Effect of kettlebell training on bone mineral density in two older adults with osteoporosis: a multiple-case study from the BELL trial

1. Introduction

This study describes clinically significant increases in bone mineral density (BMD) in two older adults with osteoporosis who participated in the BELL trial ⁽¹⁾, which involved group and home-based exercise with kettlebells performed five days a week for 16 weeks. Osteoporosis is a systemic skeletal condition associated with aging, characterised by very low BMD. Osteoporosis is diagnosed using dual-energy X-ray absorptiometry (DXA) from a T-score more than 2.5 standard deviations (SD) below that of sex-matched young adults, whereas osteopenia (pre-osteoporosis) is diagnosed when the T-score is between 1 and 2.5 SDs below sex-matched young adults ⁽²⁾.

Bone metabolism deteriorates with advancing age, which can be exacerbated by inactivity, disuse, trauma, and medical conditions such as diabetes. Currently, approximately two thirds of Australians over 50 years of age have low bone mass ⁽³⁾, with 500 daily osteoporotic fractures projected to occur in Australia by 2022 ⁽⁴⁾. More than one in three women and one in five men are likely to sustain at least one fragility fracture in their lifetime ^{(5, 6)} with the cumulative risk of fracture for postmenopausal women as high as 60% ⁽⁷⁾. More than 50% of older adults who fracture a hip are no longer able to live independently and almost one in three can die within 12 months of the fracture ^{(5, 8)}. Between 2007 and 2017, the direct costs of osteoporosis are estimated to have tripled ⁽⁹⁾. Hip fractures account for the largest proportion of this increase (43%), with 74% of the fractures occurring in people over 70 years of age ⁽⁹⁾, with direct and indirect costs of osteoporosis projected to be $3.84 billion by 2022 in Australia ⁽⁴⁾. In addition to the financial cost the social burden is considerable, with fragility fractures, especially in the hip and spine, causing severe pain, disability, loss of independence, and decreased quality of life ⁽⁶⁾. Up to 70% of the variance in fracture from loading is explained by BMD ⁽¹⁰⁾ with each 1 SD reduction in BMD representing a 1.5 to 3-fold increase in risk of fracture ^{(2, 5)}. Combined with other benefits of structured exercise, such as improved muscle function and balance, a 1-2% increase in BMD via an exercise prescription may reduce fracture risk by 5-10% ⁽¹¹⁾ and reduce the incidence of fracture by up to 50% ⁽¹²⁾. Thus, research to determine optimal exercise prescription for improving BMD remains a high priority.

Exercise prescription and lifestyle modification are interventions for osteoporosis which aim to promote bone health ⁽²⁾. Fracture index is calculated from BMD T-score, so therapeutic interventions which positively influence T-score have the potential to reduce risk of fracture. Exercise may increase bone mass and bone strength, however effects from clinical trials appear to be site specific and provide only small clinically relevant improvements in BMD (<2%), and small absolute reductions in fracture risk, if at all ^{(5, 7, 12, 13)}. Although the optimal type and dose of exercise is not definitive, current consensus is that a combination of high impact and progressive whole-body resistance training (HiPRT) is the most effective for maintaining bone mass (preventing bone loss) at the femoral neck (FN) and lumbar spine (LS) in older adults with low BMD ^{(3, 14)}. Walking, cycling, swimming and even impact and resistance exercise appear to be largely ineffective for increasing BMD when performed in isolation^{(3, 15, 16)}. Underscoring the prophylactic nature of exercise, clinical guidelines recommend that exercise should be lifelong and include regular weight-bearing and high impact activities to promote muscle strength, muscle power, and maintenance of bone mass and geometry ^{(5, 7, 13, 17)}. In the osteoporosis population however, adherence rates to exercise programs are generally poor, with a recent review suggesting that 50% of participants drop out within the first 6 months. ⁽¹⁸⁾. Results from the BELL trial, which maintained an adherence rate >90% for over 12 weeks, show that resistance training programs can be performed by older adults safely and independently with positive psychosocial affect, despite training daily and the intensity reported to be “very hard” ^{(1, 19)}.

Multiple studies from single case reports ⁽²⁰⁾ to large randomised controlled trials ⁽²¹⁾, have attempted to find the most effective types of exercise and program variables to improve BMD in older adults with osteoporosis. Results have been varied, from no significant difference after 12 months of training, to extraordinarily large increases many times greater than mean change reported in meta-analyses ^{(13, 22)}. Novel methods of utilising resistance training equipment with older adults, such as kettlebells ⁽²³⁾ and weighted vests ⁽²⁴⁾, allow researchers to determine how changing program variables may influence the bone response to exercise. These investigations are valuable as exercise interventions thus far, appear to have been more effective at maintaining rather than increasing bone mass. The aim of this case study is to report the effects of a 16-week high volume hardstyle kettlebell training program on BMD and measures of healthy aging in two adults with osteoporosis over 70 years of age.

2. Methods

4. Results

Change in LS and FN BMD are shown in Table 2. Attendance rates for group sessions, compliance with prescribed home-exercise, total training load volume, and change in outcome measures pre- to post-training are shown in Table 3. Session training load volume and arbitrary units are shown in Figures 1 and 2 for Sally and Peter, respectively.

• Download figure
• Open in new tab

Fig. 1:

Sally’s 12-week training load.

• Download figure
• Open in new tab

Fig. 2:

Peter’s 12-week training load.

Sally and Peter’s change in training load are representative of the BELL trial cohort. Internal and external training load volume depict a model of undulating periodisation which steadily increased throughout the intervention period. The final two weeks training volume was prescribed to enable all participants to attain a training load ‘personal best’ on the final day of the trial. Training at home with only limited weights, Sally and Peter both recorded a “very hard” sRPE of 9/10 Friday 1 May.

Least significant change in BMD was calculated to be 0.022g/cm² for Sally and 0.330g/cm² for Peter. For Sally, an increase in BMD of 0.017 g/cm² at the FN and 0.012 g/cm² at the LS between 2 November 2017 and 21 January 2020 (before the BELL trial intervention) was below the LSC calculated from her BMI. Between 21 January 2020 and 11 August 2020 (following the intervention), a BMD increase of 0.0443 g/cm² at the LS was more than double her LSC (2.01×) and sufficient to advance her status from osteoporosis to osteopenia. In addition, Sally’s FN BMD increased by 12.7%. Between 16 December 2019 and 16 June 2020, Peter’s total lumbar spine BMD increased by 0.064 g/cm², almost double his LSC (1.94×). Peter’s femoral neck BMD also increased by 2.5%.

Sally missed four consecutive training sessions due to a bereavement, and one further home training session. Peter missed four consecutive training sessions due to sickness.

5. Discussion

This case study is the first to show clinically significant increases in bone mass in older adults with osteoporosis following 16-weeks of hardstyle kettlebell training. These results challenge the view that a combination of high intensity and impact loads are necessary for an osteogenic exercise effect ⁽³⁾. Change in outcome measures and training load volume for both subjects in this study are representative of the results from the BELL trial, which show large improvements in grip strength, and significant increases in cardiovascular and functional capacity. Although these data are not definitive evidence of a causal relationship between kettlebell training and increased bone mass in older adults with osteoporosis, these findings warrant further investigation in sufficiently powered, randomised controlled trials.

The predictive value of low BMD for hip fracture is comparable to that of high blood pressure for stroke ⁽⁵⁾. Bone metabolism deteriorates with advancing age with insufficiently active older adults typically losing 1-3% BMD every year ⁽²⁸⁾. Rates of bone loss can be exacerbated by inactivity, disuse, medical conditions, and deficiencies in calcium and vitamin D ⁽²⁹⁾. Although osteoporosis is influenced by epigenetic and environmental factors ⁽¹⁷⁾ and mediated by factors other than mechanical loading ⁽³⁰⁾, physical activity and exercise are a major extrinsic factor responsible for the maintenance and promotion of bone mass ^{(31, 32)}. Being physically active is recommended for all adults ⁽³³⁾ however, systematic reviews remain cautious of overstating the preventative potential of physical activity for preventing osteoporosis ⁽³⁴⁾.

The most recent and comprehensive review and meta-analysis to date of exercise intensity on bone in postmenopausal women, highlights the notable absence of trials using high intensity (>80% 1RM) training ^{(35, 36)}. Only six studies met the inclusion criteria of ‘high intensity’ of which only the LIFTMOR trial reported a statistically significant increase in LS BMD. Consistent with previous meta-analyses ^{(13, 22, 37, 38)} the within-group mean increase in LS BMD in the LIFTMOR trial, was 0.023g/cm² (2.9 ± 3.1%) with no significant change in FN BMD. A change of that magnitude at the lumbar spine is 0.001g/cm² above the LSC for the underweight female in the current study, and 0.01g/cm² below the LSC for the male whose BMI was only slightly above ‘normal’. Subsequent iterations of the HiPRT protocol used in the LIFTMOR trial have been underpowered to detect a change in bone mass ⁽³⁹⁾, or resulted in a mean change of smaller magnitude (0.016g/cm²) ⁽⁴⁰⁾. In the meta-analysis by Kistler-Fischbacher et al ⁽³⁶⁾ the authors reported, 1) no evidence that high intensity training was more effective at the femoral neck, 2) no significant difference in effect size between moderate and high intensity training, 3) no significant effects of impact-only exercises regardless of intensity, 4) no benefit of resistance and impact training combined compared to resistance-only training, and 5) no association between resistance training intensity (measured as a percentage of 1RM) and change in LS and FN BMD.

Results from other trials and meta-analyses have been mixed. Single trials have reported no significant difference between training frequencies ⁽⁴¹⁾, no significant difference between resistance training and aerobic training performed over two years ⁽⁴²⁾, and no significant change following three years of resistance training ⁽⁴³⁾. Outcomes have also been shown to become non-significant when non-randomised trails are excluded from analysis ⁽¹⁵⁾, likely due to the confounding associated with non-randomised trials which can almost double reported treatment effects. Other meta-analyses have reported that only studies with combined aerobic and strength training programs result in improvements likely to reach a least significant change ⁽³⁷⁾. These are consistent with other data indicating an overall treatment effect <1% ^{(15, 44)} irrespective of concurrent HRT, bisphosphonate or calcium supplementation ⁽²²⁾. Aerobic-only and impact-only programs are largely ineffective when used in isolation, but programs which combine aerobic/impact exercise with resistance training are more likely to result in a weighted mean difference of >0.02g/cm^{2 (15)}. The findings of this study suggest there are other mediators of an osteogenic effect besides impact and high intensity loads.

The BELL and LIFTMOR trials both reported improvements in functional outcomes and bone health. The LIFTMOR trial utilised a barbell deadlift, overhead press, and back squat performed at 80 to 85% 1RM, with drop landings from a chin-up bar. Participants in the LIFTMOR trial completed a total of 75 repetitions per session for 70 sessions over an 8-month training period, compared to the 60 sessions in 3 months during the BELL trial. Mean training load was not reported in the LIFTMOR trial but given the maximum number of weighted repetitions per session (75), it is likely to have been less that the 2805 251 kg per session reported in the BELL trial. Rating of perceived exertion was high in the BELL trial with all participants recording an sRPE value of 8-10/10 (“very hard” to “maximal”) on the final day, but training involved no impact loading and only three sessions in which participants may have performed kettlebell deadlifts with loads >85% 1RM. For the last 10 weeks of the training period, participants only had access to three kettlebells to train at home; Sally had 8, 12 and 24 kg, Peter had two 8s and 22 kg. A 12-month trial of resistance training with a much larger external training load than the BELL trial, found no significant relationship between training load and BMD ⁽⁴⁵⁾, casting doubt over the influence of training load volume as a mediator. A positive relationship between BMD and training volume has been reported in men, but this may be limited to impact loads only ⁽⁴⁶⁾, thus mediators of the osteogenic effect observed in the current case study remain to be determined. Some evidence does indicate that high velocity resistance exercises may have a superior effect on FN and LS BMD than performing the same movements more slowly ⁽⁴⁷⁾. In support of this view, peak net ground reaction force in older adults was greater during a two-handed hardstyle swing with an 8 kg kettlebell than a kettlebell deadlift with 32 kg ⁽⁴⁸⁾, and data from a pilot study of low load power training in older adults, shows this combination of high velocity and light weight may be sufficient to have an osteogenic effect ⁽²⁴⁾. Given the ballistic nature of the hardstyle kettlebell exercises used in the BELL trial, the high velocity of kettlebell swings compared to standard resistance training exercises may help explain the results of this study.

Most studies measuring the effects of interventions for osteoporosis rely on DXA-derived BMD as a surrogate measure of bone strength, and a proxy for primary clinical outcomes, namely fractures ^{(7, 12)}. This can be somewhat problematic because bone strength is influenced by factors other than BMD, which the planar nature of DXA is not ideally suited to measure, such as size, shape, and structure. Furthermore, small changes in cross-sectional area can improve strength independent of BMD ⁽¹²⁾. Significant increases in cortical volume and thickness, without corresponding increases in trabecular BMC and total volume, as reported in the LIFTMOR trials ^{(21, 39)}, might be explained, in part at least, by improved edge detection resulting in a larger area of bone being reported ⁽⁴⁹⁾ or redistribution of bone minerals ⁽⁵⁰⁾. Indices of bone strength as outcomes have been limited ⁽³⁵⁾ so the effects of exercise on all the properties which improve bone strength are not currently well known, and definitive conclusions cannot yet be made regarding the effectiveness of exercise for improving bone strength and reducing fracture risk ^{(5, 7, 12)}. Increased bone mass is not confirmation of an improvement in bone quality or strength ⁽²⁾, therefore the effectiveness of an intervention should be measured by its influence on reducing incidence of fracture rather than change in BMD. Unfortunately, evidence is conflicting whether exercise reduces fracture risk ⁽⁵¹⁾ and it is presently not feasible to conduct a definitive trial to determine that ⁽³⁾.

Underscoring the potential prophylactic nature of exercise, clinical guidelines recommend that exercise should be lifelong, and include regular weight-bearing and high impact activities to promote muscle strength, muscle power, and maintenance of BMD ^{(5, 13, 17)}. Resistance exercise is recommended for older adults because it attenuates bone loss and enhances physical function ⁽⁵²⁾. Regular and consistent exercise which provides sufficient loading is especially important for bone mass and geometry ⁽⁷⁾, yet 50 % of individuals with osteoporosis who start a new exercise program have been reported to discontinue within 6 months ⁽¹⁸⁾. Programs are needed which have high rates of engagement long-term, and which provide older adults with the motivation, self-efficacy, and skills, to be able to perform at least some of their exercise independently.

A case study of a 70-year-old postmenopausal women reported a remarkable 24.3% increase in bone mass (0.196g/cm²) at the femoral neck and 29.7% (0.17g/cm²) at the lumbar spine ⁽²⁰⁾ following an extensive 12-month program of resistance training. A limitation reported in the BELL trial was that the group format was not ideally suited to the individuals with high or low functional capacity ⁽¹⁹⁾. The remarkable effects reported by Aquino et al. ⁽²⁰⁾ may be a product of the N-of-1 design, which allows the exercise program to be uniquely tailored to meet the needs and physical capacity of the individual. This is encouraging for healthcare providers training with older adults 1-on-1, but potentially challenging for those designing group-based programs. It is feasible that the rates of increase in bone mass observed in the present study, could be maintained with ongoing training, and may achieve a similar magnitude of effect to that reported by Aquino et al ⁽²⁰⁾ under similar training durations.

Due to their fear of fracture, older adults diagnosed with osteoporosis are often overly cautious and risk-averse towards engaging in impact and resistance exercise ⁽⁶⁾. This fear and anxiety can be compounded by inappropriate messages from healthcare providers where uncertainty exists around what is appropriate and safe ⁽⁵⁾. Inadequate and inconsistent advice can cause confusion and may have a negative effect on the person’s social and psychological health, wellbeing, and self-efficacy ⁽⁷⁾. For most older adults, engaging in activities which promote muscle and bone strength is safe and will help improve or at least maintain function ⁽⁷⁾, and exercise does not have to be vigorous to be beneficial. Providing reassurance to older adults with low bone mass and encouraging physical activities which promote bone health should be a priority for healthcare providers ⁽⁵⁾. Results from a qualitative study revealed the efficacy and perceived health benefits, associated with older adults participating in group kettlebell training with similar age peers ⁽¹⁹⁾. Participants reported being more physically active, having increased self-confidence, improved physical capacity, and enhanced mental health after completing the trial. Age, training status, sex, and the presence of comorbid health conditions, such as osteoarthritis and persistent pain, did not negatively impact engagement or participation in the training. Attendance was >90% with some participants still independently training together >18 months after the trial, Sally being one of them.

Regular exercise can effectively reduce a myriad of diseases associated with aging. Exercise has been described as a true geroprotector with even modest amounts of exercise capable of improving quality of life and increasing lifespan ⁽⁵³⁾. It has been suggested that 12 weeks of exercise is insufficient to observe changes in bone ⁽⁵⁴⁾ however, data from this and other studies may dispute that ⁽⁵⁵⁾ with increases in total hip BMD of 6% also reported from 16 weeks of resistance training in elderly women with decreased muscle strength ⁽⁵⁶⁾. Results from the BELL trial demonstrated that hardstyle kettlebell training can improve functional measures of healthy aging. Kettlebell training can be performed safely and independently by older adults, and when performed in a group setting, can be very rewarding, even at high rates of perceived exertion. Improvements in lean mass from exercise is protective against bone loss and risk of vertebral fractures, as anabolic signalling in the bone-muscle unit works both ways: muscle influences bone and vice versa ^{(29, 32)}. The mean increase of 0.65 kg in axial skeletal muscle mass reported in the BELL trial ⁽¹⁾, adds further support for kettlebell training being an effective mode of resistance training to promote healthy aging in older adults. Results of the present study suggests that kettlebell training might also be an effective intervention to promote bone health in older adults with low or very low BMD.

These case findings are not without limitations. Firstly, Hologic and Norland machines have different methods of deriving energy beams, different calibration techniques, and different methods for determining the regions of interest for analysis. Although comprehensive evaluation of the DXA-measured BMD values obtained using Hologic and Norland densitometers have been found to be virtually the same ⁽⁵⁷⁾ with in-vivo comparison of spinal measurements highly correlated (BMD: r = 0.990, SEE = 0.028 g/cm²), and comparable in terms of precision and accuracy ⁽⁵⁸⁾, the degree of measurement error inherent between Sally’s reports cannot be determined. Up to approximately 50% of the 0.064g/cm² change in BMD observed at Peter’s lumbar spine, could be due to the Denosumab he received in the weeks preceding commencement of training ⁽⁵⁹⁾ however, the magnitude of change would still be sufficient to reach his LSC.

6. Conclusion

Results of this study suggest that 16-weeks of high-volume moderate to high-intensity hardstyle kettlebell training performed 5 days a week, can improve bone mineral density in older adults with osteoporosis, with magnitude of change sufficient to advance osteoporosis to osteopenia. Kettlebell training may therefore be a viable mode of resistance training for improving bone health and reducing fracture risk in older adults with low and very low BMD. For the two subjects in the current study, the nature of the relationship between the training intervention and the observed increase in BMD remains unclear however, taken together with the clinically significant improvements in outcome measures, high rates of engagement, and positive participant experiences, we suggest further research in randomised controlled trial is warranted.

Funding

This trial is supported by an Australian Government Research Training Program Scholarship and will contribute towards a Higher Degree by Research Degree (Doctor of Philosophy). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. The study received no additional external funding.

Contributions

NM curated and analysed the data, interpreted the results, conducted the formal analysis, and wrote the original draft. JK and WH supported with ongoing consultation, reviewed, and provided revisions to the manuscript. All authors read and approved the final manuscript.

Ethics declarations

Rights and permissions

This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.

Supplementary data

• The BELL trial CARE (CAse REport) checklist