Prescribing albumin is just one of those things that causes controversy in the ICU. Well, maybe more so in my mind. I was first introduced to the concept of albumin during my per diem shifts in the CTICU at Cornell — the cardiac surgery fellow would always say “just give him (or her) a SPA.” After further inquiry and “research” I learned that a “SPA” was the acronym for “salt poor albumin” or 25% albumin (vs 5% albumin). Anyways, that’s how I learned about albumin.
Albumin is a colloid (a type of fluid or substance that contain macromolecules or large atoms). So albumin is basically plasma rich with protein. The idea behind prescribing albumin is to increase the patient’s oncotic pressure intravascularly in order to draw back volume from the third space and also to retain volume. It’s a neat idea that in theory makes a whole lot of sense. However, since medicine is an evidence-based profession (or should be) we need to look at the data. I’ll start off by saying that I do not prescribe albumin a lot and prescribe it a lot less than my colleagues. When I do prescribe it, the patient is usually volume up (total body fluid) and hypotensive with cirrhosis or ARDS with pulmonary edema/pleural fluid, or “chattering” the ECMO circuit (a sign that they might be hypovolemic), or a surgeon who directs me to do so. A wise cardiac surgeon once told me that yes, there are no studies that really support colloids over crystalloids for resuscitation post-cardiac surgery, but it works faster (and then he shrugged; love that guy). I will rarely give it to a patient who is hypoalbuminemic (in my mind less than 3; take this as you will — its not a real definition). This is my current practice but let’s get into it and do a semi-deep dive.
First I need to say albumin is pretty expensive (it’s derived from human plasma). For some reason I can’t see the pricing in my hospital because it’s found in the EPIC order set from a quick google search looks like 5% is around $40 and 25% is around $80.
Second, we need to talk about the SAFE study (Saline vs Albumin Fluid Evaluation), one of those “pivotal” trials in The New England Journal of Medicine that compared using crystalloid and colloids in ICU resuscitation. This study titled “A Comparison of Albumin and Saline for Fluid Resuscitation in the Intensive Care Unit” was released in May 2004 is pretty old in medicine standards but applicable nonetheless. It is a multi-centered, randomized, and double-blinded (ok good since Fauci said so) trial that took place in Australia and New Zealand. The study randomized a whopping 6997 ICU patients (3497 assigned to receive 4% albumin and 3500 assigned to receive saline) with a primary endpoint of all-cause mortality in 28 days. It’s important to note that excluded from the study were post-cardiac surgery, after liver transplantation, and burn patients. Those evaluated had trauma, severe sepsis or ARDS. The investigators assessed the SOFA scores, HR, CVP, MAP, study fluid administered, use of mechanical ventilation, use of RRT, and ICU length of stay.
- NS group received more fluid
- There were no differences in: MAP, mean length of stay in ICU and hospital, # of days on mechanical vent, RRT, or 28-day mortality
- Trauma patients who received albumin had higher mortality (unclear of the association between fluid vs TBI) but interpret with caution
- Severe sepsis patients who received albumin had lesser mortality (30.7% mortality with albumin vs 35.3% mortality with NS)
- Within the ARDS subgroup, 39.3% died who received albumin and 42.4% who died received NS
They concluded that the “study provides evidence that albumin and saline should be considered clinically equivalent treatments for intravascular volume resuscitation in a heterogeneous population of patients in the ICU. Whether either albumin or saline confers benefit in more highly selected populations of critically ill patients requires further study.”
I agree, subgroups need to be evaluated further.
The ABLIOS Study Investigators is the Italian counterpart who have a study published in 2014 in the NEJM titled “Albumin Replacement in Patients with Severe Sepsis or Septic Shock.” In this one they used 20% albumin + crystalloid vs crystalloid alone in septic shock patients with an all-cause mortality within 28-days. 1818 patients with severe sepsis were enrolled and randomly assigned the fluid. After 28 days, 31.8% in the albumin group died and 32% in the crystalloid group died. No significant difference in outcome there.
- serum albumin was higher in albumin group
- no difference in total daily fluids in both groups
- lower CV SOFA score in albumin group
- higher coagulation SOFA score (lower platelets) in albumin group (probably due to expansion of intravascular compartments and dilution)
- higher liver SOFA score (higher bili)in albumin group
- achieved targeted MAP faster in albumin group (within 6 hours)
- decreased use of vasopressors in albumin group
A lot of people know about the SAFE study, which is old and limited due to the subgroup analyses. There are so many more out there but I don’t have the patience at this moment to review and write about them. That being said, I learned that there are so many different trials with its own investigator conclusions being utilized. That’s okay, but it’s also really important to read the entire paper including its methods, outcomes, subgroups, raw data, etc. You may have your own interpretation of the study or may find that a certain subgroup benefited from the treatment rather than the average of the whole group. (Kind of like reading the bible and how they’re so many interpretations of it). This could be really helpful in your practice. Now my wise cardiac surgeon friend was right, albumin achieves faster MAPs. The Surviving Sepsis Campaign suggested the use of albumin in severe sepsis if outcomes were refractory to crystalloid. They may have a point. The mortality rate of my patient group might be better because of the lower CV-SOFA score (MAP >70, <70, dopa + DBA + epi, etc). They may have less time of vasopressors and inotropes, which in turn may help other things because of the heart rate and stuff. So, will I be more generous with albumin if they needed fluid resuscitation to begin with? Yes, because I’m reading the studies and interpreting them on my own rather than someone else telling me. I need to look up albumin and ARDS next.