CHEST
Hydroxyethyl Starch or Saline for Fluid Resuscitation in Intensive Care
Myburgh et al for the ANZICS Clinical Trials Group. NEJM 2012; 367:1901-11. doi:10.1056/NEJMoal209759
Clinical Question
- In critically ill patients requiring fluid resuscitation, does 6% hydroxyethyl starch (6% HES) compared to 0.9% sodium chloride (Saline) reduce mortality at 90 days?
Design
- Randomised, controlled, double-blinded clinical trial
- Computer generated randomisation via website providing complete concealment
- Minimisation algorithm stratified by institution and admission diagnosis of trauma
- A priori publication of trial protocol and statistical analysis plan
- Intention-to-treat analysis
- Powered at 90% to detect absolute difference of 3.5% difference in mortality at 90 days from baseline of 26%, with alpha significance of 0.05
- Planned sample size of 7000, allowing 5% loss
Setting
- 32 adult ICUs
- Mixed medical and surgical
- Australia and New Zealand
- December 2009 to January 2012
Population
- Inclusion: Attending physician judged fluid resuscitation to be needed; age over 18 years
- Exclusion: 1000ml or more 6% HES prior to ICU admission; impending or current renal replacement therapy; intracranial haemorrhage; women of child-bearing age unless known negative pregnancy state; cardiac surgery, burn injury or liver transplant.
- 19,475 patients screened; 7000 randomised
- Baseline characteristics were similar (6% HES vs Saline):
- Age: 63.1 vs 62.9 years
- Weight: 79.4 vs 78.6 kg
- Admitted from ED: 27.7% vs 27.6%
- APACHE 2 score: 17.0 vs 17.0
- Sepsis diagnosis: 29.2 vs 28.4
- RIFLE-injury at baseline: 36.0% vs 36.0%
- 6% HES prior to randomisation: 15.2% vs 15.1%
- Baseline characteristics were similar (6% HES vs Saline):
Intervention
- 6% HES (Voluven, Fresenius Kabi as 500ml bags)
Control
- 0.9% Saline (identical 500ml bags)
Both groups managed as below:
- Fluid administered at clinician’s discretion for all fluid resuscitation
- To correct hypovolaemia
- Supported by at least one objective physiological parameter
- Continued until death, discharge or day-90
- Limited at 50 ml/kg/day as per product license
- Open-label saline as required after this limit
- Change to open-label saline if renal replacement therapy required
- Maintenance and replacement fluid not controlled; fluid outside of ICU not controlled
Outcome
- Primary outcome: there was no statistically significant difference in 90-day mortality between the two groups
- 6% HES: 18.0%
- Saline: 17.0%
- Absolute risk increase: 1.0%
- Relative risk: 1.06 (95% CI 0.96 to 1.18; p=0.26)
- Secondary outcomes: comparisons are 6% HES vs Saline
- Incidence of acute kidney injury (by RIFLE scoring): reduced risk and injury, but more failure associated with 6% HES (see table).
- 6% HES associated with increased urine output but also increased serum creatinine
- Renal replacement therapy use: 7.0% vs 5.8% (RR 1.21; 95% CI 1.00 to 1.45; p=0.04) favouring Saline
- New organ failure (by SOFA scoring): less cardiovascular failure but more hepatic failure associated with 6% HES (see table)
- Duration of mechanical ventilation: 6.0 vs 5.7 days (p=0.12)
- Duration of renal replacement therapy: 5.6 vs 5.5 days (p=0.86)
- Cause-specific mortality: no statistically significant differences (see table)
- Incidence of acute kidney injury (by RIFLE scoring): reduced risk and injury, but more failure associated with 6% HES (see table).
- Tertiary outcomes:
- Fluid administered during first 4 days: 6% HES group received significantly less fluid
- 6% HES: 526 ml/day ± 425 ml/day (as mean ± SD)
- Saline: 616 ml/day ± 488 ml/day
- p<0.001
- Blood products: 6% HES group received significantly more blood products
- 6% HES: 78 ml ± 250 ml
- Saline: 60 ± 190 ml
- p<0.001
- Haemodynamic parameters were not significantly different
- Fluid administered during first 4 days: 6% HES group received significantly less fluid
Authors’ Conclusions
- “Our study does not provide evidence that resuscitation with 6% HES (130/0.4) as compared with saline, in the ICU provides any clinical benefit to the patient. Indeed the use of HES resulted in an increased rate of renal replacement therapy.”
Strengths
- Well designed, large, multi-centre randomised, controlled trial
- Biases minimised with double-blinding
- Pragmatic design – administration at clinician’s discretion
- A priori statistical plan and intention-to-treat analysis
Weaknesses
- 6% HES had been administered to 15% of patients in Saline group prior to randomisation. This small cross-over may bias towards a conclusion of ‘no difference’
- 5% loss to follow-up – this is what the authors expected, and since it is less than the primary outcome rate, it is unlikely to change the conclusion either way.
- The lower baseline mortality rate may have contributed to a false negative conclusion
- The clinician-judged need for renal replacement therapy has been criticised by some, but given clinicians were blinded, this is unlikely to have biased the conclusion
The Bottom Line
- This trial adds to the existing data: starch-based fluids do not provide a mortality benefit and lead to increased renal dysfunction requiring renal replacement therapy.
- In the context of critically ill medical and surgical patients, starch-based fluids should not be used for volume resuscitation.
External Links
- [article] Hydroxyethyl Starch or Saline for Fluid Resuscitation in Intensive Care by Myburgh et al
- [commentary] Crystalloids vs Colloids by Phillips et al
- [commentary] Starch in Sepsis from ESICM
- [further reading] Hydroxyethyl-starch solutions (HES) should no longer be used in patients with sepsis or burn injuries or in critically ill patients
Metadata
Summary author: @DuncanChambler
Summary date: 13 January 2015
Peer-review editor: @DavidSlessor
Pingback: SALT-EM Study – The Bottom Line
Pingback: SALT-EM – The Bottom Line
Pingback: 6S – The Bottom Line
One of the other weaknesses of the trial was that they were volume resuscitating patients with normal macrohemodynamic variables and metabolic status (mean base excess was +/-2). Why would anyone rightly give a volume expanding treatment to patients that were that well? One would need to see if starches save lives in truly volume depleted patients where blood product is not immediately available. This study would be near impossible to organize and execute.