Colloid v. Crystalloid ?

Selecting  Your Perfusate

Authored By:  Gerard J Myers

Did you ever consider why we use one type of perfusate over the other?  Thousands of cases are done with crystalloids only and thousands are done with colloid/crystalloid primes. Each program has equal results and I would imagine each program would argue that there outcome stats are as good as the next program.

How many perfusionists have heard surgeons or anesthetists tell us that we have to reduce our priming volumes because there were far too many patients getting transfused in your cardiac surgery programs.  Some would argue that the prime volume was the ‘main’ contributing factor to increased transfusions in their patients.  Even manufacturing focused entire marketing campaigns based on the premise that if you use their devices that require 100 or 200 mls less than a competitor, you will improve adult patient outcomes and reduce RBC transfusions.

As we all know, nothing could be further from the truth.  Prime volume is only ‘one’ contributing factor to RBC transfusion statistics during CPB, others include the anesthetist, the surgeon, ICU care givers, patient demographics, length of surgery, choice of technology for Hgb/Hct measurement, use of Hemoconcentrators, use of RAP and potentially the use of cerebral oximetry.  Therefore, if a program puts its focus on prime volume only, it is highly likely that transfusion stats will not be improved.

Perhaps the most overlooked and under taught contributions to RBC transfusion is the type of priming and resuscitation fluids used during bypass.  In other words the focus should be on hemodilution volume, not prime volume as most of our medical colleagues seem to be under the impression.

Yes my friends, it comes down to colloid versus crystalloid use during CPB.  For decades most of us have been told that our patients are better off when we use colloid based primes and colloid additions during the perioperative period, yet to date, there has not been one single study to show that patient outcomes improve when colloids are used instead of crystalloids only.

Here are only a few of the largest studies and meta analysis of retrospective randomized trials.

  • Finfer S et al; N Engl J Med. 2004 May 27; 350(22):2247-56
  • Perel P and Roberts I: Cochrane Database Syst Rev 2007 Oct 17; (4):CD000567
  • Perel P, Roberts I &  Pearson M: Cochrane Database of Systematic Reviews 2009 (3):CD000567

Yet over the past decade, the uses of colloids (synthetic and albumin) have increased during cardiac surgery.

The question needs to be asked … why?  

In spite of the fear mongering about crystalloids only causing our routine CPB patients to blow up into balloon like monsters or lead to increased transfusions during the post operative period … the literature states the opposite.  Yes, crystalloid only patients do retain more fluid, but for the vast majority of patients it does not affect post op outcomes, and actually will lead to decreased transfusions (see comprehensive reviews and studies above) and increased urine output.  The positive effects of synthetic colloids (Voluven) were debunked and placed in limbo with the recent scandal and removal of Dr. Bolt’s extensive literature on this subject.

Even the fear of increasing the incidence of oxygenator failure (high pressure excursions) when albumin is removed from the prime is unfounded since the introduction of synthetic coated oxygenators (personal experience and our evidence based literature).  Over the past 7 years (7000-8000 cases) using crystalloid only primes with synthetic coated oxygenators/circuits, we have not had any incidence of HPE.

Finally, consider the following analogy about prime volume versus hemodilution volume.

You are a center that has their prime volume down to 800 mls and the surgeon is pleased.

If that 800 mls is crystalloid only, when you go on bypass the patient’s intravascular volume is hemodiluted by approximately 250 mls because evidence indicates that 50-70% of infused crystalloid goes into the extravascular space.

On the other hand, if that 800 ml prime contains albumin (100 mls/ 5%), the patient’s intravascular hemodilutional volume will increase to 1150 ml because the oncotic effect of albumin will draw approximately an additional 350 mls of fluid from the extravascular space (ratio 3.5:1).

If that same 800 ml prime contains 500 mls of Voluven (or Hetastarch), the patient hemodilutional volume will be 1300 mls because the primes oncotic effect will draw approximately an additional 500 mls of fluid from the extravascular space (ratio 1:1).

But if the 800 ml prime contains 500 mls of Pentaspan, the intravascular hemodilutional volume will now be 1550 mls because the prime’s oncotic effect will draw approximately an additional 750 mls of fluid from the extravascular space (ratio 1.5:1).

Also consider this …

If your patient receives 500 mls of Voluven during induction, 500 mls in the prime, 500 mls during the surgery and another 500 mls during the first 24 hours post op … that patient has been hemodiluted by 4 liters or more within the first 24 hours of surgery … only from the colloid additions!!
Excessive hemodilution is what leads to RBC transfusions, not prime volumes.

To summarize, reducing the incidence of RBC transfusions is by no means a simple matter of using an oxygenator that has 100 mls less prime volume than another.

So why are we experiencing large growth in synthetic colloid use and risking the chance of prion transmissions, analyphalactic reactions, increases in pruritis post infusion  and massively increases in costs if there is not advantages in improved outcomes over crystalloids?

Gerard J Myers

Halifax, Nova Scotia

I have had several ‘discussions’ at divisional meetings (surgery, anaesthesia and perfusion) in my hospital over the years.

Anaesthesia insists colloids do not affect transfusion stats, surgery is not sure either way and perfusionists rarely comment one way or the other. There is no doubt that there may be some cases where colloid will benefit patient outcomes, but it appears to me that clinicians are trending toward using synthetic colloids during cardiac surgery like they were … well … crystalloids.

As professionals we all tend to justify or support our current fluid management practice with theories, speculation or opinions on the topic, but the bottom line is two Cochrane Data Base reviews on prospective randomized control trials have determined that there is no outcome advantages to using colloids during surgery or bypass.
As I said, if there is no outcome advantage and colloids come with some adverse reactions, then why are we so adamant on using them in our prime ????

What surprises me is that this sort of physiological look into fluids should be done ‘before’ students enter clinical. Perhaps it is ???  I’m not sure anymore.
Even the types of crystalloids we use can have some negative impacts, but most consider crystalloid as just a crystalloid … and that’s another matter.



BUMINATE 25%, Albumin (Human), 25% Solution is a sterile, nonpyrogenic preparation of albumin in a single dosage form for intravenous administration. Each 100 mL contains 25 g of albumin and is prepared from human venous plasma using the Cohn cold ethanol fractionation process. Source material for fractionation may be obtained from another U.S. licensed manufacturer. It has been adjusted to physiological pH with sodium bicarbonate and/or sodium hydroxide and stabilized with sodium acetyltryptophanate and sodium caprylate. The sodium content is 145 ± 15 mEq/L. This solution contains no preservative and none of the coagulation factors found in fresh whole blood or plasma. BUMINATE 25%, Albumin (Human), 25% Solution is a transparent or slightly opalescent solution which may have a greenish tint or may vary from a pale straw to an amber color.

The likelihood of the presence of viable hepatitis viruses has been minimized by heating the product for 10 hours at 60°C. This procedure has been shown to be an effective method of inactivating hepatitis virus in albumin solutions even when those solutions were prepared from plasma known to be infective.1-3


1. Gellis SS, Neefe JR, Stokes J Jr, et al: Chemical, clinical and immunological studies on the products of human plasma fractionation. XXXVI. Inactivation of the virus of homologous serum hepatitis in solutions of normal human serum albumin by means of heat. J Clin Invest 27:239-244, 1948

2. Gerety RJ, Aronson DL: Plasma derivatives and viral hepatitis. Transfusion 22:347-351, 1982

3. Murray R, Diefenbach WCL, Geller H, et al: Problem of reducing danger of serum hepatitis from blood and blood products. NY State J Med 55:1145-1150, 1955