The aim of this systematic review and meta-analysis was to analyze the effect of LL-RT with BFR versus HL-RT on muscle hypertrophy. As a main result, we found that muscle hypertrophy is similar between LL-RT with BFR versus HL-RT and is consistent with the results presented in a previously published meta-analysis on the topic [8]. The previous meta-analysis encompassed 10 studies, while our main meta-analysis encompassed 23 studies. This large number of studies strengthens our ability to draw inferences on the effect of LL-RT with BFR versus HL-RT on muscle hypertrophy using a variety of repetition schemes. It is noteworthy that the effect of LL-RT with BFR versus HL-RT on muscle hypertrophy seems to be unaffected by the repetition scheme adopted in LL-RT with BFR, as shown in our subgroup analyses. However, we did observe a small beneficial effect in favor of HL-RT when using upper limb BFR exercise compared to lower limb BFR exercise.
HL-RT has been recommended for enhancement of muscle strength and size [1]. Our overall analysis indicates that LL-RT with BFR elicits a similar magnitude of muscle hypertrophy compared to HL-RT. Mean percentage gains were similar between the compared training models (5.6% versus 6.3% for LL-RT with BFR and HL-RT, respectively). These results suggest that LL-RT with BFR may be a viable alternative to HL-RT to elicit muscle hypertrophy in healthy populations ranging from young to older adults. While the mechanisms underpinning the hypertrophic effects of low loads with BFR are still not fully elucidated, there appears to be some relationship with acute muscular fatigue and long-term muscle hypertrophy. Considering that LL-RT with BFR restricts arterial inflow and occludes venous return, exercise with BFR prevents the escape of metabolites from the working muscles, inducing earlier fatigue and thus conceivably increasing recruitment of motor units to maintain muscle strength levels [45]. This mechanism is theorized to be responsible for the higher myoelectric activity in low-load resistance exercise with BFR compared to low-load resistance exercise without BFR (work-matched) [46]. On the other hand, a recent meta-analysis found that LL-RT with BFR elicits lower myoelectric activity than HL-RT [47].
Lower myoelectric activity in LL-RT with BFR can be justified by motor unit cycling referring to the fact that during exercise with lower loads, some motor units are activated and deactivated to minimize fatigue, reducing the need for all motor units to be activated at the same time [48]. It should be noted that surface electromyography amplitudes do not exclusively reflect the recruitment of motor units during exercise [49]; thus, inferences regarding recruitment must be drawn with caution.
It has been hypothesized that training to concentric failure elicits greater recruitment of motor units and, consequently, enhances muscle hypertrophy [50]. However, this claim may be load-dependent. Training close to concentric muscle failure may be necessary to maximize muscle hypertrophy in LL-RT, but not necessarily in HL-RT [4]. Lixandrão et al. [7] found that LL-RT with BFR elicited significantly less quadriceps femoris hypertrophy than HL-RT, using low and high BFR pressures (40% and 80% AOP, respectively) and an arbitrary repetition scheme. On the other hand, Jesse et al. [16] found similar hypertrophy of the quadriceps femoris between HL-RT using the same BFR pressures but with sets carried out to momentary muscle failure. Jesse et al. [16] proposed that such a discrepancy could be explained by the repetition schemes prescribed in each intervention (3 sets of 15 repetitions versus sets to momentary muscle failure).
Based on the premise that the repetition scheme adopted in the intervention could potentially influence muscle hypertrophy elicited by LL-RT with BFR, we chose to stratify our analyses based on the repetition scheme adopted in LL-RT with BFR. We categorized repetition schemes into three subgroups, including two that are recommended for the prescription of LL-RT with BFR; that is, sets carried out to momentary muscle failure and a fixed repetition scheme composed of 75 repetitions performed across 4 sets (30-15-15-15) [39]. The third repetition scheme analyzed was composed of multiple sets of 15 repetitions. In all subgroups investigating repetition schemes, no statistical differences in muscle hypertrophy were observed between LL-RT with BFR and HL-RT.
In sets carried out to momentary muscle failure, LL-RT with BFR elicited a mean percentage increase of 5.2%, while HL-RT elicited a mean percentage increase of 5.7%. Similar results were observed in the analysis of the studies that employed a protocol of 75 repetitions in LL-RT with BFR (5.5% and 6.3% for LL-RT with BFR and HL-RT, respectively), as well as in the analysis that employed sets of 15 repetitions in LL-RT with BFR (5.3% and 5.8% for LL-RT with BFR and HL-RT, respectively). These results suggest that the muscle hypertrophy elicited by LL-RT with BFR is not necessarily dependent of the repetition scheme. In support of this hypothesis, Martín-Hernández et al. [51] reported that the prescription of 75 or 150 repetitions (twice the repetition volume traditionally prescribed in practice) in LL-RT with BFR promotes muscle hypertrophy similar to HL-RT, with no difference between the different volumes of repetitions tested in BFR conditions. It has been speculated that after the muscle reaches a certain level of fatigue during exercise with BFR, increasing number of repetitions is not of great relevance for muscle hypertrophy, suggesting the existence of a ceiling effect [52]. Possibly, a considerable level of fatigue can be experienced with multiple sets of 15 repetitions and loads of 20-40% of 1RM with applied BFR pressures of 40-80% of AOP. It is important to note that arbitrary repetition schemes (e.g., 3 x 10 or 3 x 20 repetitions) have been extensively used in the strength training literature with- and without LL-RT with BFR to investigate resultant hypertrophy. Thus, despite the fact that most of our included papers lacked volume-equated schemes, the repetition protocol adopted in these studies have ecological validity.
In addition, to identify potential region-specific changes in muscle growth between HL-RT and LL-RT, we introduced a subgroup analysis that considered the muscle group evaluated (upper limb and lower limb). Interestingly, we found a small, but statistically significant difference favoring HL-RT in the upper limb muscles while no differences between conditions were observed in muscle hypertrophy in the lower limbs. It is important to highlight that there is a paucity of studies comparing the effects of HL-RT and LL-RT with BFR on upper body hypertrophy (n = 4; 10 comparisons) in our meta-analysis. Moreover, only one of these studies adopted a personalized pressure (%AOP) [10] yet used a load below what is recommended (15% 1-RM) to induce similar muscle growth as HL-RT [44]. In addition, two studies used arbitrary pressures (100-160 mmHg). We speculate that the limited number of comparisons in the upper limb subgroup analysis may have contributed to the small beneficial effect observed in favor of HL-RT and the results would likely be attenuated if the number of comparisons were similar to lower limb subgroup (n = 54; 20 studies). More research comparing the hypertrophic response between LL-RT with BFR and HL-RT during upper limb exercise is needed given our findings.
This review has some notable strengths. A relatively large number of studies were included (n = 23) in the principal meta-analysis and a low-level of inconsistency was reported in all analyses performed. It is worth adding that all studies included in this review investigated local muscle growth through images obtained by ultrasound, magnetic resonance imaging, and peripheral computed tomography, improving the sensitivity to identify subtle changes in muscle mass [53]. Nevertheless, the present study has some limitations that need to be highlighted: (i) most of the measurements were taken at a single site along the length of the muscle, which may not reflect hypertrophic changes throughout the entire muscle; (ii) none of the included studies had a low risk of bias; (iii) only one of the studies included in our analyses included trained individuals [17] and the interventions ranged from 6 to 16 weeks; therefore, our results cannot be extrapolated to trained individuals or long-term adaptations.
PRACTICAL APPLICATIONS
As increased proximity to failure heightens the perceptual experiences of the exerciser regardless of the application of BFR [54], it can be assumed that long-term adherence to repetition schemes further away from failure would be greater than those repetition schemes exercising with a greater volume of repetitions or to momentary muscular failure, although this requires further research. If hypertrophy is similar between different LL-RT with BFR repetition schemes compared to HL-RT, this may have important implications for injured individuals rehabilitating from injuries whose tolerance to strenuous exercise is reduced and LL-RT with BFR is recommended.
Tolerance to LL-RT with BFR has been labeled as a barrier to long-term compliance to the intervention [24]. Therefore, reducing the required number of repetitions needed to induce positive musculoskeletal benefit seems important. Patterson et al. [44] recommends sets either be carried out to momentary muscular failure or performed for 75 repetitions over 4 sets. The current systematic review with meta-analysis indicates that a smaller repetition volume (e.g., multiple sets of 15) may induce similar hypertrophic effects as HL-RT.