Endocrine Abstracts

Rationale and purpose: Glucose tolerance often changes in patients with pancreatic lesions and usually worsens after distal pancreatectomy; however, factors affecting interindividual variation in progression are unclear. Our study aimed to investigate the glyco-metabolic status in a series of patients undergoing pancreatic resection for treatment of benign or malignant lesions and risk factors involved in the developing T3cDM.

Methods: We conducted an observational retrospective study on 29 patients (F=15; age, 67.4±2.2y, Charlson’s Comorbidity Index, 3.7±0.3) subjected to distal spleno-pancreatectomy (DP, (n=18), total duodeno-pancreatectomy (PD, (n=10) or radical proximal-distal modular pancreato-splenectomy (RAMPS, (n=1) in our university hospital. Pre- and post-surgical variables, glyco-metabolic status and the degree of glycemic control at diagnosis and during follow-up were assessed, together with pancreas autoimmunity and C-peptide levels. In a subgroup of T3cDM patients, glucose profiles obtained by continuous monitoring devices (CGM) were compared with those of an age- and sex-matched control group with T1DM.

Results: The overall prevalence of diabetes in our series was 62% (18/29), 50% of whom (9/18) developed it before surgery. In patients subjected to DP, T3cDM prevalence was 61% (11/18). Autoimmunity was negative in all cases. In the series as a whole, patients with diabetes were younger than patients without (63.3±2.6y vs 74.0±3.0y, P<0.05), but no other anthropometric, biochemical, surgical or clinical-pathological features were identified as risk factors for T3cDM. Among T3cDM patients, 60% were treated with basal-bolus insulin regimen while the remainders were treated with combination therapies. Analysis of glucose control over time showed mean HbA1c values of 8.59±0.42%. In a subgroup of insulin-treated T3cDM patients, CGM showed a higher glucose management indicator, an estimate of HbA1c, when compared to DMT1 patients (7.71±0.37% vs 6.74±0.20%, P<0.05). Moreover, the former displayed longer TAR (20.00±6.93% vs 5.00±1.96%, P=0.06) and shorter TBR (4.71±1.06% vs 1.00±0.44%, P<0.01). There were no differences in insulin dosing between the two subgroups.

Conclusions: In our series, the prevalence of T3cDM in patients who underwent DP was 50%, which coexisted with a high prevalence of pre-surgical diabetes. Lower age at pancreatic surgery emerged as the only predictor of diabetes onset. Appropriate peri-operative diabetic assessment is mandatory for all patients undergoing distal pancreatic resections, and CGM could aid physicians, patients and their care-givers in building a better insight on glycemic outcomes and complications in patients developing T3cDM.