Open, multi-centre, randomised controlled trial of cardiac output -guided haemodynamic therapy compared to usual care in patients undergoing emergency bowel surgery. This project is funded by the National Institute for Health Research Health Technology Assessment programme (project number 15/80/54)

Background – Emergency Laparotomy burden High incidence (30,000 cases/yr UK) X High mortality (20% @ 90-day in >50yrs) = Many lives saved if treatments with modest mortality benefit are routinely adopted …though showing modest effect size = very large trial!

Background – candidate treatment Perioperative cardiac output-guided haemodynamic therapy MORTALITY: Risk Ratio 0.86 [0.74-1.00] This is the updated meta-analysis included in the report of the OPTIMISE trial. This included numerous perioperative trials, the vast majority of which were in elective surgery. Pearse et al. JAMA 2014; 311: 2181-90.

Background – candidate treatment Perioperative cardiac output-guided haemodynamic therapy However: VERY few emergency laparotomy patients studied Elective ≠ Emergency Elective surgery GDHT Emergency laparotomy GDHT? Critical care GDHT Very few emergency laparotomy patients have been included in perioperative goal-directed haemodynamic therapy trials so far. The limited evidence base is recognised by the previous Cochrane review, and by NELA itself. While the evidence base for GDHT in elective surgery suggests this intervention is beneficial, the most recent studies in critical care / sepsis suggest that GDHT for resuscitation is not helpful. Emergency laparotomy patients have similarities with both these patient groups, so should be studied in their own right to see if this treatment improves outcomes

Background – current practice, equipoise? COM-based GDT “other” GDT NELA: -Cardiac output monitor in 39% -Limited evidence base  NOT an audit standard FLO-ELA clinician survey 2016: -137 respondents -85% willing to randomise NELA Centre The use of cardiac output monitoring varies widely across the NELA centres shown here. Each horizontal bar represents a NELA centre, with “goal-directed therapy” use ranging from 0% in a couple of centres to 96% in others (at the bottom). This suggests that there is familiarity with its use in this group, but that there is no clear consensus on whether it is beneficial. % of cases with GDT

Background – potential trial funding? “Efficient Study Designs” call: …particular design features – more rapid conduct, lower costs or both – but giving sufficiently robust data… …the use of routinely collected data… …simple randomised trials… focusing on a limited number of the most important outcomes…

A unique opportunity? NELA FLO-ELA Indisputable healthcare need Delivery infrastructure: POM-CTN, NIHR CRN, trainee networks NIHR HTA Efficient Study Designs Hopeful intervention with clinician equipoise Prioritisation by clinicians and public (JLA) FLO-ELA JLA = James Lind Alliance

Design Pragmatic multicentre open randomised trial supported by data from the ongoing National Emergency Laparotomy Audit (NELA). Recruitment support and local monitoring from research nurses, “real world” intervention delivery by clinicians. Inclusion criteria: Patients aged 50+ requiring emergency bowel surgery in line with NELA criteria – with/without capacity Exclusion criteria: Patient or clinician refusal

Basic standards of care set for both groups Design Intervention: cardiac output monitoring to determine the dose & timing of intravenous fluid administration according to a suggested algorithm, during and up to six hours after emergency laparotomy. Clinician discretion on choice of cardiac output monitor shown to accurately track stroke volume changes and of isotonic fluid type for boluses. Control group: Intravenous fluid administration without the use of cardiac output monitoring or algorithm. Basic standards of care set for both groups

Design Primary outcome: mortality 90 days after surgery Secondary outcomes: 1-yr mortality, critical care and hospital LoS, cost effectiveness Sample size: 7646 patients (3823 per group) to detect an absolute reduction in mortality at 90-days from 19% to 16% with 90% power Sites and recruitment: 100 UK sites taking part in NELA, recruitment over three years from mid 2017.

Efficiencies NELA & NHS Digital for all patient and outcome data Minimal supplementary data fields to track intervention Reduced research staff requirement Research active trainees and consultants support recruitment – track record Clinicians deliver intervention – familiarity Industry support for intervention costs

Centres and Recruitment The first year of recruitment will act as an internal pilot feasibility phase, to ensure we can meet target recruitment and protocol compliance. Recruitment will continue while the results of this phase are analysed.

Key challenges 1: delivering recruitment Lots of eligible patients (~25,000/yr across NELA) Within 100 of 192 NELA centres, recruiting 30% of this group for 3yrs = sample size achieved Strong track record: EPOCH as “NELA + trial” model OPTIMISE as large contemporary GDHT trial Trainees as capable of supporting / delivering RCTs Recruiting critically ill patients +/- capacity at all hours to large RCTs POM-CTN adoption & NELA network to engage local teams Trainee networks have successfully initiated and completed interventional trials including in emergency laparotomy (CrEST, ROSSINI) POM-CTN – New UK Perioperative Medicine Clinical Trial Network, network of engaged local PIs interested in delivering multicentre trials

Key challenges 2: protocol compliance Vital to maintain separation between control/intervention group despite pragmatic design Individual clinician equipoise a prerequisite to randomisation Monitoring, feedback and management of protocol adherence / contamination – including removal of sites with poor compliance rates CHECK FIGURES

Managing the risk – feasibility phase First year of recruitment to look at: Sites opening, recruitment rates, adherence and contamination Targets: 90-100 sites open and recruiting >80% of predicted recruitment <10% protocol deviations CHECK FIGURES

FLO-ELA local Principal Investigators – key roles Champion to clinical teams Gather a team covering surgery, anaesthesia, critical care Establish working practices – trainees and front line on call teams – round the clock recruitment Encourage equipoise Support rapid site set-up Recruit average 3-4 patients per site per month when established (2-3 in hours, 1-2 out of hours) Ensure protocol compliance

Support for participating centres Funded research nurse time NIHR Portfolio Band 3 tariff Support from industry Simple design – non-CTIMP, no extensive intervention or complex outcomes follow-up for research team AND – the knowledge that you have contributed to a major study which one way or another should help to change clinical practice!

Conclusion An exciting and unique opportunity to guide clinical practice on a key intervention Large pragmatic design – great community and network support is vital Watch this space!

Senior Co-investigators: Mike Grocott, Rupert Pearse @FLOELAtrial Chief Investigator: Mark Edwards (Consultant in Anaesthesia & Perioperative Medicine, Honorary Senior Clinical Lecturer, Southampton) admin@floela.org - @dr_mark_edwards Senior Co-investigators: Mike Grocott, Rupert Pearse Trial team: Monty Mythen, Dion Morton, NELA team (Dave Murray and Iain Anderson), QMUL Pragmatic Clinical Trials Unit (Brennan Kahan, Anita Patel (Health Ec), Ann Thomson), Trainee networks (Ben Harris, Marianne Johnstone), Keith Young (lay representative)