Research: Pathophysiology Pancreas volume and fat fraction in children with Type 1 diabetes S. E. Regnell 1 , P. Peterson 2 , L. Trinh 2 , P. Broberg 3 , P. Leander 4 , A. Lernmark 1 , S. M ansson 2 and H. Elding Larsson 1 1 Pediatric Endocrinology, Diabetes and Celiac Disease Unit, Lund University/Clinical Research Centre and Sk ane University Hospital, Malmo, 2 Medical Radiation Physics, Department of Translational Medicine, Lund University, Sk ane University Hospital, Malmo, 3 Department of Cancer Epidemiology, Department of Clinical Sciences, Lund University and Sk ane University Hospital, Lund and 4 Department of Radiology, Department of Translational Medicine, Lund University and Sk ane University Hospital, Malmo, Sweden Accepted 14 March 2016 Abstract Aims People with Type 1 diabetes have smaller pancreases than healthy individuals. Several diseases causing pancreatic atrophy are associated with pancreatic steatosis, but pancreatic fat in Type 1 diabetes has not been measured. This cross- sectional study aimed to compare pancreas size and fat fraction in children with Type 1 diabetes and controls. Methods The volume and fat fraction of the pancreases of 22 children with Type 1 diabetes and 29 controls were determined using magnetic resonance imaging. Results Pancreas volume was 27% smaller in children with diabetes (median 34.9 cm 3 ) than in controls (47.8 cm 3 ; P < 0.001). Pancreas volume correlated positively with age in controls (P = 0.033), but not in children with diabetes (P = 0.649). Pancreas volume did not correlate with diabetes duration, but it did correlate positively with units of insulin/kg body weight/day (P = 0.048). A linear model of pancreas volume as influenced by age, body surface area and insulin units/kg body weight/day found that insulin dosage correlated with pancreas volume after controlling for both age and body surface area (P = 0.009). Pancreatic fat fraction was not significantly different between the two groups (1.34% vs. 1.57%; P = 0.891). Conclusions Our findings do not indicate that pancreatic atrophy in Type 1 diabetes is associated with an increased pancreatic fat fraction, unlike some other diseases featuring reduced pancreatic volume. We speculate that our results may support the hypotheses that much of pancreatic atrophy in Type 1 diabetes occurs before the clinical onset of the disease and that exogenous insulin administration decelerates pancreatic atrophy after diabetes onset. Diabet. Med. 00, 000000 (2016) Introduction People with Type 1 diabetes have lower pancreas volumes than individuals without diabetes [1]. Adults who have had Type 1 diabetes for only a few months have 2631% smaller pancreases than controls, suggesting that the pancreas diminishes in size prior to the clinical onset of diabetes [2,3]. In children and adolescents, the size of the pancreas decreases with increasing duration of Type 1 diabetes in relation to age-matched controls [4]. People with Type 1 diabetes have reduced markers of exocrine pancreatic func- tion, which correlate with decreased pancreas size [5,6]. Pancreatic steatosis has been associated with several diseases involving the pancreas, including Type 2 diabetes, pancreatitis, pancreatic cancer, cystic fibrosis and haemochro- matosis [7]. Twice as much fat was found in the pancreases of people with Type 2 diabetes as in those of age- and BMI- matched controls (~ 20% compared with 10% of the organ volume, respectively), and pancreatic fat correlated negatively with indicators of b-cell function [8]. In cystic fibrosis and haemochromatosis, pancreatic parenchymal damage and atrophy are followed by adipocyte infiltration [7,9]. As yet it is unclear whether Type 1 diabetes shares the association with increased pancreatic fat that is apparent in other conditions causing pancreatic dysfunction, damage and atrophy. Pancreatic fat has been positively correlated with liver fat [10]. This relationship seems to be mediated by general obesity, and it is currently not established whether such a relationship exists when the fat contents of the organs are low. We have previously reported that children with Type 1 Correspondence to: Simon E. Regnell. E-mail: simon.regnell@med.lu.se ª 2016 Diabetes UK 1 DIABETICMedicine DOI: 10.1111/dme.13115