How Is Lumbar Fusion Associated With Compensatory Hip Motion After THA?

Although patients with THA may increase their use of hip flexion to compensate for loss of spinal motion after lumbar fusion, no studies have directly examined this relationship. Furthermore, the association between the number of fused levels on postoperative compensatory hip motion is not clear, making it difficult to anticipate these changes prior to lumbar fusion.

(1) Is lumbar fusion after THA associated with resting spinopelvic alignment or compensatory changes at the hip? (2) Is the number of fused spinal segments associated with an increase in hip compensatory motion? (3) Can this increase in hip compensatory motion be anticipated prior to lumbar spine fusion?

Between January 2017 and December 2023, we retrospectively identified 109 patients with lumbar fusion after THA. The minimum follow-up time after lumbar fusion was set at 6 months to account for implant stabilization and early dislocation events. Of these patients, 59% (64) were excluded because of incomplete questionnaires (17% [19 of 109]), neuromuscular disease (3% [3 of 109]), unavailable or obscured functional lateral radiographs before and/or after lumbar fusion (36% [39 of 109]), or lumbar fusion performed within 6 months after THA (3% [3 of 109]), leaving 41% (45 of 109) of patients for analysis. Among the included patients, 82% (37 of 45) were female, with a mean ± SD age of 74 ± 8 years. Functional lateral radiographs were obtained in three postures (free-standing, relaxed-seated, and flexed-seated) before and after lumbar fusion surgery. From these radiographs, spinopelvic alignment parameters were measured, including pelvic incidence, L1-S1 lumbar lordosis, sacral slope, pelvic tilt, pelvic-femoral angle (PFA), and pelvic incidence minus lumbar lordosis (PI-LL). The hip angle was defined as the PFA value calibrated for pelvic incidence. Lumbar, pelvic, and hip motion were assessed based on the degree of change between each posture. Multivariable regression analysis was performed to examine the association between the number of fused levels and the change in PFA in the flexed-seated position before and after lumbar fusion.

After lumbar fusion, mean ± SD Oswestry Disability Index (ODI) total scores were lower (44% ± 18% versus 22% ± 16%, mean difference -21% [95% confidence interval (CI) -29% to -14%]; p < 0.001). Median (range) ODI item Q1 (back pain) scores were also lower (3 [0 to 5] versus 1 [0 to 3], median difference -2 points [95% CI -3 to -2]; p < 0.001). In the standing position, L1-S1 lumbar lordosis was larger (29° ± 22° versus 34° ± 14°, mean difference 6° [95% CI 1° to 10°]; p = 0.01) and PI-LL was smaller (19° [-13° to 76°] versus 11° [-12° to 76°], median difference -4° [95% CI -9° to 0°]; p = 0.02), while other standing parameters were no different. In the flexed-seated position, L1-S1 lumbar lordosis, PFA, and hip angle were larger after lumbar fusion (L1-S1 lumbar lordosis: -8° ± 14° versus 8° ± 14°, mean difference 16° [95% CI 11° to 22°]; p < 0.001; PFA: 83° ± 17° versus 97° ± 16°, mean difference 14° [95% CI 10° to 17°]; p < 0.001; hip angle: 93° [51° to 127°] versus 106° [87° to 132°], mean difference 13° [95% CI 10° to 16°]; p < 0.001), indicating greater compensatory hip flexion. The number of fused levels was associated with the magnitude of increase in PFA in the flexed-seated position after lumbar fusion (standardized β = 0.39 [95% CI 0.15 to 0.62]; p = 0.002) after accounting for age, BMI, and preoperative hip angle in the flexed-seated position. Using the multivariable model, the anticipated magnitude of postoperative increase in PFA in the flexed-seated position could be described based on the number of fused levels and the preoperative hip angle in the flexed-seated position (R 2 = 0.43; p < 0.001). The following regression equation was described to anticipate the increase in compensatory motion after lumbar fusion: ΔPFA flexed-seated = (-0.39 × preoperative flexed-seated hip angle) + (2.44 × number of fused levels) + 44.7.

Our findings suggest that obtaining flexed-seated radiographs preoperatively in patients with prior THA who are scheduled to undergo lumbar fusion may be helpful. This information may assist spine surgeons in communicating the potential risk of postoperative dislocation with patients and hip surgeons and support shared decision-making before lumbar fusion. Future studies using three-dimensional imaging or motion-capture techniques will be necessary to determine how loss of lumbar motion is associated with multiplanar hip motion and dislocation risk during functional activities, particularly when considering acetabular and femoral component placement.

Level III, therapeutic study.