So it's a pleasure to introduce Jeff Kim from Texas Children's Hospital, who's going to tell us about predicting pacemaker-mediated cardiomyopathy. I'll try and get it started for you so we don't have any more. Thank you very much. Good afternoon, everyone. I am aware of the time, so I promise I'll try to go as quickly as I can. Here we go. I thought we'd start with a quick case, and this is a patient that I follow that has congenital AV block and had a transvenous pacemaker implanted at the age of six. He did quite well, came in for a regular follow-up. Oops, I'm going to see if I can get this to play, sorry. Regular follow-up at the age of 13, and the echocardiogram looked pretty good, interpreted as normal function. Came back three years later at the age of 16, and this was our echocardiogram. The LV had become dilated. The ejection fraction was about 30%, and the question is, why did this happen? Is there a way we could have predicted it? Better yet, is there a way that we could have prevented it? And to approach that, I propose the following outline. First, I do want to review a little bit the physiology and the incidence of pacemaker-mediated cardiomyopathy, then talk about predictors, and conversely, predictors of good outcome, look a little bit into special cases in congenital heart disease, and then take a stab at approach for prevention. So starting with physiology, there are now many elegant studies, and we've seen many of them that show us that RV pacing and subsequent electromechanical dyssynchrony results in late posterolateral LV activation. Oftentimes the LV free wall is activated four times later than normal. This late LV activation significantly alters the Starling curve, which results in loading and strain on the left ventricle, ultimately causing hypertrophy, ischemia, and fibrosis. In a simplified model, we could say that RV pacing leads to asynchronous electrical activation and mechanical dyssynchrony, inefficient work, ventricular loading and strain, ischemia, remodeling, fibrosis, and ultimately, long-term LV dysfunction. But not everyone develops dysfunction, so what's the true incidence of pacemaker-mediated cardiomyopathy? And these are the larger studies I could find. And if you look at it, the definition of pacemaker-mediated cardiomyopathy is a little bit discrepant, but in midterm follow-up, I think it's safe to say that the incidence is somewhere between 6% and 14%. If you look at the more recent studies and based on time dependency, probably it's on the higher end of that. So that means for us that maybe 15% of kids will develop this cardiomyopathy, 85% will not. And why does it happen in some and not in others? Are there predictive factors for mediated cardiomyopathy or good outcomes? These are factors that I found in literature. I'm not going to spend time talking much about the first two. The first, because demographic factors have been spotty and haven't been consistent in their outcomes. And the second, because baseline LV function is a little bit obvious, if you go in with bad function, even mild dysfunction, you're much more likely to get progressive pacemaker-mediated cardiomyopathy. But I do want to spend time talking about the others here, starting with genetics. So is there a genetic predisposition for pacemaker-mediated cardiomyopathy? And there's several groups that have started looking at this. This is a study from China that was recently reported in Nature Reports, and they did a prospective enrollment of all their pacemaker implants in patients with complete AV block and normal function. There were a few hundred patients. Candidate genes were then screened for cardiomyopathy-associated variants, and they were paced for a median of 2.5 years with a high burden of pacing. They found 10 patients where they thought there was true pacemaker-mediated cardiomyopathy. Of those 10 patients, when they did genetic screening, seven of them were found to have pathogenic mutations in cardiomyopathy-associated genes, and the other three had variants or multiple variants in cardiomyopathy genes. When they looked at 20 matched controls who had normal genetic findings, none had clinical heart failure and none had an ejection fraction less than 50 percent. They then transfected rat cardiomyocytes with the exact same mutations and stressed them in similar modalities as pacing and found a very high rate of apoptosis. So they started to conclude that maybe there's a genetic predisposition that we should be looking at. But how about the more classic risk factors, like the burden of pacing? There are robust data in adults, and we're all familiar with David and Most, which showed us that a threshold of greater than 40 percent pacing is a predictor of heart failure hospitalization and events. More recently, Dr. Kiel looked at the Cleveland Clinic experience and consecutive patients receiving pacemakers for AV block with preserved function. He found that the primary predictor was, again, pacing burden, and the cutoff was greater than 20 percent, had a hazard ratio of almost seven. On Kaplan-Meier analysis, indeed, over 20 percent seemed to be a threshold, showing a significant difference in freedom from cardiac dysfunction. So burden does seem to play a role as a risk factor, and it's been shown in pediatrics that we can use algorithms and modalities to try to address this in our children. In similar regards, how about pace rate? So our children that we have that have single-chamber ventricular pacing systems. In this study, they looked at children with VVI pacing, and they hypothesized that at higher pacing rates, the adverse effects of dyssynchrony may be further augmented. In their small study, they found on multivariable proportional hazard regression that a higher VVI lower rate was associated with dilated cardiomyopathy. And on survival analysis, when you divide them into quartiles, lower, set lower rate limits resulted in improved survival from pacemaker-mediated cardiomyopathy. There are several other small studies that seem to support this notion, so maybe a consideration for our kids. How about site selection or lead location as a risk factor? Several studies have suggested that pacing, RV pacing, and the location alters hemodynamics and function, potentially affecting pacemaker-mediated cardiomyopathy. This was really thought to be a big player, but in recent years, as we all know, it's come under controversy. In adults, the PaceProtect study was performed, published in 2015. It was a randomized prospective study comparing RV apex to RV high septal pacing, and they found no difference in development of pacemaker-mediated cardiomyopathy, no difference in change in LVF, heart failure symptoms, BMP, six-minute walk, or mortality. They concluded that in patients with high burden ventricular pacing and normal function, septal pacing does not provide protective effects on LV function, at least within those first two years. In pediatrics, the largest and most complete study is one that was performed by many of you in the crowd, and it was this multicenter, cross-sectional study that had 178 patients, no significant heart disease. They all had a relatively high burden of pacing and median pace duration of 5.4 years. Most of them had congenital AV block, and they were about equally split between epicardial and endocardial systems. They found overall that LV pacing did better than RV pacing. LV pacing showed preserved function as opposed to RV pacing, which showed a decrease over time. In cross-sectional analysis, LV pacing had normal ejection fractions, while RVOT and RV lateral had mean or median subnormal function. And based in percentages, LV apex and LV lateral were predominantly normal function, meaning ejection fraction greater than 55 percent, while the RV pacing sites had a higher proportion of dysfunction. Interestingly, in this study, RV pacing sites also showed no difference. So RV apex versus RV septum was not different. Dr. Karpowich then proposed that maybe all areas of the septum aren't the same, and we're just lumping them together when we should be dividing them into segments. And on this analysis, they looked at those segments. And what they found, that optimal lead implant site was variable from person to person. So there's no universal sweet spot, no one place we should be aiming for. But they did argue that optimization prior to implant with direct hemodynamic and physiologic assessment may be important. They used TPDT. Other studies have used QRS duration, pressure volume loops. More recent ones will look at elegant echo parameters. And I don't have time to go into all of these, but I do think it's something we should continue to look at. So a factor that we do have to consider in our population is congenital heart disease, because it may affect the incidence and the impact of pace-induced dyssynchrony. And just to give a taste for that, we recently saw in Dr. Balaji's presentation that children have a variable presentation of pacemaker-mediated cardiomyopathy dependent on their congenital heart disease. The risk may differ based on your underlying anatomy. The Boston Group has had several evaluations looking at specific diseases. For example, here, looking at congenitally corrected transpositions in dual-chamber pacing. When they looked at them and compared single-site versus biventricular pacing, single-site pacing had a very high percentage of cardiomyopathy, while biventricular pacing did not. And they proposed that all patients with CCTGA who develop AV block should undergo primary BIV pacing to prevent systemic ventricular dysfunction. This morning, we saw that from the simplest to the most complex congenital heart disease, the presence of congenital heart disease can affect the presence of pacemaker-mediated cardiomyopathy. So I do think that congenital heart disease alters how we look at this. We're starting to learn this, and we're seeing several studies come out. And I think as we learn it, it's going to play a role in how we talk about pacing children. So is there a way to approach prevention in these children? I think we can talk about looking at predictors and trying to avoid those. In the current era, a lot of the focus has shifted to physiologic pacing, which is HIS bundle or BIV pacing de novo. Most of the data is in adults. And the most recent study is this one, which is a combined ACC, AHA, HRS guideline meta-analysis. And they looked at BIV pacing and HIS bundle pacing versus RV pacing. And what they found in summary is that physiologic pacing, both BIV and HIS bundle pacing, reduced adverse remodeling. They had lower LVEDD, higher ejection fractions, 5% to 7% higher than RV pacing. There was variable improvement in quality of life and functional measures. And there was no real difference in mortality. More recently, Dr. Ellenbogen looked specifically at HIS bundle pacing de novo and compared that to RV pacing. And he found that HIS bundle pacing had no real change in function over time versus RV pacing, which showed a somewhat predictable decrease. And on survival analysis, HIS bundle pacing had significantly higher freedom from heart failure, death, or cardiomyopathy. They did find, however, that there was a higher risk of lead revision, about 5%, higher capture thresholds, and decreased battery longevity. And this is something that's important for us. It may not matter if you're 60 or 70, but in children where you're going to have procedure after procedure, these effects have to be considered. So is there an algorithm we could follow? And I think the answer is probably no. Not yet. I think we do have to look at them case by case. But I do think there's some generalizations we can make. And we've started doing some of this. I think for one, if patients come in and they have depressed function at get-go, or if they're in a high-risk class, meaning they have cardiomyopathy, LVNC, or congenitally corrected transposition, some of these congenital heart diseases we see, we should consider bi-view pacing them or HIS bundle pacing them de novo. If they're not high-risk and they're normal function, most of our patients, there's probably enough data that we should be avoiding the anterior RV free wall in epicardial systems. And we are asking our surgeons to get to the other side of the LAD, try to get to the left ventricle. We should consider HIS bundle pacing. We're probably not there yet, not at a point where we'd suggest it universally. And targeted septal pacing, plus or minus these acute physiologic measures. I do think we should use measures to limit pacing. And I think we're starting to show that these could help even in our group. And importantly, close follow-up and monitoring is really important. We know that these patients do respond relatively well to resynchronization. So as we see them in clinic and get their echoes and treadmills, we should be watching for this and have a relatively low threshold for putting them forward for a CRT. So in conclusion, pacemaker-mediated cardiomyopathy is increasingly recognized and is not infrequent in children or patients with congenital heart disease. Aim for more physiologic pacing. And in high-risk profiles, BIV or HIS bundle pacing should be considered early. In most patients, emphasis must be placed on personal treatment. And a lot of it is still under investigation. But close monitoring of dyssynchrony and function is important for early intervention algorithms and follow-up. Thank you for your time. ejection fraction if it's a dis-synchronous movement is going down. But is that the right measurement? What should we do, you know? That's an excellent question. And I think if you talk to a lot of our ECHO colleagues, they've moved to more elegant measures. Most of our early studies look at ejection fraction, but there's a lot of evaluation on speckle tracking and strain and other measures to do that. There are a lot of studies also showing that dis-synchrony, and I may have one here, that show that early measures of dis-synchrony can predict our more gross measures of dysfunction. So if we ask our ECHO colleagues to be looking for these measures, even before we see a gross effect on ejection fraction, it can predict the progression of disease down the line. But I do think looking just at ejection fraction, shortening fraction, those kinds of things, we'd miss the boat a little bit. Thank you very much, Jeff. It was a lovely talk. I appreciate it. Thank you. It was a great session.

pacemaker-mediated cardiomyopathy

children

genetic predisposition

physiologic pacing

prevention