2018 Frank Stinchfield Award: Spinopelvic Hypermobility Is Associated With an Inferior Outcome After THA: Examining the Effect of Spinal Arthrodesis

Abstract

Background: Many patients undergo both THA and spinal arthrodesis, and those patients may not fare as well as those who undergo one procedure but not the other. The mechanisms of how spinal arthrodesis affects patient function after THA remain unclear.

Questions/purposes: The aims of our study were to (1) determine how patient-reported outcome measures (PROMs), including the Oxford hip score as well as dislocations and complications compare after THA between patients with and without spinal arthrodesis; (2) characterize sagittal pelvic changes in these patients that occur when moving between different functional positions and test for differences between patients with and without spinal arthrodesis; and (3) assess whether differences in sagittal pelvic dynamics are associated with PROMs, complications, and dislocations after THA.

Methods: In this case-control study, we identified 42 patients (60 hips) who had undergone both THA and spinal arthrodesis between 2002 and 2016 and who were available for followup at a minimum of 12 months (mean, 6 ± 5 years) after the later of the two procedures. These cases were case-control-matched for age, gender, and body mass index with 42 patients (60 hips) who underwent only THA and had no known spinal pathology. All patients completed PROMs, including the Oxford hip score, and underwent four radiographs of the pelvis and spinopelvic complex in three positions (supine, standing, and deep-seated). Cup orientation and various spinopelvic parameters, including pelvic tilt and pelvic-femoral angle, were measured. The difference in pelvic tilt between standing and seated allowed for patient classification based on spinopelvic mobility into normal (± 10°-30°), stiff (< ± 10°) or hypermobile (> ± 30°) groups.

Results: Compared with the THA-only group, the THA-spinal arthrodesis group had inferior PROMs (Oxford hip score, 33 ± 10 versus 43 ± 6; p < 0.001) and more surgery-related complications (such as dislocation, loosening, periprosthetic fracture or infection, psoas irritation) (12 versus 3; p = 0.013), especially dislocation (5 versus 0; p = 0.023). We detected no difference in change of pelvic tilt between supine and standing positions between the groups. When standing, patients undergoing THA-spinal arthrodesis had greater pelvic tilt (25° ± 11° versus 17° ± 8°; p < 0.001) and the hip was more extended (193° ± 22° versus 185° ± 30°; p = 0.012). We found that patients undergoing THA-spinal arthrodesis were more likely to have spinopelvic hypermobility (12 of 42 versus three of 42; odds ratio, 5.2; p = 0.02) with anterior tilting of the pelvis. Of all biomechanical parameters, only spinopelvic hypermobility was associated with inferior PROMs (Oxford hip score, 35 ± 9 versus 40 ± 7 in normal mobility; p = 0.049) and was also present in dislocating hips that underwent revision despite acceptable cup orientation.

Conclusions: In patients with spinal arthrodesis who have undergone THA, spinopelvic hypermobility is associated with inferior outcomes, including hip instability. Spinopelvic hypermobility should be routinely assessed because these patients may have a narrow zone of optimum cup orientation that would require new technology to define and assist the surgeon in obtaining it.Level of Evidence Level III, therapeutic study.

Fig. 1 A-B

Flow diagram that shows the method for recruiting patients to the (A) spinal arthrodesis (SA) and THA group and (B) the THA-only group. Case-matched cohorts were recruited from separate patient populations.

Fig. 2 A-B

(A) Lateral standing and (B) flexed-seated spinopelvic radiographs of a patient with an L4-L5 spinal arthrodesis and an uncemented THA. Pelvic incidence, 53°; sacral slope (SS), standing, 33°; SS seated, 61°; AI of cup standing, 17°; AI seated, -16°; pelvic femoral angle (PFA) standing, 196°; PFA seated, 80°.

Fig. 3

Bar chart of ΔPelvic tilt_{supine standing} color-coded as per group. The mean ∆Pelvic tilt_{supine standing} was 6° ± 6° and was not different for the two groups (p = 0.7).

Fig. 4

Bar chart of ΔPelvic tilt_{standing seated} color-coded as per group. For the whole cohort, the ΔPelvicTilt_{standing seated} was -15° ± 13°.