Pancreatic cancer is a highly lethal disease, often diagnosed at an advanced stage when onlypalliative care is available. Endoscopic retrograde cholangiopancreatography (ERCP) withbrush cytology or forceps biopsy is commonly used to obtain samples from the bile duct (BD)and pancreatic duct (PD) in patients presenting with distal biliary strictures; however, itsdiagnostic accuracy remains debated. This study aimed to assess the pooled sensitivity of BDand PD brushing, as well as BD forceps biopsy, in detecting pancreatic cancer compared tohistology.

A systematic search was conducted in three major medical databases on November 05, 2024.We included studies reporting the performance measures of BD and/or PD brushing or forcepsbiopsy, primarily based on histology, but also considering radiological follow-up, laparotomy,or other accepted diagnostic methods. A random-effects meta-analysis was performed tocalculate the pooled sensitivity of each technique for various cytopathological interpretations,with results presented as percentages and 95% confidence intervals (CI).

Our systematic search identified 11340 articles, of which 83 met the inclusion criteria for thefinal analysis. These included 6,638 BD brushing procedures, 1,089 PD brushing procedures,and 3,416 BD forceps biopsies. The pooled sensitivity for PD brushing was 53% (95% CI:28% – 76%; I 2 = 70.3%), for BD brushing 28% (95% CI: 23% – 34%; I 2 = 58.8%), for BDforceps biopsy 36% (95% CI: 29% – 45%; I 2 = 0%) when only clearly malignant sampleswere considered positive. Additional analyses were performed to explore differences ofdifferent cytopathological interpretations. The post-ERCP pancreatitis rate did not differbetween index tests, and it remained under 5% for each. However, data on the safety andadverse events associated with ERCP were limited among the included studies.

PD brushing may increase the sensitivity compared to BD brushing during ERCP and providea valuable ancillary diagnostic technique in detecting pancreatic cancer in patients with distalbiliary strictures. Moreover, PD brushing may not elevate the risk for post-ERCP pancreatitisin this population. These findings support the integration of brush cytology in ERCPprocedures to enhance early detection of pancreatic cancer, ultimately facilitating moreinformed clinical decision-making.