<h3>BACKGROUND CONTEXT</h3> The AO Spine Classifications are associated with a hierarchical injury severity score. "Type A" fractures represent the most stable injuries and "Type C" represent the most severe and unstable injuries. "Three column" subaxial cervical spine and thoracolumbar injuries are designated as Type B (tension band) and Type C (fracture dislocation). Although the load sharing classification describes burst type fractures, Type B and C injuries are often minimally comminuted but grossly unstable due to "three-column" disruption. Therefore, the load sharing classification may not be appropriate to guide short- or long-segment fixation. <h3>PURPOSE</h3> To identify if construct length affects the rate of surgical complications or instrumentation revision following Type B and C subaxial or thoracolumbar injuries. <h3>STUDY DESIGN/SETTING</h3> Retrospective cohort study. <h3>PATIENT SAMPLE</h3> All patients with cervical or thoracolumbar Type B and Type C injuries between 2006-2021 were identified from a prospectively collected database. Only patients who underwent operative intervention were included. <h3>OUTCOME MEASURES</h3> Surgical complications (persistent cerebrospinal fluid leak, wound dehiscence, epidural hematoma, surgical site infection) and reoperations (instrumentation revision and/or exchange). <h3>METHODS</h3> Patients were divided based on the number of levels instrumented above and below the injured level (>2 levels (long-segment) versus < 2 levels (short-segment)). Patient demographics, surgical characteristics and clinical outcomes were collected and compared. Independent t-tests or Mann-Whitney U tests compared continuous variables, while Pearson's chi-square tests compared categorical variables. Bivariate logistic regression measured the effect of fixation construct length (long-segment fixation) on the likelihood of having a surgical complication and reoperation for instrumentation revision and/or exchange. Alpha was set at P<0.05. <h3>RESULTS</h3> A total of 180 patients representative of 187 fractures were included. Of the 187 injuries, 69 (36.9%) underwent long-segment fixation. Patients who underwent long-segment fixation were older (61.1 vs 51.5 years, p=0.005), were more likely to have comorbid ankylosing spondylitis (AS) or diffuse idiopathic skeletal hyperosteosis (DISH) (47.8% vs 21.2%, p<0.001), and were more likely to have an injury that involved a thoracic vertebra (53.6% vs 30.5%, p=0.008). Further, patients who underwent long-segment fixation were more likely to undergo posterior-only fixation (85.5% vs 60.2%, p<0.001) in comparison to anterior-only and combined anterior-posterior fixation. There was no significant difference in the proportion of long-segment versus short-segment constructs that included bilateral, unilateral or no pedicle screws within the injured vertebrae (p=0.371). Bivariate logistic regression demonstrated that a long-segment fixation construct had a non-significant increase in complication rates (OR=2.08, p=0.176) and reoperations for instrumentation revision or exchange (OR=2.99, p=0.142). We subanalyzed the 58 fractures with AS/DISH compared to the 129 fractures without AS/DISH. Injuries with AS/DISH had a mean 2.57+ 1.01 and 2.34 + 0.81 levels instrumented above and below the injury compared to injuries without AS/DISH, mean 1.97 + 1.04 and 1.90 + 0.96 levels instrumented above and below the injury (p<0.001), respectively. However, there was no difference in the rate of surgical complications (8.62% vs 7.75%, p=0.780) or instrumentation revision (6.90% vs 3.10%, p=0.256) between the two groups based on construct length. <h3>CONCLUSIONS</h3> Long-segment fixation of Type B or C subaxial cervical spine or thoracolumbar spine injuries resulted in a non-significant increase in complications and spine reoperations for instrumentation revision. Instrumentation technology has improved since the load sharing classification. This exploratory study suggests short segment fixation may be appropriate even for unstable spine fractures. <h3>FDA DEVICE/DRUG STATUS</h3> This abstract does not discuss or include any applicable devices or drugs.
Read full abstract