While we're loading up, if you have your app for multiple-choice questions, we can get to those, because I'm going to use some of those, but I have two cases. So a man with an integrated bipolar lead at the RBApex undergoes an upgrade from a Medtronic single-chamber ICD to a Medtronic CRTD for heart failure with left bundle branch block. And the programming has not changed for sensitivity. And 12 days later, the lead integrity alert is activated. The lead integrity alert is activated, and there's an electrogram that's stored. Now just to remind you, right, integrated bipolar leads sense between the tip electrode and the shock coil. The alternative dedicated bipolar leads, right, sense between the tip screw and a dedicated ring electrode, so-called dedicated, because it's dedicated to the sensing circuit as opposed to the shock coil, which is integrated into the high-voltage circuit. And we have this stored electrogram. So I have this question for you. Which of the, and you can see, right, we're showing the atrial signal and the integrated bipolar signal, and then which of the following statements is true about sensing? Intrinsic over-sensing is present, atrial over-sensing is present, both are present, neither are present. And do you guys have the voting app, and would it show up here? Okay, so you can vote, but, and, oh, you've got, good, okay, we got the, you guys are going to do that, okay. And does it tell us how many votes are in yet, or, oh, I'm, because I don't, I don't seem to have it on my slide. It's populating, okay. When you guys think it's populated enough, let us know. Okay, so about three-quarters of the people mark this as ventricular over-sensing, and a quarter as atrial over-sensing. So the correct answer is that ventricular over-sensing is present. We can tell this because there are many, there are more, so the basic rule about over-sensing, first of all, is that there are more sense markers on your device interrogation than they're actually true cardiac electrograms. And so we have a period here where there are multiple sensed events on the ventricular channel, but no actual true cardiac electrograms. On the other hand, the atrial channel has roughly the same number of electrograms throughout, and about the same amount of sensing with a little bit of atrial under-sensing here. All right, so the next question is, that we have is, what is a cause of over-sensing after we've just done a generator change? And I, and you know, sort of my kind of rule of thumb without, I'm going to do this slightly more didactically than Carson, just, you know, I always think that there are like three possibilities. One is that there's some type of connection problem, and those might be a set screw problem, it might be incomplete pin insertion, rarely it might be error in the header. The other thing that's possible is it's possible to damage a lead intraoperatively. We've all dissected leads out of rats' nests that when we've had to use more cautery neural lead than we would have liked to, and intraoperative lead damage may not be apparent at the time we do the procedure. And finally, there may be sensing differences between the old and new generators for a variety of reasons. So now that we have that, I'm not going to, what I'm going to ask you is, which of the following statements is true about the type of over-sensing we see here? Sometimes it can be reproduced by pectoral muscle exercise like Carson showed you and reviewed with you in the last session. Sometimes it can be reproduced by deep breathing. See replacement is required, or it can be resolved by disconnecting and reconnecting the RV lead. And let's have you vote. Well, maybe it's a bad question. You know, that's one of the problems with, okay. So do we have any comments from our panel here? Or should I, I'm not putting anybody on the spot. I'll just make one sort of, I like to think about these things sort of as in principles. And so the principle is, you know, you've come on to a new problem in a chronically stable RV lead and you have intervened or changed something. And so for instance, I think C is as a last resort in principle, because something that you have changed has brought about this event to happen and something that was historically stable. And then the question is, what is it about the change? And then what you're doing with your priority set here is you're thinking about the mechanisms by which you could have over-sensing and then the probability that they are an explanation and then your ability of what you need to do about them. So for example, the notion that the lead has suddenly developed a problem that it didn't have prior to the implant, the only explanation for change in true lead performance would then be something that happened surgically. And if you think that's an improbable thing because of the fastidious nature of this surgical team, et cetera, then the easy thing to do is to ask yourself, is there a platform issue or connectivity, is there an issue around the programming and settings in the device or the nature of the integrated bipolar that could be an explanation? Because that's something you can troubleshoot, right? And the other alternate explanations are more difficult to troubleshoot is my point. Another one is sort of, do you see what's going on with the impedance and thresholds? I didn't, no, that's a good point and I should have told you that. Impedance and thresholds are both fine. Does that help you? Well, I mean, it doesn't seem like there are any limitations. No, no. It's a yes-no question. I mean, can you, can you, no, no, I mean, no, no, I mean, does it, does it, is it, does it tell you what the right answer is? And the answer is no, it doesn't, because most lead failures, most pacesense lead failures and defibrillator leads, right, insulation breaches or fractures present with normal impedances and normal pacing thresholds. Now, your point is if they're abnormal and I'm hiding it from you, that would be dirty, wouldn't it? That, that, that, that's a problem. But most actual lead failures will present with an over-sensing problem and normal impedances and normal pacing thresholds. But the corollary is not true, right? So if you told me the impedance was, was suddenly low or high. No, that's correct. And if, but, but in this case, the impedance is, yes, and I think, however, the gentleman in the front row made a good point, which is that most connection problems present with abnormal impedance. And so the fact that this impedance is normal while not being very helpful in, in reducing the likelihood of a lead problem, a lead damage is in fact helpful in suggesting this is probably not a connection problem, right? How late have you seen connector problems? This is 12 days. Yes. So this is an excellent question. But we have seen, so the next question is, I guess the question would be, how long do connection, how long did doc, how long does it take for doctors to recognize connection problems? Or no, no, for patients to recognize connection problems, for devices to see the problem. But the key point is there are connections. So, so and the reason I ask that is this, most diagnosed connection problems happen within the first week after surgery. Most misdiagnosed connection problems happen between one and six months. So I can't remember a missed, so I want, and I know this because I looked at these data. And, and, and, and I, and I, not because I happen to be encyclopedic about this, because I once did this analysis personally. So I can tell you that, that connection problems are diagnosed by doctors are, and leads are, are as lead failures and leads are replaced for connection problems at least six months after implant. And, and in fact, I think there are, there's a, there's a very widely quoted paper that says, you know, failure rates of, of the recalled Fidelis leads went up dramatically after lead, after generator changes and without any actual, you know, lead validation. And I'm certain much of that is, was connection problems. However, I want to, I want to focus on this question for a moment. And we're going to get to why the correct answer is this. Sometimes this can be reduced by deep breathing, but broadly when I see an over-sensing problem and you know, everybody has their ways of thinking about over-sensing and sort of, you know, my sort of first set of rules is, does the over-sensing vary on a beat to beat basis with the cardiac cycle? Does it not? And in this case, it doesn't vary on a beat to beat basis with the cardiac cycle. So it's probably, unless it's a lead failure, it's probably not an intracardiac problem. And we're thinking about these things like lead failure, connector issues or mild potentials. Now what I want you to recognize is that as opposed to the skeletal my potentials, the muscle my, I'm sorry, the pectoral my potentials that Karsten talked about, these are diaphragmatic mild potentials. Diaphragmatic mild potentials have a characteristic visual pattern. All mild potentials are high frequency signals in the range of 80 to 200 Hertz, but diaphragmatic mild potentials are uniformly and monotonously low amplitude signals. And I want you to, you know, think about that. Sometimes, you know, pectoral my potentials can be pretty big signals. Diaphragmatic mild potentials are always small and that's why they show up in sensitive devices. Here are some other examples of pectoral mild potential, I mean, excuse me, of diaphragmatic mild potentials, just to show you the, you know, that they're monotonously high frequency, low amplitude. So it's a pattern that is good to remember. So now diaphragmatic mild potentials may occur after generator change when the new and old generators have different dynamic sensing profiles. And there are three times this can occur. One, I'm going to talk about why that might occur after CRT upgrade, even though the sensitivity is the same, with manufacturer specific changes and also with changes in filtering. So in this particular case, if you have a single chamber pacemaker in a patient with left bundle branch block, you're typically sensing all the time. And there is a dynamic sensitivity threshold after each sense beat that starts up at Medtronic devices. It's 50% of the sensed amplitude in Boston devices. It's 75% in Abbott devices. It varies, you know, there's programmable options there. But Medtronic devices post pacing always go to a more sensitive setting and a lower start of the decay profile of the sensing threshold. And the reason for that is to make sure that there's a longer blanking period after pacing beats. We want to make sure we're not missing ventricular fibrillation. So in this particular case, even though we have not changed the sensing floor at 0.3 millivolts, in sinus rhythm, the sensing floor is never reached, but during ventricular pacing it is. Now just very briefly, because the other things are rare, Dr. Narayan, who's in the audience, just recently had this case report in PACE of a different cause of inappropriate detection, inappropriate over-sensing of diaphragmatic myopotentials, when they changed manufacturers that went from one with a sensing profile that was less sensitive to one that was more sensitive for the same R-waves. So if you change your manufacturer, change your sensing profile, you can get changes in over-sensing, as shown here. And in fact, there's one manufacturer, Abbott, which has a programmable filter setting. And the nominal now is this low-frequency attenuation filter, which enhances the sensitivity to high-frequency signals so that, as several reports have shown, this one by Sylvain Plo from Bordeaux and his colleagues, you can fiddle with the sensitivity all day long and nothing will happen as long as that filter's on. The minute you turn that filter off, the over-sensing goes away. So the point is, the changes in the sensing profile may in fact be the cause of over-sensing after a generator change. It doesn't mean you have to go in and change the system. All right. Well with that enthusiasm, I'm going to move on to my second case. Do I have time? No, I don't have time. Oh, I'm sorry. Are there questions? Do I have time to do a second case? Absolutely. Okay. Oh, yes, please. You know, we're going to have you climb on up here. You're doing so well with us. No, no, this is great. So I think it's an excellent question. The question is, if we haven't tested the safety margin for VF sensing at generator change, and we often don't do that, how do we know? I'm sorry. Let me go back two steps. The most common solution to this problem is reduce the program sensitivity. More appropriate concern is, how do we know we'll still sense VF, right? And that's the question you're asking. And I don't have a perfect, I don't have a, you know, I don't have an answer from Mount Sinai for you that, you know, or that, but in general, it's unusual to get into bad oversensing problems if you set the sensitivity at 0.45 or even 0.6. Much beyond that, you are looking for trouble. But in this case, we programmed the device to 0.45, which I think, honestly, if you have an integrated bipolar lead, should probably be nominal for a number of reasons on Medtronic devices. Just like Boston nominal sensitivity on all their integrated bipolar leads is a less sensitive setting. And I think that's probably okay. So I guess my rule of thumb, if I'm going to above 0.6, and we haven't tested in a while, we probably need to retest. Chuck, does the sensed R-wave influence that at all, if the R-wave's 6 or 16? So that is a really good question. And so what we know is that the vast majority of VF undersensing in modern devices is not continuous undersensing of VF. Because all VF counting right is probabilistic, 30 out of 40, 18 out of 24. Most real VF undersensing that happens today actually has some very large signals and some very small signals. The problem is if you have a 3 millivolt VF electrogram followed by five that are down at 0.3, that the dynamic adjust sensitivity can't adjust fast enough. I'm not aware of a really good correlation between R-wave amplitude, sinus rhythm. I mean, it's always a question of the sinus rhythm R-waves are big. That usually means VF electrograms are good enough. I think if you have, there's considerable data about Medtronic devices that if you have sinus rhythm R-waves down to 3 millivolts, that certainly between three and above, that you can't tell who's going to have VF undersensing. That once they get too small for that, then I think it's really small R-waves, as you pointed out, are God's way of telling you that you should probably, you know, revise this lead. Chuck, you know what's bothering me? I'd love to. In particular, not in general. We don't have that much time, Nora. What are we obligated to ask manufacturers, representatives who are not always engineers in terms of the stuff you've just shown? You really have to know your product. You have to know your general. A lot of us, I don't know about this room or this society, a lot of us, house staff, PCPs, don't know what to ask before they even make the call. Yeah, so I think the thing to do is you don't have to be a sensing nerd like I am and know all this stuff, but you should know, as part of your troubleshooting problem, over-sensing after generator change. It may be you did everything right. It may be that there's a subtlety that you don't know about in changing the system. The question you want to ask technical support, not necessarily your rep, is, okay, tell me about the sensing profile of this device in comparison to what I had in there before. I think that's the way I would phrase the question, and that's why I wanted to bring it up, because I've seen a number of cases re-explored for this reason. Should I just quit? Well, let me just do ... No. Okay, good. I want to do this case because I like it, and I'll try to go through it slightly faster. A patient underwent implant of a primary prevention single-chamber ICD with a dedicated bipolar lead, and device detected VF was recorded four weeks later. Which clinical intervention is indicated? And we can vote on this. And yes, I'm not telling you about the pacing thresholds. It is true. I can tell you that, but that might give it away. So I'm going to be annoying about that. All right. So let's think about this. I'm sorry. There we go. So this is the correct answer, and we're going to come to it in a minute. And it's a pattern I want you to notice. So as opposed to ... So you got the theme of this lecture is oversensing. So as opposed to non-cyclical oversensing, I think here we could ... I think most of you got the idea something is being oversensed, and your suggestion for that was change ventricular oversensing. I mean, every time you see this alternation, there are two possibilities. You have true alternating cardiac electrograms like bidirectional VT, which is really rare, but that is almost never seen in both the sensing and the shock electrogram. Or you're oversensing one signal for cardiac cycle, and those are either P waves, T waves, or double counting the R waves. So the problem, though, is which is the R wave ... Oh, this didn't show up. Which is the R wave, and what is this other signal? And how do we know? Well, one of the common things that shows up when we see these problems is could this be T wave oversensing? And their answer is, well, there's some things that are similar to T wave oversensing we see here. For example, T wave oversensing commonly occurs when we have low amplitude R waves on the sensing channel, and here we have a low amplitude R wave on the sensing channel. The near field complexes alternate in frequency content, and we have that here. What is different about this pattern than T wave oversensing is that the far field electrograms also alternate, and in a typical case of T wave oversensing, the far field electrograms don't show funny looking T waves, right? It's almost always only the near field signal. Furthermore, these aren't just small R waves, right? These are tiny signals on the sensing channel, and the shock channel also has tiny signals. This does not happen in a lead that's in the RV apex. You should always, as Karsten was pointing out, look at your electrogram amplitudes. This is a lead dislodged to the atrium. Why is it we care about P wave oversensing in a lead dislodged to the atrium in an implantable defibrillator? The reason is the patient may not need pacing, so you're not going to know about it. It's not going to bother him. The lead may sit in the RV for a while, then float back into the atrium, and you won't know about it until the patient gets sinus tachycardia, and then when the sinus tachycardia rate is half the VF or VT detection rate, the patient's going to get a shock. That shock has a 50-50 chance of landing on the true R wave or landing on the true P wave. If you are in sinus tachycardia, there's a pretty good chance that your shock, if it lands on the P wave, is going to be in the ventricular vulnerable zone. It's also a pretty good chance that if you're delivering the shock in the atrium, that no matter how strong it is, the shock field in the ventricle is going to be weak, so it's like delivering a weak-inducing P wave shock. Then the worst part is the lead bipole may be sitting in the atrium sensing atrial electrical activity, so it senses atrial electrical activity fine, but it's not going to sense VF well from the atrium and is going to let the patient die. Identifying lead dislodgements in defibrillator patients is actually important. I want you to get this pattern of single over-sensed event in the sensing channel with corresponding events in the far field channel, very low amplitudes, should make you think of a lead dislodgement in the atrium. Where do you think that lead is physically? That lead is in the middle of the atrium. Lying against the wall for over-detection like that.

oversensing

implanted cardiac devices

electrogram

lead integrity alert

ventricular oversensing

lead dislodgement