The diagnostic yield of endoscopic ultrasound-guided fine-needle biopsy in autoimmune pancreatitis: a systematic review and meta-analysis

Autoimmune pancreatitis (AIP) is a rare chronic form of pancreatitis mediated by autoimmune mechanisms. Evidence regarding the diagnostic performance of endoscopic ultrasound-guided fine-needle biopsy (EUS-FNB) in AIP remains limited. This study aimed to estimate the pooled diagnostic accuracy of EUS-FNB in patients with AIP. PubMed, Embase, Cochrane Library, CNKI, and Wanfang Database were systematically searched to identify fully published studies evaluating EUS-FNB for tissue sampling in AIP. Pooled estimates were generated using a random-effects model. The assessed outcomes included diagnostic accuracy, sample adequacy, contribution to AIP diagnosis, and adverse events. Sixteen studies, including one randomized controlled trial and four prospective studies, comprising 732 patients, were included. The overall diagnostic accuracy was 79% (73%–83%). Diagnostic accuracy was higher for the Franseen needle (83%, 75%–90%) and Fork-tip needle (90%, 80%–99%) compared with the reverse/forward‑bevel (51%, 31%–71%), with a statistically significant difference ( P < 0.001). The 22G needle also demonstrated superior diagnostic accuracy (85%, 79%–91%) compared with 19G/20G needles (70%, 42%–98%) ( P < 0.001). The pooled sample adequacy rate was 97% (95%–98%). EUS-FNB contributed to the diagnosis in 69% (50%–88%) of cases where diagnosis could not be established using imaging, serology, or involvement of other organs. The overall adverse event rate was 4% (1%–8%). EUS-FNB demonstrates excellent histological tissue acquisition and moderate diagnostic accuracy in AIP, with a low rate of adverse events. The 22G Franseen or Fork-tip needles may be a reasonable option, though direct comparative evidence is lacking.