PD PHYSIOLOGY: The Basics and Comparisons to HD

John M. Burkart, MD: Hello. I'm John Burkhart, and I'm a Professor of Medicine at Wake Forest University Medical Center, and I'm going to talk about the physiology of peritoneal dialysis. Along the way, I'm going to describe the basic concepts, but also make some comparisons to hemodialysis since many of us are much more familiar with that therapy for end-stage renal disease. Now I'll start with the components of a peritoneal dialysis system. Of course we're familiar with the concept that in order to do dialysis, we have to bring blood in contact with dialysate, but it's separated by a semipermeable membrane. So here in this cartoon you see the blood compartment on your left, the dialysate compartment on your right, and the semipermeable membrane in the middle. Of course we want to remove urea, creatinine, and potassium from the blood side, and in some cases we would want to add something to the patient's body, such as buffer from the dialysate side. Now if we compare peritoneal dialysis to hemodialysis, we know that there are some basic differences. In hemodialysis the blood side is extracorporeal. We have a blood pump to move the blood through the artificial dialyzer, whereas in PD [peritoneal dialysis] the blood supply is that blood that goes to the mesenteric vessels. In contrast, when we look at the dialysate in hemodialysis, the dialysate is not sterile; it's locally produced, whereas in PD the dialysate is sterile and then shipped to the patient. Of course you know that to do dialysis you must have access to the blood, and in peritoneal dialysis the peritoneal membrane acts as that artificial, semipermeable membrane, so we have a catheter that transverses the abdominal wall; brings the dialysate in and out of the abdominal cavity; and as I show in this diagram, the mesenteric vessels that supply the intra-abdominal organs are the blood supply, in this case, to do dialysis. Now the membrane, the semipermeable membrane, is the serosal membrane that lines the peritoneal cavity. There's a thin mesothelial surface to this membrane, and these mesothelial cells are alive; they're functional; they provide surfactant to allow for peristalsis in the peritoneum, but also control inflammation in the peritoneal cavity.

The surface area of the peritoneal cavity approximates that of the body surface area. Eight percent of it is the visceral peritoneum, and about 20% is the parietal peritoneum. We're not sure where the majority of the peritoneal dialysis exchange occurs, but certainly it's thought that it may be via the visceral peritoneum. I'll come back to the fact that we have lymphatic drainage from the peritoneal cavity. This is mostly through subdiaphragmatic stoma, but there's a small amount that's absorbed directly into the tissues and then into the tissue lymphatics. Now in order for me to adequately explain the physiology of peritoneal dialysis, we'll use some mathematical models to explain the known clinical observations. So, first of all, it's important to note that every capillary has the potential to play a role in peritoneal dialysis, in diffusion and convection. But not all of them are perfused equally, and not all of them are of the same distance to the mesothelial surface. So one thing that we know is that the influence of each capillary depends on its proximity to the mesothelial surface, which, of course, is the blood-dialysate interface. The vascularity per unit of surface area, or the percentage of perfused capillaries, plays an important role in influencing the rate of peritoneal transport. Hence, peritoneal transport characteristics can change, and one of the things we'll talk about is the effective peritoneal surface area. For instance, during peritonitis when the peritoneal membrane becomes inflamed, the perfusion changes, and so the rate of transport changes.

We also think about how solute in water moves from the blood side to the dialysate side, and this happens through various pores. There are experimental models which actually show the presence of these pores, and there are mathematical models that we use to describe transport. There are 3 pore types. There are large pores which are small in number and are likely anatomical clefts between the endothelial cells. These pores participate in transport of macromolecules and are at the venial end of the capillaries. There are small pores, many, many more in number, which transport small solutes and water, and this will be important when we talk about convective removal of solute. These are likely at the arteriolar end of capillaries, and again our clefts between endothelial cells -- very important when we talk about the crystalloid- and colloid-induced ultrafiltration. And finally, there are ultrasmall pores or aquaporins. These are transcellular pores, and they are involved in the transport of water only, and these are important when we talk about crystalloid-induced ultrafiltration.

Now this cartoon is used to illustrate these pores, and as you can see, there are large pores which transport macromolecules; small pores which are clefts between the endothelial cells, used for the transport of urea, creatinine, and water; and then there are transcellular pores which are channels that run through the cell via which only water can be transported. These transcellular pores transport water only; the small pores are used for our crystalloid-induced osmosis and also the removal of small solutes.

Let's talk briefly about peritoneal dialysis solutions. As you know, these solutions contain glucose, or hydrated glucose, dextrose, and there are various concentrations that we use depending on the amount of ultrafiltration that we would like to get per dwell. We'll come back to that. The newer solution is icodextrin, which is a macromolecule, 7.5% concentration. You can see that the osmolality of the solution changes based on the percentage of dextrose that we choose to use, and there are solutions with dextrose that are available that have a lower calcium concentration. Now why the various components of the PD fluid? Why lactate, for instance, when the naturally occurring buffer is bicarbonate? You remember that peritoneal fluids are shipped, and so they have to be stored. So if we had bicarbonate in those fluids which contain calcium, the calcium and bicarbonate would precipitate during storage. Why the low pH? Well, again, these fluids need to be sterilized, and in order to minimize glucose degradation product production during sterilization and storage, we have to have the low pH, so that we have low GDP [glucose degradation product] amounts in these fluids. And again, why the high osmolality? If we're using dextrose-containing fluids, we use these higher osmolalities in order to have a crystalloid osmotic gradient, so that we can get ultrafiltration. Now let's think about that crystalloid osmotic gradient for a moment. We know that typically capillary and interstitial osmolality run around 285, maybe a little higher in uremic plasma, so you can see that in order to have a crystalloid osmotic gradient to favor ultrafiltration, we would have to have the higher osmolalities I just showed you. Whereas if we were using a colloid osmotic agent, such as icodextrin, then you want a larger number of macromolecules in the peritoneal dialysate compared to the blood. So, the plasma oncotic pressure is mainly due to albumin, icodextrin as a 7.5% solution; has more macromolecules than we typically see in plasma; and so that's illustrated in this slide. Remember, uremic plasma has an osmolality of 285-300. You can see that the 1.5%, 2.5%, and 4.25% dextrose solutions have higher osmolality so that we could get ultrafiltration. Whereas icodextrin, which has a colloid osmotic gradient, is isotonic to plasma.

Now of course when we do dialysis, we want to remove retained solutes, and we also want to remove retained fluid, so let's talk about the physics of peritoneal dialysis, and I will compare and contrast this with what we see with hemodialysis. Of course to remove solute we have 2 distinct processes: We have diffusion and convection. We're familiar with diffusion, and that is the movement of a small solute down a concentration gradient; it's influenced by the dialysate, volume, and the surface area for contact. Whereas in convection, that is the process of removing solute during ultrafiltration. So as water is removed; solute comes with the water; and in both hemodialysis and peritoneal dialysis that's an important way to augment middle-molecular-weight clearance. So, in this cartoon I tried to demonstrate the diffusion kinetics. As you can see, the blood side is on the left; the peritoneal cavity on the right; and the higher concentration of urea moves down its concentration gradient, across the semipermeable membrane, into the peritoneal cavity.

Now, we all know that the rate of diffusion is dependent on molecular weight, and here is a classic slide for hemodialysis. This slide is set up with a blood flow that varies and a fixed dialysate flow, so you can see that as you increase the blood flow, the rates of diffusion increase. But the rate is markedly higher for urea, only 60 d, than that of creatinine, 113 d, and that is faster than what we see for larger molecular weight solutes -- such as vitamin B12 or beta-2 microglobulin. So diffusion is very efficient for small-molecular-weight solutes but not as efficient for larger molecular weight solutes.

Similarly, if we were to fix the blood flow but alter dialysate flow, we see that diffusion is molecular-weight-dependent. Again, smaller molecular weight solutes diffuse faster than larger molecular weight solutes. So in hemodialysis we know that we can change the rates of diffusion by changing the time-on treatment, or the blood flow, or the dialysate flow, or the surface area, or actually influencing the membrane itself by changing the characteristics of the membrane, the pore size, the surface area, and the blood flow through the dialyzer.

Now peritoneal dialysis is similar except the patients are born with their membrane; we cannot alter it like we can with hemodialysis, but rates of diffusion, again, are molecular-weight-dependent. So in this cartoon I illustrate what would happen if at time zero you put fresh dialysate in the abdomen; over time during the dwell urea or creatinine or middle molecules would move from the blood compartment to the dialysate compartment. If we were to look at the ratio of dialysate urea to plasma urea, or dialysate creatinine to plasma creatinine, we can come up with a D/P [dialysate/plasma] ratio, and that gives us an idea of the percentage equilibration between dialysate and plasma solute. Eventually this D/P ratio becomes unity (ie, the dialysate), is saturated and equal in concentration to that in the plasma, and at that time you no longer have diffusion of that solute. You can see that that occurs faster for urea than it does for creatinine, and it occurs faster for creatinine than it does for middle molecules.

So what happens as far as diffusion during a typical PD dwell? Well it's maximal during the first hour because we have those large concentration gradients. By 4 hours, in a typical patient, urea is 90% equilibrated, while creatinine is only 65% equilibrated. Larger molecular weight solutes need longer dwell times, but importantly, the rates of diffusion vary between patients and between solutes. So if we wanted to optimize a patient's peritoneal dialysis prescription, it would be important for us to know what the rates of their diffusion are. One way we can do that is to do a standardized dwell, what's called a peritoneal equilibration test, and we look at the rates of diffusion or the D/P ratios after a time dwell in order to classify that patient's transport characteristics. So this slide shows the various transport rates that you would see in a peritoneal equilibration test. This was first standardized by Dr. Twardowski in 1987, and you can see in the orange part of the graph on the left that a patient who is a fast transporter or a rapid transporter equilibrates the urea in the dialysate faster than a slow transporter would. Similarly, because urea is a small molecule, you can see that there is not as much disparity between a rapid and a slow transporter as there is in the right side of the slide for creatinine, a larger molecule. But again, in the orange, a rapid transporter equilibrates faster than a slow transporter, and if you were to look at the D/P creatinine, you can see that at 4 hours the average patient is 65% equilibrated. These D/P creatinine ratios are what we typically use to characterize a patient's transport characteristics. So the rapid transporters are between 8 and 1, for instance, at 4 hours; the slow transporters are between about 0.45 and lower; and then the other patients fall in the average range hovering around 0.65. It's important to remember that the D/P for creatinine and the D/P for phosphate are similar; they're similar in molecular weight, so when you listen to the lectures on the dialysis prescription, you see that whatever maneuvers you would do to increase creatinine clearance would likely increase phosphate removal also. Any changes in the prescription that would decrease creatinine clearance would decrease phosphate clearance. If we look at the peritoneal equilibration test, this is the test that we use to guide our therapy. We do a standard 4-hour dwell with 2 L of 2.5% dextrose, and then we calculate those D/P ratios. You will hear about modifications of this test, such as using a 4.25% dextrose dwell and looking at sodium sieving at 1 and 2 hours. Now it's important to note in reality that the D/P ratio on these 4-hour dwells is not due to diffusion alone. Some of the creatinine that ends up in the dialysate at 4 hours is due to convection, and I'll talk about that in a moment.

How do we increase diffusion? Well, we do that by increasing the concentration gradients or increasing the surface area. There are a few ways that we can maximize concentration gradients. We can increase the instilled volume per exchange. We can increase the frequency of exchanges, so we're working on the steep end of the slope of the diffusion curve. Or we can use alternative PD therapies, such as tidal PD or continuous-flow PD where we constantly are inducing new solution, which of course has no urea or creatinine in it. As far as increasing surface area, we could use larger dwell volumes. There are ways to do this pharmacologically, such as with using nitrous oxide, but these results are not clinically consistent, and so it's not something that we use in normal everyday practice. Now I've been talking about diffusion of solutes from the blood to the dialysate, but certainly solutes move from the dialysate to the blood. For instance, we add lactate to the dialysate; that's absorbed, and it becomes our buffer. We could add insulin to the dialysate. We could add amino acids to the dialysate instead of glucose. We could add erythropoietin or other medications, but most importantly, the glucose that's in the dialysate on purpose to act as our osmotic agent is slowly absorbed from the dialysate.

Let's think about a typical PD prescription. If the patient was doing automated peritoneal dialysis, so they were doing multiple exchanges overnight and then had a daytime dwell, you could see that the daytime dwell may last up to 15 hours, and that's important. We'll come back to that. If the patient was doing continuous ambulatory peritoneal dialysis, sort of reversing the dwell cycle and doing multiple exchanges during the day and 1 long overnight dwell, that overnight dwell is only 9 hours. If you remember the diffusion curves I showed you, sometime during that long overnight dwell of CAPD [continuous ambulatory peritoneal dialysis] or the long daytime dwell of APD [ambulatory peritoneal dialysis], diffusion of small solutes will stop. So, after say 6-8 hours, you no longer will have diffusion of creatinine or urea. You will have diffusion of middle molecules, so to leave the fluid in longer is helpful for middle-molecule removal. However, if we leave the fluid any longer, we're not efficient as far as removal of urea or creatinine because we have reached a unity for the D/P ratio, and we no longer are removing any more urea or creatinine. That brings us to convection because, remember, we also remove solute in the water that is removed during ultrafiltration. Again, in this cartoon I try to illustrate that we are having a hypertonic glucose concentration on the right or the peritoneal cavity. Because of that osmotic gradient, water moves from the blood side to the dialysate side, and with that water movement we have the possibility of removing solute also. This is important because depending on what pores that water moves through, ie, small pores, we can also remove middle molecules. However, remember that if the water is moving through the aquaporins, we are removing water only and no solute. So in peritoneal dialysis this ultrafiltration is important. It provides up to 20% of small solute clearance in a typical patient.

Let's review the physics of peritoneal ultrafiltration. In contrast now in hemodialysis we adjust the transmembrane pressure, and by adjusting the pressure across the hemodialysis membrane, we influence the rate of ultrafiltration. That's dependent on the ultrafiltration coefficient of the dialyzer that we pick to use. Remember in peritoneal dialysis patients are born with their dialyzer. We cannot pick a dialyzer, and therefore we need to know what this patient's peritoneal dialysis transport characteristics are in order to optimize ultrafiltration. Remember that glucose in the peritoneal cavity provides the osmotic gradient. However, the glucose is slowly absorbed. So during the dwell, water moves from the blood side to the dialysate side so that we can equalize the crystalloid osmotic concentration in both compartments. So what happens during the dwell? Let's talk about what would happen with an optimal or maximal osmotic stimulus (ie, a 4.25% dextrose solution). We have great ultrafiltration at the start of the dwell time, about 15 mL/minute, but as the glucose is absorbed from the PD cavity, the crystalloid osmotic gradient goes away. Eventually the glucose concentration in the peritoneal cavity equals that in the blood, and we no longer have a stimulus for ultrafiltration, so intraperitoneal volume slowly increases and the fluid absorption rate slows down as the glucose is absorbed. Once osmotic equilibrium is reached, we no longer have anymore ultrafiltration. We can change the amount of fluid that's ultrafiltered depending on the chronicity of the dialysate, and we get more ultrafiltration with a 4.25 than we get with a 2.5, and more with a 2.5 than we do with a 1.5, but it depends on the dwell time. So we have to be cognizant of that, and we have to alter the dwell based on the patient's transport characteristics in order to optimize ultrafiltration. The more permeable and the more rapid the patient diffusion transport characteristics are, the more rapidly we absorb glucose. Once this glucose gradient goes away and we no longer have osmotic stimulus for ultrafiltration, what we observe is the lymphatic absorption. Now lymphatic absorption is occurring in each one of us today, right now. It's a natural thing, but it also continues to happen during dialysis. So, we have direct lymphatic absorption in the subdiaphragmatic stoma that happens via the pumping action of the movement of the diaphragm during respiration, and about 80% of intraperitoneal fluid that is absorbed, is absorbed via that mechanism. We also have some fluid that's absorbed in the capillaries, and then the lymphatics uptake that fluid and bring it back to the heart. Any fluid that's removed by the lymphatics is removed by bulk flow and can be as much as 1 mL/minute. So in this cartoon I want to illustrate the 2 opposing processes. In blue we have the transcapillary ultrafiltration induced by our osmotic gradient from the glucose: As the glucose is absorbed, the rate of the transcapillary ultrafiltration decreases, and eventually stops when the osmotic gradient goes away. However, opposing that, we have lymphatic absorption, which varies based on body position; it varies based on intraperitoneal pressure, but for the sake of this slide, let's assume it's constant at about 1 mL/minute, so you'll see that the difference between the net transcapillary ultrafiltration and the net fluid absorption equals our potential ultrafiltration or drain volume. You'll see that once the fluid rates of ultrafiltration go away, we are likely to start absorbing fluid, and if we left the fluid in there long enough, we would actually drain out less fluid than we put in. So if we go back to our typical PD prescriptions, we can see that along with the fact that diffusion stops during these long dwells of APD and CAPD, with dextrose-containing solutions, ultrafiltration is likely to stop during those dwells also. If we want to be user-friendly for the patient, we don't want to do multiple exchanges during the day or overnight, and we would leave the fluid in. However, it may not be the most efficient way to remove small solute or remove fluid. These are typical curves that you might expect in an average patient with the various dextrose-containing solutions, so you can see here on the green curve that if you use the 1.5% dextrose, after about 260 minutes of dwell you might start absorbing fluid. If you want to get more ultrafiltration or have a longer dwell, you would then choose to use 2.5% dextrose, and you can see that that gives you a longer duration of ultrafiltration. And if you want even a longer dwell, such as what you might use with APD, that 15-hour daytime dwell, you would use 4.25% dextrose.

So back to the PET [peritoneal equilibration] test. Again, remember we classified the membrane based on the D/P creatinine ratios. Those patients who are rapid transporters, rapidly absorb glucose, their osmotic gradient does not stay very long, and so they would tend to drain out less fluid as seen in the middle part of this cartoon in orange, than the slow transporter who has slow rates of diffusion; the glucose stays in the peritoneal cavity longer; you have a better ultrafiltration profile. Now remember, clearance is related to the concentration of the solute in the dialysate and the drain volume, so if we think about these 2 things, and we look at this slide, we see that if you took a rapid transporter, the red lines in these slides, you can see that, yes, diffusion occurs very quickly, and they equilibrate very quickly. But their potential drain volume occurs early during the dwell. If you leave the fluid in there longer, you will start absorbing the fluid. And so that if you want to optimize their creatinine clearance per exchange, you would only want to do perhaps a 2- to 3-hour dwell; then remove the fluid; put some more fluid back in. In contrast with the low transporters with lower rates of diffusion, higher ultrafiltration potential, you see that to optimize their clearance per exchange, you would want to do a longer dwell. If you think about this, it would suggest that a high or rapid transporter would do better with multiple short dwells, and a low transporter would do better with equally spaced dwells during the day. Again, we need to remember that the potential ultrafiltration volume during a typical dwell depends on the patient's transport type. So if you have a high transporter using a 1.5% dextrose solution, the predicted drain volume at 4 hours would be much less than that of a low transporter. Again, think about the time period between 6 and 10 hours that the overnight drain would be for CAPD or the 12- to 14-hour period for APD, and in most patients who are high transporters, if you used 1.5% dextrose, you can see they would absorb fluid during those dwells, and so they would drain out less than they put in. That's why you would need to perhaps use a 4.25 where you see you get a longer potential ultrafiltration profile, and in most patients, even for 15 hours, you can get a little bit of ultrafiltration, markedly more for the low transporter than the high transporter, but some ultrafiltration can occur.

Because of this limitation of the dextrose solutions, new solutions have been developed, and icodextrin is one that is currently available in the United States. Icodextrin is a colloid osmotic agent. It's a macromolecule that is very slowly removed from the peritoneal cavity, probably removed via absorption into the peritoneal lymphatics, and once absorbed, metabolized ultimately to glucose. It's indicated for 1 exchange per day, and you would typically use that in the long dwell. What you tend to see is slow but sustained ultrafiltration. In some patients the rate of ultrafiltration is sustained for up to 15 hours, and most of the movement of the fluid is across the small pores -- very little fluid moving across the aquaporin, and so you would tend not to see sodium sieving. These are theoretical constructs for the potential drain volume of the different solutions. We've talked about the 1.5% dextrose, 2.5%, and 4.25% dextrose. The yellow line is the theoretical drain that you would see with dwells of icodextrin. Remember this is theoretical, and in some patients the rate of ultrafiltration slows down after 10 hours, but again, the potential for slow but sustained ultrafiltration. And also, in order to get about the same or perhaps more ultrafiltration at 12-15 hours than you would get with a 4.25% dextrose, you don't have that peak intraperitoneal volume at 4 or 6 hours that you do with a 4.25% dextrose solution. So this might be a little bit more user-friendly for the patient as far as the intraperitoneal volumes that they experience during the day. In contrast to dextrose, which is rapidly absorbed from the peritoneum and therefore the ultrafiltration varies based on transport type, with icodextrin you tend to see the same potential ultrafiltration profile whether the patient is a low transporter or a high transporter. Now that is a review of the physiology. Let's just spend a couple of minutes talking about clinical implications of this physiology of peritoneal dialysis. Now I want to focus on what path the solute in the water takes when moving from blood to dialysate. Remember water can move via the small pores and via the aquaporins. Solutes, depending on their size, move via the small pores and the large pores but not via the aquaporins. So let's speak about water for a minute. When you have a crystalloid-induced ultrafiltration, you have water moving across both the aquaporin and the small pores. In about 50% of the initial water that is removed comes via the aquaporin. Now if we were to look at what happened to the peritoneal dialysate sodium during a dwell, and we were to use the maximal ultrafiltration gradient, 4.25% dextrose, you could see that early in the dwell -- and this is the green line in this slide we're talking about -- when you have maximal water via the transcellular pores moving from the blood to the dialysate, your dialysate sodium concentration would decrease, that is by dilution. It's not to say that there is no sodium being removed at that time because, of course, there is. We have sodium being removed via small pores with the water that is ultrafiltrated across the small pores. But about 50% of the water that is removed, even more initially, is via the aquaporins, and so the intraperitoneal sodium will decrease. As that ultrafiltration slows down, and as you remember, eventually stops, now we have diffusion taking over. The intraperitoneal sodium concentration is lower than that of blood, perhaps 120 mEq/L, and now sodium continues to be removed by diffusion down its concentration gradient from the blood to the dialysate, and eventually the sodium concentration of the dialysate will be fairly equal to that of blood. So early in the dwell free-water movement occurs; later in the dwell sodium is removed by diffusion. Now this has implications in controlling blood pressure in achieving euvolemia, and you will hear about that in another talk that is concerning ultrafiltration failure and volume control in PD patients, but during the short dwells we remove relatively more free water, about 50% of the UF [ultrafiltration] volume. And so if we do multiple short dwells, we will cause hypernatremia, and the patients may become thirsty, so in some APD patients, if we're trying to do a lot of ultrafiltration overnight, remember half of that UF volume is free water, the patients become slightly hypernatremic and may wake up thirsty wanting to drink more water. So we have to be careful and try to maintain ultrafiltration throughout the day. Remember, during the long dwells the D/P sodium approach is one.

A brief word about middle-molecule clearance. Now we know that we can augment this in hemodialysis by increasing the UF rate, and this cartoon depicts what happens to urea, vitamin B12, or inulin clearance as you increase UF. So for instance, for inulin -- a large-molecular-weight solute -- as you increase the UF rate, you get a marked increase in the total removal per minute, and a lot of that is due to the ultrafiltration component, which is the purple part of this. In contrast, with urea, as you increase the UF rate, although you do get an increase in the amount removed by ultrafiltration, the incremental additional amount is very small because diffusion is so efficient to begin with. So, in hemodialysis we purposely try to augment middle-molecule clearance by using high-flux dialyzers, which have a high ultrafiltration coefficient. Now let's remember the various dwell profiles that we see in peritoneal dialysis, and particularly pay attention to the 4-hour time period. Notice here that with icodextrin the potential drain volume at 4 hours is less than what you would see with 2.5% dextrose or 4.5% dextrose. Similarly, at 8 hours, the draining volume with icodextrin is less than that seen with a 4.25% dextrose.

Now let's look at beta-2 microglobulin clearance, and on this slide I look at the mean values between removal after a 4-hour dwell with icodextrin vs 1.5% or 4.25% dextrose. Now remember the 4.25% dextrose solution is likely to have a larger drain volume than what you see with icodextrin. Nevertheless, icodextrin has more middle-molecule clearance. Why is that? Well, remember the pathways of solute removal. Fifty percent of the net ultrafiltration early in the dwell with dextrose-containing solutions is free water (ie, it came via the aquaporin), and no sodium, no urea, no middle-molecule can come with it. In contrast, almost all of the ultrafiltration volume with icodextrin is via small pores, and therefore, we are able to get relatively more middle-molecule removal with icodextrin for those dwells than we would with the dextrose-containing solutions. And in this it compares 4.25% to icodextrin for an 8-hour dwell, and again we see relatively more with the icodextrin because all of that fluid came via the small pores, and we have convection occurring. Whereas with 4.25%, again, remember half the fluid early in the dwell was via the aquaporins. The bottom line about middle-molecule clearance is we don't know for sure the effect on survival, but we do know there are some subtle data in the HEMO [Hemodialysis] trial, which suggests it may be important. We know that residual renal function has a survival advantage, and there is relatively more middle-molecule removal with that. So we think that this is an important thing to consider when writing a prescription for the patient. Therefore, because middle-molecule removal is so dwell-time-dependent, we feel that it's best that you have 24 hours per day of PD dwell when treating a patient with peritoneal dialysis. And although it's influenced by ultrafiltration volume, the route of removal is important. So in summary, solute removal occurs via diffusion and by convection during ultrafiltration. Remember it depends on the pathway of the water movement. Ultrafiltration is produced by utilizing a crystalloid, which is a hypertonic osmotic gradient, or a colloid, which is an isotonic osmotic gradient in peritoneal dialysis. The peritoneal membrane varies from patient to patient. It's best to characterize that individual patient's peritoneal transport characteristics by using a PET test, which I have described. In order to optimize a patient's outcome, a tailored dialysis regimen is needed, and this can be formulated based on knowing the patient's transport characteristics and the peritoneal physiology that I just reviewed. Thank you very much for your attention.

Indirectly supported through an educational grant for the Home Dialysis Resource Center by Baxter.

Medscape Nephrology © 2007 

Cite this: PD PHYSIOLOGY: The Basics and Comparisons to HD - Medscape - Mar 19, 2007.