Magnetic Resonance Imaging or Endoscopic Ultrasonography for Detection and Surveillance of Pancreatic Neuroendocrine Neoplasms in Patients with Multiple Endocrine Neoplasia Type 1?

Our aim was to compare the clinical utility of Magnetic Resonance Imaging (MRI) and Endoscopic Ultrasonography (EUS) in identifying Pancreatic Neurondocrine Neoplasms (PanNENs) and monitoring size alterations in Multiple Endocrine Neoplasia type 1 (MEN1) patients. Thirty-one MEN1 patients with PanNENs and concurrent screening by EUS and abdominal MRI were included and 129 pancreatic lesions were detected in total. MRI detected fewer lesions than EUS (n=73 vs. 110, p=0.006). MRI sensitivity and specificity compared to EUS at 20 and 10 mm cut-offs of maximal lesion diameter were 96 and 88% (20 mm cut-off) and 90 and 82%(10 mm cut-off), respectively (concordance rates of 97 and 87% and Cohen's kappa=0.912 and 0.718, respectively). Lesions<1 cm were more often detected with EUS (p=0.025). Data from sequential concurrent imaging on lesion growth rate [n=7 (mean±SD: 2 mm/year±3.4 mm vs. 1.9 mm/year±3.6 mm)] over a period of at least two years as well as pathology data in connection to preoperative concurrent imaging were available in a small number of patients (n=7, p=0.933 for mean differences in maximal lesion diameter). MRI of the pancreas was more readily available and less expensive than EUS in an outpatient setting. In conclusion, MRI performs well compared to EUS for the detection and subsequent surveillance of MEN1-related panNENs larger than 10 mm and seems to be cost-effective. Both modalities could be used at initial assessment and MRI alone could be utilized thereafter in patient surveillance. EUS retains its value in surgical planning and the detection of small mostly functional PanNENs.