Sophie Jamal, MD, PhD, FRCPC: Good afternoon everybody, we're just going to get started if that's okay? And I'm going to welcome you to this lunchtime session looking at bone, intestinal, and vascular renal axis.

This educational activity is supported by a grant from Sanofi Renal.

What Dr Spiegel and I would like to do is describe the pathophysiology of calcium homeostasis in chronic kidney disease (CKD) compared with the general population, talk a little bit about its relationship to vascular and soft tissue calcification, apply calcium modeling to clinical situations to understand the emerging science of calcium balance, and analyze current mineral management options in patients with CKD. And then also talk about the available mineral approaches in the early management of patients with CKD to try to establish a rational approach to calcium use.

My name is Dr Sophie Jamal. I'm an endocrinologist. I do kidney research, and I'm an Associate Professor of Medicine at the University of Toronto, and the Director of the Osteoporosis Research Program. I'm a clinician scientist and my main interest is in assessing fracture risk in CKD and looking at vascular calcification in these patients.

Dr Spiegel is a Professor of Medicine, Director of the Chronic Kidney Disease program at the University of Colorado, and he's also an Attending Physician at the University of Colorado Hospital and Medical Director of the FMC Dialysis Center. Dr Spiegel’s research and clinical interests are focused on CKD, particularly the care of patients with end-stage kidney disease, and he's authored more than 50 articles and book reviews.

So hopefully between the two of us we can give you some practical approaches to the renal vascular axis.

To start with the case then, this is a 55-year-old postmenopausal woman. She has autosomal dominant polycystic kidney disease (PKD), and she asks you about calcium supplementation to reduce her risk of fractures. Her hypertension is currently being controlled with medication, her family history is significant for her mother who had a hip fracture at 62, and the patient weighs 159 lb and is 5' 5" tall. Her laboratory findings are: a renal panel that's within normal limits, a dual X-ray energy absorptiometry (DXA) scan that showed a T-score at the lumbar spine of -1.8 and at the total hip of -2.6.

The question is, how would you answer her regarding her question about the prevention of fractures? So do you think it's A: Recommend a calcium intake of 1200 mg/d; B: Begin vitamin D supplementation; C: Begin an exercise program; D: Begin bisphosphonate therapy; E: All of the above; F: A, B, and C; and then G: B, C, and D.

I'm going to go onto the next issue. What current medical concerns might influence your decision regarding calcium supplementation in this patient, this woman with PKD? So, are you worried about her PKD (A)? Are you worried about her cardiovascular risk (B), fracture prevention (C) or about the first 2 (you can pick D)? E for B and C; or F: All of the above.

So 5 years later, then, you see this woman and follow up, and her laboratory findings are the following: her creatinine is 2.5 mg/dL, with an estimated glomerular filtration rate (eGFR) of 25 mL/min/1.73 m²; her calcium is 9.4 mg/dL; phosphate of 4.6 mg/dL, and parathyroid hormone (PTH) is 145 pg/mL. Her current medications when you see her 5 years later: calcium carbonate 650 mg, vitamin D 1000 IU daily, and she takes an angiotensin-converting-enzyme (ACE) inhibitor and a beta blocker for high blood pressure. Her last chest X-ray revealed that her lungs were normal, and there was no evidence of any fractures. But she did have calcification in the aortic arch.

The question for you now is how would you manage this patient at this time? A: Continue her calcium supplementation at a reduced dose; B: Stop calcium supplementation and prescribe a calcium-based phosphate binder; C: Stop calcium supplementation and prescribe a non–calcium-based phosphate binder; D: Stop calcium and prescribe an active vitamin D agent; E: If you want B and D; and F: None of the above.

What I'm specifically going to talk about—and hopefully it'll help you to think back to this case and answer the questions—I'm going to talk to you a little bit about who needs calcium, how much calcium patients need, and, why we actually take it or need it. I'm going to talk to you about the benefits and risks of calcium supplementation, and I'm really going to give you the view of the non-nephrologists and I'm going to focus specifically on vascular calcification.

Interestingly enough, there's actually no current biochemical measurement that reflects our nutritional calcium status. So I often see people in the clinic, and they'll say to me, "I have osteoporosis. Check my serum calcium and see if it's okay.” That is actually not a helpful test to assess calcium balance. We do have an indirect measure, which is skeletal health, so we look at bone mass and fracture rates as an idea of someone's calcium balance, and these results provide adequate and upper intake recommendations for calcium.

In terms of calcium nomenclature, I think it's just important to remind you all that when we talk about elemental calcium, that's what's available for absorption. If someone is taking a supplement and it has 500 mg of calcium, that's usually going to be elemental. All calcium from food sources is elemental. So I noticed you're having a nice dessert, and I'm assuming it had a bit of milk in it; maybe there were 300 mg of elemental calcium. In supplements, some calcium is elemental, but not necessarily all of it. So your patient actually has to turn around and look to see how much calcium they're getting. And the other point that I want to make is in men and women without CKD, non-CKD individuals, the calcium absorption varies by the dose that's given, and it's really thought that at doses greater than about 600 mg, there's minimal gut absorption, which is why we really don't give more than 600 mg at one time. That's in otherwise healthy men and women.

The Institute of Medicine recently published some recommendations on calcium intake and these have been modified in response to the most recent literature that I'll take you through. What they say is that for 19‒50 years, whether you're a man or a woman, the RDA (recommended daily allowance) is about 1000 mg. For men up to age 70, it's also 1000 mg. For women 51‒70 years, it’s 1200 mg. Then as you get older, in both men and women over 70, it's 1200 mg. There's no increase if you're pregnant or lactating, so that's actually very different than what we used to recommend in terms of calcium.

Why do we give calcium? Do you know? Well, I think calcium is an essential nutrient, it is important in the mineralization of bone, and, theoretically, it increases bone density and decreases the risk of fractures. And I just want to emphasize that in otherwise healthy men and women, we're not giving calcium for hypocalcemia. So our use of calcium and calcium supplements is really from a bone perspective.

And even in patients with CKD, I want to emphasize that hypocalcemia is extremely rare. You'll hear a little bit about this from Dr Spiegel, but patients with CKD, generally speaking, are in a positive calcium balance. They have little resorption of bone, so they have little breakdown, and they have no urinary calcium excretion. And they have increased PTH due to increased phosphate, which will also increase their calcium. So overall, it's rare. I would say the only exception to not seeing hypocalcemia is in patients with stage 5 CKD who have very low levels of 1,25 (OH)₂D (1,25 D). And in those patients occasionally we can see hypocalcemia. But it is not common.

I told you that we give calcium because it's important for bone, and it has some modest effects on bone density. How does calcium actually work to improve bone density? And it's not 100% clear, but I think that the generally accepted rule or thought is that calcium increases bone density because it's a weak antiresorptive. So what does that mean? It actually prevents bone breakdown, bone resorption. And it also slightly modifies bone formation.

Here is an example of some of the data, just looking at alkaline phosphatase. You can see on the left-hand side there is the placebo group, and on the right is the calcium group, and that's the change in alkaline phosphatase. Alkaline phosphatase represents bone turnover, and you can see that the group that's given calcium has a greater decrease in alkaline phosphatase level. So this group is having greater suppression of bone turnover. That's why we think calcium works in terms of bone density.

So, calcium's good for your bones, right? Everyone's convinced?

What about calcium supplementation in children? The Cochrane collaboration is a large group of epidemiologists that are based in Hamilton. Every once in a while they will pick topics that are of important clinical interest and conduct systematic reviews. And this was a topic that obviously came up: Should we be giving calcium supplements to children? And this was published in 2006. They looked at all the data for benefits and harms of calcium supplementation in children, and this is what they concluded: While there is a small effect of calcium supplementation in the upper limb, the arm, the increase in bone density that results is unlikely to result in a significant decrease in fracture risk. And, they concluded that the results do not support the use of calcium supplementation in healthy children. So there are no data that supplements in children are helpful.

What about adults? What about postmenopausal women, for example, with osteoporosis? So these are some data looking at calcium intake by vitamin D status. What you see on the left is vitamin D levels < 50, the middle group is 50‒74, and the upper grouping is > 75. And what you can see is regardless of the vitamin D level, calcium really doesn't have a huge effect on bone density. The lowest quartile of calcium is in the blue there, with the highest quartile in the yellow, and you can see that if you look by different levels of vitamin D, there's really no difference in bone density based on calcium intake.

What about fracture risk? So this is looking at the total calcium intake. The reference range is just assuming none, and then calcium intakes between 500‒700 mg/d, 700‒900 mg/d, 900‒ 1100 mg/d, and > 1100 mg/d. Remember, we're recommending 1200 mg/d of calcium for our patients, and what you can see is that the relative risk of hip fracture is not really related to the calcium intake. So there's no decrease in hip fracture whether you take 1100 mg/d compared to no calcium.

You will say those are only a couple of studies. Well, here is a meta-analysis that was published about 5 years ago or so looking at 12 studies that looked at calcium and vitamin D on preventing fractures. If you go right down to the bottom, you'll see that overall, there is not really a benefit, that diamond is favoring control in terms of reducing fractures by taking calcium and vitamin D. So, there are no data that calcium and vitamin D are helpful in that regard.

I told you that maybe it's not that helpful. Perhaps it reduces bone resorption. It doesn't really seem to have a lot of effects in children, it doesn't seem to have a lot of effects on bone density, and we’re not convinced about the fracture. But you buy it at the health food store, you get it in Tums, so it really can't be bad for you, right? It has to be, at least, innocuous.

So what do we know about the harms of calcium supplementation? This was a study that was published about 6 years ago now in The New England Journal of Medicine looking at data from the Women's Health Initiative Study on calcium supplementation and bone mineral density. What they concluded was that among healthy postmenopausal women, calcium and vitamin D supplementation resulted in a small but significant improvement in hip density, there was no difference in hip fractures, and more concerning, there was an increased risk of kidney stones in these patients. So we're seeing maybe a little bit of harm.

I think it's important to look at those patients in the Women's Health Initiative study and the data that they reported. They enrolled about 40,000 women, and women were allowed to take, regardless of what arm they were in, 1000 mg of calcium and 600 IU of vitamin D. They were also allowed to take treatments for osteoporosis, so bisphosphonates (BP), calcitonin, the systemic estrogen receptor modulators (SERMs), and raloxifene. So what you had then in your calcium group was women who were given 1000 mg, plus 400, but they were taking or could be taking up to 1000 mg additionally. So your calcium group was now at 2000 mg of calcium, and your placebo group, or your no intervention group, could be taking about 1000 mg of calcium. So really, what we're comparing is extremely high doses of calcium with modest doses of calcium, and in that extremely high dose, the 2000 mg/d, we saw an increased risk of kidney stones. But, more importantly, we did not see an improvement, substantially, or reduction in hip fractures.

Then this study came out a couple of years later, from the British Medical Journal. Vascular events in healthy older women receiving calcium supplementation: a randomized, controlled trial. Interestingly enough, the investigators for this study had found in the literature that people who took calcium had lower blood pressure, had better HDL cholesterol levels, and they had lower LDL cholesterol levels. So their hypothesis was that if we randomized women to calcium, they would have fewer cardiovascular events than women who were randomized to placebo. And they actually found the opposite. What they found was that calcium supplementation in postmenopausal women was associated with an upward trend in cardiovascular event rates. They actually concluded, and this was probably the first time that someone came out and said this, publicly, that the potentially detrimental effect of calcium should be balanced against the likely benefit of calcium on bone.

And again, I just want to highlight what these women in the study were taking. So there were 1500 women in that BMJ study, they were 55 years and older, and they had no evidence of renal impairment. So this is different than the CKD population. They were randomized to placebo or 1000 mg of calcium citrate in divided doses. But, they were allowed to take dietary calcium. And they were actually really good at taking dietary calcium to the point that the placebo group was taking about 800 mg of dietary calcium a day, and the treatment group with their supplement was taking a total of about 1800 mg of calcium. So again, we're talking about fairly high doses where we're seeing these potentially detrimental effects.

This is what they really found: over the 5 years of follow-up, the group that received calcium in the blue had a higher risk of verified myocardial infarctions (MI), compared to the placebo group in the green.

That was 1 study, and of course the criticism with this 1 study was: Is it generalizable? Was it just a fluke? This was such an unusual finding, and it generated a lot of controversy in the literature. And these same authors then went on to look at calcium supplementations and MI using a meta-analysis technique. So what they did was they went back and identified 15 trials that had looked at calcium supplementation. I just want to point out to you that none of these trials were designed with an endpoint of MI, so all the randomized trials of calcium looked at other things. So they went back and pulled whether these people had happened to have an MI as well. These people had about 3.6 years of follow-up, and the primary endpoints that they were going to look at in this meta-analysis were MI; stroke; a composite with MI, stroke or sudden death. The secondary endpoint that they had decided to look at was all-cause mortality.

And this is their meta-analysis, really just looking at MI, and I want to point out the studies that they included. You can see that in all the studies, their significant line is favoring placebo. If you look at the total there was about a 27% increased risk of MI if you were randomized to calcium compared to placebo. That was just statistically significant. The rest of their outcomes were not statistically significant, but they were trending in the way that maybe calcium was harmful.

If you look at that from another way, which is the number needed to treat or the number needed to harm, you would need to treat 69 people for 5 years, and you would cause a heart attack. If you treated 100 people you would cause stroke, death; in 77, and you would prevent fractures in 39. And treatment of 1000 people with calcium for 5 years causes 14 MI, 10 strokes, 13 deaths, and prevents 26 fractures. So you can think of the benefits and the harms, I think that's a good way to look at it.

One of the issues that we're not going to spend a lot of time on is the argument that a lot of people have made: the issue with calcium is that it is detrimental to the cardiovascular system if you don't take it with vitamin D. So there is this concept that a lot of what we're seeing in terms of cardiovascular risk with calcium is because it's not being given with vitamin D. And if we took calcium with vitamin D, we wouldn't have this issue. So this was a meta-analysis that actually looked at vitamin D and calcium to prevent cardiovascular events, and this was published a couple of years ago in The Annals of Internal Medicine. So, they looked at 3 separate groups of patients: they looked at a group of patients just taking vitamin D, and looked at their cardiovascular risk; they looked at a group of patients taking calcium alone and their cardiovascular risk; and then they looked at studies that included patients taking calcium and vitamin D together and looked at their cardiovascular risk. And I'm just going to take you through those slides.

This is the first one, looking at vitamin D supplementation and cardiovascular risk. There were only 2 studies that reported on it, and what you can see is that there was really not a benefit or a harm in terms of vitamin D supplementation or placebo in terms of cardiovascular risk. The relative risk was 0.9 and the odds ratio crossed 1.

What about calcium supplementation? So if you look at just calcium supplementation and cardiovascular risk, there are 3 studies that were looked at. Right at the bottom you can see the odds ratio, and again it's crossing 1, with 1.14 increased risk. But it's not statistically significant.

What about vitamin D and calcium? So, theoretically, if you took vitamin D with calcium, according to the argument, you would have a lower cardiovascular risk. And here you can see there were only 3 studies that were reported, and you can see that really, again, the relative cardiovascular risk was 1.04, and again, it crossed 1.

After that study was published, there was a re-analysis of the Women's Health Initiative data also looking at calcium, vitamin D, and cardiovascular risk. So this was really to try to get larger numbers of patients to see if we could be more definitive in terms of whether calcium causes harm or does not cause harm. Again, I'm just going to show you the cardiovascular risk data.

This looked at 3 trials, larger numbers: 20,000 patients. You can see that if you go right down to the bottom and look at that blue line, that if you take calcium, you have a slightly increased risk of about 60% of having a cardiovascular event based on the meta-analyses. I want to remind you that these are healthy, generally speaking, postmenopausal women, not with CKD. None of the primary outcomes in these studies were meant to look at cardiovascular disease, typically they were looking at bone outcomes, and the patients were taking calcium supplementation.

If we look at the data I just showed you from the Women's Health Initiative, treating 1000 people with calcium for 5 years causes 4 MI, 4 strokes, and 2 deaths, and it would prevent 3 fractures.

So, I don't think we're doubting that it prevents fractures; I think the issue is the other harm that may be associated with it.

Not surprisingly, or perhaps surprisingly, the calcium intake values have been lowered in response to the recent data. So there's been a lot of controversy about the fact that calcium is harmful. Particularly in high amounts, it doesn't seem to be helpful.

I showed you the recommended daily allowance when we started. It's 1000 mg for men 19‒70 years and for women 19‒50 years. Then for women > 70 years, it's 1200 mg. Just to remind you, there's no comment and no recommendation on what to do in patients with CKD. They don't address that at all in the Institute of Medicine recommendations.

Osteoporosis Canada—and I'm a member of their scientific advisory council—reviewed all the literature and updated guidelines in October of 2010. We actually said no more than 1200 mg of calcium a day from diet and supplements, and we specifically said no other populations, such as patients with CKD. They really are beyond the scope of this recommendation, so nothing specific for patients with CKD.

The American Society of Bone and Mineral Research (ASBMR) was probably the only group that said anything about patients with CKD, and they posted this on their website in August 2010. I think they were compelled to do it because they were probably getting deluged with phone calls and emails and questions about what to do in terms of calcium supplementation. They summarized a few things. I think the first thing they said is there's been lots of large studies of calcium and vitamin D that do not show an increased risk of cardiovascular events, and you don't necessarily need to discontinue the use of calcium. You should talk to your physician and decide how much calcium you need. Then they say the beneficial effects of calcium are found in relatively low doses, so more is not necessarily better. Then the last point, which I've highlighted and I think is key for us here, talking about kidney disease, is they say that elderly individuals and others with impaired renal and kidney function who take calcium supplements may be at higher risk for cardiovascular problems. That's the only comment that is available in the literature about that.

Okay, what about you guys? What do the National Kidney Foundation/Kidney Disease Outcomes Quality Initiative (NKF/KDOQI) Clinical Guidelines say? Elemental calcium doses of 1.5 g/d—so it's higher than what Osteoporosis Canada is recommending, higher than the Institute of Medicine—or total calcium of 2.5 g/d. So it's double what Osteoporosis Canada is recommending for osteoporosis.

I've showed you some epidemiologic data. I'm convinced that it's not really beneficial to take more calcium, and taking a lot of calcium may be harmful. Why would that be?

Well, there are some data actually that demonstrate that serum calcium is associated with vascular disease, and this is just looking at a study of older women; a cross-sectional study looking at mean plaque thickness, by quintile of calcium. On the left-hand side you can see the lowest quintile of calcium, and those are values between 2.05 and 2.12 mmol/L, and then it's compared with the group with calcium in the highest quintile, 2.28-2.6 mmol/L. You can see that the mean plaque thickness is higher in patients who have higher serum calcium, so cross-sectional study, but certainly there is evidence that there may be something biologically plausible going on.

This is a study looking at serum calcium and cardiovascular events in the multiple outcome raloxifene study. This was a study of Evista^®, a SERM, just to look at fracture, and they looked at serum calcium and they found that if serum calcium was high, you were much more likely to have an increased risk of stroke, MI, or death. The risk was increased by about 17% and the high here was defined above the upper limit of normal.

We also have some data that show mortality is related to serum calcium intake, again, in normal patients not with CKD. These specifically are men, and what you can see on the Y axis is the odds ratio of death, and then serum calcium. And you can see that as serum calcium increases, the risk of death increases as well.

So, calcium supplements may very well be harmful because they increase serum calcium. Right? And, we've shown an association now between serum calcium and vascular disease; not a causal mechanism, an association.

And here are some data demonstrating actually that calcium does indeed increase serum calcium. So, this is over a 24-hour time period. The first 8 hours are elongated in terms of serum calcium over time. The yellow are patients given calcium carbonate, the diamonds are calcium citrate, and then we've got a control and potassium citrate. Obviously the control and the potassium citrate shouldn't raise your serum calcium, and you can see that when patients are given the calcium carbonate or the calcium citrate, they actually do have an increase in serum calcium levels that happens at around 4 hours post-dosing. It goes back down to normal, but it does increase above the normal range. That normal range is indicated by the dotted line, and that's ionized calcium.

To summarize, in terms of calcium supplementation and heart disease, we know that within cohorts a high normal serum calcium is associated with increased carotid artery plaque thickness. There is evidence that there is increased abdominal aortic calcification. We also know that there's an increased incidence of coronary heart disease and stroke, and there's an increased incidence of mortality.

In conclusion, I think calcium is thought to be beneficial by reducing bone resorption, and there are some weak data that I think would support that. The effects of calcium on fracture reduction remain uncertain, and I think the key point is, in otherwise healthy patients, there's really no evidence that more calcium is better. So I'll often see patients in my clinic, and their bone density is not improving or they've had a broken bone. Their natural response is to take 4 times the amount of calcium that they were taking before. That's probably not a helpful thing to do, and I think excess calcium supplementation can increase the risk for cardiovascular events.

So, where does that leave us? I think when we're seeing general patients without CKD, it's very reasonable to aim for total calcium intake of about 1000 mg/d. There are some data, fairly weak, that food sources of calcium are preferred; we may be better at absorbing calcium from a food source than a supplement. But, I think the key thing is, if we're worried about preventing fractures, we're really thinking about something other than calcium and vitamin D. If you want to prevent fractures, you need to use treatments where we have proven anti-fracture efficacy, and calcium is not one of those.

So I'm going to end there. I will have Dr Spiegel come up and talk to you. Thank you very much.

David M. Spiegel, MF, FACP: As you can see, no one really talked about CKD, so I will. No one made recommendations about CKD, so I will.

This is a controversial area, and I thought I'd start with a few things that we can all agree on. I think we can all agree that calcium balance is poorly understood in CKD. I think we all agree that adequate calcium intake though is important for normal bone health. Patients with CKD though develop vascular calcification and renal bone disease. Both of those processes are complex, and both of them are relatively poorly understood. We heard in the general population, certainly, that throwing more calcium at someone to try to improve their bone disease doesn't help it, and we don't have all that information in CKD, but I think the bulk of the evidence suggests that throwing calcium at patients with CKD to improve their bone health doesn't necessarily help their bone health, either.

So what I want to do today, in the rest of the time we have, is look a little bit at some of the physiology, some of the data we have in terms of calcium absorption in patients with CKD, a little bit on urinary calcium excretion in patients with CKD, and then some of the new data that are coming out on calcium balance in patients with CKD.

In terms of calcium absorption, we know a lot about the physiology of calcium absorption. We know, for example, that if you take in about 1 g of calcium a day, which is now recommended for most of us, a little higher for women, then, on average, about 20% of that calcium is absorbed. We'll go into this in a little more detail, in terms of the role of 1,25 D in that absorption, and that calcium moves into the extracellular space. You can see if you take it in as 1 g at a time, there's a small rise in the serum calcium level. So if 20% is absorbed, about 80% comes out in the stool. Of the calcium that's in the extracellular fluid, the ionized calcium at least is filtered, and the kidney reabsorbs most of that calcium so that we put out about 200 mg/d. So you can see we absorb about 200 mg, we put out about 200 mg, and we're in balance. Honestly, the skeleton plays a very important role in terms of regulating the serum calcium and calcium also plays a very important role in terms of maintaining skeletal health. And then you and I for the most part as we get older, we worry about it, but we don't need to think about it a lot because if all things are working well, that balance is very well maintained.

So what happens in CKD? And we know some of this. I'm going to try to convince you today that the gastrointestinal portion is relatively intact, and I'll show you that in a little bit, but what we know is that when you have CKD, the filtered load of calcium goes down. I'm going to show you that urinary calcium in you and me may be 150-200 mg/d. In patients with CKD, urinary calcium is very low, 50 mg/d on average. What I'm going to try to convince you of is that CKD results in a fundamental disruption of this normal regulation of extracellular calcium, bone calcium, and vascular calcification.

And we certainly don't understand it completely, but that's one of the hallmarks of CKD: the disruption of this normal homeostatic mechanism that relates to the vasculature, the bones, and extracellular calcium.

Let's talk a little bit about calcium absorption. So this is the gut, and we know that calcium absorption occurs through 2 mechanisms. There is an active mechanism that results in transcellular movement of calcium. Most calcium is absorbed in the small intestine. The active transport is regulated, and the primary regulator is 1,25 D, which is saturable. And 1,25 D, this regulated active transport of calcium, is critically important when dietary calcium is very low. There is also a paracellular calcium movement in the gut and it is not highly regulated, it is not particularly 1,25 D dependent, and as far as we know, it is not saturable. Dr Jamal mentioned that the maximum you can absorb is about 600 mg/d, but when you look at most of the studies, nobody gave normal individuals 4 g of calcium with a meal to see how much was absorbed. Most of those studies are giving 1000 or 1500 mg with a meal.

Only in CKD do we give doses of 2, 3, or 4 g of calcium with a meal, and those doses really haven't been studied in the general population.

So we have these 2 mechanisms of calcium absorption, and they've actually been elegantly looked at by Dr Fordtran in patients with CKD. Let me walk through a little bit of how he did this study. So he took a very small number of patients: 5 normal subjects and 4 subjects with end-stage renal disease (ESRD). He studied them before giving them calcitriol. Then, after supplementing them with calcitriol—and the way he did this study was really very elegant, my guess is none of us would want to do the study—what he did basically was he gastrointestinally intubated these patients. He put down a feeding tube and washed out the gut, and then he gave them 1 of 2 meals: 1 meal contained 120 mg of calcium and the other contained 300 mg of calcium. Then he gave them an oral tracer.

Then he gave them the meal, and then he gave them another oral tracer. He collected what came out between the 2 tracers. So, knowing what went in and what came out, you can know how much calcium was absorbed between these 2 meals. One meal had 120 mg of calcium, and the other had 300 mg of calcium. So this is not a balance study, it's kind of an absorption study.

And these are the data: so on the left-hand side is the vitamin D‒dependent calcium absorption, and he was able to ferret this out because the CKD patients who were unsupplemented had very low 1,25 D levels. He knew the 1,25 D levels in the normal people and then he supplemented the CKD patients to raise their 1,25 D levels and restudy them. So what you see here, where this red arrow is, is that when there's no 1,25 D, active calcium absorption is essentially 0. And as you increase 1,25 D levels in the gut, active calcium absorption increases. You can see the 1,25 D levels on the X-axis here. You can see that when you get very high 1,25 D levels, whether you're taking in 120 mg of calcium with your meal or 300 mg of calcium with your meal, you can absorb a lot of that. You can absorb up to 80% of the calcium in the meal if you have enough 1,25 D.

On the right hand side you see what he calls true net absorption, which is D-dependent absorption, plus passive absorption. And, first of all, look at a couple of things at the baseline. When 1,25 D levels are 0—you can see the yellow line, which is the 120 mg calcium meal—the net absorption is actually negative, because when there's no 1,25 D, there's no active absorption. And passive absorption is actually negative because there's calcium excretion into the gut. So when calcium intake with the meal is very low, you can actually lose calcium through the gut. If you give 300 mg of calcium, which is the white line, even with no 1,25 D, you see the calcium absorption is positive. Even though there's no active absorption, there's enough passive absorption with the meal of 300 mg that you are in positive calcium balance with the positive calcium absorption from that meal. Then obviously as 1,25 D levels go up, the active absorption increases, and the percentage of calcium absorbed from those meals increases. So, when you have adequate 1,25 D levels, whether you're taking in 120 or 300 mg with the meal, you're getting net calcium absorption from those meals.

When he looked at it in total, now looking at a bar graph, what you see is in the untreated hemodialysis patients, their 1,25 D level was not quite 0, it was about 8. And what you can see on the left-hand set of bars is, if they took in 120 mg, their D-independent absorption was negative; they lost about 30 mg through this passive diffusion into the gut. The D dependent was slightly positive, they had a little bit of D, but their net absorption, which is this white bar, was actually negative. So if our unsupplemented dialysis patients eat a meal that only has 120 mg of calcium, if they did that over 3 meals, that would only be about 360 mg/d of calcium. But they would be losing calcium through their gut. If they ate a meal with 300 mg of calcium, and if you add that up over 3 meals, that's about 900 mg/d. You can see the white bar again, that their net calcium absorption is positive. If they eat a meal of 300 mg, they get a net absorption of about 50 mg. That's even true in hemodialysis patients with very low 1,25 D levels.

The next set of bars are the normal individuals. You can see their 1,25 D levels are normal, about 48 pg/mL, and in normals, even on the 120 mg meal, their passive absorption is also negative, because of the fluxes into the gut. But their active absorption is much more positive because they have 1,25 D around, and then net absorption is also positive. This is why people with normal 1,25 D levels can be in neutral calcium balance on very low calcium intakes, because of the role of 1,25 D in increasing gut calcium absorption.

The last set of bars then is the dialysis patients who are supplemented with calcitriol. He supplemented them for at least 1 week before repeating these studies. He raised their 1,25 D levels to 75 pg/mL. In hemodialysis patients on the 120 mg diet, they have markedly positive absorption, about 100 mg/d from that meal, and on the 300 mg diet they also have significantly positive absorption. So when we take our dialysis patients and supplement them with active D, we essentially take their gut calcium absorption from a dialysis patient level to a normal individual level. And I think those data are very well shown here by Dr Fordtran.

He then took some more patients, and he did the same kind of procedure with them. You can see the patients are listed on the left-hand side. They were on different diets, and he measured how much calcium was in those diets so he knew how much calcium they were taking in with this meal. He measured their D levels, and then he used the equations that he came up with from that previous study. He predicted what their D-dependent absorption was and what their D-independent absorption was, and then he added them together to come up with their total absorption, which is the left part of that right circle. Then he actually measured their absorption, and basically, if you look down the 2 right-hand columns, the numbers are pretty close. Basically, his estimate of what they should absorb, both the active and the passive components, was very close to what these folks truly absorbed. So he felt that his equations in those graphs that I showed you before were pretty accurate.

He had also done some other work in patients on hemodialysis. This was a study actually done a little bit before this other study, very same technique, where he was really looking at the role of both aluminum hydroxide and calcium carbonate as phosphate binders. So, the left-hand set of bars is gut phosphorus absorption based on knowing how much phosphorus was in the meal. There was 300 mg of phosphorus in the meal, and when they gave them a placebo, 79% of that phosphorus was absorbed. When they gave them aluminum hydroxide or calcium carbonate as a binder, phosphorus absorption went down to about 62% and 65%, respectively. Also very nicely in this study though, he looked at the calcium absorption side of things. Remember now these are dialysis patients not supplemented with 1,25 D, so the right-hand set of bars is how much calcium was absorbed from these meals. Now, the placebo meal had about 260 mg of calcium. The study I showed you before was 180 mg, and this one now is about 260 mg. You can see that when there's 260 mg of calcium in a meal, there's a small amount of calcium absorption even in dialysis patients not supplemented with 1,25 D. You can see that about 40 or 50 mg of that 260 mg got absorbed.

When he gave the patients the aluminum hydroxide, and this meal happened to have 400 mg of calcium, about 20% of the calcium in that diet was absorbed. Again, these are unsupplemented hemodialysis patients. When he gave them calcium carbonate as a binder, and there was 1600 mg of elemental calcium in that calcium carbonate that he gave them, about 450 mg of that calcium was absorbed, about 28%. And part of that is largely due to this passive non–D-dependent absorption of calcium. So when you give a lot, our dialysis patients absorb a lot, and that's pretty clear from this study.

Another study he did, again very same methodology, was when calcium acetate came out. He said let me look at calcium carbonate and calcium acetate. The study was primarily looking at these two in terms of their phosphate binding capabilities, but he also looked at how much calcium was absorbed from the binder dose he gave. So, the left-hand bar is the meal that contains about 200 mg of calcium and 340 mg of phosphorus. You can see that when the meal only had 200 mg of calcium, these patients were slightly negative, almost neutral. So they presumably had some passive secretion of calcium in the gut, maybe a little active absorption, and they came out neutral. When he gave these people 1000 mg of elemental calcium, either in the form of calcium carbonate or calcium acetate, you can see that they absorbed about 355 mg of calcium or 314 mg of calcium, respectively, and the absorption was about 29% or 26%, respectively.

So these data from Dr Fordtran tell us that our patients on dialysis absorb the dietary calcium we give them. Whether we give it as a meal or whether we give it as a binder, a significant percentage is absorbed. And just because they don't have high circulating 1,25 D doesn't mean their gut calcium absorption goes to 0 or goes to 5%. It stays around 15%, 20%, 25%.

So if they absorb calcium normally from the gut, what about their urinary calcium excretion? What happens to that in CKD? And there's not been a study to directly look at that, but a number of studies have measured it.

These are 2 studies of active vitamin D analogues done in CKD patients, and one of the things they did in both of these studies is they looked at urinary calcium excretion. These are the baseline data: the left-hand set of bars is from Dr Coburn's doxercalciferol study, and the right-hand set of bars are from Dr Coyne's paricalcitol study. These are both oral medicines that were studied in CKD. And this is just the baseline urinary calcium excretion in the 2 groups. So before they got treated with any of the hormones. And what you see is the bars are not too much different. They are both, again, pretreatment groups, but if you look at the Y axis and look at what the urinary calcium excretion is, you can see that it's dramatically low. 25, 30 or 40 mg/d.

Now, when you give these people the hormone and look at what happens to their urinary calcium excretions, you can see that it doesn't change much. Again, in the doxercalciferol study on the left, the active treating group went up from about 25 to 38 mg of calcium in their urine over 24 hours. The right-hand set of bars is the paricalcitol study, and those patients really didn't change much, about 38 to 42 mg/24 hours.

Now, these studies were interpreted as: ah-ha, there's no increase in urinary calcium, and therefore these hormones don't increase gut absorption of calcium. It looked like the discussions were related to the studies.

But I'm going to suggest to you that there may be another reason why the urinary calcium levels didn't go up. These are data from Craver. This is a significant, large population, that he looked at, a European study, and some of these patients had 24-hour urine calcium levels at different stages of CKD. So the first dot and error bars are essentially normal individuals. They call them CKD stage 1, and you can see that their urinary calcium is relatively normal, about 150 mg with these very wide error bars. The second dot is CKD stage 2, again relatively normal, but as you move through CKD, what you see is that urinary calcium falls dramatically. What I'd like you to look at is CKD stage 4 vs CKD stage 2. Now remember the error bars give you...obviously the smaller the number, the bigger the error bars you're going to get if you're measuring a sample of the population. But there are actually fewer patients in CKD stage 4 than in the CKD stage 2 group, so you would think its error bar might be wider. But I'm going to suggest to you that the reason that the error bars get so tight in CKD is because the kidney is limited to what it can do.

When we teach physiology in terms of what happens in CKD, we all put up a slide similar to this, and on the Y axis we could have water excretion or sodium excretion, and I'm going to suggest to you that on the Y axis now we could also put urinary calcium excretion. And as you develop progressive kidney disease, the ability of the kidney to modulate how much calcium you can put out dramatically decreases. And if you have a fixed urinary calcium excretion and you measure a population at some advanced stage, you're going to get a group that all ends up with the same urinary calcium excretion and a very narrow standard deviation in your curve. If you’re looking at people with normal kidney function, their urinary calcium excretion is really going to more likely reflect their dietary calcium intake, where there's a lot of variability. And I think that's the reason that patients with CKD all had urinary calcium excretion that fell into a narrow range. My guess is, in those active vitamin D studies, the reason the urinary calcium didn't increase is because the patients by and large were not able to increase their urinary calcium excretion.

So, the final question then is: What is the calcium balance in patients with CKD? And this is a study we recently completed. It's in press in Kidney International and it's actually available online. It's not yet in the print version yet. But what we hypothesize is hopefully what you would all hypothesize based on the data I showed you, and that is that patients with late-stage 3 and stage 4 CKD are no longer able to maintain calcium balance when they take in a large calcium load. We also hypothesize that the calcium that they take in will not result in an increase in serum calcium because that's tightly regulated. So we suggested that if they take in a lot of calcium, they'll go into positive calcium balance, but we won’t see that in the serum calcium level because, as you know and as Dr Jamal had said, the serum calcium does not tell us about calcium balance.

So this is the way we did the study. We took 2 groups of patients, late-CKD stage 3 and stage 4, and their GFRs were about 25‒30 mL/min/m². We took normal controls with a GFR of at least > 60 mL/min/m². All the subjects were supplemented before they entered the study with vitamin D to get their vitamin D levels above 30. Then they were on 1 of 2 diets: they were either on a diet containing 800 mg of calcium or a diet containing 2000 mg of calcium. Dr Jamal would say, "Why did you pick such a high number?" But we know how much calcium we traditionally give our CKD patients and our dialysis patients, and obviously if you do a study you want to give good separation, so you can look for differences, too. So those are our 2 calcium diets. Then after they were on each diet for about 9 days, then they went into a clinical research center. Then they were off the diet for at least 1‒4 weeks, on their regular diets, and then they were switched over to the other diet. So it's kind of a crossover study. All the diets had 1600 mg of phosphorus, and all the food for the diet was prepared through the clinical research center. So instead of the patients going for takeout to Chili's, they went for takeout to the clinical research center at the University hospital, and picked up their diet a couple of times. And they only ate food that was on their diet, and if they didn't eat it all, they were supposed to bring it back so the dietitian could carefully calculate how much calcium and how much phosphorus they were taking in.

So the study schema is that the patients were on this diet for 7 days as outpatients, and then, for the last 2 days of the 9-day period, they were admitted to the clinical research center where they had two 24-hour urine collections for calcium, phosphorus, and creatinine. Their blood was also measured for calcium, phosphorus, PTH, 25 D, and 1,25 D, and we collected the stool during that time period also. And we tried to do a formal balance.

So we had 12 patients who completed this study, 6 with CKD, and you can see their estimated GFR here, and 6 with normal kidney function. Then I'll show the rest of the data—the results of this trial.

So this is their serum calcium level, and it shouldn't be surprising to you. The circles are CKD subjects, and the triangles are controls, but they are all overlapping as you might gather. Almost all the patients had normal serum calcium. You can see a few of the CKD patients had slightly low serum calcium levels, but whether they were on the 800 mg or switched over to the 2000 mg calcium diet, their serum calcium levels didn't change. And all this shows us is that individuals regulate around the calcium set point, and they do that whether on 800 or 2000 mg. Again demonstrating that the serum calcium level does not tell you about calcium balance. It does not tell you about dietary calcium intake.

When we looked at their phosphorus levels, they were more scattered, as you might imagine, and, in general, the CKD patients had slightly higher phosphorus levels than the control group. A few of the CKD patients were pushing the upper limit of normal, but most of them had phosphorus levels, even at this relatively late stage of CKD, within the normal range. You can see on the 800 vs the 2000 mg some serum phosphorus levels went up and some went down. Overall, there was no change in the serum phosphorus level, again, as you might expect in this population of patients.

When we looked at the PTH, again the circles now are the CKD patients, you can see they had higher PTH than the control group. The Y-axis is split, here, because of the significant difference between the 2 groups. Even though some of the lines go down and some look relatively level—again this is a pair crossover study—overall the PTH went down slightly. The majority of the patients had a decrease in their PTH when going from a calcium intake of 800 to a calcium intake of 2000 mg, exactly what you might imagine because the serum calcium, or the calcium load, is the principle regulator of PTH. If you want to know whether the serum calcium changes, at least in healthy people, you look at the PTH level, because the PTH level is more sensitive to the calcium probably than the lab is. And you all know that inverse sigmoidal curve, so the PTH went down when you calcium loaded people.

And, we found that the 1,25 D level also went down when people go from an 800 mg calcium diet to a 2000 mg calcium diet, which again is exactly what you might expect. If you have more calcium coming into the system, PTH goes down. PTH is a regulator of 1,25 D, so 1,25 D goes down. So again, exactly what we expected in terms of 1,25 D levels.

This is the urinary phosphorus excretion. It may surprise you—these are the different groups on the X-axis here—but the urinary phosphorus excretion was identical in all of these patients. Now, at first blush you may say that this is surprising: I thought patients with CKD would have lower urinary phosphorus excretions than the normal controls. Remember, they were all taking in the same amount of phosphorus. What I'm going to submit to you, and we really don't know, we don't have the data, but even though we always assume that patients with CKD are in positive phosphorus balance, I would suggest to you that that is a very, very late occurrence in CKD. And what really happens is that most patients in CKD are in neutral phosphorus balance until very late.

Now, they stay in neutral phosphorus balance, as you know, because they have increasing levels of fibroblast growth factor (FGF23), and increasing levels of PTH, both of which are phosphaturic, and they maintain balance. Dr Slatapolsky showed us a long time ago that what happens in CKD is that tubular reabsorption of phosphorus goes down as you have CKD, or the fractional excretion of phosphorus goes up, however you want to think about it. These are the data from our patients, and they show exactly the same thing. So again, the triangles are the normal subjects, normal GFR, and you can see their tubular reabsorption of phosphorus is greater than 80%, which is what ours is. Once you develop CKD, your tubular reabsorption of phosphorus drops dramatically so that you're able to excrete more of the filtered load of phosphorus, and able to stay in balance. Presumably also in CKD, the serum phosphorus levels tends to rise a little bit, even within the normal range, and that slight increase in serum levels also result in an increased filtered load.

This is the urinary calcium excretion in our patients. The left 2 bars are the control subjects. The light gray bar is control subjects on the 800 mg calcium diet, and they excreted between 200 and 300 mg of calcium in their urine. I was surprised it was so much, but it's probably an effect of the low number of patients in the study. When those patients ingested 2000 mg of elemental calcium, you can see their urinary calcium excretion went up. It wasn't statistically significant but it went up to over 300 mg/d. The right 2 bars are urinary calcium excretions in patients with CKD, and you can see these findings are entirely consistent with what I've shown you before. In the 800 mg calcium group, the urinary calcium excretion was on the order of 50 or 60 mg/d, and when they increased to 2000 mg of calcium intake, their urinary calcium didn't go up very much. It went up slightly from 80 to maybe 92 or 93 mg. So again, no dramatic increase in their urinary calcium excretion, and CKD patients put out significantly less calcium than the control group.

And this is the calcium balance. So the first set of dots is the individual patients. In the lines are the 95% confidence intervals. So you can see the control subjects taking in 800 mg of calcium per day, and you can see where the dots fall. Again, small n, but the 95% confidence interval crosses the neutral axis, so our study found that normal individuals taking 800 mg of calcium are in neutral calcium balance. That's not surprising or should not be surprising, I think, for anyone. The third set of dots are CKD patients taking in 800 mg per day, and again you can see they fall very close to the neutral calcium line. And again the 95% confidence interval crosses that line. So CKD patients taking in 800 mg per day, not supplemented with 1,25 D are in neutral calcium balance. The second set of dots is control subjects taking in 2000 mg/d, and you can see there's a fair amount of scatter in those points. Overall, those patients were in positive calcium balance, slightly. Now remember, this study was only after 9 days, and most true balance studies are 3, 4, 6, or 8 weeks. So the question is, if these control subjects would have stayed on 2000 mg of calcium per day, would they have come into balance? Maybe they would have, we don't know, because it was a short study in that sense. The set of dots on the far right side is CKD patients on the high calcium diet. What you can see is they're in marked positive calcium balance, and significantly higher than the control subjects on the same calcium intake. So, I think these data strongly suggest that when our CKD patients are on 800 mg, they're in neutral balance. When you go up to 2000 mg—which as Dr Jamal said is way above the recommendation, for at least most societies in normal individuals—I think now we have some data in CKD patients suggesting that when they take in 2000 mg, they're in positive calcium balance.

So the preliminary conclusions from our study: subjects with advanced kidney disease are unable to increase their urinary calcium excretion to help them maintain balance; late-stage 3 and stage 4 patients with CKD are in marked positive calcium balance after 1 week of ingesting 2000 mg of elemental calcium; control subjects were at even in moderate positive calcium balance on that 2000 mg calcium diet; and all the subjects were in neutral calcium balance when taking 800 mg/d.

So I showed you an N of 12 in our study. Are there any supporting studies? There are.

These are data from a poster. Dr Kim Hill, Dr Moe, and Dr Munro Peacock from Indianapolis are doing a calcium balance study also, and it was presented on a poster at last year's American Society of Nephrology (ASN 2011). I think the data collection is done. I think they're analyzing their data. They did a little more sophisticated study than we did; they actually kept their patients in the hospital for 1 week and collected all the stool for a whole week. The preliminary results from their study were presented in the poster, and also in some oral presentations. In advanced CKD patients—and you can see the GFR here, it was in the 30 mL/min/m² range—these CKD patients were, again, in neutral phosphorus balance when they measured it. In taking in 2400 mg of calcium, some in diet and some as a binder, these patients were also in marked positive balance of about 500 mg/d; relatively close to what we got in our data. So again, I think her data are going to be very supportive of our data, suggesting that when patients take in 2000 or 2400 mg of elemental calcium, they're in marked positive calcium balance.

These are data that many of you have seen. This is a paper published by Dr Bushinsky where he modeled calcium balance in patients on dialysis, based on Dr Fordtran's data, and other data, and basically you see 2 lines. The top line is patients who are supplemented with 1,25 D, and the bottom line is patients who are not supplemented. Where these 2 lines cross this dashed line of zero is neutral calcium balance. The data suggest that if patients are not on 1,25 D or not on an active D analogue, their calcium intake needs to be about 1000‒1200 mg to be in neutral balance. If they’re on an active D analogue, an intake of about 1000 mg/d puts them in neutral calcium balance. Again, very close to what we found in our study, and I think what Drs Moe and Peacock will find in their study.

So I think the implications of this are, again, adequate calcium intake is needed in CKD, but in general, based on the best evidence we have, 800 mg/d is probably adequate. CKD subjects fail to augment their urinary calcium excretion when given a calcium load. That may be due to decreased filter load of calcium or it may be due to enhanced tubular reabsorption of calcium from the effects of their high PTH, and therefore calcium intake, I think, needs to be limited in patients with CKD to prevent calcium loading. And we need to keep in mind whether the patients are on active D or not on active D in terms of what that calcium intake might be. I think in general somewhere between 800, 1000, or maybe 1200 mg/d is probably not only adequate, but probably enough calcium, and I think you may run risks when you start getting higher doses. Again, this calcium is probably best given as dietary calcium rather than a large bolus at one point in time because of the effects on the serum calcium intake.

So Dr Jamal has gone over this a little bit, but what are the effects of taking in too much calcium?

Here are very nice data from Dr Shanahan where she looked at children who had CKD, and looked at their blood vessels. The beauty of studying children is because most of them don't do all the bad things to their blood vessels like we do. They haven't had long-term effects of diabetes, they haven't smoked, and they haven't had high blood pressure. So these were children, the green dots are normal children, how much calcium they have when they look at their blood vessels. These were kids who were in accidents who had to go to surgery and they got a piece of the blood vessel. The predialysis were children who were getting their peritoneal dialysis (PD) catheter placed for the first time. Some of them had been on calcium supplements, you can see already there's a slight increase in the amount of calcium in the blood vessel of these children even before they start chronic dialysis. And the blue dots are children who are already on dialysis, who are often getting calcium supplements, and you can see that their vessels are heavily loaded with calcium. This is long before you see evidence based on electron-beam computed tomography (EBCT) scan or high resolution computed tomography (CT) scan. So this is small calcium deposition in the vessels before it's visible by the scanning techniques we have, these were all pathologic studies of the vessels done at the time this PD catheter was implanted.

We know that the health of your vessels dictates your longevity. This is a study by Dr Block, but there are several other studies in the nephrology literature looking at how much calcium is in the blood vessels, what percentage of your blood vessels are calcified, and all of them show basically the same thing. The more heavily the blood vessels are calcified in patients with CKD or patients with dialysis, the worse their long-term survival.

And this is a meta-analysis by Dr Jamal. I know you can't see all the data, but it basically looks at all of the studies of calcium binders vs non-calcium binders to look at whether there's progression or no progression of vascular calcification. Again, most of the studies are small and a lot of them cross the neutral line. That orange dot at the very bottom is the composite endpoint, but if anything, it favors non–calcium-based binders in terms of not progressing vascular calcification.

This slide has nothing to do with calcium; this is the 4S study (Scandinavian Simvastatin Survival Study). This was the first study looking at cholesterol-modifying agents in terms of secondary prevention risk for cardiovascular disease. The reason I show you this study is, in terms of cardiovascular studies, it's relatively small, 4000 patients. But in terms of dialysis studies it's still relatively large; we don't have many studies with this number of patients. But kind of a small study, 4000 patients, and they looked at statins for secondary prevention of MI. But what I want you to look at is the X-axis, which is the time in years. If you look at 1 year, if this would have been a 1-year study, it would have been negative. If this would have been a 2-year study, it would have been negative in terms of survival, whether statins increased survival. Even if this would have been a 3-year study of statins, it probably would have been a negative study. So when you have 4000 patients, and you put one on statin and another not on statins, and you look at survival over 3 years, it's probably going to be negative. Only when you look 4, 5, and 6 years out do you see the benefit of the medication. That's important to keep in mind when you do randomized, prospective trials; you have to have adequate exposure either to what's bad or what's good to see a difference.

So then, again, here are data from Dr Jamal looking at the studies that are available of calcium binders vs non-calcium binders in terms of cardiovascular events or survival. And what you see is that most of the studies cross the neutral line, but in this column called months, I showed you how long the analysis was. So one is 12 months, one is 5 months, and 44 months was the study that Dr Block and I did. That is the only one that's longer than 3 years. All of these studies are small or have problems: 12 months, 12 months, and 20 months. So if you put these in the context of the 4S study, really only 1 of them was adequately long enough to even begin to look at cardiovascular risk. You can see the big orange dot, which is the composite, cross the neutral line, not statistically significant. It fell on the side of favoring non–calcium-based binders. I would suggest to you that not looking properly for an effect is not equal to having no effect or no evidence. And so when people say it's been looked at and we know that calcium vs non-calcium doesn't affect cardiovascular risk, I would argue that we just haven't looked at it properly, we haven't done the proper studies to know definitively. So it's still an area of controversy, although I think the physiology is very clear.

So in summary then, calcium loading of the vasculature is common in patients with CKD and ESRD. Vascular calcification is a strong indicator of mortality, and vascular calcification is an active and complex process that we don't understand completely. I think the weight of evidence suggests that calcium loading accelerates vascular calcification and worsens outcomes.

So my takeaway message for you is based on the physiology we know and based on what Dr Jamal told you. It is also based on the physiology of Dr Fordtran’s study, our study, and Dr Moe's study. I think we can safely say that there's probably no rational justification for administering greater amounts of calcium to individuals with minimal urinary calcium excretion; that's our patients with CKD, who have relatively intact gut calcium absorption, then one would recommend for the general population. And in the bottom graph here I put the Institute of Medicine's recommendations which are 800 mg in men as the average requirement, and 1000 in women, the RDA. I think I should define RDA for you: the acronym stands for recommended daily allowance. But what it means is that that amount of calcium will meet the nutritional needs of 97.5% of the general population. So 1000 mg in men will meet the needs for 97.5% of men, and I would suggest to you that's true of our CKD patients, and true of our dialysis patients. And 1200 mg will meet the needs of 97.5% of women, and I think that's true of the general population and true of our ESRD population. And the upper level is 2000 mg, which means that when you start getting levels up there, you certainly may be causing risk.

That's my last slide, and now we're going to do the cases again.

So this is our 55-year-old postmenopausal woman, with autosomal dominant PKD, and she asks you about calcium supplementation to reduce her risk of fractures. Her hypertension is controlled, but her mother had a hip fracture at 62, so that's why she's certainly worried. She weighs 159 lb, she's 5' 5" tall. Her laboratory panels: She's got a normal kidney function, her DXA scan is the same as what we told you before. We didn't change the case on you.

And the question is: How would you answer her question regarding the prevention of fractures? Recommend A: Calcium intake of 1200 mg/d; B: Begin a vitamin D supplement; C: Begin an exercise program; D: Begin a bisphosphonate; E: Do all of those plus more; F: A, B, and C; G: B, C, and D?

So, most of you would do A, B, and C. Before this talk, most of you would have recommended calcium intake at 1200 mg, put her on vitamin D, started an exercise program, and some of you would have done that plus adding a bisphosphonate. I guess some people would have just done exercise. So let's turn to Dr Jamal for her answer to this.

Sophie Jamal, MD, PhD, FRCPC: I think probably the most appropriate thing would be A, B, and C, which is answer F. So recommend a calcium intake of 1200 mg/d, that's actually very reasonable, a total of diet and supplement. Start vitamin D and add an exercise program.

David M. Spiegel, MD, FACP: So the next question was: What current medical concerns might influence your decision regarding calcium supplementation in this patient? Are you worried about her A: PKD? Are you worried about her B: Cardiovascular risk? Are you worried about C: Fracture prevention? Are you worried about A and B, which is her PKD and her cardiovascular risk? Or E: B and C? Or are you worried about all of them plus more stuff?

So again, I think things changed just a little bit, maybe, I think what we saw is at the beginning everybody was worried about all of those, at the end I think a lot of people focused more on the cardiovascular risk, although, I think that was part of F, also, so Dr Jamal, do you want to comment?

Sophie Jamal, MD, PhD, FRCPC: I think probably all 3 are important. You want to think about her PKD, obviously her cardiovascular risk, and her fracture prevention.

David M. Spiegel, MD, FACP: And I guess the take-home message from you would be when you're thinking about calcium supplements—we often think of the good we do as physicians, that's what we want to do, but we have to remember the harm of everything or the potential harm of everything we do—calcium is not just a biological goody that's out there, it can have downsides also.

Sophie Jamal, MD, PhD, FRCPC: Absolutely.

David M. Spiegel, MD, FACP: Okay, and then the final question is 5 years later, her creatinine is 2.4 mg/dL, her GFR is 24 mL/min/m², her calcium is 9.4 mg/dL, in the normal range, phosphorus 4.6 mg/dL, right at the upper limit of normal, her PTH is above the normal range, and she's on calcium carbonate, vitamin D, an ACE inhibitor and a beta blocker. Her chest X-ray revealed that her lungs were normal and there were no fractures, but she had calcification in her aortic arch.

And the question is now: how would you manage her care at this time? Would you A: Continue the calcium supplementation maybe at a reduced dose? Would you B: Stop her calcium supplementation and prescribe a calcium-based phosphate binder? Would you C: Stop her calcium supplementation and prescribe a non–calcium-based phosphate binder? D: Stop her calcium supplementation and prescribe an active vitamin D agent? E: Do B and D? Or F: None of the above? I can't wait to see the results here.

We probably should have given you a couple hours to answer this one. So let's see, at the beginning, most people said they would stop the calcium and prescribe a non–calcium-based binder, and basically that went up to, at the end, to more people would stop the calcium and prescribe a non–calcium-based binder. Some people would elect to use, a D analogue, that didn't change much. Fewer people didn't know what to do at the end then at the beginning. So I think this is the controversy. If you look for what's FDA approved to use in CKD, obviously no phosphate binder is FDA approved in CKD. So if there's anybody here from the FDA, you all failed. So we don't have any medicines that are really approved for phosphorus control in CKD. If you ask me what I would do with this patient, I would do exactly what most people would do. I would stop the calcium supplement and I would put this patient on a phosphate binder, because we think that phosphorus, whether it's positive phosphorus balance or the tendency toward phosphorus for attention sets into play this whole hormonal milieu where FGF23 goes up and PTH goes up. We think that plays an early role in this disease process. I generally direct my therapy toward trying to reduce dietary phosphorus and try to put patients in a more negative phosphorus balance, or at least neutral phosphorus balance, early in CKD to try to prevent this process from taking place. We know that phosphorus plays a role, because I'm sure if we took this lady and measured her tubular reabsorption of phosphorus, at this stage of CKD, it would already be depressed, showing that the kidneys are working harder or these hormones are in place trying to help her excrete phosphorus. So I would stop her calcium supplement and put her on a non–calcium-based binder, knowing that it's not FDA approved. I think this is controversial, and we need more studies in this area.

So I think that's the last of our slides. Let's open it up for questions. I want to thank you all for your participation up to this point, if you need to leave. Thank Dr Jamal again for her wonderful discussion, and I think we have questions on the microphone and there are also pieces of paper you can use.

Male speaker: Thank you for that nice presentation. I'd like to first make a comment. You can think of calcium as an exposure to a potentially toxic agent, and I think you've provided ample evidence, both of you, that all the Bradford Hill criteria for avoidance of calcium or at least excessive calcium, have been met. But given that notion, and the notion that we want to prevent cardiovascular disease as it's affected by mineral and bone metabolism disorders, and the efforts for the Quality Incentive Program (QIP) in others, to have some markers of vascular and calcification etc. We're still left with serum phosphorus and calcium, and you've just shown very nice evidence that neither of those are probably showing what we should be looking for. So do you have any suggestions, either of you, on other markers that we should be tracking to better determine and influence the outcomes in our patients?

David M. Spiegel, MD, FACP: Well that was a simple question doc. I will try to answer it.

Sophie Jamal, MD, PhD, FRCPC: I think that's definitely a nephrology question.

David M. Spiegel, MD, FACP: So nephrology question. We obviously don't know what we need to be tracking in our patients. I can tell you what we do, in our clinic. We measure fractional excretion or tubular reabsorption of phosphorus, and we try to normalize that in our CKD patients. And we try to do it initially with dietary phosphorus counseling, and there's a lot of work coming out that the type of dietary phosphorus you take in may make a big difference in how much is absorbed, moving from a protein- or meat-based phosphorus source to a plant-based may decrease phosphorus absorption. Having people try to avoid prepared foods that have these phosphorus additives that are very highly absorbed may make a difference, and then we go off-label and use non–calcium-based binders early in our patients. We try to normalize their tubular reabsorption of phosphorus. Is that right or wrong? We just don't have the data. But a great question, and, when you get the answer you'll let us know.

Male speaker: Thank you for the presentation. One of the questions is about vitamin D. When you give a patient vitamin D, just because its physiologic function will increase the calcium, so, not really understanding what the difference would be between giving vitamin D, and just giving the patient PhosLo, in terms of cardiovascular disease? I mean it seems that even giving vitamin D would cause harm because of the increase in calcium absorption.

David M. Spiegel, MD, FACP: So I'll take a shot at that, and then Dr Jamal. So when we're talking vitamin D, we're talking about the vitamin, not the hormone. So the conversion of D to 1,25 D is very efficient, so there can be a lot of people who are, we think, vitamin D-deficient, but who have adequate 1,25 D levels. So I think what we're seeing—and this has been studied in terms of vitamin D dosing, when you look at increasing and supplementing people with the vitamin D and look at their urinary calcium excretion, which is probably the best measure in normals—about calcium intake is it doesn't increase in the range of vitamin D supplementation that's being recommended. So, the data suggest that when you supplement people with the vitamin, up to this level of 30, you don't increase their urinary calcium excretion, which presumably means you're not significantly increasing their gut calcium absorption. When you get to very high levels of vitamin D, then I think you can run into that problem...

Sophie Jamal, MD, PhD, FRCPC: I agree.

David M. Spiegel, MD, FACP: So I think giving the vitamin supplement is very different than giving a calcium supplement.

Sophie Jamal, PhD, MD, FRCPC: There were a couple of questions talking about bisphosphonate use or antiresorptives in ESRD. There were about 3 or 4 of them, so I guess that's a pressing issue for people. I think the issue—and this is where I do a lot of my work—in patients who have stage 5 CKD and are on dialysis—is they're often breaking bones. But the issue really is: what's the etiology of their bone disease? And to remind you, there are 2 diametrically opposed types of bone disease. There's high-turnover bone disease, such as hyperparathyroid bone disease, and then there's low-turnover bone disease, such as adynamic bone disease. And a concern with giving an antiresorptive agent, whether it be the bisphosphonates or Prolia^®, is that if this person has low-turnover bone disease and you were giving them a drug that further suppresses turnover, there's a potential to do harm. So there's a potential, there are no definitive data on that, but it seems to make sense that you wouldn't want to give that to someone whose fracturing with low-turnover disease. And the bottom line is that today, the only way to tell whether someone has high- or low-turnover bone disease in stage 5 CKD is bone biopsy. So, you have patients who have fractures. The question I always get asked is: what dose of bisphosphonate should I give them, because they're on dialysis? I think the real issue is: what's going on at the level of their bone? There are a few exceptions to the rule. Patients with PTH levels 10 times the upper limit of normal, and bone Alk-phos that’s undetectable. In your practice, you don't see those patients a lot. Most people have PTH levels that are in the middle to high range, with Alk-phos in that range as well. So, that's the general sort of issue. With regard to predialysis CKD, there have been several analyses done by Dr Miller and myself looking at things like alendronate, risedronate, and Prolia^® as well, that show that it actually does work in patients with predialysis CKD. But they are a very specific group of patients; they're patients who typically have normal serum creatinine, with an elevated clearance, so little old ladies who typically have normal PTH and normal vitamin D. So there's a distinction that may need to be made between renal bone disease with aging, and then bone disease that's associated with more significant conditions. I think it's just something to be cautious of.

David M. Spiegel, MD, FACP: What is your strong recommendation for calcium in the dialysate? And I think that's an important question. We all struggle with that. And I think we struggle here as a community. We don't know exactly what's right. In my mind, dialysis procedure itself theoretically should be calcium neutral. And the question is how you achieve that, and it's probably different in each individual patient. Remember, you have diffusive calcium movement and you're also going to have convective calcium movement. There are people who have written about this, and we don't have great data or large studies. I think some of the data we have suggests that if your serum calcium is in the mid-9 range, a dialysate calcium of 2.5 mEq/L is relatively neutral, which means if your serum calcium is lower than that, then theoretically a lower dialysate calcium would keep you relatively neutral. I think, again based on no real good scientific evidence, some of the mistakes we tend to make is that we have patients with low serum calcium levels or lower serum calcium levels, and other than post-parathyroidectomy, sometimes we put them on a higher dialysate calcium to try to bring that calcium up. I think what we really need to do is understand why the patients have a low serum calcium. And I'm not talking about post-parathyroidectomy patients, but we have patients with calciums of 8.2 or 8.3, and we put them on a higher bath to try and raise that calcium. I think we really have to remember that we're putting the patients in very large positive balance during that dialysis treatment. It's unlikely that we're making their bones healthier. Their serum calciums often don't change dramatically, and then the concern is where is that calcium going? So I think the dialysis procedure, if I could dial it in, I would say calcium neutral dialysis so the patient doesn't lose or gain calcium during the dialysis procedure. Then I would modify their diet to try to keep them in calcium balance. I know that I don't really understand this relationship between blood calcium, bone disease, and vascular calcification completely. What I know is that the way we've been treating patients for all these many years with higher calcium baths or higher calcium supplements—I mean our patients are getting vascular calcification and they're dying; cardiovascular death. And I hope that if we begin to think about this properly and do the studies, we can begin to modify that disease process. But I certainly don't have the answer to what to do there.

David M. Spiegel, MD, FACP: So this says: In patients with CKD and low vitamin D levels, but who have a history of recurrent kidney stones, calcium oxalate, for example, would you treat this patient with ergocalciferol or vitamin D? Now this said CKD patients, but let's say normal individuals, with low vitamin D but a history of stones, would you treat them with a vitamin D supplement?

Sophie Jamal, MD, PhD, FRCPC: So I think in otherwise healthy men and women who have had a history of stones, I typically will measure 24-hour urine for calcium and oxalate, obviously, and if diet can be modified, we do that. Certainly we would change our recommendation for calcium supplementation in those patients, because the concern there is that maybe their threshold for absorption is modified with the supplement. But I would absolutely treat those patients with vitamin D. So if their levels were low, so 25 D in our non-CKD patients and then even in the CKD patients, not a lot of data, actually probably no data, the 25 D replacement is helpful, D3. But it is found on skeletal muscle, and perhaps for falling, frailty; it may be helpful in that regard. It certainly doesn't give high blood calcium, so it may not be particularly harmful to give them 25 D, and then the issue is the 1,25 D. I think that's very low dosing, give cautiously, because the concern obviously there is you're going to be limited with hypercalcemia.

David M. Spiegel, MD, FACP: By the calcium level. And even though we've talked about vitamin D supplementation, I think we need to keep in mind that we don't have the large randomized prospective trial there, either, as you pointed out. We don't have data that supplementing the vitamin in these patients improves longevity or decreases cardiovascular risk. Those studies are ongoing now, but we don't have the data yet.

Sophie Jamal, MD, PhD, FRCPC: So I have one about the studies on calcium intake and cardiac events, whether they’re ethnicity-specific, Hispanic, non-Hispanic, and generally speaking, they're very limited to Caucasians. A little bit of the Women's Health Initiative data, included some African-Americans and Hispanics, but, generally speaking, it’s really not generalizable. And just to remind you, most of the large data, a bunch of the studies that I showed you, are really accumulations of small studies that were limited, so our generalizability to a larger group is really limited.

David M. Spiegel, MD, FACP: Is there any evidence that giving patients active vitamin D analogues in CKD accelerates the progression of CKD? And I think there were some early studies, mainly from Europe, with some of the longer-acting preparations, that that was the case. We don't know exactly why the reason was. Our dialysis patients, we see them all the time and we check calciums very frequently, and hopefully—we could probably spend another whole lecture on the role of 1,25 D in dialysis patients—if you get hypercalcemic, we back off on the dose. And I think what happens in CKD patients is they're seen very infrequently, and so the feeling in those studies is they got put on these D analogues, maybe they were hypercalcemic for some period of time, and that may have been the cause that accelerated their CKD. So, there are some earlier data out there, but I think in the D analogue trials in CKD there was no evidence that they progressed. But again, they were smaller studies and short periods of time.

Sophie Jamal, MD, PhD, FRCPC: So I have 1 more question. Has there been any assessment of bone quality, then T-score ordered vs disordered, as it relates to calcium supplementation in non-CKD or CKD. So, just to be clear, bone density, which gives you T-score measurements, does not give you an idea of bone quality. the only way that we're going to really be able to assess bone quality, unfortunately, right now is with bone biopsy, and there have been no data that have looked at specifically calcium supplementation and bone biopsy in non-CKD patients. I don't think that that's been specifically looked at in CKD either.

David M. Spiegel, MD, FACP: So in CKD and dialysis patients, it's recommended that you not get a DXA scan because, again, it gives you a measure of bone calcium but not turnover, not quality, and we think our patients fracture because they have poor bone quality as opposed to low bone density. So again, we don't know exactly what to do with the DXA scan when we get it. A lot of the drugs are contraindicated even though they are being used in that area. I think the take home message is that to really know what's happening at the level of bone right now, we still need a bone biopsy, in advanced CKD and in dialysis. And I think one of the other caveats that KDIGO (Kidney Disease: Improving Global Outcomes) came out with is this: CKD stage 3a may not really be CKD. A lot of elderly white women, when you measure their creatinine and calculate their GFR, they come up with a GFR of 55 or 52 mL/min/m². So you put them at CKD stage 3, and now people are worried about preventing their fracture risk. My guess is most of those patients, if they don't have hypertension and they don't have proteinuria, are more likely to fracture then they are to have progressive CKD. So I think the recommendation is to treat those patients whose numbers fall into CKD 3a more like the general population, in terms of managing their bone disease.

Sophie Jamal, MD, PhD, FRCPC: I think the second the post-hoc analyses from the randomized trials have concluded—patients exactly like that—so it's this stage 3a that then were randomized to receive alendronate, risedronate, even teriparatide, and had improvements in bone density and fracture risk reduction.

I have a last question for you. In your stage 3/4 CKD patient with type 2 diabetes, what would you recommend to manage his hyperphosphatemia?

David M. Spiegel, MD, FACP: So stage 3/4 CKD?

Sophie Jamal, MD, PhD, FRCPC: With type 2 diabetes.

David M. Spiegel, MD, FACP: The fortunate thing is most of those are not going to be hyperphosphatemic, because hyperphosphatemia really starts very late, or occurs very late in stage 4 CKD and even stage 5 CKD. So most patients with CKD are going to have a phosphorus level in the normal range, which doesn't necessarily mean it's not a problem. And again, I manage those patients, whether they're hyperphosphotemic or whether they have a decreased tubular reabsorption of phosphorus, with dietary phosphorus, counseling, and then non–calcium-based phosphate binders, off-label.

Sophie Jamal, MD, PhD, FRCPC: So I'm assuming if they were stage 4 or 5, you would do the non–calcium-based binders.

David M. Spiegel, MD, FACP: I would do the same. I think one thing we don't do in our patients very well, and I don't do very well in my practice, is we counsel our patients to avoid dietary products and we counsel our patients to avoid a lot of sources of calcium. Sometimes their diets are poor enough where they actually may be taking in very low calcium intakes, and I think there is that concern also that needs to be out there, that if their calcium intake is between 400 and 500 mg/d, that may be on the low side. It may be affecting their bone disease, and may be affecting their secondary hyperparathyroidism. And so we want to make sure they have adequate calcium intake, which is probably on the order of 800 mg/d, but we don't want to go excessive, don't go higher than 1000 mg either.

Anybody have any other questions?

Male speaker: The cardiovascular risk is so high early on in stage 3b and stage 4, even before, as you mentioned, the phosphorus levels are getting higher, possibly on the basis of vascular calcification. Do you think that dietary interventions early on at that stage, to keep the phosphorus level lower, may benefit them potentially, even though their phosphorus is in the normal range? Or not? It's gone up, say they were 2.5 when they were healthy, and as they develop progressive CKD, they become 4.2. We know there's evidence that vascular calcification goes up early on, even at that level. Should we be pushing harder, earlier on, maybe to lower FGF23 levels which might have some vascular effect? Or just comment on the process.

David M. Spiegel, MD, FACP: You know, there are a lot of provocative data out there, we know the recent data from Myles Wolf and his group showing that FGF might be a direct cardiovascular toxin, but the association of treating with a phosphate binder and lowering FGF levels hasn't been clearly shown either. There are a lot of things that make sense when we look at what we know, but that haven't been shown in any kind of prospective trial yet, that X connects to Y, and Y connects to Z, so to speak, so I think we're all struggling. As far as I know, absolutely, I think that's very reasonable to begin dietary modification early in CKD, try to educate the patient about the dietary phosphorus sources, and then the question is when you start a binder, off-label, nobody knows for sure. We use tubular reabsorption as our guide, but nobody knows. But dietary phosphorus modification early, absolutely. I mean, none of us should be drinking the dark colas and all that kind of stuff. Might as well just get an IV infusion of phosphorus.

Well, thank you very much for your attention, thanks to Dr Jamal, and thanks to Sanofi for funding this.