Patients in groups with more severe vascular abnormalities measured by endoscopic ultrasound recorded sharply different rebleeding rates after tissue glue treatment for esophagogastric varices, according to a retrospective study of 56 people. Rebleeding within 12 months rose from 5.56% in the mild-abnormality group to 46.15% in the severe group, with a 25.00% rate across the study population.
The findings point to a possible way of sorting patients by risk for individualized treatment decisions, but they do not show that the ultrasound-based system itself prevents bleeding or that using it to select treatment improves outcomes.
A clear gradient in outcomes
The investigators classified patients into mild, moderate and severe groups using vascular parameters derived from endoscopic ultrasound, or EUS. This measurement-based system was used to create a hierarchy of vascular abnormality, while the study tracked 12-month rebleeding as its primary outcome.
The difference between the groups followed a stepwise pattern. The reported rebleeding rate was 28.00% in the moderate group, compared with 5.56% in the mild group and 46.15% in the severe group. The reported P value for the group comparison was 0.033.
A similar pattern appeared at six months, when variceal remission was reported in 83.33% of the mild group, 64.00% of the moderate group and 38.46% of the severe group. The comparison had a reported P value of 0.036.
One blood-flow measure stood out
When the researchers analyzed several factors together, portal vein flow velocity was identified as an independent predictor of rebleeding in the reported analysis. The study gave an odds ratio of 0.60, with a 95% confidence interval from 0.39 to 0.84 and a P value of 0.009. That result describes an association within the analysis, not proof that changing flow velocity would change a patient's risk.
The researchers also tested whether the EUS-based classification added information to the Child–Pugh, MELD and Baveno criteria. The reported net reclassification improvement was 0.243, with a P value below 0.001. In practical terms, the authors say the new stratification provided additional predictive information beyond those criteria.
Internal checks using bootstrap resampling and cross-validation reported model stability without overfitting, with an optimism-corrected area under the curve of 0.88. These checks assess performance within the available data, rather than testing the system in a separate patient population.
What remains untested
The study used retrospective clinical and EUS data, so its findings are associations and cannot establish that the stratification causes any change in bleeding or remission. The comparison was between EUS-defined severity groups, not a randomized treatment trial, and no separate untreated control group was reported.
The authors say larger-cohort external validation is still needed. The supplied analysis also does not report the detailed EUS definitions, thresholds, model coefficients, calibration details or handling of missing data. The study reports the total of 56 patients but not the sizes or baseline characteristics of the three severity groups.
The article is reported as a version-of-record publication. Its datasets are available from the corresponding author on reasonable request. The funding section names a Beijing Municipal Health Commission research program parallel project, and the authors reported no conflicts of interest or financial ties to disclose.
The retrospective study received ethics approval, and informed consent was waived because the patient information was anonymized.
Paper data and sources
Original title: Endoscopic ultrasound vascular parameter-based stratification for precise treatment of esophagogastric varices: a study on endoscopic ultrasound-guided tissue glue obliteration.
Authors: Zhao Peng, Ping Li
Journal/Repository: Surgical endoscopy
Status: Peer-reviewed
First online: 2026-08-20
DOI: 10.1007/s00464-026-13290-7
Original paper · Full text