The clinical utility of the 2017 Fukuoka Guidelines and 2015 American Gastroenterological Association Guidelines for the management of intraductal papillary mucinous neoplasms.

640 Background: Mucinous cystadenocarcinoma are malignant and mucinous pancreatic cysts (PC) have malignant potential. The management of PC remains controversial despite consensus guidelines. This study aims to evaluate the clinical utility of the 2017 Fukuoka Guidelines (FG) and 2015 American Gastroenterological Association Guidelines (AGA-G) for the management of PC. Methods: 212 patients who underwent EUS for PC between 2010 and 2017 were identified. The FG and AGA-G were used to define worrisome and high-risk cyst features. Receiver Operating Characteristic (ROC) curve was used to define sensitivity (SN), specificity (SP), positive predictive value (PPV) and negative predictive value (NPV). Results: 141 of 212 patients had IPMNs.EUS-FNA was performed in 76.5% with no reported complications. Median follow-up was 4.2 years. The majority of the IPMNs were in the pancreatic head (44.7%) or body (39.7%) while only 15.6 % were in the tail. Using the FG, 46.1% had at least one worrisome feature (FG-W) and 7.1% had at least one high risk feature (FG-HR). Using the AGA-G, 28.4% had at least one HR feature (AGA-HR1) and 1.4% patients had two or more risk factors (AGA-HR2). A change in cyst character (increase of > 5 mm in 2 years, development of a solid component, or new pancreatic duct dilation) was noted in 43.2% patients. The median time to cyst change was 21 months. For prediction of cyst changes, the FG-W had a SN of 45.8%, SP of 55.4%, PPV 45%, and NPV 56%. FG-HR had a SN of 14.3%, SP of 53.2%, PPV 1.7%, and NPV 91.8%. AGA-HR1 had a SN of 35.3%, SP of 51.5%, PPV 20%, and NPV 69.9%. AGA-HR2 had a SN of 0%, SP of 54.2%, PPV 0%, and NPV 97.3%. No difference was seen in cyst change or development of high risk or worrisome features with CEA > 192 vs. < 192 (p = 0.99). During follow up, 14 patients died, but only one patient died of pancreatic cancer. Conclusions: FG and AGA-G are difficult to validate because malignant cyst transformation is rare. There was no correlation between any cyst characteristics on EUS and cyst changes. FG-W had the best performance in predicting changes. Surgical candidates should be carefully selected, as these guidelines have a limited clinical utility.

Never Say ‘Never Say “Never’’ ’?

Is there anything wrong with using a concept C in the course of arguing against the use of C? You might think not, so long as the argumentation is framed as a kind of reductio; or climbing a ladder to be kicked away. On the other hand, you might find such “hypocritical” arguments self-undermining: if they’re sound, then we shouldn’t accept any of their premises that make use of C. Understanding the status of hypocrisy is an urgent question for conceptual ethics (and therefore engineering), insofar as concepts of philosophical interest tend to be central to our conceptual schemes, in the sense that they tend to turn up in the analyses of lots of other concepts. Developing these points (of view), the present chapter exposes a worrisome feature of the anti-hypocrisy argument just sketched: it seems to be indeterminate whether or not the argument is itself hypocritical.