Quadriceps rate of torque development predicts disability in individuals with tibiofemoral knee osteoarthritis

Purpose: Knee osteoarthritis (OA) is the leading musculoskeletal cause of years lived with disability. While maintaining quadriceps strength is imperative for limiting disability, many activities of daily living (i.e. rising from sitting, ascending stairs) also require rapid force generation, quantified as rate of torque development (RTD), rather than strength alone. Quadriceps RTD may also influence disability in individuals with knee OA. As quadriceps strength declines as the disease progresses, quadriceps RTD may influence disability differently depending upon radiographic OA disease severity. Determining if quadriceps RTD influences disability in individuals with knee OA would identify a specific therapeutic target for interventions aiming to reduce disability. Our purpose was to determine if quadriceps RTD predicts disability in individuals with knee OA after accounting for quadriceps strength. Secondarily, we separately determined if quadriceps RTD predicts disability in those with mild-moderate bilateral knee OA and moderate-severe bilateral knee OA. Methods: Forty-seven individuals (55% female; 62.7 ± 6.7 years; 172.3 ± 9.1 cm; 82.8 ± 15.8 kg) with symptomatic knee OA, defined as K-L score of 2–4 and normalized WOMAC function score ≥31, were included. Disability outcome measures included self-reported function assessed via the WOMAC function subscale, and three measures of physical performance including: 1) 20-meter fast paced walk (WALK), 2) 30-second chair stand (CHAIR), and 3) timed stair climb (STAIR). Quadriceps strength and RTD were assessed during a maximal isometric contraction at 70° on an isokinetic dynamometer. Quadriceps strength was defined as the peak torque value, and RTD was calculated as the slope of the torque-time curve at various time intervals following torque onset (RTD0–30 ms, RTD0–50 ms, RTD50–100 ms, RTD100–200 ms). Early and late RTD intervals were used to assess neural and structural components of RTD, respectively. Quadriceps strength and RTD were calculated for the involved (i.e more symptomatic) and uninvolved limb. Pearson product moment correlations were used to assess bivariate associations between each measure of disability and quadriceps strength and quadriceps RTD. If a quadriceps RTD outcome significantly associated with a measure of disability a linear regression model was used to determine the unique contribution of quadriceps RTD to the specific measure of disability after accounting for quadriceps strength. After adding quadriceps strength into the regression model, the change in R2 (ΔR2) for the RTD outcome was determined. Participants were then separated into two groups based on bilateral disease severity, which was determined by summing bilateral K-L scores. Individuals with a total K-L score ≤ 5 were classified as mild-moderate bilateral OA. Individuals with a total K-L score ≥ 6 were classified as moderate-severe bilateral OA; this ensured both limbs were at least a K-L grade of 3. The previous statistical analysis was utilized again within each severity group. Alpha level was set a priori (P ≤ 0.05). Results: In the entire cohort, involved limb RTD100–200 significantly predicted 7% of WALK (ΔR2 = 0.074; P = 0.028). In the entire cohort, uninvolved limb RTD0–50 significantly predicted 15% (ΔR2 = 0.148; P = 0.003) of CHAIR; RTD50–100 significantly predicted 12% (ΔR2 = 0.122; P = 0.008) and 8% (ΔR2 = 0.077; P = 0.035) of WALK and CHAIR, respectively; RTD100–200 significantly predicted 8% (ΔR2 = 0.079; P = 0.037) and 7% (ΔR2 = 0.074; P = 0.050) of WALK and CHAIR, respectively. In the mild-moderate cohort uninvolved limb RTD0–30 significantly predicted 22% (ΔR2 = 0.217; P = 0.038) and 28% (ΔR2 = 0.275; P = 0.006) in WOMAC and CHAIR, respectively; RTD0–50 significantly predicted 32% (ΔR2 = 0.320; P = 0.002) in CHAIR; RTD50–100 significantly predicted 18% (ΔR2 = 0.184; P = 0.030) in CHAIR. None of the quadriceps RTD measures significantly predicted any outcome of disability in the cohort of participants with moderate-severe bilateral OA after accounting for quadriceps strength (P > 0.05). Conclusions: In individuals with knee OA, quadriceps RTD significantly predicts various measures of disability after accounting for quadriceps strength. Interestingly, quadriceps RTD predicted the greatest amount of unique variance in disability in the mild-moderate bilateral OA group, but quadriceps RTD did not predict disability in the moderate-severe group. Interventions to decrease disability may need to be specific to radiographic disease severity; improving quadriceps RTD may be beneficial for decreasing disability in individuals in the early stages of radiographic disease.