Clinical Utility of EUS ‐ FNA / FNB for Diagnosis and Grade Assessment in Pancreatic Neuroendocrine Tumors < 20 mm: A Systematic Review and Meta‐Analysis

OBJECTIVE: Management of pancreatic neuroendocrine tumors (pNETs) < 20 mm remains controversial. Endoscopic ultrasound-guided tissue acquisition is crucial for diagnosis and grading, but data on small pNETs and Ki-67 sample adequacy are limited. This systematic review and meta-analysis evaluated sample adequacy, diagnostic yield, and histologic concordance with surgical specimens. METHODS: A systematic search of MEDLINE, EMBASE, and SCOPUS was conducted. Studies reporting diagnostic yield or concordance between EUS-guided fine-needle aspiration (FNA) or biopsy (FNB) and surgical histology in pNETs < 20 mm were included. The primary outcomes were diagnostic yield, concordance with surgical histology, and adequacy of samples for Ki-67-based tumor grading. Pooled estimates were calculated using random-effects models. RESULTS: 27 studies were included: 17 reported sample adequacy, 9 diagnostic yield, 13 grade concordance, and 5 both yield and grading. Pooled diagnostic yield was 84.4% (95% CI: 0.80-0.87) and pooled concordance rate was 84.2% (95% CI: 0.803-0.882). For sample adequacy to establish tumor grade, the pooled rate was 71.6%. Notably, EUS-FNB demonstrated significantly higher adequacy (79.8%) compared to FNA (52.5%). CONCLUSION: EUS-guided tissue acquisition offers high diagnostic accuracy and reliable tumor grading in pNETs < 20 mm. FNB outperforms FNA in obtaining adequate tissue for grading, underscoring its importance in clinical decision-making.