To evaluate whether endoscopic ultrasound-guided tissue acquisition (EUS-TA) isassociated with an increased risk of peritoneal carcinomatosis in patients with hilarcholangiocarcinoma, and to assess its impact on time-to-peritoneal dissemination andoverall survival. A secondary aim was to analyse whether outcomes differed between operated and non-operated patients, and to compare results according to the type of needleused (FNA vs FNB).

This retrospective cohort study included 100 patients with hilar cholangiocarcinoma. Eightpatients with peritoneal carcinomatosis at baseline were excluded from the analysis ofdissemination risk. Clinical, radiological, and procedural variables were collected, includinguse of EUS-TA, surgical management, and needle type. Primary endpoints were: (1)development of peritoneal carcinomatosis, (2) time to peritoneal dissemination, and (3)overall survival. Analyses were performed using Fisher’s exact test for categorical variablesand Kaplan–Meier curves with log-rank testing for time-to-event outcomes. Subgroupanalyses were conducted for operated vs non-operated patients and for FNA vs FNB.

Among the 92 patients without baseline peritoneal disease, peritoneal carcinomatosisoccurred in 6/32 EUS-TA patients (11.7%) and 7/60 without EUS-TA (18.8%), with nosignificant difference (p = 0.36). Median time to peritoneal dissemination was similarbetween groups: 13.5 months (IQR 6.75–18.5) for EUS-TA vs 17 months (IQR 14–26) fornon-EUS (log-rank p = 0.165). Median overall survival was 301 days (IQR 95–566) withEUS-TA and 151 days (IQR 51.5–474) without EUS-TA (log-rank p = 0.60).In the operated subgroup, dissemination occurred in 2/11 (18.2%) with EUS-TA vs 5/30(16.7%) without EUS-TA (p = 0.20). In non-operated patients, dissemination rates were19.0% (4/21) with EUS-TA and 6.7% (2/30) without EUS-TA (p = 1.00). Time-to-dissemination and overall survival curves showed no significant differences within eithersubgroup.In the needle-type analysis, dissemination rates were similar between FNA (4/22, 18.2%)and FNB (2/10, 20.0%) (p = 1.00), with no differences in survival outcomes (all log-rank p >0.4).

EUS-guided tissue acquisition does not increase the risk of peritoneal carcinomatosis inpatients with hilar cholangiocarcinoma, nor does it affect time-to-dissemination or overall

survival. These findings remained consistent across operated and non-operated subgroups,suggesting that surgical status does not modify the oncological impact of EUS-TA.Additionally, the needle type (FNA vs FNB) did not influence dissemination or survival,supporting the safety of both sampling techniques.