LAUSR

Reply: I read with interest the study by Seo et al, 1 which aimed to validate the 2012 Fukuoka Consensus Guidelines (FCG) and to compare the diagnostic performances of computed tomography and magnetic resonance imaging for distinguishing benign from malignant intraductal papillary mucinous neoplasms (IPMNs) in a large surgical series of 158 patients. The study concluded that mural nodules were the most important predictor of malignancy for IPMN, and there was no significant difference between the diagnostic performance of computed tomography and magnetic resonance imaging in differentiating benign from malignant IPMN. An important point I would like to raise regarding this study is that contrary to their title and main objective, there was no attempt by the authors to ‘‘validate’’ the FCG. This study attempted to identify and evaluate only the value of preoperative predictors of malignant IPMN such as tumor size, main pancreatic duct diameter, presence of mural nodule, and lymphadenopathy. In my opinion, any validation study of the international consensus guidelines should report the accuracy, specificity, sensitivity, positive predictive value (PPV), or negative predictive value (NPV) of the guidelines. Two systematic reviews, which similarly have included only surgical-resected cases, have recently been performed to validate the original Sendai Consensus Guidelines (SG) and FCG, respectively. In the first study, which included 501 SCGþ ve and 189 SCG-ve branch-duct IPMNs, the PPV of SCGþ ve group was only 29.9% and the NPVof SCG-ve was 90%. In the more recent review validating the FCG in 1382 patients, pooled analysis demonstrated that there were 362/1000 (36.2%) malignant FCGþ ve IPMNs (high risk and worrisome features) and 342/382 (89.5%) benign FCG-ve IPMNs. The PPV of the high-risk group and the worrisome risk groups alone were 104/158 (65.8%) and 75/261(28.7%), respectively. It would be extremely useful and important if the authors could report the proportion of benign and malignant IPMN classified as high risk, worrisome features or low risk according to the FCG in their patient cohort. Further information on the PPV of FCGþ ve IPMN and NPV of FCG-ve IPMN according to the 2012 guidelines would also add to the value of this study. Finally, it is important to add that in studies including only surgically resected IPMN, the true denominator is unknown as many benign and low-risk IPMNs are managed conservatively. Hence, the accuracy of the guidelines based on surgical series have to be interpreted carefully, especially with regard to the NPV of FCG-ve tumors.