Oral Posters: Cervical
Presented by: J. Goldstein - View Audio/Video Presentation (Members Only)
N. Lakomkin(1), J. Cheng(2), J. Bosco(3), J. Goldstein(4)
(1) Icahn School of Medicine at Mount Sinai, New York, NY, United States
(2) Yale School of Medicine, Neurological Surgery, New Haven, CT, United States
(3) NYU Hospital for Joint Diseases, Orthopaedic Surgery, New York, NY, United States
(4) NYU Hospital for Joint Diseases, New York, NY, United States
Introduction: Surgical fusion is frequently performed in the treatment of adult spinal deformity and is characterized by substantial rates of postoperative adverse events. Although discharge disposition has been identified as a potential risk factor for subsequent complications in a variety of surgical subspecialties, its role in fusion for adult spinal deformity remains unclear. The purpose of this study was thus to identify the perioperative variables associated with non-routine discharge and examine the association between discharge disposition and adverse events following fusion. We hypothesized that discharge to an outside facility would be significantly associated with increased incidence of post-discharge complications.
Methods: The 2013-2014 National Surgical Quality Improvement Program (NSQIP) database was used to identify all patients who presented with an international classification of diseases (ICD) 9/10 diagnosis for spinal deformity. Patients undergoing fusion for deformity were subsequently isolated using the relevant current procedural terminology (CPT) codes, as per prior studies. All patients presenting with sepsis, disseminated cancer, or trauma were excluded. Demographic data, preoperative comorbidities/laboratory values, and perioperative characteristics including operative time, hospital length of stay (LOS), and American Society of Anesthesiologists (ASA) score were collected for each patient. Discharge destination, represented as disposition to home, inpatient rehabilitation, or a skilled nursing facility was noted. Postoperative complications were assessed as subgroups consisting of major complications (sepsis, stroke, cardiac arrest, major infection, etc.), minor complications (pneumonia, superficial wound infection, and urinary tract infection), and as a cumulative group. These adverse events were then stratified based on whether they occurred before or after discharge. Chi squared analysis and a backward, binary logistic regression model were performed to identify the perioperative factors associated with non-home discharge. An identical analysis was subsequently employed to examine the association between discharge disposition and post-discharge adverse events. The area under the curve (AUC) was used to assess the predictive capacity of each model.
Results: A total of 3,190 patients were included in the series, presenting with a mean age of 59 years. Of these, 2,383 (74.7%) were discharged to home, while 807 (25.3%) were discharged to a skilled nursing facility or inpatient rehabilitation center. In a multivariable logistic regression model controlling for significant confounders, female sex (OR=1.48, p< 0.001), longer operative time (OR=1.003, p< 0.001), age>65 (OR=3.34, p< 0.001), functional dependence (OR=1.98, p=0.001), preoperative congestive heart failure (OR=5.08, p=0.012), ASA 3-4 (OR=1.65, < 0.001), and prolonged LOS (OR=1.14, p< 0.001) were found to be associated with non-routine discharge. This model represented excellent predictive capacity, with an AUC of 0.823. In separate regression models controlling for variation in perioperative variables as well as pre-discharge complications, non-home discharge was an independent risk factor for developing minor (OR=1.77, p=0.031), major (OR=1.63, p=0.039), and all (OR=1.55, p=0.024) post-discharge adverse events. These models exhibited strong predictive capacity, with AUC values of 0.70, 0.70, and 0.68.
Conclusions: A variety of perioperative variables, including newly diagnosed congestive heart failure and operative time were found to be predictive of discharge to rehabilitation or skilled nursing. Non-routine discharge disposition was an independent risk factor for all types of post-discharge complications. These findings may be valuable in the risk stratification and discharge decisions regarding adult deformity patients.