Background Low back pain (LBP) and osteoarthritis (OA) of the lumbar spine are common causes of disability. The contribution of joint hypermobility (range of motion greater than normal at most joints) to LBP and lumbar spine OA is not well known.
Objectives This cross-sectional study examined the relationship of joint hypermobility with LBP and lumbar spine OA in a large cohort in the United States.
Methods Of the 2146 participants with Beighton (hypermobility) data collected from 2003 to 2010, 1864 had complete LBP and lumbar spine radiographic data available for analyses. For the Beighton criteria, one point was given for each completed maneuver: passive dorsiflexion right/left fifth finger 90+ degrees, passive apposition right/left thumbs to forearm, right/left elbow hyperextension 10+ degrees, right/left knee hyperextension 10+ degrees, and palms on floor during forward trunk flexion with knees extended. The total score ranged from 0 (unable) to 9 (performed all maneuvers). A Beighton score ≥4 was defined as hypermobility. Presence of LBP was based on pain, aching or stiffness of the low back on most days. Each lumbar spine radiographic level was graded for disc space narrowing (DSN) and osteophytes (OST) in a semi-quantitative fashion (0–3) according to the Burnett Atlas. Radiographic lumbar spine OA (rOA) was defined as the presence of DSN and OST grade 1+ at the same lumbar level. Symptomatic lumbar spine OA (sxOA) was defined as presence of rOA with LBP. Associations of LBP and lumbar spine OA with hypermobility (Beighton ≥4) and each individual Beighton maneuver were estimated using separate logistic regression models, controlling for gender, race, age, body mass index (BMI), and history of low back injury. Interactions were examined between hypermobility and each covariate (p<0.10 considered statistically significant).
Results Participant characteristics were: mean age 66 years (standard deviation [SD]±10), mean BMI 31 (SD±7) kg/m2, 65% women, 33% African American, 2% low back injury, 6% hypermobility, 40% LBP, 59% rOA, and 25% sxOA. Adjusted results are summarized in the Table. Although not statistically significant, the adjusted odds of rOA were 18% lower among participants with vs. without hypermobility, while the odds of LBP and sxOA were 30% and 20% higher, respectively. Fifth finger dorsiflexion was inversely associated with LBP (35% lower odds) and sxOA (21% lower odds). The elbow maneuver was positively associated with LBP (41% higher odds), but inversely associated with rOA (28%); a similar, but not statistically significant, pattern was seen for the knee maneuver. The trunk flexion maneuver was inversely associated with LBP, rOA, and sxOA. No interactions of hypermobility and covariates were noted.
Conclusions Joint hypermobility (specifically elbow and knee maneuvers) appears to be positively associated with LBP but inversely related to lumbar spine rOA. The trunk maneuver may reflect the musculotendinous (hamstring) flexibility more than ligamentous laxity, suggesting that greater hamstring flexibility is linked to a reduced occurrence of LBP and lumbar spine OA. Longitudinal studies may elucidate the role of joint hypermobility in lumbar spine outcomes.
Disclosure of Interest None declared