RESUMEN
PURPOSE: It was the aim of this study to investigate the complementary diagnostic performance of a combined pelvic and thoracoabdominal magnetic resonance imaging (MRI) examination and positron emission tomography (PET) enhanced by image fusion in patients with suspected rectal cancer recurrence. PATIENTS AND METHODS: Thirty-one patients with clinically suspected recurrence from rectal cancer were retrospectively included, who had received MRI (high resolution pelvic MRI combined with thoracoabdominal MRI performed during continuous table translation) and (18)F-FDG-PET within 30 days. MRI alone, PET alone, and MRI and PET combined including fusion images were analysed by two observers in consensus. The likelihood of malignancy of all detectable lesions was rated on a 5-point Likert scale. The standard of reference consisted of histopathology and follow-up imaging. Confidence ratings were analysed with a jackknife free response receiver-operator characteristic paradigm (JAFROC). Further test characteristics were derived by considering "probably malignant" and "definitely malignant" lesions as positive test results. RESULTS: The reference standard comprised 150 malignant lesions (48 local, 102 distant). JAFROC analysis revealed overall figures-of-merit of 0.73 for MRI, 0.63 for PET, and 0.83 for the combined approach (differences significant). The sensitivities of MRI, PET and the combined approach were 85.4, 52.1, and 95.8â% for local recurrence and 61.8, 47.1, and 81.4â% for distant recurrence, respectively. CONCLUSION: The combination of local high-resolution MRI, thoracoabdominal continuously moving table MRI and FDG-PET supported by image fusion improves lesion detection in recurrent rectal cancer.
