Risk factors for revision surgery following primary adult spinal deformity surgery in patients 65 years and older

OBJECTIVE Surgical correction of adult spinal deformity (ASD) is a complex undertaking with high revision rates. The elderly population is poorly studied with regard to revision surgery, yet senior citizens constitute a rapidly expanding surgical demographic. Previous studies aimed at elucidating appropriate risk factors for revision surgery have been limited by small cohort sizes. The purpose of this study was to assess factors that modify the risk of revision surgery in elderly patients with ASD. METHODS The PearlDiver database (2005-2012) was used to determine revision rates in elderly ASD patients treated with a primary thoracolumbar posterolateral fusion of 8 or more levels. Analyzed risk factors included demographics, comorbid conditions, and surgical factors. Significant univariate predictors were further analyzed with multivariate analysis. The causes of revision at each year of follow-up were determined. RESULTS A total of 2293 patients who had been treated with posterolateral fusion of 8 or more levels were identified. At the 1-year follow-up, 241 (10.5%) patients had been treated with revision surgery, while 424 (18.5%) had revision surgery within 5 years. On univariate analysis, obesity was found to be a significant predictor of revision surgery at 1 year, while bone morphogenetic protein (BMP) use was found to significantly decrease revision surgery at 4 and 5 years of followup. Diabetes mellitus, osteoporosis, and smoking history were all significant univariate predictors of increased revision risk at multiple years of follow-up. Multivariate analysis at 5 years of follow-up revealed that osteoporosis (OR 1.98, 95% Cl 1.60-2.46, p < 0.0001) and BMP use (OR 0.70, 95% Cl 0.56-0.88, p = 0.002) were significantly associated with an increased and decreased revision risk, respectively. Smoking history trended toward significance (OR 1.37, 95% CI 1.10-1.70, p = 0.005). Instrument failure was consistently the most commonly cited reason for revision. Five years following surgery, it was estimated that the cohort had 68.8% survivorship. CONCLUSIONS For elderly patients with ASD, osteoporosis increases the risk of revision surgery, while BMP use decreases the risk. Other comorbidities were not found to be significant predictors of long-term revision rates. It is expected that within 5 years following the index procedure, over 30% of patients will require revision surgery.