Background/Purpose: Knee buckling (“giving way”) is a common symptom in knee osteoarthritis (OA).  Yet, little is known regarding risk factors for buckling.  Knee buckling has been found to be associated with muscle weakness, a treatable cause, in one study and this finding needs to be confirmed. In addition, vibratory deficits have been shown to be present in knee OA, and may create alterations in knee position, increasing vulnerability to buckling but this has not been examined.  Here, we evaluate the association of muscle strength and vibratory perception with buckling in older adults with, or at high risk for, knee OA.

Methods: MOST is a NIH-funded longitudinal study of persons with symptomatic knee OA or at increased risk of OA.  At the 60-month visit, participants underwent evaluation of isokinetic muscle strength at the right leg, bilateral evaluation of vibratory perception threshold (VPT) at predetermined anatomic sites, and were asked about buckling and its frequency in the past 3 months. Knee buckling was defined in each subject as (a) any buckling and (b) repeated buckling (≥ 2 episodes). VPT was evaluated using a biothesiometer.  The applicator tip of the instrument was placed on preselected anatomic bony prominences and the voltage increased by 1 volt/sec until the participant acknowledged sensation.  Mean VPT between the limbs was used for analyses.  Quadriceps strength was measured as the maximum torque during active isokinetic extension using a dynamometer and scaled to body size by dividing the maximum torque by BMI.   VPT and strength were categorized into groups based on ±1 SD of the gender-specific mean of the sample.  A person-based analyses using logistic regression to estimate adjusted odds ratios for the association of VPT and strength with buckling in the past 3 months was performed.   We adjusted in analyses for age, sex, BMI, race, clinic site, and WOMAC knee pain.

Results: We evaluated 2,291 subjects (60% women, age±SD=68±8 yrs, BMI±SD=31±6 kg/m²).   16.7% of participants reported buckling and 13.4% reported repeated buckling in the past 3 months.  Results are summarized in the Table.  A borderline association was found between muscle strength and any buckling episode, however, a significant association was observed with repeated buckling. High quadriceps strength was associated with a significantly decreased odd of repeated buckling (Adj OR: 0.59[0.36,0.99].  There was no significant association found between VPT and buckling.   

Conclusion: In this large cohort of participants with knee OA or at high risk for knee OA, greater quadriceps strength was found to be associated with decreased odds of repeated knee buckling.   There was no significant association found between vibratory perception and knee buckling.  Future studies should continue to evaluate for other potential risk factors for buckling in knee OA.

Table

*VPT adjusted for muscle strength and muscle strength adjusted for VPT

**adjusted for all covariates