Early Medical Complications and Delayed Discharge after Spinopelvic Fusion: A Comparative Analysis of 887 NSQIP Cases from 2006 to 2016

INTRODUCTION: The effect of pelvic fixation on postoperative medical complications, blood transfusion, length of hospital stay, and discharge disposition is poorly understood. Determining factors that predispose patients to increased complications after spinopelvic fusion will help surgeons to plan these complex procedures and optimize patients preoperatively. METHODS: We conducted a retrospective cohort study using data from the ACS-NSQIP database between 2006 and 2016 of patients who underwent lumbar fusion with and without spinopelvic fixation. Data regarding demographics, complications, hospital stay, and discharge disposition were collected. RESULTS: A total of 57,417 (98.5%) cases of lumbar fusion without spinopelvic fixation (LF) and 887 (1.5%) cases of lumbar fusion with spinopelvic fixation (SPF) were analyzed. The transfusion rate in the SPF group was 59.3% vs 13% in the LF group (p < 0.001). The mean length of stay (LOS) and discharge to skilled nursing facility (SNF) were significantly different (LOS: SPF 6.5 days vs LF 3.5 days p < 0.001; SNF: SPF 21.3% vs LF 10.4% p < 0.001). After controlling for demographic differences, the overall complication rates were not significantly different between the groups (p = 0.531). The odds ratio for transfusion in the SPF group was 2.9 (p < 0.001). The odds ratio for increased LOS and increased care discharge disposition were elevated in the SPF group (LOS OR: 1.3, p < 0.012, Discharge disposition OR: 1.8, p < 0.001). CONCLUSIONS: Patients who underwent SPF had increased complications, transfusion rate, LOS, and discharge to SNF or subacute rehab facilities as compared with patients who underwent LF. SPF remains an effective technique for achieving lumbosacral arthrodesis. Surgeons should consider the implications of the associated complication profile for SPF and the value of preoperative optimization in a select cohort of patients.