Therapeutic Endoscopic Intervention in Cirrhosis: Routine Clinical, Laboratory and Ultrasound Markers in a Retrospective Cohort Study
Background and Aims: Upper gastrointestinal endoscopy remains the reference standard for detecting and grading esophageal varices in cirrhosis, but repeated surveillance is invasive and resource-intensive. Most non-invasive strategies focus on ruling out high-risk varices. We assessed whether routinely available clinical, laboratory and ultrasound parameters are associated with clinically relevant endoscopic findings, focusing on therapeutic endoscopic intervention as a pragmatic real-world endpoint. Methods: This retrospective single-center cohort study included adults with confirmed liver cirrhosis treated at a German tertiary care hospital between 2014 and 2021. Only the first examination episode per patient was analyzed. Gastroscopy, abdominal ultrasound and laboratory testing were eligible when performed during the same hospital stay. The primary endpoint was therapeutic intervention during upper gastrointestinal endoscopy. Secondary endpoints were presence of esophageal varices and Paquet grade II/III varices. These summaries were data-derived, unweighted and not internally or externally validated. Discrimination was assessed using receiver operating characteristic analysis, and logistic regression was used to estimate odds ratios with 95% confidence intervals (CI). Results: Among 257 patients, esophageal varices were present in 235 (91.4%), and therapeutic endoscopic intervention was performed in 38 (14.8%). The intervention point-based summary, including ascites, hemoglobin <9.9 g/dL and alanine aminotransferase >33.5 U/L, was calculable in 248 patients and showed moderate apparent discrimination for intervention, with an AUC of 0.706. Each additional point was associated with higher odds of intervention (OR=2.80, 95%CI: 1.73–4.53; p<0.001). In multivariable analysis, hemoglobin <9.9 g/dL and alanine aminotransferase >33.5 U/L remained associated with intervention, whereas ascites did not. Secondary point-based summaries for esophageal varices and Paquet grade II/III varices showed apparent AUCs of 0.775 and 0.623, respectively, but were limited by complete-case sample size and endpoint distribution. Conclusions: Routine clinical, laboratory and ultrasound parameters were associated with clinically relevant endoscopic findings in cirrhosis, particularly therapeutic endoscopic intervention. Low hemoglobin and elevated alanine aminotransferase were associated with intervention need, but the findings likely reflect clinical presentation and therapeutic endoscopic relevance rather than portal pressure alone. The exploratory point-based summaries are descriptive, data-derived tools and should not replace established non-invasive strategies or endoscopic assessment. Prospective external validation is required before clinical use.
Authors
- Dominik Marschner (ORCID: https://orcid.org/0009-0003-1639-8762)
- Andreas Lienig
- Christian Benignus (ORCID: https://orcid.org/0000-0002-6167-6798)
- Wolfram G. Zoller (ORCID: https://orcid.org/0009-0006-9941-6787)
- Julian Müller Kühnle
Institutions
- University Medical Center Freiburg (DE)
- Klinikum Ludwigsburg (DE)
- Klinikum Stuttgart (DE)
Publication Details
- Journal
- Journal of Gastrointestinal and Liver Diseases
- Published
- 2026-11-21
- DOI
- https://doi.org/10.15403/jgld-7486
- Primary Topic
- Liver Disease and Transplantation
- Type
- article
- Field-Weighted Citation Impact
- 0.00