Hello All:
Several people asked me to share what I learned while researching options for using dialysis equipment to do therapeutic plasma exchange (TPE). Here is what I found. Forgive me for being so long-winded, but this has been quite an adventure with a dizzying amount of information!
All but one pediatric center we contacted used a cell separator, not dialysis equipment, regardless of whether the service was provided by nephrology or by another department such as blood bank.
Obviously, the main attraction of a dialysis-based system is that you can use existing equipment, avoiding the cost of buying, maintaining and training your staff to use another machine. Some have suggested that a plasma filter approach carries less risk of thrombocytopenia than a cell separator approach. The plasma filter can only do TPE and not leukopheresis, which is sometimes requested for leukemic patients. (There is an interesting article in a recent Ped Neph suggesting that leukopheresis may no longer be necessary in the age of rasburicase...but that’s another topic.)
At present there are only 2 plasma filters available in the US. One is the Plasmaflo, made by Asahi as a stand-alone product. The other is made by Gambro as a component of the Prisma system. It comes as part of a set called the TPE 2000.
In theory, you simply place the plasma filter in the circuit where the dialyzer or hemofilter would go, and perform a hemofiltration procedure (no dialysate). The large pores in the filter allow plasma proteins to come out. As the plasma is pulled off, you infuse albumin or FFP to replace it at the same rate, in a process very similar to CVVH. It sounds very simple, but there are a number of technical problems.
First, it’s important to note that removing plasma is physiologically different from removing ultrafiltrate. When you remove water from the vascular compartment, extravascular fluid can diffuse in to buffer the volume change. Even then, it is tricky—we all have experience with intradialytic hypotension. When you remove plasma from the vascular compartment, there is no buffer, so there is an even higher risk of cardiovascular instability. It’s important, then, that plasma removal and replacement occur at precisely the same rate. This is why CRRT equipment is ideally suited to doing TPE.
Most people we talked to did not think the UF controller in a dialysis (not CRRT) machine was precise enough to do this. No one cited any clear data on this, but multiple people expressed intense fear of using the built-in UF controller for plasma exchange. And let’s face it, we’ve all experienced the post-dialysis weight change not correlating with the reported UF. Another problem with using the built-in UF pump is that plasma has different optical properties from ultrafiltrate, and it will set off the blood leak detector, which stops the blood pump. So you might have to take apart your machine and somehow disconnect or disable the blood leak detector.
When using a dialysis machine, the preferred method everyone described was to use a separate external pump to do the removal and replacement. This is a special duplex pump in which the removal and replacement circuits are very tightly linked, assuring that they occur at the same rate. Finding such a pump is very difficult. We could only find one supplier, who seemed a little shady and said they MIGHT be able to get us a used pump for $3000. (“I’ll have to check with my guy,” she explained.) She also explained that it comes with no warranty and no assurance that parts or service will be available for it.
We found one adult nephrologist who said this was all just marketing propaganda from the pump company, and the UF controller would work just fine. But his center doesn’t do that—they use the pump!
Another disadvantage to using a dialysis machine is that your machine probably will not start up or run without being connected to water, acid and bicarb. You won’t be using these for TPE, but you still have to set them up, which is a waste and an added expense. If you are doing this outside the dialysis unit, you’ll have to lug your portable RO along and have a plumbing connection in the room. Again you can see why CRRT equipment is better suited to TPE.
TPE has some unique anticoagulation issues. If you use heparin, it will come out with the plasma, so you will have to use more than with hemodialysis. With citrate you have to be careful about toxicity, especially when replacing with FFP, which is already loaded with citrate. Coagulopathy is also a concern, since you’re removing plasma proteins involved in coagulation. For reasons I don’t fully understand, plasma filters apparently have a lower tolerance for high TMP, and occlude more easily than other types of membranes.
In summary, using hemodialysis equipment for TPE gets pretty wild and wooly. In all cases, you have to improvise a homemade setup, mixing equipment from different manufacturers and possibly hacking your equipment to work differently. You may run into problems with tubing or other component compatibility. You’re pretty much on your own, as there is hardly any published experience out there. The manufacturers cannot provide much training or technical support when you are using their equipment for a purpose they did not intend, disabling the safety features, etc. Obviously, this entails some risk and the potential for unforeseen problems.
Wouldn’t it be nice if somebody made a system that was actually designed and built to do both renal replacement and TPE? It turns out two companies do. The Gambro Prisma and the Baxter Accura are approved for TPE procedures.
To use the Prisma, you need Gambro’s plasma filter set, the TPE 2000, which fits on the Prisma just like the M60 sets which are used for CVVHD. This is an FDA-approved system, with all components from one manufacturer and hence sure to be compatible. Gambro also provides staff training for this system. There is only one catch...you will need a $3000 software upgrade. Look at the label on the back of your Prisma – if you do not see the letters “TPE” you will need to pull out your wallet. Since the tubing is integral to the system, you cannot use a smaller tubing set to reduce the circuit volume.
Baxter offers their Accura CRRT machine for plasma exchange, using the Asahi plasma filter. No software or other upgrades are required. Again, you enjoy the benefits of an “official” use of the machine, with the manufacturer’s blessing and technical support. The Accura will run blood flow rates up to 450 ml/min, which is helpful since TPE is usually an intermittent therapy. You can also use pediatric tubing. Of note, Baxter also makes a dedicated filter-type machine that does TPE only. It’s called the Autopheresis-C and it uses the Asahi filter.
A very similar, off-label approach is to use the Baxter BM25 with the Asahi plasma filter in the circuit. The Ottawa group has reported their experience with >60 treatments using this approach (see abstract below). They found it to be very satisfactory. This is the only pediatric group that reported using something other than a cell separator for TPE.
Some of the newer dialysis machines are designed to work in a CRRT mode as well. These hybrid machines might have some interesting implications and you could theoretically use the Asahi filter with them to do TPE. Perhaps you would not need the external pump in this case. But again, you are on your own here. If you are the first to try this be careful, and then report your results so others can benefit.
Finally, there was an inquiry about the cost-effectiveness of all of this. I really can’t answer that one. I think it depends on the economics of your institution and the bizarre and incomprehensible practices of Medicaid and insurers. We approached it from the viewpoint of adding a needed service with minimal investment. There is no center within hundreds of miles which will do TPE on children, so having it here will benefit the children we serve. Whether we will profit from it is anyone’s guess.
We nephrologists are the experts in extracorporeal blood circuits and vascular access, and we have many patients who need TPE, so it’s logical and reasonable for us to provide it. In fact, we can probably do it better than other services. We just need to remember that this is NOT just glorified hemofiltration. It’s a completely different process on many levels. So however we approach it, we must do our homework, be careful and be diligent.
Thanks to everyone who responded or expressed interest in this topic.
Best wishes,
Rob
Robert S. Gillespie, M.D., M.P.H.
Pediatric
Nephrologist
New info -
2/2007: A colleague from outside the USA offers this
suggestion...
We
do plasmapheresis with dialysis equipment only. We use
Gambro PF 2000N which is a plasmapheresis filter that fits on the AK200
dialysis machines (Gambro) that we use. There is also a smaller filter
1000N, used for smaller kids. The machine is run without dialysate, and
the plasma removed is collected into an open measuring vessel.
Replacement (FFP or albumin, depending on the case) is given into the
tubing directly by syringe ( by the nurse), drawing from the FFP bags
and pushing through a 3 way valve into tubing containing the returning
blood. A nephrologist is always present during each procedure, and the
I's and O's are meticulously balanced, while monitoring vitals. You did
not mention the PF 2000 in your info page. No additional software is
needed.