Search This Blog

Follow by Email

Tuesday, September 8, 2009

Using a Dialysis Machine to do ECMO

(There is more on ECMO on my Oct 1, 2009 Posting)

Many of the recent case reports indicate that sophisticated machines are required to treat the patients infected by the current novel H1N1 strain of influenza. Basic ventilators such as the existing Pandemic Ventilator Project designs may not be adequate for these H1N1 patients that develop ARDS. Pandemic Ventilator Project type units, however could possibly be utilized on other existing patients to free up more sophisticated equipment for patients requiring advanced therapies. I have also found a design for high frequency oscillatory ventilator that I posted (here).

Another technology that almost certainly will be in shortage during the pandemic is access to ECMO (Extra Corporeal Membrane Oxygenation) machines. ECMO machines oxygenate the blood directly using a gas permeable membrane. These machines can keep people with severely damaged lungs alive long enough for their bodies to repair their damaged lung tissues. There is very little of this equipment around. Many centers do not have any ECMO machines, or have only one.

Consider this:
An ECMO machine pumps blood from the patient, adds an anticoagulant, runs it past a gas exchange membrane to remove CO2 and add O2, regulates the blood temperature with a heat exchanger, removes air bubbles via drip chambers, checks incoming and return pressures, and has safety systems to ensure air is not infused, or pressure limits are not exceeded.

A dialysis machine pumps blood from the patient, adds an anticoagulant, runs it past a dialyzing membrane to stabilize electrolytes and remove toxins and fluid, regulates the blood temperature by controlling dialysate temperature, removes air bubbles via drip chambers, checks incoming and return pressures, and has safety systems to ensure air is not infused, or that pressure limits are not exceeded.

Hemodialysis System

They are pretty similar eh?

Note that terminology for blood access is opposite in ECMO vs hemodialysis.

  • In ECMO, the port where the blood is drawn into the pump is termed the Venous line and the port where the blood is returned to the body is termed the Arterial line.
  • In Hemodialysis, the port where the blood is drawn into the pump is termed the Arterial line and the port where the blood is returned to the body is termed the Venous line.
  • In CRRT, (a form of hemodialysis) the port where the blood is drawn into the pump is termed the access line, and the port where the blood is returned to the body is termed the Return line.

The Hemodialysis picture is from METU BIOMAT, and the ECMO picture is from Medscape. (Note there is an error in the Medscape ECMO drawing, both pressure ports are named "Post-Membrane Pressure Monitor". The lower one should be named "Pre-Membrane Pressure Monitor) Note also that fluids and heparin are normally infused post pump in hemodialysis, as this method is usually considered a safer method. Air removal, and monitoring safety systems are also not in the ECMO picture. Both VV-ECMO and Hemodialysis can use a Jugular Venous Dual Lumen Catheter for access.

There are two types of ECMO. VA-ECMO or Venous-Arterial ECMO, has a more complicated method of attaching to the patients circulation system. VA-ECMO operation is similar to the use of a heart-lung bypass machine in that it replaces the function of both the heart and lungs of a patient. VV-ECMO or Venous-Venous ECMO, has a less complicated method of blood system access. It is done using high flow central line catheters similar to the ones used for dialysis. It replaces only the lung function of the patient.

Some patients with H1N1 are getting lung damage and progressing to ARDS. They may require ECMO because their lungs are so damaged that they can no longer provide enough gas exchange to maintain other body functions. A ventilator may not be adequate in these situations. The heart is not usually compromised. These patients could benefit from VV-ECMO if a machine was available. As stated earlier, available ECMO machines would probably be in very short supply during the pandemic.

You can see that the equipment for ECMO is very similar to the equipment required to perform dialysis. In fact CRRT or SCUF are sometimes done in order to control electrolyte and fluid volume levels by adding a dialyser to an ECMO machine without needing any additional equipment.

It seems to me that one could do VV-ECMO treatments using a dialysis machine with a diffusion membrane oxygenator attached in line on the blood tubing set. Some extra gas and oxygen regulators and controls may also be required. If a standard hemodialysis machine is used, it can be run at a low dialysis flow rate (available on machines such as the Fresenius 2008K) to run in a SLED (Sustained Low Efficiency Dialysis) mode continuously. I would like to hear comments from people that have worked with ECMO equipment to hear if they think this is at all feasible.

This Just in (Sept 15, 2009)

Article in New York Times about ECMO use in H1N1 pandemic and potential shortage of ECMO machines.

Lancet article about the efficacy of ECMO for severe influenza treatment.

Bloomberg article on using ECMO for near death swine flu cases.

Belfast Telegraph article about ECMO;jsessionid=80D2A25F7E4033BF410D32971134D6DA?postingType=posting&mode=thanks&postingId=14493924

Update, Sept 18 2009

I have been thinking about this doing ECMO using a dialysis machine for a few days now. So far I have not had any comments either for or against on this blog.

I have done some further research into the equipment required for ECMO and some of the problems with ECMO therapies. It appears that maintaining systemic coagulation using heparin is sometimes a problem. Patients may not properly respond to the heparin therapy, they may have allergies, or there may be bleeding problems associated with systemic coagulation. These are problems that are also very common in hemodialysis and CRRT therapies. One solution to this problem is to use regional citrate anticoagulation. Citrate is infused into the blood circuit at the blood access port to initiate anticoagulation and calcium is infused at the blood return port to cancel the effect of the infused citrate.

This can be more complex than straightforward heparin infusion because the infusion of these chemicals also alters the calcium, pH, fluid volume and sodium levels of the patient. In CRRT and SLED therapies these parameters are monitored and controlled by adjusting the sodium and bicarbonate levels of the dialyzing and infusion fluids. Patient fluid volumes are also easily controlled by the dialysis machine.

Regional citrate anticoagulation has been shown to significantly extend the filter (dialyser) life compared to heparin coagulation by reducing clotting. It is sometimes used when the patient has HIT (Heparin Induced Thrombocytopenia). Regional citrate anticoagulation can also reduce other complications that would occur when using systemic anticoagulation protocols.

During a pandemic, it may be difficult to obtain enough membrane oxygenators to do ECMO. It is reasonable to assume that regional citrate anticoagulation could also extent the serviceable life of the membrane oxygenator by reducing clotting in the device. It will be important to make the best use of whatever supplies one has on hand. If it is indeed possible to use a dialysis machine to do ECMO, and also employ regional citrate anticoagulation with it, this could be a good way to save more lives with the possibly limited supplies available.

Here is a link to a PubMed abstract of an ASAIO journal article about using regional citrate anticoagulation with ECMO.

Some More Info if this Intrigues You...

JAMA article shows that most patients with severe H1N1 that are treated with ECMO survive

Some general Info on ECMO systems and complications

Here is a link to a Patent for an ECMO system

CDC info on the use of ECMO and CRRT on novel A H1N1 patients.

Response to the first comment by Anonymous (see below)

Thank you for your comments. I have been waiting to get some feedback on this issue. Just to clarify, ELSO is Extracorporeal Life Support Organization centered at the University of Michigan.

Now you have question about the origin, purpose and legitimacy of the Pandemic Ventilator Project. It was started on Feb 22, 2007 to promote alternative methods of supplying additional ventilators during a pandemic. In order to reduce the death toll of people either ill from a pandemic or those who would be denied life support so that the ventilator they are using could be used to save a pandemic victim (due to triage protocols). Now when you question legitimacy, I am not quite sure what you are after. I am not trying to defraud or manipulate anyone, and my motives for the project are entirely humanitarian. It is not a commercial venture; in fact I have spent a fair bit of my own time and money on it. All of my work and postings are available for you to view and see for yourself. Now if by legitimacy, you mean authority, I really have none. The opinions I express are my own. It is up to the reader to determine if my arguments are rational and my sources of information are valid.

Now when you warn against an untrained person just setting up ECMO on a dialysis machine when no prior testing or feasibility studies have been done you are absolutely correct. When I proposed this idea, it was for people that are qualified to do ECMO treatments to try to find innovative alternative ways to provide this potentially life saving treatment even if there were a shortage of existing ECMO equipment during a pandemic. I was hoping that knowledgeable people could look at the idea and see if they could make it work safely rather than dismiss it out of hand. Perhaps a someone could find a solution to this problem with the pumps that you mentioned.

Now when you assuredly state that there will be NO shortage of ECMO systems in the US, I do not think you can say that for sure. When we have Dr. Michael Osterholm, director of the Center for Infectious Disease Research and Policy at the University of Minnesota (CIDRAP) worried about a shortage of ECMO machines ,, and with Dr Dr. Giles Peek of Glenfield Hospital in Leicester, England talking about how few the number of ECMO machines are available in Britain. The World Health Organization is also warning developed countries "to anticipate this increased demand on intensive care units, which could be overwhelmed by a sudden surge in the number of severe cases."

There is agood chance that the current H1N1 pandemic will remain mild and within the ability of our current infrastructure and surge capacity to manage, But I do not believe anyone can definitely say that this will be the case.

What you say about legal liabilities is unfortunately sadly true. The heroic measures undertaken by individuals during the polio epidemic to build their own ventilators to save the lives of children could never happen in today’s legal liability climate. The only hope for that is if legislatures provide legal liability exemptions to the individuals that decide who gets which machine and treatment in a pandemic. Under today’s legal climate it is more prudent for a physician let his patient die by denying access to a potentially life saving treatment than to risk a lawsuit by using an uncertified device.

I must say in defence of any nephrology professionals that read this, hemodialysis is also a type of life supporting treatment that is done extracorporeally. Most of the complications that can occur in VV-ECMO can also occur in hemodialysis. Hemodialysis is routinely done in a safe mode by trained individuals. There were over 300,000 patients safely dialyzed for more than 150 million hours of treatment in more than 4000 centers in the US last year alone.

Clarence Graansma


  1. This is my first exposure to this site so I am uncertain as to it's origin, purpose, or legitimacy.
    But anyone reading this should understand that extracorporeal life support (ECMO, ECLS) is an established life support for critically ill patients. There are over 120 ECMO centers that report to ELSO. Over 40,000 patients have been reported to ELSO since 1989. There are guidelines for who should be trained to do ECMO and how that training should occur. There are guidelines for how hospitals should go about establishing an ECMO program.
    Although there are many similarities between Dialysis pumps and ECMO pumps...they are significantly different. It would be foolish for an institution to try to provide extracoporeal life support with a modified Dialysis circuit. The legal liabilities would be incredible. We are not going to have a shortage of ECMO Systems in the U.S. The ECMO community is well aware of the issues with H1N1 and are adequately preparing to provide the necessary support. This may require transporting a very ill H1N1 patient. But that can be done safely and there are processes in place to make that happen.

  2. I am an ICU specialist. ECMO is not of proven value. Even the biggest study to date a had deeply flawed methodology, and is unreliable. My experience is that people can safely survive very low levels of oxygen well below the threshold at which ECMO fans start ECMO. Provided people do not panic and keep doing simple things, normal ventilation will suffice for most patients. What kills patients is when their doctors try too hard to achieve normal oxygenation in this most abnormal situation. Certainly I have looked after a niumber of patients who have tolerate oxygen saturations around 75% for a week or more when many people would give up or suggest ECMO. It is unnecessary.

  3. As for ecmo thank God for it. My 22 year old daughter has been on it for the last eight days due to complications from the swine flu.Both of her lungs at the start were completely messed up now her left lung looks completely normal and her right lung is recovering but much much slower. Prior to being put on ecmo she spent 10 days in the icu on a ventilator with no improvements whatsoever!

  4. Dear Clarence Graansma,
    This is a very interesting topic and I like the way you are thinking outside the box to explore options for the "what ifs". As a perfusionist who has been involved in ECMO (neonatal, pediatric and adult)for over 20 years, I believe that we could very well have a severe ECMO resource (staffing & equipment)crisis.

    My first thought regarding improvising a hemodialysis apparatus to do ECMO would be possible pump blood flow limitations. An adult on ECMO (VV or VA) may require blood flow rates of 3 - 6 liters/min. Dialysis is much lower flow rate, I'm sure. Could a larger Diameter pump raceway tubing be used on these systems?

  5. But as your links indicate ... the dialysis machine appears capable to perform ECMO on babies.

  6. Thanks for the feedback Ed.

    Dialysis machine blood pump flows typically max out at about 500 to 600 ml/min. I am most familiar with the Fresenius 2008K. You can get the operators manual and technical manuals here.

    The 2008K maximum is 600 ml/min. The allowable tubing sizes that can be put in the standard pump head range from 2mm to 10mm. It should be possible to use large diameter tubing and then program the pump as if it were small diameter tubing. This would increase the actual blood flow through the pump however the display would still read the lower value. There is still a practical limit though. The machine also has load sensing circuitry to detect stalls, overloading and jamming of the pump head. I could test this and see what the maximum flow I could get out of the machine without triggering the overload function. These machines also allow more than one pump to be used at a time. Dual pumps are often used for single needle dialysis modes. The pumps then alternate from each other. It would not be difficult to “enhance” the circuitry to enable two pumps to run at the same time (in parallel) while still maintaining the safety systems that stop the pumps in the event of pressures out of range or air being found in the return line.

  7. I did some testing on a dialysis machine today. Although the specifications in the manual say that pump tubing diameter sizes from 2mm to 10mm can be used, the pump only has settings for 2.4, 4.8, 6.4 and 8 mm diameters. The pump roller guides and pump raceway and tubing retainers will accept tubing up to 10mm in diameter though. When the diameter setting of the pump is changed to a smaller size, the maximum rate setting is also reduced. I did some testing to see what the maximum output at the various settings would be using 8mm tubing. These are the results

    Tubing ID……Max setting….Actual Flow Rate (ml/min)

    The maximum actual flow seems to be at the 4.8 setting. The output of the pump is proportional to the area of the tubing. 10mm tubing would have an area 156% greater than 8mm tubing. We should be able to expect a maximum flow rate for one pump of 1.1 liters/min using 10mm tubing. With 2 pumps we could get 2.2 liters/min.

  8. Dreamer,
    That's good to know. These machines would be suitable for ECMO application in neonates, infants, and pediatric patients <10 kg. I am not sure what the afterload limitations would be for dialysis units, but I think that using the new polymethylpentene hollow fiber oxygenators would be most suitable rather than the old workhorse silcone oxygenators which have a very high resistance across them (and are most often used by ELSO centers).

  9. The idea sounds novel . But flow limitation will make it impractical. Alhough studies have shown that hemolysis is not factor between vortex and roller pumps, I am not too sure about these tiny pumps hitting on even tinier tubings in a dialysis machine.

    Perhaps in infants and neonates, but even then, most times we tend to aim for flows between 800mls and a liter. Dialysis machine flow rates are too little and too close a margin to play around. You may not have the flow rate when you really want it, unless the machine itself can be modified drastically to take on bigger tubings and bigger pumps.

    If that were the case, it would be easier to assemble a simple ECMO machine.

  10. Good comment ... simple ECMO systems(contrasted to the relatively complex traditional Bartlett-style roller pump ECMO systems) are rapidly being adopted for adults with H1N1 by many of my perfusionist colleagues across the country. These ECMO systems use the latest generation of centrifugal pumps, the low resistance PMP membrane oxygenators and dual lumen veno-venous cannula. The idea is to make the system easy for the bedside nurses and simple and safe enough minimize staffing depletion by allowing a single in-house perfusionist to be available for trouble-shooting.

  11. I definitely like the idea of thinking of unique ways to do things. That is how we make progress. BUT, I can assure you the capacity to provide ECMO support in the United States will far surpass the need for it even during a severe H1N1 crisis. The established ECMO community is geared up and ready for this and the Perfusion community in general is adapting and ready to support patients. The resources are there if institutions are willing to look for them and willing to accept the help. Trying to create a make-shift system with a Dialysis machine could be disasterous...although as I said before...I like the thought process.

    And in response to the "ICU specialist" who doesn't believe in are doing a disservice to your patients by denying them the opportunity for this technology when appropriately applied. ECMO is not appropriate for every patient...just like mechanical ventilation is not appropriate for every patient. And certainly high pressure ventilation is dangerous and often deadly. But there is a patient population that benefits greatly from ECMO when properly applied. ECMO does have a proven value and has for years. Just about any study you find on anything can be picked apart by someone. I think you will find 1000's of health care professionals in 100s of institutions that have supported over 40,000 patients that would all disagree with your opinion...and I imagine those patients and family members would disagree with you also.

  12. UAE,
    I am a PICU resident and in my country (united Arab emirates) we don't have ECMO machine and we lost some children, who were completely healthy, because we don't have it. I like this idea, and i would like to know more about how to set the dialysis machine to work as ECMO for example for 3kg infant.

  13. This comment has been removed by the author.

  14. You shared a fabulous knowledge with us. Thanks a lot for this useful information. I also want to share some great knowledge relevant to Ventilator Machines

  15. I guess I have selected a mind blowing and interesting blog. look at this site

  16. Hi to everybody, here everyone is sharing such knowledge, so it’s fastidious to see this site, and I used to visit this blog daily orthodontic treatment glenfield

  17. In today context, this is now a real dire need indeed.


Commenting has been limited for this post

Note: Only a member of this blog may post a comment.