Study Design. Retrospective cohort study.
Objective. To assess the rate, causes, and risk factors of unplanned hospital readmission after spine fusion for the treatment of adult spinal deformity.
Summary of Background Data. Hospital readmissions in the elderly are common, and with increasing emphasis on the quality of health care, readmission rates are used to assess hospital performance. Spine surgery has seen rapidly increased utilization during the past 2 decades. Surgical treatments of complex spinal deformity are known to have higher rates of complications than other types of spine surgery. However, there are no reports describing the rates and causes of hospital readmission after deformity surgery.
Methods. Patients were identified at a single institution from 2006 through 2011 that received a spine fusion for the treatment of adult spinal deformity. All hospital readmissions within 90 days of discharge were reviewed for cause. Unplanned readmission rates were calculated via Kaplan-Meier failure analysis. Rates were compared across patients receiving different lengths of spine fusion (short: 2–3 vertebra, medium: 4–8, long: 9 or more). Risk factors were assessed using a Cox proportional hazards multivariate model.
Results. Eight hundred thirty-six patients were enrolled (111 short, 402 medium, and 323 long fusions). The overall unplanned readmission rate was 8.4% at 30 days and 12.3% at 90 days. Patients with long spine fusion had higher rates of readmission than patients with medium or short length fusions. Surgical site infection accounted for 45.6% of readmissions. Risk factors for readmission include longer fusion length, higher patient severity of illness, and specific medical comorbidities.
Conclusion. Unplanned hospital readmissions after spine fusion for adult spinal deformity are common, and are most often due to surgical site infection. Patient medical comorbidities are an important part of assessing risk and can be used by providers and patients to better assess individual risk prior to treatment.
Level of Evidence: 3