Predictive Value of Preoperative Magnetic Resonance Imaging Findings for Survival and Local Recurrence in Patients Undergoing En Bloc Resection of Sacral Chordomas

Scott L Zuckerman,Behrang Amini,Sun-Ho Lee,Ganesh Rao,Claudio E Tatsui,Laurence D Rhines
DOI: https://doi.org/10.1093/neuros/nyy578
IF: 5.315
2019-12-01
Neurosurgery
Abstract:Background: Few studies have assessed the prognostic importance of preoperative magnetic resonance imaging (MRI) scan findings in patients undergoing en bloc resection of sacral chordomas. Objective: To (1) report in-depth imaging characteristics of sacral chordomas, (2) describe the patterns of local and distant recurrence, and (3) determine imaging predictors of overall survival (OS) and local recurrence (LR). Methods: This retrospective case series was obtained from a prospectively maintained spine database from 1995 to 2016. All patients underwent en bloc resection of sacral chordomas. Detailed demographic, outcome, and MRI data were collected. A survival analysis was completed with multivariable cox regression to assess the impact of MRI findings on OS and LR. Results: A total of 41 patients underwent en bloc resection of sacral chordoma with preoperative MRI scans. Tumor characteristics included lobulated: (93%) and soft tissue tail (54%). The following areas had tumor invasion lobulated: sacroiliac (SI) joint (15%), ilium (5%), piriformis (61%), gluteus (46%), subcutaneous fat (32%), and lumbosacral venous plexus (22%). After multivariable analysis, only subcutaneous fat extension was an independent predictor of decreased OS (hazard ratio 5.30, 95% confidence interval 1.47-19.19, P = .011). Though not significant after multivariable analysis, the following factors were significant predictors of LR after univariate logrank testing: above the L5/S1 disc space (P = .004), SI joint invasion (P = .036), and piriformis extension (P = .022). Conclusion: The presence of subcutaneous fat extension was an independent predictor of decreased OS. Other MRI findings with potential for future evaluation include size, presence of soft tissue tail, extension above L5/S1, and SI joint and piriformis invasion.
What problem does this paper attempt to address?