Gadolinium Contrast Enhancement Improves Confidence in Diagnosing Recurrent Soft Tissue Sarcoma by MRI

Rationale and Objectives To determine how utilization of postgadolinium magnetic resonance imaging (MRI) influenced reader accuracy and confidence at identifying postoperative soft tissue sarcoma (STS) recurrence among readers with various levels of expertise. Materials and Methods This retrospective study was institutional review board approved and Health Insurance Portability and Accountability Act compliant. Postoperative MRI from 26 patients with prior STS resection (13 patients with confirmed recurrence, 13 without recurrence) was reviewed. Four blinded readers of varying expertise (radiology resident, fellow, attending, and orthopedic oncologist) initially evaluated only the precontrast images and rated each MRI for recurrence on a 5-point confidence scale. Assessment was repeated with the addition of contrast-enhanced sequences. Diagnostic accuracy based on confidence ratings was evaluated using the area under the receiver operating characteristic curve (AUC). Changes in confidence ratings were calculated using Wilcoxon signed-rank test. Results All readers demonstrated good diagnostic accuracy both with and without contrast-enhanced images (AUC >0.98 for each reader). When contrast-enhanced images were made available, the resident recorded improved confidence with both assigning ( P = 0.031) and excluding recurrence ( P = 0.006); the fellow showed improved confidence only with assigning recurrence ( P = 0.015); and the surgeon showed improved confidence in excluding recurrence ( P = 0.003). The addition of contrast-enhanced images did not significantly influence the diagnostic confidence of the attending radiologist. Conclusions Diagnostic accuracy of MRI was excellent in evaluating postoperative STS recurrence, and reader confidence improved depending on expertise when postgadolinium imaging was included in the assessment.