Internal Versus External Drainage With a Pancreatic Duct Stent For Pancreaticojejunostomy During Pancreaticoduodenectomy for Patients at High Risk for Pancreatic Fistula: A Comparative Study.

BACKGROUND: The aim of this study was to determine whether internal or external drainage with a pancreatic duct stent is the optimal pancreaticojejunostomy method to prevent pancreatic fistula (PF) after pancreaticoduodenectomy (PD) for subgroups of patients at high risk for PF. MATERIALS AND METHODS: A total of 495 patients who underwent PD were reviewed. Univariate and multivariate analyses were used to identify risk factors for PF after PD. We further compared the incidence of PF and outcomes between the internal and external drainage groups for subgroups of patients at high risk for PF. RESULTS: There was no difference in the incidence of complications according to the Clavien-Dindo classification or the rate of PF after PD in both groups (P = 0.961 and P = 0.505, respectively). The incidence of mortality was 3.8% in the internal drainage group and 3.9% in the external drainage group (P = 0.980). Univariate and multivariate analyses identified male gender (odds ratio [OR] = 2.93; 95% confidence interval [CI], 1.78-4.83; P = 0.000), pancreatic duct diameter (<3 mm) (OR = 2.58; 95% CI, 1.57-4.23; P = 0.000), and soft pancreatic texture (OR = 2.92; 95% CI, 1.71-4.98; P = 0.000) as independent risk factors for PF after PD. No differences in the incidence of PF for the subgroups of patients with one, two, or three risk factors were observed between the internal and external drainage groups (P = 0.334, P = 1.000, and P = 0.936, respectively). No differences in total complications, delayed gastric emptying, postpancreatectomy hemorrhage, biliary fistula, infection complications, reoperation, perioperative mortality, or postoperative hospital stay were noted. In addition, liquid loss and tube-related complications occurred in the external drainage group. CONCLUSIONS: Internal drainage is the optimal method to prevent PF after PD for subgroups of patients at high risk for PF because the surgical procedure is simple and prevents liquid loss and tube-related complications associated with external drainage. However, no differences in the incidence of PF and other complications after PD were observed between the two approaches.

Outcomes of preoperative endoscopic nasobiliary drainage and endoscopic retrograde biliary drainage for malignant distal biliary obstruction prior to pancreaticoduodenectomy.

AIM: To compare the outcomes of preoperative endoscopic nasobiliary drainage (ENBD) and endoscopic retrograde biliary drainage (ERBD) in patients with malignant distal biliary obstruction prior to pancreaticoduodenectomy (PD). METHODS: Data from 153 consecutive patients who underwent preoperative endoscopic biliary drainage prior to PD between January 2009 and July 2016 were analyzed. We compared the clinical data, procedure-related complications of endoscopic biliary drainage (EBD) and postoperative complications of PD between the ENBD and ERBD groups. Univariate and multivariate analyses with odds ratios (ORs) and 95% confidence intervals (95%CIs) were used to identify the risk factors for deep abdominal infection after PD. RESULTS: One hundred and two (66.7%) patients underwent ENBD, and 51 (33.3%) patients underwent ERBD. Endoscopic sphincterotomy was less frequently performed in the ENBD group than in the ERBD group (P = 0.039); the EBD duration in the ENBD group was shorter than that in the ERBD group (P = 0.036). After EBD, the levels of total bilirubin (TB) and alanine aminotransferase (ALT) were obviously decreased in both groups, and the decreases of TB and ALT in the ERBD group were greater than those in the ENBD group (P = 0.004 and P = 0.000, respectively). However, the rate of EBD procedure-related cholangitis was significantly higher in the ERBD group than in the ENBD group (P = 0.007). The postoperative complications of PD as graded by the Clavien-Dindo classification system were not significantly different between the two groups (P = 0.864). However, the incidence of deep abdominal infection after PD was significantly lower in the ENBD group than in the ERBD group (P = 0.019). Male gender (OR = 3.92; 95%CI: 1.63-9.47; P = 0.002), soft pancreas texture (OR = 3.60; 95%CI: 1.37-9.49; P = 0.009), length of biliary stricture (≥ 1.5 cm) (OR = 5.20; 95%CI: 2.23-12.16; P = 0.000) and ERBD method (OR = 4.08; 95%CI: 1.69-9.87; P = 0.002) were independent risk factors for deep abdominal infection after PD. CONCLUSION: ENBD is an optimal method for patients with malignant distal biliary obstruction prior to PD. ERBD is superior to ENBD in terms of patient tolerance and the effect of biliary drainage but is associated with an increased risk of EBD procedure-related cholangitis and deep abdominal infection after PD.