How much fluid should patients having abdominal surgery receive?

The RELIEF Trial: Restrictive versus Liberal Fluid Therapy for Major Abdominal Surgery. Myles et al. New England Journal of Medicine DOI: 10.1056/NEJMoa1801601

“I’m a lover of my own liberty, and so I will do nothing to restrict yours.” – Mahatma Gandhi


Traditionally, perioperative fluid regimes have tended to be liberal but this has been associated with pulmonary complications, renal injury, sepsis and poor wound healing. Such concerns have led to smaller studies exploring restrictive regimes and although these protocols have suggested fewer perioperative complications, and been incorporated into ERAS guidelines, there are concerns regarding hypotension and reduced organ perfusion with restrictive fluid. The authors felt the evidence in major abdominal surgery was inconclusive.

Primary hypothesis

A restrictive fluid regime will have a lower rate of complications and a higher rate of disability-free survival than a liberal regime.


3000 elective patients undergoing major abdominal surgery of at least 2 hours duration who were at risk of complications were included.

Patients were randomly assigned to either a liberal or restrictive intravenous fluid group during surgery and for 24 hours post-operatively.

Liberal: 10ml/kg bolus on induction of anaesthesia, 8mls/kg/hr during surgery and 1.5mls/kg/hour for 24 hours post-operatively. Fluids could be reduced intra-operatively after 4 hours and the post-operative fluid rate could be altered if required.

Restrictive: 5mls/kg bolus, 5mls/kg/our intra-operatively and 0.8mls/kg/hour for 24 hours post-operatively. Fluids could be altered post-operatively if required but a vasopressor was 1st line for hypotension if the patient was euvolemic.

Oliguria was not used as a marker of filling in either group.

Outcome measures

Primary outcome: disability-free survival up to 1 year post-op

Secondary outcomes: AKI, 30 day mortality, major infective complications, serum lactate, peak CRP, transfusion rates, length of stay, unplanned ICU admission and quality of recovery.


The study ran from May 2013 to September 2016 in 47 centres across 7 countries with 3000 patients.

The mean age of patients was 66 years and the commonest comorbidities were HTN (60%), CAD (15%) and COPD (14%). 43% underwent colorectal surgery and 64% of patients underwent cancer surgery.

1 year data was available for 96.7% of patients


Liberal group Restrictive group
Administered fluids at 24 hours: 6146mls  (5000-7410) 3671mls (2885-4880) p=<0.001
Median weight gain 1.6kg (0-3.6) 0.3kg (-1 –1.9) ND

Primary outcome:

The rate of disability free survival at 1 year was 81.9% in the restrictive group and 82.3% in the liberal group (CI 0.88-1.24; P=0.61).

* Patients recruited in NZ had improved disability-free survival in the restrictive group.

Secondary outcomes:

  • Acute kidney injury

There was a statistically higher rate of AKI in the restrictive group (8.6% vs. 5%) (p=<0.001) which correlated with more oliguric patients in the restrictive group.  AKI was not associated with a higher rate of renal replacement.

  • Blood transfusion

There was significant higher rate of RBC transfusion in the liberal group despite similar baseline Hb values and intra-op blood loss (p=0.015)

  • Other outcomes

There was no difference in any of the other secondary outcomes including infection between the groups.


Recruitment – over 2000 additional patients fulfilled criteria but were not recruited.

Generalisability – the study included a range of surgical procedures, including laparoscopic procedures, which likely all have different perioperative fluid requirements.

Fluid was only controlled up to 24 hours so we are unaware of overall fluid balance by the end of hospital stay.


The RELIEF study does suggest a moderately positive perioperative fluid balance is safe but the importance of responding to physiological parameters, including urine output, to judge fluid balance is vital. Restrictive regimes with resulting AKI should be avoided, but responding to urine output might have modified this by identifying at risk patients. In addition, iatrogenic haemodiluation with subsequent blood transfusion, that was evident in the liberal fluid application, should also be avoided.

Summary by Dr Alexa Strachan


Second Opinion

Here’s what they thought of this paper over at The Bottom Line

In the interview below, Paul Myles the lead author discusses the paper (courtesy of TopMedTalk).


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s