Whole-body MRI added to gadoxetic acid-enhanced liver MRI for detection of extrahepatic disease in patients considered eligible for hepatic resection and/or local ablation of colorectal cancer liver metastases. (February 2020)

• Record Type: Journal Article
• Authors: Sivesgaard, Kim; Larsen, Lars P; Sørensen, Michael; Kramer, Stine; Schlander, Sven; Amanavicius, Nerijus; Mortensen, Frank V; Pedersen, Erik M
• Is Part Of: Acta radiologica. Volume 61:Number 2(2020)
• Page Start: 156
• Page End: 167
• Publication Date: 2020-02
• Subjects: Colorectal neoplasms -- neoplasm metastasis -- magnetic resonance imaging -- positron emission tomography computed tomography; Radiology, Medical -- Periodicals; Radiography, Medical -- Periodicals; Radiotherapy -- Periodicals; 616.0757
• DOI: 10.1177/0284185119855184
• Languages: English
• ISSNs: 0284-1851

Abstract

Background: Fluorodeoxyglucose positron emission tomography/computed tomography (FDG-PET/CT) can detect extrahepatic disease before local treatment of colorectal liver metastases. Purpose: To investigate if whole-body magnetic resonance imaging (MRI) added to gadoxetic acid liver MRI could replace FDG-PET/CT for detection of extrahepatic disease in patients with colorectal liver metastases eligible for hepatic local treatment. Material and Methods: This health-research-ethics-committee-approved prospective consecutive diagnostic accuracy study with written informed consent analyzed 79 cases included between 29 June 2015 and 7 February 2017. Whole-body MRI covering the thorax, abdomen, and pelvis and FDG-PET/CT including contrast-enhanced CT was performed 0–3 days before local treatment of liver metastases. Whole-body MR images were read independently by two readers. FDG-PET/CT images were read independently by two pairs of readers. Histopathology and follow-up imaging were used as reference standard. Sensitivities and specificities were calculated and compared including noninferiority analysis. Results: Seventy-five malignant lesions and 419 benign lesions were confirmed. Sensitivities of both PET/CT reader pairs 1 and 2 (56.7 and 67.9%) and MRI reader 2 (63.0%) were significantly higher than that of MRI reader 1 (42.7) (P<0.04). Specificities of both PET/CT reader pairs 1 and 2 (92.5 and 92.4%) and MRI reader 1 (91.1%) were significantly higher than that of MRI reader 2 (86.3%) ( P < 0.02). Sensitivity of MRI reader 2 was non-inferior compared to PET/CT reader pair 1. Specificities of both MRI readers were non-inferior to both PET/CT reader pairs. Conclusion: For detection of extrahepatic disease in patients with colorectal liver metastases, whole-body MRI was non-inferior to FDG-PET/CT for some reader combinations. However, reader-independent non-inferiority could not be demonstrated.