Restrictive vs Liberal Blood Transfusion for Acute Upper Gastrointestinal Bleeding: Rationale and Protocol for a Cluster Randomized Feasibility Trial

Abstract

Acute upper gastrointestinal bleeding (AUGIB) is the commonest reason for hospitalization with hemorrhage in the UK and the leading indication for transfusion of red blood cells (RBCs). Observational studies suggest an association between more liberal RBC transfusion and adverse patient outcomes, and a recent randomised trial reported increased further bleeding and mortality with a liberal transfusion policy. TRIGGER (Transfusion in Gastrointestinal Bleeding) is a pragmatic, cluster randomized trial which aims to evaluate the feasibility and safety of implementing a restrictive versus liberal RBC transfusion policy in adult patients admitted with AUGIB. The trial will take place in 6 UK hospitals, and each centre will be randomly allocated to a transfusion policy. Clinicians throughout each hospital will manage all eligible patients according to the transfusion policy for the 6-month trial recruitment period. In the restrictive centers, patients become eligible for RBC transfusion when their hemoglobin is < 8 g/dL. In the liberal centers patients become eligible for transfusion once their hemoglobin is < 10 g/dL. All clinicians will have the discretion to transfuse outside of the policy but will be asked to document the reasons for doing so. Feasibility outcome measures include protocol adherence, recruitment rate, and evidence of selection bias. Clinical outcome measures include further bleeding, mortality, thromboembolic events, and infections. Quality of life will be measured using the EuroQol EQ-5D at day 28, and the costs associated with hospitalization for AUGIB in the UK will be estimated. Consent will be sought from participants or their representatives according to patient capacity for use of routine hospital data and day 28 follow up. The study has ethical approval for conduct in England and Scotland. Results will be analysed according to a pre-defined statistical analysis plan and disseminated in peer reviewed publications to relevant stakeholders. The results of this study will inform the feasibility and design of a phase III randomized trial.

Perioperative blood transfusion in gynecologic oncology surgery: Analysis of the National Surgical Quality Improvement Program Database

“…Similarly, the Transfusion Trigger Trial for Functional Outcomes in Cardiovascular Patients Undergoing Surgical Hip Fracture Repair (FOCUS) [21] trial showed no difference in mortality in a cardiovascular high-risk cohort undergoing hip surgery randomized to restrictive (transfused for Hb < 8 g/dL) or liberal (transfused for Hb < 10 g/dL) approach. A third randomized controlled trial of patients with acute upper gastrointestinal bleeding revealed that a restrictive strategy (transfused for Hb <7 g/dL) was associated with decreased mortality and adverse events compared to rates in the liberal group (transfused for Hb < 9 g/dL) [22]. Expert opinion supports a restrictive transfusion protocol utilizing hemoglobin triggers of 7 – 8g/dL[23–25].…”

How big should the pilot study for my cluster randomised trial be?

“…The decision about number of participants in each cluster will, in turn, determine the total number of participants. Okwen et al, 8 Jairath et al, 16 and Leblanc et al 18 are examples of this. In our simulations, we only applied this situation to studies involving naturally large clusters, because when clusters are naturally small entire clusters would usually be invited to participate leaving no flexibility in number of participants per cluster.…”

“…The pilot study examples [5][6][7][8][14][15][16][17][18][19][20] are representative of a range of settings and vary in the included numbers of clusters. All involve randomisation of clusters into two or three groups, seven with equal allocation and three in which there was one extra cluster either in the intervention or control group.…”

“…Four of the examples were conducted in hospital in-patients facilities and include as clusters: designated elderly care hospital bays in wards in England (one designated bay from each of eight hospitals randomised into two groups of four) 5 ; adult non-maternity in-patient wards from two NHS Trusts in London (18 hospital wards randomised into three groups of six) 15 ; UK general hospitals admitting emergency Acute Upper Gastrointestinal Bleeding cases (six academic centres randomised into two groups of three) 16 and in-patient rehabilitation centres in New Zealand (four centres randomised into two groups of two). 20 Three studies were conducted in community-based health care facilities or pharmacies; government primary health care and dispensary centres in Tanzania (36 dispensaries randomised into two groups of 18) 8 ; primary care Family Medicine Groups (FMGs) from two selected health regions in Canada (four FMGs randomised to two groups of two, with one additional late entry added to control group) 18 ; and Scottish community pharmacies (12 pharmacies randomised into two groups of six).…”

Update on the transfusion in gastrointestinal bleeding (TRIGGER) trial: statistical analysis plan for a cluster-randomised feasibility trial

“…The Transfusion in Gastrointestinal Bleeding (TRIGGER) trial is a cluster-randomised trial designed to evaluate the feasibility of implementing a restrictive vs. liberal RBC transfusion policy for patients admitted to hospital with AUGIB [8]. TRIGGER is a pragmatic trial which aims to reflect real world settings as closely as possible.…”