Yield of ERCP tissue sampling of malignant biliary strictures by brush, forceps, and needle aspiration methods

J. Jailwala, S. Sherman, K. Gottlieb, S. Ikenberry, F. Gress, D. Earle, J. Flueckiger, G. Lehman

Research output: Contribution to journalArticlepeer-review

5 Scopus citations


Brush cytology of biliary strictures is common practice at ERCP but has low sensitivity for cancer detection. This series reviews the yield of 3 biliary stricture tissue sampling techniques applied at ERCP METHODS: Patients with biliary strictures underwent triple tissue sampling at 1 ERCP session. One Geenen cytology brush was passed back and forth across the stricture and removed with its sheath. The Howell aspiration needle was passed to the lower rim of the stricture and thrust into the tissue 2 times. Biopsy forceps were passed to the lower rim of the stricture and 3-4 bites taken. From 1990-95, 104 cancer patients underwent ERCP with triple tissue sampling. Final cancer diagnosis was based on all sampling methods plus surgery, autopsy, and clinical follow-up. Tissue specimens were reported as benign, malignant or atypia (mild, moderate, high grade). Nearly all atypia cases eventually were shown to have cancer and, therefore, were tallied as cancer. RESULTS: This table shows the frequency of positive sampling in patients with malignant strictures undergoing 3 simultaneous sampling methods. Sampling Technique Pancreatic Carcinoma n=46 Cholangio- carcinoma n=30 Ampullary Cancer n=13 Metasteses n=15 All Cancers n=104 Brush 52% 47% 38% 47% 48% (a) FNA 37% 37% 38% 40% 38% (b) Biopsy 52% 50% 77% 47% 54% (c) Brush+FNA 61% 57% 46% 53% 57% (d) Brush+Biopsy 72% 70% 77% 67% 71%(e) FNA+Biopsy 63% 60% 85% 60% 64% (f) Brush+FNA+ 78% 73% 85% 73% 77% (g) Biopsy p < .05 for: a vs. e, f, g; b vs. c, d, e, f, g; c vs. e, f, g, d vs. g, f vs. g The cumulative sensitivity of triple tissue sampling decreased to 62% if only high grade atypia was considered as cancer and 52% if no atypia was considered as cancer. SUMMARY: 1) Tissue sampling sensitivity varied according to the type of cancer; 2) Combining a second or third method increased sensitivity CONCLUSION: 1) General use of at least 2 tissue sampling methods is recommended; 2) Improvements in tissue sampling methods, such as better devices and/or more specimens/session, are needed to bring biliary sampling up to the 90% sensitivity level commonly found in UGI or colon biopsies.

Original languageEnglish (US)
Number of pages1
JournalGastrointestinal endoscopy
Issue number4
StatePublished - Jan 1 1996

ASJC Scopus subject areas

  • Radiology Nuclear Medicine and imaging
  • Gastroenterology

Fingerprint Dive into the research topics of 'Yield of ERCP tissue sampling of malignant biliary strictures by brush, forceps, and needle aspiration methods'. Together they form a unique fingerprint.

Cite this