Document resource
Introduction Endoscopic ultrasound fine needle biopsy (EUS FNB) is the modality of choice for tissue diagnosis in the gastrointestinal tract. A new EUS FNB needle (Cook) has been developed, but there is a paucity of data from clinical application. The aim of this study was to assess the diagnostic performance of this new needle for tissue acquisition.Methods Patients who underwent EUS-FNB using the new needle from 10 centres in the UK were included. Demographic detail, lesion characteristics and histological outcome were recorded for all lesions and solid pancreatic lesions respectively. Diagnostic performance was assessed including calculation of sensitivity, specificity, positive and negative predictive value and accuracy. Comparison was made with a similar cohort of patients who underwent EUS-FNB using a previously described alternative EUS FNB needle (Medtronic)Results 520 patients (median age 69 years, 286 males) were included for analysis; 326 with the new needle and 194 controls. 378/520 (72.3%) were pancreatic lesions with 263/378 (69.6%) diagnosed with malignancy. There were no differences in median age, sex distribution, proportion with malignancy or adverse event rate between the two groups. There was a higher number of passes with the new needle (median 3 vs. 2, p<0.01). In all lesions, the diagnostic performance was not significantly different to the control needle (see table 1) with both demonstrating sensitivity of 87.3% and 83.6% respectively for all lesions and 87.8% and 88.7% respectively for solid pancreatic lesions. Overall accuracy was 89.9% and 87.0% for the new needle and control needle respectively and for solid pancreatic lesions 89.9% and 90.7% for the new needle and control needle.Conclusions This represents the largest study to date assessing this new needle. The needle is not inferior to the existing control needle and shows comparable diagnostic performance. This was a retrospective analysis, but with significant patient volumes from multiple centres. Further studies to assess ease of use and comparison of cost will help guide future needle selection.Abstract O76 Table 1New needle and control needle diagnostic performanceNeedleSensitivitySpecificityPositive predictive valueNegative predictive valueAccuracyClearcore all lesions87.310010067.489.9Sharkcore all lesions83.610010063.187.0Clearcore pancreatic lesions87.810010061.589.9Sharkcore pancreatic lesions88.710010065.690.7