DR. EVENEPOEL: I completely agree that we should avoid this kind of storm during the dialysis session. There are so many changes in the different ions that could trigger some arrhythmia. I completely agree that we should avoid important gradients, whether the ion is calcium or potassium. Now we’re discussing calcium. I think we surely shouldn’t use the dialysis time just to get rid of the calcium that we build in the time before the dialysis. That’s one point. Another point, the data we’re certainly missing so far are, of course, what is the calcium balance in dialysis patients, and especially what is the fractional gastrointestinal absorption in renal failure patients? Most estimates with regard to the latter are derived from calcium isotope tracer studies which, it should be emphasized, do have important inherent limitations.10 According to these isotope tracer studies, the fractional intestinal calcium absorption rates in dialysis patients was estimated at 25%. Tracer techniques may overestimate real calcium absorption substantially since they do not account for impaired calcium release from the food (digestion and solubilization), which may be especially relevant in pathological conditions such as chronic kidney disease (CKD). Again, I would point to the model being put forward by David in his paper where he modeled the calcium balance according to calcium intake and according to vitamin D intake.1 This is based on the assumption of a fractional gastrointestinal calcium absorption of 25%, which I just mentioned could be an overestimation. If it were truly less then, again, calcium balance wouldn’t be that positive, even in the intradialytic time period. We should, again, be very careful in stating that the calcium balance is for sure positive in dialysis patients on 1,25D therapy when they’re consuming, let’s say, one gram of elemental calcium. I think this is certainly something that should be put in with a question mark.
DR. BUSHINSKY: Absolutely. If I might interject, there are now several studies, one by David Spiegel,11 and a second by Katie Hill,12 and they both show basically the same thing. David did a formal calcium balance study in six patients with CKD and six healthy controls, and compared those on an 800 mg per day calcium diet and those on a 2000 mg per day calcium diet. Those on the lower calcium diet were in neutral calcium balance. These are CKD patients, so they have some renal function, and those on a high calcium, or 2 g calcium intake, were in positive calcium balance. Katie Hill demonstrated essentially the same results in another study. She took eight patients with CKD, put them on about 950 mg of calcium, and supplemented 500 mg of calcium carbonate or placebo three times per day. Thus the patients received about 950 mg versus about 2450 mg of calcium a day. Again, on the lower calcium diet, the patients were in neutral calcium balance. On the higher calcium intake, the addition of 500 mg of calcium three times per day, led to a marked increase in calcium balance of about 500 mg per day, suggesting the calcium absorption was about 33%, even higher than the 20% that Jack Coburn found and that we used in our mathematical modeling. If anything, Katie Hill found more calcium was absorbed, approximately 33%, with these stage 3 and 4 CKD patients than we normally assume.
DR. BLOCK: There are a couple of interesting issues here. One is, we don’t really know with modern patients what the calcium balance is because we don’t really know the fractional absorption in the setting of CKD with or without vitamin D therapy. The second issue is, David, that you so beautifully, I thought, wrote in both of your papers,1,2 that regardless of balance, the kinetics of where calcium is moving with regard to extracellular fluid calcium concentration, because ionized calcium is so tightly regulated, that the kinetics of where the calcium is moving is probably equally as important as whether patients are in neutral, negative, or positive calcium balance. To take a very simplistic approach to just trying to achieve so-called neutral balance may tell you nothing about where that calcium is actually living within the individual.
DR. BUSHINSKY: Thank you, Geoff. Let me just briefly comment that the typical physiologic environment of our patients, the metabolic acidosis, the administration of 1,25 D or analogs, and their high PTH all move calcium from the bone, where we want it to be, to the soft tissues, where we don’t want it to be.1,2 You’re absolutely right. In our CKD and dialysis patients, not only might they be in positive calcium balance, but the calcium is maldistributed. More is in the soft tissues, and less is in the bone.
DR. BLOCK: I think we’re all aware of—and I think pretty impressed by—the work of some of the basic scientists, like Catherine Shanahan and Giachelli,13 who have shown that intracellular calcium content of patients with CKD, even pediatric patients with CKD, is staggering. Long before we can see or detect the calcium accumulation in soft tissue, the intracellular calcium content is accumulating tremendously.
I want to move for a second over to you, Allan. Given your incredible experience running the USRDS and now with the Kidney Peer Initiative, I would like to hear your comments on this analysis by Steve Brunelli,14 who actually looked at it in a retrospective way. Because this notion of low dialysate calcium gathered some momentum, there was a number of facilities within the large dialysis organizations that actually switched the majority of their patients in their facility to a low dialysate calcium. I was really impressed with this retrospective analysis, of looking at outcomes in facilities that switched to a low dialysate calcium compared to those who stayed on what is generally recommended, the 1.25 mmol/L calcium bath, because it actually supports exactly what you and David and Pieter were just saying with regard to what happened with arrhythmogenicity. What are your thoughts on that analysis?
DR. COLLINS: What I liked about what Steve did is he looked at the causal path to say all right, if this calcium has a multiplier effect by just what we were talking about, through the electrolyte abnormalities that are going to occur by the end of the run—and we’re discounting the bone and mineral issues here—but if it really does have this causal path, then not only should we see all cause mortality, we should see very specific types of morbidity from these arrhythmic events. What did we find? We found exactly that. I said, “Gee, that really sort of gives you a wake-up call that this fooling around with the calcium and the bath has to be put in the context of, what are the overall things that are going on that create the major morbidity and mortality?” Of the hospitalization events, atrial fibrillation is probably number three now as a cause of cardiac hospitalization, beyond heart failure and acute myocardial infarctions, we’ve got atrial fibrillation that occurs. These QT prolonging electrolytes, low electrolytes to prolong the QT interval, creates this conduction, accelerated conduction abnormalities in the heart as opposed to, we’re so used to the high potassium side of the equation, which actually moves to asystole. All these arrhythmic problems that we actually have may be traceable to this combination of too aggressive use of low calcium.
One other element I’d like you to consider—and David, you’d probably want to comment on this as well—is the amount of ultrafiltration that’s going on during the dialysis treatments and the amount of solute drag that’s occurring with the Donnan effect is also going to affect how much ionized calcium is coming across the membrane. With the European dialysis population—where 30% of them are now on hemodiafiltration (HDF)—where there’s a lot of flux that’s going on across the dialyzer, 25 years ago, when we were doing hemofiltration quite a bit, we got into pretty substantial calcium balance problems when we were using convective transport to cross the membrane. Obviously, ultrafiltration itself is a convective transport system. Do you want to comment on that, David, on how that plays a role in the net calcium balance here?
DR. BUSHINSKY: Absolutely, Allan. When we did our analysis in 2010, we recognized—as you so beautifully said—that every time you remove a liter of ultrafiltrate, you’re removing the calcium with that liter of ultrafiltrate.1 We built that into the model, and we routinely ultrafilter 3L of fluid during each dialysis session. That’s a fair amount of calcium that we’re removing. If you couple that with a low calcium dialysate, you’re removing a substantial amount of calcium from the body, and this calcium is a signaling molecule. You can’t look at calcium as, for example, you look at urea, where, at the end of the week, you want the balance to be neutral. You can’t have these large swings of blood calcium that, if it’s low, will stimulate PTH and arrhythmias and if it’s high, it will suppress PTH and also stimulate arrhythmias. You have to maintain a signaling molecule relatively constant throughout the week.
DR. BLOCK: Pieter, that’s a perfect segue to you because I think HDF is quite a bit more common in Europe. What are your thoughts on this since, as David said, we can’t make sure calcium transfer and mass calcium over the course of a week are neutral. What do you think of that from a European perspective?
DR. EVENEPOEL: Of course, the European perspective is that HDF is much more common than in the US with a penetrance of about 40% if I’m correct. I completely agree with the point that calcium can be exchanged not only by diffusion but also by convective transport. Studies evaluating calcium balances during HDF are, to the best of my knowledge, not in existence, but you can imagine that net losses may be substantial when using a low calcium dialysate in the setting of predilution hemodiafiltration with convection volumes exceeding 20 L. It’s something certainly that is a lack, at least in my opinion, of what’s really going on with regard to calcium balance or even just looking at calcium levels, ionized calcium levels in HDF patients as compared with hemodialysis (HD) patients, but surely that’s something that should be investigated into more depth. I completely agree this is something that should not be neglected, the amount of calcium being lost just by means of convective transport. I don’t think there are any data that are really giving us some insight in this topic.
DR. COLLINS: You’re right. I’m not sure that I’ve heard much from our European colleagues with this HDF. What exact calcium bath is being used?