Search

Background & Aims Little is known about whether the 2006 Sendai guidelines or 2012 Fukuoka guidelines are being used to determine the level of risk posed by suspected pancreatic mucinous cystic neoplasms (PCNs). We evaluated whether the guidelines accurately predicted which patients with suspected PCNs, which was based on cross-sectional imaging findings, would be found to have advanced neoplasia in surgery. Methods We performed a retrospective study of data collected from 194 patients with cystic lesions of the pancreas, which were assessed by cross-sectional imaging analyses, who underwent surgery for suspected PCNs at the Hospital at the University of Pennsylvania from 2000 through 2008. Imaging data were used to classify patients according to the Sendai guidelines as high risk or low risk and according to the Fukuoka guidelines as high risk, worrisome, or low risk. Pathology analyses of samples collected during surgery were used as the reference. A logistic regression model was created to identify factors associated with advanced neoplasia. The Sendai and Fukuoka guideline criteria were analyzed by univariate and multivariable logistic regression analyses. Results Advanced neoplasias were found in 36 patients (18.5%; 22 invasive cancers and 14 high-grade dysplasias). The median size of cysts was 33 mm. All patients found to have invasive cancers were accurately assigned to the Sendai guidelines high risk or Fukuoka guidelines high risk groups. However, 3 patients in the Sendai guidelines low risk and 2 patients in the Fukuoka guidelines low risk groups were found to have high-grade dysplasia. The Sendai guidelines identified patients with advanced neoplasia with 91.7% sensitivity, 21.5% specificity, 21% positive predictive value, and 91.9% negative predictive value. A designation of Fukuoka guidelines high risk identified patients with advanced neoplasia with 55.6% sensitivity, 73% specificity, 32% positive predictive value, and 87.9% negative predictive value. Overall, there was no statistically significant difference between the guidelines in predicting which patients had advanced neoplasia. On multivariate analysis, the presence of a mural nodule (odds ratio [OR], 2.88; 95% confidence interval [CI], 1.33–6.27; P = .008), dilated main pancreatic duct >10 mm (OR, 7.44; 95% CI, 2.36–23.52; P = .001), or enhancing solid component (OR, 2.92; 95% CI, 1.16–7.64; P = .02) were associated with detection of advanced neoplasia in pancreatic cysts. Conclusion On the basis of a retrospective analysis, the Sendai and Fukuoka guidelines accurately determine which patients with pancreatic cysts have advanced neoplasia. The guidelines accurately recommended surgical resection for all patients found to have invasive cancer, although some patients with high-grade dysplasia were missed. The updated Fukuoka guidelines are not superior to the Sendai guidelines in identifying neoplasias. Cyst size was not associated with advanced neoplasia. [ABSTRACT FROM AUTHOR]