Feasibility and yield of a new EUS histology needle: Results from a multicenter, pooled, cohort study
Background: EUS-guided FNA is an efficacious technique for sampling intraintestinal and extraintestinal mass lesions. However, cytology has limitations to its final yield and accuracy, which may be overcome if histological specimens are provided to the pathologist. Objective: To evaluate feasibility, yield, and diagnostic accuracy of a newly developed 19-gauge, fine-needle biopsy (FNB) device. Design: Multicenter, pooled, cohort study. Setting: Five medical centers. Patients: This study involved 109 consecutive patients with 114 intraintestinal or extraintestinal mass lesions and/or peri-intestinal lymph nodes. Intervention: EUS-guided FNB (EUS-FNB) with a newly developed, 19-gauge, FNB device. Main Outcome Measurements: Percentage of cases in which pathologists classified the sample quality as optimal for histological evaluation and the overall diagnostic accuracy compared with a composite criterion-standard diagnosis. Results: We evaluated 114 lesions (mean [± standard deviation] size 35.1 ± 18.7 mm; 84 malignant [73.7%] and 30 [26.3%] benign). EUS-FNB was technically feasible in 112 lesions (98.24%). Sample quality was adequate for full histological assessment in 102 lesions (89.47%). In 98 cases (85.96%), diagnosis proved to be correct according to criterion-standard diagnosis. Sensitivity, specificity, positive predictive value, negative predictive value, and overall accuracy for diagnosis of malignancy were 90.2%, 100%, 100%, 78.9%, and 92.9%, respectively. Limitations: Use of a surrogate criterion-standard diagnosis, including clinical follow-up when no surgical specimens were available, mainly in benign diagnoses. Conclusion: Performing an EUS-FNB with a new 19-gauge histology needle is feasible for histopathology diagnosis of intraintestinal and extraintestinal mass lesions, offering the possibility of obtaining a core sample for histological evaluation in the majority of cases, with an overall diagnostic accuracy of over 85%.