Endoscopic variceal ligation as primary prophylaxis for upper GI bleeding in children

Efficacy and Safety of Endoscopic Primary Prophylaxis of Bleeding in Children With High-Risk Gastroesophageal Varices

OBJECTIVES

Primary prophylaxis of bleeding is debated in children with gastroesophageal varices; one of the reasons is the limited number of studies concerning its efficacy and safety. We report our experience with endoscopic primary prophylaxis.

METHODS

From 2006 to 2019, 145 children (median age, 3.5 years; cirrhosis, n = 116) with high-risk gastroesophageal varices underwent primary prophylaxis (banding, n = 114; sclerotherapy n = 31, primarily in smaller children).

RESULTS

We observed the eradication of varices in 93% of children after a mean of 6 months, at least one recurrence of varices in 45% after eradication, and gastrointestinal bleeding in 17% of children. Irrespective of the cause of portal hypertension, grade 3 esophageal varices, presence of gastric varices along the cardia and a lower composite score of endoscopic severity were associated with a worse probability of eradication, a longer time to eradication and a lower risk of a first recurrence and of bleeding following the procedure, respectively. Ten-year probabilities of overall survival and of bleeding-free survival were 95% and 75%, respectively.

CONCLUSIONS

Endoscopic primary prophylaxis of variceal bleeding is reasonably effective and safe in children with high-risk gastroesophageal varices. Worse results are observed in children with more advanced endoscopic features. This pleads for endoscopic screening in children with portal hypertension and early detection of varices warranting primary prophylaxis.

Associations Between Endoscopic Primary Prophylaxis and Rebleeding in Liver Cirrhosis Patients with Esophagogastric Variceal Bleeding

Aim

To identify the association between endoscopic primary prophylaxis and the risk of rebleeding in patients with liver cirrhosis receiving endoscopic therapy.

Methods

This cohort study involved in 944 liver cirrhosis patients with esophagogastric variceal bleeding (EGVB) receiving endoscopic therapy. All participants were divided into two groups: rebleeding group (n = 425) and non-rebleeding group (n = 519) according to the occurrence of rebleeding in patients. Rebleeding indicated any bleeding after endoscopic therapy for the first bleeding of esophagogastric varices in liver cirrhosis patients. Univariate and multivariate logistic analyses were employed to identify the association between endoscopic primary prophylaxis and rebleeding in patients with liver cirrhosis after endoscopic therapy.

Results

In total, 425 patients rebleeded at the end of the follow-up. The risk of rebleeding in patients with endoscopic primary prophylaxis decreased by 0.773 times (OR = 0.227, 95%CI: 0.139–0.372, P < 0.001) after adjusting covariables. Subgroups were divided according to the Child-Pugh (CP) score, and the results revealed that the risk of rebleeding in patients with endoscopic primary prophylaxis decreased by 0.858 times in Grade A patients (OR = 0.142, 95%CI: 0.066–0.304, P < 0.001) and 0.804 times in Grade B patients (OR = 0.196, 95%CI: 0.085–0.451, P < 0.001) compared with patients without endoscopic primary prophylaxis, but showed no difference in Grade C patients.

Conclusion

Endoscopic primary prophylaxis was associated with a decreased risk of rebleeding in liver cirrhosis patients with EGVB after endoscopic therapy, which suggested that clinicians should pay more attention to endoscopic primary prophylaxis to prevent the occurrence of rebleeding in these patients.

A Model for Early Endoscopic Detection of High-Risk Gastroesophageal Varices in Children With Biliary Atresia

OBJECTIVE

In children with biliary atresia and portal hypertension, progression to gastroesophageal varices carrying a risk of bleeding depends on age, total serum bilirubin concentration and initial endoscopic features. We report an attempt to use these factors for early detection of high-risk varices (HRVs).

METHODS

Based on different combinations of these factors, a model was set to estimate the probabilities of emergence of HRVs at various time intervals. A 10% probability was chosen to set the date of the next endoscopy in children who did not display HRVs initially. A total of 113 children without HRVs who underwent their first endoscopy before age 8 in 2013-2020 were included. A comparison was made with children seen during the period 1990-2012 when this model was not used.

RESULTS

In all, 65 of the 113 children underwent one to five additional endoscopies at dates set according to the model. The emergence of HRVs was recorded in 22 children after a mean interval of 14 months and was managed by endoscopic primary prophylaxis in all but one who underwent liver transplantation. Three other children bled before the next planned endoscopy. Compared with 175 children of the same age ranges without HRVs in the period 1990-2012, the use of the model was associated with a faster detection of HRVs with a lower number of endoscopic procedures (P  = 0.0022 and P  = 0.023, respectively).

CONCLUSION

The results suggest that the model reported may be a useful tool for the early detection of HRVs to allow primary prophylaxis of bleeding.

Factors associated with bleeding after endoscopic variceal ligation in children

BACKGROUND

Endoscopic variceal ligation (EVL) is a widely accepted treatment for esophagogastric varices in patients with portal hypertension (PHT). It is used for urgent treatment and prophylactic treatment of esophagogastric varices in pediatric as well as adult patients. However, major life-threatening adverse events such as early rebleeding can occur. Although early rebleeding after EVL among children and adolescents has been reported, the risk factors remain obscure. This study evaluated the risk factors for early rebleeding after EVL in children and adolescents.

METHODS

The subjects were children and adolescents (<18 years) with PHT who underwent EVL for esophagogastric varices. Early rebleeding was defined as hematemesis, active bleeding, or blood retention in the stomach, confirmed by esophagogastroduodenoscopy from 2 h to 5 days after EVL.

RESULTS

A total of 50 EVL sessions on 22 patients were eligible for this study. There were four episodes of early rebleeding. No other major adverse event has occurred. Multivariate analysis showed that EVL implemented at cardiac varices just below the esophagogastric junction (EGJ), within 5 mm from the EGJ, is the independent factor for a higher risk of early rebleeding: odds ratio 18.2 (95% confidence interval: 1.40-237.0), P = 0.02.

CONCLUSIONS

Children and adolescents who undergo EVL for cardiac varices just below the EGJ have a higher risk of early rebleeding than those who do not.

Band ligation versus sham or no intervention for primary prophylaxis of oesophageal variceal bleeding in children and adolescents with chronic liver disease or portal vein thrombosis

BACKGROUND

Portal hypertension commonly accompanies advanced liver disease and often gives rise to life-threatening complications, including bleeding (haemorrhage) from oesophageal and gastrointestinal varices. Variceal bleeding commonly occurs in children and adolescents with chronic liver disease or portal vein thrombosis. Prevention is, therefore, important. Randomised clinical trials have shown that non-selective beta-blockers and endoscopic variceal band ligation decrease the incidence of variceal bleeding in adults. In children and adolescents, band ligation, beta-blockers, and sclerotherapy have been proposed as primary prophylaxis alternatives for oesophageal variceal bleeding. However, it is unknown whether these interventions are of benefit or harm when used for primary prophylaxis in children and adolescents.

OBJECTIVES

To assess the benefits and harms of band ligation compared with sham or no intervention for primary prophylaxis of oesophageal variceal bleeding in children and adolescents with chronic liver disease or portal vein thrombosis.

SEARCH METHODS

We searched the Cochrane Hepato-Biliary Group Controlled Trials Register, CENTRAL, PubMed, Embase, and two other databases (April 2020). We scrutinised the reference lists of the retrieved publications, and we also handsearched abstract books of the two main paediatric gastroenterology and hepatology conferences from January 2008 to December 2019. We also searched clinicaltrials.gov, the United States Food and Drug Administration (FDA), the European Medicines Agency (EMA), and the World Health Organization (WHO) for ongoing clinical trials. We imposed no language or document type restrictions on our search.

SELECTION CRITERIA

We aimed to include randomised clinical trials irrespective of blinding, language, or publication status, to assess the benefits and harms of band ligation versus sham or no intervention for primary prophylaxis of oesophageal variceal bleeding in children with chronic liver disease or portal vein thrombosis. If the search for randomised clinical trials retrieved quasi-randomised and other observational studies, then we read them through to extract information on harm.

DATA COLLECTION AND ANALYSIS

We used standard Cochrane methodology to perform this systematic review. We used GRADE to assess the certainty of evidence for each outcome. Our primary outcomes were all-cause mortality, serious adverse events and liver-related morbidity, and quality of life. Our secondary outcomes were oesophageal variceal bleeding and adverse events not considered serious. We used the intention-to-treat principle. We analysed data using Review Manager 5.

MAIN RESULTS

One conference abstract, describing a feasibility multi-centre randomised clinical trial, fulfilled our review inclusion criteria. We judged the trial at overall high risk of bias. This trial was conducted in three hospital centres in the United Kingdom. The aim of the trial was to determine the feasibility and safety of further larger randomised clinical trials of prophylactic band ligation versus no active treatment in children with portal hypertension and large oesophageal varices. Twelve children received prophylactic band ligation and 10 children received no active treatment. There was no information on the age of the children included, or about the diagnosis of any child included. All children were followed up for at least six months. Mortality was 8% (1/12) in the band ligation group versus 0% (0/10) in the no active intervention group (risk ratio (RR) 2.54, 95% confidence interval (CI) 0.11 to 56.25; very low certainty of evidence). The abstract did not report when the death occurred, but we assume it happened between the six-month follow-up and one year. No child (0%) in the band ligation group developed adverse events (RR 0.28, 95% CI 0.01 to 6.25; very low certainty of evidence) but one child out of 10 (10%) in the no active intervention group developed idiopathic thrombocytopaenic purpura. One child out of 12 (8%) in the band ligation group underwent liver transplantation versus none in the no active intervention group (0%) (RR 2.54, 95% CI 0.11 to 56.25; very low certainty of evidence). The trial reported no other serious adverse events or liver-related morbidity. Quality of life was not reported. Oesophageal variceal bleeding occurred in 8% (1/12) of the children in the band ligation group versus 30% (3/10) of the children in the no active intervention group (RR 0.28, 95% CI 0.03 to 2.27; very low certainty of evidence). No adverse events considered non-serious were reported. Two children were lost to follow-up by one-year. Ten children in total completed the trial at two-year follow-up. There was no information on funding.  We found two observational studies on endoscopic variceal ligation when searching for randomised trials. One found no harm, and the other reported E nterobacter cloacae septicaemia in one child and mild, transient, upper oesophageal sphincter stenosis in another. We did not assess these studies for risk of bias. We did not find any ongoing randomised clinical trials of interest to our review.

AUTHORS' CONCLUSIONS

The evidence, obtained from only one feasibility randomised clinical trial at high risk of bias, is very scanty. It is very uncertain about whether prophylactic band ligation versus sham or no (active) intervention may affect mortality, serious adverse events and liver-related morbidity, or oesophageal variceal bleeding in children and adolescents with portal hypertension and large oesophageal varices. We have no data on quality of life. No adverse events considered non-serious were reported. The results presented in the trial need to be interpreted with caution. In addition, the highly limited data cover only part of our research question; namely, children with portal hypertension and large oesophageal varices. Data on children with portal vein thrombosis are lacking. Larger randomised clinical trials assessing the benefits and harms of band ligation compared with sham treatment for primary prophylaxis of oesophageal variceal bleeding in children and adolescents with chronic liver disease or portal vein thrombosis are needed. The trials should include important clinical outcomes such as death, quality of life, failure to control bleeding, and adverse events.