All right. Good morning, everyone. Thanks for attending our second in-person session for the Chest Pain MI Registry. Again, I'm Kate Malish. I'm the product manager for your registry. Today's presentation is titled Type 2 MI and MONOCA Strategies for Success. So Dr. Contos, who you met yesterday, and Tabitha Isbell-Lewis, or Lewis-Isbell, sorry, they are going to discuss the proper diagnosis of MI, and more importantly, for your purposes, the proper documentation specifically of Type 2 MI. As you know, for the Chest Pain MI Registry, that's very important because it's into our inclusion criteria, whether or not the patient is going to be included in the registry. So we met Dr. Contos yesterday, but just to reiterate, he is the Steering Committee Chair of the Chest Pain MI Registry. He's also a professor at the Virginia Commonwealth University Medical Center in Richmond, Virginia. And Tabitha Lewis-Isbell is a nurse by trade, and she currently works in the Outcome Services Division as the Registry Product Manager at John Peter Smith Hospital in Fort Worth, Texas. So this is really a great presentation. So I hope everyone enjoys it. I'll be down there monitoring the Q&A. If anybody has any questions about how to submit a question, feel free to come up and ask. Okay, I'm going to turn it over to them. Well, thank you all for attending this session and for inviting me here. So for the first part of the lecture, I'll spend outlining the problem that we have with Minoka and Type 2 MI, and then Tabitha will go through the proper ways to try to figure out how to translate what we see on the clinical side into a documentation point of view. My talk is going to be focused on the following things. Why do we use cardiac troponin? What's the definition of MI? Because that's really critical to what we're doing here today, looking at myocardial infarction with non-obstructive coronary disease, and then Type 2 MI with the majority of the talk focusing on some of the background information, such as the frequency we see it, what are the risk factors for it, what are outcomes and potential treatments. So why do we use troponin? Well, I think everyone realizes now that it's probably the best marker for identifying patients with myocardial necrosis. It's nearly 100% sensitive for acute MI with serial draws. It's very cardio-specific, so it doesn't go up with non-cardiac disease states. So although we commonly talk about it being elevated in people with renal failure, this is typically because those patients have an underlying cardiomyopathy resulting in low-grade troponin release. It does provide a good marker for identifying patients with myocardial infarction because it can stay elevated for days to weeks after the acute insult, depending on the size of infarction. This can be very useful for patients presenting late but can often be a point of confusion when you have patients who have recurrent chest discomfort a few days after they come in with an MI. Does that troponin elevation actually represent a new infarction or residual from the prior one? And I think this is one of the few places where CKB still can be useful to help sort those two conditions out. Importantly, it helps to identify patients who are higher risk. And even better, it can actually trigger appropriate therapy. We now have specific therapies that are more beneficial in patients who are troponin positive than troponin negative. For diagnosis of MI, this has been put together by a group of experts over the last 15 to 20 years with a number of iterations, with the most recent one called the fourth universal definition. It typically requires three elements for diagnosis of MI. First, you have to have a rise and fall of cardiac biomarkers. Typically, this should be troponin if possible, and I can't imagine very many places that don't have this currently available. At least one of these values has to be elevated, so it has to be above the 99th percentile of your upper reference level for that particular assay. And finally, there has to be evidence of ischemia. And this is probably the hardest part, I think, where we try to sort through it. Some of these elements are fairly objective. Other ones aren't. Probably the least objective is the presence of symptoms. Oftentimes, as everyone knows, patients with ACS can present with a variety of symptoms, and the elephant on the chest actually occurs in only a minority of patients. Instead, we can have patients with pressure, tightness, squeezing, burping, or even just shortness of breath. And this is, I think, where the biggest diagnostic dilemma comes in trying to sort out whether those symptoms really represent myocardial ischemia or some other non-ischemic state. Other things that are important that can be useful for meeting the criteria for MI are the presence of ischemic ST changes or new Q waves, imaging or imaging evidence of new loss of viability of myocardium or a new regional wall motion abnormality on echo, or identification of intercoronary thrombus by angiography or, in rare cases, when patients undergo autopsy. Obviously, it's the first two on the left, symptoms and ECG changes that are typically the ones that we're using to define criteria for evidence of myocardial ischemia. MI is broken down into five different classifications, which we'll spend the majority of time just talking about type 1 and type 2. Type 1 is the ones, the patients that are included in our registry. They have spontaneous MI related to a primary coronary event, which can either be related to plaque erosion or plaque rupture, fissuring or dissection, and results in acute thrombus formation. And this is a typical ACS-related presentation. Type 2 MIs are those that are related to ischemia from either increased O2 demand or decreased oxygen supply. And this can occur in a variety of different conditions, such as coronary vasospasm, embolization, anemia, arrhythmias, or hyper or hypotension. The other definitions are ones we typically don't use that much. Type 3 would be sudden cardiac death, which occurs before we can actually obtain blood samples. And this is often used primarily for epidemiological studies. And then type 4 and type 5 MI are related to post-procedural myocardial infarction, either from PCI or from bypass surgery. So I'll spend a little bit of time talking about MINOCA. So originally MINOCA, or myocardial infarction, was called myocardial infarction with normal coronary arteries. It's now more appropriately called myocardial infarction with non-obstructive coronary disease, because many of these patients do have some mild obstructive lesions. To meet this criteria, you have to have acute MI that meets the criteria for universal definition, as described before, no significant coronary lesion more than 50% in a major epicardial vessel, and no other clinically specific causes for the acute presentation. The European Society of Cardiology has these recommendations for evaluation of these patients. Come up with a specific diagnostic algorithm to differentiate true myocardial infarction from other alternative diagnosis of causes for troponin elevation. Future performing cardiac MRI in all of these patients who do not have an obvious cause, and in general we really should consider this a working diagnosis rather than a final diagnosis. I think oftentimes we'll have patients come in, we'll do a quick cath, no obstructive coronary disease, at least we don't think so, and we kind of write it off as a myocardial infarction without obstructive coronary disease. In fact, many of these patients can have other causes for their acute MI. How frequent is it? Well, in this one particular study, they looked at a little bit over 5,000 patients who presented over about a four-year time period who came in with stemming. About 10 to 15% of these were culprit-free, in which they didn't find a lesion over 50%. Many of these patients were troponin positive, so essentially this was a group of patients in which we might consider them to be false positive cath lab activations related to just abnormal ECG findings, but a significant minority actually had evidence of MI. A wide variety of different diagnostic tests were done to try to sort out what was going on, and you can see on the bottom what the differential was, at least in this single-setter study, in which most of them, although still a minority, about 31% had underlying cardiomyopathy, with about one in six or one in seven having myopericarditis, and about one in ten with myocardial infarction, with other diagnoses, as you can see there, with either valve disease, pulmonary embolism, or aortic dissection. It's important to consider that this is not a benign disease. If we look at the summary of about six different studies comparing outcomes in patients with MINOCA versus those who had typical MI, you can see that although the mortality is significantly lower across all these groups, it is not negligible, and in fact, when compared to normal age group cohorts, it is significantly higher in this group. So again, not a benign process, although acutely these patients do quite well. We do need to consider what we're going to do long-term for further evaluation and treatment. So what should we do with these patients? Well, this is a proposed diagnostic algorithm. Again, when patients come in, the initial coronary angiography doesn't show an obvious culprit lesion. The first thing that we've done at our institution is very analogous to this, go back and review the angiogram. Is there some obstructive small branch vessel that may have been occluded that was initially missed, or potentially there's evidence of spontaneous coronary artery dissection, which can be difficult to see, at least on the first go-round. In those that don't have any obvious lesion, you can consider doing a couple different angiographic tests, either using cardiac ultrasound or IVUS to look down the lesion, or some institutions are using optical coherent tomography, like a mini-CT where you can go down and actually look at plaque composition. This has the advantage of being about 10 times more sensitive than intravascular ultrasound, although it does require contrast for reviewing, for imaging the arteries. In those patients who do have evidence of destructive coronary disease, then we can sort it out on the right into some sort of final diagnosis related to either spontaneous coronary dissection, coronary vasospasm or embolism, or atherosclerotic disease that may have had a plaque rupture, acute thrombus formation that resolved before the sub-patient subsequently underwent the coronary angiogram. On the left side is what we typically do with the majority of our patients now in which there's nothing on the coronary angiogram to indicate a culprit lesion. Most patients will try to get a cardiac MRI. This helps really to define whether or not is this truly an infarction or is this something else such as stress cardiomyopathy or myocarditis. It's important to differentiate because your long-term treatment is going to be markedly different in that patient with the stress cardiomyopathy versus underlying ACS, where in the latter case they'll substantially benefit from anti-platelet treatment, aggressive anti-lipid treatment, as well as other secondary preventions, measures to prevent subsequent myocardial infarction. Here's a study that was performed recently looking at a cohort of women who came in with what was presumed to be non-obstructive coronary disease. They did a number of different things, which described as previously, particularly using OCT and cardiac MRI in the majority of patients. By using the combination of these two techniques, they could come up with a diagnosis in about 85% of patients either related to unsuspected coronary ACS, stress cardiomyopathy, or myocarditis performed in the majority of them with only about 15% not having a true diagnosis at the end of it. What about type 2 MI? Again, type 2 MI is related to an imbalance between oxygen supply and demand. Typically, we have underlying coronary disease, which causes obstruction, which is typically not clinically apparent until we put some demand on the patient. This can either be as a result of dropping their hemoglobin, the patient has a GI bleed, or for example, someone who has a tachyarrhythmia such as atrial fibrillation. This results in the supply-demand mismatch resulting in myocardial infarction or myocardial damage. It's important to note that you actually don't need to have coronary disease to have type 2 MI, and to me, this is one of the more confusing things about this definition, because I would always assume you have to have some sort of obstruction to result in this imbalance. But in patients who may come in with, say, severe sepsis, this is not always true. There's a wide variation in the frequency of type 2 MI. This is from a recent review in the Journal of American College of Cardiology looking at different studies that assessed it, and you can see it went anywhere from only about 2% to 3% all the way up to 70% in that particular study. Why the reasons for the variation? Well, there are important differences based on the patient population studied. Are we looking at only those coming through the emergency department, those who are just admitted, or are we focusing particularly on an ICU group? It's going to be very substantially based on the sensitivity and diagnostic thresholds of your troponin assay that you're using, and also whether or not you consider using acute non-ischemic myocardial injury is also included, as this is another important cause of troponin elevations. How well or how thorough you investigate the patient is also important for determining whether or not you classify this correctly as a type 1 or type 2. And obviously we do have limitations in our diagnostic criteria which result in variation in interpreting how we interpret these diagnostic criteria by adjudicators. Even if you take a group of cardiologists and train them on specific definitions, you get substantial variation in whether a patient would be considered a type 2 MI or myocardial injury. We're pretty good at sorting out that it's a type 1 MI with a very high rate of agreement, but when we get down to myocardial injury, it was almost like a flip of a coin to determine whether it was or not. So again, I think all of you have probably run into this situation in your hospital where if you ask two different cardiologists what's going on, you'll get three different diagnoses. And this is not uncommon in our institution. Even I run into difficulties because of the wide variety of non-ACS causes of troponin elevation. There's just a few of them that are shown here. It's important again to realize that troponin is heart-specific, but it's not disease-specific. It doesn't necessarily tell you what's going on that's resulting in the troponin elevation. Obviously a point of diagnostic difficulty in many cases, and I know many of our cardiologists of our institution have thought that maybe since we're not going to be evaluating low-risk chest pain patients in the ER, instead we're going to have a practice entirely devoted to trying to sort out troponin elevations. How frequently do we see this? Well, it's going to vary in large part on how sensitive you want to be or how frequently you want to send troponins. This is an interesting study that was performed looking at troponin sampling in 8,500 patients in different emergency departments. And what they did was they looked at two different groups, a more specific group of hospitals, those performed in the UK, England, and Scotland, and compared it to a U.S. cohort in which we have a much lower bar for trying to attempt to rule out myocardial infarction. And they found that the selection of patients for troponin sampling was significantly affected and it affected the diagnosis. So, for example, if we look at the middle bar overall, if we look at all the troponin elevations, about 15% of those ended up being type 1 MIs with 3.9% type 2 and only 5.9% myocardial injury. In contrast, because the frequency of troponin sampling was substantially higher in U.S. hospitals, we see almost reverse of those numbers with type 1 MI actually being in the minority of all the three groups with myocardial injury being substantially larger. And I think getting more information on what's going to happen in the U.S. is going to be important for us to try to figure out how to manage these individual patients. Switching to something like high sensitivity troponins that many hospitals have done now can also affect the frequency that you're going to see type 2 MI. This is another look at it in a study where on the left they looked at what happened when they switched to a high sensitivity troponin. And you can see the number of patients diagnosed with either type 1 MI, type 2, or myocardial injury increased in each of these cohorts. Overall, despite the fact that we used a more sensitive assay, we still had more type 1 and type 1 MIs. On the right side, though, what ended up happening, the proportion of patients who had type 1 MI, although still the majority dropped, going from 71% using a conventional assay, decreasing to only 55% of all the troponin elevations when we used the more sensitive assay. So we'll see more troponin elevations diagnosed, more type 1 MIs, but the overall proportion of type 1 compared to the others will decrease. Why do we have type 2 MIs? Well, again, it's frequently related to a supply-demand mismatch. This is from a meta-analysis of approximately from 40 studies, 120,000 patients or so, and looked at the underlying causes. And many of these are what you would be familiar with that would come in that typically are not necessarily going to be admitted to a cardiology service but will be admitted for their underlying medical conditions. Things like sepsis, heart failure, arrhythmia, and anemia were predominant, occurring in about a third of these patients with substantial overlap with other conditions such as respiratory failure and COPD. What are the risk factors for type 1 and type 2 MI? This is a comparison, again, using that same data set, looking at it, which about almost 78% had type 1 MI and 11% had type 2. Overall, you can see on the left, regular standard cardiac risk factors were pretty similar between the two cohorts with a slightly higher rate of tobacco use in the type 1 MIs than which one would expect. I think these patients tend to be a little bit younger. But as noted before, underlying comorbidities were substantially higher in the type 2 MI group. Looking on the right, both for CKD, atrial fibrillation, COPD, and heart failure were all higher in the type 2 MI cohort. This was a combination of predominantly European and Australian studies. In a recent study using a U.S. cohort using the National Patient Sampling Database, we actually saw similar results. Again, cardiac risk factors were typically similar or maybe slightly more in the type 1 MI group with those with underlying comorbidities on the right, CKD, AFib, COPD, and heart failure, again, being predominantly significantly more frequent in those with type 2 MIs. How do you diagnose it? Well, the first, obviously, is based on symptoms. And again, this is where it becomes somewhat difficult to do because of the overlap and atypical nature in many of our patients. But in general, patients who are presenting with type 2 MI typically don't have chest pain as their frequent symptoms. It's only about 58% compared to about 88% for type 1 MIs. In contrast, presentation with shortness of breath is significantly more in those with type 2 MIs. And those other nonspecific symptoms were present in about two-thirds of patients with type 2 MI. Looking at our peak troponin can also be useful. The higher the troponin, the more likely it's related to a type 1 MI. However, there's often substantial overlap in this summary of three different studies with type 1 MI in the red, type 2 MI in the blue, and myocardial injury in the purple. You can see that in all three cohorts, type 1 MIs all had larger or higher peak troponin values. However, in contrast with the other two groups having lower troponins, however, you can see by looking at the bar graphs, there was substantial overlap between the two. So if a troponin value is very high, it's much more likely to be type 1 MI. If it's in the lower range, it's really not that useful to help differentiating between the two. These patients are not a low-risk cohort. Their mortality is often frequently higher than those with type 1 MIs. So again, I think it gets us back to although we don't typically think kind of don't worry about them quite as much, it's important to recognize that they are not a low-risk cohort. And in fact, the patients who we would often consider in many places are called troponinosis or troponin leak also have a substantially higher mortality comparable to those with type 2 MI. So I think often differentiating the two between type 2 MI and marker injury may, at least from a treatment point of view and follow-up point of view, may not be quite as important because of the higher mortality overall. And again, significantly higher than those with type 1 MI. Why is the mortality higher? Well, as I noted earlier, they have a substantial higher rate of underlying comorbidities. And these comorbidities appear to drive a lot of the mortality, which in many cases may result in more than 50% of the cause of death in these patients with cardiovascular disease being 50% or lower frequently in different cohorts that we study. So what can we do to try to reverse this or mitigate this overlying risk? Well, we don't have any good evidence-based data. In fact, there's very limited clinical data on what medications may be beneficial in these patients. This is one study that did look at the rates of underlying coronary disease between type 1 and type 2 MI. This is from the very large Sweetheart Registry in Sweden. They have the advantage that every person who comes in who undergoes coronary angiography is subsequently enrolled in their database and their registry and they can keep track of this. So when they went back and analyzed this cohort of almost 42,000 patients, about 90% of these patients had type 1 MI, of which the vast majority of these had underlying coronary disease. In contrast, in the type 2 MI, only a small minority underwent coronary angiography. In that cohort, about half of them had underlying coronary disease. But the important finding in this group is that if we look at the four different curves, on the bottom are type 1 and type 2 MI, those who did not have obstructive coronary disease. The bar in the middle, the orange bar, is type 1 MI who did have obstructive disease, and the very top bar, the one with the highest mortality, was type 2 MI with underlying coronary disease. So it may be at least assessing underlying coronary disease may be helpful because it identifies a group of patients, type 2 MI, that have a substantially higher risk than those who don't have obstructive coronary disease. When we look at a number of different studies that have performed coronary angiography, about a third of these patients do have underlying coronary disease, and about a quarter of those who have coronary disease have multivessel disease. So again, once we find coronary disease, many of these are at high risk cohort. Rather than doing coronary angiography, noninvasive testing may be useful. However, again, there's very limited data on this. This was really the only study that I could find that looked at it closely, and it was a fairly small cohort of about 234 patients who underwent stress NPI. What they found is those that stress perfusion imaging was abnormal in more than half the patients at 58%, with a significant number of them having abnormal perfusion, abnormal systolic function, or the combination of both. Importantly, if you look long-term, mortality was substantially higher in those that had abnormal stress NPI versus those that had normal stress NPI. Importantly, I think when I looked at it, the curves don't separate for the first few months which I think at least provides indirect evidence that an evaluation can be useful but it doesn't necessarily have to be done on an acute inpatient setting debate to identify higher risk patients. Even though these patients have a high rate of underlying coronary disease evidence based guideline therapy is infrequently used and I think this is a big gap that we're missing because these patients again are very high risk for long term events and this is something simple we can do in many of our patients. If we look at this meta analysis comparing overall medications we can see substantially lower rates of beta blockers, ACE or ARBs, anti-platelet therapy and even something as simple as statins for reducing overall risk was significantly lower in this cohort. So how do we put it together. This is at least one start one algorithm from the fourth definition of MI. I think the first step that what we should do is the first step is look at our troponin values. Is it greater than the 99th percentile. If so that meets criteria for for myocardial injury either acute or chronic depending on further steps in our algorithm. Step two once we have an elevated troponin we want to know is there a rise and fall and if so it falls into our acute myocardial injury group. If not this falls into a chronic myocardial injury group on the right which would often kind of patients with underlying cardiovascular disease either heart failure or an air disease or valvular heart disease. And these patients are a higher long term risk but clearly in most cases don't require acute cardiovascular care other than treatment of their underlying condition and those that do have a rise and fall in the troponin what we need to do now is assess for evidence of myocardial ischemia. If they don't have evidence of myocardial ischemia they would fall in the acute myocardial injury group. These would be patients that present with conditions such as acute myocarditis stress cardiomyopathy or even patients with pulmonary embolism. These are not actually considered myocardial infarctions anymore but acute myocardial injury. If they do have evidence of ischemia then we're now down to trying to determine whether it's type 1 versus type 2 mi mi groups. One of the difficulties I think again between separating type 2 mi and acute heart the acute myocardial injury is the substantial overlap between patients who may have heart failure in which how do we know that this acute myocardial ischemia which would put him into a type 2 mi or a myocardial injury at least for me the way I kind of look at it is I like to have regional evidence of underlying myocardial ischemia to me that becomes a much more obvious type 2 mi or in the absence of regional ischemia where it's a global event. I usually at least for me consider that to be acute myocardial injury although you'll get substantial disagreement from different people where these patients should fall into. So when we do have type 2 mi here's a proposed framework for classifying it first put it into context. What is the main reason for the patient presenting with the coronary episode and then second look at the mechanism on the top or patients that have pure supply reduction often not due to known obstructive coronary disease. So for example coronary vasospasm cornea embolism or corny coronary endothelial dysfunction and finally spontaneous coronary artery dissection. I think many of this when these patients present often will come in with acute ST elevation and we'll consider these as just as an obvious type 1 mi. It was somewhat of a surprise to me when I went back and looked at the definition that these are actually truly considered type 2 mi. We're looking at this as far as our registry for exactly how to classify these patients but currently they would be considered type 2 mi. On the other end if they do have obstructive coronary disease we're looking for patients that either have increased demand reduced supply or the combination of that if they have underlying concomitant coronary disease we should consider following the coronary second second secondary prevention management guidelines for further risk reduction. This provides a kind of another SIMP algorithm for how we should go through evaluating them. First obviously we want to treat the underlying precipitant of the supply demand mismatch. If they don't have no one coronary disease we can see we should consider doing further evaluation either anatomic or functional evaluation and then depending on the results treat them either primary prevention or secondary prevention. My diagnostic strategy which I which I tend to do in most cases is first do that which follow I think those guidelines very closely clinical assessment for ischemia and then echo for the vast majority of them just because knowing LV function helps to drive subsequent therapy strongly consider an ischemic evaluation often can be done as an outpatient with a coronary angiography and a minority of these patients and then implementing secondary risk factor modification with guidelines more to aim towards secondary management goals rather than primary goals to finish type 2 M.I. myocardial injury are common they're increasing as we go to high sensitivity troponins how we manage these patients is difficult because they're very limited evidence based therapy reasonable recommendations I think are again assess LV function look for the presence and absence of coronary disease and initiate appropriate secondary risk factor modification and treatment. Now for that background book now gets into how you're supposed to classify these for your ultimate diagnosis. Thank you Dr. Contos that was excellent. Good morning everybody. My name is Tabitha and I'm so excited to be here with everybody I cannot believe this is the last day of our conference. Thank you to the ACC for putting this on. I feel a little bit starstruck being up here seeing Kate John Denise Kim all these people that we see on our calls all the time just walking around and getting to talk to them it's like oh I get to see a movie star. Anyway I am here to discuss why to document the non STEMI type. So here are the key takeaways the importance of the non STEMI documentation the chest pain of my inclusion criteria. Well our facility does to aid in non STEMI type documentation clean data in equals clean data out data drives change and collaboration is key. So I work at a facility called John Peter Smith Hospital located in Fort Worth Texas. We're also known as the Tarrant County Hospital District. We're a level one trauma facility serving the indigent population. We have 573 beds that includes over 25 primary care specialty health centers. We have a board of managers that consists of 11 representatives that are appointed by the Tarrant County Commissioner's Court and the Tarrant County Commissioner's Court actually approves our hospital district property tax rate and our annual budget. So we are a tax funded health support system. We received many awards designations and certifications because we are held accountable to our external stakeholders which are the taxpayers. It's important to have clean data. It's also helpful to have recognition through facilities like the ACC because it shows the taxpayers that we are giving good quality care with their money to our patients. So why is non-STEMI type documentation important. It determines which patients get submitted to the registry. So a non-STEMI type 1 patient gets included but the type 2, 3, 4, 5 does not get included but the type 2 has to be clearly documented for exclusion. It also affects the metrics and the benchmarking reports. If a bunch of type 2 patients are getting into your data you may have fallouts because the treatment is different and this can decrease your performance. It can affect your awards and your certifications because if your data is lower than the benchmark you won't be eligible for the awards. It can also affect billing and denial because if the documentation and the testing does not support the documented diagnosis then there's a risk for payment denial. It can also affect the hospital base value purchasing program with CMS 30 day readmission and mortality rates which can affect the reimbursement and the public reporting. So the coding will assign that ICD-10 code and in the coding world a non-STEMI is a non-STEMI type 1. So then it can fall into the 30 day readmission and mortality and you can have a potential for up to a 2% decrease in payment. Clean data in, clean data out. And I know at home my team members are laughing right now as they hear this because that is my why in my job. I want to make sure that my teams have clean data and the reason is is because data drives change. And if your teams don't have that clean data they're not going to trust the data and the data. And as our CEO always says the most important is our true north which is our patients because type 1 and type 2 are treated differently. So for registry inclusion when we're finding the patients to include in the registry these are the four qualifying diagnoses based on the documentation. You've got the STEMI, non-STEMI, unstable and low risk and your STEMI and your non-STEMI are the type ones. So after several years of cleaning up our data and making some changes we finally started receiving some ACC performance awards and this was very exciting to our teams. And so I'm here to share our journey on what we did to help clean up our non-STEMI type documentation which actually played a huge role in our data. So these two graphs show us benchmark in the registry. The top graph is the overall AMI composite and the bottom graph is the overall defect free care which shows if we gave that perfect care for each AMI patient. And as you can see on the left hand side back in 2016 when we started we were below the benchmark especially on that overall defect. We were down at the 10 percent range which is not very good. But in the late 2017 early 2018 we made some huge strides with that new ICD 10 code of non-STEMI type 2. But then look over on the far right side for the quarter 2 22 of this year we fell again. And this was my face when I saw that as I'm sure that we all do when we see any change in our data. So when I was performing our deep dives it was because we were getting too many non-STEMI diagnoses again in our data that should have been probably ruled out or deemed a type 2. So we did have some new physicians that had been on boarded so we just needed to work on some re-education again. And I know that we all get scared with change but that is the that is how health care works and change can be good especially if it benefits our patients. So when the ICD 10 non-STEMI type 2 code when that got started in October 1st 2017 that made a huge impact on our data because before this was introduced all non-STEMI were included in the registry and the CMS data. After this change the type twos were excluded. So we noticed that education to our physicians were key for exclusion. We also started doing concurrent data versus retrospective data. We also built an AMI daily list in our EMR system which we use EPIC. We look at this daily to find any potential patients that could be falling into our list and we follow them the whole time that they are in house. We leave sticky notes in the chart so EPIC actually allows us to leave a sticky note in the chart to communicate to our staff members but then it's not part of the medical record. So we'll leave a sticky note in there saying that this is an ACC chest pain AMI patient to remind them of the metrics. Then if we start seeing some muddy documentation or conflicting documentation for type 1 or type 2 we'll involve the clinical documentation integrity specialist who will actually help aid physician with their documentation. They will work with the physician because they have some coding background to know what to help clear up that documentation for us and they'll do that while the patient is still in house. We also created a discharge list in EPIC based on the type of discharge list in EPIC based on the ICD 10 codes not just the primary or the secondary code but all of them just so we can kind of watch and see and follow the documentation as well as the coding practices in the coding department. Then we will query the physician if we need any additional documentation using the EPIC secure chat we can actually attach a patient link to that query. So all the physician has to do is actually click on the patient link which helps save some time versus just an email. Then if a patient gets discharged and they are coded with a non-STEMI diagnosis but we're seeing some conflicting documentation in the chart then we will send that back to the coding department so that way they can query the discharging physician for any potential cleanup. And as the ACC says all the time submit early and often we submit our data every Friday and we look at it every Monday or Tuesday so that way we can catch things as quickly as possible. This is a copy of the sticky note that we actually leave our patients in the chart while they're in-house and we can actually modify this you know depending on what's going on with the patient. So if we're actually seeing some muddy documentation or some conflicting on the one or the two we will leave that note right there in the middle that's bolded saying that if it's ruled out or a type 2 to please clearly document for an exclusion. But it reminds us of the metrics in the cardiac rehab referral to get that ordered. Now I'm going to review a few of the case studies that we actually reviewed with our physicians from our quarter 2 2022 data which actually caused our data to fall and these are actually real cases. So a patient presented with left sided chest pain pressure that resolved after aspirin and nitroglycerin past medical history of hypertension anemia and smoker the EKG showed LVH and T wave inversions troponins were elevated. The plane was aspirin heparin beta blocker trend troponins and a cardiology consult to the cath lab today. The cath lab indication is non-STEMI the cath lab results and geographically normal with 0 percent stenosis in TEMI 3 flow but nobody ruled it out non-STEMI diagnosis made it all the way to the discharge summary and it got coded non-STEMI. So the patient presented with left sided chest pain pressure non-STEMI. So the fallouts on this particular patient were LVF was not evaluated. Cardiac rehab phase 2 was not ordered and a P2Y12 inhibitor was not prescribed at discharge. But when you actually think of a type 1 or type 2 if you look at that type 1 the patient had 0 percent stenosis so really it didn't meet criteria for the diagnosis. So the opportunity here is that the clinical evidence does not support the type 1 so they should specify the type or rule it out by the time that discharge happens. Case study number 2 patient presents with left sided chest pressure for one day denies shortness of breath diaphoresis and nausea significant past medical history of heart failure cardiomyopathy hypertension hyperlipidemia LAD stent cabbage heroin use nine days ago and smoker. The ECG was consistent with previous ones. Cardiac cath six months ago showed a dilated cardiomyopathy with an LVF of less than 20 percent patent RCA CERC and Lima with Timmy 3 flow. The troponins were elevated. The plan was aspirin heparin cardiology consult to the cath lab the following day. The cath lab indication is non-STEMI. The cath lab results showed the patent RCA CERC Lima to the LAD with Timmy 3 flow and on comparison to the previous angiogram there were no significant changes and unlikely to be cardiac and etiology. Nobody ever ruled this out. Everybody kept calling it a non-STEMI all the way to the end and on the discharge summary and it got coded non-STEMI. So the registry fallouts were no LVF was evaluated cardiac rehab phase 2 was not ordered. And this patient was a Medicare patient came back to our facility in less than 30 days as an inpatient. So it was also a CMS 30 day readmission fallout. The opportunities that the clinical evidence doesn't support AMI so specify the type or rule it out by discharge. And then the last case study I have is the patient presented with chest pressure like someone is sitting on me since this morning with nausea diaphoresis headache syncopal episode times 2 past medical history of an MI with a stent hypertension hyperlipidemia COPD and a smoker past medical history is an MI with a stent. I just say that I think I did. I'm sorry ECG is normal. Antroponins are negative. So the plan is that the coronary artery disease the initial work up was unremarkable. However the symptoms were concerning for progressive C.A.D. The stress test was ordered and it was abnormal with evidence of inducible ischemia. The cath lab indication is unstable angina abnormal stress test C.A.D. The cath lab results showed a 70 percent eccentric distal right posterior AV groove segment stenosis that was considered significant by IV ultrasound and treated with the drug eluding stent. But for whatever reason the hospitalist started calling it a non-STEMI after the cath went all the way to the discharge and it was actually coded a non-STEMI but always look at your inclusion and your exclusion criteria. Our facility only submits the AMI population. We do not submit the unstable angina or the low risk chest pain. So for us there were no registry fallouts because to meet the non-STEMI their troponins were negative they didn't they didn't have any positive troponins. But the opportunity here is that the clinical evidence does not support AMI and there's a high risk for medical necessity denial with an estimated loss of thirty six thousand six hundred dollars. That's a lot of money. All three of these cases provide different reasons supporting the need for the non-STEMI type documentation. So data drives change. We have a monthly AMI committee run by our fearless chest pain coordinator Chrissy and it's a very collaborative group. We involve EMS, ED, cardiology, the cath lab, nursing administration, the hospitalist group, we have teaching teams at the hospital. So we just got them on board as well. Our MET team which some people their hospitals will call them the RRT team, the education department, quality and nursing resource from every single floor. Then we also have ad hoc members as well cardiac rehab, the CDI team, coding, IT, our EPIC, whoever we need to bring on from time to time. These teams we review all STEMI and inpatient STEMI cases, we review the data trends from the registry and our scorecards, we discuss processes and any potential needs for change and we bring the non-STEMI type 1 versus type 2 documentation here. All registry data and scorecards are placed on our quality division scorecard page for any staff to review so that way they can just click it. This shows transparency and it gives them a why to what we do and each quality service member we have a link under our signature on our email so they can just click the link and it takes them right to the page to look at any registry or scorecard data. And then education, education, education. So our hospitalist group discharges 60 to 75 percent of our AMI discharge and the rest of them get split up between our cardiothoracic team and our medicine teaching teams. So we do bimonthly registry data specific to their group because the registry allows us to put the discharge physician in there we can break it out per team. We give them a list of the OFIs or the fallouts with ways to improve and we do non-STEMI type 1 versus type 2 education here. We also do resident education and we enlist our quality educator to assist with the education. Each registry has a physician champion so that way if there's something that we need help with or a peer to peer type thing we bring it to him and he will go take care of that for us. And then we have three abstractors that abstract the chest pain in my registry and we meet monthly. We use the ACC IRR tool and so when we use that we meet monthly we talk about our IRR percentages and I know all of us people that abstract we are very competitive with our IRR percentages and we discuss the frequently asked questions to make sure that we are on the same page. And if there's ever a question we use the contact us function that the ACC provides and we let them be our mediator. And then do not forget to celebrate wins when you have physicians that are working with you on their type 1 type 2 documentation send them a note say thank you for working on this quality initiative. A thank you goes a far a long way. When we received our ACC awards Chrissy and I we printed off the award framed it and took one to every single department that takes care of our AMI population. We even drove out to our EMS and gave them as one as well to say thank you for taking such great care in helping us achieve this award. So I believe in sharing of knowledge which is why I'm here. You never know if one of us or one of our family members will end up in one of our facilities so feel free to reach out with me reach out to me for suggestions or questions. Thank you. All right. Thank you so much for that presentation. Unfortunately we're going to have to excuse Dr. Contos. He has a flight to catch. So rest assured we have all of these wonderful questions and we're going to do our best to to get them answered for you. I will email Dr. Contos with a list of questions and have him get back to me. There are a couple of questions for Tabitha if I can find it. It involves you know how to get buy in from your team of physicians the team that you work with to actually do something as great as your facility did. Can you talk about that. Actually I can. That's a common question that I've seen throughout this conference. So as I started seeing that I actually emailed all my registry champions last night my physicians and said ask them because I knew that you guys would ask me that. So from their words they said keep us looped in the data. We want to know what's going on and then give us case by case review with fallouts with education so that way we can learn from it. One physician said that they really like the didactic meetings rather than the physicians having to go look up the data because they don't have time to go look up the data. They want you to come to them and talk to them verbally and tell them. And then one physician actually suggested tying it to quality metric with incentive pay. I talked about our CDI taking it to our CDI specialist because they can query the physicians. Well that hospitalist group actually has a metric tied to how many times they they answer their queries. So when they do that that's why I send it to them because they're more likely to answer them and they have four days to respond. But what the CDI team tells me is you really need to work with the physicians because if a physician is on PTO they shouldn't be penalized for that. So work with your physicians on that particular metric. And then my husband is actually a nurse practitioner here in the audience and I was asking him about this as well last night. And he said when you bring me data I want clean data. I want the metrics explained and I want metrics that I can actually change and so empower me to be able to change that. Excellent. Thank you so much. Unfortunately that we're a little over time so we're going to have to end the Q&A session but thank you so much for all of your questions.

Type 2 MI

documentation

Chest Pain MI Registry

cardiac troponin

non-STEMI type documentation

diagnosis and treatment

clean data

physician education