Acute thoracolumbar burst fractures (AO types A3/A4) with and without concomitant posterior ligamentous complex injury: treatment outcomes in surgically and nonsurgically managed patients. A multi-center prospective study
摘要
Treatment of neurologically intact thoracolumbar burst fractures (AO types A3/A4) is controversial. Furthermore, the impact that concomitant posterior ligamentous complex (PLC) injuries have on patient reported outcome measures (PROMs) is not well-established. This analysis compared outcomes between patients with A3/A4 injuries with and without concomitant B1/B2 injuries and between patients treated with operative versus nonoperative management.
MethodsData from a prospective multicenter cohort study (ClinicalTrials.gov: NCT02827214) was used. Adults with AO type A3/A4 burst fractures (T10-L2), with or without concomitant B1/B2 injuries, were included. PROMs included Oswestry Disability Index (ODI), Pain NRS, EQ-5D, and AOSpine Patient Reported Outcome Spine Trauma (PROST) score. Isolated A3/A4 injuries were compared to combined A3/A4 and B1/B2 injuries—patients were then sub-analyzed by treatment.
Results198 patients were included (34 with combined A3/A4 and B1/B2 injuries). Patients with combined injuries had similar baseline but different 1-year postoperative (12.9 ± 12.8 vs. 7.3 ± 9.9;p = 0.024) ODI scores compared to patients with isolated injuries. After subdividing by treatment, both operatively and nonoperatively treated patients with combined injuries had higher rates of suspected/indeterminate and injured PLC statuses (p < 0.001) compared to those with isolated injuries. There were no differences in 1-year postoperative ODI scores between surgically managed patients with and without concomitant PLC injuries. Patients treated nonoperatively with combined injuries had significantly worse one-year postoperative ODI (20.8 vs. 7.6;p = 0.018) and two-year postoperative Pain NRS (3 vs. 1.1;p = 0.04), and EQ-5D (0.8 vs. 0.9;p = 0.03) scores.
ConclusionPatients with isolated A3/A4 injuries performed similarly compared to patients with concomitant B1/B2 injuries after surgical treatment. However, combined injuries treated nonoperatively performed worse on multiple metrics of pain and disability at final follow-up. Combined injuries were associated with a significantly higher rates of suspected/indeterminate or injured PLC status. Thus, suspicion of PLC injuries should prompt serious consideration of surgical intervention in the setting of burst fractures without neurologic deficits.