Questions remain about what to do when coronary artery disease is found in candidates for kidney transplant and other people with chronic kidney disease. When would coronary angiography and revascularization make sense? In this episode of AP Cardiology, host Andrew Perry, MD, is joined by Alec Moorman, MD, and cardiology fellow Alex Taylor, MD, both at the University of Washington, Seattle.
A transcript of the podcast follows.
Perry: This is AP Cardiology and this is your host, Andrew Perry. Thank you both for joining with me today. I have Dr. Alec Moorman and Dr. Alex Taylor visiting with me. Can I have you both introduce yourselves, starting with Dr. Moorman?
Moorman: Hello. Thank you for having me. I’m a clinical associate professor at the University of Washington and I spend a fair amount of my clinical time doing consultative work for our kidney transplant population.
Perry: Perfect, and then Alex.
Taylor: Great. My name’s Alex Taylor. I’m one of the second-year cardiology fellows and during my first year of fellowship I was in Dr. Moorman’s clinic, the Eastside Specialty Center.
Perry: Perfect, and they’re both with me here to talk about coronary artery disease in patients with chronic kidney disease, which is a common comorbidity and a common patient population that you encounter. How about we launch in with a patient case with Alex?
Taylor: To kind of have a starting point for our conversation today, we’d like to talk about a 58-year-old man with a history of end-stage renal disease due to type 2 diabetes and hypertension. He is currently on hemodialysis, which he’s been on for about five years. He also has hyperlipidemia, a prior tobacco use history with a 20 pack — a-year history. He quit about 15 years ago. He’s sent to see you in consultation in the cardiology clinic by his kidney transplant team after a screening pharmacologic nuclear stress study showed a reversible perfusion defect in his distal anterior wall and apex. The summed difference score with that stress study was 6. A resting echo showed that his left ventricular ejection fraction is 57%. He had no resting wall motion abnormalities. When you see him, he is asymptomatic, without any anginal chest pain.The question for discussion is should he undergo coronary angiography and possible PCI of the lesions found? Then secondarily, if he has evidence of obstructive coronary disease on cath, will a coronary revascularization decrease his perioperative risk?
Perry: Before we launch into that discussion, Dr. Moorman, might you just comment on…? You mentioned, Alex, that the patient had a stress test with a summed difference score of 6. Can you just, for some of our listeners who won’t know what that means, is that like a low risk, high risk, medium risk? Where does that kind of fall?
Moorman: Yeah. That would be right on the border of what we consider a moderate burden of ischemia. This would be an ischemia burden of approximately 10%.
Perry: Okay, so moderate amount of ischemia on a patient undergoing a workup for a kidney transplant. What are your thoughts about sending this patient towards a diagnostic angiography?
Moorman: Excellent question. Historically, this is something that most kidney transplant programs would do. They would do diagnostic coronary angiography for further risk stratification in somebody who has an abnormal stress test. Now, this patient’s asymptomatic and has normal left ventricular ejection fraction, which is a good prognostic sign for somebody undergoing a kidney transplant. It’s not definite, of course, that this patient would benefit from that further information of knowing the coronary anatomy, and that’s the crux of our discussion today.
Taylor: I think in these cases the concern of the transplant teams is that, based on some prior data, it’s known that having more… basically, patients who have coronary disease have worse outcomes after transplant than those who don’t have coronary disease, but there’s been no randomized data showing that treating obstructive coronary stenosis actually improves outcomes in those patients. I think that’s where there are differences in practice. Because I think there’s some observational data that treating obstructed CAD actually does improve outcomes, but that data just doesn’t exist in the randomized setting.
Perry: Interesting. In specific, what trials or data studies do we have, Dr. Moorman, that supports this idea?
Moorman: There is no randomized trials in the kidney transplant population in the modern era. There was a small study done in the 1990s, a single-center study where they randomized, I think, 24 patients or so, to coronary revascularization versus medical management. The patients who were revascularized had better outcomes, but this was in an era before what we can consider contemporary or modern medical therapy.
There have been randomized clinical trials for pre-op management, but not in patients with chronic kidney disease. The most well-known and talked about is the CARP study, which was a VA study in patients who were undergoing vascular surgery and they were randomized to routine coronary revascularization before elective vascular surgery versus medical management.
Perry: Those are pretty high-risk vascular surgeries, from what I recall in that trial, Alec
Moorman: Yes. That’s peripheral bypass or aortic aneurysm repair, so high-risk surgery in a high-risk population, and with a survival that’s actually similar to patients with chronic kidney disease and end-stage renal disease. The three-year mortality was about 30%, which is comparable to the ISCHEMIA-CKD’s population.
Perry: Yeah, very high. Otherwise, unfortunately, it seems like for a lot of these patients with chronic kidney disease, they are excluded from a lot of those trials in revascularization studies, so looking at does preoperative revascularization improve outcomes or do patients with stable ischemic heart disease have improved outcomes? That kind of leads us towards ISCHEMIA-CKD. The ISCHEMIA trial then had a subsection of that or, probably better called, like an additional study, looking at patients with chronic kidney disease. That was like a historically unique trial in that perspective, right?
Moorman: Correct. Our landmark studies in stable ischemic heart disease that preceded the ISCHEMIA trials were COURAGE and BARI 2D. COURAGE had only 16 patients enrolled with a GFR less than, I believe it was, 45, so very few patients with moderate to severe kidney disease in COURAGE. In BARI 2D, which was a study of patients with diabetes who were randomized to revascularization versus medical therapy, that study excluded patients with a creatinine greater than 2, so very low numbers of CKD patients in our stable ischemic heart disease trials. It’s important to remember that we can’t extrapolate to CKD patients because the biology of their coronary disease is different. They have more medial calcifications, much worse calcifications of their arteries, and also a higher incidence of microvascular disease, which of course can’t be managed with revascularization.
Perry: Kind of coming back to our question about patients with chronic kidney disease who are particularly patients undergoing a transplant evaluation, it seems like on a general high level that, generally speaking, referring patients to diagnostic angiogram and revascularization strategies is probably not going to improve benefits. However, are there select patients within that population or other strategies in which we might consider revascularization for those patients? Because I was specifically thinking of patients with more severe ischemia, or even maybe a history or science suggestive of triple vessel disease.
Moorman: Right, so we should think of these patients… I would argue we should think of our CKD patients’ indications for revascularization the same as our patients without CKD and we should be revascularizing people with left main coronary disease, or low EF, or three-vessel disease. Those are class 1 indications for revascularization in anybody. The idea that if you select patients with a particularly high ischemia burden, that you might benefit that select group with revascularization, that was the whole point of the ISCHEMIA trials, so to better understand that, I will say that it’s common practice nationwide in pre-kidney transplant population that you revascularize. I’m not saying this is the right thing to do. I’m saying historically this has been what’s done, is patients are revascularized before kidney transplant if obstructive coronary disease is found. That was not driven by great evidence. That’s more consensus or expert opinion. The kidney transplant physicians point out that their patients are higher-risk and this surgery is higher-risk than many other populations that are studied, so that’s their argument for why you can’t extrapolate to their patients.
Taylor: As a side note, what are the barriers to actually doing a randomized trial of preoperative coronary revascularization in the kidney transplant population? Why hasn’t that been pursued historically?
Moorman: Good question. I think that the barriers are similar to some of the challenges we saw in conducting the ISCHEMIA trials… and I’m sorry. I should have mentioned at the outset in my introduction that I was the local investigator for these trials at the University of Washington. The barriers are 1) just collaborating with different centers who all… they’re subject to the UNOS guidelines, but each center has some local differences in how they run a transplant program. To do a randomized trial of any great size, you’d either have to do that in one single high-volume center or collaborate. I think that some physicians would think that there wasn’t enough equipoise. They would be reluctant to randomize their patients, which is a problem that we saw with the ISCHEMIA trials. In other words, they might acknowledge that there is this equipoise in the literature, but they wouldn’t feel comfortable randomizing their patient population. Then the patients, I’m not sure it would be easy to get patients to layer on being involved in a randomized trial, in addition to all the hoops they’re already jumping through to try and get a kidney transplant. I would worry, and I think some ethicists would worry, that would patients feel, like obligated to participate in the trial? Even if you told them this in the consent process that being involved in the trial is not mandatory, there might be concerns about inducement or feeling like they have to be in the trial if they want to eventually get a kidney transplant at the center. I think it would be challenging, but I think it’s greatly needed. I have thought about this and talked about it with our transplant docs. I think it would just be harder to execute.
Perry: Makes sense. I’m sure this is a common referral, maybe just taking a step back, but thinking about the patients who are referred to you as part of their kidney transplant workup, what are the routinely… like how do you evaluate those patients just in terms of like basic things, like history, physical, and then like additional tests that are pretty common for you? I don’t know, Alex, you might be able to comment on some of that first.
Taylor: Yeah. What I’ll say is the patients are typically sent to us after they’ve already had some sort of a non-invasive stress test, and it tends to be a mix of either a pharmacologic nuclear test or, in some cases, some form of stress echo. What we’ve encountered is that the sensitivity and specificity of non-invasive stress testing is actually pretty low in the CKD population, and I think it’s particularly an issue with the nuclear stress tests where they just don’t perform as well. I think that and, probably in part related to the increased prevalence of probably a microvascular disease in the CKD population… what are other things that contribute to that worse sensitivity and specificity?
Perry: Sure. Just one thing along that is, I think, looking at the data from the ISCHEMIA-CKD paper, is that all these patients had a positive stress test, but about a fourth of them had no evidence of obstructive coronary disease on their angiogram. Dr. Moorman, could you elaborate on those?
Moorman: Yeah. I think we saw that for a few reasons. 1), and Alex just mentioned this, but the sensitivity and specificity of myocardial perfusion scans in the CKD population is worse than it is with the general population, and that’s been published. Robert Wood Johnson Medical Center, New Jersey published their data, and the likelihood ratios, both the positive and the negative likelihood ratios, of a myocardial perfusion scan are less than 50%, so that’s pretty awful.
Perry: That’s pretty bad.
Moorman: Yeah, it’s pretty bad. It’s worse than a coin flip, right? I think part of that is true false positives leading to coronary angiography, but part of it is also the microvascular disease, where they have myocardial ischemia, but they don’t have obstructive disease in the epicardial coronaries that you see on a cath and that you could potentially treat with a mechanical conduit. I think it’s a mixture of both those things. Then for the CKD, the ISCHEMIA-CKD study, there was no coronary CTA to exclude the patients with non-obstructive epicardial disease from being randomized, which we were able to do in the main ISCHEMIA study.
Perry: Yeah, very interesting. Okay.
Moorman: If I could go back to your original… if I could add a few things. You asked about the workup and the history and the physical.
Moorman: I wanted to mention that it… and this was maybe going to come up later, but patients with chronic kidney disease are much more likely to have absent or atypical symptoms of angina, similar to patients with diabetes. That can be one of the challenges when assessing these patients. Because most of them have multiple risk factors for coronary disease, but using angina as a screening tool is not sufficient.
Perry: Yeah. Another interesting bit from that ISCHEMIA-CKD paper, I think, about half of the patients enrolled in there reported no angina on their baseline questionnaire, just to further validate what you’re saying there. I do have some other questions or topics that are related to this, but I think they are also particularly highlighted better by the second case. Let’s turn over to Alex to lead us into the second case here.
Taylor: For our second case, this is a 62-year-old woman with a history of CKD stage 4, so her GFR is 18. Hypertension, hyperlipidemia, prediabetes, and then an elevated body mass index of 34 kilograms per meter squared. She is sent to cardiology by her primary care physician because she has been having some chest pains. Her medications at the time of the consultation include lisinopril 5 milligrams daily, pravastatin 20 milligrams daily, as well as metoprolol succinate 25 milligrams daily. What she describes to you is that over the past few months she has noticed a pressure in her left chest when she carries a bag of groceries and the pressure goes away when she stops to rest. She has symptoms about once per week. Her pulse is 70 beats per minute to the blood pressure of 136 over 84, and so you decide to send her for a treadmill stress echocardiogram to further figure out what’s going on with her symptoms. She exercised for six minutes in the Bruce protocol and stopped when she started having her typical chest pain. The echo images show that she has an inducible hypokinesis of the inferior wall at peak exercise. The question posed here is should she undergo coronary angiography and possible PCI for this abnormal stress test result?
Moorman: I would argue no. This is not a high-risk stress test and she has stage 4 CKD, so there is real risk of giving this patient contrast and potentially accelerating her kidney disease. She’s not currently on optimal or guideline-directed medical therapy for her presumed CAD. I think my practice, even before CKD, but validated by the CKD study, is that I would medically manage a patient like this, at least to start.
Perry: I think we talked about this a bit earlier, but the data for secondary prevention therapy with statins and aspirin is not quite as strong in patients who have significant CKD. Do you ever find that you’re… Are there situations where you don’t prescribe statins for secondary prevention in CKD patients, or do you typically add it on?
Moorman: For patients who have proven atherosclerosis, like clinically important atherosclerosis I would say, like evidence of ischemia on a stress test, or history of MI or stroke, I still would manage that patient with the highest-intensity statin therapy they could tolerate. It is true that the evidence for benefit with statin therapy in patients with CKD and ESRD is worse. There have been two large randomized trials, the AURORA study and the 4D study, and these were trials of atorvastatin or rosuvastatin in patients with advanced kidney disease. There was no benefit for reducing MACE or mortality with statins, so there is kind of this paradox. Because CKD patients are very high risk for dying of heart disease, but the evidence for statins benefiting them is not strong.
Perry: I wanted to probe a little bit about this issue. It seems to me that maybe it’s an issue of competing risk for these patients, and so they have very advanced kidney disease. Statin therapy is probably going to take a few years for these to take effect, and if you’re at very advanced levels of chronic kidney disease, GFR is less than 20, for example, are you really going to get the benefit to it? It also is kind of like a bit of a cognitive dissonance for me because at the same time, it’s like these patients are at such high risk for coronary disease and strokes. Like you mentioned, their event rates are high. Mortality is like 30% at a year…
Perry: … or for like events. More thoughts about that? I guess what do people talk about this, like in the literature? What do people say when this issue comes up?
Moorman: Yeah, you’re right. One of the theories is competing risk and that it’s just their life expectancy is so short that they’re not going to derive the benefit. I think there is also probably…this might be related to the fact that they’re less likely to have like, lipid-rich or soft plaque versus the dense medial calcifications, or the Monckeberg’s arteriosclerosis is the original term in the pathology literature. Their endothelium and their coronary disease might just be different than the patients with coronary disease who are enrolled in the primary and secondary prevention statin trials. I’ll say that we love randomized clinical trials. I think they are kind of the pinnacle or the best study that we can do, but there is also the risk of type 2 error with these types of studies and there’s retrospective or like observational-type studies that still suggest benefit to statin therapy in this population when you pool data or look at the community data. Again, my practice is that I still use statins in secondary prevention scenarios or patients with proven coronary disease, especially when they have multiple reasons, like if they also have type 2 diabetes, for example.
Taylor: Yeah, I think in some of those trials looking at statin therapy in patients with CKD, when they did a subgroup, or when they looked at subgroups who had high LDL cholesterol, I think there was some benefit if CKD patients have concomitant hyperlipidemia. There may still be some benefit to statin therapy.
Perry: Gotcha. Interesting. As I also think about patients with chronic kidney disease, there’s kind of like the typical patient with chronic kidney disease that I see as a cardiology trainee — so these are my patients with hypertension, diabetes, and then they have chronic kidney disease — and so I frequently forget about this other population of patients that you probably see a little bit more of. These are patients with primary kidney pathologies, so like are nephrotic syndromes or nephritic syndromes, and those issues. I wonder about in terms of should be thinking about them differently, and in the respects of their benefits to statin medications as well, and how we approach their risk and mitigating their risk for cardiovascular disease?
Moorman: Yeah. I do think we should think about them differently in terms of their medical management and also the way that we interpret tests and think about whether to go down that cath and revascularization pathway. Some examples that come to mind are, like I frequently see patients who are referred to me who have like autosomal dominant polycystic kidney disease. They are 40 years old. They’re like stage 5 CKD, getting worked up for transplant, and then I get a perfusion scan that has a mild abnormality in the inferior wall, cannot rule out diaphragmatic attenuation artifact. That’s the sort of patient that I will not cath, because they don’t have the additional risk factors that you mentioned. They’re young. There’s a high risk of a false positive nuclear stress study in that context, so these patients who have primary kidney disease issues without the additional cardiovascular risk factors I do treat differently. For primary prevention, I do think… back to your statin question… I don’t reflexively start statins for primary prevention in patients with CKD. I would be very cautious about using the ASCVD risk calculator or putting an ESRD patient on a statin for high LDL if they didn’t have any other risk factors.
Perry: Sure. Yeah. In fact, one interesting thing as I was preparing for this, I looked again at the ASCVD risk calculator and there’s really nothing in there about chronic kidney disease, or like your baseline, or like your serum creatinine levels. Like it’s completely just not even included in that score, which is kind of interesting. We’ve kind of had like a hodgepodge of things. As kind of like some summary comments or thoughts, maybe I could turn to Alex for just some “Things I’d like you learn,” or some takeaways from your, maybe, time in clinic, and some pearls and highlights from managing of those patients. Then maybe we’ll turn to Dr. Moorman as well for other pearls or some summarizing comments as well.
Taylor: I think my biggest takeaways from seeing these patients in clinic would be just, I guess, trying to put the stress test result in the context of the patient. If you have a high suspicion for an artifact from a nuclear stress setting, for example, following up with a tiebreaker, which, for example, could be a stress echocardiogram. There is potentially a role for a coronary CTA here, but if we run into the issue of giving contrast to these patients who are already kind of teetering on dialysis, and it’s not clear through a scope. But a venous contrast is as bad as the risk of giving an artery for a diagnostic angiogram, but not something, in some cases, that we contemplated doing. I think in terms of… I think the other striking thing is I probably saw, over the course of my year with Dr. Moorman, 30 or 40 patients who had advanced chronic kidney disease and had abnormal stress tests. I would say a very small fraction, almost none of them, actually had anginal symptoms. The ischemia was really just something that was kind of uncovered from just doing a… because usually just we kind of figured out that they had ischemia because they’d gotten a kind of a screening stress test as part of a pre-transplant evaluation, and so honestly, I didn’t do much. I didn’t really find myself adding antianginal therapy very frequently, just because they were asymptomatic and it wasn’t clear to me that they really needed any more medical therapy. Then similarly, we pretty rarely started statins for these patients. I would typically check their LDL to make sure it wasn’t very high, but typically it’s like modestly elevated, somewhere between 130 and 150. I kind of found myself in this situation where we know they’re at elevated cardiovascular risk, but we don’t really know if any medical therapy is actually going to improve their outcome. Then the question from the kidney transplant team is always, “Well, we have this stress test result that’s abnormal, so we should just cath the patient.” I think, as we have kind of talked about, it’s really not clear that doing that is really going to change the patient’s outcome at all. I think it probably helps the transplant team in their kind of risk stratification process. If someone has multivessel disease, then perhaps they’re going to be less likely to offer the patient a transplant. I think, obviously, those are kind of my takeaways.
Perry: Gotcha. Thank you. Dr. Moorman?
Moorman: A few takeaway points. First, the ISCHEMIA-CKD study really validates a conservative approach for most of these patients and that we can stick with medical therapy. I have been working with our transplant team to educate them on that and to try and get them more comfortable. They have been accepting, but we do have to acknowledge that this wasn’t a perioperative management study, but we can, I think, use the data to inform our decisions. One thing to realize is that one of the reasons that they want the cath is for prognostic information. They want to know if this patient’s high risk, not just for the patient and see what their risk is at the time of surgery, but also because they think of the organ as a precious resource. You don’t want to have the patient have a bad outcome and have that kidney end up being wasted, for lack of a better term. That’s where the transplant doctors are coming from. But on the other hand, the data shows that even high-risk patients, patients who have very high RCRI scores, high risk of cardiovascular events around the time of surgery, they still have a better survival if they have a transplant compared to the expected survival if they stay on dialysis. At our center here, our transplant team is becoming more comfortable with the idea of sending higher-risk patients to surgery, knowing that they will have some non-ST elevation MIs in the post-op period. But even when they see those, those patients often do well. I have now seen a couple of patients who we sent for kidney transplant who did have perioperative NSTEMIs, but they still did well, and their transplant took, and they’re off dialysis, and they have a great quality of life. I think it’s important that we move away from this concept that an abnormal stress test necessitates a cath and an abnormal cath necessitates revascularization.
Perry: I think that was a really informative discussion. I really appreciate both of you lending your time, expertise, and knowledge on the subject. Thank you again for participating.
Moorman: All right. Thanks for having me.
Andrew Perry, MD, is a cardiology fellow at the University of Washington Medical Center in Seattle.