Scolaris Content Display Scolaris Content Display

Cochrane Database of Systematic Reviews Protocol - Intervention

Pharmacological interventions for prevention and treatment of upper gastrointestinal bleeding in newborn infants

Authors' declarations of interest

Version published: 02 July 2015 Version history

https://doi.org/10.1002/14651858.CD011785

This is not the most recent version

view the current version 02 July 2019
Collapse all Expand all

Abstract

This is a protocol for a Cochrane Review (Intervention). The objectives are as follows:

To assess whether different pharmacological interventions (PPIs, H2RAs, antacids, sucralfate or bismuth salts) administered to preterm and term neonates for the prevention or treatment of upper gastrointestinal bleeding reduce morbidity and mortality compared with each other, placebo, no treatment or supportive care.

Background

Description of the condition

Upper gastrointestinal bleeding (arising proximal to the ligament of Treitz in the distal duodenum) is typically a mild, self‐limiting condition that can affect both preterm and term neonates. Common symptoms include vomiting of blood‐stained or coffee ground‐like material (hematemesis) and black, tarry stools (melena) (Green 2003). Upper gastrointestinal bleeding can be diagnosed clinically by the presence of blood‐stained aspirates through indwelling nasogastric or orogastric tubes, hematemesis or endoscopically by examining the gastric mucosa for bleeding lesions (Green 2003). Neonates who have swallowed maternal blood should not be confused with neonates who have gastric bleeding.

Epidemiological data on upper gastrointestinal bleeding in neonates is limited, but it is generally considered a rare condition (Lazzaroni 2002). Lazzaroni 2002 has described an association between upper gastrointestinal tract mucosal lesions and upper gastrointestinal bleeding. The prevalence of gastric lesions is high in neonates admitted to intensive care (Kuusela 2000), with 20% of infants admitted and 53% of mechanically ventilated infants developing signs of gastrointestinal bleeding. Despite this, Kuusela 2000 found that many neonates with such lesions do not exhibit symptoms of upper gastrointestinal bleeding.

Although the symptoms and signs of upper gastrointestinal bleeding in neonates are similar to those in adults, the aetiology of this disease differs slightly between the two groups (Chawla 2007). Common causes of upper gastrointestinal bleeding in neonates include stress‐induced gastritis or stress‐induced ulcers, trauma (e.g. from intubation or other upper airway trauma, nasogastric tubes and nasogastric tube lavage), vitamin K deficiency, coagulopathies, milk protein intolerance and drugs (e.g. non‐steroidal anti‐inflammatory drugs (NSAIDs) and corticosteroids) (Boyle 2008; Chawla 2007). The risk associated with mechanical ventilation in neonates has been particularly emphasised by Kuusela 2000. It is rare for genetic disorders (e.g. Zellweger syndrome) and structural abnormalities of blood vessels (e.g. hereditary haemorrhagic telangiectasia, hemangiomas and Ehlers‐Danlos syndrome) to cause upper gastrointestinal bleeding in neonates. Haematemesis may occur in newborn infants who have swallowed maternal blood, and so these infants should be distinguished from infants with true cases of upper gastrointestinal bleeding (Apt 1955).

Description of the intervention

Acid suppression agents are used in the management of upper gastrointestinal bleeding in the adult population (Mejia 2009). In neonates, however, evidence‐based recommendations supporting similar benefits are not readily available. The main principle underlying this form of pharmacological intervention is to protect the gastric mucosa from damage and promote healing. In adults with peptic ulcers, gastric acidity has been associated with ongoing mucosal tissue damage and reduced clot formation (Kolkman 1996). People with gastric ulcers with a pH < 6 are predisposed to increased fibrinolysis of overlying clots (Green 1978).

Drugs commonly used to treat conditions associated with upper gastrointestinal bleeding include:

  1. Proton pump inhibitors (PPIs) ‐ acid suppressors (e.g. omeprazole);

  2. H2 receptor antagonists (H2RAs) ‐ acid suppressors (e.g. ranitidine, cimetidine and famotidine);

  3. Antacids ‐ acid neutralisers (e.g. sodium/calcium carbonate, magnesium/aluminium hydroxide and almagate);

  4. Mucosal protective agents (e.g. sucralfate and bismuth salts).

PPIs, such as omeprazole, are prodrugs that require protonation to exert their inhibitory effect on H+‐K+‐ATPase. This results in decreased gastric acid production (Sachs 2007). PPIs further protect the mucosal lining of the stomach by reducing pepsin secretion (Brunner 1995). H2RAs, such as ranitidine, also act to suppress gastric acid production. They achieve this through binding to H2 receptors on the basolateral side of parietal cells, which results in inhibition of the effects of histamine. H2RAs have a faster onset of action but shorter duration of action compared to PPIs (Australian Medicines Handbook 2013).

Significant adverse effects of PPIs and H2RAs are infrequent or rare and include hypotension, rashes, thrombocytopenia and insomnia (Australian Medicines Handbook 2013). H2RAs have also been shown to increase the risk of necrotizing enterocolitis (NEC) in very low birth weight (VLBW) infants (Terrin 2012). In mechanically ventilated patients, PPIs have been shown to increase the risk of gastrointestinal haemorrhage, pneumonia and Clostridium difficile infection compared to patients administered H2RAs (MacLaren 2014).

Unlike PPIs and H2RAs, antacids exert their effects within the gut lumen. They both neutralise gastric acid and inhibit pepsin activity (Maton 1999). Sodium bicarbonate, calcium bicarbonate, magnesium hydroxide, aluminium hydroxide and almagate are examples of antacids (Mejia 2009). Possible adverse effects of antacids include hypophosphataemia, hypermagnesaemia, intestinal obstruction and osteomalacia (Australian Medicines Handbook 2013).

Whilst PPIs, H2RAs and antacids predominantly influence acid secretion, sucralfate and bismuth treatment aims to protect the gastric mucosa (Australian Medicines Handbook 2013). Both of these drugs are excreted in faeces (Mejia 2009). Sucralfate may cause adverse constipation and nausea, whilst bismuth more commonly causes blackening of faeces and darkening of the teeth and tongue (Australian Medicines Handbook 2013).

How the intervention might work

In adult patients, PPIs are indicated for both the prevention and treatment of upper gastrointestinal bleeding (Australian Medicines Handbook 2013). Treatment with PPIs has been shown to decrease the rate of re‐bleeding compared to H2RAs or placebo (Leontiadis 2009). This effect has been demonstrated to occur independent of dose, route of administration or geographic location. In adults with upper gastrointestinal bleeding, high‐dose omeprazole infusion has been shown to reduce signs of upper gastrointestinal bleeding and the need for endoscopy therapy (Lau 2007).

Since gastric acid production is not solely regulated by histamine, H2RAs are not as effective as PPIs at reducing acid production. Consequently, H2RA treatment is no longer recommended for treatment in adult patients with acute ulcer bleeding. However, H2RAs may prove to be useful in treating upper gastrointestinal bleeding in neonates (Barkun 2010). Although PPIs have superior acid suppression ability, the shorter onset of action of H2RAs may provide faster relief of upper gastrointestinal bleeding symptoms (Mejia 2009).

In adults, antacid use has largely been replaced by PPIs and H2RAs. Despite this, however, acid neutralisation may have a greater role in the prophylaxis or treatment of upper gastrointestinal bleeding in neonates.

In response to the acidic environment of the gut, sucralfate promotes mucosal healing via angiogenesis, growth factor delivery and granulation tissue formation at ulcer sites (Tarnawski 1995). Healing mechanisms of bismuth include stimulating prostaglandin and bicarbonate production in addition to inhibiting the growth of Helicobacter pylori in the gut mucosa (Mejia 2009). These mechanisms of mucosal protection may have greater importance in the prophylaxis of upper gastrointestinal bleeding in neonates.

Why it is important to do this review

Current recommendations for the prevention and treatment of upper gastrointestinal bleeding in neonates are based on those used for adults largely due to a lack of studies in neonates. International consensus guidelines recommend intravenous and oral PPIs, but not H2RAs, for the prevention and treatment of acute upper gastrointestinal bleeding in adults (Barkun 2010). However, a recent study by Sreedharan 2010 demonstrated the usefulness of treatment with PPIs in reducing stigmata of recent haemorrhage but not in reducing mortality and rebleeding rates. A previous Cochrane review described benefits of PPI treatment for acute upper gastrointestinal bleeding in adults, but not in neonates (Leontiadis 2006). Uncertainty in the application of these guidelines to neonates is further increased by the diverse profile of adverse effects of these drugs seen in adults, as previously described. The lack of consensus on pharmacological prophylaxis and treatment of upper gastrointestinal bleeding in neonates forms the basis for this Cochrane review.

Objectives

To assess whether different pharmacological interventions (PPIs, H2RAs, antacids, sucralfate or bismuth salts) administered to preterm and term neonates for the prevention or treatment of upper gastrointestinal bleeding reduce morbidity and mortality compared with each other, placebo, no treatment or supportive care.

Methods

Criteria for considering studies for this review

Types of studies

We will include randomised controlled trials (RCTs), quasi‐RCTs and cluster‐RCTs.

Types of participants

In the context of prevention, we will consider preterm and term neonates at risk of upper gastrointestinal bleeding. We will define neonates at risk of upper gastrointestinal bleeding as those who have been exposed to known upper gastrointestinal bleeding risk factors including preterm birth, birth asphyxiation, mechanical ventilation, neonate with hypotension, neonatal cyanosis and neonatal seizures (Kuusela 2000; Ombeva 2013).

In the context of treatment, we will consider preterm and term neonates with upper gastrointestinal bleeding. Upper gastrointestinal bleeding will be identified by the presence of blood‐stained or coffee ground‐like material in gastric aspirates or vomits (haematemesis), microscopically by the presence of blood in gastric aspirates or vomits or endoscopically by examining the gastric mucosa for bleeding lesions (Green 2003). We will exclude neonates who have swallowed maternal blood.

Types of interventions

Pharmacological intervention (H2RAs, PPIs, antacids, sucralfate, bismuth salts, or in any combination) administered enterally or parenterally at any dose or frequency with or without co‐administration with other treatment/prophylaxis modalities (e.g. ceasing oral feeds; ceasing NSAIDs) with the intention of preventing or treating upper gastrointestinal bleeding, compared with each other, placebo, no intervention or supportive therapy. We will exclude treatment trials for upper gastrointestinal bleeding aimed at correcting underlying bleeding disorders, such as Vitamin K administration for vitamin K deficiency or fresh frozen plasma transfusion, as these have already been described in previous Cochrane reviews (Ardell 2010; Puckett 2000).

We plan to analyse the following comparisons:

a) Comparisons between drug classes

  1. PPIs vs. H2RAs;

  2. PPIs vs. antacids;

  3. PPIs vs. bismuth;

  4. PPIs vs. sucralfate;

  5. PPIs vs. placebo, no treatment or supportive care;

  6. H2RAs vs. antacids;

  7. H2RAs vs. bismuth;

  8. H2RAs vs. sucralfate;

  9. H2RAs vs. placebo, no treatment or supportive care;

  10. Antacids vs. bismuth;

  11. Antacids vs. sucralfate;

  12. Antacids vs. placebo, no treatment or supportive care;

  13. Bismuth vs. sucralfate;

  14. Bismuth vs. placebo, no treatment or supportive care;

  15. Sucralfate vs. placebo, no treatment or supportive care.

b) Comparisons between specific drugs of different classes

  1. Specific H2RAs vs. specific PPIs;

  2. Specific antacids vs. specific PPIs;

  3. Bismuth vs. specific PPIs;

  4. Sucralfate vs. specific PPIs;

  5. Specific antacid vs. specific H2RAs;

  6. Bismuth vs. specific H2RAs;

  7. Sucralfate vs. specific H2RAs;

  8. Specific antacids vs. bismuth;

  9. Specific antacids vs. sucralfate;

  10. Bismuth vs. sucralfate.

c) Comparison between specific drugs within a drug class

  1. Specific H2RAs vs. specific H2RAs;

  2. Specific PPIs vs. specific PPIs;

  3. Specific antacids vs. specific antacids.

d) Comparisons of combinations of specific drugs

  1. H2RAs plus one or more of the following: antacids, sucralfate or bismuth salts vs. PPIs;

  2. H2RAs plus one or more of the following: antacids, sucralfate or bismuth salts vs. placebo, no treatment or supportive care;

  3. PPIs plus one or more of the following; antacids, sucralfate or bismuth salts. vs. H2RAs;

  4. PPIs plus one or more of the following; antacids, sucralfate or bismuth salts. vs. placebo, no treatment or supportive care.

Types of outcome measures

Primary outcomes

  1. All cause mortality to near term age or discharge;

  2. Duration of upper gastrointestinal bleeding in infants with upper gastrointestinal bleeding (days);

  3. Any upper gastrointestinal bleeding in infants at risk of upper gastrointestinal bleeding.

Secondary outcomes

  1. Gastric lesions detected by endoscopy and macroscopically;

  2. All cause infant mortality (i.e. < one year of age);

  3. Anaemia requiring blood transfusion;

  4. Volume of blood transfused for treatment of upper gastrointestinal bleeding (mL/kg);

  5. Number of blood transfusions for treatment of upper gastrointestinal bleeding;

  6. Necrotising enterocolitis (NEC) as defined by Bell stages 1‐4 (confirmed = Bell stage 2 or greater) (Bell 1978);

  7. Time to full feeds (days);

  8. Duration of total parenteral nutrition (days);

  9. Proven infection (culture positive from a normally sterile site);

  10. Neonatal cholestasis ('serum conjugated bilirubin concentration > 17.1 µM and total serum bilirubin < 85.5 µM' OR 'serum conjugated bilirubin concentration > 20% of total serum bilirubin if total serum bilirubin is > 85.5 µM')

  11. Ventilator‐associated pneumonia (new or progressive infiltrate with positive respiratory specimens after 48 hours of mechanical ventilation);

  12. Neonatal chronic lung disease (also known as bronchopulmonary dysplasia) (defined according to the 2001 National Institute of Child Health and Development criteria (Jobe 2001): treatment with oxygen > 21% for at least 28 days);

  13. Duration of ventilation (days);

  14. Duration of respiratory support (days);

  15. Duration of hospital stay (days);

  16. Neurodevelopmental disability (defined as neurological abnormality including cerebral palsy on clinical examination or global developmental delay (two or more standard deviations (SDs) below population mean on Bayley Scales of Infant Development or Griffiths Mental Development Scales at any time after term corrected at one year, 18 months, two years and five years postnatal age);

  17. Haemorrhagic shock (at least 15% blood loss);

  18. Thrombocytopenia (platelet count < 150,000/µL);

  19. Retinopathy of prematurity or other severe adverse events.

Search methods for identification of studies

We will use the standard search strategy of the Cochrane Neonatal Review Group as outlined in the Cochrane Library. Unpublished studies will be eligible for review. We will not apply any language restrictions.

Electronic searches

We will search the Cochrane Neonatal Group Specialized Register.

We will search MEDLINE and PREMEDLINE (1946 to 2015) via OVID interface using the strategy listed in Appendix 1.

We will adapt this strategy to search the following electronic databases:

  1. Cochrane Central Register of Controlled Trials (CENTRAL, the Cochrane Library, 2015);

  2. EMBASE (1980 to current) via Scopus interface;

  3. CINAHL (1982 to current) via EBSCO interface.

We have also adapted the search strategy for Google Scholar. The following search strategy will be used for Google Scholar because of the search limit of 32 words: random AND "upper gastrointestinal bleed" OR "upper gastrointestinal hemorrhage" AND neonate OR infant AND "histamine h2 antagonist" OR "histamine h2 receptor antagonist" OR "proton pump inhibitor" OR bismuth OR sucralfate OR antacid.

Searching other resources

We will conduct additional searches of the following resources:

1. Ongoing trials in the following trial registries:

2. Conference abstracts from:

  • Proceedings of the Pediatric Academic Societies (American Pediatric Society, Society for Pediatric Research and European Society for Pediatric Research) from 1990 to current from the journal Pediatric Research and Abstracts Online;

  • Proceedings of the European Academy of Paediatric Societies (EAPS) (The European Society for Paediatric Research (ESPR), the European Academy of Paediatrics (EAP) and the European Society of Paediatric and Neonatal Intensive Care (ESPNIC) from 2003 to current from Abstracts Online;

  • Proceedings of the Perinatal Society of Australia and New Zealand (PSANZ) from 1996 to current (handsearch).

3. Reference lists: After reading the identified individual studies, we will screen the reference lists of these papers to identify further relevant studies about upper gastrointestinal bleeding.

4. Personal communications:

  • If any unpublished trials are identified, we will contact the corresponding investigator for information on any unpublished trials potentially eligible for inclusion. Unpublished studies will be eligible for review;

  • We will contact the corresponding authors of some identified RCTs for additional information about their studies when data provided in the studies are deemed insufficient.

5. Pharmaceutical companies: We will contact companies that develop pharmacological interventions (H2RAs, PPIs, antacids, sucralfate or bismuth salts) for preventing or treating upper gastrointestinal bleeding for possible unpublished studies using their product in neonates.

Data collection and analysis

We will use the standard methods of the Cochrane Neonatal Review Group and Cochrane, as documented in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011).

Selection of studies

At least two review authors will independently assess studies identified for inclusion. We will resolve any differences through discussion and consensus. If necessary, we will seek assistance from a Cochrane editor.

Data extraction and management

At least two review authors will independently extract data using specifically designed data extraction forms. We will resolve discrepancies through discussion and consensus. We will use the retrieved information to determine trial eligibility, to extract methods and data from eligible trials and for requesting additional unpublished information from authors of original reports. Using RevMan 2015 we will enter and cross‐check data.

Assessment of risk of bias in included studies

At least two review authors will independently assess study quality and risk of bias using the following criteria documented in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011).

Sequence generation (to assess selection bias)

For each included study we will analyse the method used to generate the allocation sequence to determine whether it should produce comparable groups. We will assess the method as:

  • Low risk (random component in the sequence generation process, e.g. random number table; coin tossing; throwing dice);

  • High risk (non‐random component in the sequence generation process, e.g. date of birth; date of admission; hospital record number);

  • Unclear risk.

Allocation concealment (to assess selection bias)

For each included study we will analyse the method used to conceal the allocation sequence to determine whether intervention allocation could have been foreseen in advance of or during recruitment, or changed after assignment. We will asses the method as:

  • Low risk (participants and investigators enrolling participants could not foresee assignment of allocations, e.g. telephone allocation or opaque, sealed envelopes);

  • High risk (participants or investigators enrolling participants could possibly foresee assignment of allocations, e.g. unsealed envelopes);

  • Unclear risk.

Blinding of participants and personnel (to assess performance bias)

For each included study we will analyse the method used to blind participants and personnel from knowledge of which intervention a participant received. We will assess the method as:

  • Low risk (e.g. blinding of participants and key study personnel);

  • High risk (e.g. no blinding or incomplete blinding of participants and key study personnel);

  • Unclear risk.

Blinding of outcome assessment (to assess detection bias)

For each included study we will analyse the method used to blind outcome assessors from knowledge of which intervention a participant received. We will asses the method as:

  • Low risk (e.g. blinding of outcome assessment ensured);

  • High risk (e.g. no blinding of outcome assessment);

  • Unclear risk.

Incomplete outcome data (to assess attrition bias)

For each included study we will analyse the completeness of data, including attrition and exclusions from the analysis. We will state whether attrition and exclusions were reported, the numbers included in the analysis at each stage (compared with the total numbers of randomised participants), reasons for attrition or exclusion, where reported, and whether missing data were balanced across groups or were related to outcomes. We will assess completeness of data as:

  • Low risk (e.g. no missing outcome data, < 20% missing data);

  • High risk (e.g. reason for missing outcome data likely to be related to true outcome, > 20% missing data);

  • Unclear risk.

Selective reporting (to assess reporting bias)

For each included study we will investigate the possibility of selective outcome reporting bias. We will assess the method as:

  • Low risk (e.g. study protocol is available and all of the study's pre‐specified (primary and secondary) outcomes have been reported in the pre‐specified way);

  • High risk (e.g. not all of the study's pre‐specified primary outcomes have been reported);

  • Unclear risk.

Other bias

For each included study we will analyse any other possible sources of bias. We will assess the method as:

  • Low risk (the study appears to be free of other sources of bias);

  • High risk (e.g. the study had a potential source of bias related to the specific study design used);

  • Unclear risk.

When necessary, we will request additional information and clarification of published data from trial authors. We will use discussion and consensus to resolve discrepancies and, if necessary, provided levels of agreement between review authors or details of resolution of differences, or both.

Measures of treatment effect

We will analyse the results of the studies using RevMan 2015. Each continuous outcome will be reported as a weighted mean difference (WMD) with 95% confidence intervals (CIs). We will report each categorical outcome as a risk ratio (RR) and risk difference (RD) with 95% CIs. For results that are statistically significant, we will use 1/RD to calculate the number needed to treat for an additional beneficial outcome (NNTB) or number needed to treat for an additional harmful outcome (NNTH).

Unit of analysis issues

When assessing individually RCTs, the unit of analysis will be the participating infant. However, in cluster‐RCTs the neonatal unit or the cluster that is randomised will be the unit of analysis. We plan to include cluster‐RCTs in the analyses using an estimate of the intra‐cluster correlation co‐efficient (ICC) derived from the trial (if possible), or from another source as described in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). If ICCs from other sources are used, we plan to report this and conduct sensitivity analyses to investigate the effect of variation in the ICC. If we identify both cluster‐RCTs and individually RCTs, we plan to synthesise the relevant information. We will consider it reasonable to combine the results from both if there is little heterogeneity between study designs, and the interaction between the effect of intervention and the choice of randomisation unit is considered to be unlikely.

Dealing with missing data

We will obtain missing data from the trial authors when possible. If this is not possible, we will conduct analyses on available data (i.e. ignoring the missing data). In addition, we will conduct another analysis by using the imputation method (both best‐ and worst‐case scenarios) and the last observation carried forward to the final assessment (LOCF) method for dichotomous and continuous outcome data respectively. For dichotomous outcomes, we will conduct both best‐ and worst case scenarios and intention‐to‐treat (ITT) analysis with imputation. We will compare results obtained from two analysis options to have a better understanding of the robustness of results relative to the different analytic approaches. We will consider an imputation approach of best‐case (i.e. all missing participants in the intervention group did not experience poor outcomes (e.g. death) and all missing participants in the control group experienced poor outcomes) and worst‐case (i.e. all missing participants in the intervention group experienced the event and all missing participants in the control condition did not) scenarios. We will conduct sensitivity analysis to compare results based on different imputation assumptions (i.e. best‐case vs. worst‐case scenarios). We will analyse missing continuous data on an endpoint basis, including only participants with a final assessment, or using LOCF if trial authors report these data.

Assessment of heterogeneity

We will use RevMan 2015 to assess the heterogeneity of treatment effects between trials. We will undertake this assessment using the following two formal statistical models:

  1. The Chi2 test, to assess whether observed variability in effect sizes between studies is greater than would be expected by chance. Since this test has low power when the number of studies included in the meta‐analysis is small, we will set the probability at the 10% level of significance;

  2. The I2 statistic, to ensure that pooling of data is valid. We will grade the degree of heterogeneity as either none (< 25%), low (25% to 49%), moderate (50% to 74%) or high (75% to 100%). Where there is evidence of apparent or statistical heterogeneity, we will assess the source of the heterogeneity using sensitivity and subgroup analyses, looking for evidence of bias or methodological differences between trials.

Assessment of reporting biases

If we identify at least 10 studies that include a specific intervention (comparison) reporting on the same outcome, we will assess reporting and publication bias by examining degree of asymmetry of a funnel plot in RevMan 2015.

Data synthesis

We will perform statistical analyses according to the recommendations of the Cochrane Neonatal Review Group (http://neonatal.cochrane.org). We will analyse all infants randomised on an ITT basis and treatment effects in the individual trials. We will used a fixed‐effect model to combine the data. For any meta‐analyses analysing categorical outcomes, we plan to calculate typical estimates of RR and RD, each with 95% CIs. For any meta‐analyses analysing continuous outcomes, we plan to calculate the weighted mean difference (WMD) with 95% CIs if outcomes are measured in the same way between trials, and standardized mean difference (SMD) with 95% CIs to combine trials that measure the same using different scales. If high heterogeneity exists, we will not report a typical effect. When meta‐analysis is judged to be inappropriate, we will analyse and interpret individual trials separately.

We will construct 'Summary of findings' tables with the assistance of GRADEpro 2014 software.

We will include the following outcomes in the 'Summary of findings' tables that address comparisons for the prevention and treatment of upper gastrointestinal bleeding: all cause mortality to near term age or discharge; (for 'prevention' 'Summary of findings' table) any upper gastrointestinal bleeding in infants at risk of upper gastrointestinal bleeding; (for 'treatment' 'Summary of findings' table) duration of any upper gastrointestinal bleeding in infants with upper gastrointestinal bleeding (days); gastric lesions detected by endoscopy and macroscopically; anaemia requiring blood transfusion; NEC as defined by Bell stages 1‐4; ventilator associated pneumonia (new or progressive infiltrate with positive respiratory specimens after 48 hours of mechanical ventilation); and duration of hospital stay (days).

Subgroup analysis and investigation of heterogeneity

If sufficient data are available, we will explore potential sources of clinical heterogeneity through the following a priori subgroup analyses:

  • Gestational age (< 32 weeks, 33 to 36 weeks, or > 36 weeks gestation);

  • Weight for age z‐score;

  • Risk factor present for the development of upper gastrointestinal bleeding (e.g. stress‐induced gastritis or ulcers; trauma or mechanical ventilation; NSAID use; preterm birth; birth‐asphyxiation; neonatal cyanosis; neonatal seizures);

  • Higher vs. lower dose of pharmacological intervention (median dosing recommendation as the threshold).

Sensitivity analysis

If sufficient data are available, we will explore methodological heterogeneity through sensitivity analyses. We will perform these by including only trials with adequate allocation concealment, randomisation or blinding of treatment and < 10% loss to follow‐up.