Small neuroendocrine pancreatic tumors in multiple endocrine neoplasia type 1 (MEN1): least significant change of tumor diameter as determined by endoscopic ultrasound (EUS) imaging.

Endosonography enables detection and localization of small pancreatic neuroendocrine tumors (PETs) which cannot be detected by computed tomography, magnetic resonance imaging, or somatostatin receptor scintigraphy. Knowledge about the prognosis of very small PETs in MEN1 is limited, and if there are no clinical symptoms, endocrine activity or mechanical problems and thus no clear indication for surgical therapy, an appropriate decision for the management of such patients might be to control their follow-up by endosonographic imaging. Therefore, the reproducibility of the measurement of the diameter of very small PETs by endosonographic imaging was investigated in this prospective study. We included 33 PETs smaller than 15 mm in their largest diameter detected by endosonographic imaging (Pentax FG 32 UA) in ten patients with genetically confirmed MEN1-disease. Three repeated measurements of each tumor were performed. Reproducibility was expressed as mean coefficient of variation of intra-observer variability. Mean tumor diameter was 6.9 +/- 3.4 mm (range 2.8 - 14.2 mm). Mean coefficient of variation was 5.5 +/- 4.6 % (range 0.0 - 19.4 %): in tumors < 5 mm (n = 13) 7.1 +/- 6.3 %, in tumors > 5 mm (n = 20) 4.4 +/- 2.6 %. Least significant change (p < 0.05) was calculated as 15.4 % (tumors < 5 mm: 19.9 %; tumors > 5 mm: 12.3 %). In conclusion, endosonographic imaging enables the measurement of small PETs with an acceptable reproducibility. Changes of tumor diameter of more than 20 % have to be taken as statistically significant.