Pancreatic Volume in Diabetes Mellitus
We read with great interest the article by Dr Sonmoon Mohapatra and colleagues,1 “Diabetes mellitus is associated with exocrine pancreatopathy: conclusions from a review of literature,” published in Pancreas on September 2016. In this review, the authors provide an overview of studies on morphological, structural, and functional changes in the exocrine pancreas in type 1 and type 2 diabetes mellitus (T1DM and T2DM). This is a relevant topic and merits further discussion. We noticed a few discrepancies in “Table 1, Pancreatic Volume by MR or CT in DM,” that may interfere with the interpretation of data.
First, values are in disagreement with the original data published by Bilgin et al.2 This study describes data on 5 patients with T1DM and 52 patients with T2DM, instead of 28 T1DM and none with T2DM, as shown in Table 1 of the article of Dr. Mohapatra and colleagues. Moreover, pancreas size, measured as the anteroposterior diameter of the organ, was compared between 3 groups: (a) group 1, 29 patients with diabetes (4 T1DM and 25 T2DM) and pancreatic exocrine insufficiency; (b) group 2, 28 patients with diabetes only (1 T1DM and 27 T2DM); and (c) group 3, 21 nondiabetic subjects with normal exocrine function. Therefore, it is not possible to assume differences in pancreas size between T1DM and controls, because T1DM and T2DM patients were mixed together in statistical analysis.
Second, in the study by Saisho et al,3 which compared pancreatic volume in T2DM patients and healthy subjects, we believe the correct mean (standard deviation, SD) values for pancreatic volume in the control group of 1721 subjects is 72.4 (25.8), instead of 74.9 (27). The value presented in Table 1 [74.9 (27)] refers to a subgroup of 660 healthy subjects matched for age and body mass index with the group of 165 T2DM patients.
Third, another study on the topic was published in 2015 and was not included in the systematic review.4 Macauley et al investigated volume, morphology, and composition of the pancreas by magnetic resonance in T2DM patients (n = 41) and healthy controls (n = 14) and found a 33% reduction in pancreas volume of diabetic patients compared with controls [55.5 (17.92) vs 82.6 (17.95); P < 0.0001], in line with other studies included in this systematic review.
In conclusion, study selection and data extraction is the cornerstone of a systematic review and might be carefully performed to guarantee correct interpretation of data.