Time spent in moderate to vigorous physical activity predicts thigh morphological asymmetries in young adults 1-7 years following intra-articular knee injury.

Purpose (the aim of the study):
Post-traumatic osteoarthritis (PTOA) is a common outcome after an intra-articular knee injury. Following a knee injury there are several suggested morphological maladaptations in the reconstructed leg in comparison to the uninjured contralateral leg. Researchers have previously implied these adaptations are pathologic, with no known effective therapeutic. However, at present, no study has investigated whether the mechanical loading of the injured leg can predict the magnitude of morphological asymmetries. Consequently, the aims of this study were (1) to measure muscle cross-sectional area (mCSA),muscle density (MD), and bone mineral density (BMD) asymmetries between the injured and uninjured leg, (2) to investigate whether time spent in moderate to vigorous physical activity (MVPA) can predict morphological asymmetries, and (3) to explore the potential association between cumulative knee joint loading and morphological asymmetries.

Methods:
Thirty-four participants (18 male, 16 female; age 26 ± 6, BMI 25.2 ± 4.2) were recruited from local hospitals and sports clubs. Peripheral quantitative computerised tomography (pQCT) scans taken at 66% of the calf and 50% of the thigh, providing quantification of mCSA, MD, and BMD. Limb symmetry was calculated using: ((Uninjured Leg - Injured Leg) ÷ (0.5*(Uninjured Leg + Injured Leg)) *100%

Three-dimensional kinematics (Qualysis Miqus) and ground reaction forces (Instrumented treadmill - Bertec) were recorded during walking (0.8m/s, 1.3m/s, 1.7m/s) and running. Resultant accelerations recorded from the wrist (GENEActiv Original) were recorded at 100Hz +/- 8g. Knee flexion moment impulse (KFMi) was a surrogate of knee joint load and calculated in OpenSim, and normalised to body mass. Generalized Estimation Equation-based Physical Activity and Accelerometery R package (GGIR) was used to process accelerometer data and measure time spent in light, moderate, and vigorous physical activity. Euclidean Norm Minus One(ENMO) was calculated using: √(x+ y+ z) -1 during the walk and run and KFMi was averaged for each respective activity intensity. Total KFMi per week was then calculated by multiplying KFMi for each intensity by time spent in each activity domain per week. Paired T-tests were used for morphological comparisons, a simple linear regression was used for prediction of morphological asymmetries, and a Pearson correlation was used to explore the association between cumulative joint loading and morphological asymmetries.

Results:
mCSA for the calf and thigh was significantly lower in the injured leg in comparison to the uninjured leg (Figure 1). MD for the calf and thigh was significantly lower in the injured leg in comparison to the uninjured leg (Figure 1). BMD for the calf and thigh was not significantly different (Figure 1). Time spent in MVPA significantly predicted mCSA asymmetry at 50% of the thigh (Figure 2). KFMi minutes did not significantly associate with mCSA asymmetry at 50% of the thigh (Figure 3).

Conclusions:
Our results suggest physical activity level can predict the magnitude of these morphological deficits, which implies the therapeutic role physical activity could have on reducing risk factors for PTOA (e.g.,knee extensor/flexor weakness). However, the role of mechanotransduction in PTOA prevention needs further work when considering our framework of cumulative joint loading, which accounts for knee joint load, duration of each load, and the total number of loads, was not associated.