Study characteristics

Methods

Large multicentre (two centres) double‐blind RCT

Participants

Country: UK. Included 1147 participants (578 on PPI; 569 on placebo). PU 43.9%; oesophageal varices 2.5% of total participants. Excluded people with "bleeding of such severity that immediate surgery was indicated", bleeding that developed in inpatients, and people on warfarin

Interventions

1. Omeprazole 80 mg IV immediately, then 3 doses of 40 mg IV at 8‐hour intervals, then 40 mg oral every 12 hours for 101 hours or until surgery, discharge or death

2. Identical regimen with IV and oral placebo (the IV placebo was mannitol)

Post‐intervention drug treatment at discretion of physician. Initial endoscopic haemostatic treatment at discretion of endoscopist (administered only to a minority of high‐risk participants)

Outcomes

40‐day mortality; rebleeding; surgery; stigmata of recent haemorrhage at index endoscopy; time to discharge; number of participants requiring blood transfusion

40‐day mortality, rebleeding and surgery were also reported by peptic ulcer site.

Notes

Only a few of the high‐risk participants with PU received initial endoscopic treatment. Timing of assessment of rebleeding and surgery not clear. There is a typo causing uncertainty about the results for initial endoscopic haemostatic treatment (page 145): “Thirty nine patients received endoscopic treatment (15 in omeprazole group, 17 in placebo group)". 39 is different that the sum of 15 plus 17; therefore, one of the three numbers is a typo.

Risk of bias

Bias

Authors' judgement

Support for judgement

Random sequence generation (selection bias)

Low risk

No description of sequence generation was provided by the authors, but given that block randomisation was used ("treatments were randomised in blocks of 10") and the overall conduct quality of the study, we judged that the sequence generation method was probably adequate.

Allocation concealment (selection bias)

Unclear risk

No description of the methods applied to ensure allocation concealment

Blinding of participants and personnel (performance bias)
All outcomes

Low risk

The authors stated that the study was "double blind" and that mortality assessors were blinded, although they did not specify who else was blinded. The authors also reported that the appearances of study treatment and placebo were identical. We judged that participants and personnel were probably blinded.

Blinding of outcome assessment (detection bias)
All outcomes

Low risk

The authors stated that the study was "double blind" and that mortality assessors were blinded (mortality was the predefined main outcome), although they did not specify who else was blinded. The authors also reported that the appearances of study treatment and placebo were identical. Although there is no clear statement for assessors for outcomes other than mortality, we judged that probably all outcome assessors were blinded.

Incomplete outcome data (attrition bias)
All outcomes

Low risk

Initially 1154 participants were randomised. The authors reported that 4 participants were not given the study treatment (1 in the omeprazole group and 3 in the placebo group), and in 3 participants, the treatment given could not subsequently be clearly identified. Hence, 1147 were "successfully randomised" (578 in the omeprazole group and 569 in the placebo group). The authors provided the reasons for protocol violations in 98 participants (62 participants were prescribed concomitant H₂RA, 18 entered the trial for second time, 15 started treatment more than 12 hours after admission, 3 were under age or possibly pregnant). The reasons for protocol violations were not separately reported for each study arm. However, both intention‐to‐treat and per‐protocol analyses were reported, and for each outcome, the results of those two analyses did not differ substantially. Overall, we judged that it is unlikely that incomplete outcome data would have biased the results.

Selective reporting (reporting bias)

Low risk

A preregistered protocol is not available, but the authors have reported all the key outcomes (mortality, rebleeding and surgery) that are expected to have been recorded in an upper GI bleeding study.

Other bias

Low risk

No evidence of other bias

Study characteristics

Methods

Multicentre (two centres) double‐blind RCT

Participants

Country: UK. 414 participants in total (102 on PPI; 103 on placebo; 103 on tranexamic acid; 106 on tranexamic acid plus PPI). In the PPI and placebo groups combined: peptic ulcer bleeding 37.2%; bleeding from oesophageal varices 7.1%. Excluded "bleeding so severe as to require immediate surgical intervention"

Interventions

1. Lansoprazole 60 mg orally (stat), followed by 30 mg orally plus dummy medication four times daily for four days or discharge or until a clinical endpoint occurred

2. Placebo ‐ double dummy technique

3. Tranexamic acid 2 g orally (stat), followed by 1 g orally plus dummy medication four times daily for four days or discharge or until a clinical endpoint occurred

4. Tranexamic acid and lansoprazole ‐ both active drugs as above for four days or discharge or until a clinical endpoint occurred.

Post‐intervention drug treatment not mentioned. Initial endoscopic haemostatic treatment (adrenaline injection) offered for participants with active bleeding

Outcomes

30‐day mortality; 30‐day surgery; rebleeding (timing unclear); stigmata of recent haemorrhage at index endoscopy; number of participants requiring blood transfusion

Notes

In meta‐analysis, we included group 1 (lansoprazole alone) as active treatment group and group 2 (placebo) as control group. Participants who received tranexamic acid or the combination of PPI and tranexamic acid were not included in meta‐analysis. Timing of assessment of rebleeding not clear

Risk of bias

Bias

Authors' judgement

Support for judgement

Random sequence generation (selection bias)

Low risk

No description of sequence generation was provided by the authors, but given that block randomisation was used ("randomised in blocks of 4") and the overall conduct quality of the study, we judged that the sequence generation method was probably adequate.

Allocation concealment (selection bias)

Unclear risk

The authors did not report methods of allocation concealment.

Blinding of participants and personnel (performance bias)
All outcomes

Low risk

The authors described this study as "double blind, double dummy". Although they did not specifically state who was blinded, it is reasonable to presume that participants and personnel were blinded.

Blinding of outcome assessment (detection bias)
All outcomes

Low risk

The authors described this study as "double blind, double dummy". Although they did not specifically state who was blinded, it is reasonable to presume that outcome assessors were blinded.

Incomplete outcome data (attrition bias)
All outcomes

Low risk

Authors provide a clear disposition of participants at each stage after randomisation, with separate results for each of the 4 study arms. Of the 414 participants enrolled, 55 not endoscoped. Of the 298 endoscoped, 61 were not GI bleed. 248 participants were eligible and 50 not eligible for further evaluation due to protocol violations (39 more than 72 hours after start of bleeding, 9 more than 8 hours from 1st dose to endoscopy, 2 participants with no trial data). 20 further participants were not evaluable (14 participants missed 2 or more doses, 4 endoscopy received before trial treatment, 3 prohibited drugs during trial and 1 previously in trial). At each stage after randomisation, the reasons for withdrawal were adequately balanced amongst the 4 treatment arms. Clinical outcomes were analysed on an intention‐to‐treat basis

Selective reporting (reporting bias)

Low risk

A preregistered protocol is not available, but the authors have reported all the key outcomes (mortality, rebleeding and surgery) that are expected to have been recorded in an upper GI bleeding study.

Other bias

Low risk

No evidence of other bias

Study characteristics

Methods

Single centre, open RCT

Participants

Country: Turkey. Included 58 participants (30 on PPI, 28 on H₂RA). Peptic ulcer bleeding 77.6%. Excluded people with "massive bleeding requiring immediate surgery".

Interventions

1. Omeprazole 80 mg IV bolus (over 5 min) as soon as possible after admission, followed by 40 mg IV bolus once a day for 6 days; then, omeprazole 20 mg orally once a day for 3 weeks
2. Ranitidine 100 mg IV bolus (over 5 min) as soon as possible after admission, followed by 100 mg IV bolus 3 times a day for 6 days; then, famotidine 40 mg orally once a day for 3 weeks

Outcomes

Mortality, rebleeding, stigmata of recent haemorrhage at index endoscopy, number of participants requiring blood transfusion

Notes

This study was published as an abstract in English, and as full text in Turkish. The authors kindly provided a complete translation of the full text in English.

Risk of bias

Bias

Authors' judgement

Support for judgement

Random sequence generation (selection bias)

Unclear risk

The authors reported no details on sequence generation.

Allocation concealment (selection bias)

Unclear risk

The authors reported no details on allocation concealment.

Blinding of participants and personnel (performance bias)
All outcomes

High risk

No mention of blinding status or measures to ensure blinding in the full‐text publication. The primary author stated by email that "patiens [sic] were randomised double‐blindedly [sic]" but no specific measures to ensure blinding were mentioned. The PPI and the H₂RA were distinctly different with regards to preparation (omeprazole was available as dried powder flacon (small bottle with a cap) with a 10 mL solvent, while ranitidine was available as 50 mg/2 ml ampoules) and administration frequency (omeprazole was administered once daily, while ranitidine was administered 3 times a day). Therefore, without specific measures to ensure blinding (i.e. double dummy methods), it is unlikely that participants and personnel would have been or remained blinded. Thus, we considered the study to be at high risk of performance bias for all outcomes.

Blinding of outcome assessment (detection bias)
All outcomes

High risk

The authors did not mention blinding status or measures to ensure blinding. Given that the PPI and the H₂RA were distinctly different with regards to preparation and administration frequency, without specific measures to ensure blinding (i.e. double dummy methods), the outcome assessors would have been unblinded.The risk of detection bias is low for the outcome of mortality, but high for all other outcomes (including the outcome of rebleeding, given that the criteria for rebleeding required some subjective judgements).

Incomplete outcome data (attrition bias)
All outcomes

Low risk

The authors provided clear disposition for all randomised participants. No participant was lost within the first 6 days (all participants underwent scheduled endoscopy on day 1 and day 6). 20% of the participants in the PPI group were not endoscoped in the 1^st month (one participant died of massive upper GI bleeding due to end‐stage liver disease, two participants moved to another city, one participant was operated on for appendicitis, two participants refused repeat endoscopy). 29% of the participants in the H₂RA group were not endoscoped in the 1^st month (one participant died of pancreatic carcinoma with massive upper GI bleeding, one participant had coronary bypass surgery, one participant had unstable heart disease, 3 participants refused the endoscopy procedure and 2 participants with unknown reasons could not be re‐examined). Athough the authors omitted these participants "from the rest of the study", their detailed reporting allowed us to include them in our intention‐to‐treat analysis (30‐day endoscopy was not expected to influence any of the outcomes of interest of this review).

Selective reporting (reporting bias)

Low risk

A preregistered protocol is not available, but the authors have reported all the key outcomes (mortality, rebleeding and surgery) that are expected to have been recorded in an upper GI bleeding study.

Other bias

Low risk

No evidence of other bias

Study characteristics

Methods

Single centre, double‐blind RCT

Participants

Setting: Hong Kong. 631 participants randomised (314 in the PPI group and 317 in the placebo group). Peptic ulcer bleeding 59.7%; variceal bleed 3.8%. Excluded long‐term aspirin users (these participants were enrolled in another concurrent trial) and those with refractory shock requiring emergency endoscopy

Interventions

1. Intravenous omeprazole 80 mg bolus at randomisation and continuous intravenous infusion at 8 mg/hour until endoscopy

2. Intravenous placebo bolus followed by continuous intravenous infusion until endoscopy

Participants with high‐risk stigmata requiring endoscopic treatment were transferred to a gastroenterology ward and treated with PPI 8 mg/hour infusion for 72 hours followed by 8 weeks of oral omeprazole 40 mg once daily; (if they were H pylori positive, they were given one‐week triple eradication therapy, followed by 7 weeks of oral omeprazole 40 mg once daily).

Participants who did not require endoscopic therapy returned to the general medical ward; their medical treatment after endoscopy was not described clearly, but most likely it was oral omeprazole 40 mg once daily. It was not stated if their treatment after discharge was the same as for the participants discharged from the gastroenterology ward.

Outcomes

30‐day mortality, rebleeding, surgery, proportion of participants requiring endoscopic therapy, length of hospital stay and mean units of blood transfusion reported according to treatment group. Stigmata of haemorrhage reported only in peptic ulcer participants

Notes

Risk of bias

Bias

Authors' judgement

Support for judgement

Random sequence generation (selection bias)

Low risk

A "computer‐generated list of random numbers in blocks of 20" was used

Allocation concealment (selection bias)

Low risk

Consecutively numbered sealed packages were prepared centrally in the pharmacy and sent to the wards with lowest numbered pack to be opened by the resident treating the participant.

Blinding of participants and personnel (performance bias)
All outcomes

Low risk

Participants and all investigators were blinded to the treatment groups.

Blinding of outcome assessment (detection bias)
All outcomes

Low risk

Participants and all investigators were blinded to the treatment groups.

Incomplete outcome data (attrition bias)
All outcomes

Low risk

Clear reporting of disposition of participants per study arm at each stage of the study. Only 5/319 (PPI arm) and 2/319 (placebo arm) participants were excluded from the analyses (3 received the wrong diagnosis, 1 had small‐bowel obstruction, 1 had a history of total gastrectomy, 1 had cholangitis, 2 withdrew voluntarily). 2 of the 314 participants did not undergo endoscopy. Of the 319 participants in the placebo group, 2 were excluded from the analysis, 1 because of wrong diagnosis and the other withdrew voluntarily.

Selective reporting (reporting bias)

Low risk

The authors have measured and reported all outcomes listed in the pre‐registered protocol (NCT00164866).

Other bias

Low risk

No evidence of other bias

Study characteristics

Methods

Single centre, open RCT

Participants

Country: Croatia. Participants with acute upper GI bleeding admitted to intensive care unit.

Interventions

1. IV pantoprazole 80 mg bolus after randomisation and IV 40 mg three times per day IV for 5 days.

2. The control group received no treatment until endoscopy and thereafter was treated with pantoprazole 40 mg IV bolus followed by IV 40 mg three times per day for 5 days.

Outcomes

Mortality, rebleeding, surgery, length of stay in ICU, mean number of blood units transfused and lesion stabilisation (on repeat endoscopy at 5 days). No mention of follow‐up duration and time of outcome measurement

Notes

Risk of bias

Bias

Authors' judgement

Support for judgement

Random sequence generation (selection bias)

Unclear risk

Abstract publication. The authors did not provide details of sequence generation.

Allocation concealment (selection bias)

Unclear risk

Abstract publication. The authors did not provide details of allocation concealment.

Blinding of participants and personnel (performance bias)
All outcomes

High risk

Abstract publication. The authors did not mention blinding status or measures to ensure blinding. Given that the control group did not receive placebo treatment, the study was probably non‐blinded.

Blinding of outcome assessment (detection bias)
All outcomes

High risk

Abstract publication. The authors did not mention blinding status or measures to ensure blinding. Given that the control group did not receive placebo treatment, the study was probably non‐blinded. The risk of detection bias is low for the outcome of mortality, but high for all other outcomes (including the outcome of rebleeding, given that the criteria for rebleeding required some subjective judgements).

Incomplete outcome data (attrition bias)
All outcomes

Unclear risk

Abstract publication. Authors do not provide disposition of participants at each stage of study. Hence it is difficult to assess the completeness of the data. The authors do not report withdrawal or dropout data.

Selective reporting (reporting bias)

High risk

Abstract publication. A preregistered protocol is not available. Results for mortality were not reported numerically; the authors simply state that "a difference in mortality rates has not yet been demonstrated". (Selective under‐reporting of data on mortality: reported but with inadequate detail for the data to be included in a meta‐analysis)

Other bias

Low risk

No evidence of other bias

Study characteristics

Methods

Single centre, open RCT

Participants

Country: Poland. Included 102 participants (50 on PPI and 52 on placebo). Peptic ulcer bleeding 75.5% of total; no participants with oesophageal varices (hepatic insufficiency was an exclusion criterion).

Interventions

1. Omeprazole delivered via IV bolus; the dosing regimen is unclear: stated as "40 mg" or "80 mg" or "120 mg" (presumably representing total daily doses)

2. Ranitidine delivered via IV bolus; the dosing regimen is unclear: stated as "150 mg" or "200 mg" or "300‐400 mg" (presumably representing total daily doses)

Unclear if participants within each treatment arm were allocated to each dosing group by a random method or not. Duration of treatment depending on continuation of bleeding. Initial endoscopic haemostatic treatment not mentioned.

Outcomes

Mortality; surgery; stigmata of recent haemorrhage at index endoscopy; number of participants requiring blood transfusion. Timing of outcome assessment not clear

Notes

Timing of assessment of rebleeding not clear. Initial endoscopic haemostatic treatment not mentioned. Dosing of pharmacological treatments not clear. Rebleeding rates could not be extracted because the study was designed to assess time needed for bleeding cessation.

Risk of bias

Bias

Authors' judgement

Support for judgement

Random sequence generation (selection bias)

Low risk

Series of random odd and even numbers generated by computer

Allocation concealment (selection bias)

Unclear risk

No details of allocation concealment were reported

Blinding of participants and personnel (performance bias)
All outcomes

High risk

The authors described the study as "open". It was not a blinded study.

Blinding of outcome assessment (detection bias)
All outcomes

High risk

The authors described the study as "open". It was not a blinded study. The risk of detection bias is low for the outcome of mortality, but high for all other outcomes (including the outcome of rebleeding, given that the criteria for rebleeding required some subjective judgements).

Incomplete outcome data (attrition bias)
All outcomes

Low risk

Reported disposition of all randomised participants and outcomes for all participants randomised. No withdrawals or dropouts were observed during the study period.

Selective reporting (reporting bias)

High risk

A preregistered protocol is not available. Rebleeding not reported as an outcome (the study was designed to assess cessation of bleeding at day 5 with clinical and laboratory criteria).

Other bias

Low risk

No evidence of other bias