Spine Stiffness Leads to High Pelvic Mobility: Uncoupling Native Mechanics and Explaining Why Patients With Stiff Spines Have Increased Dislocation Risk.

BACKGROUND Patients with stiff spines are at increased risk of instability after THA because of pelvic stiffness. Comprehensive study of patients with a stiff spine without hip arthritis could provide insight into native compensatory mechanisms and provide guidance on the mechanics to account for after arthroplasty. QUESTIONS/PURPOSES The primary aim of this study was to characterize static and dynamic compensatory mechanics that occur in the presence of either a stiff hip or stiff spine. The secondary study aims were to assess which spinopelvic imaging modalities would best uncouple compensation mechanisms and to test the effect of length of spinal fusion (that is, number of fused segments) on the existing compensatory mechanics. METHODS This was a prospective, case-control study performed at two academic tertiary referral centers. The cohort studied included three groups: (1) the control group of asymptomatic volunteers without signs of hip osteoarthritis or history of spinal surgery (n = 52); (2) the hip group of patients with osteoarthritis treated with THA between 2018 and 2019 (n = 512), excluding those with age < 18 years (n = 2), BMI > 40 kg/m2 (n = 9), different diagnosis than osteoarthritis (n = 117), history of spinal or lower limb disease or surgery (n = 206), neurologic comorbidities (n = 17), absence of study consent (n = 20), or without spinopelvic radiographs (n = 17), in which the included patients (n = 124) were matched for age, sex, and BMI to the control group, resulting in the final hip group of 52 patients; and (3) the spine group were patients seen in clinic between 2023 and 2024 (n = 121), 1 year after spinal fusion, excluding those with BMI > 40 kg/m2 (n = 10), hip osteoarthritis or surgery (n = 16), neuromuscular disease (n = 1), spinal fusion not including lumbar spine (n = 1), or without spinopelvic radiographs (n = 41), leaving 52 patients. The whole cohort comprised 60% (93 of 156) females, and the mean ± SD age was 64 ± 11 years. All underwent standing, relaxed-, and deep-seated radiographs to determine static characteristics: lumbar lordosis, pelvic tilt, pelvic-femoral angle, and pelvic incidence. Dynamic characteristics included difference in pelvic tilt, lumbar lordosis, and pelvic-femoral angles between standing and relaxed- or deep-seated positions, thereby determining which imaging modality best uncoupled compensatory mechanisms. Correlation between the number of fused segments and spinopelvic parameters was assessed using Spearman correlation coefficient. RESULTS When standing, the spine group had a higher mean ± SD pelvic-femoral angle than the control (197° ± 7° versus 186° ± 10°, mean difference -11° [95% confidence interval (CI) -14° to -7°]; p < 0.001) and hip group (197° ± 7° versus 183° ± 11°, mean difference -14° [95% CI -18° to -10°]; p < 0.001) and a higher pelvic tilt compared with the control (20° ± 9° versus 15° ± 8°, mean difference -5° [95% CI -8° to -2°]; p = 0.003) and hip group (20° ± 9° versus 15° ± 7°, mean difference -5° [95% CI -9° to -2°]; p = 0.004). Dynamically, the spine group exhibited the least lumbar flexion (ΔLL) in both relaxed- (12° ± 11° versus 22° ± 12° versus 16° ± 12°; p = 0.002) and deep-seated transitions (25° ± 14° versus 43° ± 13° versus 43° ± 13°; p < 0.001). Between standing and deep-seated, change in pelvic tilt was greater in the spine group compared with the hip (20° ± 16° versus -6° ± 16°, mean difference -28° [95% CI -33° to -22°]; p < 0.001) and control group (20° ± 16° versus 4° ± 17°, mean difference -19° [95% CI -26° to -13°]; p < 0.001). Deep-seated, the spine group flexed the hip more than the hip group (109° ± 15° versus 70° ± 21°, mean difference -40° [95% CI -47° to -34°]; p < 0.001) and control group (109° ± 15° versus 85° ± 18°, mean difference -23° [95% CI -30° to -16°]; p < 0.001). Standing to deep-seated assessments better uncoupled compensatory mechanisms, as these detected differences between control and spine group (for instance, ∆LL standing/deep-seated 43° ± 13° versus 25° ± 14° [mean difference 19° (95% CI 14° to 25°); p < 0.001] versus ∆LL standing/relaxed-seated 16° ± 12° versus 12° ± 11° [mean difference 4° (95% CI 0° to 9°); p = 0.15]). The number of segments fused was associated with deep-seated lumbar lordosis (ρ = 0.55; p < 0.001) and pelvic tilt (ρ = -0.31; p = 0.02). CONCLUSION In this study, patients with a stiff spine have hyperextended hips when standing and hyperflexed hips in a deep-seated position and exhibit a fivefold greater change in pelvic tilt between these positions compared with controls. The greater pelvic tilt change may cause an acetabular cup to be brought in a functionally suboptimal orientation, leading to impingement or dislocation. Deep-seated radiographs can uncouple compensatory mechanisms and are recommended to better identify patients with spinal stiffness. LEVEL OF EVIDENCE Level II, diagnostic study.