Synthetic colloids have gotten a beating lately. First, the Cochrane group released their meta analysis on evidence supporting colloid use. They found none. Then HES colloids got hit hard by the 6S and CHEST studies. Are there any good indications left for synthetic colloid use? Right now, I’m not sure I can think of any.
The Cochrane review group looked at colloids vs crystalloids and found
“no evidence from RCTs that resuscitation with colloids reduces the risk of death, compared to resuscitation with crystalloids, in patients with trauma, burns or following surgery. As colloids are not associated with an improvement in survival, and as they are more expensive than crystalloids, it is hard to see how their continued use in these patients can be justified outside the context of RCTs.”
Pretty strong words from a Cochrane review.
We often tend to think of fluids as just different types of harmless water. Fluids aren’t medication. Lately there’s been more focus on different types of fluid and how they affect the patients. From the more physiological constitutes of Ringer’s Lactate over “Normal” Saline, to the effects of colloids vs. crystalloids. Different, but not dangerous, most of us thought. So even though colloids are more expensive and the Cochrane review didn’t show any advantage for colloids, surely they couldn’t be harmful?
6S and CHEST
At the same time as the Cochrane review came out, the 6S – Scandinavian Starch for Severe Sepsis/Septic Shock - trial was published. It was a well conducted trial with around 800 severly septic patients randomised to one arm with a moderate amount of colloid (and further fluid needs in crystalloids) and the other arm receiving just crystalloids. I think the numbers surprised the authors as well. While the increased complications from bleeding and increased number of patients needing renal replacement therapy was suspected, the increased mortality in the colloid arm was certainly an unpleasant surprise.
In the 6S trial, 51% died in the colloid arm vs 43% in the crystalloid arm (figure A). That’s a big difference – a 20% increase in mortality in the colloid group. Also note it takes quite long before the colloid mortality increases – over 30 days. The mortality numbers are pretty high in both arms, reflecting the sepsis severity in the study population.
A few months later, the Australians and Kiwis published their huge study on colloids: CHEST – The Crystalloid versus Hydroxyethyl Starch Trial - on 7000 ICU patients. This study also showed a significant increase in patients needing renal replacement therapy in the colloid arm, but could not show any increased mortality as found in the 6S study.
Both the 6S study and CHEST is available in full text for free, links below.
So what does this mean?
It means colloids don’t offer any advantange for the patient – although it is tempting for the doctor to use it as the patient gets a quick rise in BP and less overall edema in the short run. But not only is there no advantage with colloids, the 6S and CHEST study show sigificant DISadvantages for the patients receiving HES colloids.
One reason the CHEST study didn’t show any mortality difference might be due to the fact that they studied a mixed ICU population, while the 6S trial only included really sick, severely septic patients. The CHEST group also mentions that their patients were not as severly ill as the 6S population. This was also reflected in the 6S study, where sepsis patients in shock were at greater risk for harm from HES than patients not in shock (see figure B). In the non-shocked group, the results look inconclusive. So the really sick patients might be the ones that take the greatest hit from synthetic colloids.
This difference is probably also the reason we haven’t been able to show any mortality difference in the general population in the operating theatre. Because there is none. These, mostly healthy, patients can probably take colloids onboard without any real disadvantage. But then again, these are healthy patients and don’t really need colloids. So there’s no real rationale for using it – except for the satisfaction you get as a doctor in seeing a quickly responding blood pressure.
Even though some oncotic pressure is needed to keep fluid intravascular, it doesn’t seem like plasma expanders help much in the long run. Maybe it is because the low oncotic pressure is due to leaking vessels – so your added colloid also ends up sieving out into the extracellular fluid. And in the cases where intravascular fluid is lost not by leaking, but by bleeding – the patient needs blood, not just colloid substitution. So looking for a good way to apply colloids might have a flawed basis to it?
This downfall of HES products has led to a renewed interest in the “natural colloids” albumin and plasma as resuscitation fluids. They don’t seem to harm patients like HES does. But Albumin and plasma still haven’t show any survival advantage. The SAFE study couldn’t find any albumin advantage, and the Cochrane review lumps albumin and plasma together with other colloids, also say they have no proven advantage over crystalloids. Some researchers are looking into gelatin based colloid fluids again, as they don’t have the starch polymers that stay in the body long after the fluid is gone. Or there might be subgroups of patients that could benefit from albumin or plasma? The Canadians are looking into it, and the PRECISE trial is still under way. Time will tell.
What to do?
Throw your synthetic colloids out of the ICU, and possibly out of the hospital.
Or maybe that’s too soon? Even if you think so, you should at least keep HES away from your severly septic patients, and most possibly from other critically ill patients as well. And think twice before giving them to anyone else. So, every time you feel the itch to give a quick fix of colloids – think again.
UPDATE OCT 11th 2013:
PRAC – the European Pharmacovigilance Risk Assessment Committee, under the European Medicines Agency (EMA) has issued a statement that HES solutions should no longer be used in septic patients, burns patients or critically ill patients.