Effects of fluid resuscitation with colloids vs. crystalloids on mortality in critically ill patients presenting with hypovolemic shock

In critically ill patients, there was no significant difference in 28-day mortality in patients resuscitated with colloids compared to crystalloids. However 90-day mortality was significantly higher in the crystalloid group, Level of evidence: 1B (CEBM, RCT of good quality)

Appraised by S Evans

Citation: Annane D, Siami S, Jaber S et al. Effects of fluid resuscitation with colloids vs crystalloids on mortality in critically ill patients presenting with hypovolemic shock: The CRISTAL randomized trial. JAMA 2013;310:1809-1817.

Lead Author: Djillali Annane, [e-mail]

Three-part clinical question:

Patients: Adult patients accepted to the intensive care unit (ICU) requiring fluid resuscitation for acute hypovolaemia.

Intervention: Patients received either crystalloids (isotonic/hypertonic saline or any buffered solutions) or colloids (hypo-oncotic (e.g. gelatins, 4%/5% of albumin)) and hyper-oncotic (e.g. dextrans, hydroxyethyl starches and 20%/25% of albumin) solutions. Within each treatment group, investigators could use whichever fluids were available at their institution. The amount of fluid and duration of treatment were left at the discretion of the investigators within certain restrictions. These were that the daily total dose of hydroxyethyl starch could not exceed 30 ml/kg and investigators were required to follow any local regulatory agency recommendations governing use. Patients were subsequently managed exclusively within the fluid category to which they were randomised until ICU discharge. Exceptions were that regardless of treatment group, maintenance fluids were isotonic crystalloids and if physicians wished, they could administer albumin in response to demonstrated hypoalbuminemia (serum albumin <20 g/dL).

Outcomes:

1. Primary outcome: Mortality at 28 days.

2. Secondary outcomes: Mortality at 90 days, ICU discharge and hospital discharge; number of days alive and not receiving renal replacement therapy, mechanical ventilation or vasopressor therapy; days without organ system failure (i.e. Sequential Organ Failure Assessment (SOFA) score <6); days not in the ICU/hospital.

3. Study design: International, multi-centre, prospective, parallel group, randomised clinical trial. The CRISTAL Trial was designed to test whether colloids altered mortality compared with crystalloids for fluid resuscitation in critically ill patients.

The study patients: There were more than 5000 patients screened for eligibility in 57 ICUs (France, Belgium, Canada and North Africa) from February 2003 to August 2012, and 2857 were included.

Eligibility: Adults who had not received fluids for resuscitation during their ICU stay and now required fluid resuscitation for acute hypovolaemia defined by a combination of:

Hypotension (systolic arterial pressure <90 mmHg, MAP <60, orthostatic hypotension, Delta pulse pressure of ≥13%.

Evidence for low filling pressures and low cardiac index assessed invasively/non-invasively

Signs of tissue hypoperfusion/hypoxia (≥2 of: GCS <12, mottled skin, UO <25 ml/h, CRT ≥3 s, arterial lactate >2, BUN >56 mg/dL, fractional excretion of sodium <1%

Exclusion criteria: received fluid therapy in ICU, anaesthesia-related hypotension, advanced chronic liver disease, chronic renal failure, acute anaphylactic reaction, inherited coagulation disorders, had DNAR pregnant, burned >20% of body surface area, allergy to any study drug, refused consent, dehydrated, brain death/organ donor.

Table 1.

The evidence (Control group = crystalloids; Experimental group = colloids).

Outcome	Time to outcome	CER	EER	RRR	ARR	NNT
Death	28 days	0.270	0.254	6%	1.6	ns
	95% Confidence Intervals			−6% to 18%	1.6% to 4.8%
Death	90 days	0.342	0.307	10%	3.5%	29
	95% Confidence Intervals			0% to 20%	0.1% to 6.9%	14 to 1447
Death	ICU discharge	0.281	0.251	11%	3%	ns
	95% Confidence Intervals			−1% to 22%	0.2% to 6.2%

Notes: CER = control event rate, EER = experimental event rate, RRR = relative risk reduction, ARR = absolute risk reduction, NNT = number needed to treat.

EBM questions:

Do the methods allow accurate testing of the hypothesis? Yes. This was a large, international, multi-centred, computer-randomised, controlled trial. The only potential problems were that the clinicians were not blinded; however, the outcome assessors were which minimised the risk of assessment bias. Secondly, the extended recruitment period of nine years; however, this was not shown to modify the direction and size of estimates.

Do the statistical tests correctly test the results to allow differentiation of statistically significant results? Yes.

Are the conclusions valid in light of the results? Yes. The authors conclude that in ICU patients with hypovolaemia, the use of colloids compared with crystalloids did not result in a significant difference in 28–day mortality, and the results support this. However, within secondary outcomes, 90–day mortality was significantly higher with the use of crystalloids (p = 0.03). There were also significantly more days alive within 7 and 28 days without mechanical ventilation (p = .01/.01) and without vasopressor therapy (p = 0.04/0.03) in the colloid group. The authors offer three explanations for these results and feel further exploratory studies are required before conclusions are drawn. It is also worth noting that the trial was only powered to detect a 5% difference in risk of death at 28 days and therefore these secondary outcomes may not be statistically or clinically significant.

Did results get omitted and why? No. Of the 2857 patients randomised, none were lost to follow up or discontinued intervention.

Did the authors suggest areas of future research? Yes. They suggested a further study looking at the 90-day mortality in colloids compared with crystalloids.

Did they make recommendations based on the results and were they appropriate? Yes. That resuscitation with colloids offers no benefit over crystalloids.

Is the study relevant to my clinical practice? Yes. This study did not demonstrate beneficial effects of using colloids over crystalloids in hypovolaemic patients and therefore should not be used as a standard routine practice.

What level of evidence does this study represent? 1B (CEBM). A good quality RCT. Due to the broad variety of drugs in each class used and unpredictable total amount of fluid to be administered during ICU stay, it was unfeasible to blind clinicians because study treatments had to be available immediately for emergency resuscitation and there was no practical way to stock sites with adequate supplies of masked fluid solutions. However, the blinded outcome assessor and the robustness of the primary outcome minimised the risk of assessment bias.

What grade of recommendation can I make on this result alone? A

What grade of recommendation can I make when this study is considered along with other available evidence? A. There is no mortality effect and no evidence for choosing to administer colloids over crystalloids in hypovolaemic ICU patients and this is in concordance with other randomised controlled studies.

Should I change my practice because of these results? No. Considering the outcome of the study, there is no current evidence to change practice. Given that colloids are considerably more expensive than crystalloids, it is hard to see how their continued use can be justified over crystalloids. However, if a further study powered for 90-day mortality showed a statistically significant difference then one would strongly consider a change of practice if supported by other randomised controlled trials.

Should I audit my current practice because of these results? No. Substantial numbers were required to demonstrate a statistical difference in the primary outcomes and local audit is therefore unlikely to be worthwhile.