Restrictive versus Liberal Fluid Therapy in Abdominal Surgery

Doctors about to start abdominal surgery on patient.

By Connor Emdin
August 28, 2018

Abdominal surgery can result in significant fluid loss, arising from multiple sources, including fasting prior to surgery, evaporation during surgery, from blood loss and from other sources.1 Traditionally, individuals undergoing abdominal surgery received liberal fluid resuscitation (up to 7L on the day of surgery) which frequently exceeded their losses and led to weight gain of 3-6kg.2 Excessive fluid loading can lead to elevated rates of postoperative heart failure, arrhythmias, and wound infection due to local tissue edema, and in small clinical trials, restrictive hydration strategies (targeting net zero fluid balance) have been associated with fewer complications than liberal fluid replacement.3 Consequently, clinical guidelines now recommend more restrictive fluid therapy for abdominal surgery.4,5

The Study

However, despite these clinical guidelines, no large-scale randomized trial has demonstrated superior outcomes with restrictive versus liberal fluid replacement. Australian investigators therefore conducted the Restrictive versus Liberal Fluid Therapy in Major Abdominal Surgery (RELIEF) trial to assess the risks and benefits of restrictive versus liberal fluid replacement regimens.1 Three thousand patients undergoing major abdominal surgery were randomly assigned to either a restrictive or a liberal fluid regimen. The restrictive fluid regimen aimed for net-zero fluid replacement, with a 5 ml/kg bolus of fluid with induction of anesthesia, 5 ml/kg per hour infusion during surgery, and 0.8 ml/kg per hour infusion postoperatively. In the liberal fluid regimen strategy, induction of anesthesia occurred with an 8 ml/kg bolus of fluid, followed by infusion of fluids at 8 ml/kg per hour during surgery and 1.5 ml/kg per hour postoperatively. Fluids were continued up to 24 hours and it was expected that the restrictive group would receive around half the volume received by the liberal group.

Patients enrolled were those undergoing major abdominal surgery, with an expected inpatient stay of at least three days. Those enrolled were selected to be at higher risk of surgical complications, and the baseline characteristics reflect that. The average age was 66 years, 60% of patients had hypertension, ~15% had coronary artery disease, 16% had COPD, and 7% had moderate or severe renal disease (which was not defined by GFR). The primary outcome of the trial was disability-free survival one year after surgery. Secondary outcomes included acute kidney injury and the composite of death or septic complications at 30 days.

The Results

Outcomes for 2983 participants were available for analysis. The cumulative total for intravenous fluid 24 hours after surgery was 6.1 L in the liberal fluid group and 3.7 L in the restrictive fluid group.

Overall, there was no significant difference in the primary outcome of disability-free survival at one year or the secondary outcome of sepsis or death (see Table). However, the restrictive fluid group had a higher rate of acute kidney injury than the liberal fluid group (5.0% in the liberal group versus 8.6% in the restrictive group, p<0.001) and had a higher rate of renal replacement therapy (0.3% in liberal group versus 0.9% in the restrictive group, p=0.048). Higher rates of surgical site infection were also seen in the restrictive group (16.5% versus13.6% in liberal group, p=0.02).

Table. Outcomes of restrictive fluid transfusion versus liberal fluid transfusion in patients undergoing major abdominal surgery.

Outcome Liberal Fluid Regimen (n=1493) Restrictive Fluid Regimen (n=1490) Hazard Ratio p-value
Disability-free survival at one year 1232 (82.3%) 1223 (81.9%) 1.05 (0.88, 1.24) 0.61
Sepsis or death at thirty days 295/1487 (19.8%) 323/1481 (21.8%) 1.10 (0.96, 1.27) 0.19
Sepsis at thirty days 129/1487 (8.7%) 157/1481 (10.6%) 1.22 (0.98, 1.52) 0.08
Acute kidney injury at thirty days 72/1439 (5.0%) 124/1443 (8.6%) 1.71 (1.29, 2.27) <0.001
Renal replacement therapy at ninety days 4/1462 (0.3%) 13/1460 (0.9%) 3.27 (1.01, 13.8) 0.048
Median Duration of ICU Stay (days) 1.4 (0.9, 2.9) 1.8 (1.0, 3.1) NA 0.13

As would have been suggested by the structure of the protocol, the restrictive group also had a higher rate of vasopressor use (78.2% in the liberal group versus 81.7% in the restrictive group, p=0.02). Low rates of urine output were common in this study, and in the liberal arm, 27.4% of patients suffered oliguria or anuria, whereas 38.7% were oliguric or anuric in the restrictive arm.

There are several factors about this study that influence the interpretation of the results. According to the protocol, bolus fluids could be given to replace blood loss, (ml for ml replacement) but in contrast to most standard clinical practices, oliguria was not an indication for additional volume. Indeed, based on a comment in the supplementation data, it appears that investigators were encouraged not to consider declining urine output as an indicator of hypovolemia or decreased renal perfusion.1 In keeping with this, in the restrictive group, vasopressors could be used first line for treatment of hypotension in patients without signs of hypovolemia whereas fluid boluses were indicated in the liberal arm.

Of note, the median weight gain in the liberal arm was 1.6kg, as opposed to 0.6kg in the restrictive arm. As the editorialists point out, this degree of weight gain was considerably lower than that used in the prior trials of restrictive versus liberal fluid resuscitation.6 Furthermore, despite a plan to encourage patients to resume intake in the early postoperative period, this occurred less frequently than anticipated, with half the patients not eating on postoperative day 2.


The primary outcome of the study suggests that there is no mortality benefit to highly restrictive fluid replacement protocol as compared to more liberal fluid replacement. Although the primary outcome was similar between the two groups, the liberal fluid transfusion group had a lower rate of acute kidney injury, renal replacement, and surgical site infection. Indeed, the study findings emphasize an important clinical point—measures to avoid over-hydration should not occur at the cost of the appropriate management of hypovolemia and oliguria, i.e, fluid replacement.6 The results cannot be interpreted as favoring the prior standard of ‘liberal’ fluid resuscitation (where patients gained up to 6kg by post-op day 1) but do indicate that a highly restrictive protocol for fluid replacement is associated with increased risks of AKI, but without overall mortality benefit.

Update and Board Review in Internal Medicine Online

Register for a self-paced CME course with Harvard Medical School:
Update and Board Review in Internal Medicine Online: Part One
Update and Board Review in Internal Medicine Online: Part Two


  1. Myles PS, Bellomo R, Corcoran T, et al. Restrictive versus Liberal Fluid Therapy for Major Abdominal Surgery. N Engl J Med. 2018;378(24):2263-2274. doi:10.1056/NEJMoa1801601.
  2. Tambyraja AL, Sengupta F, MacGregor AB, Bartolo DCC, Fearon KCH. Patterns and clinical outcomes associated with routine intravenous sodium and fluid administration after colorectal resection. World J Surg. 2004;28(10):1046–51–discussion1051–2. doi:10.1007/s00268-004-7383-7.
  3. Brandstrup B, Tønnesen H, Beier-Holgersen R, et al. Effects of intravenous fluid restriction on postoperative complications: comparison of two perioperative fluid regimens: a randomized assessor-blinded multicenter trial. Ann Surg. 2003;238(5):641-648. doi:10.1097/01.sla.0000094387.50865.23.
  4. Gustafsson UO, Scott MJ, Schwenk W, et al. Guidelines for perioperative care in elective colonic surgery: Enhanced Recovery After Surgery (ERAS®) Society recommendations. Clin Nutr. 2012;31(6):783-800. doi:10.1016/j.clnu.2012.08.013.
  5. Feldheiser A, Aziz O, Baldini G, et al. Enhanced Recovery After Surgery (ERAS) for gastrointestinal surgery, part 2: consensus statement for anaesthesia practice. Acta Anaesthesiol Scand. 2015;60(3):289-334. doi:10.1111/aas.12651.
  6. Brandstrup B. Finding the right balance. N Engl J Med. 2018:378(24):2335-2336

Connor Emdin_headshot_150x127Connor Emdin is a post-doctoral research fellow in Sek Kathiresan’s lab at the Broad, specializing in the genetics of cardiovascular disease. He completed his doctorate in cardiovascular epidemiology at the University of Oxford from 2009-2013.

Follow Connor on Twitter: @connoremdin

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