Resident Physician University of South Florida Tampa, Florida, United States
Introduction: To treat Adult Spinal Deformity, long construct fusions are often required to stabilize the spine and bring it back into physiologic alignment. The biomechanical forces required to maintain this alignment in the diseased spine can exceed the material strength resulting in rod fractures (RFs) especially in the setting of non-union. We sought to identify factors associated with RFs.
Methods: We conducted a retrospective review of all degenerative long construct surgeries at our tertiary care center from 2016-2021. Long construct cases were defined by fusions with upper instrumented vertebra of L2 or higher and lower instrumented vertebra of S1 or the pelvis. These cases were reviewed for demographics, preoperative parameters, surgical characteristics, postoperative parameters, and complications encountered.
Results: We identified 216 patients who fit our inclusion and exclusion criteria. Thirty-five patients were found to have a RF. There was no difference in average age (RF 63.1+3.1 years, no RF 61.5+1.6 years, p=0.411), nor body mass index (RF 28.7+1.68kg/m2, no RF 29.3+0.86kg/m2, p=0.590). Twenty-one (60%) patients with RF were males (p=1.00). Constructs with lateral (p=0.690) or anterior (p=0.177) interbody fusions in addition to their posterior hardware were at no less risk of rod fracture. Prior thoracolumbar surgery was not a risk factor (p=0.947). Patients with rod fractures had higher numbers of levels instrumented (RF 10.5+0.78 no RF 8.7+0.48, p< 0.01). If S2-alar-iliac (S2AI) screws were used as the inferiormost instrumentation, there was a higher risk of rod fracture than if S1 or iliac screws were used (p=0.016). A multivariate logistic regression model was created that found only number of levels instrumented to be predictive of rod fracture (Odds ratio = 1.18, 95% CI = 1.055-1.316, p=0.004). There was no difference in follow-up time (RF 791+174 days, no RF 666+68 days, p=0.148).
Conclusion : The only factor we found to be predictive of rod fracture was increasing amounts of instrumentation: odds of rod fracture increased by 1.18 for each additional level of instrumentation. With longer constructs perhaps more structure is needed to sustain the biomechanical forces, something offered by multiple rod constructs.
