Benjamin Boxer thesis.pdf (3.23 MB)
Download file

The role of physical activity and exercise on bone and cartilage physical properties, metabolism, and joint symptoms

Download (3.23 MB)
posted on 21.11.2022, 10:07 authored by Benjamin Boxer

Osteoarthritis (OA) and osteoporosis (OP) are highly prevalent and debilitating conditions that share some common risk factors. Exercise and/or physical activity can be beneficial to both conditions, however, there still remain many unanswered questions regarding how different types, durations and frequencies of exercise/physical activity affect bone and cartilage metabolism and joint health. The aim of this thesis was to examine the role of physical activity and exercise in bone and cartilage physical properties, metabolism, and joint symptoms.

The first study of the thesis examined the acute response to a bout of high impact exercise (IE) and a resistance exercise (RE) in healthy young men. Participants completed a control (CON) and IE or RE trial. In the IE trial 120 maximum effort jumps were performed, in the RE trial 120 lifts were performed. Both included rest pauses. Venous blood samples were taken before and immediately and 24 hours after exercise and at corresponding times in the CON trial. The concentration of serum sclerostin ([sSOST]), cartilage oligomeric matrix protein ([sCOMP]) Procollagen I N-Terminal Propeptide ([PINP]) and C-Terminal Telopeptide of Type I Collagen ([CTx-I]) was measured. Repeated measures ANOVAs were used to assess whether there was trial*time interaction. 

Participants were 26 men (mean (SD) age: 23.4 (2.9) years: body mass index 23.1 (1.7) kg/m2). High forces were recorded during IE; countermovement jump peak force at take-off and vertical ground reaction force were 1.6 (0.3) and 6.2 (2.1) times body weight respectively and one repetition maximums (1RM) for the RE participants were all greater than one times body weight. In the impact exercise trial, changes in [sSclerostin] differed significantly from those in control (p for trial*time interaction = <0.0001); with the post exercise value being significantly higher than baseline and 24 hours later (+35.6% and +35.6% respectively; both P<0.05); there was no significant time*trial interaction in the resistance exercise trial (p > 0.05). [sPINP] also differed significantly between exercise and control in the impact exercise trial (p for trial*time interaction = 0.006) with post values being significantly higher than baseline (+11.5% post exercise, p = 0.001). After adjustment for haemoconcentration [sPINP] also changed significantly in the resistance exercise (p for trial*time interaction = <0.0001) but decreased after exercise (-13.2% p < 0.001). sPIICP and sCOMP also showed different responses between impact exercise and control trials (p for time*trial interaction = 0.030 and 0.049 respectively), with increases of 4.2 and +20.2% immediately after the exercise. Impact exercise was associated with an acute increase in [sSclerostin] which was accompanied by an increase in [sPINP] after impact exercise and a decrease after resistance exercise once adjusted for haemoconcentration. This was also accompanied by changes in [sCOMP] and [sPIICP] however, no change in [sCTx-I]. However, it is uncertain whether an acute increase in [sSclerostin] relates to an increase in synthesis or an increased distribution of pre-existing sclerostin from bone to systemic circulation; future research should examine the underlying mechanism.  

Study two investigated changes in hip shape after unilateral high impact exercise interventions in older adults and applied a previously developed model associated with OA risk. Dual X-ray absorptiometry scans were analysed using statistical shape modelling, an image analysis technique employing principal component analysis, and statistically derived modes of variation were generated, describing how much each image varies from the mean shape of all images. Femoral neck BMD increased in exercise compared to control leg (p <0.05). In men there was a non-significant decrease in LHM2 mean score (p = 0.034 before adjustment, p = 0.68 after) possibly indicating a general widening of the femur. MrOSM modes 2, 3 and 4 were negatively associated with OA and 9 positively associated. When applied to our model there were no significant changes in any MrOSM after adjustments were made for multiple comparisons. However, before adjustment there was a decrease in MrOSM 2 in the exercise leg of men (p = 0.007 before adjustment, p = 0.056 after) suggesting a potential detrimental effect of exercise on hip shape, increasing OA risk. Both interventions increased femoral neck BMD and however, had no effect on hip shape after adjustments for multiple comparisons. Unilateral high impact exercise seemed to have limited effect on proximal femur shape despite increasing density.

The COVID-19 restrictions in the UK have influenced physical activity engagement and sedentary behaviour due to stay at home orders and the closure of leisure facilities which may have impact upon joint symptoms. Study three was an online survey examining physical activity (Community Healthy Activities Model Program for Seniors (CHAMPS)), sedentary behaviour (Longitudinal Aging Study Amsterdam (LASA)), and joint symptoms (Knee Injury and Osteoarthritis Outcome Score (KOOS)) before and during the UK 2020 COVID-19 lockdown and one year later in adults ≥60 years old. One hundred and eighty-eight participants were included in analysis, 95 completing the follow up survey one year later. There was no significant change in physical activity over lockdown (p>0.05) but a significant decrease in sedentary behaviour during week days and increase during the weekend (p<0.05). These changes in physical activity and sedentary behaviour were accompanied by some small inconsistent changes in joint symptoms (KOOS pain over lockdown and KOOS quality of life (QOL) one year later) indicative of a moderate worsening. In conclusion, participants maintained their physical activity levels over the first COVID-19 lockdown in the UK although sedentary time decreased on weekdays and increased at weekends with some modest worsening of joint symptoms.

This research has examined the role of physical activity and exercise on several aspects of bone, cartilage, and joint health. Impact exercise in the form of jumping increased bone signalling molecules sclerostin which was accompanied by increases in bone formation and cartilage metabolism. RE had no effect on sclerostin but decreased bone formation. The first COVID-19 lockdown in the UK did not significantly decrease individuals PA levels as we had hypothesised but did have a small significant effect on sedentary behaviour and some self-reported joint symptoms. Finally, six months of unilateral high impact exercise seemed to have limited effect on proximal femur shape despite increasing density. 

Overall, this research indicated that physical activity and exercise had a small but mostly positive effect on bone and cartilage physical properties, metabolism, and joint symptoms each being an independent contributor to bone and joint health. Cumulatively these small effects could amount to substantial benefits to bone and cartilage health and future research should incorporate all elements; physical properties, metabolism, and joint symptoms, to investigate the impact of exercise and other lifestyle factors on the joint as a whole.



  • Sport, Exercise and Health Sciences


Loughborough University

Rights holder

© Benjamin Boxer

Publication date



A Doctoral Thesis. Submitted in partial fulfilment of the requirements for the award of the degree of Doctor of Philosophy of Loughborough University.




Katherine Brooke-Wavell ; Jonathan Folland

Qualification name


Qualification level


This submission includes a signed certificate in addition to the thesis file(s)

I have submitted a signed certificate