Association of fatty pancreas with increased risk of pancreatic adenocarcinoma.

Abstract

e16729 Background: Pancreatic adenocarcinoma (PADC) remains one of the most fatal malignancies with poor outcomes and prognosis. Several risk factors have been associated with its development such as smoking, age, obesity, chronic pancreatitis, diabetes mellitus and a family history of PADC. Furthermore, recent pathologic studies demonstrated that fatty infiltration of the pancreas (FP) is positively correlated with PADC development. We sought to systematically review the literature and perform the first meta-analysis to study the risk of PADC among patients with FP. Methods: We conducted a systematic search of the Pubmed, EMBASE, and Cochrane databases from inception through November-2019 for studies correlating FP with PADC. Relevant data was extracted and analyzed using comprehensive meta-analysis software. Random-effects model was used for all variables. Heterogeneity was assessed using the I2 measure and Cochrane Q-statistic. Publication bias was assessed using Egger’s test. Meta regression models accounting for independent variables such as age, sex, smoking, family history of PADC, chronic pancreatitis and method of FP diagnosis were constructed to explain heterogeneity. Results: Five observational case-control studies published between 2014 and 2019 including a total of 761 patients (320 PADC patients and 441 controls) were included. FP was associated with increased PADC with an OR 4.6 (CI 2.4-8.9) compared to controls with a considerable heterogeneity (I2= 69%). Meta regression analysis accounting for modality used to diagnose FP was able to explain 100% of the noted heterogeneity. Conclusions: While we noted FP to be significantly associated with increased PADC, heterogeneity in FP diagnostic approach resulted in significant inter-study variation. A consensus on a clear definition of FP with a standardized diagnostic approach is needed to better appraise literature on this emerging disease entity. Further prospective studies are needed to validate our results and explore the possible role for PADC screening in FP in addition to known factors such as family history and new-onset diabetes mellitus.