CRRT Management Question

[Ronin786]

Common case scenario, patient comes in with some form of shock. Cardiogenic vs septic, gets "resuscitated" and now in acute renal failure, volume overloaded and requiring CRRT. Patient is anuric.

Simple and stupid question potentially but I just can't explain it to myself. Why would a patient become hypotensive with a desired fluid loss of zero. If they're getting 100cc of volume via drips an hour and having 100cc of volume removed an hour, how does that lead to intravascular fluid shifts and hypovolemia? Does machine configuration play into it, i.e. CVVH vs CVVHD vs CVVHDF? Also does CRRT on ECMO make a difference? I just can't wrap my head around how the removal of 1.5cc of volume per minute(!) can result in someone becoming hypovolemic when they're getting that volume replaced instantaneously through their IVs. Hoping someone smarter than me can explain it.

---

Members don't see this ad.

 

[DNA888]

Hi! If net intake is 100 cc/hour and UF on CRRT is -100 cc/hour then net volume loss from the body (excluding insensible losses of course) is exactly zero. This really isn't contentious in any way; just simple mathematics (100/hour - 100/hour = 0/hour). CRRT configuration is not relevant here.

There is a lot of confusion surrounding the hemodynamic impact of RRT. The root cause of RRT-associated hemodynamic derangements is the 'net' volume removed per hour. For example, in a patient undergoing HD with no concurrent infusions, UF is typically 1 L/hour. So 1L of fluid is removed from the intravascular compartment in 1 hour. This fluid is replaced by fluid from the interstitial compartment at the capillary level. However, the rate of refill (aka 'plasma refill rate') may not be as high as the UF rate. Say the plasma refill rate is only 500 cc/hour. This means that after 1 hour of HD, the patient's intravascular volume has been reduced by 500 cc (1L - 500 cc). You can now see how this may cause hemodynamic instability in some. This is the benefit of CRRT. Even if the UF is as high as 200 cc/hour, it is still (typically) less than the plasma refill rate so the patient does not develop periods of reduced intravascular volume.

That being said, the blood flow rate (in HD or CRRT) has absolutely nothing to do with hemodynamics. Typical value for this is ~400 cc/min in both HD and CRRT. The fact is that 400 cc/min of blood is also being returned at the same time through the return cannula so there is no net deficit at any point of time. So it doesn't matter if it is 400 cc/min or 4000 cc/min

Finally, I have seen people 'turning off' an already running CRRT circuit in cases of severe hemodynamic instability (maxed on 4 pressors etc) as the patient is "unstable". There is no rationale for this (although, it may be risky to initiate CRRT in such a patient as the initial extracorporeal shift of blood volume may cause some trouble). If, for example, the UF from CRRT is 50 cc/hour and the patient is getting 50 cc/hour from infusions (pressors, propofol etc.) then the net volume loss is zero. So there is no theoretical mechanism by which CRRT can hurt hemodynamics (stopping CRRT in such a case it may certainly cause issues).

 

[Beeftenderloin]

Don’t forget about those precious insensible fluid losses in your cold clamped down patient (probably) on multiple pressers.

 

[Ronin786]

Finally, I have seen people 'turning off' an already running CRRT circuit in cases of severe hemodynamic instability (maxed on 4 pressors etc) as the patient is "unstable". There is no rationale for this (although, it may be risky to initiate CRRT in such a patient as the initial extracorporeal shift of blood volume may cause some trouble). If, for example, the UF from CRRT is 50 cc/hour and the patient is getting 50 cc/hour from infusions (pressors, propofol etc.) then the net volume loss is zero. So there is no theoretical mechanism by which CRRT can hurt hemodynamics (stopping CRRT in such a case it may certainly cause issues).

That's exactly my point. Yet I've been trying to do that in a number of patients in my new ICU and I've personally seen the hemodynamic effects and I'm trying to rationalize it. I think it's because of how their circuits are set up, but I can't figure it out.

 

[DNA888]

That's exactly my point. Yet I've been trying to do that in a number of patients in my new ICU and I've personally seen the hemodynamic effects and I'm trying to rationalize it. I think it's because of how their circuits are set up, but I can't figure it out.

Establishing cause and effect can be challenging. Like I mentioned, if you are initiating CRRT in a super unstable patient, there is a potential risk. We use Prismaflex at our shop and depending on the model, you may lose up to ~200 cc of blood in the CRRT circuit upon initiation - (https://econnect.baxter.com/assets/...therapies/PRISMAFLEX_System_M60_M100_M150.pdf).

That may cause issues if they are very preload dependent. You may potentially avoid trouble by giving volume at the same time as CRRT initiation. But I think in most cases this is not a major issue.

However, in a patient who has been steady on CRRT (with matching I and O) and develops new hemodynamic instability, I would look for other causes.

 

[Akarat]

That's exactly my point. Yet I've been trying to do that in a number of patients in my new ICU and I've personally seen the hemodynamic effects and I'm trying to rationalize it. I think it's because of how their circuits are set up, but I can't figure it out.

In the case of no UF (volume removal), one of the possible reasons for hemodynamic instability has to do with effective osmole removal. This risk is seen more so with intermittent hemodialysis and less so with CRRT. For instance, if the patient's serum sodium is higher than the dialysate or replacement fluid that is being used, then the fluid from the ECF compartment will shift into the ICF compartment. The reason why there's a higher risk for hypotension with intermittent hemodialysis is because it removes effective osmoles at a faster rate than CRRT. With intermittent hemodialysis, clearance is limited by blood flow rate which usually doesn't get much better after about 300+ mL/min (the slope of the curve plateaus). Dialysate flow rate is usually set at 600-800 mL/min. In contrast, with CRRT, the dialysate or replacement fluid is the limiting factor. The usual prescription for dialysate or replacement fluid is like what 25 mL/kg? I usually use 30 mL/kg. For a 70 kg man, if we use 30 mL/kg, then the dialysate or replacement fluid rate on CRRT is 2100 mL/hr. Note that intermittent hemodialysis flow rate is mL/min while CRRT is mL/hr. Therefore, the rate of 2100 mL/hr can be restated as 35 mL/min (compare to the 600-800 mL/min flow rate on intermittent hemodialysis). That's the reason why the dialysate or replacement fluid is the limiting factor in CRRT.

Have you ever noticed that patient's who get hypotensive during dialysis seem to tolerate it better when you're doing volume removal with no clearance (PUF for intermittent or SCUF for CRRT)? No effective osmoles are being removed, and only fluid is being removed. Again, this is assuming that the only difference between the clearance + UF scenario vs UF only scenario is whether or not clearance is being performed. You can use this knowledge of effective osmoles to your advantage. If I have to clean the blood and pull fluid from the blood, then I usually increase the Na bath to at least 140 mmol/L if not higher. The down side is that the patient can become hypernatremic which triggers the thirst sensation which then causes the patient to drink water (Nooo, I just removed fluid from you and now you're drinking fluid!).

 

[Ronin786]

I might have an idea of what the problem was, I don't think nurses have been accounting for CT output. The patient in mind had 70-100cc/hr of CT output.

 

[Ronin786]

In the case of no UF (volume removal), one of the possible reasons for hemodynamic instability has to do with effective osmole removal. This risk is seen more so with intermittent hemodialysis and less so with CRRT. For instance, if the patient's serum sodium is higher than the dialysate or replacement fluid that is being used, then the fluid from the ECF compartment will shift into the ICF compartment. The reason why there's a higher risk for hypotension with intermittent hemodialysis is because it removes effective osmoles at a faster rate than CRRT. With intermittent hemodialysis, clearance is limited by blood flow rate which usually doesn't get much better after about 300+ mL/min (the slope of the curve plateaus). Dialysate flow rate is usually set at 600-800 mL/min. In contrast, with CRRT, the dialysate or replacement fluid is the limiting factor. The usual prescription for dialysate or replacement fluid is like what 25 mL/kg? I usually use 30 mL/kg. For a 70 kg man, if we use 30 mL/kg, then the dialysate or replacement fluid rate on CRRT is 2100 mL/hr. Note that intermittent hemodialysis flow rate is mL/min while CRRT is mL/hr. Therefore, the rate of 2100 mL/hr can be restated as 35 mL/min (compare to the 600-800 mL/min flow rate on intermittent hemodialysis). That's the reason why the dialysate or replacement fluid is the limiting factor in CRRT.

Have you ever noticed that patient's who get hypotensive during dialysis seem to tolerate it better when you're doing volume removal with no clearance (PUF for intermittent or SCUF for CRRT)? No effective osmoles are being removed, and only fluid is being removed. Again, this is assuming that the only difference between the clearance + UF scenario vs UF only scenario is whether or not clearance is being performed. You can use this knowledge of effective osmoles to your advantage. If I have to clean the blood and pull fluid from the blood, then I usually increase the Na bath to at least 140 mmol/L if not higher. The down side is that the patient can become hypernatremic which triggers the thirst sensation which then causes the patient to drink water (Nooo, I just removed fluid from you and now you're drinking fluid!).

One of the people I most admired in fellowship was an ICU/Nephrologist. Smarter guy I've ever met. I think it's amazing combo of training.

 

[mw18]

Volume can hide in the splanchnics and cause an effective hypovolemia.

 

[DNA888]

In the case of no UF (volume removal), one of the possible reasons for hemodynamic instability has to do with effective osmole removal. This risk is seen more so with intermittent hemodialysis and less so with CRRT. For instance, if the patient's serum sodium is higher than the dialysate or replacement fluid that is being used, then the fluid from the ECF compartment will shift into the ICF compartment. The reason why there's a higher risk for hypotension with intermittent hemodialysis is because it removes effective osmoles at a faster rate than CRRT. With intermittent hemodialysis, clearance is limited by blood flow rate which usually doesn't get much better after about 300+ mL/min (the slope of the curve plateaus). Dialysate flow rate is usually set at 600-800 mL/min. In contrast, with CRRT, the dialysate or replacement fluid is the limiting factor. The usual prescription for dialysate or replacement fluid is like what 25 mL/kg? I usually use 30 mL/kg. For a 70 kg man, if we use 30 mL/kg, then the dialysate or replacement fluid rate on CRRT is 2100 mL/hr. Note that intermittent hemodialysis flow rate is mL/min while CRRT is mL/hr. Therefore, the rate of 2100 mL/hr can be restated as 35 mL/min (compare to the 600-800 mL/min flow rate on intermittent hemodialysis). That's the reason why the dialysate or replacement fluid is the limiting factor in CRRT.

Have you ever noticed that patient's who get hypotensive during dialysis seem to tolerate it better when you're doing volume removal with no clearance (PUF for intermittent or SCUF for CRRT)? No effective osmoles are being removed, and only fluid is being removed. Again, this is assuming that the only difference between the clearance + UF scenario vs UF only scenario is whether or not clearance is being performed. You can use this knowledge of effective osmoles to your advantage. If I have to clean the blood and pull fluid from the blood, then I usually increase the Na bath to at least 140 mmol/L if not higher. The down side is that the patient can become hypernatremic which triggers the thirst sensation which then causes the patient to drink water (Nooo, I just removed fluid from you and now you're drinking fluid!).

Great point about the effective osmole removal! I did not consider that. Thanks for bringing it up. This would certainly be relevant for IHD, especially if baseline serum tonicity is significantly higher than that of the dialysate. Although I would imagine that this effect would be very blunt with CRRT and as such should not cause "acute" hemodynamic instability although slow loss of ECF volume may be seen. Plus, as you indicated, this effect is dependent on the baseline serum tonicity. On the contrary, if the baseline tonicity is lower than that of the dialysate (e.g. hyponatremia), RRT may instead lead to a gain of effective osmoles and therefore expansion of the ECF compartment. (Correct me if I'm wrong.)

Unrelated question - do you usually use a different Na bath in a standard patient? I have seen 140 for most cases. Also, I couldn't grasp how they can become hyperatremic with a Na bath of 140. My understanding is that the process of diffusion basically strives to equilibrate concentrations on either side, so a dialysate Na of 140 over time should drive the blood Na to 140? Thanks!

 

[Akarat]

Great point about the effective osmole removal! I did not consider that. Thanks for bringing it up. This would certainly be relevant for IHD, especially if baseline serum tonicity is significantly higher than that of the dialysate. Although I would imagine that this effect would be very blunt with CRRT and as such should not cause "acute" hemodynamic instability although slow loss of ECF volume may be seen. Plus, as you indicated, this effect is dependent on the baseline serum tonicity. On the contrary, if the baseline tonicity is lower than that of the dialysate (e.g. hyponatremia), RRT may instead lead to a gain of effective osmoles and therefore expansion of the ECF compartment. (Correct me if I'm wrong.)

Unrelated question - do you usually use a different Na bath in a standard patient? I have seen 140 for most cases. Also, I couldn't grasp how they can become hyperatremic with a Na bath of 140. My understanding is that the process of diffusion basically strives to equilibrate concentrations on either side, so a dialysate Na of 140 over time should drive the blood Na to 140? Thanks!

For a standard patient, I usually use a Na bath of 137 mmol/L. Maybe I'm making mountains out of mole hills but I believe that there's a lot of fine-tuning that one can do to the dialysate bath that ultimately has a significant clinical effect on the patient. The following excerpt from a journal article illustrates some of the intricacies that I am referring to with regards to choosing a Na bath:

RELATIONSHIP OF DIALYSATE SODIUM WITH [INTERDIALYTIC WEIGHT GAIN (IDWG)] AND HYPERTENSION​

Although our current approach to reduce volume overload with dietary salt restriction may decrease left ventricular hypertrophy,52 hypertension,53,54 and IDWG,53 the benefits may be muted due to the difficulty of long-lasting lifestyle and behavioral modification resulting in poor compliance. In recent years, attention has shifted to the “other salt,” dialysate sodium, a prerequisite for the delivery of HD.​

​

The link between high dialysate sodium concentration with increased IDWG and worsening hypertension has been reported since the very early interventional studies comparing high dialysate sodium (≥145 mEq/L) with low dialysate sodium (132-133 mEq/L) to reduce intradialytic symptoms including hypotension.55,56 Subsequent studies showed that while changes in dialysate sodium concentration from 130 to 136 mEq/L reduced the incidence of cramps, sodium diffusion increased with subsequent thirst, IDWG, and increased mean arterial pressure.27,57 Several studies have consistently shown that higher sodium dialysate concentration resulted in increased postdialysis plasma sodium level.31,58-60 While data for actual sodium loading are scant, the principles of the elegant early study by Van Stone et al60 examining higher dialysate sodium effects to overcome HD-associated side effects remain valid concepts. When dialysate sodium concentration exceeds serum sodium level, water shifts from the intracellular to the extracellular space, resulting in fluid removal from both compartments. In the presence of lower dialysate sodium concentration, the opposite occurs, with water moving into the intracellular space. Thus, identical ultrafiltration volumes with high dialysate sodium concentration results in lower plasma volume removal compared to lower dialysate sodium concentration.​

​

Mendoza JM, Arramreddy R, Schiller B. Dialysate Sodium: Choosing the Optimal Hemodialysis Bath. Am J Kidney Dis. 66(4):710-720.​

You're correct that using a Na bath of 140 mmol/L would not make one hypernatremic. I should have used a more accurate description of what I meant. Essentially, if serum Na goes above 140 mmol/L, it then triggers the thirst mechanism which is still intact in those with ESRD. This makes the patient more prone to drinking excessive amounts of fluid which would then cause volume overload and all of its bad consequences (higher BPs, edema, worsening of left ventricular hypertrophy, etc.). It's true that the patient will eventually drink but I would like to think that I at least made some mitigation in avoiding excessive interdialytic weight gain.