Introduction Endoscopic Ultrasound (EUS) is a well established modality, in particular, fine needle aspiration (FNA) is used to sample suspicious lesions. The uptake in EUS is not well documented with the number of procedures performed and their indication unknown. Understanding demand for a procedure is essential for allocating resources in an endoscopy service. Our primary aim was to examine the trends in EUS in our unit and to examine EUS-FNA sensitivity and specificity in pancreatic and non-pancreatic lesions.
Method Retrospective review of all EUS procedures from 2002 until 2013 was performed. Data collected included patient demographics, indication for EUS procedure (oesophago-gastric (OG) or pancreatobiliary (PB)) and obtained FNA samples. Pathology results of FNA specimens from 2011–2013 (n = 143) were analysed further with reference to number of passes into any lesion sampled. Sensitivity and specificity rates of FNA were calculated for pancreatic and non-pancreatic lesions respectively.
Results The number of EUS procedures rose from 54 in 2002 to 287 in 2013. EUS FNA started in 2006 with the purchase of a linear echoendoscope which was associated with increasing number of EUS. There was a further increase in numbers in 2011 with the appointment of an additional consultant with an interest in EUS. There was a dip in total numbers 2008–2009 due to technical issues. Overall, there was a significant increase in the number of PB EUS comparing 2002 with 2013 (9/54 vs. 207/287, p < 0.0001). There was a significant increase in the number of FNA performed comparing 2002 with 2013 (0/54 vs 48/287, p = 0.0006). EUS-FNA sensitivity for solid pancreatic lesions was 61.5%, for non-pancreatic lesions it was 66% but specificity was 100% in both groups of lesions. In solid pancreatic masses, diagnostic cytology was associated with increased numbers of passes into the lesion compared to non-diagnotic cytology (median 4 passes vs. median 2 passes, p = 0.001). There was no difference number of passes for non-pancreatic lesions with respect to diagnostic cytology rates.
Conclusion There has been a significant increase in the utility of EUS in our unit, particularly PB EUS. Our current EUS FNA sensitivity is not good enough but will change practice to have a minimum number of passes for each lesion type. Reassessment of this service will then be conducted to determine whether any other improvements are required.