Background context: Several studies have examined the occurrence of medical complication after spine surgery. However, many of these studies have been done using large national databases. Although these allow for analysis of thousands of patients, potentially influential covariates are not accounted for in these retrospective studies. Furthermore, the accuracy of these retrospective data collection in these databases has been called into question.
Purpose: Using multivariate analysis on a prospectively collected data registry to determine significant risk factors for medical complication after spine surgery.
Study design: Retrospective multivariate analysis of prospectively collected registry data. The registry is a prospectively collected database of all patients who underwent spine surgery in our two institutions from January 1, 2003 to December 31, 2004.
Methods: Extensive demographic and medical information were prospectively recorded as described previously by Mirza et al. Complications were defined in detail a priori, and they were prospectively recorded for at least 2 years after surgery. We analyzed risk factors for medical complication after spine surgery using univariate and multivariate analyses.
Results: We analyzed data from 1,591 patients who met out inclusion criteria. The cumulative incidences of complication after spine surgery per organ system are as follows: cardiac, 8.4%; pulmonary, 13%; gastrointestinal, 3.9%; neurological, 7.35%; hematological, 10.75%; and urological complications, 9.18%. The occurrence of cardiac or respiratory complication after spine surgery was significantly associated with death within 2 years (relative risk, 4.11 and 10.76, respectively). Surgical invasiveness and age were significant risk factors for complications in five of the six organ systems evaluated. Individual organ system-specific elative risk values with 95% confidence intervals and p values are listed in Tables 3 and 4.
Conclusions: Risk factors identified in this study can be beneficial to clinicians and patients alike when considering surgical treatment of the spine. Future analyses and models that predict the occurrence of medical complication after spine surgery may be of further benefit for surgical decision making.
Copyright © 2012 Elsevier Inc. All rights reserved.