A novel distinguishing system for the diagnosis of malignant pancreatic cystic neoplasm

PurposeTo explore a simple and reliable non-invasive distinguishing system for the pre-operative evaluation of malignancy in pancreatic cystic neoplasm (PCN).MethodsThis study first enrolled an observation cohort of 102 consecutive PCN patients. Demographic information, results of laboratory examinations, and computed tomography (CT) presentations were recorded and analyzed to achieve a distinguishing model/system for malignancy. A group of 21 patients was then included to validate the model/system prospectively.ResultsBased on the 11 malignancy-related features identified by univariate analysis, a distinguishing model for malignancy in PCN was established by multivariate analysis: PCN malignant score = 2.967 × elevated fasting blood glucose (FBG) (≥6.16 mmol/L) ± 4.496 × asymmetrically thickened wall (or mural nodules ≥ 4 mm) ± 1.679 × septum thickening (≥2 mm) − 5.134. With the optimal cut-off value selected as −2.8 in reference to the Youden index, the proposed system for malignant PCN was established: septum thickening (>2 mm), asymmetrically thickened wall (or mural nodules > 4 mm), or elevated FBG (>6.16 mmol/L, accompanying commonly known malignant signs), the presence of at least one of these 3 features indicated malignancy in PCN. The accuracy, sensitivity and specificity of this system were 81.4%, 95.8% and 76.9%, respectively. MRI was performed on 32 patients, making correct prediction of malignancy explicitly in only 68.8% (22/32). The subsequent prospective validation study showed that the proposed distinguishing system had a predictive accuracy of 85.7% (18/21). Moreover, a higher model score, or aggregation of the features in the proposed system, indicated a higher grade of malignancy (carcinoma) in PCN.ConclusionElevated FBG (>6.16 mmol/L), asymmetrically thickened wall (or mural nodules > 4 mm) and septum thickening (>2 mm) are of great value in differentiating the malignancy in PCN. The developed distinguishing system is reliable in the diagnosis of malignant PCN.