Statistical analysis plan for Fluid Optimisation in Emergency Laparotomy (FLO-ELA) Trial; a multi-centre randomised trial of cardiac output-guided fluid therapy compared to usual care in patients undergoing major emergency gastrointestinal surgery

Abstract Background FLO-ELA is a multi-centre, parallel-group, open-label, randomised controlled trial (RCT) looking at the effect of minimally invasive cardiac output monitoring in the management of patients aged 50 years and above undergoing emergency gastrointestinal surgery. The primary outcome is days alive and out of hospital within 90 days of randomisation (DAOH-90). This article describes the statistical analysis plan (SAP) for this trial. The estimand and analytical approach presented here may be appropriate for clinical trials in a similar setting (critical care surgery). Methods and design FLO-ELA is a superiority trial, incorporating an early-stage pilot phase, conducted at 63 UK hospitals. Participants are randomised on a 1:1 basis to the intervention or control group using a minimisation algorithm which balances on the following two factors (i) age, split into 3 groups, and (ii) American Society of Anesthesiologists (ASA) category, 5 categories on an ordinal scale reflecting participants’ baseline co-morbidities. In this SAP, we define the estimands and strategies for handling intercurrent events for each study outcome, then give a detailed description of planned analyses including general analytical principles, analytical framework for primary, secondary and process measure outcomes, approach to missing data and supplementary analyses looking at heterogeneity in treatment effect between key subgroups and potential differences in outcomes pre and post onset of Covid pandemic in the UK. Trial registration ISRCTN 14729158. Registered on 02 May 2017.