Article Text


Basic science (liver)
PMO-116 EUS guided sampling of pancreatic malignancy
  1. D Lloyd1,
  2. A Higginson2,
  3. B Stacey3,
  4. H Shepherd1,
  5. H Gordon1
  1. 1Department of Gastroenterology, Royal Hampshire County Hospital, Winchester
  2. 2Department of Radiology, Portsmouth Hospitals NHS Trust, Portsmouth
  3. 3Department of Gastroenterology, Southampton University Teaching Hospital Foundation Trust, Southampton, UK


Introduction Endoscopic ultrasound (EUS) guided sampling of malignant pancreatic lesions is increasingly performed to confirm malignancy prior to chemotherapy and/or radiotherapy. Historically lesions have been sampled by fine needle aspiration (FNA) yielding cells for cytological analysis. Cook Medical has recently developed the ProCore needle for EUS guided fine needle biopsy (FNB). This study compared the diagnostic yield of Procore FNB and FNA in patients undergoing EUS guided sampling of suspected pancreatic malignancy in a tertiary EUS centre.

Methods All patients with suspected pancreatic malignancy undergoing EUS guided tissue sampling over a 1-year period from 1st January 2011 to 31st December 2011 were retrospectively identified from endoscopy records. Note was made of whether FNA or FNB were performed. Electronic records were reviewed to determine the results of FNA/FNB. Standard statistical tests were used to compare the diagnostic yield of FNA and FNB.

Results EUS guided sampling was performed on 51 suspected malignant pancreatic mass lesions. FNA was performed on 27 occasions; FNB was performed on 40 occasions. Fifteen lesions were sampled by both FNA and FNB. FNA yielded a sample sufficient for cytological analysis in 19 (70%) cases. Of these samples, cytology confirmed malignancy in 17 (89%) of cases. FNB yielded a sample sufficient for histological analysis in 27 (68%) cases. Of these, histology confirmed malignancy in 26 (96%) cases. There was no statistically significant difference in either the yield of analysable tissue or the yield of positive cytology/histology between FNA and FNB. In cases where both FNA and FNB were performed, both modalities confirmed malignancy in eight cases (53%) and both modalities failed to yield diagnostic tissue in three cases (20%). In two cases FNA was positive with insufficient tissue from FNB, and in two cases FNB was positive with FNA yielding insufficient tissue. The overall yield of FNB was only one in five of patients undergoing repeat sampling where the initial sample had been non-diagnostic, compared to two in three for FNA.

Conclusion An advantage of Procore FNB is that cytology specimen preparation in endoscopy is not required and the larger sample allows more extensive histological assessment. The overall positive yield of FNB in patients who underwent EUS guided sampling of a suspected pancreatic malignancy was 65% compared to 63% for FNA. The limitation to higher yield appears to be acquisition of sufficient tissue for histological analysis. The yield of FNB in repeat sampling is low suggesting that combined FNA & FNB should be performed in such situations, ideally with on site microscopy assessment to ensure adequate tissue acquisition.

Competing interests None declared.

Statistics from

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.