Lumbar Multifidus Muscle Activation Differs In Older Adults With Symptomatic Knee Osteoarthritis With And Without Low Back Pain

PURPOSE: Compare lumbar multifidus (LM) activation in older adults with symptomatic knee osteoarthritis (KOA) with and without low back pain (LBP) and the association between muscle activation and subjective function. METHODS: Ten older adults with KOA and LBP (5F/5M, 63.2 ± 10.2 years, 95.6 ± 27.3 kg, 172.3 ± 12.4 cm), and 12 adults with KOA (5F/7M, 65.3 ± 9.1 years, 85.0 ± 23.3 kg, 170.4 ± 11.3 cm), completed the Knee Injury and Osteoarthritis Outcome Score (KOOS), and the Western Ontario and McMaster Universities Arthritis Index (WOMAC). LBP was defined as having greater than a score of 12 on the Oswestry Disability Index. Resting LM thickness and dynamic LM thickness during a modified Sorensen test were obtained using ultrasound at the L4-L5 facet joint. Three images were measured and averaged for resting and dynamic images. LM activation ratio (AR) was calculated as the ratio of contracted LM thickness to resting LM thickness. Group differences in AR were compared using an independent t-test and Cohen’s d effect sizes were calculated. Associations between LM AR and subjective function for each group were evaluated using Pearson’s correlation coefficient (r). RESULTS: Patients with KOA and LBP had lower LM AR when compared to those with only KOA (MD = .22 ± .23%; d = .94; p = .04). Greater LM AR was associated with greater disability on WOMAC subscales (pain: r = .66, p = .02; stiffness: r = .69, p = .01; physical function: r = .66, p = .02) and total WOMAC total (r = .68, p = .01) in the KOA group but not in the KOA + LBP group (pain: r = -.35, p = .33; stiffness: r = -.31, p = .38; physical function: r = -.09, p = .81; total score: r = -.20, p = .57). Greater LM AR was also associated with greater disability on KOOS subscales in the KOA group (symptoms: r = -.63, p = .03; ADL: r = -.61, p = .03; QoL: r = -.77, p < .01) but not in the KOA + LBP group. LM AR was not significantly correlated with KOOS pain subscale in both KOA (r = -.51, p = .09) and KOA + LBP groups (r = .43, p = .21). CONCLUSIONS: Patients with KOA + LBP demonstrate lower LM activation than those with only KOA. Further, differences in AR-subjective function associations between the groups suggests that once patients with KOA develop LBP, it is too late to clinically modify their low back muscle activation. Thus, clinical importance should be given to improving low back muscle strength and activation in KOA patients.