Charlson Score is a Robust Predictor of 30-Day Complications Following Spinal Metastasis Surgery

Study Design. Retrospective chart review. Objective. To identify predictors of 30-day complications after the surgical treatment of spinal metastasis. Summary of Background Data. Surgical treatment of spinal metastasis is considered palliative with the aim of reducing or delaying neurologic deficit. Postoperative complication rates as high as 39% have been reported in the literature. Complications may impact patient quality of life and increase costs; therefore, an understanding of which preoperative variables best predict 30-day complications will help risk-stratify patients and guide therapeutic decision making and informed consent. Methods. We retrospectively reviewed 200 cases of spinal metastasis surgically treated at Stanford Hospital between 1999 and 2009. Multiple logistic regression was performed to determine which preoperative variables were independent predictors of 30-day complications. Results. Sixty-eight patients (34%) experienced one or more complications within 30 days of surgery. The most common complications were respiratory failure, venous thromboembolism, and pneumonia. On multivariate analysis, Charlson Comorbidity Index score was the most significant predictor of 30-day complications. Patients with a Charlson score of two or greater had over five times the odds of a 30-day complication as patients with a score of zero or one. Conclusion. After adjusting for demographic, oncologic, neurologic, operative, and health factors, Charlson score was the most robust predictor of 30-day complications. A Charlson score of two or greater should be considered a surgical risk factor for 30-day complications, and should be used to risk-stratify surgical candidates. If complications are anticipated, medical staff can prepare in advance, for instance, scheduling aggressive ICU care to monitor for and treat complications. Finally, Charlson score should be controlled for in future spinal metastasis outcomes studies and compared to other comorbidity assessment tools.