# Portal hypertension and bleeding esophageal varices caused by portal vein thrombosis in children: experience of a single center with emphasis on treatments and results

## TL;DR

The study finds that endoscopic and surgical treatments for portal vein thrombosis in children have similar outcomes, suggesting a conservative approach is preferable.

## Contribution

The study compares endoscopic and surgical treatment outcomes for portal vein thrombosis in children, advocating for conservative management.

## Key findings

- Endoscopic and surgical treatments for portal vein thrombosis in children yield comparable results.
- A conservative treatment approach is recommended due to similar outcomes between treatment types.
- Platelet counts increased in the surgical group, but this did not affect overall treatment outcomes.

## Abstract

•Portal Vein Thrombosis (PVT) is a cause of gastrointestinal bleeding in children.•Results of endoscopic and surgical treatment are comparable.•We concluded that a conservative approach should be adopted whenever possible.

Portal Vein Thrombosis (PVT) is a cause of gastrointestinal bleeding in children.

Results of endoscopic and surgical treatment are comparable.

We concluded that a conservative approach should be adopted whenever possible.

Background and aim: Portal Vein Thrombosis (PVT) is the most common cause of Portal Hypertension (pH) in children. This study aimed to present the experience of a large university center by analyzing patients with portal hypertension caused by portal vein thrombosis in search of prognostic factors, types of treatments, and results.

Methods: We retrospectively analyzed 110 pediatric patients treated from 2000 to 2019, who were divided into two groups: endoscopic procedures (67 patients) and surgical assistance protocols (43 patients).

Results: Among the patients who underwent the surgical protocol, 22 underwent Meso-Rex Shunt (MRS) and 21 underwent other surgical techniques: distal splenorenal shunt (DSRS ‒ 11 patients), azigoportal dissociation with partial splenectomy (6 patients), and mesocaval shunt (4 patients). Five patients with meso‑Rex developed graft thrombosis (5/22, 22.72 %). There were no differences between the groups regarding endoscopic findings (first and last), no correlation with the need for surgery, and no definition of the initial endoscopic characteristics that could predict worsening disease. The only observed difference was the increase in platelet count in the surgical group (median 25.02 thousand/mL, interquartile interval 118.00), which was greater than that observed in the endoscopic group (-42.50 thousand/mL, interquartile interval 56, p < 0.001, Mann-Whitney test). Only umbilical catheterization could predict the need for exclusive endoscopic treatment.

Conclusions: As the evolution of children treated endoscopically was not different from that of children treated surgically, we conclude that a more conservative approach should be adopted, whenever possible.