Management and Outcomes of Bleeding Ectopic Varices in Patients With Liver Cirrhosis: A Systematic Review and Pooled Analysis
Angela Zhu, Kiran Rasaratnam, Ammar Majeed, William Kemp, Warren Clements, Stuart K. RobertsABSTRACT
Background
Bleeding ectopic varices is a rare but life‐threatening complication in patients with clinically significant portal hypertension and liver cirrhosis, with limited evidence available to guide treatment decisions.
Aim
To evaluate the safety and efficacy of current therapies for bleeding ectopic varices.
Methods
A systematic review with pooled analysis was performed based on searches conducted in Embase, MEDLINE and Cochrane Library databases. Case series, retrospective and prospective studies including adults with a history of liver cirrhosis or portal hypertension presenting with bleeding ectopic varices were included. Individual patient data on interventions, rebleeding, complications and mortality were pooled and summarised using descriptive analysis. No conventional meta‐analysis was conducted due to the heterogeneous nature of included studies.
Results
44 studies (15 non‐consecutive case series, 25 single‐arm retrospective cohort studies, 3 double‐arm retrospective cohort studies and 1 single‐arm prospective study) involving 458 patients met selection criteria. Alcohol‐related liver disease (38.7%), viral hepatitis (18.8%) and primary sclerosing cholangitis (15.7%) were the most common aetiologies. Child‐Pugh A, B and C cirrhosis accounted for 29.7%, 46.7% and 23.5% of patients, respectively, and the pooled median Model for End Stage Liver Disease score was 14.0 (IQR 10.0–20.8). The most reported sites of ectopic varices were duodenal (27.7%), rectal (27.3%) and stomal/peri‐stomal (23.1%). Endoscopic sclerosant injection appeared to have a lower rebleeding rate than band ligation in the treatment of bleeding rectal varices (3.3% vs. 33.3%). Across variceal subtypes, transjugular intrahepatic portosystemic shunt (TIPS) combined with variceal embolisation, was observed to have a lower rebleeding rate than TIPS alone. The pooled all‐cause mortality following small bowel, rectal and stomal variceal bleeding were 25.0%, 42.9%, and 66.6%, respectively, over a median follow‐up period of 354 (IQR 120–780) days. Small sample sizes, missing data and significant heterogeneity between studies limited data quality.
Conclusion
Bleeding ectopic varices frequently recurs and carries high mortality. An individualised approach in conjunction with early multidisciplinary team discussions is essential in achieving favourable patient outcomes.