Circulation October 18, 2016 Issue

Carolyn: Welcome to Circulation on the Run, your weekly podcast summary and backstage pass to the journal and its editors. I'm Dr. Carolyn Lam, associate editor from the National Heart Centre and Duke National University of Singapore. Have you ever wondered what the clinical implications of very brief episodes of device-detected atrial tachyarrhythmias are? Well, we will be discussing this with novel data from the RATE registry in just a moment. First, here's your summary of this week's journal.

    The first study provides the first evaluation of the Sweden nationwide abdominal aortic aneurysm screening program. Of almost 303,000 men invited for screening, 84% attended. The prevalence of screening detected abdominal aortic aneurysm was 1.5%. After a mean of 4.5 years, 29% of patients with aneurysms had been operated upon with a 30-day mortality rate of 0.9%. The introduction of screening was associated with a significant reduction in aneurysm-specific mortality. The number needed to screen to prevent 1 premature death was 667, while the number needed to operate on to prevent 1 premature death was 1.5.

    Furthermore, the authors showed that their screening program was highly cost-effective in the contemporary setting in Sweden. These findings confirm results from earlier randomized controlled trials in a large population-based setting, and may be important for future healthcare decision-making. This and the diverse requirements for efficient population screening for abdominal aortic aneurysm, from program management to maintaining skills in open repair are discussed in an excellent accompanying editorial by Dr. Cole from Imperial College London.

    The next study looks at thoracic epidural anesthesia and suggests that caution may be needed in patients with or at risk for right ventricular dysfunction. You see, thoracic epidural anesthesia involves blockade of cardiac sympathetic fibers, which may affect right ventricular function and interfere with the coupling between the right ventricle and right ventricular afterload. Dr. Wink and colleagues from the Leiden University Medical Center therefore used combined pressure volume conductance catheters to study the effects of thoracic epidural anesthesia on right ventricular function and ventricular pulmonary artery coupling in 10 patients scheduled for lung resection.

    Thoracic epidural anesthesia resulted in a significant reduction in right ventricular contractility, stroke work, dP/dt max and ejection fraction. This was accompanied by a reduction in effective arterial elastance such that ventricular pulmonary coupling remain unchanged. Clamping of the pulmonary artery increased right ventricular contractility but decreased ventricular pulmonary coupling. These effects of increased afterload were the same before and after thoracic epidural anesthesia. In conclusion, therefore, thoracic epidural anesthesia impaired right ventricular contractility but did not inhibit the native positive ionotropic response of the right ventricle to increase afterload. These findings are clinically relevant for daily practice in cardiothoracic surgery because pulmonary hypertension is frequently encountered, and right ventricular function is an important determinant of early and late outcomes.

    The next study suggests that the use of point of care hemostatic testing may have a place in the management of patients undergoing cardiac surgery. Dr. Karkouti and colleagues of the Toronto General Hospital hypothesized that point of care hemostatic testing within the context of an integrated transfusion algorithm would improve the management of coagulopathy in cardiac surgery, thereby reducing blood transfusion. They therefore conducted a pragmatic multi-center stepped-wedge cluster randomized controlled trial of a point of care based transfusion algorithm in 7,402 consecutive patients undergoing cardiac surgery with cardiopulmonary bypass in 12 hospitals in Ontario, Canada. They found that the trial intervention reduced rate of red cell transfusion with an adjusted relative risk of 0.91 and a number needed to treat of 24.7.

    The intervention also reduced rates of platelet transfusion and major bleeding but had no effect on other blood product transfusions or major complications. These findings that point of care testing improved management of coagulopathy in cardiac surgery support the consideration of their broader adoption in clinical practice.

    The next study provides experimental evidence that brings us one step closer to therapeutic targeting of arterial leukocyte recruitment in the context of atherosclerosis. In this study from first author Dr. Ortega-Gómez, corresponding author Dr. Soehnlein and colleagues from LMU Munich, authors focus on cathepsin G, which is stored in neutrophil and azurophil granules and discharged upon neutrophil activation. They studied site-specific myeloid cell behavior after high-fat diet feeding or TNF stimulation in the carotid artery, the jugular vein, and cremasteric arterioles and venules in APOE E and Cathepsin G-deficient mice.

    Their studies revealed a crucial role for Cathepsin G in arterial leukocyte adhesion, an effect that was specific for the arteries and not found during venular adhesion. Consequently, Cathepsin G deficiency attenuated atherosclerosis but not acute lung inflammation. Mechanistically, Cathepsin G was immobilized on arterial endothelium, where it activated leukocytes to firmly adhere, engaging endocrine clustering, a process of crucial importance to achieve effective adherence under high-sheer flow.

    Therapeutic neutralization of Cathepsin G specifically abrogated arterial leukocyte adhesion without affecting myeloid cell adhesion in the microcirculation. Repetitive application of Cathepsin G-neutralizing antibodies really allowed the inhibition of atherogenesis in the mice. Taken together, these findings presented evidence of an arterial-specific recruitment pattern centered on Cathepsin G adhesion, thus representing a potential novel strategy and target for the treatment of arterial inflammation. Well, that wraps it up for the summary of this week's journal. Now, for our featured discussion.

    Our feature paper for today discusses the clinical implications of brief device-detected atrial tachycardias and really novel findings from the RATE registry. I'm so happy to be here with the first and corresponding author, Dr. Steven Swiryn from Feinberg School of Medicine, Northwestern University. Hi, Steven.

  Steven: Good morning.

  Carolyn: We also have with us Dr. Mark Link, associate editor from UT Southwestern. We all know that prolonged episodes of atrial tachycardia or atrial fibrillation are associated with increased risk and that if we anticoagulate those with a high CHA2DS2–VASc score, we can lower the risk of stroke. Now, the European Society of Cardiology guidelines also say that recent data reinforced the assumption that even brief episodes of silent atrial fibrillation may convey an increased risk of stroke. We also know that prior studies have looked at device-detected atrial fibrillation. Steven, I'd really love if you could start by telling us what makes your study different. What was the main thing you were trying to look at?

  Steven: Well, one reason it's attractive to use the device population, patients with pacemakers or defibrillators, to look at these issues is because devices have a very high likelihood of detecting episodes of atrial fibrillation whereas symptoms or single 12 EKGs miss a lot of atrial fibrillation, so the sensitivity is much higher, although not perfect. The problem is that very brief episodes of atrial fibrillation are very poorly detected by devices. The specificity of automatic detection is very low, such that all previous studies until the RATE registry have excluded any episode of atrial fibrillation detected by a device less than 5 minutes in duration because they're unreliable. A lot of them turn out to be false positive detections. Our study was designed to evaluate whether even very brief episodes of an atrial tachyarrhythmia might also be associated with risk of clinical events and might or might not warrant anti-coagulation.

  Carolyn: Ah, that's interesting, so you really helped to answer how brief is "brief" when we need to talk about device-detected atrial fibrillation. Could you expand on how you actually defined "short episodes" here?

  Steven: Right. A short episode for the purpose of the RATE registry was defined as an episode where the electrogram that we scrutinized had both the onset and the offset of the episode within the same electrogram tracing, so although we can't put a specific time duration on it because that wasn't part of the criterion, it's typically less than 20 seconds or so, although not always, whereas a long episode was defined as an electrogram where either the onset and/or the offset was not captured by the device memory and therefore we don't know the duration. Some of those may not have been very long, and some of those may have been extremely prolonged episodes. That allows us to actually scrutinize the electrogram. We looked at 37,530 individual electrograms using 8 teams of adjudicators, each with a physician and a field clinical engineer from the device company so that we could actually say definitively, "Yes, this was atrial fibrillation," or, "No, it wasn't."

  Carolyn: This is the first study to really look under that 5-minute limit of atrial tachycardias. What did you find?

  Steven: Well, we found that in contrast to prolonged episodes, short episodes of atrial tachyarrhythmias were not associated with an increased risk compared to those without atrial fibrillation of pre-defined clinical events, including death from any cause, heart failure, stroke, hospitalization for atrial fibrillation, and a few other smaller events.

  Carolyn: This was over a 2-year follow-up period, is that right?

  Steven: The median follow-up was slightly less than 2 years, that's right.

  Carolyn: What I really was struck with was also the second finding, the propensity to develop longer episodes. Could you expand on that?

  Steven: We reasoned that in the clinic, one might be faced with a short episode was we defined them, and then you don't know what's going to happen for the next 2 years to bring to bear the results of our study. We looked at if your first episode was short, what was your likelihood over the full follow-up of the study of progressing to longer episodes. About 50% of patients who had their first episode as only a short episode progressed to a longer episode over the full follow-up and therefore were in the long category for the rest of the results. Half of them never got a longer episode.

    It was, as one might imagine, if you had your first short episode very early in the study and had a longer follow-up, you were more likely to end up in the long category, and if you had very frequent short episodes, you were also more likely to end up in the long category by the time the full follow-up was over with. Having an initial short episode is not a guarantee that you're never going to get a long episode and that you'll never acquire a consideration of anti-coagulation.

  Carolyn: That was a very important message to me as well because it meant that although I can be secure or reassured by these data for very short episodes, I needed to look out for the development of longer episodes, at least that's what your registry showed over 2 years of follow-up. I'm curious, Mark, what were your take-home messages because that leaves us with a bit of a conundrum. What do we do about anti-coagulation in these patients?

  Mark: I think this study is a big help to the practicing electrophysiologist and practicing cardiologists. It's a very ledger number of patients with a lot of episodes of afib. It's reassuring to me that the shorter episodes of afib as defined by the study, the individuals did not have a higher incidence of stroke compared to those with no episodes, so it's reassuring and very important clinically as I go through my practice.

    I do look forward to more analyses and more data from this study because although now we know that episodes less than 20 seconds are in all likelihood not going to need anti-coagulation, we still don't know about those from 20 seconds to 5 minutes. Hopefully with more analysis of this study we'll get that answer also.

  Carolyn: Steven, do you agree with that?

  Steven: We would love to have that. At first glance, you would think that devices would give you all of the data you needed because after all, they're monitoring the patient 100% of the time, but there are difficulties with that because device memory is limited, and you don't get electrograms that go on until the termination of atrial fibrillation even if the device were accurate in determining when that termination was because depending on how the device was programmed and depending on whether it was a more modern device later in the trial or earlier and had more or less memory, it cuts off after a limited amount of time, and you don't see necessarily how long the duration is.

    Now, you can use device-based data. The device gives you its estimate of how long the episode is, but those are not as reliable as adjudicating the electrograms and actually looking at them. Those data would be a little softer than the main results if we get there.

  Mark: That was the data that was used for all of the other studies, was [transassert 00:14:51]. It would be comparable to those other studies. I still think it would be very important data that I'd love to see.

  Steven: Okay, well, I agree. I think it would be very interesting to look at that and a number of other things. We have a number of other things we could do with this database. There are a number of substudies that are in progress. For example, one interesting one is there were some instances we found, because we actually looked at these electrograms, there's something that we termed "competitive atrial pacing," where the device will pace at times when we as clinicians would not want to pace. For example, pacemaker-mediated tachycardia would be an instance of that, but then you can pace in the atrium inappropriately. There's a rhythm called repetitive non-reentrant ventricular atrial systole, which, although it's exotic to all of us, actually turned out to be fairly common where there's pacing in the atrium that occurs for various reasons when we want it to.

    We actually saw instances where the device itself induced atrial fibrillation. It wasn't that common, but we did see it. We have a substudy that we're working on about the subjective competitive atrial pacing to see how much of that there was and of what, if any, consequence that was. That's one of the things that's been done. Because we scrutinized these so carefully, we tracked morphology and atrial rate at least as a crude estimate, and we have those data, so we could actually evaluate whether if something looks very, very rapid and disorganized as opposed to more organized electrograms at a slower rate, did that make any difference. We don't have any results for those analyses yet. I agree with Mark that the intermediate durations would be interesting to look at.

  Carolyn: I agree too, and I'm really grateful for you sharing those thoughts. Very grateful for both of you for your time today. I just have to congratulate you. I completely agree this paper fills an important knowledge gap, and congratulations once again.

  Steven: Thank you very much.

  Mark: Thank you.

  Carolyn: Thank you for listening. You've been listening to Circulation on the Run. Please tune in next week.