Development of a prediction model of pancreatic fistula after duodenopancreatectomy and soft pancreas by assessing the preoperative image

Abstract

Background: Pancreatic consistency is one of the most widely accepted risk factors of clinically relevant postoperative pancreatic fistula (CR-POPF) after pancreatectoduodenectomy (PD). The present study aims to identify preoperative characteristics from the preoperative computed tomography (CT) associated with an increased risk.

Methods: Retrospective observational cohort study of patients who underwent PD surgery (January 2010-2019) were enrolled. All patients with available preoperative imaging were included; 103 met the inclusion criteria. Several parameters were measured on preoperative abdominal CT: retrorenal adipose tissue; abdominal perimeter; total adipose tissue, visceral and subcutaneous; skeletal muscle mass; main pancreatic duct (MPD) diameter; pancreatic thickness; remnant pancreatic volume; pancreatic attenuation (pancreas-to-spleen ratio). Primary endpoints were the association of radiological variables with soft pancreatic consistency and POPF development. All variables possibly associated with POPF and soft pancreas were subsequently included into a multivariable logistic regression model.

Results: Soft pancreas consistency was found in 43 patients (41.7%) and CR-POPF was higher (51.2% vs. 18%, p < 0.001). Multivariable analysis identified MPD ≤ 3 mm (OR = 7.2, 95%CI 2.3-23, p = 0.001), a remnant pancreatic volume ≥ 20 cm³ (OR = 6.4, 95%CI 2-21, p = 0.041), pancreas-to-spleen < 0.8 (OR = 3.2, 95%CI 1.2-8.4, p = 0.039), and retrorenal adipose tissue ≥ 12 cm³ (OR = 5.3, 95%CI 1.8-15.7, p = 0.013). Multivariable analysis showed MPD ≤ 3 mm (OR = 8.25, 95%CI 2.2-30.8, p = 0.002) and total adipose tissue ≥ 190 cm³ (OR = 3.2, 95%CI 1.1-9.1, p = 0.0027) were independent predictors of CR-POPF.

Conclusion: The preoperative assessment of MPD, remnant pancreatic volume, pancreas-to-spleen ratio, total adipose tissue, and retrorenal adipose tissue are associated with soft pancreas texture and the risk of CR-POPF.

Keywords: Anastomotic leak; Computed tomography; Pancreatic fistula; Pancreatoduodenectomy; Soft pancreas.