Effect of walking with blood flow restriction in elderly women with osteoporosis/osteopenia

Abstract Introduction The preservation of bone mass in elderly women is associated with better levels of practice of systematic physical exercises. Aerobic training combined with blood flow restriction seems to be a new alternative that determines this process, but knowledge gaps are still observed when referring to exercise associated with blood flow restriction (BFR) and adaptations on bone variables. Objective To analyze the chronic effects of aerobic training with and without BFR on bone mineral density and bone biomarker osteocalcin concentrations in older women. Methods Thirty women were randomized into the following groups: walking on a treadmill at low intensity with BFR; moderate treadmill walking with no BFR; only BFR (no exercise) for 20 minutes, twice a week, for 24 weeks. Bone mineral density was measured before and 24 weeks after intervention. Blood serum osteocalcin concentrations were measured before, 12 and 24 weeks after intervention. Results There were no differences between groups in bone mineral density (femoral neck, p = 0.31; total femur, p = 0.17; lumbar spin, p = 0.06) and osteocalcine (W(2) = 0.27; p = 0.87) ouctomes after 24 weeks of intervention. Conclusion There was no difference between walking training, blood flow restriction only, or walking+blood flow restriction on bone mineral density and osteocalcin concentrations after 24-weeks of intervention in older women with osteopenia/osteoporosis.


Introduction
The reduction of estrogen production, characteristic of the post-menopausal life stage, accelerates the reduction in bone mineral density (BMD) that makes women more susceptible to postmenopausal osteoporosis and to an increased risk for fracture.This fact constitutes a public health problem, with high rates of morbidity and mortality in the older people. 1 The analysis of bone biomarkers contributes to earlier information about bone remodeling compared to the changes observed from a bone densitometry scan.
As a result, biochemical markers are more sensitive to the effects of physical exercise on bone formation and resorption in both acute or chronic training protocols and are therefore considered more effective to examine the effects of exercise on bone remodeling.2,3 The American College of Sports Medicine recommends several types of exercise for the preservation of bone health in adulthood, especially in women after menopause.4 Several studies have observed the effect of different exercises (aerobic, high impact, strength), 5-7 or their combination, on BMD in elderly women, however, bone mass gains are less than 2% for the lumbar spine.8 Aerobic training, without any other intervention, has little or no effect on BMD of the spine and the femoral neck.In a meta-analysis, Palombaro 9 recommends walking, or any low impact exercise, combined with other forms of training to preserve bone mass in women after menopause.The implementation of high intensity exercise programs is not feasible in some situations, for instance for older persons with joint restrictions such as osteoarthritis, herniated discs, and vertebral fractures.10 Thus, much of the scientific community has been looking for alternatives that use low intensity exercises for such individuals to improve bone health.
In this perspective, in order to reduce load-related stress, the Japanese, nearly 50 years ago, developed Participants were excluded from the study if: 1) were unable to achieve a minimum training frequency of at least 85%; 2) were injured; or 3) if they requested to be withdrawn from the study.

Experimental design
This study was a prospective clinical trial that randomly allocated participants into one of the following experimental groups: 1) a walking group (W), which walked on a treadmill for 20 minutes at 60% of VO  for LS and TS to be considered significant was 0.03 g/ cm 2 , and 0.035 g/cm 2 for FN.23 The adapted Cirilo bench step protocol was used with manual increments in the height of the steps according to the height of each volunteer, in order to estimate VO 2 peak (ml/kg/min) indirectly and to establish the training speeds on the treadmill.24 The W+BFR group trained with a low load (40% of VO 2 max.) and the W group trained with a moderate load (60% of VO 2 max.).This test was performed pretraining, and at 1, 3, 5 and 6 months (end of the training intervention).Since the volunteers were classified as untrained, the test rhythm used was 116 steps per minute, as measured by a metronome (Tagima®, Japan).

Training protocols
The W+ BFR group trained with low load (40% of VO2 peak) and the W group trained with a moderate load (60% of VO 2 peak.).The BFR group was subjected to BFR without performing any type of exercise, twice a week, The selection of models for bone variables was based on Quase Likehood Independence Criterion -QIC.26 The normality of the residues was verified using Q-Q charts and considered plausible in each instance.
The Bonferroni post hoc test was used when a significant reason was identified for the isolated effect of the factors analyzed or for interaction between them.moderate; ρ = 0.5 -0.7: high; ρ = 0.7 -0.9: very high; ρ = 0.9 -1.0: almost perfect, 27 with a significance level of 5%, for all comparisons.

Results
The participants´ sociodemographic information and baseline of lumbar spine, femoral neck and total femur score for osteoporosis classification are presented in Table 1.There were no significant differences in BMD or osteocalcin (p > 0.05) between the groups prior to the training intervention.
As outlined on Table 2, there were no significant group versus time interactions and no significant group or time main effects for the BMD measured at the femoral neck, total femur and lumbar spine regions.

Discussion
To our knowledge this was the first study that performed a 24 week aerobic training intervention in combination with the BFR in older women with osteopenia/osteoporosis.The main findings of the present study were: 1) there were no differences between groups in BMD of LS, FN and TF; 2) there were no differences between groups in osteocalcin levels; and 3) osteocalcin levels increased at week 12 of the intervention and was accentuated following the 24 week intervention for each protocol (W+BFR, W and BFR).There was a moderate effect, following 24 weeks of training for the W+BFR group.
Hatori et al. 30 reported significant increases in BMD following 28 weeks walking in a similar subject cohort as that in the present study.However, the walking intensity was higher than that in our study, being above the anaerobic threhsold.
Martyn-St James and Carroll 19 observed in their metaanalysis that there was a low, but significant effect of walking, when performed daily, on BMD of the femoral neck (0.014 g/cm 2 ) but none in the lumbar spine, suggesting that other forms of exercise, which provide greater load, may be necessary to preserve or increase BMD in women after menopause.
As can be seen in Figure 3, there were no differences between groups for osteocalcin concentration (W(2) = 0.27; p = 0.87) following the training intervention.In the present study, although there were no significant gains, there was a maintenance of bone mass at the three sites (LS, FN and TF) in every intervention group after 24 weeks.This is notable since the W+BFR and BFR groups used low loads (40% VO 2 peak) and maintained bone mass to the same extent as the W group, which trained with moderate intensity (60% VO 2 peak).The maintenance of BMD in the present study could be due to the increased activation of hypoxiainduced transcriptional factor signaling pathway and the simultaneous activation of vascular endothelial growth factor (VEGF), which induces angiogenesis, consequently, the supply of oxygen and nutrients for osteogenesis.31 Although not addressed in the present Some possible limitations in the present study are the low pressure of blood flow restriction, the short intervention time, the sample size, and the absence of a control group that had not performed any intervention.

Conclusion
The three protocols herein were effective for maintaining BMD and increasing osteocalcin following 24 weeks of the intervention, indicating that BFR is an effective alternative for the preservation of bone mass.
Further research is needed with the BFR technique combined with exercise programs longer than 24 weeks in older women and there is a need to study the blood pressure restriction ideal for osteogenesis without any risk to the health of the elderly.

Figure 1
outlines all the procedures involved in the screening process at the study entry.Sample size was determined a priori using G*Power (ES = 0.25, Power = 0.80, α = 0.05, three groups, and three measurements) following the recommendations from Faul et al. 20 and Beck.21 According to our calculations, a total sample size of 36 individuals would be required for this study.A total of 60 women volunteered for the current study, however, five participants did not meet all the inclusion criteria, nine were not able to follow the training schedules, three requested to be withdrawn from the study before initiating the training, and thirteem during training for reasons unrelated to the study procedures, leaving a final sample size of 30 participants (Figure 1).This study was approved by the Research Ethics Committee of the Health Sciences Center of the Universidade Federal da Paraíba (No. 2086608; CAAE: 67125317.1.0000.5188)and registered in the Brazilian Clinical Trials Registry Platform (RBR-3d957w).All volunteers signed an informed consent form.
walking with blood flow restriction group (W+BFR), which performed 20 minutes of walking on a treadmill at 40% of VO 2 max; and 3) a blood flow restriction group (BFR), which did not perform any exercise and only received the BFR stimuli.Participants were required to visit the laboratory for a total of 72 different occasions over the course of six months.During the first visit, participants were provided with an explanation about the study procedures, filled out standardized questionnaires, and had their ABI assessed.If the participants' ABI was within the range stated in the inclusion criteria, participants completed the next steps, which consisted of determing the total arterial occlusion pressure for the lower-body and an estimated peak oxygen consumption (VO 2 peak) test.After a period of 24 hours, body weight and height were assessed and participants completed a total body dual x-ray absorptiometry (DXA) scan, followed by a blood draw, used to determine the plasm osteocalcin baseline levels.During the third visit, participants in the W and W+BFR groups were familiarized with the training protocols.Following the first 12 weeks of training, osteocalcin was assessed and BFR training pressures were adjusted.After 24 weeks, the same analysis was completed as done after week 12, with the addition of the DXA scans (Figure 2).
Osteocalcin serum concentrations were assessed before training, following 12 weeks of training, and at the end of 24 weeks of training.Venous blood samples were collected by a nurse at 7am (±1), following an eight hour fasting period.Blood samples were allowed to clot at room temperature, then centrifuged.The plasma was separated and transferred to polystyrene tubes and frozen at -80 ºC until the final analysis.All blood samples were analyzed by the Hermes Pardini Laboratory (Vespasiano, MG, Brazil).The serum concentration of osteocalcin was assessed using the Elecsys reagent, according to a protocol (EP5 A2) from the Clinical and Laboratory Standards Institute (CLSI), on the Modular analytics E170 (ROCHE) equipment.The sensitivity of the assay used was 0.5 to 300 ng/mL.The intra and inter assay coefficient of variation for this analysis were 4.76% and 8.00%, respectively.
for 20 minutes, for six months.Volunteers in the W+BFR and BFR group used inflated pressure tourniquets on both proximal portions of the thighs, and throughout each 20 minute training session.25 The cuff pressure used during the firs month of training was 20% of total occlusion pressure (TOP), and was then incrementally increased at two months (30% TOP), three months (40% TOP), and then remained the same pressure the in the remainder of the intervention at 50% TOP.22 Statistical analyses All data were analyzed using the Statistical Package for the Social Sciences (SPSS -20.0,IBM, New York) and the results plotted in GraphPad Prism software (5.03).Initially, data normality (Shapiro-Wilk test) and homogeneity of variances (Levene test) were verified.The performance of the groups, over time, for the variables BMD (LS, FN and TF) and osteocalcin were analyzed using Generalized Estimated Equations (GEE) considering the autoregressive covariance matrix (AR-1) and the link log function with gamma distribution model.

FISIOTERAPIA
EM MOVIMENTO Physical Therapy in Movement Bittar ST et al.Fisioter Mov.2023;36:e36116 6 Associations between groups and categorical variables were verified by Fisher's exact test.Cohen's effect size (ES) was estimated to outline the differences in the means of the groups with unequal sample sizes within a pre-post-control design and interpreted as follows: d < 0.20 (trivial); d = 0.20 -0.59 (small); d = 0.60 -1.19 (moderate); d = 1.20 -1.99 (large); d = 2.00 -3.99 (very large) e; d ≥ 4.00 (almost perfect effect), according to Hopkins et al. 27 For the purposes of calculating ES, the BFR group was considered as a control group.For the interpretation of d Cohen, the probability of superiority, measured in percentage was used.28,29To verify the relationships between osteocalcin plasma levels and BMD (CL, CF, FT) following the 24 weeks of intervention, Spearman's correlation test (Rho ρ) was used, according to the following classification: ρ = 0 -0.01: very low; ρ = 0.1 -0.3: low; ρ = 0.3 -0.5:

1 )
Note: W+BRF = walking with blood flow restriction group; W = walking group; BRF = blood flow restriction group.*Data are presented as mean ± standard deviation.FISIOTERAPIA EM MOVIMENTO Physical Therapy in Movement Bittar ST et al.Fisioter Mov.2023;36:e36116 7 Trivial effects were observed for the osteocalcin concentration in the W+BFR and W groups when compared to the BFR group, with an effect size of d = 0.027 and d = 0.079 after 12 weeks of intervention and a trivial effect for both experimental groups (W+BFR: d = 0.065; W: d = 0.136) 24 weeks post-intervention.The correlations (Spearman) between BMD (LS, FN and TF) variables versus osteocalcin were not significant.
However, it was possible to observe that regardless of the group, all individuals presented values significantly greater 24 weeks post-intervention when compared to the pre-assessment (p = 0.002) and 12 weeks postintervention (p = 0.006).Thus, regardless of the type of training, positive effects on osteocalcin were observed after 12 and 24 weeks of intervention (W (2) = 10.95;p < 0.01).

Table 1 -
Sociodemographic information about the participants and baseline of lumbar spine, femoral neck and total femur score for osteoporosis classification

Table 2 -
Absolute bone mineral density values across groups before and after 24 weeks post-training 16,25,34,35r studies that used the BAP as a biomarker of bone formation.16,25,34,35However, in the present study we opted to use osteocalcin.BAP is specific for bone formation but does not eliminate the cross reaction with the hepatic isoform (15-20%), being more specific and indicative than total alkaline phosphatase, but less specific than osteocalcin.36Tosatisfy the control condition warranted by randomized clinical studies, we adjusted the effect sizes so that the probability of a participant randomly selected from the W+BFR and W training groups having a plasma osteocalcin level higher than a control group participant (BFR) was 50%.After 24 weeks of intervention, the W+BFR and W groups had a 50% and 52.82% probability, respectively.Beekley et al., 16 who aerobically trained healthy men aged 21 to 28 years for three weeks obtained a 61.14% probability with the Kaatsu training group to have a plasma BAP level higher than that of an individual in the group control.The magnitude of the effect in relation to the present study can be explained by sex and age differences.Another possible explanation for the small effect sizes in the present study could be the lower blood flow restriction pressure used in the BFR intervention groups, when compared with those in the literature.16,25 Again, the age of the subjects herein warranted a sharper control of risk with BFR.22