Critically ill patients with hypovolemic shock had the same rate of survival when resuscitated with crystalloid as with colloid solutions, a large randomized controlled trial has found.
In a randomized, international multicenter trial lasting 9 years and enrolling nearly 3,000 patients, 28-day mortality did not differ significantly between those treated with colloid solutions, such as gelatins, hydroxyethyl starches, or albumin, and those treated with crystalloid solutions, such as salines. Mortality at 90 days was found to be somewhat better for colloids than crystalloids, though investigators cautioned that the 90-day finding would require further study.
The question of whether to resuscitate patients with hypovolemic shock with colloids or crystalloids has long been controversial, and many large randomized trials have attempted, over the past decade, to define any differences in mortality and other outcomes between the two classes of fluid therapies.
The study, published in JAMA (doi:10.1001/jama2013.280502), supports previous studies in that no significant mortality differences were found at 28 days. Unlike some earlier studies, which showed adverse renal outcomes associated with colloid use (JAMA 2013;309:678-88), this study did not find a difference in renal outcomes.
For their research, Dr. Djillali Annane, of the University of Versailles, in Garches, France, and colleagues at 57 intensive care units in France, Belgium, Tunisia, and Canada, recruited 2,857 patients with hypovolemic shock over a 9-year period ending in 2012. Of these 1,414 were randomized to colloids and 1,443 to crystalloids. At 28 days the colloids group had 359 deaths (25.4%) and the crystalloids group, 390 (27%). At 90 days the colloids group had 434 deaths (30.7%) and the crystalloids group, 493 (34.2%).
Dr. Annane and colleagues called the 90-day mortality findings surprising. However, "given the null findings at 28 days and the fact that the confidence limit approaches 1, the finding of improved mortality with colloids should be considered exploratory until replicated in a study focusing on this outcome," the investigators wrote.
Colloid use did show other improved outcomes compared with crystalloid use. Patients on colloids had significantly more days alive without mechanical ventilation within 7 days, and more days without vasopressor therapy within 7 days.
Renal outcomes were similar, with 156 patients (11%) in the colloids group requiring renal replacement therapy compared with 181 patients (12.5%) in the crystalloids group.
The fact that the study did not find a higher rate of renal effects associated with colloids, the investigators said, could be due to the fact that the trial excluded patients with severe chronic renal failure, that total dose of starches never exceeded doses recommended by regulatory agencies in the study countries, or that the majority of the crystalloids patients received chloride-rich normal saline, which might increase the risk of kidney injuries compared with a chloride-restricted fluid therapy.
Dr. Annane and colleagues noted the study’s long recruitment period and open-label design as weaknesses.
In an editorial accompanying Dr. Annane and colleagues’ study, Dr. Christopher W. Seymour and Dr. Derek C. Angus, of the University of Pittsburgh, questioned the two-fluid-classes design of the trial, which, they argued, might not have been ideal to settle the question of ideal fluid therapies in hypovolemic shock. Rather, they wrote, "there are a number of complexities, including the type of shock requiring resuscitation, the resuscitation targets, and the use of adjunctive vasoactive therapies."
In addition, they wrote, "any given fluid choice could have both beneficial and harmful effects, with trade-offs that vary depending on the other complexities listed above. Thus, perhaps the most important message from the latest round of trials is that simply performing larger two-group trials with greater rigor will not bring the field to consensus. Instead, alternative study designs should be considered, perhaps with multiple study interventions and use of adaptive trial design methods."
Two of the 22 coinvestigators reported financial ties to industry. Dr. Seymour disclosed receiving institutional grants from the American Heart Association, the Society of Critical Care Medicine, and the MedicOne Foundation. The study was funded by the French Ministry of Health.