Background/Purpose: Varus knee thrust is an abrupt change in frontal plane alignment of the tibiofemoral joint observed during gait. Thrust has been previously linked to radiographic knee OA progression, worsening cartilage and bone marrow lesions, and knee pain. Knee thrust is a dynamic event that likely influences the entire kinematic chain of the lower limb and causes trunk lean yet the effects of thrust on the hip and ankle and the lower back are not known. As low back, hip, and ankle pain are impactful problems among older adults, and as varus thrust is potentially modifiable through gait retraining interventions, determining a relation of thrust to pain at these regions is of interest. Our objective was to determine the effects of knee thrust on the two-year incidence of frequent ankle, hip, and low back pain in older adults with or at risk for OA.

Methods: The Multicenter Osteoarthritis Study (MOST) is a prospective cohort study of older Americans that have or are at risk for knee OA. At the 60-month clinic exam, 60 Hz frontal plane video recordings were acquired as participants completed two self-paced walking trials over a 4.9 meter walkway. A trained reader, blinded to disease status, assessed the presence of varus thrust on a majority of steps (intra-rater κ = 0.73). Frequent ankle, hip, and low back pain was self-reported by MOST participants using a homunculus. Among participants with no pain at the 60-month exam, incident frequent pain in these regions was defined at the 84-month exam as “pain, aching, or stiffness on most days in the past 30 days.” To assess the relation of thrust to incident pain, we used logistic regression adjusting for age, sex, race, BMI, and gait speed. In limb-based analyses (hip and ankle pain), generalized estimating equations accounted for two limbs from the same subject.

Results: 1375 subjects (mean age 67.3 ± 7.7; mean BMI 30.3 ± 5.7; 89.3% White; 58.7% female) contributed to the person-based analysis. 2158 knees from 1087 subjects comprised the sample for the limb-based analyses. Varus thrust was observed in 31.3% of knees. Nonsignificant results suggested that knees with thrust may have 1.30 times the odds (95% CI: 0.97, 1.73) of incident hip pain (p = 0.08), and that persons with thrust in at least one knee have 1.47 times the odds (95% CI: 0.96, 2.24) of incident low back pain (p = 0.08) compared to knees without thrust. No association was found between knee thrust and risk of incident ankle pain (p = 0.22) (see Table).

Conclusion:  While small numbers of incident pain cases may have restricted our ability to detect significant relationships, nonsignificant findings suggest a possible association between varus thrust and two-year risk of developing frequent hip and low back pain. Further investigation into the relationship between aberrant movement at the knee and painful injury at these regions may help explain these findings.