Original article
Clinical endoscopy
EUS-FNA is superior to ERCP-based tissue sampling in suspected malignant biliary obstruction: results of a prospective, single-blind, comparative study

https://doi.org/10.1016/j.gie.2013.12.031Get rights and content

Background

Both EUS and ERCP sampling techniques may provide tissue diagnoses in suspected malignant biliary obstruction. However, there are scant data comparing these 2 methods.

Objective

To compare EUS-guided FNA (EUS-FNA) and ERCP tissue sampling for the diagnosis of malignant biliary obstruction.

Design

Prospective, comparative, single-blind study.

Setting

Tertiary center.

Patients

Fifty-one patients undergoing same-session EUS and ERCP for the evaluation of malignant biliary obstruction over a 1-year period.

Interventions

EUS-FNA and ERCP tissue sampling with biliary brush cytology and intraductal forceps biopsies.

Main Outcome Measurements

Diagnostic sensitivity and accuracy of each sampling method compared with final diagnoses.

Results

EUS-FNA was more sensitive and accurate than ERCP tissue sampling (P < .0001) in 51 patients with pancreatic cancers (n = 34), bile duct cancers (n = 14), and benign biliary strictures (n = 3). The overall sensitivity and accuracy were 94% and 94% for EUS-FNA, and 50% and 53% for ERCP sampling, respectively. EUS-FNA was superior to ERCP tissue sampling for pancreatic masses (sensitivity, 100% vs 38%; P < .0001) and seemed comparable for biliary masses (79% sensitivity for both) and indeterminate strictures (sensitivity, 80% vs 67%).

Limitations

Single-center study.

Conclusion

EUS-FNA is superior to ERCP tissue sampling in evaluating suspected malignant biliary obstruction, particularly for pancreatic masses. EUS-FNA appears similar to ERCP sampling for biliary tumors and indeterminate strictures. Given the superior performance characteristics of EUS-FNA and the higher incidence of pancreatic cancer compared with cholangiocarcinoma, EUS-FNA should be performed before ERCP in all patients with suspected malignant biliary obstruction. (Clinical trial registration number: NCT01356030.)

Section snippets

Methods

At our center, same-session EUS and ERCP are routinely offered for all patients with suspected pancreaticobiliary pathology. All patients with suspected malignant biliary obstruction based on clinical presentation of painless jaundice with elevated levels on liver tests in a cholestatic pattern and evidence of biliary obstruction, stricture, or pancreatic/biliary mass on preprocedure imaging (contrast CT or magnetic resonance imaging) were invited to participate in the study. Patients with

Results

Between May 2011 and June 2012, a total of 77 patients with clinical suspicion for malignant biliary obstruction provided informed consent to participate in this prospective study. After the initial EUS, 26 patients patients were excluded from the study for the following reasons: (1) EUS-FNA provided on-site diagnosis of a resectable neoplasm, patient referred for expedited surgery without ERCP drainage (n = 14); (2) biliary stricture not present on ERCP (stones or other cause of biliary

Discussion

We directly compared EUS-FNA with ERCP-based tissue sampling in a large series of unselected patients with suspected malignant biliary obstruction. Surprisingly, there is a paucity of studies that compare these sampling techniques. Oppong et al15 performed a retrospective analysis of EUS-FNA compared with ERCP brushings in a series of 37 patients with suspected malignant obstruction. In their study, ERCP was performed before EUS-FNA, procedures were performed in a single session in only 56% of

References (30)

Cited by (114)

  • Endoscopic Diagnosis of Cholangiocarcinoma

    2022, Techniques and Innovations in Gastrointestinal Endoscopy
    Citation Excerpt :

    EUS-FNA identified 100% of dCCA and 83% of pCCA.99 Several other studies showed similar findings.89–92,98–101 Moreover, the presence of an associated mass lesion significantly improves the diagnostic yield.

View all citing articles on Scopus

DISCLOSURE: All authors disclosed no financial relationships relevant to this publication.

If you would like to chat with an author of this article, you may contact Dr Shah at [email protected].

See CME section; p. 152.

View full text