviews, then the most recent, highest-quality (AMSTAR 2), and most comprehensive. Sensitivity analyses evaluated how selection priorities affected conclusions, ensuring robustness despite possible overlap.

Subgroup Analysis and Heterogeneity Assessment

We examined predefined subgroups, such as comparisons of LIBFR vs LIRT and LIBFR vs HIRT, as well as sex differences. Where subgroup data were not explicitly reported in the systematic reviews, we relied on available subgroup analyses or attempted to extract and interpret the data narratively.

Sensitivity Analysis

We performed sensitivity analyses to investigate the robustness of effect estimates when applying our hierarchical criteria for selecting representative data from overlapping reviews. The purpose was to determine whether different priority approaches led to different interpretations of the findings.

Summarizing the Evidence and Grading of Recommendations Assessment, Development, and Evaluation

For GRADE assessments, we evaluated certainty only for comparisons and outcomes informed by KOA-specific data. Consequently, indirectness was not downgraded due to mixed populations in reviews retained for descriptive purposes, because those reviews did not contribute non-KOA data to pooled effects or to GRADE.

We used GRADE to assess the certainty of evidence across the included systematic reviews. The GRADE approach considers risk of bias, inconsistency, indirectness, imprecision, and publication bias.³⁵ We generated evidence summary tables using GRADEpro GDT software (https://www.gradepro.org/). When reviews did not report GRADE, we conducted our own GRADE judgments following standard guidelines and using the information reported by the systematic reviews.

Results

Selection of Reviews

The database searches identified 251 records: 10 from PubMed via MEDLINE, 15 from EMBASE, 46 from the Cochrane Database of Systematic Reviews, 12 from Web of Science, 7 from CINAHL and Sports Discuss, 26 from PEDro, 37 from Epistemonikos, and 98 from gray literature sources. After removing 35 duplicates, 216 references remained for screening. Following the title and abstract review, 198 records were excluded, leaving 18 full-text systematic reviews that met the inclusion criteria (Figure 1—PRISMA flowchart).

Figure 1.

PRISMA 2020 flow diagram for new systematic reviews, which included searches of databases, registers, and other sources.

Description of Included Reviews

We included 18 systematic reviews investigating LIBFR in individuals with KOA or including it as a subgroup, summarizing their characteristics in Table 1. Most reviews comprised RCTs and compared LIBFR with LIRT or HIRT. The interventions lasted 4 to 12 weeks at frequencies of 2 to 3 sessions per week. Overall, these reviews covered 5656 participants. The earliest review is dated to March 2016 and the latest to October 2022.^36,37

Table 1.

Summary of Characteristics of Systematic Reviews Evaluating Blood Flow Restriction in Knee Osteoarthritis..

Systematic review	Search date	Study design	N RCTs (n participants)	Pathologies	Intervention/comparison	Relevant outcomes	Risk of bias assessment	Assessing the certainty of evidence
Reina-Ruiz et al³⁷	October 31, 2022	RCTs in English (language of scientific publications worldwide) and Spanish (authors’ native language)	14 (533)	Neurological and musculoskeletal pathologies: • Knee OA; • Patellofemoral pain; • ACL reconstruction; • Sporadic inclusion body myositis; • Anterior knee pain; • Severe multiple sclerosis; • Meniscus repair surgery; • Rheumatoid arthritis; • Non-reconstructive knee arthroscopy	BFR intervention group x Group control Types of exercises: aerobic and strength. Frequency: sessions 1 to 3 times/week; intervention duration from 1 session to 16 weeks	- Pain (KOOS, EAV, DAS-28, Kujala, MDI, NPRS, Börg pain scale, WOMAC); - Objective functionality (TUG, TSA, FSST, STS5, 40 m walking speed, T25FW, SSWV, 6-minute walk test, modified SEBT, sit and stand test, Stair Climb Power); - Subjective functionality (SPPB, LLFDI, IKDC, LEFS, KOOS, Lysholm, Tegner, IBMFRS, HAQ, Berg balance scale); - Quality of Life (SF-36, MSIS-29, VR-12, WOMAC)	PEDro Scale	None
Reina-Ruiz et al³⁸	January 1, 2015 to August 21, 2021	RCTs in Spanish, English, Italian, French, and Portuguese	20 (787)	Individuals with chronic pathologies: • Knee OA; • Chronic nonspecific low back pain; • Elderly individuals in coma; • Chronic kidney disease; • End-stage kidney disease; • Coronary disease; • Stage 2 chronic kidney disease; • ACL reconstruction; • Patellofemoral pain; • Sporadic inclusion body myositis; • Severe multiple sclerosis; • Open heart surgery; • Rheumatoid arthritis; • Non-reconstructive knee arthroscopy	BFR intervention group (30-80% vascular occlusion) x Group control Types of exercises: aerobic and strength. Frequency: sessions 2 to 3 times/week; duration: 4 to 16 weeks	- Muscle strength (isokinetic dynamometry, 1RM test, CAR, MMT-8, kinetic communicator); - Muscle hypertrophy (magnetic resonance imaging, ultrasound, computed tomography, measuring tape); - Physiological muscle changes (Fatigue: MFIS, FSS; Effort: RPE; Time: accelerometer); - Blood and cardiovascular outcome (Vascular thickness: cephalic vein and radial artery ultrasound; Lung capacity: ergospirometry test-aerobic and anaerobic thresholds, maximum oxygen consumption; Blood pressure: systolic and diastolic pressure; Cardiac adaptation: electrocardiogram, end-diastolic and end-systolic dimensions, and left ventricular ejection fraction)	PEDro Scale	None
Wang et al³⁹	January 1 to May 31, 2020	RCTs in English	5 (182)	• Knee OA	LIBFR x HIRT (⩾ 60% 1RM) OR LIRT (< 60% 1RM) or walking, placebo, or other intervention without BFR	- Pain (KOOS pain subscale, numeric pain scale, WOMAC pain subscale) - Muscular strength (isokinetic knee extensor strength, isotonic leg press muscle strength, knee extensor strength) - Muscle hypertrophy (cross-sectional area of the quadriceps) - Performance of physical function (TUG, sit-stand test, gait speed, SPPB) - Self-reported function (Lequesne Score, WOMAC, LLFDI) - Adverse events (intolerance to pressure cuff discomfort; discontinuation of intervention due to exercise-induced knee pain, especially in the high-intensity resistance exercise group)	Cochrane RoB (risk of bias in RCTs)	GRADE
De Araújo et al⁴⁰	November 2021	Clinical trials (randomized and non-randomized)	9 (331)	• Knee OA • Healthy men • Postknee arthroscopy • Patellofemoral pain • Musculoskeletal injuries of the lower limbs • Myositis	LIBFR X HIRT or LIRT X control group without physical exercise (CG)	- Muscle strength - Functional capacity	TESTEX scale	None
Grantham et al³	January 2020	RCTs in English	5 (199)	• Knee OA	LIBFR (20% and 30% of 1RM) X LIRT (30% 1RM) or HIRT (80% 1RM) Intervention for at least 4 weeks	- Pain (numeric pain scale, visual analog scale, WOMAC pain subscale, KOOS pain subscale); - Self-reported functional capacity (Lequesne and WOMAC questionnaires, LLFDI) or quality of life (SF-36); - Objective physical function (TUG, stair climb power, sit-stand test, and walking speed); - Muscle strength (manual dynamometer, isokinetic, isotonic strength and power, leg press 1 RM) - Muscle hypertrophy (computed tomography and magnetic resonance imaging)	PEDro Scale	GRADE
Bobes Álvarez et al²³	April to June 2020	RCTs in English and Spanish	10 (364)	• ACL reconstruction • Knee OA	LIBFR (20-30% 1RM) x LIRT (30% 1RM) and HIRT (60-70% 1RM)	Primary outcome: - Muscle strength (isokinetic, isotonic strength, MVC, leg press strength test) Secondary outcomes: - Muscle hypertrophy (quadriceps cross-sectional area—CSA)—computed tomography and magnetic resonance imaging; - Pain (WOMAC, KOOS, and numerical pain scale) - Functionality (LLFDI, SPPB, gait speed, sit-stand test, TUG) and/or Quality of Life (IKDC, WOMAC, SF-36, and Lequesne questionnaire)	Cochrane RoB-2 PEDro Scale	None
Li et al²¹	January 23, 2020. Search from January 2000 to January 2020	RCTs in English Gray literature, abstracts, or poster presentations were excluded	9 (369) 135 men and 234 women	• Knee OA • Rheumatoid arthritis • Patellofemoral pain • ACL reconstruction	LIBFR (179 subjects) x LIRT (94 subjects) and HIRT (96 subjects)	- Muscular strength (maximum strength of the lower-limb extensors) - Pain (KOOS, WOMAC, SF-36, visual analog scale)	Cochrane RoB-2	None
Nitzsche et al⁴¹	January 2021	RCTs	10 (386)	• ACL reconstruction (n = 4) • Knee OA (n = 4) • Anterior knee pain (n = 1) • Various injuries to the lower limb (n = 1)	LIBFR x LIRT and HIRT	Primary outcome: - Muscle strength (assessed by isokinetic dynamometer, dynamic strength, and maximum voluntary contraction) Secondary outcomes: - Muscular hypertrophy (cross-sectional area assessed by computed tomography, magnetic resonance imaging, and ultrasound) - Pain (KOOS, WOMAC, numerical pain scale, Kujala score for patellofemoral pain)	Cochrane RoB-2 TESTEX scale	None
Pitsillides et al⁴²	November 1 to December 15, 2019	RCTs in English	3 (117)	• Knee OA	LIBFR x Moderate to HIRT (60-80% 1RM)	Primary outcomes: - Pain (numerical pain scale, WOMAC) - Muscle strength (handheld dynamometer, leg press 1RM, chair extension 1RM, isokinetic dynamometer) Secondary outcomes: - Physical function (Lequesne, TUG, sit-stand test, WOMAC, SPPB, LLFDI) - Quality of life (WOMAC, SF-36)	Cochrane RoB-2 PEDro Scale	None
Dos Santos et al⁴³	July 2021	RCTs in English	5 (174)	• Knee OA • Rheumatoid arthritis	LIBFR (20-50% 1RM) x LIRT (20-50% 1RM) or moderate to HIRT (> 60% 1RM) Intervention duration: 2 weeks or more	- Maximum muscle strength (isometric dynamometer, isokinetic dynamometer, and RM tests) (specific tests for quadriceps strength); - Muscle hypertrophy (computed tomography or magnetic resonance imaging); - Functionality (TUG and 400 m walk test)	Cochrane RoB-2 PEDro Scale	None
Cuyul-Vásquez et al¹³	May 2019	RCTs or controlled clinical trial published in English or Spanish	8 (340) LIBFR: 160 participants RT: 180 participants men: 140 women: 200 The average age was 49.5 years	• Knee OA • Patellofemoral pain • Anterior knee pain • Knee arthroscopy	LIBFR (20% and 30% 1RM) x LIRT (30% 1RM) and HIRT (70-80%1RM) Note: 1 study used an external weight of 5 kg. Another study used stretching as a supporting modality, in both control and experimental groups	- Pain (numerical pain scale, visual analog scale, KOOS, WOMAC) - Self-reported assessment of knee function (WOMAC, Lequesne, Kujala Score for patellofemoral, LLFDI, and KOOS) - Lower-limb function (SPPB)	Cochrane RoB-2	GRADE
Ferlito et al⁴	October 2018 to April 2020	Single-blind and double-blind RCTs No language restriction	5 (190) Average age 59.89 ± 7.47 years	• Knee OA	LIBFR (20-30% 1RM) x LIRT (20-30% 1RM) and HIRT (60-80%1RM) The duration of the BFR training intervention ranged from 4 to 12 weeks, with a frequency of 2 to 3 training sessions per week	- Knee function (TUG, timed position test, stair climb power, WOMAC, Lequesne, LLFDI, SPPB, and walking speed); - Pain (WOMAC pain subscale, KOOS pain, numeric pain scale); - Muscle strength (isotonic bilateral leg press, knee extensor isokinetic, 1RM leg press, leg press power 40% 1RM, isometric voluntary contraction, 1RM leg press test, and 1RM knee extension test); - Muscle hypertrophy (cross-sectional area)	PEDro Scale	None
Van Cant et al⁴⁴	January 2019	RCTs	8 (335) BFR: 157 participants; Control interventions: 178 participants Age between: 24 and 62 years	• Knee OA (n = 164) • Patellofemoral pain (n = 79) • Postoperative knee arthroscopy (n = 24) • ACL reconstruction with flexor tendon graft (n = 68)	LIBFR (30% 1RM) leg press, knee extension, and straight knee raise (SRL) X LIRT (30% 1RM) and HIRT (70-80% 1RM) Duration of the BFR intervention ranged from 2 to 16 weeks, with a frequency of 2 to 6 sessions per week Most studies based their training on 4 sets with 30 reps in the first set and 15 reps in the last 3 sets, with 30 seconds of rest between sets	- Pain (numerical pain scale, visual analog scale); - Knee symptoms and function (AKPS, KOOS, VR-12, Lequesne, WOMAC); - Self-selected physical capacity (walking speed, 5-time sit-stand test, FSST, TSA, TUG) - Muscle strength (maximum isotonic strength through a 1RM test, knee extensor strength with an isometric and isokinetic dynamometer); - Muscle hypertrophy—cross-sectional area (computed tomography, magnetic resonance imaging, thigh perimetry, and ultrasound)	PEDro Scale	None
Centner et al²⁴	February 1, 2020	RCTs in English and German	9 (269) Subgroups: Knee OA: 3 (117) - ACL: 6 (152)	• Knee OA • ACL rupture/reconstruction	LIBFR (at rest or during physical activity) X control group Subgroups - Knee OA: duration of intervention was 6 weeks (1 study) and 12 weeks (2 studies) - ACL: 3 studies use BFR from 0-16 weeks postoperatively; 2 studies use BFR pre-rehabilitation; 1 study using BFR during immobilization (without associated exercise)	Subgroups Knee OA - Muscle hypertrophy (computed tomography and double X-ray absorptiometry) - Muscle strength (maximum voluntary contraction and 1RM test) - Functionality (WOMAC, TUG, sit-stand test, SEBT)	PEDro Scale	None
Baker et al²²	January 1, 1990 and January 1, 2019	RCTs, cohort studies, cross-sectional studies using BFR therapy with exercise interventions in adults aged 50 and over. English language	30 (413) Men (n = 101); Women (n = 278); Sex not reported (n = 34) Average age of 63 years	Overall, 80% of articles included healthy participants, but individuals in a coma, those with spinal cord injury (incomplete quadriplegia), and those with OA were included	LIBFR (20-minute walk; passive mobilization in comatose individuals) X LIRT and HIRT Study durations ranged from cross-sectional to 16 weeks, full size ranged from 9 to 48. Twenty-five of the 30 studies reported outcome measures of muscle strength, and 8 studies included measurements from the 30-second sit-to-stand test and/or TUG. The sample size ranged from 6 to 56 with a total of 694 participants (men n = 205; women n = 455; gender not reported n = 34), and the average age was 66 years. Overall, 74% of articles included healthy participants, but individuals with OA, osteoporosis, cardiovascular disease, and sporadic myositis were included	- Muscular hypertrophy (cross-sectional area, volume, mass, thickness, or circumference of the limb) - Muscle strength (maximum voluntary contraction force, maximum voluntary isometric contraction force, torque, muscle activation, or 1 repetition maximum) - Physical function (sit and stand test for 30 seconds, timed 8 feet, and TUG)	Cochrane RoB-2	None
Clarkson et al⁴⁵	November 25, 2018	RCTs in English	13 (332)	- Most commonly examined population were older adults (aged 60 and over) Other populations examined: - Healthy adults - Women with knee OA - Postmenopausal women; - Adults postknee arthroscopy; - Individuals with end-stage renal disease; - Individuals with sporadic myositis	LIBFR (aerobic, RT: 1 study used BRF during hydrotherapy exercises) X Controls: control groups (RT, no exercise or alternative traditional exercise prescriptions). Within participant controls single-limb interventions were excluded. Most interventions ranged from 6 to 12 weeks, with only 1 being implemented for less than 3 weeks and no study duration longer than 12 weeks. Training sessions occurred 2 times per week in 7 studies, 3 times per week in 2 studies, 4 times per week in 2 studies, 1 study 5 sessions per week, and another study completing 9 sessions per week and resting on weekends. Session lengths ranged from 8 minutes to 60 minutes. LIBFR (20-30% of 1RM) with 75 reps (usually as 1 set of 30 reps followed by 3 sets of 15 reps, which is common). The BFR interventions in the other studies used training variables closer to traditional HIRT (3-4 sets of 10-15 repetitions using loads between 40% and 50% 1RM) LIBFR (aerobic exercise) lasted 10 to 20 minutes and were performed at walking speeds on a treadmill of 4 to 4.5 km/h with approximately 45% maximum heart rate reserve and equivalent to 11-14 on the Borg perceived exertion scale. Cuff occlusion pressures ranged from 60% to 80% of the maximum occlusion pressure; or 50% to 150% of systolic blood pressure; or from 70 mm Hg to 270 mm Hg	Objective measure of physical function indicative of ADL. Subjective measures associated with physical function (questionnaires or surveys) were excluded - Physical function indicative of ADL (6-minute walk test, variations of the sit-stand test, balance tests, or grip strength tests, which present similarities in their execution with everyday activities). Measures excluded were laboratory tests (maximal strength tests or graded exercise tests using measures of oxygen utilization, ventilatory threshold, or lactate), as these do not reflect ADL	Cochrane RoB-2	None
Barber-Westin and Noyes¹⁸	Systematic literature review from 1974 to 2017	RCTs and non-randomized in English 8 level 1 RCTs 1 level 2 non-randomized study	9 (335) 165 BFR individuals (93 women, 72 men); average age of 42.2 years. 170 controls (100 women, 70 men); average age of 40.6 years	- Knee OA (n = 4) - ACL reconstruction (n = 3) - Postknee arthroscopy (n = 1) - Patellofemoral pain (n = 1)	LIBFR (occlusion pressure 160-200 mm Hg; 60%-70% of maximum occlusion pressure; occlusion maintained during exercise sets and deflated during rest between sets) x LIRT and HIRT While some studies focused solely on quadriceps strengthening, others included hamstring exercises. Although not evaluated as part of the BFRT protocol, hip exercises (abduction-adduction, lateral leg raise, and Theraband resistance) and gastrocnemius-soleus exercises were also included in 2 studies	- Muscle strength of quadriceps and hamstrings (isokinetic, isometric, or maximum leg press); - Muscle hypertrophy—cross-sectional area of the quadriceps and hamstrings (magnetic resonance imaging or ultrasound, muscle biopsy)	PEDro Scale Methological Index for Non-Randomized Studies (MINORS) for non-randomized controlled trials	None
Hughes et al³⁶	January 1, 1990 to March 31, 2016	RCTs in English	20 Meta-analysis: 13 studies Average age (58 ± 14) years	- Knee OA (n = 3) - Ligament injury (n = 3) - Sporadic myositis (n = 1) - Sarcopenic elderly (n = 13)	LIBFR (10-30% of 1RM), resistance training with elastic band, and walking training (45% of heart rate reserve at 67 m/min-1) x LIRT and HIRT Samples of 10 to 41 participants Cuff width: 3 to 18 cm Occlusion pressure: 60 to 270 mm Hg based on maximum occlusion pressure or systolic blood pressure. Intervention duration: 2 to 16 weeks Frequency of 2 to 6 training sessions per week. Some studies performed repeated sessions on the same day when BFR was used alone and in combination with simple muscle exercises	- Muscular strength (maximum isotonic strength, 1RM test—most studies, isometric and isokinetic maximum strength); - Muscle hypertrophy (cross-sectional area, muscle mass, muscle volume, muscle thickness, and fat cross-sectional area); - Physical function (reaction time tests, strength to climb stairs, single-leg balance, timed supports, 10 m walk time, maximum step distance, functional reach test, sit-stand test)	Cochrane RoB TESTEX scale	None

RCTs (randomized clinical trials); LIBFR (low-intensity resistance training with blood flow restriction); LIRT (low-intensity resistance training); HIRT (high-intensity resistance training); RT (resistance training); BRF (blood flow restriction); ACL (anterior cruciate ligament); KOOS (Knee Osteoarthritis Outcome Score), VAS (visual analog scale), DAS-28 (Diseases Activity Score); Kujala, MDI (Myositis Damage Index); NPRS (numerical pain ratio scale), Pain Börg Scale, WOMAC Pain (Western Ontario and McMaster Universities Osteoarthritis Index); TUG (Time Up and Go Test); TSA (timed stair ascent); FSST (four square step test); STS5 (sit-to-stand 5 times); T25FW (timed 25-foot walk test); SSWV (self-selected walking velocity); SEBT (star excursion balance test) modified; SPPB (Short Physical Performance Battery); LLFDI (Late Life Function and Disability Instrument); IKDC (International Knee Documentation Committee); LEFS (Lower Extremity Function Scale); KOOS (Knee Osteoarthritis Outcome Score); IBMFRS (Inclusion Body Myositis Functional Rating Scale); HAQ (Health Assessment Questionnaire); SF-36 (Short Form-36 Health Survey); MSIS-29 (Multiple Sclerosis Impact Scale-29); VR-12 (Veterans RAND 12-Item Health Survey); CAR (central activation ratio); MMT 8 (Manual Muscle Test–eight muscles); MFIS (modified fatigue impact scale); FSS (fatigue severity scale); RPE (Rating Perceived Exertion Scale); CSA (cross-sectional area); MVC (maximum voluntary contraction); BES (best evidence synthesis); SLR (straight leg raises); AKPS (Anterior Knee Pain Scale); ADL (activities of daily living).

Several included systematic reviews addressed LIBFR across broader clinical populations.^22,37,38 These reviews were retained to illustrate the breadth and variability of LIBFR application and were appraised with AMSTAR 2. Importantly, they were used for descriptive synthesis only and did not contribute to the quantitative meta-analyses or to the GRADE certainty ratings, which were informed exclusively by KOA-specific data.

Initially, these reviews examined various populations (eg, patellofemoral pain, anterior cruciate ligament (ACL) reconstruction, multiple sclerosis, sarcopenia, rheumatoid arthritis). To reduce heterogeneity and concentrate on KOA, they conducted separate meta-analyses for KOA only, excluding data from other populations. As a result, individuals with distinctly different conditions were omitted, ensuring that findings more accurately reflect the efficacy and safety of interventions for KOA. Any remaining heterogeneity within the KOA subgroup was also assessed and reported.

Of the 18 reviews, only 4 focused solely on KOA, with 3 performing meta-analyses.^3,4,39 Most studies evaluated outcomes related to pain, muscle strength, muscle hypertrophy, physical function, QoL, and adverse events. In several cases, we performed further meta-analyses (using Review Manager) to refine the data and remove results combined with other pathologies, maintaining focus on KOA populations.

In total, these reviews included 57 primary studies. Although many searched multiple languages, English prevailed, with fewer studies reported in Spanish, German, French, Italian, and Portuguese.

Divergent Recommendations Across Reviews

We observed disagreements among the included reviews regarding the clinical recommendation for LIBFR. Of the 18 reviews, 14 (77.7%) suggested LIBFR as a beneficial intervention, while 3 (16.6%) found no advantage. One review³⁹ did not recommend the use of LIBFR for most outcomes but highlighted that LIBFR might be safer compared with HIRT regarding adverse events. Interestingly, the reviews that did not recommend LIBFR tended to be those that performed meta-analyses, whereas those favoring the approach relied more on narrative syntheses. Table 2 summarizes these findings and highlights the differences in recommendations and the presence or absence of meta-analytic data.

Table 2.

Systematic Reviews With Conclusions in Favor of or Against LIBFR.

Systematic review	Recommended or did not recommend BFR	Meta-analysis (yes or no)
Reina-Ruiz et al³⁷	Recommended	No
Reina-Ruiz et al³⁸	Recommended	No
Wang et al³⁹	Not recommended for most outcomes, but recommended it for adverse events compared to HIRT	Yes
De Araújo et al⁴⁰	Recommended	No
Grantham et al³	Not recommended	Yes
Bobes Álvarez et al²³	Recommended	No
Li et al²¹	Recommended	Yes
Nitzsche et al⁴¹	Not recommended	Yes
Pitsillides et al⁴²	Recommended	No
Dos Santos et al⁴³	Recommended	Yes
Cuyul-Vásquez et al¹³	Not recommended	Yes
Ferlito et al⁴	Recommended	Yes
Van Cant et al⁴⁴	Recommended	Yes
Centner et al²⁴	Recommended	No
Baker et al²²	Recommended	No
Clarkson et al⁴⁵	Recommended	No
Barber-Westin and Noyes¹⁸	Recommended	No
Hughes et al³⁶	Recommended	Yes

Conflicts of Interest and Funding

Among the 18 reviews, 16 reported no conflict of interest, while 2 did not report on this issue. Four reviews documented independent funding sources through university grants, government research bodies, or strategic academic partnerships.^21,38,39,45 The remaining reviews either lacked funding or did not report funding sources. Table 3 presents these details.

Table 3.

Conflict of Interest and Funding Source for Systematic Reviews.

Systematic review	Conflict of interest(Yes, No, Not reported)	Funding source
Reina-Ruiz et al³⁷	No	No
Reina-Ruiz et al³⁸	No	Erasmus + Strategic Partnership for Higher Education Program: 2018-1-PL01-KA203-051055.
Wang et al³⁹	Not reported	Supported by the National Natural Science Foundation of China (grant 81871848).
De Araújo et al⁴⁰	No	No
Grantham et al³	No	No
Bobes Álvarez et al²³	No	No
Li et al²¹	No	The study was conducted with financial aid from the University Sains Malaysia (1001/PPSK/8012364).
Nitzsche et al⁴¹	No	No
Pitsillides et al⁴²	No	Not reported
Dos Santos et al⁴³	No	No
Cuyul-Vásquez et al¹³	No	No
Ferlito et al⁴	No	No
Van Cant et al⁴⁴	No	No
Centner et al²⁴	No	Not reported
Baker et al²²	No	No
Clarkson et al⁴⁵	Not reported	This research was supported only by local funds made available by the School of Exercise and Nutrition Sciences, Faculty of Health, Deakin University, Victoria, Australia.
Barber-Westin and Noyes¹⁸	No	Not reported
Hughes et al³⁶	No	Not reported

Overlap of Studies

We observed moderate overlap among primary studies in these systematic reviews (overlap = 57.8%; CA = 14.9%; CCA = 9.9%). Five core trials were repeatedly cited in KOA-focused analyses, possibly skewing overall conclusions.^16,46-49 To mitigate this, we used a hierarchical approach, prioritizing (1) the most recent³⁹ and (2) the most comprehensive³ reviews when reporting estimates, and also applied AMSTAR 2 to confirm the robustness of heavily repeated trials. This strategy helped assess whether frequently cited studies distorted conclusions, examine methodological quality, and preserve potentially relevant evidence. Although prioritizing recency and comprehensiveness reduces duplication, we remain mindful of unique insights from other reviews and acknowledge the potential impact of overlap.

Methodological Quality of Included Reviews (AMSTAR 2)

We judged the methodological quality of each included review using AMSTAR 2. The overall confidence ratings were as follows:

Moderate confidence: 9 reviews.^{21-24,37-39,42,45}

Low confidence: 7 reviews.^{3,4,13,36,41,43,44}

Critically low confidence: 2 reviews.^18,40

Table 4 provides a narrative summary of our AMSTAR 2 judgments. Reviews with moderate confidence generally provided a more accurate and comprehensive summary of the evidence. Those rated as low or critically low had methodological shortcomings that limited the reliability and interpretability of their findings.

Table 4.

AMSTAR 2 Summary for Reviews Evaluating Blood Flow Restriction in Knee Osteoarthritis.

Systematic review	1	2^a	3	4^a	5	6	7^a	8	9^a	10	11^a	12	13^a	14	15^a	16	Overall confidence in review results
Reina-Ruiz et al³⁷	Y	PY	Y	PY	Y	Y	PY	Y	Y	N	N/A	N/A	Y	N	N/A	Y	Moderate
Reina-Ruiz et al³⁸	Y	PY	Y	PY	Y	Y	PY	Y	Y	N	N/A	N/A	Y	N	N/A	Y	Moderate
Wang et al³⁹	Y	Y	Y	PY	Y	Y	Y	Y	Y	N	Y	Y	Y	Y	Y	N	Moderate
De Araújo et al⁴⁰	Y	PY	Y	PY	Y	Y	N	Y	Y	N	N/A	N/A	N	N	N/A	N	Critically low
Grantham et al³	Y	Y	Y	Y	N	N	Y	Y	Y	N	Y	Y	Y	Y	N	Y	Low
Bobes Álvarez et al²³	Y	PY	Y	Y	Y	Y	Y	Y	Y	N	N/A	N/A	Y	Y	N/A	Y	Moderate
Li et al²¹	Y	Y	Y	PY	Y	Y	Y	Y	Y	N	Y	Y	Y	Y	Y	N	Moderate
Nitzsche et al⁴¹	Y	Y	Y	PY	N	N	Y	Y	Y	N	Y	Y	Y	Y	N	Y	Low
Pitsillides et al⁴²	Y	PY	Y	PY	N	N	Y	Y	Y	N	N/A	N/A	Y	N	N/A	Y	Moderate
Dos Santos et al⁴³	Y	Y	Y	PY	Y	Y	Y	Y	Y	N	Y	Y	Y	Y	N	Y	Low
Cuyul-Vásquez et al¹³	Y	Y	Y	Y	Y	Y	PY	Y	Y	N	Y	Y	Y	Y	N	Y	Low
Ferlito et al⁴	Y	Y	Y	PY	Y	Y	Y	Y	Y	N	Y	Y	Y	Y	N	Y	Low
Van Cant et al⁴⁴	Y	Y	Y	Y	Y	N	Y	Y	Y	N	Y	Y	Y	Y	N	Y	Low
Centner et al²⁴	Y	PY	Y	PY	Y	Y	Y	Y	Y	N	N/A	N/A	Y	Y	N/A	Y	Moderate
Baker et al²²	Y	PY	Y	PY	Y	N	Y	Y	Y	N	N/A	N/A	Y	N	N/A	N	Moderate
Clarkson et al⁴⁵	Y	PY	Y	PY	Y	N	Y	Y	Y	N	N/A	N/A	Y	N	N/A	N	Moderate
Barber-Westin and Noyes¹⁸	Y	PY	Y	PY	N	N	N	Y	Y	N	N/A	N/A	Y	N	N	N	Critically low
Hughes et al³⁶	Y	Y	Y	PY	Y	Y	Y	Y	Y	N	Y	Y	Y	Y	N	Y	Low

Scores for the 16 questions from the AMSTAR 2 critical assessment tool. N: no; N/A: not applicable, as no meta-analysis was performed; PY: partial yes; Y: yes.

Items representing critical domains.

Risk of Bias of Prioritized Reviews

Since we prioritized both Grantham et al³ and Wang et al³⁹ for meta-analysis results, we carefully examined the risk of bias in their included primary RCTs (Figure 2). Across both reviews, the risk of bias ranged from very low to moderate, indicating that no trial reached a critical level of concern. Nonetheless, some domains—particularly blinding and selective outcome reporting—showed moderate risk, which may influence the direction or magnitude of the estimated effects.

Figure 2.

(A) Summary of risk of bias in the included studies. (B) Risk of bias as percentages across all included studies.

In Grantham et al,³ sensitivity analyses and subgroup analyses were conducted to explore heterogeneity and evaluate the robustness of findings, specifically comparing LIRT, HIRT, and LIBFR, as well as differences by gender (male vs female). Although these subgroup analyses provide insights into potential effect modifiers, the presence of moderate-risk studies still calls for cautious interpretation of pooled results.³

Overall, the variability in methodological quality across the included RCTs underscores the need for well-designed, high-quality trials in future research. Such studies, combined with further sensitivity analyses (eg, excluding lower-quality data), would enhance our understanding of the true impact of these interventions on KOA outcomes.

Effects of the Interventions

Primary Outcomes

Pain

Pain measurement tools used in the primary studies included numerical rating scales, visual analog scales, and the pain subscales of WOMAC and KOOS. All systematic reviews performing meta-analyses, including Grantham et al³ and Wang et al,³⁹ concluded that there were no statistically significant differences in pain reduction between LIBFR and either LIRT or HIRT in KOA. The certainty of the evidence ranged from low to moderate. Overall, LIBFR did not appear superior to conventional resistance training for reducing pain in individuals with KOA.

Muscle Strength

Muscle strength was assessed via handheld or isokinetic dynamometry and 1RM testing. In general, the included reviews reported no significant differences between LIBFR and LIRT or HIRT for strength gains. The certainty of the evidence varied (very low, low, moderate). However, 1 review²¹ found that LIBFR led to greater strength gains compared with LIRT (SMD = 0.75; 95% CI = [0.40 to 1.10]; I² = 56%; 3 studies; 139 participants) with low-certainty evidence. This isolated finding suggests that, under some conditions or specific protocols, LIBFR may provide a strength advantage over low-load exercise without BFR.

Functionality/Physical Function

Functionality outcomes included performance-based tests such as the TUG, 400-m walk test, gait speed tests, sit-to-stand tests, and stair-climbing tasks. The meta-analyses showed no significant differences between LIBFR and LIRT or HIRT for improving these functional metrics. The certainty of evidence was classified between very low and moderate. Thus, from an overall perspective, adding BFR to low-intensity exercises did not yield clear functional performance improvements beyond those achieved with standard approaches.

Self-Reported Function

Scales such as WOMAC, Lequesne, SF-36, and the Late Life Function and Disability Instrument (LLFDI) were used to evaluate self-reported function. Neither Wang et al³⁹ nor Van Cant et al⁴⁴ identified notable differences between LIBFR and traditional resistance training approaches for self-reported functional improvement. Certainty of evidence was low to moderate. Thus, patient-perceived improvements in their functional status were not significantly altered by the addition of BFR.

Muscular Hypertrophy

Muscle volume and cross-sectional area assessments were conducted via imaging techniques such as computed tomography or MRI. Van Cant et al⁴⁴ observed that LIBFR produced greater muscular hypertrophy compared with LIRT (SMD = 0.81; 95% CI = [0.10 to 1.52]; I² = 52%; 2 studies; 72 participants), with very low-certainty evidence. Dos Santos et al⁴³ also reported increases in muscle mass in both HIRT (+8%; P < .0001) and LIBFR (+7%; P < .0001). These findings suggest that when hypertrophy is a key target, adding BFR to low-load training might enhance muscle growth to a degree that is more comparable to high-load training.

Secondary Outcomes

Quality of Life

None of the included reviews conducted a meta-analysis specifically for QoL outcomes in KOA. Thus, evidence regarding LIBFR’s impact on QoL remains sparse and inconclusive.

Adverse Events

Wang et al³⁹ reported 4 primary studies noting participant exclusions due to adverse events, with the most common being exercise-induced knee pain. Pooled analyses (RR = 0.45; 95% CI = [0.20 to 1.01]; P = .05; I² = 0%) indicated no significant difference in the rate of adverse events between LIBFR and traditional exercise interventions. However, when comparing LIBFR with HIRT specifically, some data suggest fewer adverse events in the LIBFR group, possibly due to lower mechanical stress on the joint.³⁹

Certainty of Evidence—Grading of Recommendations Assessment, Development, and Evaluation

Of the 9 systematic reviews with meta-analyses, 3 applied GRADE.^3,13,39 For the other 6, we conducted our own GRADE assessments.^{4,21,36,41,43,44} The GRADE analyses frequently downgraded the certainty due to inconsistency (heterogeneity in results), imprecision (small sample sizes), and publication bias. None of the included evidence sets were downgraded for indirectness, as the populations and interventions generally matched our inclusion criteria. The certainty of evidence according to GRADE, together with the sensitivity analysis of LIBFR in knee osteoarthritis, is summarized in Table 5.

Table 5.

Summarizing the Evidence (GRADE) and Sensitivity Analysis of LIBFR in Knee Osteoarthritis.

Systematic review	Pain	Muscle strength	Functional disability	Functionality/physical function	Self-reported function	Hypertrophy	Adverse events
Wang et al³⁹	LIBFR x RT SMD (-0.04); 95% CI (-0.31 to 0.24); I² = 0%; 6 studies; 206 participants (99 LIBFR and 107 RT) CERTAINTY OF EVIDENCE: MODERATE (downgraded for: Limitations in study design—lack of blinding or failure to describe allocation confidentiality)	LIBFR x RT SMD (0.30); 95% CI (-0.1 to 0.91); I² = 83%; 5 studies; 275 participants (134 LIBFR and 141 RT) CERTAINTY OF EVIDENCE: VERY LOW (downgraded for: Limitations in the study design—lack of blinding or failure to describe allocation confidentiality; inconsistency heterogeneity I²> 50%; imprecision—Lower 95% CI encompassed an effect size of 0.5).		LIBFR x RTSMD (0.12);95% CI (-0.55 to 0.78);I² = 82%;5 studies;220 participants (108 LIBFR and 112 RT)CERTAINTY OF EVIDENCE: VERY LOW (downgraded for: Limitations in the study design—lack of blinding or failure to describe allocation confidentiality; inconsistency heterogeneity I²> 50%; imprecision—Lower 95% CI encompassed an effect size of 0.5)	LIBFR x RTSMD (0.14);95% CI (-0.24 to 0.52);I² = 0%;3 studies;107 participants (53 LIBFR and 54 RT)CERTAINTY OF EVIDENCE: MODERATE (downgraded for: Limitations in study design—lack of blinding or failure to describe allocation confidentiality)		LIBFR x RTRR (0.45);95% CI (0.20 to 1.01);I² = 0%;4 studies;180 participants (86 LIBFR and 94 RT)CERTAINTY OF EVIDENCE: MODERATE(downgraded for: Limitations in study design—lack of blinding or failure to describe allocation confidentiality)LIBFR x HIRTRR 0.26; [95% CI 0.09 to 0.72], P < .01; I² = 0%.^a
Grantham et al³	LIBFR x RTSMD (0.023);95% CI (-0.271 to 0.317);I² = 0%;4 studies;163 participantsCERTAINTY OF EVIDENCE: MODERATE(downgraded for: heterogeneity)SubgroupsLIBFR x LIRTSMD (0.136);95% CI (-0.234 to 0.506);I² = 0%;3 studies;113 participantsCERTAINTY OF EVIDENCE: MODERATE(downgraded for: imprecision)LIBFR x HIRTSMD (-0.170);95% CI (-0.654 to 0.313);I² = 0%;2 studies;66 participantsCERTAINTY OF EVIDENCE: MODERATE		LIBFR x RTSMD (-0.198);95% CI (-0.595 to 0.199);I² = 0%;2 studies;82 participantsCERTAINTY OF EVIDENCE:LOW(downgraded for: heterogeneity and imprecision)SubgroupsLIBFR x HIRTSMD (-0.237);95% CI (-0.721 to 0.248);I² = 0%;2 studies;66 participantsCERTAINTY OF EVIDENCE: MODERATE(downgraded for: imprecision)	LIBFR x RTSMD (-0.148);95% CI (-0.567 to 0.271);I² = 9.375%;2 studies;82 participantsCERTAINTY OF EVIDENCE: MODERATE(downgraded for: imprecision)SubgroupsLIBFR x HIRTSMD (0.056);95% CI (-0.427 to 0.539);I² = 0%;2 studies;66 participantsCERTAINTY OF EVIDENCE: MODERATE(downgraded for: imprecision)
	(downgraded for: imprecision)WomenLIBFR x RTSMD (-0.076);95% CI (-0.410 to 0.258);I² = 0%;3 studies;138 participantsCERTAINTY OF EVIDENCE: MODERATE(downgraded for: heterogeneity)MenLIBFR x LIRTSMD (0.363);95% CI (-0.256 to 0.981);CERTAINTY OF EVIDENCE: LOW
Li et al²¹	LIBRF x LIRTSMD (-0.09);95% CI (-0.43 to 0.24);I² = 0%;4 studies;139 participants (67 LIBFR and 72 LIRT)CERTAINTY OF EVIDENCE: MODERATE—Critical(downgraded for: publication bias)LIBFR x HIRTSMD (0.11);95% CI (-0.41 to 0.63);I² = 0%; 2 studies;57 participants (28 LIBFR and 29 HIRT)CERTAINTY OF EVIDENCE: MODERATE—Critical(downgraded for: publication bias)	LIBFR x LIRTSMD 0.75; 95% CI (0.40 to 1.10)^a;I² = 56%;3 studies; 139 participants (67 LIBFR and 72 LIRT)CERTAINTY OF EVIDENCE: LOW—Critical(downgraded for: inconsistency, imprecision and publication bias)LIBFR x HIRTSMD (-0.41);95% CI (-0.94 to 0.13);I² = 40%; 2 studies;57 participants (28 LIBFR and 29 HIRT)CERTAINTY OF EVIDENCE: MODERATE—Critical(downgraded for: publication bias)
NITZSCHE et al⁴¹	LIBFR x LIRTSMD (-0.04);95% CI (-0.42 to 0.33);I² = 0;3 studies; 113 participants (54 LIBFR and 59 LIRT)CERTAINTY OF EVIDENCE: MODERATE—Critical(downgraded for: publication bias)LIBFR x HIRTSMD (0.10);95% CI (-0.39 to 0.59);I² = 0;2 studies;66 participants (33 LIBFR and 33 HIRT)CERTAINTY OF EVIDENCE: MODERATE—Critical(downgraded for: publication bias)	LIBFR x LIRTknee extension strength in CKCSMD (0.23);95% CI (-0.14 to 0.60);I² = 0%;3 studies; 113 participants (54 LIBFR and 59 LIRT)CERTAINTY OF EVIDENCE: MODERATE—Critical(downgraded for: publication bias)Knee extensor strength in OKCSMD (0.25); 95% CI (-0.32 to 0.81); I² = 56%; 3 studies; 113 participants (54 LIBFR and 59 LIRT)CERTAINTY OF EVIDENCE: LOW—Critical(downgraded for: inconsistency and publication bias)LIBFR x HIRTknee extension strength in CKCSMD (-0.35); 95% CI (-1.06 to 0.36); I² = Not Applicable; 1 study; 32 participants (16 LIBFR and 16 LIRT)Knee extensor strength in OKCSMD (0.26); 95% CI (-0.78 to 1.30); I² = 77%; 2 studies; 66 participants (33 LIBFR and 33 HIRT)CERTAINTY OF EVIDENCE: LOW—Critical (downgraded for: inconsistency and publication bias)				LIBFR x LIRTSDM (0.25); 95% IC (-0.44 to 0.94);I² = Not Applicable;1 study; 32 participants (16 LIBFR and 16 LIRT)LIBFR x HIRTSMD (-0.11);95% CI (-0.80 to 0.58);I² = Not Applicable; 1 study; 32 participants (16 LIBFR and 16 LIRT)
DOS SANTOS et al⁴³		LIBFR x LIRTSMD (5.24);95% CI (3.57 to 6.91); I² = Not Applicable; 1 study; 32 participants (16 LIBFR and 16 LIRT)LIBFR x HIRTSMD (0.14); 95% CI (-0.25 to 0.54); I² = 56%; 3 studies; 101 participants (49 LIBFR and 52 HIRT)CERTAINTY OF EVIDENCE: LOW—Critical(downgraded for: inconsistency and publication bias)LIBFR (+ 23%; ES = 0.86; P = .23)LIRT (+ 7%; ES = 0.21; P = .23); 1 study		LIBFR x HIRTSMD (0.01);95% CI (-0.38 to 0.40);I² = 0%; 3 studies; 101 participants (49 LIBFR and 52 HIRT)CERTAINTY OF EVIDENCE: MODERATE—Important(downgraded for: publication bias)		LIBFR x HIRTMD (-56.42);95% CI (-188.35 to 75.51); I² =Not Applicable;1 study; 32 participants (16 LIBFR and 16 LIRT)HIRT (+ 8%; ES = 0.54; P < .0001) ^a LIBFR (+ 7%; ES = 0.39; P < .0001) ^a
Cuyul-Vásques et al¹³	LIBFR x RTSMD (-0.46);95% CI (-1.21 to 0.28);I² = 80%;4 studies;153 participants (74 BFR and 79 RT)CERTAINTY OF EVIDENCE: LOW (Critical)(downgraded for: inconsistency and imprecision)			LIBFR x RTSMD (-0.23);95% CI (-0.71 to 0.26);I² = 0%; 2 studies;66 participants;CERTAINTY OF EVIDENCE: LOW (Important)(downgraded for: risk of bias and imprecision)
Ferlito et al⁴	LIBFR x LIRTMD (-1.70);95% CI (-8.65 to 5.24);I² = 13%;2 studies;81 participants (38 BFR and 43 LIRT)CERTAINTY OF EVIDENCE: MODERATE—Critical(downgraded for: publication bias)	LIBFR x HIRTSMD (-0.00);95% CI (-0.54 to 0.54);I² = 57%;3 studies;125 participants (61 BFR and 64 HIRT)CERTAINTY OF EVIDENCE: LOW (Important)(downgraded for: inconsistency and publication bias)		LIBFR x LIRTSMD (0.01);95% CI (-0.42 to 0.44);I² = 25%;2 studies;116 participants (59 LIBFR and 57 LIRT)CERTAINTY OF EVIDENCE: MODERATE—Important(downgraded for: publication bias)LIBFR x HIRTSMD (-0.20);95% CI (-0.45 to 0.06);I² = 0%;3 studies;245 participants (118 BFR and 127 HIRT)CERTAINTY OF EVIDENCE: MODERATE—Important(downgraded for: publication bias)
Van Cant et al⁴⁴	LIBFR x LIRTSMD (-0.84);95% CI (-1.74 to 0.06);I² = 81%;3 studies;114 participants (55 LIBFR and 59 LIRT).CERTAINTY OF EVIDENCE: LOW (Critical)(downgraded for: inconsistency and publication bias)LIBFR x HIRTSMD (0.33); 95% CI (-0.15 to 0.82); I² = 0%; 2 studies; 66 participants (33 LIBFR and 33 HIRT)CERTAINTY OF EVIDENCE:MODERATE—Critical (downgraded for: publication bias)	LIBFR x LIRTSMD (0.59); 95% CI (-1.08 to 2.27); I² = 96%; 3 studies; 228 participants (110 LIBFR and 118 LIRT)CERTAINTY OF EVIDENCE: VERY LOW (Critical) (downgraded for: inconsistency and publication bias)LIBFR x HIRTSMD (-0.09); 95% CI (-1.04 to 0.86); I² = 81%; 2 studies; 98 participants (49 LIBFR and 49 HIRT)CERTAINTY OF EVIDENCE: VERY LOW (Critical) (downgraded for: inconsistency and publication bias)		LIBFR x LIRTSMD (-0.45);95% CI (-1.28 to 0.39);I² = 92%;2 studies;316 participants (159 LIBFR and 157 LIRT)CERTAINTY OF EVIDENCE: VERY LOW (Important) (downgraded for: inconsistency and publication bias)	LIBFR x HIRTSMD (0.00); 95% CI (-0.46 to 0.46); I² = 61%; 2 studies; 196 participants (98 LIBFR and 98 HIRT)CERTAINTY OF EVIDENCE: LOW (Important)(downgraded for: inconsistency and publication bias)	LIBFR x LIRTSMD (0.81); 95% CI (0.10 to 1.52)^a; I² = 52%; 2 studies; 72 participants (35 LIBFR and 37 LIRT)CERTAINTY OF EVIDENCE: LOW (Important)(downgraded for: inconsistency, imprecision and publication bias)LIBFR x HIRTSMD (-0.27); 95% CI (-0.96 to 0.43); I² = Not Applicable; 1 study; 32 participants (16 LIBFR and 16 HIRT)
Hughes et al³⁶		LIBFR x LIRTMD (0.16); 95% CI (-0.27 to 0.59); I² = 0%; 2 studies; 84 participants (39 LIBFR and 44 LIRT)CERTAINTY OF EVIDENCE: MODERATE—Critical(downgraded for: publication bias)LIBFR x HIRTMD (1.045);95% CI (0.343 to 1.747);I² = Not Applicable;1 study

LIBRF (low-intensity resistance training with blood flow restriction); RT (resistance training); LIRT (low-intensity resistance training); HIRT (high-intensity resistance training); 95% CI (confidence interval); MD (mean difference); SMD (standardized mean difference); I² (heterogeneity); CKC (closed kinetic chain); OKC (open kinetic chain).

Significant effect estimate.

Subgroup Analysis

We relied on Grantham et al³ for more comprehensive subgroup assessments. Comparing LIBFR and LIRT or LIBFR and HIRT revealed no differences in pain, functional disability, or mobility/balance outcomes. The evidence ranged from low to moderate certainty. Similarly, no differences by sex were observed, although the data were limited. In both women and men with KOA, LIBFR did not differ significantly in pain reduction when compared with other forms of exercise.³

Sensitivity Analysis

Sensitivity analyses comparing Grantham et al³ and Wang et al³⁹ against other reviews found that most effect estimates aligned with these 2 key reviews. Differences in selected primary studies or weighting did not significantly alter the conclusions. Removing outlier studies⁴⁸ affected heterogeneity but did not change the direction or statistical significance of results.

While some reviews^21,44 demonstrated more favorable outcomes of LIBFR on muscle strength and hypertrophy compared with LIRT; Grantham et al³ and Wang et al³⁹ reported more neutral effects. For adverse events, Wang et al³⁹ suggested fewer complications in LIBFR compared with HIRT.

Discussion

Summary of Main Findings

This overview synthesized evidence from systematic reviews evaluating LIBFR in individuals with KOA. Across reviews, LIBFR generally appeared feasible and capable of producing measurable improvements in muscle-related outcomes—particularly quadriceps strength and, in some contexts, hypertrophy-related proxies—when compared with LIRT. In contrast, the superiority of LIBFR over comparator exercise approaches for pain relief and functional outcomes was not consistently demonstrated. The overall interpretability of these findings is constrained by heterogeneity in intervention protocols, short intervention windows in many primary trials, variability in review methods and eligibility criteria, and incomplete harm reporting.^3,4,39,44

Interpreting the Dissociation Between Strength Gains and Pain/Function Outcomes

A key clinical and methodological observation is the recurrent dissociation between more responsive strength outcomes and less consistent pain and function outcomes. This pattern is plausible because muscle capacity is a proximal physiological target of LIBFR, whereas pain and disability in KOA are multifactorial and influenced by joint irritability, peripheral and central sensitization processes, psychosocial context, sleep and mood, habitual activity patterns, and overall load tolerance. Consequently, LIBFR may generate neuromuscular adaptations that are detec on objective testing (eg, dynamometry) without necessarily modifying the broader determinants that sustain pain and functional limitation.^1,3,4,39

In addition, functional outcomes are task-dependent. Improvements in quadriceps strength assessed under controlled conditions do not guarantee transfer to activities such as stair negotiation, sit-to-stand, or prolonged walking unless training includes task-specific progression, graded exposure to functional loading, and concomitant gains in confidence and movement tolerance. Where comparator programs are well structured, pain and function may improve in both arms due to shared “active ingredients” of exercise participation, leaving limited room for incremental between-group differences. Finally, patient-reported outcomes typically show higher within-person variability and contextual influences, which—combined with heterogeneous LIBFR protocols—can dilute between-group contrasts even when objective physiological changes are present.

Short Intervention Duration and Timing Effects

Many KOA exercise trials—and therefore the reviews that summarize them—use relatively short intervention periods. This timing may favor detection of early changes in strength (including neural adaptation and test familiarization) while underestimating changes in pain, disability, and performance-based function that may require sustained reductions in irritability, improved load tolerance, and repeated successful exposure to meaningful tasks. Longer-term trials with adequate follow-up are essential to determine whether strength gains with LIBFR translate into durable functional improvements, reduced symptom fluctuations, or improved participation-level outcomes.^3,4,39

Protocol Variability and Challenges in Defining “Optimal” Parameters

Substantial variability in LIBFR protocols remains a central barrier to synthesis and clinical translation. Reviews differ in how occlusion pressure is prescribed and reported, cuff width and materials, continuous vs intermittent occlusion, exercise selection, set-rep schemes, progression rules, supervision, and total training dose. Such heterogeneity limits the ability of an overview of reviews to infer “optimal” parameters, particularly when underlying trials are small and diverse. Rather than supporting a single best protocol, the current evidence more reliably supports a pragmatic conclusion: LIBFR can be used to deliver a meaningful muscle stimulus at low external loads, potentially appealing when higher-load resistance training is poorly tolerated. Future work should prioritize consensus around a minimum reporting set, enabling dose-response analyses and more coherent pooling across studies.^36,44,50

Heterogeneity Across Systematic Reviews and Implications for Confidence

Between-review heterogeneity also contributes to inconsistent conclusions. Reviews often differ in KOA case definitions, disease severity, co-interventions, comparator selection, outcome selection and time points, and risk-of-bias handling. Differences in review methods can lead to apparently conflicting findings despite a partially shared primary-trial base. To manage this, our synthesis separated aims: reviews with mixed clinical populations were retained for descriptive mapping of protocol implementation, whereas meta-analyses and GRADE were informed by KOA-specific data only. This approach minimizes indirectness in the core inferences relevant to KOA while still capturing how the broader literature applies LIBFR across clinical contexts.^3,4,39

Safety, Tolerability, and Limitations of Adverse-Event Reporting

Safety reporting remains an important limitation in the evidence base. While several reviews describe LIBFR as generally well tolerated, harms are often reported inconsistently, and cuff-specific adverse events are not uniformly defined, captured, or linked to occlusion pressure and cuff characteristics. Across KOA-specific trials, reported events were predominantly musculoskeletal (exercise-induced knee pain) and tolerability-related (cuff discomfort/intolerance), and LIBFR was associated with fewer adverse events than HIRT. Serious thrombotic or major hemodynamic events were not signaled; however, higher-risk patients are typically excluded and rare events cannot be ruled out. This limits the ability to evaluate pressure-tolerability relationships and to establish safety margins across different protocols. More rigorous harm reporting is needed, including standardized definitions, systematic collection, and transparent reporting by group and by pressure prescription.^39,50

Implications for Practice

From a clinical perspective, the current evidence supports LIBFR as an option to facilitate strength-oriented training at low external loads, which may be relevant when patients cannot tolerate heavier loading. However, given that superiority for pain and function is not consistently observed and overall certainty is frequently low, LIBFR should be viewed primarily as an alternative dose-delivery strategy rather than a universally superior approach for symptom control. When LIBFR is used, clinicians should emphasize individualized prescription, careful monitoring of discomfort and cuff-related symptoms, and integration with task-specific functional training and graded exposure strategies aimed at improving load tolerance and participation-level function.^1,3,4,39

Implications for Research

Future trials should extend intervention duration and follow-up; standardize and transparently report LIBFR parameters; improve adverse-event collection and reporting (including cuff-specific events and tolerability by pressure level); and examine clinically relevant subgroups using adequately powered designs. Harmonized reporting standards would substantially improve the ability of future syntheses to address protocol optimization and safety margins.^39,50

Limitations

This overview is limited by the low to very low certainty of evidence for most outcomes, driven by risk of bias, imprecision, and inconsistency across primary trials summarized within the included systematic reviews. Although some included reviews enrolled mixed populations, we minimized indirectness in our core inferences by restricting quantitative synthesis and GRADE to KOA-specific data; thus, mixed-population reviews primarily informed the descriptive mapping of protocol variability rather than the certainty of pooled effects in KOA. Primary-study overlap across reviews was moderate (CCA), which may reduce the independence of the evidence base and can overstate the impression of replication at the review level. Intervention protocols varied substantially (eg, cuff width, pressure prescription/reporting, load, and progression), limiting any ability to define optimal parameters. Finally, adverse events—particularly cuff-specific events and pressure-related tolerability—were inconsistently reported, constraining conclusions regarding safety and dose-response relationships.^33,39,50

Conclusion

Low-intensity blood flow restriction training may improve quadriceps strength in individuals with KOA. Effects on pain and physical function appear broadly similar across low-intensity exercise with or without blood flow restriction and higher-intensity resistance training, but confidence in these estimates is limited by low- or very low-certainty evidence and substantial heterogeneity in protocols and populations. Compared with HIRT, LIBFR may be associated with fewer adverse events, largely reflecting less exercise-induced knee pain, although cuff discomfort can occur and evidence is insufficient to exclude rare serious events. Future trials should use standardized and transparently reported blood flow restriction parameters, consistent adverse-event definitions, and longer follow-up to clarify patient-important benefits and safety.

Supplemental Material

sj-docx-1-amd-10.1177_11795441261446451 – Supplemental material for Effects of Blood Flow Restriction Therapy in Individuals With Knee Osteoarthritis: Overview of Systematic Reviews

Supplemental material, sj-docx-1-amd-10.1177_11795441261446451 for Effects of Blood Flow Restriction Therapy in Individuals With Knee Osteoarthritis: Overview of Systematic Reviews by Felipe A. Machado, Gustavo J. Almeida, Francisco R. A. Santos, Graziella F. B. Cipriano, Gerson Cipriano, Jefferson R. Cardoso and Wagner R. Martins in Clinical Medicine Insights: Arthritis and Musculoskeletal Disorders

Supplemental Material

sj-docx-2-amd-10.1177_11795441261446451 – Supplemental material for Effects of Blood Flow Restriction Therapy in Individuals With Knee Osteoarthritis: Overview of Systematic Reviews

Supplemental material, sj-docx-2-amd-10.1177_11795441261446451 for Effects of Blood Flow Restriction Therapy in Individuals With Knee Osteoarthritis: Overview of Systematic Reviews by Felipe A. Machado, Gustavo J. Almeida, Francisco R. A. Santos, Graziella F. B. Cipriano, Gerson Cipriano, Jefferson R. Cardoso and Wagner R. Martins in Clinical Medicine Insights: Arthritis and Musculoskeletal Disorders

Supplemental Material

sj-docx-3-amd-10.1177_11795441261446451 – Supplemental material for Effects of Blood Flow Restriction Therapy in Individuals With Knee Osteoarthritis: Overview of Systematic Reviews

Supplemental material, sj-docx-3-amd-10.1177_11795441261446451 for Effects of Blood Flow Restriction Therapy in Individuals With Knee Osteoarthritis: Overview of Systematic Reviews by Felipe A. Machado, Gustavo J. Almeida, Francisco R. A. Santos, Graziella F. B. Cipriano, Gerson Cipriano, Jefferson R. Cardoso and Wagner R. Martins in Clinical Medicine Insights: Arthritis and Musculoskeletal Disorders

Supplemental Material

sj-docx-4-amd-10.1177_11795441261446451 – Supplemental material for Effects of Blood Flow Restriction Therapy in Individuals With Knee Osteoarthritis: Overview of Systematic Reviews

Supplemental material, sj-docx-4-amd-10.1177_11795441261446451 for Effects of Blood Flow Restriction Therapy in Individuals With Knee Osteoarthritis: Overview of Systematic Reviews by Felipe A. Machado, Gustavo J. Almeida, Francisco R. A. Santos, Graziella F. B. Cipriano, Gerson Cipriano, Jefferson R. Cardoso and Wagner R. Martins in Clinical Medicine Insights: Arthritis and Musculoskeletal Disorders

Footnotes

Acknowledgements

The authors thank the information specialist/librarian (FRAS) for guidance on the search strategy. The authors also thank colleagues who provided methodological feedback during protocol development.

ORCID iD

Felipe A. Machado

Ethical Considerations

This study is an overview of previously published systematic reviews and did not involve the collection of primary data from human participants. Therefore, ethics approval was not required.

Informed Consent

Not applicable. This study did not involve direct contact with human participants or the collection of individual-level participant data.

Consent for Publication

Not applicable.

Author Contributions

Felipe A. Machado: Conceptualization; Methodology; Software; Data curation; Investigation; Validation; Formal analysis; Supervision; Visualization; Project administration; Writing—original draft; Writing—review & editing.

Gustavo J. Almeida: Conceptualization; Methodology; Software; Data curation; Investigation; Validation; Formal analysis; Supervision; Writing—original draft; Writing—review & editing.

Francisco R. A. Santos: Data curation.

Graziella F. B. Cipriano: Conceptualization; Methodology; Supervision; Writing—original draft; Writing—review & editing.

Gerson Cipriano Jr: Conceptualization; Methodology; Software; Investigation; Validation; Formal analysis; Supervision; Writing—original draft; Writing—review & editing.

Jefferson R. Cardoso: Methodology; Software; Validation; Formal analysis; Visualization; Writing—review & editing; Writing—original draft.

Wagner R. Martins: Conceptualization; Methodology; Software; Data curation; Investigation; Validation; Formal analysis; Supervision; Writing—original draft; Writing—review & editing.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

All data extracted and synthesized in this overview are available within the article and its supplementary materials.

Systematic Review Registration

PROSPERO CRD42022367209.

Supplemental Material

Supplemental material for this article is available online.

References

Kolasinski

Neogi

Hochberg

, et al. 2019 American College of Rheumatology/Arthritis Foundation Guideline for the management of osteoarthritis of the hand, hip, and knee. Arthritis Care Res. 2020;72:149-162.

Duarte

Santos

dos Rodrigues

, et al. Exercícios físicos e osteoartrose: uma revisão sistemática. Fisioter em Mov. 2013;26:193-202.

Grantham

Korakakis

O’Sullivan

, et al. Does blood flow restriction training enhance clinical outcomes in knee osteoarthritis: a systematic review and meta-analysis. Phys Ther Sport. 2021;49:37-49.

Ferlito

Pecce

SAP

Oselame

, et al. The blood flow restriction training effect in knee osteoarthritis people: a systematic review and meta-analysis. Clin Rehabil. 2020;34:1378-1390.

Pazit

Jeremy

Nancy

, et al. Safety and feasibility of high speed resistance training with and without balance exercises for knee osteoarthritis: a pilot randomised controlled trial. Phys Ther Sport. 2018;34:154-163.

Cheon

Kim

Suh

, et al. Relationship between decreased lower extremity muscle mass and knee pain severity in both the general population and patients with knee osteoarthritis: findings from the KNHANES v 1-2. PLoS ONE. 2017;12:e0173036.

Glass

Torner

Frey Law

, et al. The relationship between quadriceps muscle weakness and worsening of knee pain in the MOST cohort: a 5-year longitudinal study. Osteoarthr Cartil. 2013;21:1154-1159.

Ruhdorfer

Wirth

Eckstein

Association of knee pain with a reduction in thigh muscle strength: a cross-sectional analysis including 4553 osteoarthritis initiative participants. Osteoarthr Cartil. 2017;25:658-666.

Williams

Pierre-Louis

Osteoarthritis of the knee. Physician Assist Clin. 2024;9:59-69.

10.

Shan

Suzuki

, et al. Intermediate and long-term quality of life after total knee replacement: a systematic review and meta-analysis. J Bone Jt Surg Am. 2015;97:156-168.

11.

Fransen

McConnell

Harmer

, et al. Exercise for osteoarthritis of the knee: a Cochrane systematic review. Br J Sports Med. 2015;49:1554-1557.

12.

Gibbs

Gray

Wallis

, et al. Recommendations for the management of hip and knee osteoarthritis: a systematic review of clinical practice guidelines. Osteoarthr Cartil. 2023;31:1280-1292.

13.

Cuyul-Vásquez

Leiva-Sepúlveda

Catalán-Medalla

, et al. The addition of blood flow restriction to resistance exercise in individuals with knee pain: a systematic review and meta-analysis. Brazilian J Phys Ther. 2020;24:465-478.

14.

Finnoff

Hall

Kyle

, et al. Hip strength and knee pain in high school runners: a prospective study. PM R. 2011;3:792-801.

15.

Long-Rossi

Salsich

GB.

Pain and hip lateral rotator muscle strength contribute to functional status in females with patellofemoral pain. Physiother Res Int. 2010;15:57-64.

16.

Ferraz

Gualano

Rodrigues

, et al. Benefits of resistance training with blood flow restriction in knee osteoarthritis. Med Sci Sports Exerc. 2018;50:897-905.

17.

American College of Sports Medicine. Progression models in resistance training for healthy adults. Med Sci Sports Exerc. 2009;41:687-708.

18.

Barber-Westin

Noyes

FR.

Blood flow–restricted training for lower extremity muscle weakness due to knee pathology: a systematic review. Sport Heal. 2019;11:69-83.

19.

Ogasawara

Loenneke

Thiebaud

, et al. Low-load bench press training to fatigue results in muscle hypertrophy similar to high-load bench press training. Int J Clin Med. 2013;04:114-121.

20.

Schoenfeld

Peterson

Ogborn

, et al. Effects of low- vs. High-load resistance training on muscle strength and hypertrophy in well-trained men. J Strength Cond Res. 2015;29:2954-2963.

21.

Shaharudin

Kadir

MRA

. Effects of blood flow restriction training on muscle strength and pain in patients with knee injuries: a meta-analysis. Am J Phys Med Rehabil. 2021;100:337-344.

22.

Baker

Stannard

Duren

, et al. Does blood flow restriction therapy in patients older than age 50 result in muscle hypertrophy, increased strength, or greater physical function? a systematic review. Clin Orthop Relat Res. 2020;478:593-606.

23.

Álvarez

Santamaría

PIK

Fernández-Matías

, et al. Comparison of blood flow restriction training versus non-occlusive training in patients with anterior cruciate ligament reconstruction or knee osteoarthritis: a systematic review. J Clin Med. 2021;10:1-23.

24.

Centner

Mauch

Paul

, et al. Evidence-based effects of blood flow restriction training in the rehabilitation of knee osteoarthritis and anterior cruciate ligament reconstruction: a systematic review. Sport Orthop Traumatol. 2020;36:131-142.

25.

Becker

Oxman

. Overviews of reviews. In: Higgins

JPT

Thomas

Chandler

Cumpston

Page

MJ WV

, eds. Cochrane Handbook for Systematic Reviews of Interventions: Cochrane Book Series. Wiley:607-631.

26.

Silva

Grande

Martimbianco

ALC

, et al. Overview de revisões sistemáticas—um novo tipo de estudo. parte i: Por que e para quem? Sao Paulo Med J. 2012;130:398-404.

27.

Silva Antonio José Grande Valter Pedrosa

, et al. Overview de revisões sistemáticas. 2014;19:29-41.

28.

Cumpston

Page

, et al. Updated guidance for trusted systematic reviews: a new edition of the Cochrane Handbook for Systematic Reviews of Interventions. Cochrane Database Syst Rev. 2019;10:ED000142.

29.

Page

McKenzie

Bossuyt

, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ. 2021;372:n71.

30.

Luyten

Denti

Filardo

, et al. Definition and classification of early osteoarthritis of the knee. Knee Surgery, Sport Traumatol Arthrosc. 2012;20:401-406.

31.

Mahmoudian

Lohmander

Jafari

, et al. Towards classification criteria for early-stage knee osteoarthritis: a population-based study to enrich for progressors. Semin Arthritis Rheum. 2021;51:285-291.

32.

Migliore

Scirè

Carmona

, et al. The challenge of the definition of early symptomatic knee osteoarthritis: a proposal of criteria and red flags from an international initiative promoted by the Italian Society for Rheumatology. Rheumatol Int. 2017;37:1227-1236.

33.

Pieper

Antoine

Mathes

, et al. Systematic review finds overlapping reviews were not mentioned in every other overview. J Clin Epidemiol. 2014;67:368-375.

34.

Shea

Reeves

Wells

, et al. AMSTAR 2: a critical appraisal tool for systematic reviews that include randomised or non-randomised studies of healthcare interventions, or both. BMJ. 2017;358:j4008.

35.

Meader

King

Llewellyn

, et al. A checklist designed to aid consistency and reproducibility of GRADE assessments: development and pilot validation. Syst Rev. 2014;3:82.

36.

Hughes

Paton

Rosenblatt

, et al. Blood flow restriction training in clinical musculoskeletal rehabilitation: a systematic review and meta-analysis. Br J Sports Med. 2017;51:1003-1011.

37.

Reina-Ruiz

ÁJ

Martínez-Cal

Molina-Torres

, et al. Effectiveness of blood flow restriction on functionality, quality of life and pain in patients with neuromusculoskeletal pathologies: a systematic review. Int J Environ Res Public Health. 2023;20:1-25.

38.

Reina-Ruiz

ÁJ

Galán-Mercant

Molina-Torres

, et al. Effect of blood flow restriction on functional, physiological and structural variables of muscle in patients with chronic pathologies: a systematic review. Int J Environ Res Public Health. 2022;19:1-27.

39.

Wang

Chen

Cheng

, et al. Efficacy and safety of blood flow restriction training in patients with knee osteoarthritis: a systematic review and meta-analysis. Arthritis Care Res (Hoboken). 2022;74:89-98.

40.

De Araújo

Bittar

Pinto

, et al. Effect of resistance training whit blood flow restriction on muscle strength and functional capacity of clinical populations: a systematic review. Motricidade. 2022;18:323-329.

41.

Nitzsche

Stäuber

Tiede

, et al. The effectiveness of blood-flow restricted resistance training in the musculoskeletal rehabilitation of patients with lower limb disorders: a systematic review and meta-analysis. Clin Rehabil. 2021;35:1221-1234.

42.

Pitsillides

Stasinopoulos

Mamais

Blood flow restriction training in patients with knee osteoarthritis: systematic review of randomized controlled trials. J Bodyw Mov Ther. 2021;27:477-486.

43.

Dos Santos

do Espírito Santo

Ramis

, et al. The effects of resistance training with blood flow restriction on muscle strength, muscle hypertrophy and functionality in patients with osteoarthritis and rheumatoid arthritis: a systematic review with meta-analysis. PLoS ONE. 2021;16:e0259574.

44.

Van Cant

Dawe-Coz

Aoun

, et al. Quadriceps strengthening with blood flow restriction for the rehabilitation of patients with knee conditions: a systematic review with meta-analysis. J Back Musculoskelet Rehabil. 2020;33:529-544.

45.

Clarkson

May

Warmington

. Chronic blood flow restriction exercise improves objective physical function: a systematic review. Front Physiol. 2019;10:1058.

46.

Bryk

dos Reis

Fingerhut

, et al. Exercises with partial vascular occlusion in patients with knee osteoarthritis: a randomized clinical trial. Knee Surg Sports Traumatol Arthrosc. 2016;24:1580-1586.

47.

Harper

Roberts

Layne

, et al. Blood-flow restriction resistance exercise for older adults with knee osteoarthritis: a pilot randomized clinical trial. J Clin Med. 2019;8:265.

48.

Segal

Davis

Mikesky

AE.

Efficacy of blood flow-restricted low-load resistance training for quadriceps strengthening in men at risk of symptomatic knee osteoarthritis. Geriatr Orthop Surg Rehabil. 2015;6:160-167.

49.

Segal

Williams

Davis

, et al. Efficacy of blood flow-restricted, low-load resistance training in women with risk factors for symptomatic knee osteoarthritis. PM R. 2015;7:376-384.

50.

Patterson

Hughes

Warmington

, et al. Blood flow restriction exercise position stand: considerations of methodology, application, and safety. Front Physiol. 2019;10:533. doi:10.3389/fphys.2019.00533.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.01 MB

0.03 MB