Outcome prediction in lumbar disc herniation surgery

Introduction: It is a well recognized fact that a significant proportion of patients operated on for lumbar disc herniation exhibit a poor outcome, regardless of the apparent technical success of the operative procedure itself. The aim of this study was to identify a set of widely available variables that accurately predict outcome after discectomy and to develop a predictive model based upon those variables. Patients and methods: Basic demographic, clinical and radiological variables were evaluated in a group of 70 patient operated on for lumbar disc herniation. Outcome was assessed using VAS and RM scales 6 months postoperatively and correlated to aforementioned variables. Results: Preoperative pain intensity and duration, age and type of disc herniation were all shown to be statistically significant predictors of outcome, unlike sex, type of radiological investigation and preoperative tension sign testing results. Multivariate regression analysis including only variables previously identified as good outcome predictors revealed that the pain intensity exhibited the strongest correlation with outcome, followed by pain duration, type of disc herniation and age. Even though MR scan was more sensitive in detecting disc extrusion than CT (sensitivity of 100% versus 65%, respectively), the presence of preoperative MR scan did not influence the outcome. Conclusion: The study identified a set of widely available and easily attainable variables as fair predictors of outcome after lumbar discectomy. Subsequent logistic regression resulted in a predictive model whose accuracy is to be determined in another prospective study.