Diagnostic yield of EUS-guided through-the-needle biopsy in pancreatic cysts: a meta-analysis

Abstract

Background and aims: There is currently limited evidence supporting the use of EUS-guided through-the-needle biopsy for sampling pancreatic cystic lesions. The aim of this meta-analysis was to provide a pooled estimate of the diagnostic performance of through-the-needle biopsy for sampling of pancreatic cysts.

Methods: A bibliographic search on the main databases was performed in September 2019. Pooled effects were calculated using a random-effects model by means of the DerSimonian and Laird test. The primary outcome was sample adequacy. Additional endpoints were diagnostic accuracy, optimal histologic core procurement, mean number of needle passes, pooled specificity, and sensitivity. Adverse event rates were also analyzed.

Results: Eleven studies enrolling 490 patients were included. Eight articles compared through-the-needle biopsy with cytology/cystic fluid analysis. Most patients were female, and the body/tail was the most frequent location of cystic lesions. Sample adequacy with through-the-needle biopsy was 85.3% (78.2%-92.5%), and subanalysis performed according to cyst morphology, size, and location confirmed the result. Through-the-needle biopsy clearly outperformed FNA both in terms of sample adequacy (odds ratio, 4.83; 95% confidence interval, 1.63-14.31; P =.004) and diagnostic accuracy (odds ratio, 3.44; 95% confidence interval, 1.32-8.96; P =.01). The pooled diagnostic accuracy rate, sensitivity, and specificity of through-the-needle biopsy were 78.8%, 82.2%, and 96.8%, respectively. A mean of 3.121 (2.98-3.25) passes through the cyst was needed to obtain adequate histologic samples. The incidence rates of mild bleeding and pancreatitis were 4% and 2%, respectively.

Conclusion: Our meta-analysis speaks in favor of the use of through-the-needle biopsy as a safe and effective tool in EUS-guided tissue acquisition of pancreatic cysts.