Why do we give fluids in resuscitation? Part 2: Oncotic Pressure

While I still have not answered why we give fluids in resuscitation, the end goal is to have fluid in the intravascular space. Do the ends justify the means? I will discuss that in part 5. For now, we will continue to look at the physiology of intravascular volume.

Let us first get rid of two dogmas that are causing a lot of problems in resuscitation, and they both deal with albumin.

1. Albumin does not work or worsens outcomes.

2. Albumin only lasts for a couple of hours

If you have said either of these things, please keep an open mind, and hopefully, you will realize that these two statements have led to over-resuscitation and worse patient outcomes.

Before we look at these dogmas, let's look at the purpose of albumin. Albumin makes up 75% of the oncotic pressure in the intravascular space. The oncotic pressure makes up a small percentage (0.5%) of the pressure in the intravascular space compared to the osmoles of electrolytes. That does not mean that it is not important. Looking again at the Starling forces equation shows that oncotic pressure plays a large part in driving fluid into the intravascular space. The oncotic difference between the intravascular and interstitial space is what creates this force. By either increasing intravascular oncotic pressure or decreasing interstitial oncotic pressure will shift the equation and fluid will want to remain in the intravascular space.

Starling equation:

Jv = LpS [(Pc - Pi] - σ(πc-πi)]

Jv = Rate of fluid entering or leaving the intravascular space

Lp = Hydraulic permeability coefficient

S = Surface area of membrane

Pc = Capillary hydrostatic pressure

Pi = Interstitial hydrostatic pressure (small)

πc = Capillary oncotic pressure

πi = Interstitial oncotic pressure (negligible)

σ = reflection coefficient for protein

Albumin works better clinically than anticipated, this phenomenon is due to the Gibbs-Donnan Effect.

Gibbs-Donnan Effect:

Albumin is negatively charged. This causes a change in the osmolarity gradient between the interstitial and intravascular space. Cations are more inclined to stay in the intravascular space (ratio 100:95) and anions in the interstitial space (ratio 100:105) which is shown in Table 1.

Table 1: Gibbs-Donnan Effect on Osmolarity

A normal plasma protein level of 0.9mmol/L should generate an oncotic pressure of 17.4mmHg (1mmol/L = 19.3mmHg), but actually generates an oncotic pressure of 25mmHg. Due to the Gibbs-Donnan effect attracting cations to its negative charge it adds 0.4mmol/L to the overall concentration, creating 1.3mmol/L, which accounts for the oncotic pressure increase from 17.4mmHg to 25mmHg. The oncotic pressure will pull water into the intravascular space and the Gibbs-Donnan effect increases the oncotic pressure and total water being pulled in. This is why a normal albumin level is so important to the body’s fluid equilibrium (See Figure 1). When the protein level drops, which can be seen in critical illness, dilutional resuscitation, and malnutrition, the oncotic pressure and Gibbs-Donnan effect are lost. This leads to third-spacing into the interstitial space with intravascular depletion. Physiologically, once the oncotic pressure drops below 11 mmHg the benefits are lost and spontaneous third-spacing occurs. Clinically, this loss of oncotic pressure benefit is around 2g/dL.

Figure 1: The Gibbs-Donnan Effect on Intravascular Fluid

The figure above shows an increased intravascular volume compared to the interstitial space. This is done just to show the pull the proteins have in getting fluid into the intravascular space and improving volume.

Addressing Dogmas:

1. Albumin does not work or worsens outcomes.

The most quoted study is the SAFE trial of 2004. This trial has been used for almost 2 decades as an excuse to not give colloids to patients. One complaint I have with this article is that they used 4% albumin.

4% albumin has an osmolality of 260 mOsm/kg. This is converted to mOsm/L under the assumption that there are 70g protein and 930g plasma water per liter. Dividing 260 by 0.93 converts 4% albumin to 280 mOsm/L.

0.9% normal saline has an osmolarity of 308 mOsm/L with 154meq Na and 154meq Cl.

280 mOsm/L was compared to 308 mOsm/L, although they were both defined as isotonic.

The normal osmolality is 275-295 mOsm/kg which converts to an osmolarity around 300-312 mOsm/L and isotonic solutions should fall within these values. So, comparing a more hypotonic solution against an isotonic solution for resuscitation and expecting a fair comparison is interesting.

The result was that there was no difference but when looking at a post hoc analysis for TBI, called the SAFE-TBI study, it was found that 4% albumin had worse outcomes. For me, this comes as no surprise since it is more hypotonic, and we know that hypotonic solutions should not be used. That is why I was surprised that 4% albumin was used. The SAFE trial did trend towards improved outcomes in septic shock with 4% albumin but was not powered sufficiently.

Also, hetastarch got thrown in with albumin in the past, and then it was found to worsen renal function. I think this played a part in a bad name for albumin as well.

2. Albumin only lasts for a couple of hours

People love to use this argument. Why give it if it is only going to last in the intravascular space for a couple of hours. It is true that the half-life of albumin is estimated at 12-16hrs. It will remain in the intravascular space much longer than most people state when making this argument. If this half-life is keeping providers from using albumin, then isotonic crystalloids should not be used either. In the same study, normal saline had a half-life of 110 minutes and lactated ringers had a ½ life of 50 minutes. Albumin had a significantly longer half-life compared to crystalloid and so this argument should stop being perpetuated.

Time to look at albumin differently:

Let’s look at albumin as an oncotic medication and not a resuscitative fluid. Its job is to help keep fluid in the intravascular space. It is not a replacement for plasma water, it just makes it more efficient. As we can see from the Starling equation, if we ensure an appropriate oncotic pressure the fluid will stay in the intravascular space longer. Physiologically this would imply that patients would have shorter times in shock, less time on vasopressors, and less overall crystalloid needed in resuscitation.

Are there trials that show clinically what you would expect physiologically?

The 2014 ALBIOS trial looked at using 20% albumin to keep the serum albumin level >3g/dL with crystalloid against crystalloid alone. It did not show a mortality benefit, but the albumin group reached goal MAP faster, which means less time in shock, lower lactate levels in the first 24 hours, and lower norepinephrine doses. The albumin group also had a lower net fluid balance compared to the crystalloid-only group.

The 2018 SWIPE trial look at 4-5% albumin vs 20% albumin for resuscitation. The results of the SWIPE trial showed that there was significantly less fluid needed with the 20% albumin group versus the 4-5% albumin group. There was also a trend towards improved mortality with the 20% albumin group with more patients being discharged alive from the ICU in the 20% group but this significance was lost at discharge out of the hospital alive.

Both trials show that albumin can be used. It should not be used as a resuscitative fluid but should be used as an adjunct to appropriately use fluids. In short, to keep fluids in the intravascular space you need to optimize oncotic pressure without increasing hydrostatic pressure.

Summary:

If you want to be a resuscitationist and an expert on fluid resuscitation, understanding the physiology of fluid is a must. When I am resuscitating patients, I make sure the albumin level is >2.5g/dL. This is to ensure it is above the 11mmHg necessary to avoid third-spacing. I will go to a target of 3g/dL occasionally, for example, on ECMO patients to reduce chatter. The reason I do not normalize the albumin level is due to cost and the risk of increasing hydrostatic pressure and worsening pulmonary edema. Albumin is expensive, but I believe it is a necessary medication for this critically ill patients.

References:

1. Nguyen MK, Kurtz I. Determinants of plasma water sodium concentration as reflected in the Edelman equation: role of osmotic and Gibbs-Donnan equilibrium. American Journal of Physiology-Renal Physiology. 2004;286(5):F828-F37.

2. Yartsef A 2015;Pageshttps://derangedphysiology.com/main/cicm-primary-exam/required-reading/cellular-physiology/Chapter%20121/gibbs-donnan-effect on 12/1/20.

3. Yartsef A. 2020; Pages https://derangedphysiology.com/main/cicm-primary-exam/required-reading/body-fluids-and-electrolytes/manipulation-fluids-and-electrolytes/Chapter%20013/osmotic-pressure-and-oncotic-pressure on 7/22/21.

4. Hahn RG, Lyons G. The half-life of infusion fluids: An educational review. Eur J Anaesthesiol. 2016;33(7):475-482. doi:10.1097/EJA.0000000000000436

5. 5 - Disturbances of Free Water, Electrolytes, Acid-Base Balance, and Oncotic Pressure, Editor(s): Peter D. Constable, Kenneth W. Hinchcliff, Stanley H. Done, Walter Grünberg, Veterinary Medicine (Eleventh Edition), W.B. Saunders, 2017, Pages 113-152, ISBN 9780702052460, https://doi.org/10.1016/B978-0-7020-5246-0.00005-X.

6. Finfer S, Bellomo R, Boyce N, et al. A comparison of albumin and saline for fluid resuscitation in the intensive care unit. N Engl J Med. 2004;350(22):2247-2256. doi:10.1056/NEJMoa040232

7. Caironi P, Tognoni G, Masson S, et al. Albumin replacement in patients with severe sepsis or septic shock. N Engl J Med. 2014;370(15):1412-1421. doi:10.1056/NEJMoa1305727

8. Mårtensson J, Bihari S, Bannard-Smith J, et al. Small volume resuscitation with 20% albumin in intensive care: physiological effects : The SWIPE randomised clinical trial. Intensive Care Med. 2018;44(11):1797-1806. doi:10.1007/s00134-018-5253-2