Risk factors for nonroutine discharge in adult spinal deformity surgery

BACKGROUND CONTEXT: Surgery for adult spinal deformity (ASD) is increasingly common. Although outcomes of ASD surgery have been studied extensively, to our knowledge, no data exist regarding factors predicting nonroutine discharge in this population. Nonroutine discharge is defined as discharge to a health care facility after surgery rather than to home. PURPOSE: To determine which patient and surgical factors predict nonroutine discharge after ASD surgery. DESIGN: This is a retrospective study. PATIENTS SAMPLE: We conducted a retrospective single-center study of 303 patients who underwent arthrodesis of five or more spinal levels to treat ASD between 2009 and 2014. OUTCOME MEASURES: Patients were stratified into two groups according to discharge disposition: home or nonroutine. METHODS: Objective preoperative characteristics, intraoperative course, and postoperative recovery were analyzed to identify pre- and perioperative factors associated with nonroutine discharge. Univariate analysis was performed first. All factors with P values <.2 on univariate analysis were included in a logistic regression model. Additionally, to understand the relationship between subjective patient-reported outcome measures and nonroutine discharge, we compared the two groups with respect to mean Oswestry Disability Index and Scoliosis Research Society-22r domains using Student t-tests. RESULTS: On univariate analysis, objective measures that differed significantly (P <.05) between the two cohorts were age (≥65 years), osteoporosis, Charlson Comorbidity Index score of ≥2, prolonged hospital stay (>8 days), and blood transfusion. Given the above logistic regression inclusion criteria, we controlled for the performance, and type, of osteotomy (P =.055). On multivariate analysis, older age, osteoporosis, prolonged hospital stay, blood transfusion, and 3-column osteotomy were independently associated with nonroutine discharge. Subjective patient-reported outcome measures, including Oswestry Disability Index and Scoliosis Research Society-22r physical function and pain domain scores, were significantly worse in the nonroutine discharge cohort (P <.05). CONCLUSION: To our knowledge, this is the first study to evaluate pre- and perioperative factors associated with nonroutine discharge after ASD surgery. Elderly patients who undergo complex surgery and receive blood transfusions are at particularly high risk of nonroutine discharge. Surgeons should consider these factors during surgical planning and preoperative patient counseling.