Professor Kamyar Kalantar-Zadeh has conducted numerous clinical studies on outcomes and epidemiology in kidney disease patients. In this video, recorded exclusively for COMPACT RENAL, Professor Kalantar-Zadeh provides his expert opinion on the value of observational studies related to iron therapies in CKD. He also highlights good examples of both observational and interventional studies in this area.
The recent KDIGO meeting on controversies in anemia and iron therapy in chronic kidney disease recommended that we would benefit from more observational studies, in particular in CKD patients not yet on dialysis, transplant patients, and dialysis patients and especially peritoneal dialysis patients.
Observational studies have the advantage of showing us real-world scenario circumstances, especially if they come from a large number of patients across large populations, say national databases looking at all dialysis patients or most dialysis patients in a given country, be it Germany, United States, Japan, any other place. So you see exactly how the practice pattern is implemented and is followed and you see the consequences.
So therefore, looking at those large observational studies, we can learn a lot in terms of what is happening and what sort of outcome measures we can consider and analyze. So we need more of those studies.
What are the values and limitations of both observational studies and interventional studies?
Observational studies and interventional studies are usually compared to each other and both of them provide us with important piece of information. Observational studies are as valuable as, if not sometimes more valued, than interventional studies. Both observational and interventional studies have certain advantages and certain drawbacks we need to be cognizant of.
Observational studies, not infrequently, especially if they are conducted in a large number of patients, they represent the real-world scenario practice pattern. We can learn a lot from those studies. Of course, their limitation, among others, is the capability to adjust for confounders, which could be multiple.
Now as compared to interventional studies that are considered the randomized studies, where patients are randomized and therefore confounders are equally distributed among the two or more groups, we should be careful not to consider them as the ultimate perfect studies because, not infrequently interventional studies are limited by the inclusion and exclusion criteria which make them less representative of the real-world scenario.
So therefore, a lot of data that are coming from those studies should be considered and evaluated under qualified conditions of those interventional studies.
So as an example, the number of large observational studies that looked at use of IV iron, in dialysis patients yield very valuable results. Why? Because this data came from large dialysis populations in the United States, where the practice pattern was giving different types of IV iron at different doses and different frequencies.
So we’re able to compare these patterns and look at the outcomes, including heart outcomes, mortality, and hospitalization. So these studies include some studies in which I have been involved, showed for example that a higher dose of iron, up to certain level, may be associated with better survival. Whereas very high dose was not, and whether or not these are confounded by medical indication could be discussed.
The interventional studies, some of them are also very valuable, some in the past [were] smaller studies. Currently, among ongoing interventional studies, one is the so-called PIVOTAL study that is conducted in England. A large number of dialysis patients are randomized to receive intravenous iron through different protocols, and we look forward to those results.
What are good examples of observational and interventional studies on iron management in CKD?
In terms of examples of iron management in CKD, some of our own studies in the past have shown, for instance, when we looked at large number of dialysis patients in the United States, that receiving iron up to a certain dose, for example up to 300 mg or 400 mg per month, was safe and in fact was associated with better survival, incrementally, up to the point that the higher administration of intravenous iron was then associated with worse survival.
I would argue that I’m not even sure that association of higher dose and worse survival was real or based on confounding by medical indications. That means that patients who are sicker, they required very high doses. But at least we saw incremental improvement up to 400 mg per month. And I think that’s the case for most, if not all, intravenous iron products, be it iron sucrose, iron carboxymaltose, or anything else.
Now in terms of interventional studies, there have been some, in the past, smaller studies, and I think the one that we look forward to see the results of, one is actually the one currently, [being] conducted in England by Dr Iain Macdougall, the PIVOTAL study, where dialysis patients are receiving intravenous iron at different forms, at different doses, and we look forward to comparing hard outcomes in this large national study in the entire United Kingdom.
There’s also the FIND study in CKD patients not yet on dialysis. So these are the maybe two larger studies. Other older studies are on smaller number of patients and we have to be careful interpreting those results in terms of extrapolating them to a larger group of patients in a real-world scenario.
The results of observational and interventional studies are usually consistent with each other in terms of iron management. However, it’s important to appreciate that in the real-world scenario, there are different patterns of IV iron administration.
One way to look at them is to divide them into repletion approach, right . . . that means administering iron and holding it . . . and maintenance iron, giving iron in small doses on a weekly to monthly basis, let’s say 50 mg a month of a given intravenous iron, or 50 mg every two weeks. Or maybe if you go higher, 50 mg every week even. Or in the United States, sometimes we give even higher doses. In other countries, probably it’s not practiced very often.
And in both the United States as well as other countries, there are markers such as ferritin that are used to monitor administration of iron. Some colleagues, they take ferritin values, maybe, more seriously than others. Some colleagues or some practice patterns are mostly based on iron saturation or transferrin saturation ratio, ignoring ferritin. And we and others, we use some ferritin up to certain level as a safety level.
For instance, according to KDOQI and KDIGO guidelines, ferritin up to 500 is considered safe to continue or maintain IV iron therapy. According to some colleagues, up to 800 . . . this is the old KDOQI guidelines . . . ferritin up to 800 nanograms per ml should be considered. And according to some other studies, including the DRIVE study, for instance, serum ferritin up to 1200 nanograms per ml may pose no safety issue for maintaining iron therapy when patient would benefit from [this].
So these are different levels of conservative versus aggressive, more aggressive approach to iron therapy. And there is no reason for me to believe that one is more or less superior than the other. I think this at the discretion of nephrologists and centers to decide how to do this and how to manage iron therapy.
How would you conduct a good observational study of IV iron management in CKD patients?
In terms of what a good observational study is in CKD patients or CDK patient populations, I think what is more important is, in observational studies especially, how representative that given population is. If your study is only about one or two dialysis centers as compared to a study that represents a large number of dialysis centers, let’s say hundreds of dialysis centers with thousands of dialysis patients, I feel the data from those large studies, for observational studies, are more reliable and more representative of real-world scenario.
The other one is that what kind of data we have from these observational studies. Do we have all the needed variables? Do we know, for example, which patients receive IV iron, which patients required the other medications for anemia management, what other medications were given, which patients had more infectious events, and so forth and so on?
So the richness of a given observational study allows us also to be able to implement more of the methods to adjust for confounders. So therefore, these criteria could be used to evaluate and rate observational studies.