Sclerotherapy versus beta‐blockers for primary prophylaxis of oesophageal variceal bleeding in children and adolescents with chronic liver disease or portal vein thrombosis

Daniela Gattini; Lorena I Cifuentes; Romina Torres-Robles; Juan Cristóbal Gana

doi:10.1002/14651858.CD011659.pub2

Escleroterapia versus betabloqueantes para la profilaxis primaria del sangrado de várices esofágicas en niños y adolescentes con enfermedad hepática crónica o trombosis de la vena porta

Declaraciones de intereses de los autores

Versión publicada: 10 enero 2020 Historial de versiones

https://doi.org/10.1002/14651858.CD011659.pub2

Contraer todo Desplegar todo

Resumen

disponible en

Antecedentes

La hipertensión portal suele acompañar a la enfermedad hepática avanzada y a menudo da lugar a complicaciones potencialmente mortales, incluida la hemorragia (sangrado) de las várices esofágicas y gastrointestinales. La hemorragia de las várices ocurre comúnmente en niños con enfermedad hepática crónica u obstrucción de la vena porta. Por lo tanto, la prevención es importante. Tras los resultados de numerosos ensayos clínicos aleatorizados que demostraron la eficacia de los betabloqueantes no selectivos y la ligadura endoscópica de las várices para disminuir la incidencia de la hemorragia de las várices, la profilaxis primaria de este cuadro en adultos se ha convertido en el estándar de atención establecido. Sin embargo, la escleroterapia es la única opción profiláctica endoscópica actualmente disponible en los lactantes con menos de 10 kg de peso corporal debido al tamaño del ligador endoscópico.

Objetivos

Determinar los efectos beneficiosos y perjudiciales de la escleroterapia comparada con cualquier tipo de betabloqueante para la profilaxis primaria del sangrado de las várices esofágicas en niños y adolescentes con enfermedad hepática crónica o trombosis de la vena porta.

Métodos de búsqueda

Se hicieron búsquedas en el registro de ensayos controlados del Grupo Cochrane Hepatobiliar, CENTRAL, PubMed, Embase Elsevier, LILACS (Bireme) y el Science Citation Index Expanded (Web of Science) en febrero 2019. Se examinaron las listas de referencias de las publicaciones identificadas y se realizó una búsqueda manual en los principales libros de resúmenes de las conferencias de gastroenterología y hepatología pediátrica (NASPGHAN y ESPGHAN) de enero 2008 a diciembre 2018. Se realizaron búsquedas de ensayos clínicos en curso en ClinicalTrials.gov, FDA, EMA y la OMS. No hubo restricciones de idioma ni de tipo de documento.

Criterios de selección

Se programó incluir ensayos clínicos aleatorizados, independientemente del cegamiento, el idioma o el estado de publicación, que evaluaran la escleroterapia versus cualquier tipo de betabloqueante para la profilaxis primaria de la hemorragia de las várices esofágicas en niños y adolescentes con enfermedad hepática crónica o trombosis de la vena porta. Se programó incluir estudios cuasialeatorizados y otros estudios observacionales identificados con las búsquedas de ensayos clínicos aleatorizados para obtener el informe de los efectos perjudiciales.

Obtención y análisis de los datos

Se programó recopilar y resumir los datos de los ensayos clínicos aleatorizados como se describe en el protocolo, con el uso de las metodologías Cochrane estándar.

Resultados principales

No se encontraron ensayos clínicos aleatorizados que evaluaran la escleroterapia versus betabloqueantes para la profilaxis primaria del sangrado de las várices esofágicas en niños y adolescentes con enfermedad hepática crónica o trombosis de la vena porta.

Conclusiones de los autores

Faltan ensayos clínicos aleatorizados que evalúen los efectos beneficiosos o perjudiciales de la escleroterapia versus betabloqueantes para la profilaxis primaria del sangrado de várices esofágicas en niños y adolescentes con enfermedad hepática crónica o trombosis de la vena porta. Por lo tanto, se necesitan ensayos con una potencia estadística suficiente y un diseño apropiado, que evalúen los efectos beneficiosos y perjudiciales de la escleroterapia versus betabloqueantes en los resultados clínicos relevantes para el paciente como la mortalidad, el fracaso en el control de la hemorragia y los eventos adversos. Hasta que se realicen tales ensayos y se publiquen los resultados, no se pueden establecer conclusiones con respecto a los efectos beneficiosos o perjudiciales de las dos intervenciones.

PICO

Population

Intervention

Comparison

Outcome

El uso y la enseñanza del modelo PICO están muy extendidos en el ámbito de la atención sanitaria basada en la evidencia para formular preguntas y estrategias de búsqueda y para caracterizar estudios o metanálisis clínicos. PICO son las siglas en inglés de cuatro posibles componentes de una pregunta de investigación: paciente, población o problema; intervención; comparación; desenlace (outcome).

Para saber más sobre el uso del modelo PICO, puede consultar el Manual Cochrane.

Resumen en términos sencillos

disponible en

Escleroterapia versus betabloqueantes para la profilaxis primaria del sangrado de várices esofágicas en niños

Antecedentes

La hipertensión portal se define como un aumento de la presión arterial dentro de un sistema de venas (un tipo de vaso sanguíneo) llamado sistema venoso portal, que drena sangre del tracto gastrointestinal (intestino) y el bazo hacia el hígado. La hipertensión portal suele acompañar a la enfermedad hepática avanzada y a menudo da lugar a complicaciones potencialmente mortales, como hemorragia (sangrado) del esófago y várices gastrointestinales (venas dilatadas o inflamadas).

Después de numerosos ensayos clínicos aleatorizados (estudios en que los pacientes son asignados de forma aleatoria a uno de dos o más grupos de tratamiento) que demuestran la efectividad de los medicamentos llamados betabloqueantes no selectivos, y la ligadura endoscópica de las várices (en que se ata o se liga una vena dilatada con una banda elástica) para disminuir la incidencia de la hemorragia de las várices, el tratamiento para la prevención de la hemorragia de las várices en adultos (llamado profilaxis primaria) se ha convertido en el estándar de atención en adultos. Sin embargo, se desconoce si estas intervenciones son beneficiosas o perjudiciales cuando se utilizan en niños. Debido al tamaño del ligador endoscópico, la escleroterapia (la inyección endoscópica de irritantes tisulares que causan la obliteración de los vasos sanguíneos) es la única opción profiláctica endoscópica actualmente disponible en los lactantes con menos de 10 kg de peso corporal.

Objetivos

Se realizó una revisión sistemática de ensayos clínicos aleatorizados para evaluar los efectos beneficiosos y perjudiciales de la escleroterapia versus betabloqueantes para la prevención del sangrado de las várices esofágicas en niños con enfermedad hepática crónica o trombosis de la vena porta (bloqueo o estrechamiento de la vena porta [el vaso sanguíneo que lleva sangre al hígado desde los intestinos] debido a un coágulo de sangre). Se buscaron estudios hasta febrero 2019.

Resultados clave

No se encontraron ensayos clínicos aleatorizados para la inclusión en esta revisión. Por consiguiente, no se dispone de resultados de estudios de ensayos clínicos aleatorizados para concluir si la escleroterapia versus betabloqueante puede ser beneficiosa o no cuando se administra como profilaxis primaria de las várices esofágicas en niños y adolescentes con una enfermedad hepática o una trombosis de vena porta. Se necesitan ensayos bien diseñados que incluyan resultados clínicos importantes como la muerte, el fracaso en el control del sangrado y los efectos secundarios.

Authors' conclusions

Implications for practice

Randomised clinical trials assessing the benefits or harms of sclerotherapy versus beta‐blockers for primary prophylaxis of oesophageal variceal bleeding in children with chronic liver disease or portal vein thrombosis are lacking. Unless such trials are conducted and the results are published, we cannot make any conclusions regarding the benefits or harms of sclerotherapy and beta‐blockers.

Implications for research

This systematic review identified the need for well‐designed, adequately powered randomised clinical trials to assess the benefits and harms of sclerotherapy versus beta‐blockers for primary prophylaxis of oesophageal variceal bleeding in children with chronic liver disease or portal vein thrombosis. Those randomised clinical trials should include patient‐relevant clinical outcomes such as mortality, failure to control bleeding, adverse events, and quality of life. The trials should follow the CONSORT Statement (www.consort-statement.org/), Interventional Trials (SPIRIT) (Chan 2013), and Patient‐Centered Outcomes Research Institute recommendations (PCORI 2012).

Background

Description of the condition

There are scarce data on the prevalence and burden of liver disease in children. However, the natural history of portal hypertension in children is different than in adults. In adults, the cause of portal hypertension is mostly intrahepatic, whereas the cause of portal hypertension in children is mostly extrahepatic. The main underlying condition which results in the development of portal hypertension in children is extrahepatic portal vein obstruction (Di Giorgio 2019), followed by cirrhotic aetiologies, such as biliary atresia (Shneider 2016; Chapin 2018).

A hepatic venous pressure gradient (HVPG) of 10 mmHg or more, in adults with portal hypertension, has been associated with the formation of oesophageal varices. We know only a little about HVPG gradient values in children, as data are limited. One study in children with portal hypertension found that HVPG was not associated with the presence of varices or history of variceal bleeding, suggesting the possibility of intrahepatic shunting in children with advanced liver disease (Ebel 2019).

Variceal bleeding (haemorrhage) is common in children with portal hypertension that has developed because of chronic liver disease or portal vein obstruction (Lykavieris 2000; Miga 2001; van Heurn 2004; Lampela 2012; Duche 2013; Di Giorgio 2019). One study of 125 children with biliary atresia and signs of portal hypertension or history of gastrointestinal bleeding reported that 88 (70%) children developed oesophageal varices (Duche 2010). In children with biliary atresia, the observed incidence of variceal bleeding was from 17% to 29% over a 5‐ to 20‐year period (Miga 2001; van Heurn 2004; Duche 2013; Angelico 2019; Parolini 2019). One prospective study in 50 children with oesophageal varices, primarily due to cirrhosis, and who did not receive active treatment for prevention of variceal bleeding, showed that 42% of the children had upper gastrointestinal bleeding after a median follow‐up period of 4.5 years (Goncalves 2000). Available studies in children with portal hypertension due to portal vein thrombosis suggested that up to 50% of these children had a major variceal bleeding by 16 years of age (Lykavieris 2000). Variceal bleeding in children with portal hypertension has been associated with significant morbidity and variable mortality rates. Mortality rates in children with variceal bleeding of cirrhotic aetiologies were between 2% and 19% (Eroglu 2004; van Heurn 2004; Carneiro 2018). In contrast, variceal bleeding in children with portal vein thrombosis and no parenchymal liver disease seemed to carry less than 3% risk of mortality (Lykavieris 2000; Di Giorgio 2019).

Specific endoscopic variceal patterns are predictive of the risk of gastrointestinal bleeding. Grade 1 varices are small varices that extend just above the mucosal level; grade 2 varices project in less than one‐third of the luminal diameter and cannot be compressed with air insufflation; and grade 3 varices are large varices that occupy more than one‐third of the luminal diameter. In children with biliary atresia, grade 3 varices and grade 2 varices with oesophageal red markings, and the presence of gastric varices, have been reported to be independent risk factors for bleeding (Duche 2013). However, in children, there are limited data regarding endoscopic pattern of gastroesophageal varices predicting high risk of bleeding in children with portal vein thrombosis and other causes of portal hypertension (Shneider 2016; Chapin 2018).

Description of the intervention

Several publications have reported results on the use of sclerotherapy, band ligation, and beta‐blockers for primary prophylaxis of variceal bleeding in children (Goncalves 2000; Duche 2008; Lampela 2012; Duche 2013; El‐Karaksy 2015; Duche 2017; Galand 2018; Angelico 2019). Various surveys regarding primary prophylaxis for variceal bleeding in children, conducted by paediatric gastroenterologists, have shown different approaches to the management of children with portal hypertension (Groszmann 2004; Gana 2011a; Verdaguer 2016; Jeanniard‐Malet 2017). This suggests that many paediatric specialists apply the guidelines for management of portal hypertension in adults to children. However, there is an important variation of care provided by physicians, probably secondary to the lack of good‐quality studies. There are several differences in the pathophysiology of portal hypertension between adults and children. Therefore, precautions must be taken before extrapolating facts and data from adults to children or adolescents. Children are more dependent on chronotropy for maintenance of systemic blood pressure during hypovolaemia than adults who depend mainly on vasoconstriction. The main concern is that by limiting tachycardia in children, the use of a non‐selective beta‐blocker may impair tolerance to hypovolaemia and may lead to a more adverse outcome from variceal bleeding. The efficacy, pharmacokinetics, and adverse event profiles of the drugs may also differ significantly between children and adults. Finally, the principal aetiologies of liver disease and portal hypertension are different in these populations.

Although band ligation is frequently used in children, it is of limited use in infants. The currently available equipment for band ligation is too large to be introduced into the oesophagus of infants smaller than 10 kg of bodyweight, and sclerotherapy, due to the size of the endoscopic ligator, is the only endoscopic treatment currently available in infants weighing less than 10 kg of bodyweight.

How the intervention might work

The endoscopic sclerotherapy procedure involves the passage of an oesophagoscope and accomplishes vascular obliteration by injection of a sclerosant. Sclerosants are tissue irritants that cause vascular thrombosis and endothelial damage, leading to endofibrosis and vascular obliteration when injected into or adjacent to blood vessels. When a sclerosant is injected directly into the vein, blood clots are formed and the bleeding is stopped. When a sclerosant is injected into the area beside the distended vein, it stops the bleeding by thickening and swelling the tissue around the vein (Al‐Khazraji 2019).

The mechanisms of action of non‐selective beta‐blockers include reduction of cardiac output, reduction of portal venous flow, and antagonism of the noradrenaline (norepinephrine)‐induced constriction of intrahepatic myofibroblasts, activated stellate cells, and vascular smooth muscle cells. Different meta‐analyses in adults with cirrhosis have shown that non‐selective beta‐blockers reduce the incidence of variceal bleeding by 50% (Hayes 1990; D'Amico 1999). In children, we have no randomised clinical trials or systematic reviews of such trials, assessing the benefits and harms of beta‐blocker primary prevention (Bozic 2015). In spite of this, beta‐blockers are chosen by many physicians as primary prophylaxis (Verdaguer 2016; Jeanniard‐Malet 2017).

Why it is important to do this review

Primary prophylaxis with band ligation or beta‐blockers is the standard of care for the management of adults with chronic liver disease and portal hypertension with medium or large varices to prevent variceal bleeding (Garcia‐Tsao 2017). Variceal bleeding is common in children with oesophageal varices secondary to chronic liver disease or portal vein obstruction, and it has been associated with mortality. Therefore, prevention in children with portal hypertension is important (Gana 2010; Ling 2011; Gana 2011b; Shneider 2012).

Currently, there are no evidence‐based recommendations for the prophylactic management of children at risk of variceal bleeding. Various surveys conducted by paediatric gastroenterologists have found a wide range of practices regarding primary prophylaxis of variceal bleeding in children (Groszmann 2004; Gana 2011a; Shneider 2016; Verdaguer 2016; Jeanniard‐Malet 2017). In one survey of 30 paediatric gastroenterologists in the US, approaches to the management of children with portal hypertension, using screening endoscopy and primary prophylaxis of variceal bleeding, varied considerably among the respondents. In this survey, 63% of gastroenterologists performed surveillance of oesophageal varices and 84% offered primary prophylaxis of variceal bleeding in children (Groszmann 2004). In one survey in Canada, 70% of paediatric gastroenterologists reported that they would consider screening for oesophageal varices in children with liver disease and evidence of cirrhosis or portal hypertension (such as splenomegaly, thrombocytopenia, or portosystemic collaterals on sonography). However, only 58% of respondents who would screen for varices would provide primary prophylactic treatment (Gana 2011a). One survey of 35 Chilean paediatric gastroenterologists showed that 29 (83%) gastroenterologists had screened children for oesophageal varices because of clinical evidence of portal hypertension and 12 (34%) gastroenterologists had screened every child with chronic liver disease. Twenty‐eight (80%) respondents had used primary prophylaxis, mainly beta‐blockers, but also band ligation and sclerotherapy (Verdaguer 2016). One survey conducted in 38 French‐speaking hospitals showed that more than 75% of the centres had used endoscopy to screen children diagnosed with chronic liver diseases with suspected portal hypertension (Jeanniard‐Malet 2017). Among these 28 centres, 20 (71%) had performed primary prophylaxis for portal hypertension within their institution, 1 (4%) had done so only for children with cystic fibrosis because of its particular medical recruitment, and 7 (25%) had referred the children to a tertiary centre. In cases of grade 2 varices with red marks and grade 3 varices, more than 90% of the centres had performed sclerotherapy or endoscopic variceal ligation. Approximately 20% of centre had used beta‐blockers (Jeanniard‐Malet 2017). This suggests that there is an important variation of care provided by physicians, probably secondary to the lack of good‐quality studies.

Different treatments have been proposed for the primary prophylaxis of bleeding of oesophageal varices in children. This systematic review, comparing sclerotherapy versus beta‐blockers for primary prophylaxis of oesophageal variceal bleeding in children and adolescents with chronic liver disease or portal vein thrombosis (Gana 2015), is one of six reviews that were planned to examine the utility of these treatment modalities (Gana 2019; Gana 2020; Gattini 2020; Cifuentes 2021a; Cifuentes 2021b).

Objectives

To assess the benefits and harms of sclerotherapy versus any type of beta‐blocker for primary prophylaxis of oesophageal variceal bleeding in children and adolescents with chronic liver disease or portal vein thrombosis.

Methods

Criteria for considering studies for this review

Types of studies

Randomised clinical trials, regardless of publication status, language, or blinding. The trials should have compared sclerotherapy versus any type of beta‐blocker, administered as primary prophylaxis of oesophageal variceal bleeding in children with chronic liver disease or portal vein thrombosis. We planned to examine quasi‐randomised and observational studies if retrieved with the searches for randomised clinical trials. We aimed to report the occurrence of adverse events narratively.

Types of participants

Children or adolescents (up to 18 years old) with chronic liver disease or portal vein thrombosis, irrespective of the aetiology, severity of disease, and duration of illness, in whom the presence of oesophageal varices was confirmed by oesophagogastroduodenoscopy. The review planned to focus on children who had not had gastrointestinal bleeding from oesophageal varices (primary prophylaxis of variceal bleeding).

Children with a previous surgical portal‐systemic shunt procedure or insertion of a transjugular intrahepatic portal‐systemic shunt (TIPS), previous band ligation or sclerotherapy of oesophageal varices, or history of upper gastrointestinal bleeding are a distinct group of trial participants in whom the diagnosis or natural history of oesophageal varices has been modified. These children constituted an exclusion criteria of this review unless study data were subdivided following patient groups.

Types of interventions

Experimental

Sclerotherapy of oesophageal varices: any type of sclerotherapy, dosage, and duration of treatment.

Control

Beta‐blockers: any type of beta‐blocker, dosage, and duration of treatment.

Types of outcome measures

Primary outcomes

All‐cause mortality.
Upper gastrointestinal bleeding.
Serious adverse events and liver‐related morbidity (i.e. proportion of participants who developed ascites, hepatorenal syndrome, hepatocellular carcinoma, or hepatic encephalopathy). A serious adverse event, defined according to the International Conference on Harmonisation (ICH) Guidelines for Good Clinical Practice (ICH‐GCP 1997), was any untoward medical occurrence that resulted in death, was life‐threatening, required hospitalisation or prolongation of existing hospitalisation, resulted in persistent or significant disability or incapacity, or was a congenital anomaly or birth defect. All other adverse events were considered non‐serious adverse events.
Health‐related quality of life, determined exclusively by validated scales, classification, and measurement systems such as the Paediatric Quality of Life Inventory (PedsQL), Child Health Questionnaire (CHQ), and the Europena DISABKIDS tool (assesses health‐related quality of life in children and adolescents with a chronic condition).

Secondary outcomes

Oesophageal variceal bleeding.
Non‐serious adverse events (any adverse event that did not meet the above criteria for serious adverse events).

The follow‐up times for all outcomes were as defined in each trial; and up to five years' follow‐up after treatment (the primary time point for collecting data for analysis).

Search methods for identification of studies

Electronic searches

We searched the Cochrane Hepato‐Biliary Group (CHBG) Controlled Trials Register (maintained and searched internally by the CHBG Information Specialist via the Cochrane Register of Studies Web; February 2019), Cochrane Central Register of Controlled Trials (CENTRAL) in the Cochrane Library (2019, Issue 2), PubMed (1809 to February 2019), Embase (Elsevier; 1947 to February 2019), LILACS (Bireme; 1982 to February 2019), and Science Citation Index Expanded (Web of Science; 1975 to February 2019) (Royle 2003). We scrutinised the reference lists of the retrieved publications. We also searched the trial registries ClinicalTrial.gov (clinicaltrials.gov/), European Medicines Agency (EMA) (www.ema.europa.eu/ema/), World Health Organization International Clinical Trial Registry Platform (www.who.int/ictrp), and the Food and Drug Administration (FDA) (www.fda.gov), for ongoing trials. We applied no language or document type restrictions. Search strategies with the time spans of the searches are listed in Appendix 1.

Searching other resources

We tried to identify additional references by manually searching the reference lists of articles from the computerised databases and relevant review articles. Furthermore, we performed a manual search from the main paediatric gastroenterology and hepatology conferences such as North American Society for Pediatric Gastroenterology, Hepatology and Nutrition (NASPGHAN), and European Society for Paediatric Gastroenterology, Hepatology and Nutrition (ESPGHAN) abstract books from 2008 to 2018. We applied no language or document type restrictions.

Data collection and analysis

We followed the guidelines provided in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). We planned to use Review Manager 5 for our analyses (Review Manager 2014).

Selection of studies

We planned to retrieve publications if they were potentially eligible for inclusion based on an abstract review, or if they were relevant, review articles for a manual reference search. Two review authors (LC and DG) independently assessed the publications for eligibility using the inclusion criteria. Abstracts were only to be included if sufficient data were provided for analysis. Any disagreements were to be resolved by consensus between three review authors (LC, DG, and JCG).

Data extraction and management

Two review authors (LC and DG) planned to independently complete a data extraction form on all the included studies and retrieve the following data.

General information: title, journal, year, publication status, and trial design.
Sample size: number of participants meeting the criteria and total number screened.
Baseline characteristics: baseline diagnosis, age, sex, race, disease severity, and concurrent medications used. Severity of liver disease of the studied population may have been considered using the Child‐Pugh score (Pugh 1973), the paediatric end‐stage liver disease (PELD) scores for children aged less than 12 years (McDiarmid 2002), and model for end‐stage liver disease (MELD) for children aged 12 years and older (Kamath 2001).
All‐cause mortality, upper gastrointestinal bleeding, oesophageal variceal bleeding, and quality of life determined exclusively by validated scales.
Adverse events: serious and non‐serious.

One review author (JCG) was to arbitrate in case of disagreements in data extraction.

Assessment of risk of bias in included studies

Following recommendations in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011) and methodological studies (Schulz 1995; Moher 1998; Kjaergard 2001; Wood 2008; Savović 2012a; Savović 2012b; Savović 2018), two review authors (LC and DG) independently planned to assess the risk of bias of each included trial using the following definitions.

Allocation sequence generation

Low risk of bias: sequence generation was achieved using computer random number generation or a random number table. Drawing lots, tossing a coin, shuffling cards, and throwing dice were adequate if performed by an independent person not otherwise involved in the trial.
Unclear risk of bias: the method of sequence generation was not specified.
High risk of bias: the sequence generation method was not random. Such studies were to be excluded from the assessment of benefit.

Allocation concealment

Low risk of bias: the participant allocations could not have been foreseen in advance of, or during, enrolment. Allocation was controlled by a central and independent randomisation unit. The allocation sequence was unknown to the investigators (e.g. if the allocation sequence was hidden in sequentially numbered, opaque, and sealed envelopes).
Unclear risk of bias: the method used to conceal the allocation was not described so that intervention allocations may have been foreseen in advance of, or during, enrolment.
High risk of bias: the allocation sequence was likely to be known to the investigators who assigned the participants. Such studies were to be excluded from the assessment of benefit.

Blinding of participants and personnel

Low risk of bias: blinding of participants and personnel performed adequately using a placebo. We defined lack of blinding as unlikely to affect the evaluation of mortality (Savović 2012a; Savović 2012b).
Unclear risk of bias: insufficient information to assess blinding.
High risk of bias: no blinding or incomplete blinding.

Blinding of outcome assessors

Low risk of bias: blinding of outcome assessors performed adequately using a placebo. We defined lack of blinding as unlikely to affect the evaluation of mortality (Savović 2012a; Savović 2012b).
Unclear risk of bias: insufficient information to assess blinding.
High risk of bias: no blinding or incomplete blinding.

Incomplete outcome data

Low risk of bias: missing data were unlikely to make treatment effects depart from plausible values. Sufficient methods, such as multiple imputation, were employed to handle missing data.
Unclear risk of bias: there was insufficient information to assess whether missing data in combination with the method used to handle missing data were likely to induce bias on the results.
High risk of bias: the results were likely to be biased due to missing data.

Selective outcome reporting

Low risk of bias: the trial reported the following predefined primary outcomes: all‐cause mortality, gastrointestinal bleeding, and serious adverse events. If the original trial protocol was available, the outcomes should have been those called for in that protocol. If the trial protocol was obtained from a trial registry (e.g. www.clinicaltrials.gov), the outcomes sought were those enumerated in the original protocol if the trial protocol was registered before or at the time that the trial was begun. If the trial protocol was registered after the trial was begun, those outcomes were not considered to be reliable.
Unclear risk of bias: not all predefined outcomes were reported fully, or it was unclear whether data on these outcomes were recorded or not.
High risk of bias: one or more predefined outcomes were not reported.

Overall bias risk assessment

Low risk of bias: all domains in a trial are at low risk of bias.
High risk of bias: one or more of the bias domains in a trial are at unclear or high risk of bias.

We expected a lack of blinding of participants in the trials considering the different treatment modalities (endoscopic versus oral), and this could lead to a bias that might have necessitated analysis.

We planned to generate a 'Risk of bias' graph and 'Risk of bias' summary to show a summary of this assessment.

Measures of treatment effect

For dichotomous outcomes, we planned to calculate the risk ratio (RR) with 95% confidence intervals (CI). For continuous outcomes such as health‐related quality of life, we planned to calculate the mean difference (MD) with 95% CI if all studies reported it using the same scale, and standardised mean difference (SMD) with 95% CI if the studies used different scales for its reporting.

Unit of analysis issues

The unit of analysis was to be the participant undergoing treatment (i.e. primary prophylaxis of oesophageal variceal bleeding in children with chronic liver disease or portal vein thrombosis) according to the intervention group to which the participant was randomly assigned. In the case of cross‐over trials, we planned to use the outcome data after the period of first intervention because the assigned treatments could have residual effects (Higgins 2011). Due to the clinical situation, we did not expect to find cluster‐randomised trials. In case of trials with multiple intervention groups, we planned to collect data for all trial intervention groups that met our inclusion criteria. We planned to divide the control group into two to avoid double‐counting in case this was a common comparator.

Dealing with missing data

We planned to perform an intention‐to‐treat analysis whenever possible; otherwise, we planned to use the data available to us and contact the original investigators to request the missing data.

Regarding the dichotomous primary outcomes, whenever possible, we planned to conduct the following two sensitivity analyses.

Extreme case analysis favouring the experimental intervention ('best‐worse' case scenario): none of the dropouts or participants lost from the experimental arm, but all of the dropouts and participants lost from the control group experienced the outcome; including all randomised participants in the denominator.
Extreme case analysis favouring the control ('worst‐best' case scenario): all dropouts or participants lost from the experimental arm, but none from the control arm experienced the outcome; including all randomised participants in the denominator.

For the continuous primary outcome, health‐related quality of life, we planned to impute the standard deviation from P values, according to guidance in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). If the data were likely to be normally distributed, we planned to use the median for meta‐analysis when the mean was not available; otherwise, we planned to provide a median and interquartile range of the difference in medians. If it was not possible to calculate the standard deviation from the P value or the CIs, we planned to impute the standard deviation using the largest standard deviation in other trials for that outcome. This form of imputation can decrease the weight of the study for calculation of MDs and may bias the effect estimate to no effect for calculation of SMDs (Higgins 2011).

Assessment of heterogeneity

We planned to identify heterogeneity by visual inspection of the forest plots, by using a standard Chi² test and a significance level of α = 0.1, in view of the low power of such tests. We planned to use the Chi² test for heterogeneity to detect between‐trial heterogeneity. In addition, we planned to specifically examine the degree of heterogeneity observed in the results with the I² statistic according to the following classification: from 0% to 40%, heterogeneity may not be important; from 30% to 60%, heterogeneity may be moderate; from 50% to 90%, heterogeneity may be substantial; and from 75% to 100%, heterogeneity may be considerable (Higgins 2003).

If there was heterogeneity, we planned to determine the potential reasons for it by examining the individual trial and subgroup characteristics.

Assessment of reporting biases

We planned to assess reporting biases with funnel plots of the relative risk estimates from the individual trials (plotted on a logarithmic scale) against trial size or precision (variance) or the estimators.

Data synthesis

Meta‐analyses

We planned to conduct this systematic review according to the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). We planned to use the statistical software Review Manager 5 provided by Cochrane to analyse data and produce summary estimates of the treatment effect (Review Manager 2014). We planned to present results with a random‐effects meta‐analysis because we expected that the included trials would be heterogeneous. We planned to present results of continuous outcomes as MD or SMD, with 95% CI (Higgins 2011).

Subgroup analysis and investigation of heterogeneity

We planned to perform the following subgroup analyses because we expected that we would observe heterogeneity in:

trials at low risk of bias compared to trials at high risk of bias, because trials at high risk of bias may overestimate or underestimate a treatment effect (Higgins 2011);
trials at risk of vested interest bias compared to trials conducted without risk of vested interest bias because trials at high risk of vested interest bias may overestimate or underestimate a treatment effect (Lundh 2017);
primary prophylaxis of small varices compared to primary prophylaxis of only medium or large varices because of the different risk of bleeding according to the variceal size (Garcia‐Tsao 2017);
children with chronic liver disease compared to children with extrahepatic portal vein obstruction, because the differences in the physiopathology of the cause of the portal hypertension between children with liver disease versus alteration in the portal inflow (Chapin 2018);
severity of liver disease (Child‐Pugh A, B, or C, and PELD or MELD), because of the different risk of bleeding according to the impaired liver function (Garcia‐Tsao 2017);
children with cholestatic compared to children with non‐cholestatic liver disease, because of the different risk of bleeding according to different aetiologies (Chapin 2018).

Sensitivity analysis

In addition to the sensitivity analyses specified under Dealing with missing data, in order to assess the robustness of the eligibility criteria, our intention was to undertake sensitivity analyses in an attempt to explain our findings as well as any observed heterogeneity.

We planned to conduct Trial Sequential Analysis to assess imprecision in our primary and secondary outcomes (Thorlund 2017; TSA 2017), and compare the result of our assessment with the assessment of imprecision using GRADEpro GDT.

Trial Sequential Analysis

We planned to apply Trial Sequential Analysis (Thorlund 2017; TSA 2017) because cumulative meta‐analyses are at risk of producing random errors due to sparse data and repetitive testing of the accumulating data (Wetterslev 2008). To minimise random errors, we planned to calculate the required information size (i.e. the number of children needed in a meta‐analysis to detect or reject a certain intervention effect) (Wetterslev 2008).

The required information size calculation should also account for the heterogeneity or diversity present in the meta‐analysis (Wetterslev 2008; Wetterslev 2009). In our meta‐analyses, we planned to base the required information size on the event proportion in the control group; assumption of a plausible relative risk reduction of 20% on the relative risk reduction observed in the included trials with low risk of bias; a risk of type I error of 2.0% because of four primary outcomes and 3.30% because of two secondary outcomes, a risk of type II error of 10%, and the assumed diversity of the meta‐analysis (Wetterslev 2009).

The underlying assumption of Trial Sequential Analysis is that testing for significance may be performed each time a new trial is added to the meta‐analysis. We planned to add the trials according to the year of publication, and, if more than one trial was published in a year, we planned to add trials alphabetically according to the last name of the first author. Based on the required information size, we planned to construct trial sequential monitoring boundaries (Wetterslev 2008; Thorlund 2017). These boundaries determine the statistical inference one may draw regarding the cumulative meta‐analysis that has not reached the required information size; if the trial sequential monitoring boundary for benefit or harm is crossed before the required information size is reached, firm evidence may be established and further trials may be superfluous.

However, if the boundary is not surpassed, it is most probably necessary to continue conducting trials in order to detect or reject a certain intervention effect. This can be determined by assessing if the cumulative Z‐curve crosses the trial sequential monitoring boundaries for futility.

'Summary of findings' tables

We planned to create a 'Summary of findings' table to present information about certainty of evidence, magnitude of effects of the interventions, and summarise data on outcomes using GRADEpro GDT on all‐cause mortality, upper gastrointestinal bleeding, health‐related quality of life, serious adverse events, oesophageal bleeding, and non‐serious adverse events. We also planned to provide the time and range of follow‐up. The GRADE approach appraises the certainty of a body of evidence based on the extent to which one can be confident that an estimate of effect or association reflects the item being assessed. The certainty of a body of evidence considers within‐study risk of bias, indirectness of the evidence (population, intervention, control, outcomes), unexplained inconsistency (heterogeneity) of results (including problems with subgroup analyses), imprecision of results, and risk of publication bias.

We defined the levels of evidence as 'high', 'moderate', 'low', or 'very low' as follows.

High certainty: we are very confident that the true effect lies close to that of the estimate of the effect.
Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different.
Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect.
Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.

Results

Description of studies

We found no randomised clinical trials that qualified for inclusion in the review.

Results of the search

We identified 1296 records in the initial electronic search that assessed interventions for primary prophylaxis of oesophageal variceal bleeding in children. For this current systematic review that evaluated the benefits and harms of sclerotherapy compared to any type of beta‐blocker for primary prophylaxis of oesophageal variceal bleeding in children with chronic liver disease or portal vein thrombosis, we identified 45 records for abstract revision. Of these studies, we excluded 44 because they did not meet the requirements to answer our review question (Figure 1). We assessed one full‐text publication for eligibility, but it did not meet the inclusion criteria of our review, since all included participants were adults (see Characteristics of excluded studies table). We found no ongoing studies related to our review question.

Figure 1

Study flow diagram. Date of search: 28 February 2019.

We planned to consider quasi‐randomised and other observational studies that might have been retrieved with the searches for randomised clinical trials for report of harm. We found no such studies.

Included studies

We found no studies that matched our inclusion criteria.

Excluded studies

We retrieved and assessed one full‐text report for eligibility. This study was excluded since all included participants were adults (Figure 1; Characteristics of excluded studies table).