So, I've been tasked to talk about prevention of sudden cardiac death in dialysis patients. No disclosures. So, as a way of background, in the U.S., there's over 110,000 patients who start dialysis every year, and approximately 410 are currently treated with dialysis. And the long-term survival of these patients remain poor and has not improved over the past decade, despite 95% of patients meeting dialysis adequacy goals. From the recent United States renal data system, about 20% of patients die within the first year of starting dialysis, and by year five, there's about a 60% mortality rate. And cardiovascular disease accounts for about 50% of all these deaths, with a risk that it's about 10 to 100 times higher than the general population. And then, partly it's due to traditional risk factors, such as hypertension, diabetes is pretty highly prevalent in this population, as well as nontraditional risk factors, such as inflammation, oxidative stress, and cardiac fibrosis. And sudden cardiac death is the most common cause of death in dialysis patients, accounting for 50% of all cardiovascular deaths. And just to put this more in context, so the risk of sudden death in dialysis patients is actually pretty high when they start initiating dialysis, with about 10 per 100 person years within their first three months. That comes down a little bit to six per 100 person years after the first three months. By way of comparison, the general population, the risk of sudden death is about 0.1 per 100 person years. Patients with heart failure with reduced ejection fractures, about six per 100 person years. And post-MI patients, it's about eight per 100 person years in the first six months. So when you start initiating dialysis, you're pretty very high risk for sudden death. And it's pretty multi-factorial in terms of why these patients have sudden cardiac death. Some of it is due to dialysis factors, which I'll go into. Patient-related factors, including age, inflammation, other comorbidities, and QT-prolonging drugs, as well as cardiac factors, including systolic-diastolic dysfunction, LVH, and most myocardial fibrosis. But it's important to know that less than 25% of dialysis patients who experience sudden cardiac death actually have an EF that's less than 35%, while more than 70% have LVH. And dialysis patients exhibit more diffused myocardial scar patterns by CMR compared to non-dialysis patients. This is a study from Scheininger, it's a number of years old now, just showing different MRIs on patients to see that they have pre-significant LVH, as well as delayed enhancement, indicating a possible scar. So what is the mechanism, though, of sudden death in this population? I'll view a few studies. One was by Satcher, who implanted ILRs in about 71 patients with chronic kidney disease on dialysis. And bradyarthritis was the most common rhythm that was noted at the time of the sudden cardiac death. Out of the 71 patients, 16 died during the course of the study, 4 was deemed to be secondary to sudden cardiac death, and all of them were due to bradycardia asystole. There were 10 documented ventricular arrhythmias that were recorded, and all of them were non-sustained VT. It's also to note that atrial fibrillation was very highly seen in these ILR recordings in these dialysis patients. The only thing that was significant in terms of what predicted sudden cardiac death and conduction syndrome disorders were potassium levels are greater than 5, as well as a longer interdialactic period. And the non-sustained ventricular arrhythmias were more associated with potassium levels less than 4. Another study, 50 patients underwent ILR implants who had chronic kidney disease on dialysis. Once again, bradyarrhythmias was the primary mode of death in all these patients. And the incidence of arrhythmias was highest after the first dialysis session of the week. So once again, after this longer interdialactic period was when the highest risk was seen. And finally, the last sort of ILR study that was recently published looked at 66 patients. Forty-four, 67%, had clinically significant arrhythmias that was defined as bradycardia signs bradycardial less than 40, high degree AV block or complete heart block, non-sustained as well as sustained ventricular arrhythmias. And the majority of the arrhythmias that were seen were bradycardia, and there was only one episode of sustained VT. And once again, as the prior study showed, the highest rate of arrhythmias was seen during the first HC session of the week. So once again, after this long interdialactic period for patients who are undergoing three times per week dialysis. So we know that beta blockers, ACE inhibitors, are sort of standard therapy for patients with heart failure with reduced ejection fraction, but does it play a role in patients' dialysis? So this was a study from the hemodialysis study, and this was a post hoc analysis looking at the association of beta blockers with sudden cardiac death among 1,700 patients who were enrolled in this study. And basically the take-home point was that baseline beta blocker use was not associated with a decrease in sudden cardiac death, and both multivariable as well as propensity-matched analysis. And there was interaction for those who had ischemic heart disease, but overall the message was that beta blockers did not decrease the risk of sudden cardiac death in the dialysis patient population. In terms of ACE and ARBs, some of the individual studies were mixed, but a meta-analysis of 837 patients who were randomized to an ACE and ARB versus placebo among dialysis patients, there was what they found was actually a reduction in LV mass, but there was actually no difference in cardiovascular mortality in this meta-analysis. One thing that has been shown to potentially decrease cardiovascular mortality is actually decreasing the ultrafiltration rate. Once again, this hemodialysis study, this was a study, the primary randomization of the study was for 1,800 patients randomized to UF at different levels, 10 millimeters per hour per kilogram versus 10 to 13 to greater than 13. And what they found is that rapid ultrafiltration, so greater than 13 millimeters per hour per kilogram, was associated with a significantly increased risk of cardiovascular hospitalization as well as cardiovascular mortality. This is kind of suggesting that we shouldn't rapidly remove volume from these patients because it potentially leads to hypotension, which could potentially lead to potential ischemia in this population as well. So what about ICD therapy? So in terms of all the primary prevention ICD trials exclude patients on dialysis, so most of this data I'm going to be presenting are sort of single-center, retrospective sub-studies of some multi-center studies, sub-studies, sorry, of the MATEDs and MATED2 studies. So this single-center retrospective study from Emory looked at mortality in end-stage renal disease patients who had ICDs, and the take-home point was that the one-year mortality was high among patients who were on dialysis, 18.1% versus 7.7%, as well as three-year mortality. Among patients, once again, who have ICDs from the Medicare data sets, so this was a study of 9,500 patients on dialysis who underwent ICD placement, once again mortality rates were very high, 448 per 1,000 patient years, as well as post-implant infection rates. And there was really no difference in terms of those who had ICDs versus those who did not compared to a propensity match control from the Get With The Guidelines heart failure cohort. The only population that seemed to potentially benefit was the secondary prevention ICD cohort, which had a 40% lower mortality risk compared to those who did not undergo ICD placement. And ICD therapy just may be ineffective in this population. So this is data from the NCBR ICD registry, which was matched to the Social Security death master file to assess the rate of death after primary prevention ICD placement according to chronic kidney disease stage. So compared to those who had a GFR greater than 60, here in purple, those who were on dialysis had a hazard ratio of almost five, an increased rate of death after ICD placement. That corresponded to a one-year mortality rate of about 22.4%. There's another study from the NCBR ICD registry where they took 108 dialysis patients with ICDs and did propensity score matching to 195 patients on dialysis without ICDs, and they found no difference in one or three-year mortality between the two groups. And the three-year mortality rate was kind of shocking, 74% in the ICD group compared to 76.6% in the non-ICD group. And this is data from the Israeli ICD registry looking at, you know, even if you get an appropriate ICD shock, what is, you know, what is your survival moving forward? So once again, so advanced kidney disease was patients with GFRs less than 30 or who were on dialysis. So they had a much higher mortality rate overall. After their first appropriate ICD shock, once again, those who had advanced kidney disease continue to have significantly increased mortality. And here, these are the graphs showing patients who died without actually receiving an appropriate shock. And so within five years, 50% of patients on dialysis who had GFRs less than 30 died without even receiving an appropriate ICD shock. And finally, this is a meta-analysis of MADITs, MADIT-2, SCUD-HEF, as well as four other randomized clinical trials and primary prevention of ICD therapy. This meta-analysis included 2,800 patients, and once again, since dialysis patients were not included in these studies, they separated into a GFR less than 60 or greater than 60. And the take-home point is that if you had a GFR less than 60, there's actually no survival benefit even if you receive an ICD. So this is, you know, not even patients who are just on dialysis or a GFR less than 30, just even less than 60, there's really no difference in terms of long-term survival. So in conclusion, SCUD event rates are high in dialysis patients, with small studies suggesting that the primary mechanism is bradycardia asystole, beta-blockers, ACE inhibitors are not effective in reducing sudden cardiac risk, and retrospective subgroup analyses suggest that ICD therapy is ineffective in this patient population. Moving forward, I do think that we do need bigger studies to better understand the mechanism of sudden cardiac death in these patients, to maybe understand the triggers for these rhythmic deaths, if this truly is bradycardia asystole, and some things to look at in terms of potassium shifts during dialysis, looking at the type of dialysate you're using during dialysis. Looking at the frequency of dialysis is three times weak enough for some of these patients, if most of these events are happening during a long interdialytic period. Looking at hypotension during dialysis itself, which could potentially lead to myocardial ischemia, as well as medications, including avoiding QT prolonging medications in this patient population. So this HEARTLINK study is a study that we're part of at Hopkins, it's a prospective NIH study that we're running right now. The plan is actually to recruit 1,000 dialysis patients who are undergoing ILR placement. They're going to have pretty detailed electrolytes, the testing before and after dialysis sessions, and really the aim is to estimate the risk of all-cause mortality and the risk of heart failure hospitalization with the presence of ventricular and atrial arrhythmias, to assess the short and long-term increase in mortality and heart failure hospitalization associated with the arrhythmic burden, as well as assess the characteristics of the heart rhythm at the time of death. And the second aim is really to evaluate the specific characteristics of the dialysis session and the development of these arrhythmias. Thank you. That was great. Not necessarily happy, but great. No, yeah, very sobering. So I actually have two questions. One was, do you have an idea of when the bradycardia happened with these patients? Yeah, so from the ILR studies, it seems most of them occurred between their last AC session of the week and the first one the following. So between that two or three days, or actually around the time of the day. So they didn't really comment in terms of the time of day, whether it's at night while they're sleeping, or they just really, the way these studies characterize it, just basically which between dialysis sessions these events occurred. So I'm not sure in terms of the timing of the day itself, but it's been consistently shown that the longer interdialysis period that you have, the higher rate you're seeing of these bradycardia events. And the other question is, do you have an idea whether most of the ICD studies that you were commenting on use single chamber or dual chamber devices, because that may make a big difference as well. So I don't know the percentage for all of them. So the main analysis was from MADETS, MADETS 2, SCUD-HEF, most of these are going to be single chamber devices. I will say, I mean, even with the single chamber devices, you are going to have backup pacing at least to 40 beats per minute. And it's kind of hard to believe that you would die suddenly from, you pace at 40, maybe for a long period of time. But I think that's the reason why we have that big backup pacing capacity. But certainly, you know, clinically and just from antidotes, you know, when you see patients in dialysis centers who code, a lot of times I think it's more PEA as opposed to true sinus bradycardia that would respond to pacing, to be quite honest. And I would imagine that if you were to pace them, you probably wouldn't be capturing anything. This is a really fascinating analysis. The next thing is it kind of flew by there, but the rate of infections was incredible. And it's particularly important because I've seen recommendations from my EP colleagues that maybe we should go with sub-Q devices because they have an, you know, the intravascular shunt. But if most of the rhythms are bradycardias, that would not be an appropriate choice. Yeah, I mean, yeah. So I didn't put a lot of information about the infectious rate, but certainly the infectious rates are very high in this population. And also somebody who removes devices, you know, this could be a pain to do so. But you know, our practice, a lot of times we do kind of go to these sub-Q ICDs. But I think that's the reason why I think it's important to really understand the mechanisms for their sudden cardiac death. You know, these studies were kind of small, 50, 70 patients, and hopefully with a much larger kind of group, we'll kind of be able to really define their, the mechanism. But if it truly is bradycardia, then sub-Q ICDs are probably not the answer. But then you're kind of stuck, you know, putting transvenous system patients with fistulas and DABALs and dialysis catheters. It's not, yeah. At the back, back mic. You, okay. That's right, it's me. So I come from the autonomics world. This is, thank you very much, first. It's bad news, but at least it's definitively bad news. And this looks like a non-paceable bradycardia, I suppose, that's the simplest thing. So that's actually vasovagal syncope. And we know that in the dialysis world, probably there's a lot of vasodepressor syncope. And if you take people like this and you give them a prolonged period of hypotension that's vasodepressed hypotension and not bradycardic hypotension, it'll look like they're dying an asystolic death, but actually they're dying a vasodepressed death feeding back on their heart. So when you're, this actually is a practical suggestion. When you're looking at your rhythms, there's a widely recognized schema for sorting out what's vaguely mediated from what's not. And you can pick this up on an ILR pretty straightforwardly. If you do that, that'll actually change the way you approach the management of these people. Because it might be, and I have nothing to do with the Biotronic, but it might be that the Biotronic CLS pacemaker would actually help these people if that's what the mechanism is. Yeah, that's a good, thank you for that. That's a wonderful contribution. Thank you very much. This is really elegant. All of the talks were really thoughtful, I think, and helped us to look at things in a different way. So I thank all the speakers very much and the audience also for excellent questions. Thank you.

sudden cardiac death

dialysis patients

cardiovascular disease

bradycardia

asystole