Fatty pancreas is independently associated with subsequent diabetes mellitus development: A 10-year prospective cohort study

BACKGROUND & AIMS While the association between fatty pancreas and metabolic syndrome has been suggested in retrospective studies, long-term prospective data on the effect of fatty pancreas on various metabolic outcomes is lacking. We aimed to prospectively investigate the association between fatty pancreas and development of major metabolic outcomes. METHODS 631 subjects from a population study using fat-water magnetic resonance imaging to quantify pancreatic and liver fat content during 2008-2010 were prospectively followed until December 2020 (mean follow-up time: 11.1±1.1 years). Subjects with significant alcohol intake and diabetes (DM) at baseline were excluded. Incidence of newly diagnosed DM, hypertension, dyslipidemia, ischemic heart disease, cardiovascular accidents, pancreatic cancer and mortality were evaluated. RESULTS Among the 631 subjects (mean age 48±11 years), 93 (14.7%) had fatty pancreas. The fatty pancreas group had a higher incidence of DM (33.3% vs 10.4%, p<0.001), hypertension (37.7% vs 22.7%, p=0.003) and dyslipidemia (37.7% vs 14.6%, p<0.001) during long-term follow-up. Individuals with both fatty liver and pancreas had the highest DM incidence, followed by fatty liver only and fatty pancreas only groups (p<0.001). Fatty pancreas was independently associated with DM (adjusted hazard ratio [aHR] 1.81, 95%CI 1.10-3.00, p=0.020) but not hypertension or dyslipidemia on multivariate analysis. Each percentage rise of pancreatic fat increased the risk of incident DM by 7% (aHR 1.07, 95%CI 1.01-1.13, p=0.016). No participants developed pancreatic cancer during follow-up. CONCLUSION Fatty pancreas is independently associated with subsequent DM development but not hypertension or dyslipidemia.