Pancreatic cystic lesions (PCLs) are increasingly detected due to the widespread use of cross-sectional imaging. Endoscopic ultrasound (EUS) is the preferred modality for evaluating their nature and malignancy risk, yet fluid analysis and cytology offer limited sensitivity. Through-the-needle biopsy (TTNB) has emerged as a more accurate diagnostic tool, though associated with higher adverse event (AE) rates. In 2021, our center implemented a selective TTNB protocol excluding frail or elderly patients and suspected IPMNs, and standardizing the procedure to two passes, complete cyst aspiration, and selective antibiotic prophylaxis. This study aimed to compare AE rates before and after protocol implementation, evaluate safety factors including antibiotic use, and assess TTNB adequacy and diagnostic accuracy. 

We retrospectively analyzed consecutive patients referred for TTNB at AOUI Verona between March 2016 and March 2025, dividing them into two groups: before (Group A) and after (Group B) protocol adoption. Patients not punctured due to technical issues, lack of indication, or presumed pseudo-cystic nature were excluded.

Of 970 patients evaluated by EUS, 190 underwent TTNB (100 in Group A, 90 in Group B). Lesions were mainly located in the pancreatic body or tail, with significantly larger size in Group B. Overall AE rate was 6.3%, significantly higher in Group A (11%) than in Group B (1%). Antibiotic prophylaxis was not associated with AE occurrence. TTNB adequacy was 88.9%, and diagnostic accuracy 75.3%. Among 68 surgical cases, TTNB was accurate in 79.4%.

A selective and standardized TTNB approach significantly reduces AEs while maintaining high adequacy and diagnostic accuracy.