New-Onset Diabetes After Pancreaticoduodenectomy: 3D Volumetric Analysis and Evaluation of Preoperative Metabolic Load

Objectives: Advanced 3D volumetry is increasingly utilized to predict postoperative outcomes. It is widely hypothesized that remnant pancreatic volume and progressive atrophy dictate the loss of endocrine function after pancreaticoduodenectomy (PD). This study aims to investigate this anatomical paradigm by evaluating whether 3D volumetric tissue loss or the baseline metabolic load drives new-onset diabetes mellitus (NODM). Materials and Methods: This retrospective study included 159 non-diabetic patients who underwent PD. Preoperative and postoperative (6-month) computed tomography (CT) images were evaluated using Synapse 3D 6.7 version volumetric analysis software to calculate total pancreas volume, remnant volume, and postoperative progressive atrophy rate. All measured volumes were indexed to body weight. Factors associated with NODM were analyzed using multivariate logistic regression and ROC analysis. Results: Postoperative follow-ups revealed that 30.2% (n = 48) of the patients developed NODM. Interestingly, advanced 3D volumetric analysis revealed no significant differences between the diabetic and non-diabetic cohorts regarding body weight-indexed total pancreas volume, preoperative remnant volume, or postoperative progressive atrophy rate (p = 0.475). Instead, multivariate analysis identified the preoperative metabolic load, represented by Body Mass Index (BMI), as the sole independent predictor of NODM (p = 0.006, OR = 1.125), independent of surgical tissue loss. ROC analysis demonstrated a moderate predictive threshold for BMI > 26.5 kg/m² (AUC = 0.644). Conclusions: In this cohort, the anatomical volume and progressive atrophy of the remnant pancreas, despite precise 3D quantification, were not statistically associated with the development of NODM after PD. Postoperative endocrine failure appears to be strongly associated with the patient’s baseline metabolic stress rather than solely the physical extent of surgical tissue loss. These findings suggest that while anatomical tissue preservation remains biologically important, preoperative metabolic risk stratification should be a clinical priority.