A thesis submitted in fulfilment of the requirements for the degree of Doctor of Philosophy, to the Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, Johannesburg 2018 / Knee osteoarthritis (OA) is the most prevalent form of OA and it is present in over 33% of adults aged 50 years and above. Patients with end-stage knee OA have poor health outcomes including severe knee pain, functional limitations and poor quality of life (QoL) with decreased physical activity (PA) and increased sedentary behaviour (SB). In spite of the cost-effectiveness of total knee arthroplasty (TKA) in improving patients’ health outcomes (as measured using patient-reported outcome measures (PROMs)), the objectively measured PA shows little or no change after surgery and SB has received very little attention following TKA. However, published studies have only been conducted in populations from high-income countries and no studies have assessed PA and SB in knee OA patients from low-middle income countries including South Africa. Furthermore, the detailed patterns by which patients with knee OA accumulate PA and SB before and after TKA have not been described. Studies have mainly focused on measuring overall PA or moderate to vigorous PA (MVPA) and/or patients’ adherence to the PA guidelines with very little attention to low intensity activities of the movement continuum (SB and light activity- LPA). Furthermore, different activity monitors have been used with very few of them capable of measuring low intensity activities. Assessing activity behaviours incidental in activities of daily living (ADL) (such as sitting, standing and walking) before and after TKA may be clinically useful as activities of older adults undergoing TKA mainly constitute these low intensity activities. With no previously published systematic review on changes in SB following TKA, the objective of the first study of this thesis was to integrate available evidence on changes in SB in patients with knee OA after a primary TKA. A systematic literature search from January 2002 to 31 October 2017 was performed across seven electronic databases, for longitudinal and cross-sectional studies published in English on objectively (through accelerometry) and/or subjectively measured changes in SB following TKA. Ten studies reporting on SB with a total of 1,028 participants were included in the review. Three studies reported changes in SB with two showing a reduction in SB and one, with high risk of bias, showing an increase in SB after TKA. Seven studies showed no change in SB following TKA. The second study of this thesis was a longitudinal design comprising of two parts (Study 2A and B). Participants wore two activity monitors (ActiGraph GT3X+ and ActivPAL) to measure PA and SB for seven consecutive days (24 hours/day) at baseline (preoperative), six weeks and six months after TKA. Therefore, the second objective (Study 2A) of this thesis was to objectively measure changes in volume and pattern of PA and SB (using ActiGraph GT3X+ accelerometer) in patients with knee OA from baseline to six months after TKA and to assess changes in PROMs following TKA. Eighty-nine patients (13 males, 76 females between 55 and 80 years of age) scheduled for primary TKA took part in the study. Physical activity and SB were measured with an ActiGraph GT3X+ accelerometer for seven consecutive days (24 hours/day) and range of motion (ROM) was measured prior to TKA, and six weeks and six months after TKA. The University of California Los Angeles (UCLA) Activity index and Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) were used to assess self-reported activity and physical function respectively. Of the initial 89 patients recruited, 57 completed the six months followup and 45 had valid activity data at the 6 months follow-up. The proportion of time (% of waking day) patients spent in light physical activity (LPA) increased from baseline [29.0 (26.6-31.4)%] to 6 months [34.8 (31.3-38.3)%; p=0.008]. However, time spent in moderate to vigorous PA (MVPA) did not change from baseline [median (interquartile range): 2.0 (7.8) min/day] to six months after TKA [3.4 (11.6) min/day, p>0.05]. Approximately 9%, 5% and 18% of the patients met the PA guidelines at baseline, and six weeks and six months after TKA respectively. The proportion of time (% of waking day) patients spent in SB decreased after TKA [baseline: mean (95% CI): 70.1 (67.5-72.7)%; six months: 64.0 (60.6-67.9)%; p=0.009]. The interruptions to SB increased between baseline and six months after TKA [mean (95% CI): 85.0 (80.0-90.0) to 93.0 (88.0-98.0) breaks/day, p=0.014]. There was a significant improvement in WOMAC score [median (interquartile range): 71.0 (27.0) vs. 4.0 (11.3), p<0.001], UCLA score [median (interquartile range): 2.0 (1.0) vs. 5.0 (1.0), p<0.001] as well as ROM [mean range: (0.0 - 90.0)° vs (0.0 - 110)°, p<0.05] between baseline and six months after TKA. Study 2A showed that LPA increased and SB decreased as measured using ActiGraph GT3X+. In addition, self-reported functional capacity (FC) or functional ability (as measured with PROMs) improved after TKA. The third objective (Study 2B) of this thesis was to objectively assess changes in the times spent sitting, standing and walking following TKA and to examine their associations with the changes in PROMs after TKA. The same patients in Study 2A also wore a second activity monitor, the ActivPAL (which accurately measures low intensity activities and posture) for the same periods of time as described in Study 2A above. Patients spent significantly more of their waking wear time walking at six months after TKA (mean% (95% CI): 10.8% (9.4-12.1)), than preoperatively (mean% (95% CI): 8.3% (7.7-10.0)), p=0.039), however, the percentage of daily time spent standing did not change at six months after TKA (mean% (95% CI): 34.2% (29.8-38.6)) compared to percentage time preoperatively (mean% (95% CI): 32.4% (28.6-35.5)), p=0.530). Patients decreased their average daily time spent sitting from preoperative to six months after TKA by 33.7 mins/day (95% CI: -18.9 – 106.3, p=0.099). Patients took significantly more steps per day at six months after TKA [mean (95% CI: 3670 (2886-4020)] steps/day compared to preoperatively 2570 (2366-3189) steps/day, p<0.001. Participants also increased their cadence (steps/min) six months after surgery [mean (95% CI): 33 (31-34) vs. 38 (33-39), p=0.004]. There were no associations between objectively measured changes in the time spent sitting, standing and walking and changes in PROMs (p>0.05). The studies presented in this thesis have novel aspects that extend the body of knowledge on activity behaviours of patients with knee OA undergoing TKA. The studies in this thesis report the first systematic review on changes in SB of knee OA patients following TKA. This thesis is the first to objectively measure the detailed patterns of PA and SB in patients with knee OA undergoing TKA from a low-middle income country (South Africa). Furthermore, this thesis is also the first to use two accelerometers to generate detailed activity behaviour in patients with knee OA undergoing TKA. Lastly, this thesis is the first to assess the association between changes in times spent sitting, standing and walking in relation to changes in health outcomes in knee OA patients after TKA. In conclusion, the systematic review showed that SB has been superficially described and there is insufficient evidence to suggest that time spent in SB decreases following TKA. Majority of the studies reported no change in SB after TKA. The longitudinal study showed that, following TKA, there was a decrease in the overall time spent in SB and an increase in the number of breaks in SB that appeared to be replaced by LPA. Participants’ volume and average daily cadence increased following TKA. In addition, participants decreased their time spent sitting by over half an hour at six months after TKA. However, there were no associations between changes in the times spent sitting, standing and walking and changes in measures of participants’ health outcomes (PROMs) following TKA. Both objective and subjective measures should be used to accurately assess improvements in patients’ health outcomes following TKA. This comprehensive analysis of detailed daily activity behaviours can be used to employ feasible interventions for increasing the duration of LPA (standing and walking) and decreasing sedentary time (sitting/lying) to improve quality of life and overall health following TKA. / XL2019