Time Stamps

Sponsor: We’re thrilled to have AMBOSS as our sponsor. AMBOSS for Clinicians is a next-gen all-in-one reference tool providing clinical info in half the time of any other resource. AMBOSS is blazing fast and presents evidence-based answers created and reviewed by a team of 100+ US-trained physicians.

Use the code “COREIM-AMBOSS” to get a free month of access on https://go.amboss.com/CoreIM-E9


Show Notes

Diagnosis and Duration of Atrial Fibrillation

  • The association between the frequency of atrial fibrillation episodes and thrombotic risk is less certain.

Atrial Fibrillation in Non-Cardiac Surgery or Acute Illness and Long and Short Term Outcomes

  • Should we think of patients who develop afib after acute illness or non-cardiac surgery differently from those who have pre-existing atrial fibrillation in the community?
    • Atrial fibrillation diagnosed after sepsis or non-cardiac surgery (see sources 1, 2, and 3) confers an elevated risk of long term recurrent atrial fibrillation and thromboembolism.
  • Does putting patients with atrial fibrillation on anticoagulation after critical illness or non-cardiac surgery lower the risk of stroke?
    • Non-cardiac surgery: There is evidence that the risk reduction of anticoagulation for thrombo-embolic stroke is similar between patients with atrial fibrillation diagnosed after non-cardiac surgery and in patients with pre-existing atrial fibrillation.
    • Sepsis: There is a paucity of evidence (see sources 1 and 2) on the effect of anticoagulation for thrombo-embolic stroke for patients with atrial fibrillation diagnosed during sepsis.
    • There are no clear guidelines on anticoagulation in these populations and practice patterns vary. The decision to initiate anticoagulation must be individualized based on a patient’s unique risk profile (using information such as CHA2DS2-VASc, AF burden, and bleeding risk).
  • Why is atrial fibrillation after cardiac surgery thought of differently?
    • Due to direct cardiac manipulation, atrial fibrillation following cardiac surgery is common and often resolves with cardiac healing. It is distinct from atrial fibrillation following non-cardiac surgery and warrants cardiac consultation to determine the optimal management.
  • Is there an increased short-term risk of stroke for those newly diagnosed in the hospital with atrial fibrillation following sepsis or non-cardiac surgery?
    • Non-cardiac surgery: Atrial fibrillation diagnosed after non-cardiac surgery is associated with an increased risk of 30 day or in-hospital ischemic stroke. However, it is important to consider the risk of bleeding post-operatively as well (see below).
    • Sepsis: For those with newly diagnosed atrial fibrillation during severe sepsis, thromboembolic stroke risk is likely higher during the sepsis episode than for those with pre-existing or no atrial fibrillation. However, anticoagulation during sepsis did not reduce  this risk and did increase clinically significant bleeding risk. This suggests it would be prudent to ensure a patient has fully recovered from their sepsis episode before considering anticoagulation.

Timing of Postoperative Anticoagulation

  • Surgical considerations that may delay initiation of anticoagulation postoperatively include more extensive surgeries, trauma patients, and surgeries in confined areas where hematoma formation may be catastrophic (cranial, spinal, cardiac surgeries).
  • The PAUSE trial supports resumption of DOACs 1 day after a low-bleeding risk procedure and 2-3 days after a higher risk intervention, provided hemostasis is achieved.
  • The CHEST 2022 guidelines on perioperative anticoagulation management support the following in stable patients:
    • Resuming warfarin 12-24 hours after surgery
    • Resuming DOACs, enoxaparin, or bridging anticoagulation
      •  ~24 hours after low to moderate bleeding risk surgeries/procedures
      • ~ 48-72 hours after high bleeding risk surgeries/procedures

Utility of Post-Discharge Cardiac Monitoring in Newly Diagnosed Atrial Fibrillation Following Non-Cardiac Surgery or Acute Illness

  • Continuous cardiac monitors record all telemetry information and are then sent in for analysis.
  • Loop recorders (also known as event recorders) only save events triggered by a patient input or device detection.
    • Some devices are hybrids, recording all data continuously while also isolating triggered events. 
  • Holter monitors are continuous monitors that often record for 24-72 hours. This can be useful to assess rate control in patients with permanent or frequently paroxysmal atrial fibrillation, but may not be enough time to detect infrequent episodes of atrial fibrillation.
  • Patch monitors serve as both continuous and event monitors and generally record for 14 days. This can allow for improved detection of fibrillation episodes and help characterize the duration and frequency of these episodes.
    • It is possible that 14 days may not be enough time and infrequent episodes of atrial fibrillation may still be missed by this monitor.
  • Longer-term loop recorders may be required to detect more infrequent episodes of atrial fibrillation.
    • These include external/event loop recorders (which may be worn for 30 days) and implantable loop recorders (which may be worn for up to 3 years).
    • There are no guidelines regarding when to discontinue anticoagulation if no atrial fibrillation is detected. Longer monitoring and repeat discussions with the patient are often part of an ongoing risk-benefit discussion
  • There is some evidence that wearable technology may be helpful in the detection of occult atrial fibrillation.

Discussing Atrial Fibrillation and Anticoagulation Risks and Benefits with Your Patient

  • Anticoagulation adherence is low, about 50-60% in those on DOACs with newly diagnosed atrial fibrillation.
  • Risk-benefit discussions regarding anticoagulation often involve new terms and multiple statistics. These factors, in addition to recent illness and a stressful hospitalization, often make it difficult for patients to fully engage in these discussions.
  • Pausing to check in with the patient their understanding of atrial fibrillation can help tailor how to best present the nuances of atrial fibrillation and how much to present. 
  • Clear explanation with the patient of their post-discharge plan and communication with their outpatient providers can help ensure a safe transition of care.


A: Welcome to Gray Matters, where we unpack how medical management is rarely black or white.

J: …and we go on deep dives along the way! I’m Jason Freed, a hematologist at Beth Israel Deaconess medical center.

A:  I’m Ali Trainor, a pulmonary and critical care fellow at Beth Israel Deaconess Medical Center and Mass General Hospital, and today we’re also joined by our friend and colleague Nick Villano

N: Hi, I’m Nick Villano, and I’m a hospitalist at Beth Israel Deaconess Medical Center.

J: So Nick, I understand you have a case that’s going to take us back into the gray zone?

N: Yeah I do Jason. I’m pretty excited to talk about this case, it’s about afib and some other common issues that come up all the time for us. And it actually had me go down 5 deep dives , so get ready for a lot of tangents here.

A: Let’s hear it!

N: So, this happened when I was on the inpatient medicine consult service. At 3pm on a Sunday we were consulted by general surgery. The patient was a 72 year old woman with a history of GERD, diabetes, and poorly-controlled hypertension. She was 2 days post-op from open cholecystectomy and washout for perforated cholecystitis and peritonitis. So at this point she was recovering and on antibiotics. But today, the primary surgery team noticed new atrial fibrillation noted on telemetry. And that’s why they consulted us, the medicine consult team.

J: Ugh post-op afib. I feel like we are about to be squinting at tele strips to see if it’s really there. 

N:  Oh trust me, I know. The surgery team had already done their squinting and called us in for backup. And of course by the time we got there…she was already back in sinus rhythm. 

J: Nicely done. Another therapeutic consult. In all seriousness, though, what were you able to capture? Was there at 12 lead or was it just on the telemetry?  

N: See this is where the first challenge came in. It was only 2-3 runs of about 10-15 seconds on telemetry that had alarmed for atrial fibrillation. But when I saw it, it was just hard to make out anything. Very erratic baseline. If you had told me she was just brushing her teeth I would’ve believed you.

A: The timeless struggle in diagnosing Afib. And once fib is on the chart, it almost never comes off. And afib has a lot of implications for the patient!

N: I know, so I’m sitting there wondering what to say to the resident on the consult service with me, not to mention the primary surgical team that’s waiting for an answer. Was this Afib? How clinically important is something that brief?… So for our first deep dive I really wanted to explore how do you diagnose new atrial fibrillation and what can the duration of the afib actually tell you?  So I sat down with Dr. Pooja Jagadish to talk about some of the takeaways from a recent article of hers on postoperative atrial fibrillation.

Dr. Pooja Jagadish: The definition of clinical atrial fibrillation to diagnose is either 30 seconds on telemetry or some sort of wearable like an apple watch or equivalent device, like EMA cardio or a full 12 lead EKG. If you’ve got episodes that are shorter than this, they don’t meet criteria for clinical atrial fibrillation

A: So to confirm, you technically need 30 seconds on tele for the diagnosis.

J: OK you only need 30 seconds of afib to be afib, but is like 31 seconds of afib the same as 31 days of afib? 

N: And does only 29 seconds of AF make you feel better? 

J: Right, like is afib really that binary? Or does the frequency and or duration of Afib episodes itself tell us something about their risk for stroke?

Dr. Jason Matos: So from the little data that we have, the number of AFib episodes in inpatient is probably not as relevant as longest episode. For example, five 30 second episodes would not be as significant to us as a single four or six hour episode. And that, that’s typically how I think about it.  And so the simplest thing and what I lean toward most would just be the longest episode someone has over sheer number of episodes.

N: That was Dr. Jason Matos, another one of our cardiology experts. So Dr. Matos says, there is actually some good data that correlates a patient’s longest episode of atrial fibrillation to that patient’s stroke risk. And that’s the ASSERT trial. As a caveat this study is looking at patients with chronic AF in  the outpatient setting.  But it followed these patients for 2 and a half years and found that patients with longer episodes of fibrillation on cardiac monitor…had a higher risk of stroke.

A: So you’re saying that independent of the CHADS2vasc score, how long you’re in AF impacts your stroke risk?

N:  Yep the main takeaway from the trial is that stroke risk increased with increasing duration of the longest afib episode. Patients whose longest episode of afib was less than 6 minutes … had a 0.6% yearly incidence of stroke or systemic embolism … those with any episodes greater than 6 minutes … had a 1.7 % yearly incidence of stroke or systemic embolism. And in subgroup analysis, patients with any single episode of fibrillation 18 hours or longer had a stroke or systemic embolism incidence of nearly 5% per year. 

J: Dang I mean you know me but those numbers are astonishing to me. But looking at those 3 groups you mention here with distinct stroke risk and their risk is almost ten fold different between the patients with short episodes (less than 6 mins) and really long episodes (greater than 18 hours).

A: And then there’s your patient – she only had 10- 15 seconds of possible afib on tele – that doesn’t meet the ESC definition of Fib and we are even more in the grey. 

N: That’s right. The evening we were consulted we just saw some very brief episodes of possible fib buried in artifact … But then, the next morning, we sat down to check her tele and … we see 2 hours of clear Afib on monitor.

A: OK, so it seems we are out of the grey zone, at least in terms of diagnosis.

N: Yeah what can i say …Her heart finally showed its cards.

J: That’s a good joke, I feel like making the diagnosis of new-onset atrial fibrillation just raises more questions for me. I mean she had Afib in the setting of a recent major surgery and an infection. Is she really at ongoing risk for atrial fibrillation and stroke when she gets better and doesn’t have those things anymore?

N: I’ll be honest, I always thought of the garden variety afib that happens in the community as being in its own bucket … basically like “pre-existing” AF … and I’ve lumped atrial fibrillation after surgery or during acute illness into a separate bucket … kind of like “sick afib.” 

J: I’ve never used the term sick afib, but i think have similar mental categories. they got afib while they were sick in the hospital. 

N: Just to be clear, we’re talking about afib after non-cardiac surgery. Afib after cardiac surgery is super common, usually self-limited, and is treated very differently, so we won’t be getting into that here. But so for our 2nd deep dive I really wanted to put that sick afib after sepsis and non-cardiac surgery under a microscope. Does sick Afib put you at risk for ongoing fibrillation and stroke the same way we think of pre-exisiting afib in the community?  I mean does my patient really have Afib or … is this like AFib with a big asterisk? 

Dr. Greg Katz: So is post-op AFib the same as paroxysmal AFib? I tend to think that postop AFib is really the same as prox AFib. There’s an inciting event of the surgery that makes somebody hyper adrenergic and inflamed and sick. I don’t know that that’s any different than somebody who gets pneumonia or a UTI or has alcohol withdrawal. And so it says to me that anything that raises catacoalmines that causes AFib, there is something different about your, your atria that makes you more likely to have these irregular heartbeats. 

N: That was Dr Greg Katz, another one our cardiology experts. … So I found this really surprising and made me question if even should be thinking of “sick afib” patients differently. And our discussants referred me to multiple studies that actually show that … if you were diagnosed with afib during a hospitalization for sepsis or non-cardiac surgery, your future risk of afib and stroke really is higher than that of the general population without afib. 

 J: ok thats the compared to the general population? How does stroke risk for someone who gets afib in the hospital compare to those who have pre-existing afib like regular afib/community afib?

N: So this was the big moment for me, although it’s not identical, the stroke risk for someone with sick afib is really similar to that of someone with pre-existing afib. (see sources 1, 2, and 3)

J: Okay what i’m realizing here in the short course of this episode is that Nick,  you create a new term that I liked, sick afib, but now you are telling me it doesn’t really matter if its sick afib or not, we should just think about it as afib — afib in sepsis or noncardiac surgery carries similar risk of stroke and more fibrillation. 

N: Yeah I won’t be too heartbroken if that sick AF thing doesn’t catch on. 

A: That begs the natural next question then–  does anticoagulation work as well in these patients compared to pre-existing/community atrial fibrillation as sometimes as we call it, AF? 

Dr. Jason Matos: So the most cited, and I think useful retrospective study that looks at this is from a Danish group, a large nationwide database that compared atrial fibrillation patients. Those that developed it from postoperative non cardiac surgery and those who just have AF ub the community, they looked at thromboembolism rates between both of those groups of atrial fibrillation patients and they looked at the relative risk reduction of anticoagulation use for thrombo embolism, both these groups of patients. And what they found is things are pretty similar. If you anticoagulate a postoperative non-cardiac surgery AFib patient and you anticoagulate a patient, a community dwelling non-VA AF patient, your relative risk of thromboembolism reduction is similar in both those two groups. So I think this is the best data we have right now, albeit retrospective, that suggests we should be treating the postoperative non cardiac surgery population similar to community dwelling non-VA atrial fibrillation patients 

A: So both data and guidelines to consider anticoagulation in patients who develop afib after noncardiac surgery. What about afib that I see develop most often in the ICU when someone has sepsis? Does being on a blood thinner actually help decrease stroke the same way as other atrial fibrillation populations? 

Dr. Jason Matos: Unfortunately, the data for sepsis is not as nowhere near as robust. And sometimes in patients with sepsis, anti anticoagulation is tricky. The best we can do is surmise that AFib developing after non cardiac surgery is not from direct cardiac manipulation, so must be some sort of adrenergic surge and maybe something similar is happening in the sepsis population. So the best we can do is take this really good, I think, or at least very good non-cardiac surgery data and do our best to apply to the sepsis population, albeit with some caveats and some exceptions obviously in particular cases. 

N: Since there is a lack of robust data in the sepsis population. We pressed Dr. Matos on his approach on since there is a lack of robust data in the sepsis population and on how he decides regarding anticoagulation in patients who got afib during sepsis

Dr. Jason Matos: And I always go with the same triangulation of Chad Vasco burden of atrial fibrillation. Is there a history of a prior stroke and what’s the bleeding risk? And we do our best to come up with the best, the right decision for that patient. Right. In quotation marks.

A: But if we return to your patient, she was post-abdominal surgery, right? We do have some data that favors anticoagulation.

N: Right. She had had peritonitis from a perforated gallbladder, but was clinically doing much better after cholecystectomy, washout, and antibiotics. We had clear post-operative atrial fibrillation diagnosed on a 12 lead EKG with a 2 hour long episode of fibrillation. And since we now know that her longterm risk of recurrent AFib and stroke is significant, the question was now on the risks and benefits of anticoagulation. 

A: Bleeding versus clotting, always straightforward.

N: Nothing could be simpler. So I was going through my typical risk-benefit considerations for anticoagulation, but then I started thinking … hold on, is her risk of stroke during this hospitalization like really unusually high? I mean, she is having a lot of long episodes of afib with recent sepsis and surgery … her heart must be pretty irritated. Should we be jumping on anticoagulation immediately?

J: That’s a good thought! Does this milieu of inflammation and increased atrial fibrillation burden mean we’re in a real danger zone when it comes to stroke risk, like right now?

N: So basically that was my springboard into another rabbit hole. And I found something pretty surprising down there. Basically, if you have new onset afib, whether it’s in the setting of sepsis or non-cardiac surgery, your risk of stroke during that hospitalization is higher … BUT your risk of bleeding is also higher.

A: Yeah we think about this a lot in the ICU. Most of the time we don’t anticoagulate right away for new onset AF in the context of sepsis or critical illness because there have been several retrospective studies showing that stroke risk in the acute setting isn’t changed by being on anticoagulation and it might just put you at higher bleeding risk.

N:  Yeah absolutely, we all worry about strokes but it makes sense to wait for sepsis to resolve before thinking about blood thinners.

A: What about after surgery? How did her recent surgery factor into your decision making around anticoagulation? 

N: That’s pretty much exactly what we were wondering. At this point, we hadn’t made a final decision around anticoagulation yet. I did feel more comfortable with how to think about her risk of stroke, but I was still not really sure how to think about her bleeding risk after surgery. That’s why for our third deep dive I wanted to ask when is it safe postoperatively to start anticoagulation, and how is that decision even made?

A: Yeah, if I think someone needs to be on anticoagulation post-operatively I usually just page surgery and ask when they’re comfortable with us starting it.

N: Luckily my distress page was answered by Dr. Patrick Georgoff, an acute care surgeon and host of the Behind the Knife podcast.

Dr. Patrick Georgoff: So really answer your question. The answer is no really, usually within four to 12 hours after surgery, any type of true risk of major bleeding should be taken care of based on, uh, sound surgical principles. And after that, you’re really worried about, uh, what we, you know, referred to kind of in a broader context as medical bleeding.

N: Ok so, from the technical surgical aspect of things,  we wait at least 4-12 hours before starting blood thinners … but … depending on location of the surgery … it might be longer.

Dr. Patrick Georgoff:  So there’s a certain set of patients that you really want to, um, think about them as being high risk. And that’s, uh, a trauma patients, uh, who may still have injuries afterwards. Like maybe they have a liver laceration and that can instill bleed, um, or other patients that have surgery on their brain, perhaps spinal cord or their heart.  Again, that’s because the consequences are devastating if they should bleed and really squish those organs, you squish a brain is so good. You squish the spinal cord is no good. You squish the heart. That’s no good. So then when we talk about bigger kind of categories of patient patient who gets a thoracic procedure, maybe a wedge resection on their lung or a patient who has an open hemi-collectomy, so kind of bigger surgeries, but that fall into the general surgery category. Well, those folks usually were talking about within 2, 3, 4 days you’ve seen that they’re doing well. And at that point you say, okay we let’s, let’s get that, that Eliquis back on.

N: It’s interesting to get inside the surgeon’s head, since pretty much everything (in there/surgical) is a gray area for me.

Dr. Patrick Georgoff: And I think one of the, the biggest things that we need to hammer home for the audience is that, you know, all of everything about anticoagulation, restarting anticoagulation, postoperatively is highly surgeon specific it’s patient specific it’s case specific it’s institution specific. 

N: Yeah every situation is different. But conveniently, The American College of Chest Physicians gave us somewhere to start. CHEST released its 2022 guidelines on perioperative anticoagulation management while we were working on this episode. 

J: Oh I’m glad you brought that up. Walk us through the relative part of the guidelines

N: So when it comes to starting AC postoperatively, they break it up by surgical bleeding risk. For low to moderate bleeding risk surgery, you can consider resuming DOACs as soon as post-op day 1 and for higher bleeding risk surgery, it’s best to wait until at least post-op day 2 or 3. 

A: And what about warfarin?

N: CHEST recommends that you start within 24 hours of surgery since it takes a few days before the person will even reach therapeutic levels.

A: So to summarize, there are a few factors that go into when you can start or resume anticoagulation postop including the site of the surgery, how invasive it was, which AC you’ll be using, if it’s a high or low risk procedure. And although not specifically mentioned in the guidelines, you should always be working collaboratively with the surgical team because there is nuance here. 

J:  These guidelines are helpful. What I like to do is present the surgeon with the range from the guidelines, whether they are a higher stroke or clotting  risk, and then I get their sense of bleeding risk specific to the individual surgery and how it went. 

N: These questions are so case-specific, but don’t get me wrong, it’s nice to have some guidance to go off!

J: Yeah, even a loose guideline is better than solid gray. But I’m guessing the surgery team didn’t have a problem with anticoagulation in your case since she was doing better and at least 2 days out from the OR?

N: Yeah that’s right, the decision was up to us. So, we have a 72 year old woman … with newly discovered afib … after an infection and a surgery that the CHEST guidelines consider high risk for bleeding. She is out of the ICU and doing better, and surgery has no issues with starting a blood thinner. 

A: Ok so you got the okay to start AC post surgery, so what did you do?

N: Well, we calculated her CHA2DS2-VASc to be 4, which suggests about a 4.8% yearly stroke risk. Her HAS-BLED was 2, suggesting a 4.1% yearly bleeding risk. Taking in her long episode of afib and our new knowledge that she is at risk for recurrent afib and stroke, I felt she would likely benefit from AC. 

J: So what happened? 

N: She was pretty hesitant about being on an anticoagulant …  she had a loved one who had a pretty serious bleed on a blood thinner. She also said those two hours she was afib, she was just really stressed from a phone call she got from her son … said she was just gonna avoid stressful situations like that in the future. And understandably, she was tired, overwhelmed, and she really didn’t want to take another pill.

J: I get where she is coming from but With the duration of afib and her CHA2DS2-VASc of 4, her stroke risk isn’t nothing. what do you do now?

N: I spoke with cardiology and they recommended we set her up with outpatient cardiology follow up and do some long-term monitoring to look for atrial fibrillation, which seemed to make sense. I mean, remember, some patients do not have recurrent AF after surgery or sepsis, and this could help us know if she is one of those people. I explained the plan and the need for follow up with her, but also led me to some other question so for our next deep dive I want to take a look at how we choose the best cardiac monitor for a patient and how is that cardiac monitoring data really used in atrial fibrillation?

A: ok I’m glad we’re talking about this. Can we review the types of monitors please?  I’m not sure which one I should be reaching for for newly found afib on hospital discharge.

N: Same here. I definitely had to go back for a refresher. I think it’s helpful when you think of these monitors to ask yourself … do you want it to record everything (e.g. continuous cardiac monitoring) or do you want it to only record something specific (e.g with a event recording). And then ask yourself … how long do you want to monitor the patient?

J: The way I think about it is like in generations of technology. There’s the old fashioned Holter monitor where it records every single thing for 24-48 hours. But those didn’t have enough memory to record anything longer than that. 

A: So that might not be useful in this situation because if she has paroxysmal afib it might not happen in those 24-48 hours. 

N: Right, a holter monitor may be more useful if you’re in afib most of the time and you want to know how good your heart rate control is. 

J: So the next technology is the event monitor AKA the loop recorder. It’s still limited by memory, but the way it gets around the problem is this… it’s recording everything, but then if there’s no event, it loops over like you are recording over an old cassette. But then if you hit the button because you are feeling palpitations it saves like a minute of that event and the 30 seconds beforehand and doesn’t record over it. And also the more modern ones have an internal system that can recognize arrhythmias so it saves those events too, even if you don’t hit the button. So in this way it can monitor you for 30 days without needing to have 30 days of memory. 

A: So the loop recorder AKA event monitor could tell you how many times you went into afib but it won’t really tell you the total time you’re in afib?

N: Right, So the third option is sort of like combining the benefits of the holter and event monitors. And that’s the patch monitor. It lets you record everything that happens  for 2 weeks, and also saves events triggered by the patient or that the device detects.   So if you go into afib in those 2 weeks you will know the total burden of atrial fibrillation, the episode durations, the rates… So that’s exactly what my patient went home with – a 14 day patch monitor

A: Ok thanks for going over that. So your patient didn’t want to be on a blood thinner. We diagnosed her with afib in the hospital, so how is gathering more data with the monitor gonna help us?

Dr. Jason Matos: So if you can imagine a patient that’s on a medicine service that develops 15 minutes of atrial fibrillation, you’re like, I, I don’t know, have a CHA2DS2-VASc. The guidelines said anticoagulate, and you, you put a heart monitor a two week or four week continuous heart monitor on, and then they have a 19 hour run of atrial fibrillation, whether it be symptomatic or not. I think then you really have some ammunition there with the patient, uh, to say, you know, I, I think we really should be, um, though the warning shot was in the hospital…  I think we’re at the point now where the, the, the scales are tipping to an anticoagulation.

A: So cardiac monitoring will give us more data on the burden. And seeing a longer duration and multiple other runs makes us think she really would benefit from AC and we should recommend this patient start anticoagulation.

J: But say you sent the patient home with a patch monitor and you don’t see any arrhythmia in that 2 week span. Are you now comfortable with the patient not being on anticoagulation? Have we undiagnosed afib? Is a 2 week patch enough?

Dr. Jason Matos: That being said, if you’d have someone with a two week monitor, uh, that shows none or zero, I still think the jury’s out. And sometimes we repeat two week monitors will, will repeat in another six months. 

A: Yeah, maybe they didnt have that catecholamine surge, or a really stressful phone call,  in that 2 week span to capture their afib. So for patients you are concerned about, it makes sense to have longer monitoring periods.

Dr. Jason Matos: And if you’re 42 years old, two to four weeks of atrial fibrillation monitoring, or even if you’re 65 years old, two to four weeks of atrial fibrillation monitoring is just not sensitive enough. Um, and we don’t know what that cutoff is, but in, in certain patients where the, the, we feel like the risk is high enough, we’ll implant a Loop recorder for say one year, at least, and see if any, uh, subclinical atrial fibrillation arises, which would prompt in anticoagulation.  

Dr. Pooja Jagadish: And it really has to be shared decision making based on risk factors and understanding the burden of atrial fibrillation in the long run, perhaps with a loop recorder, which can record up to two years that can tell you either yes or no. If the patient has atrial fibrillation at all, they stay on anticoagulation, but if they don’t and they’ve been off of anticoagulation for a month, a year, two years, that potentially changes practice and can help you make a decision with your patients in the outpatient setting about what to do with the anticoagulation, especially if they’re lower risk.

A: So we know from studies that some patients won’t get afib again. But there’s no clear cutoff for when cardiologists think it’s safe to say that afib is gone and that it’s safe not to be on AC. 

J: I wonder where wearable technology like iwatches will fit into all this. Because it basically can be lifetime monitoring.   

Dr. Pooja Jagadish: In the world of apple watches and smart tools that patients can use at home, there’s a lot that patients can do without you the physician. So let’s talk about the apple watch study, for example, that was done on about 420,000 individuals 0.5% received an irregular pulse notification, and that correlated to 34% of monitored participants having a diagnosis of atrial fibrillation. 

J: That’s fascinating but there’s currently no way to send your patient home with an apple watch.

A: Yeah, sadly no, but wrapping up here I think the takeaway is that when we give patients a shorter-term cardiac monitor like a patch monitor, we should be realistic about what to expect from that. A 2-4 week monitor can help support a decision to anticoagulate if AF is detected. But if none is seen, there’s still a lot of uncertainty there!.

J: So to bring it back to your patient, Nick. She did not want a blood thinner and was discharged with a patch monitor. So what happened? What did the patch show? Did she eventually go on a  blood thinner? 

N: So, unfortunately … it’s not the ending I hoped for. She missed the post-discharge cardiology appointment. I even called the patch tech and he said she never even wore her patch so we don’t even have that data. 

A: Ugh that’s so tough, It sounds like you really tried to set things up well for her on discharge but there’s so many reasons why she might have been lost to follow up.

J:  I’m curious Nick what were you thinking about at this point seeing that she didn’t get the patch you had arranged for?

 N: Well…I don’t know. I couldn’t help but think … was there something I could have done differently or communicated better… something to help her understand the importance of following through with the plan? Should I have tried harder to convince her to take a blood thinner from the start? I mean, I thought we had a great conversation about it. 

A: I wouldn’t be so hard on yourself because I think we all do our best to educate patients but sometimes what we say may not be what the patient hears

N: That’s what lead me to the final deep dive to examine the best practices for communicating something complicated like afib and anticoagulation with our patients.

J:  I mean, I think one of the first questions is how much we should really expect our patients to understand?

N:  Yeah she was sick and very tired of being sick. When we walked in the room and we mentioned her heart, I could tell she was already overwhelmed. 

Dr. Greg Katz: Like you can’t just tell somebody something once and then expect them to remember it. How much do you remember?  What about what you had for breakfast today? Probably not that much. And so if you’re stressed out and you’re not feeling well and you’re recovering from surgery and you have anesthesia in your system and you’re in a weird setting and you’re sleep deprived, like, are you gonna really remember what your doctor said about this irregular heartbeat that maybe you heard about on a commercial during a football game? Probably not. 

J: You can do all this cerebral work on rounds talking about the duration afib, risk of stroke and  pros and cons of anticoagulation but all of it goes in vain, if the patient doesn’t understand it.

A: Totally and with afib it’s really two things that are difficult to explain here, one is weighing the risks and benefits of blood thinners…and the other is atrial fibrillation itself.

Dr. Greg Katz: I try to be calm with patients when I explain it to them. And I phrase it like, you know, your adrenaline levels went up related to the surgery and there’s inflammation, and there’s all this stuff going on in your body. And as a consequence of that, but your ticklish heart went into this irregular heartbeat. And it’s a super common irregular heartbeat. There’s 5 million or so Americans who have AFib. And it’s really something that’s very well able to be managed with medical treatments. 

A: I loved the ticklish heart analogy and am stealing that for the future!

Dr. Greg Katz: I  don’t view the conversation as the be all end all of our discussion and it’s different if you’re the consultant and all you’re gonna do is see this patient in the hospital. Once maybe you need to communicate a little bit more about the gravity of this situation and about the lifelong importance of monitoring and potential blood thinners and all of that stuff.

A: I like that – focusing more on communicating the gravity of the situation that the patient can also take ownership of the long-term attention this condition needs

Dr. Greg Katz: And so somebody comes in with a score of four. I tell them you have about a 4% annual risk of stroke. And we lower that considerably by putting you on blood thinners, we don’t make it zero, but we lower it quite a bit. And, you know, talking with people about what those actual numbers are over the course of a year sometimes helps them to have a sense of like, do I wanna take this blood thinner or not? But you know, sometimes I think that when you only talk about a one year risk, you actually undersell how important it is and you also miss the asymmetric risk profile of stroke. And what I mean by asymmetric risk profile is if I put you on a blood thinner, maybe it’s a little bit annoying to take. Maybe you bruise a little bit more easily. It’s possible. You even have some minor GI bleeds. It’s possible you have a major bleed, but if I don’t put you on a blood thinner, you have a risk of stroke that is modifiable, but we’re not modifying it. And strokes are really, really bad, or at least they can be really, really bad. And so when I say asymmetric risk profile, I mean the downside of having a stroke, because we didn’t put you on a blood thinner has potential to be so much more catastrophic than the downside of having of being on a blood thinner. And it’s not to say that some patients don’t have life threatening bleeds because they do, but with direct oral anticoagulants, number one, the risk of intracranial bleeding is really, really low. And that’s the bleeding that you’d worry most about.

A: So a 4% chance that I have a stroke this year…. Is that a lot? I mean as a physician I’m having trouble deciding how to advise the patient based on that number, so is it even  fair to ask patients to understand all of the complex physiology and risk modeling that goes into these decisions?

J: Ali, I totally agree. I think to a certain extent, many patients walk away instead with, if I don’t take this medicine, I will have a stroke, but if I do take this medicine, I won’t have a stroke, which is obviously not that binary, but to a certain extent, we benefit from them having a misunderstanding because that’s the understanding they need like conceptually to do the thing that’s overall in aggregate right for them. 

If they learned that actually it’s a 4% per year risk and we’re only reducing the risk by about 2/3rds with blood thinners.

So just to do the math, 4% annual risk of stroke becomes 1.5%. 

So the Absolute risk reduction is 2.5% which is a number needed to treat of 40. 

So 40 people like you have to take this medicine for an entire year to save one from a stroke. 

And then to make it more complicated, you know the number needed to harm is like a hundred.  

A: Jason, please for the love of god stop with the numbers, you’ve made your point….my head is spinning.

J: Ok fair, but I think the point is if you actually tried to help people really truly understand what the numbers meant. It’s almost too much. It’s almost counterproductive.

N:  Yeah and there is so much even we don’t know. What does their cumulative risk look like over 5 years or 10 years? We don’t have great ways of estimating that. What if this is the patient that does have a catastrophic intracranial bleed?

Dr. Jason Matos: I mean it’s super complicated, right? Because a lot of people like to quote the one year risk. But as I explain to patients in the office, every one year you’re getting older and your stroke risk is going up. So each year we’re doing the calculus. It’s a little, unless you have a major bleeding event in that interim time, your embolic stroke risk, your risk benefit for anticoagulation is favoring the anticoagulation as time accumulates and as a patient ages. So it’s a tricky decision. But I would say that if a patient has a reasonable life expectancy that they’re going to accumulate more high risk years on anticoagulation and they’ll more stand to benefit,

N: I’ve had patients who really appreciate being looped in. They want to hear the nuance of their x% risk of stroke, they want to know this has comes from solid population studies … and others give me that look that says “you’re the doctor, why the heck are you asking me?” But it’s hard to know which kind of person you’re talking to. How do you two gauge how much a patient wants to hear? 

A: I kind of borrow from the SPIKES mnemonic and ask the patient, “how much do they want to know?”

J: Yeah I do something similar, I start with “Tell me what you know”. And from that I get a sense from how in depth they go I have guess of their health literacy, their interaction with the health system, etc. 

A: Yeah it seems like we’re starting with similar idea of sort of a “check-in”. Is this someone who wants all the details? Or Is this someone who just wants the clinician to tell me what to do. 

J: Yeah I also like pausing and checking throughout and asking “Tell me your thoughts on that.” Sometimes I even say the words PAUSE, “i’m gonna PAUSE here and see what questions you have” in the middle of an explanation rather than finishing the whole explanation. It helps me titrate the amount of information I give. 

N: In hindsight … I really can’t remember if I did a check-in with her… If I were to have another chance at talking to her, I’d try to get a better sense of where she was at … mentally and in her understanding of her disease … I mean she told me that she had a stressful conversation with her son and I think soon after I just jumped into the risks and benefits of blood thinners … maybe I could have approached it differently. … Just really stress that it’s okay if she feels overwhelmed and that she doesn’t need to make any decisions today, but that this is a condition that really does need close follow up.

J: You know your experience with this patient wasn’t unique. We know that compliance with DOACs is low, like only 50-60%, so I think maybe we all have some work to do in communicating with our patients about the importance of anticoagulants in this situation. 

A: Nick, I’m so glad you brought this case to Gray Matters. If everyone listening reflects a bit more on how to improve their Afib conversations I think that’s so meaningful. Let’s PAUSE (ha see what i did there) and recap some of the the things we are taking away from the deep dives. Nick, do you want start us off.

N: In the first deep dive we learned that when diagnosing AF you need either a 12 lead ECG or at least 30 seconds of a fib on telemetry. It’s also important to note that increasing duration of AF is correlated with increasing stroke risk with about 3 buckets of patients, those whose afib was less than 6 minutes, greater than 6 minutes and greater 18 hours or longer

J: In deep dive 2 we learned to basically think of the stroke risk in patients who develop afib after sepsis or non-cardiac surgery is very similar to patients who develop atrial fibrillation in the community.

A: In deep dive 3 we talked about timing of starting anticoagulation post-operatively which depends on the bleeding risk associated with the surgery. You can also take a look at the CHEST guidelines for help, but also always talk with the surgeons regarding their input. 

N: In deep dive 4, we talked about the utility of monitoring for recurrence of AF to help you decide if the patient should be on anticoagulation. It can be helpful if you DO see AF, BUT if you DON’T see AF it’s hard to know if we can ever be confident that this patient truly does not have AF and doesn’t need a/c. 

J: In deep dive 5 we saw there is so much nuance in discussing a new diagnosis that’s hard to wrap your head around like afib, but one of my big take aways was the importance of pausing to check for understanding. 

A: Nick, thank you so much for answering our plea for someone to come join us so it’s not just Jason and me going off tangents that potentially only we’re interested in.

N: Thanks for having me! – Happy to join the tangent train any time

J: Yes, and listeners if you have a case that you want to bring to Grey Matters please let us know. If you found this episode helpful, please share with your team and colleagues and give it a rating on Apple podcasts or whatever podcast app you use! It really does help people find us! 

A: If you have a case you’d like to bring on air, please email us at hello@coreimpodcast.com.  Thank you to Daksh Bhatia for the audio editing. Opinions expressed are our own and do not represent the opinions of any affiliated institutions.


Tags: , , , , , ,