Journal Logo

Myths in Emergency Medicine

Myths in Emergency Medicine

Colloids Can't Cut It Caring for the Critically Ill

Runde, Dan MD

Author Information
doi: 10.1097/01.EEM.0000481757.59258.7c

    The history of science in general and medicine in particular is littered with the carcasses of things that we were absolutely sure were true, right up until the moment where we found out that they we were completely wrong. We rarely bleed the bad humours out of our patients anymore either (the occasional errant central line placement notwithstanding), and unless you've got a sweet black market pipeline into Eli Lilly, you're likely not hitting up your septic patients with activated protein C (oh, Xigris, we hardly knew ye). Another recent addition to the ignominious list of “Things We Shouldn't Do Anymore Even Though They Sounded Like Good Ideas” is using colloid solutions for the initial resuscitation of sick patients.

    First, a little background: Colloid solutions, like albumin, dextran, gelatins, and starches (we'll get back to those in a minute) are composed of fluids with larger molecules that don't readily cross barriers between compartments. By giving them intravascularly, we increase the oncotic pressure, drawing fluid into circulation and expanding the plasma volume while administering less fluid through the IV. That's the primary proposed benefit to colloids: You get more plasma volume expansion per milliliter of colloid than you do from using crystalloids. You could almost say colloids deliver a bigger bang for your buck except they are hugely expensive, so the opposite is true.

    Colloids are not exactly new, but compared with normal saline and Lactated Ringer's solution (which has been around since the 1880s), they look like the hip, young arrivals on the resuscitation scene. People were so excited by colloids that in 1985, in fact, Anesthesia published a pro-colloid article, “The End of the Crystalloid Era?” that called the use of crystalloid fluids illogical, and proposed that only blood or colloid fluids be given to peri-operative patients. (1985;40[9]:860.)

    The pro-colloid freight train continued to roll along well into the 1990s and the carefree first decade of the new millennium, providing us with exciting new options with names like Hetastarch and Gelofusine. Intensivists would secretly smile to themselves as they indoctrinated each fresh crop of rotating medical students on the ins and outs of albumin infusion, and patients were likely thrilled to be receiving the latest and greatest in medical care.

    Then, at some point, someone decided to do a study to see how much good we were doing treating our critically ill patients with colloid fluids. It happened that they stumbled upon a surprising result: We weren't doing much good at all. A prime example of this is a Cochrane Review article updated in 2013.

    Colloids versus Crystalloids for Fluid Resuscitation in Critically Ill Patients

    Perel P, Roberts I, Ker K

    Cochrane Database Syst Rev

    2013 Feb 28;2:CD00056

    Perel and colleagues looked at how various types of colloid solutions compared with crystalloids when they gave them to critically ill patients. This included trauma and burn victims, patients undergoing surgery, and those with complications of sepsis. What did they find?

    Patients received no mortality benefit from albumin, one of the most commonly used colloids, in 24 trials with nearly 10,000 patients. Likewise, no mortality benefit was seen when gelatins were compared with crystalloids in 11 trials with 500 patients. And a comparison of dextran and crystalloids in nine trials and more than 800 patients? You guessed it: no mortality benefit. The same results were seen in nine trials of just under 2,000 patients comparing dextran and hypertonic saline with crystalloids.

    Anyone starting to see a theme here? Wait! What about giving albumin with hypertonic saline, you ask? Turns out there was a 100 percent increase in the rate of survival for patients who got that magical mixture! Unfortunately, it was only a single trial, with 14 total patients (one in the colloid group and two in the crystalloid group died), and the results weren't statistically significant (95% CI 0.06-4.33, p=0.53).

    More than 13,000 critically ill patients were studied, and colloid fluids don't seem to make a difference as far as who lives or dies. That's not great, but it doesn't seem like colloids are really hurting anyone, so comme ci, comme ça, right? Doctors should use whatever they are most comfortable with, and residents should just give whatever it is that their attending prefers. Not so fast. Multiple studies have found using colloids for sick patients can have some nasty consequences for the kidneys, such as the just-published meta-analysis in the British Journal of Surgery.

    Meta-analysis of Colloids versus Crystalloids in Critically Ill, Trauma and Surgical Patients

    Qureshi SH, Rizvi SI, et al.

    Br J Surg


    This analysis of 59 randomized control trials involving almost 17,000 patients again demonstrated no mortality benefit for colloid fluids. What they did find, however, was that colloid administration increased the risk of causing acute kidney injury severe enough that the patient had to be started on renal replacement therapy [OR 1.35, 95% CI 1.17-1.57]. On the bright side, they performed a subgroup analysis that suggested that colloids might only tank the kidneys of critically ill septic patients (as opposed to those with trauma or burns), so perhaps a “hooray” is in order?

    On top of this, colloids are dramatically more expensive than their crystalloid cousins. Prices vary from region to region and fluid to fluid, but colloids always cost more. One study from the early 1990s decided to break down the cost in an interesting way.

    Colloids versus Crystalloids in Fluid Resuscitation: An Analysis of Randomized Controlled Trials

    Bisonni RS, Holtgrave DR, et al.

    J Fam Pract


    The authors looked at the cost-effectiveness of each fluid per life saved. The results: The cost of saving a life using crystalloid solution was $45, and to save that same life using a fancy-shmancy colloid was $1,500 (technically $1,493.60 for the nitpickers out there). That is roughly 3,200 percent more.

    And saving the worst for last, consider hydroxyethyl starch, or Hetastarch, as it is more commonly known. Remember when I said colloids didn't have any impact on mortality? Well, I was lying. Hetastarch does have an impact on mortality, just not the kind that you or your patient want. Authors of a paper published in JAMA found a stunning result.

    Association of Hydroxyethyl Starch Administration with Mortality and Acute Kidney Injury in Critically Ill Patients Requiring Volume Resuscitation: A Systematic Review and Meta-analysis

    Zarychanski R, Abou-Setta AM, et al.

    JAMA 2013;309(7):678

    Hydroxyethyl starch was associated with increased mortality (and kidney injury) when given to critically ill patients [RR 1.07, 95% CI 1.00-1.14]. This prompted an editorial in the same issue highlighting that Joachim Boldt, MD, conducted nearly all of the studies showing benefit for Hetastarch. (JAMA 2013;309[7]:723.) Dr. Boldt, mind you, had a number of his studies retracted for ethical and scientific misconduct. They concluded that “the harms of hydroxyethyl starch most likely outweigh the benefits and suggest that these products should not be used for acute volume resuscitation of critically ill patients.”

    Here's the reader's digest version of everything I just said:

    • Colloids don't save lives when compared with crystalloids.
    • They are wildly more expensive.
    • They can damage your kidneys (but maybe only if you're septic).
    • Hetastarch will kill you (and also doesn't recycle).

    Only eight days after that JAMA editorial was published, Perel released the updated Cochrane Review cited above that included a new conclusion suggesting that hydroxyethyl starch may increase patient mortality. That study offered this conclusion: “As colloids are not associated with an improvement in survival and are considerably more expensive than crystalloids, it is hard to see how their continued use in clinical practice can be justified.”

    Share this article on Twitter and Facebook.

    Access the links in EMN by reading this on our website or in our free iPad app, both available at

    Comments? Write to us at

    Copyright © 2016 Wolters Kluwer Health, Inc. All rights reserved.