Impact of preoperative internal versus external biliary drainage on surgical and oncological outcomes after pancreaticoduodenectomy: a systematic review and meta-analysis.

BACKGROUND

Preoperative biliary drainage is required for complicated malignant obstructive jaundice before pancreaticoduodenectomy (PD). However, the impact of internal (IntBD) versus external biliary drainage (ExtBD) on post-pancreatectomy clinical and oncological outcomes remains controversial.

METHODS

Systematic search of MEDLINE, EMBASE, CENTRAL, Web of Science, and ClinicalTrials.gov (through January 2025) identified studies comparing IntBD and ExtBD before PD. Meta-analyses were performed using random-effects modeling to calculate risk ratios (RRs), mean differences (MDs), and respective 95% confidence intervals. The primary outcome was overall postoperative morbidity; secondary outcomes included specific surgical complications and oncological endpoints.

RESULTS

6832 patients from 31 retrospective cohort studies and 1 randomized trial were analyzed. ExtBD was associated with a lower overall complication rate after PD (RR: 1.31 [1.08-1.59]; P = 0.006). Post-pancreatectomy specific complications were reduced after ExtBD: pancreatic fistula (RR: 1.56 [1.29-1.90]; P < 0.00001), delayed gastric emptying (RR: 1.84 [1.19-2.85]; P = 0.006), and post-pancreatectomy hemorrhage (RR: 1.56 [1.12-2.19]; P = 0.01). Infectious complications, including biliary tract (RR: 2.40 [1.54-3.76]; P = 0.0001) and surgical site infections (RR: 1.39 [1.20-1.61]; P < 0.00001), were lower, length of hospitalization after PD was shorter with ExtBD (MD: 2.53 [0.20-4.86]; P = 0.03). Narrative analysis of long-term follow-up suggests similar distant metastases rates, but higher local recurrence (RR: 0.41 [0.32-0.55]; P < 0.00001) and lower 5-year overall survival (RR: 0.67 [0.50-0.90]; P = 0.007) of patients with periampullary cholangiocarcinoma treated with preoperative ExtBD compared with IntBD. However, long-term outcomes are biased through oncologically unbalanced cohorts.

CONCLUSIONS

ExtBD was associated with lower postoperative morbidity after PD, whereas oncological outcomes remain uncertain. Given the low methodological quality and predominantly retrospective design of the included studies, these findings should be considered hypothesis-generating. Prospective trials are needed to clarify the optimal preoperative drainage strategy.