Accuracy and quality assessment of EUS-FNA: A single-center large cohort of biopsies

Abstract

Introduction. Thorough quality control (QC) study with systemic monitoring and evaluation is crucial to optimizing the effectiveness of EUS-FNA. Methods. Retrospective analysis was composed of investigating consecutive patient files that underwent EUS-FNA. QC specifically focused on diagnostic accuracy, impacts on preexisting diagnoses, and case management. Results. 268 patient files were evaluated. EUS-FNA cytology helped establish accurate diagnoses in 92.54 (248/268) of patients. Sensitivity, specificity, PPV, NPV, and accuracy were 83, 100, 100, 91.6, and 94, respectively. The most common biopsy site was the pancreas (68). The most accurate location for EUS-FNA was the esophagus, 13/13 (100), followed by the pancreas (89.6). EUS-FNA was least informative for abdominal lymph nodes (70.5). After FNA and followup, eight false negatives for tumors were found (3), while 7.5 of samples still lacked a definitive diagnosis. Discussion. QC suggests that the diagnostic accuracy of EUS-FNA might be improved further by (1) taking more FNA passes from suspected lesions, (2) optimizing needle selection (3) having an experienced echo-endoscopist available during the learning curve, and (4) having a cytologist present during the procedure. QC also identified remediable reporting errors. In conclusion, QC study is valuable in identifying weaknesses and thereby augmenting the effectiveness of EUS-FNA. © 2012 Benjamin Ephraim Bluen et al.