Chest Pain (to D/c or not to D/c)

This forum made possible through the generous support of SDN members, donors, and sponsors. Thank you.
H

HoosierdaddyO

Full Member
7+ Year Member
Joined
Sep 9, 2015
Messages
656
Reaction score
617
After having a lengthy conversation with a few other attendings, it was really funny and interesting to see how 4 different doctors approached chest pain disposition all differently. Some use a hard cutoff heart score, others a gestalt of 1 or 2 trops and dispo home with followup... some just admit everyone and I mean everyone. I think for academics purposes, looking to see if I can start a discussion on chest pain work ups and what you are doing for these peeps :)?!
Members don't see this ad.
 
HoosierdaddyO said:
After having a lengthy conversation with a few other attendings, it was really funny and interesting to see how 4 different doctors approached chest pain disposition all differently. Some use a hard cutoff heart score, others a gestalt of 1 or 2 trops and dispo home with followup... some just admit everyone and I mean everyone. I think for academics purposes, looking to see if I can start a discussion on chest pain work ups and what you are doing for these peeps :)?!
Click to expand...

A HEART score of 0-3 still gives you a MACE rate of 0.9-1.7%. And 4-6 points gives you 12-16.6%. I feel we can get that level of discrimination on gestalt alone. The folks who would get 4-6 points on the HEART score were always easy, there wasn't too much disagreement on admitting them. And on the lower end, a HEART score is not the savior because from the point of view of an experienced, conservatively practicing emergency physician, the problem was never the lack of understanding that the risk to these patients is in the ballpark of 1-2%. I don't think most conservative attending were overestimating the risk. The problem was what to do about that risk. The traditional retort was that a typical emergency physician might see about 300 chest pain patients a year, and if somehow they all had a HEART score of 0 were discharged there would still be 2-3 bad outcomes per year. And I think there is still not a good come back to that. This is not a knowledge or innovation problem. This is a medicolegal and philosophical problem.
 
  • Like
Reactions: 1 users
Keep in mind that the studies define a Major Adverse Cardiac Event not only as death, but also any myocardial infarction or cardiac revascularization (PCI or CABG). So if someone followed up with cardiology, had an outpatient stress test followed by a heart catheterization, and subsequently got a stent, then they were considered a MACE for follow-up purposes. Diving into the data, the mortality rate for HEART <4 was <0.1%.
 
  • Like
Reactions: 9 users
Members don't see this ad :)
southerndoc said:
Keep in mind that the studies define a Major Adverse Cardiac Event not only as death, but also any myocardial infarction or cardiac revascularization (PCI or CABG). So if someone followed up with cardiology, had an outpatient stress test followed by a heart catheterization, and subsequently got a stent, then they were considered a MACE for follow-up purposes. Diving into the data, the mortality rate for HEART <4 was <0.1%.
Click to expand...
Exactly this. And data out of Kaiser demonstrates that in their patient population, a HEART score of 5 or less has an incidence of mortality or MI of 0.5% at 30 day follow-up. That’s not ready for prime-time, but I occasionally dc chest pain pts with HEART scores >3 who have excellent f/u ,very non-cardiac type chest pain, and normal delta EKGs and trops.
 
  • Like
Reactions: 1 user
Always consider cardiac equivalents. Hospitalists love to ask “why is this patient with an elevated troponin being admitted? Do they even have chest pain?”. Well no, it’s a 75 year old female with fatigue/nausea/dizziness/shortness of air. ACS presents as more than just chest pain, especially in certain patient populations.

Which reminds me: I’ll never forget a hospitalist telling me to “be more careful” who I order a trop on for fear of a positive result when it turned out the patient I admitted to them ended up requiring PCI.

Also ask your hospitalists if they know what a heart score is. Some still don’t. And make sure to discuss what the test means with your patients. It’s helpful to pull out mdcalc on your phone and walk them through it.

Be aware of your patient population and their ability to follow up. Do they have a heart score of 4 but low risk gestalt and they can get into cards or a stress test in 24-48 hrs? Awesome. Will they end up lost to follow up with a heart score of 3 and a high clinical suspicion for disease? I’ll admit them.
 
Ever since we switched to high sensitivity troponin, our chest pain admission rates have skyrocketed. I think the learning point is that clearly not sensitive enough, as a few people are still managing to get discharged.

On a serious note, I find the more annoying patients are not the borderline Heart score patients, but the patient with the high Heart score patietns that are there for chronic chest pain, for their #th visit for the year, with technically abnormal EKG that are unchanged for the last 50 EKGs in the system, and negative stress/PCI in the last # month/week/day.
 
HEART score gives a wonderful framework to having a brief but informed conversation with your patient (i.e. shared decision making) about their hypothetical risk, and alternatives to their diagnostic workup.

Just as an example text--
"What we've done today is enough testing to show you didn't have a heart attack yesterday/today. We can also say, statistically, there is a 98+% chance you'll be absolutely fine for the next month. But that isn't 100% and I want you to understand that. With my HIGH risk patients, I want to admit them to the hospital, have a cardiologist see them, likely get a stress test here before they are allowed to go home. But you are a low/moderate risk patient, and depending on your preference, we could discharge you home with just the testing we've done in the ER, and feel pretty good about it. But I absolutely need you to see your PCP in the next couple of days, and discuss having an outpatient stress test or other tests to make sure you're OK long-term."

Frankly, 19 out of 20 patients with HEART score <=3 want to run out of the ED as fast as possible. Hell, many of my HEART score 6 patients with a single negative troponin want to leave, and we have to pump the breaks and ensure they have a complete understanding of the situation.

In the end, its relatively easy to catch the 95% of people with positive EKGs, positive troponins, classic angina, etc. For patients that understand the conservation we're having, I basically try to explain that in others to get to 100% sensitivity, we have to admit a TON of people, and do a TON of extra testing (including invasive cath), and the math says we are actually harming people more than helping to get to that almost-100% number. Most people grasp this intuitively when I explain it, and seem understanding of going for 95-98% accuracy...

Anyway, is all of this protective of lawsuits? I would posit that if you document your HEART score correctly, and have a real shared decision making conversation with your patient YES it is.
 
god-grant-me-the-serenity-to-accept-the-things-i-cannot-change-the-courage-to-change-the-things-i-can-and-the-wisdom-to-know-the-difference13.jpg

Remember, all a Heart score <4 means is that they have the same baseline risk as almost every other patient we see.
If you followed your abdominal pain, shortness of breath, and even other random complaints like ankle sprains, you'll have a MACE of 0.1-1.7%. Because that encompasses everyone.
Also, you should only be using the Heart score on patients after you have used your gestalt to say "could this be cardiac chest pain". You don't need to to troponins on the 20 year old patients. You might do them on some, but there's no rule that says "chest pain, must do troponin".
I love it when the 8 year olds come in for chest pain and the triage nurse just default orders all that ****. No 8 year old needs a D-Dimer.
 
  • Like
Reactions: 1 user
The practice variations are pretty interesting. I will admit I am pretty conservative with chest pain. I admit a lot of chest pain.

Young (<50) and healthy (relatively) with HEART < 3 is going home in almost if not all cases, unless there's somethnig weird.

Older than 50 I will often admit regardless of HEART score unless they are the picture of health and it's not a weekend/holiday and I know they can get into see their PCP/Cards within 72 hrs for a stress.

I use HEART to backup a discharge that I'm already planning to do based on gestalt, not to tell me what to do.

Whether or not stress is really evidence based, it's still standard of care (I think? Correct me if I'm wrong please) to have them stressed within 72 hours. The best systems I've worked in have had ER/Cards co-managed chest pain centers where patients will get a couple trops and EKGs and if that's negative will get stressed and then dc home if negative. This is the safest thing for both patient and physician. It's not my job to aggressively discharge chest pain because "the hospital won't get paid."
 
Also keep in mind that the HEART score included patients with a positive troponin. If you take those people out (because you'd be admitting them), the actual MACE rate drops significantly.

There was a great ERCast episode about this. In 2018, Kaiser Permanente followed 30,000 patients and found that folks with a HEART score of 0-3 had a 6-week MACE risk of 0.4%, and those with a score of 4-6 had a 2.5% MACE risk (not 12-17% like the original study found). Much much lower numbers than the original study.

If we change things from "6 week MACE risk" to "30 day risk of all-cause mortality or MI", HEART score 0-3 risk was 0.2%, HEART score 4-6 risk was 1%.

Why? Perhaps Americans go to the ED for chest pain much more frequently, and the threshold for "go to the ED" is much lower.

HEART scores 0-3 were sent back to their PCP without any special testing ordered. Scores 4-6 were told to follow-up/referred to cardiology for outpatient testing at the discretion of the cardiologist.

My hospitalists will push back on chest pain OBS with a HEART score of 4-5 due to medical history/age alone (with negative ECG and troponin), and this study seems to say that's ok. But I can also get folks into cardiology within the same week and almost everyone has a PCP in my hospital system.

This study was a practice changer for me.

The HEART Score for Suspected Acute Coronary Syndrome in U.S. Emergency Departments
 
Last edited:
Zebra Hunter said:
Exactly this. And data out of Kaiser demonstrates that in their patient population, a HEART score of 5 or less has an incidence of mortality or MI of 0.5% at 30 day follow-up. That’s not ready for prime-time, but I occasionally dc chest pain pts with HEART scores >3 who have excellent f/u ,very non-cardiac type chest pain, and normal delta EKGs and trops.
Click to expand...

Why do you think this is not ready for prime time? 30,000 US patients in a health care system that has about as tight of follow-up as we can accomplish in this country.
 
Zebra Hunter said:
Exactly this. And data out of Kaiser demonstrates that in their patient population, a HEART score of 5 or less has an incidence of mortality or MI of 0.5% at 30 day follow-up. That’s not ready for prime-time, but I occasionally dc chest pain pts with HEART scores >3 who have excellent f/u ,very non-cardiac type chest pain, and normal delta EKGs and trops.
Click to expand...

The Kaiser data has some concerns. Most HEART studies had around a 35% score of <4. Kaiser's study approached 50%. Makes me wonder if they were applying the HEART score to people who didn't need it (i.e., 18 year olds with GERD, traumatic chest pain, etc.).
 
  • Like
Reactions: 1 user
Had one attending who’s decision making was “do you think they are having cardiac chest pain or not.” And that was it. Regardless age, medical history, etc.
 
Members don't see this ad :)
southerndoc said:
The Kaiser data has some concerns. Most HEART studies had around a 35% score of <4. Kaiser's study approached 50%. Makes me wonder if they were applying the HEART score to people who didn't need it (i.e., 18 year olds with GERD, traumatic chest pain, etc.).
Click to expand...

Good point. Are you comparing this to the original HEART validation studies in the Netherlands, or the HEART pathway, or what? I have a feeling we just have a lot more nonsense visits for chest pain in the US.
 
Low risk HEART goes home, generally with a repeat trop at 3 hours unless patient feels satisfied with single trop.
Moderate risk gets admitted, though many still prefer home with follow up and I'll generally oblige without much fight assuming spouse or whoever else is there seems cool with it.
I can get rapid cards or PCP follow up.
Our cardiologists are also generally happy to chat about patients if needed.
Hospitalists are RVU based and most will accept an ACS r/o dispo without much fuss, but I also try to keep my BS admit rate low enough that we all feel I'm selling a straight deal.
 
HoosierdaddyO said:
After having a lengthy conversation with a few other attendings, it was really funny and interesting to see how 4 different doctors approached chest pain disposition all differently. Some use a hard cutoff heart score, others a gestalt of 1 or 2 trops and dispo home with followup... some just admit everyone and I mean everyone. I think for academics purposes, looking to see if I can start a discussion on chest pain work ups and what you are doing for these peeps :)?!
Click to expand...

Oh boy....now a riveting discussion on chest pain dispositions. This will surely bring out the flames, links to studies, and links to studies that are not applicable.
 
pkwraith said:
On a serious note, I find the more annoying patients are not the borderline Heart score patients, but the patient with the high Heart score patietns that are there for chronic chest pain, for their #th visit for the year, with technically abnormal EKG that are unchanged for the last 50 EKGs in the system, and negative stress/PCI in the last # month/week/day.
Click to expand...

These people are hard to take care of. They have 99 problems and the heart is one. They are ticking time bombs and will explode at some point, maybe on your shift. Nobody wants to discharge the CABG, ESRD, CAD, STENT, PAD, PVD. Nobody can really risk stratify these patients with any amount of confidence.
 
southerndoc said:
Keep in mind that the studies define a Major Adverse Cardiac Event not only as death, but also any myocardial infarction or cardiac revascularization (PCI or CABG). So if someone followed up with cardiology, had an outpatient stress test followed by a heart catheterization, and subsequently got a stent, then they were considered a MACE for follow-up purposes. Diving into the data, the mortality rate for HEART <4 was <0.1%.
Click to expand...

That's the problem with MACE as a primary endpoint in these trials.
 
Fox800 said:
There was a great ERCast episode about this. In 2018, Kaiser Permanente followed 30,000 patients and found that folks with a HEART score of 0-3 had a 6-week MACE risk of 0.4%, and those with a score of 4-6 had a 2.5% MACE risk (not 12-17% like the original study found). Much much lower numbers than the original study.

If we change things from "6 week MACE risk" to "30 day risk of all-cause mortality or MI", HEART score 0-3 risk was 0.2%, HEART score 4-6 risk was 1%.
Click to expand...

Kaiser patients are a particular cohort that is not representative of the population at large. I work at a Kaiser hospital and a non-Kaiser hospital. The amount of medical knowledge, overall knowledge, and desire to be good healthy people is substantially higher with Kaiser patients. I would never apply a Kaiser study to the people I see at my other hospital, who in general don't give a crap about their health and just want Test'n'Pills for all their medical complaints.
 
I use a combination of HEART and ADAPT with delta trop @2h. I think the data looks good and it's a reasonable approach. If its constant chest pain with atypical features for over 6 hours and I have a low clinical suspicion, then I won't even get the 2nd trop. If my clinical suspicion is high, then I'll admit. If it's a borderline case, I'll consult cards and throw their name in the chart. That's really all you can do. Luckily, all my cardiologists and hospitalists are fairly reasonable at my current hospital. For the low HEART score patients that get d/c, I can set them up with an outpatient stress test within 4-5 days.
 
Groove said:
I use a combination of HEART and ADAPT with delta trop @2h. I think the data looks good and it's a reasonable approach. If its constant chest pain with atypical features for over 6 hours and I have a low clinical suspicion, then I won't even get the 2nd trop. If my clinical suspicion is high, then I'll admit. If it's a borderline case, I'll consult cards and throw their name in the chart. That's really all you can do. Luckily, all my cardiologists and hospitalists are fairly reasonable at my current hospital. For the low HEART score patients that get d/c, I can set them up with an outpatient stress test within 4-5 days.
Click to expand...
Why >6 hrs instead of 3hrs? If you would be comfortable doing a 3hr delta trop on a pt with 1 min of chest pain per the HEART score study, why aren't you comfortable doing a single trop with constant pain >3 hrs? In my low risk/low gestalt patients with constant pain >3 hrs, they get one trop only.
 
I'm not big on serial troponins. If I'm concerned enough about cardiac etiology to get more than one I'm generally going to admit them to the hospital. I find now that I probably send 75% of chest pains home.
 
  • Like
Reactions: 1 user
BoardingDoc said:
Why >6 hrs instead of 3hrs? If you would be comfortable doing a 3hr delta trop on a pt with 1 min of chest pain per the HEART score study, why aren't you comfortable doing a single trop with constant pain >3 hrs? In my low risk/low gestalt patients with constant pain >3 hrs, they get one trop only.
Click to expand...

I get what you're saying. The AHA has said that with >6 hours of continuous symptoms, a negative troponin has basically a 100% negative predictive value for MI. The initial HEART score study did not include a delta troponin. It was the HEART Pathway study that threw this in, which is what a lot of places have bought into. Just depends on your risk tolerance I presume.

The original HEART score study had a 2.2% MACE rate for the low risk group with a single troponin. HEART pathway improved that to <1% by adding a 3-hour delta trop. A negative trop at >6 hours should effectively be 0% chance of MI, not sure how this would change the MACE rate but it would still be ridiculously low.
 
  • Like
Reactions: 1 user
BoardingDoc said:
Why >6 hrs instead of 3hrs? If you would be comfortable doing a 3hr delta trop on a pt with 1 min of chest pain per the HEART score study, why aren't you comfortable doing a single trop with constant pain >3 hrs? In my low risk/low gestalt patients with constant pain >3 hrs, they get one trop only.
Click to expand...

We have all had pts who have chest pain for days, either allegedly constant or probably comes and goes, whose second troponin is positive. Happens to me about 1/20. So I usually two trop patients most of the time.
 
thegenius said:
Kaiser patients are a particular cohort that is not representative of the population at large. I work at a Kaiser hospital and a non-Kaiser hospital. The amount of medical knowledge, overall knowledge, and desire to be good healthy people is substantially higher with Kaiser patients. I would never apply a Kaiser study to the people I see at my other hospital, who in general don't give a crap about their health and just want Test'n'Pills for all their medical complaints.
Click to expand...

I guess my shop is closer to the Kaiser model in the sense that most people have a PCP and I can get them into cardiology, and they aren't using drugs. Otherwise I see just as many old vasculopaths (CAD + HTN + DM + hyperlipidemia) as any other place. Just less cocaine and meth in the mix. They have a HEART score of 4-5 just walking through the parking lot for their med refill. I don't see how being a Kaiser Patient would un-hypertensionize and un-diabetesize them.

I don't see how these folks are drastically different. Risk factors are risk factors are risk factors. With a negative work-up, are you more inclined to admit these folks to OBS at your non-Kaiser shop if they have two negative troponins, just because they don't have a PCP and are more squirrely?

If you can't assure follow-up and they have horrible protoplasm and you think they need inpatient cards evaluation then that's one thing. More and more we're seeing that outpatient stress testing doesn't change outcomes.
 
Last edited:
  • Like
Reactions: 1 user
Fox800 said:
I guess my shop is closer to the Kaiser model in the sense that most people have a PCP and I can get them into cardiology, and they aren't using drugs. Otherwise I see just as many old vasculopaths (CAD + HTN + DM + hyperlipidemia) as any other place. Just less cocaine and meth in the mix. They have a HEART score of 4-5 just walking through the parking lot for their med refill. I don't see how being a Kaiser Patient would un-hypertensionize and un-diabetesize them.

I don't see how these folks are drastically different. Risk factors are risk factors are risk factors. With a negative work-up, are you more inclined to admit these folks to OBS at your non-Kaiser shop if they have two negative troponins, just because they don't have a PCP and are more squirrely?
Click to expand...

If evaluating chest pain patients just came down to identifying risk factors, look at the EKG and interpreting a troponin result, then I agree it doesn't matter what health system you are in.

But with the above example MacKenzine McJenneyson PA could evaluate a patient with chest pain, plug in the numbers into a computer and follow the result of "Admit or Discharge". And we all have our opinion on that based on numerous threads here over the past year.

Hospital 1: 2.2 pph, cohort is largely high school educated only pts of a variety of ethnicities, English is not primary language in 20% of cases, they have difficulty describing their symptoms, often go to the ED and don't see their regular doctor, you have charts that document different histories, you cannot have a normal discussion about risks and benefits about testing because they don't understand, they don't know what medicines they take except "a whole bunch, some are blue, one is green", etc., shared decision making conversations are nearly impossible.

Hospital 2: 1.2 pph, cohort is largely college-educated pt who are vast majority one ethnicity, English is primary language in 90% of cases, they may or may not have difficulty describing their symptoms, see their doctor on a regular basis, know what medicines they take, you ask them questions and they answer them normally, if you say "the current guidelines say to do 'x'", they appear to understand the general rationale behind it. You can undergo shared decision making because you can have a normal conversation with them.

Comparing Hospital 1 and 2, there are numerous differences in delivering health care - the major differences are 1) you have time at one hospital to actually see patients, and 2) degree of education. I think that equates to a significant difference in delivering patient care.

Fox800 said:
If you can't assure follow-up and they have horrible protoplasm and you think they need inpatient cards evaluation then that's one thing. More and more we're seeing that outpatient stress testing doesn't change outcomes.
Click to expand...

Totally agree with that...we (the health care system) order tests for a pre-test probability of disease <1%, and the testing characteristics are fairly lousy with a sensitivity and specificity between 75-90% depending on what you order.

Low risk chest pain should just undergo medical therapy / lifestyle modifications and not even get stress tests.

In fact, we should just not have stress tests. You either get a cath or medical therapy. I am very slowly discharging more and more patients with chest pain as compared to a few years ago, and I haven't ordered a stress test in the ED in over a year.
 
  • Like
Reactions: 1 user
After the first trop on low risk patients, I tell them about options including repeat trop to get more certainty. I'd say 30+% opt to go home after the initial trop.
Fox800 said:
I get what you're saying. The AHA has said that with >6 hours of continuous symptoms, a negative troponin has basically a 100% negative predictive value for MI. The initial HEART score study did not include a delta troponin. It was the HEART Pathway study that threw this in, which is what a lot of places have bought into. Just depends on your risk tolerance I presume.

The original HEART score study had a 2.2% MACE rate for the low risk group with a single troponin. HEART pathway improved that to 6 hours should effectively be 0% chance of MI, not sure how this would change the MACE rate but it would still be ridiculously low.
Click to expand...
 
  • Like
Reactions: 1 user
My decision to get a delta trop is usually based on the timing of the pain. If the patient is moderate/high risk by HEART (which I use) then they are admitted. The only time I get a delta trop on a patient who is going to be admitted for chest pain anyways is if I am really worried about them and trying to push for an emergent cath in a patient that seems like an OMI to me but doesn't have a clear cut STEMI (or STEMI equivalent that the cardiologists believes in) on EKG.

If the patient is low risk by HEART and potentially dischargeable if they have had more than 6 hours of constant pain, I think one trop is enough. If it's less than 6 hours I need a negative trop at 6 hours since onset to discharge. Example: If the patient's chest pain started 3 hours prior to arrival in the ER, I get an initial trop with the initial labs and then a repeat trop 3 hours later, so that 2nd trop is 6 hours + from chest pain onset.

Once in a blue moon I'll get a delta trop on a HEART score 3-type player even with more than 6 hours of chest pain just to strengthen the case for discharge a la the HEART score pathway. Of note I always recommend cardiology follow up in 48 hours for stress test for all discharged chest painers. I doubt 5% of patient's comply with this recommendation, but that is what I officially document in my MDM (that I verbalized this recommendation to the patient who verbalized back comprehension) and written discharge instructions.
 
thegenius said:
But with the above example MacKenzine McJenneyson PA could evaluate a patient with chest pain, plug in the numbers into a computer and follow the result of "Admit or Discharge". And we all have our opinion on that based on numerous threads here over the past year.
Click to expand...
Except that the utility of the doctor brain isn't following that algorithm. It's knowing who to use it on. Otherwise you'd be getting labs/XR/EKG on everyone. The 8 year old who got punched in the chest. The crossfitter who went for a new PR. Everyone.
The pathway is to see who is low risk on the people you think are having cardiac chest pain, not everyone. Otherwise you can get heart scores >4 on ankle sprains.
 
  • Like
Reactions: 1 user
BoardingDoc said:
Why >6 hrs instead of 3hrs? If you would be comfortable doing a 3hr delta trop on a pt with 1 min of chest pain per the HEART score study, why aren't you comfortable doing a single trop with constant pain >3 hrs? In my low risk/low gestalt patients with constant pain >3 hrs, they get one trop only.
Click to expand...

See @Fox800's post. If their CP was at least 6 hours ago and the trop is negative, then my NPV is almost 100% for low risk pts. I should preface that by saying that personally, I like my 2nd troponin to be at least 6 hours after sx. So, I'd be ok getting my first at 4h and my 2nd at 6h. Again, it really just depends on the pt. If they are atypical, HEART <=3, 3d continuous pain, etc... then I will simply HEART them, acquire a single troponin and d/c them with f/u if indicated. However, we all get the pt's that are not so straightforward. They simply aren't great historians, or they haven't been to the doctor in years and I suspect they may have more CAD risk factors than they admit, the pain might be intermittent, etc.. Any number of things might obfuscate the presentation without significantly altering my underlying gestalt or pre-test probability into the realm of admission/formal r/o. For those patients, as long as the pain was 4-6h ago, I will incorporate a HEART and ADAPT approach and if their HEART < =3 and delta trop negative, then I feel fairly comfortable discharging these pts especially considering we have very close f/u and a system where I can get these pt's stress tests within 4-5 days.

Anybody else gets admitted for formal rule outs. If they have chest pain again while I'm working them up, they also get admitted. In the end, I'm trying to balance overloading the system with needless chest pain admissions while safely caring for the pt's. At this point in time with all the excellent research in the past 10 years, I really feel like we've got solid evidence based guidelines and approaches for dealing with a great number of these patients that would otherwise have all been admitted in times past.

I should also say we don't have highly sensitive troponins at my shop. I'm not sure I want them either. We use a combination of POC Troponins in the ED and send down troponins. That's another topic entirely. For instance, our POC machines upper limit is 0.08 and our central lab assay upper limit is 0.034. You can't even get a decimal place of sensitivity on the POC machines. In fact, if you start digging into the literature, the only time these machines approach equal sensitivity/specificity is if you artificially lower the upper limit to 0.04. Also, many of the studies showing equal sensitivity were funded by the manufacturers for the POC machines. I will admit the accuracy has improved in the machines over the first generations, however there is still a noticeable increase in quality error rates and the machine inherently is at risk for more preanalytical errors. So, I still get an uncomfortable variance in values with the machines. I really dug into it a few years ago and essentially, depending on the manufacturer, anything between 0.0-0.02 is standard error and not accurately representative of an actual troponin value whatsoever. Meaning, if your initial value is 0.00 and you're next value is 0.02, that might be a standard of expected error intrinsic to the machine and not representative of an increased level of troponin in the actual assay. So, you can see how frustrating it can be when we are trying to calculate deltas with a POC device. Not that I don't do it...but it really just depends on the clinical case. For instance, if I get a 0.03 on the POC, many times I send a sample down to our central lab to correlate OR I will simply send down my next 2h sample for a central assay IF I'm getting deltas. Most of the times if I already decide to do a delta r/o, I'm sending one or both of the trop's downstairs to our central lab. Otherwise, I'm exclusively using our POC machines which I both love and hate for the above reasons.

Anyway, after reading in here, It sounds like we're all doing variations of the same thing. Everyone is going to have their own comfort levels. It's not like HEART is perfect by any means. There is still some reproducibility issues as the first question of History is entirely subjective and so is the EKG interpretation for that matter. I might say they're low suspicion, another might say moderate to high suspicion. I say it's a normal EKG, another might say non specific repolarization disturbance. I say HEART 0, someone else says HEART 3, etc..
 
Last edited:
  • Like
Reactions: 1 user
Anyway, speaking of chest pain...I need to go get ready for work and just logged in to look at the tracking board. I'm by myself until 9a and there's already 22 pt's in the ED... I need some alka-seltzer.
 
  • Like
Reactions: 1 user
Groove said:
Anyway, speaking of chest pain...I need to go get ready for work and just logged in to look at the tracking board. I'm by myself until 9a and there's already 22 pt's in the ED... I need some alka-seltzer a new job.
Click to expand...
 
  • Like
Reactions: 1 user
Resurrecting this thread for a specific question.

Is there good data that answers the following question
Among patients that present to the ED with active chest pain or had chest pain earlier, and with an EKG that is at the patient's baseline or is not ischemic and not dynamic, is there 30-day MACE data if they are discharged after having two negative troponins at least three hours apart? This question applies to all comers independent of their HEART Score.

Basically the question is asking if you have two negative troponins three hours apart, for any risk of chest pain (low, medium or high on the HEART SCORE), what is their 30 day risk of having a bad outcome.

I don't know the answer because we tend to bundle these into low, medium, and high risk per the HEART score. But that is not what this question is asking. Thanks
 
thegenius said:
Resurrecting this thread for a specific question.

Is there good data that answers the following question
Among patients that present to the ED with active chest pain or had chest pain earlier, and with an EKG that is at the patient's baseline or is not ischemic and not dynamic, is there 30-day MACE data if they are discharged after having two negative troponins at least three hours apart? This question applies to all comers independent of their HEART Score.

Basically the question is asking if you have two negative troponins three hours apart, for any risk of chest pain (low, medium or high on the HEART SCORE), what is their 30 day risk of having a bad outcome.

I don't know the answer because we tend to bundle these into low, medium, and high risk per the HEART score. But that is not what this question is asking. Thanks
Click to expand...

You're describing the HEART pathway. I'm assuming you could just pull the data for all of the patients in that study.
 
Fox800 said:
You're describing the HEART pathway. I'm assuming you could just pull the data for all of the patients in that study.
Click to expand...

The HEART Pathway uses an objective history, history of CAD, and absence of ischemic changes in its determination of low vs. not low risk. The low-risk patients get a 2nd troponin at 3 hours and discharge of negative. Moreover, the objective history assigns points for pain characteristics such that you really need to use the app or have the pathway integrated into your EHR to do it properly.

thegenius said:
I looked at the HEART study results (https://www.ahajournals.org/doi/pdf/10.1161/CIRCOUTCOMES.114.001384) and it doesn't look like that data is reported. It's assuredly within their raw data set but what I was looking for wasn't reported.
Click to expand...

The closest decision rules to what you describe come from Europe and use hs-Tn. The High-STEACS (High-Sensitivity Troponin in the Evaluation of Patients With Acute Coronary Syndrome) pathways uses a troponin cutoff of <5 ng/L at 0 and 3 hours with only a non-ischemic ECG and no clinical information.


In addition, a modification of the Manchester Acute Coronary Syndromes model called the T-MAC (T for troponin only) also kinda accomplishes what you want, but with hs-Tn.

 
  • Like
Reactions: 1 user
thegenius said:
Resurrecting this thread for a specific question.

Is there good data that answers the following question
Among patients that present to the ED with active chest pain or had chest pain earlier, and with an EKG that is at the patient's baseline or is not ischemic and not dynamic, is there 30-day MACE data if they are discharged after having two negative troponins at least three hours apart? This question applies to all comers independent of their HEART Score.

Basically the question is asking if you have two negative troponins three hours apart, for any risk of chest pain (low, medium or high on the HEART SCORE), what is their 30 day risk of having a bad outcome.

I don't know the answer because we tend to bundle these into low, medium, and high risk per the HEART score. But that is not what this question is asking. Thanks
Click to expand...

The other thing I wonder is how many patients who are admitted for chest pain have a meaningful change in their outcome at 30 days and 1 year? Does an inpatient stress test actually change anything in terms of outcomes versus an outpatient test regardless of heart score? Should patients with a positive stress test but who have resolution of pain and no significant EKG changes or troponin elevation even get a cath? I don't know, but I guess both the benefit and the drawback of the HEART score is that we can push these decisions and the associated medicolegal liability onto our hospitalist colleagues. The frustrating thing is that I don't necessarily think that a hospitalist is better equipped to make these decisions than we are, but the 'high risk chest pain' admission is still standard of care at so many places.
 
SamtheWise said:
The other thing I wonder is how many patients who are admitted for chest pain have a meaningful change in their outcome at 30 days and 1 year? Does an inpatient stress test actually change anything in terms of outcomes versus an outpatient test regardless of heart score? Should patients with a positive stress test but who have resolution of pain and no significant EKG changes or troponin elevation even get a cath? I don't know, but I guess both the benefit and the drawback of the HEART score is that we can push these decisions and the associated medicolegal liability onto our hospitalist colleagues. The frustrating thing is that I don't necessarily think that a hospitalist is better equipped to make these decisions than we are, but the 'high risk chest pain' admission is still standard of care at so many places.
Click to expand...

Nothing we do really changes anything when it comes to outpatient stress tests.
 
southerndoc said:
EM:RAP recently had a good summary of all the latest research. Basically, stress tests do nothing.
Click to expand...

Did you find any holes in Dr. Morgenstern’s analysis, or are you discharging all of your HEART > 3 patients after 2 or 3 trops?
 
I basically follow the herd on this one. If everybody starts espousing something, like the HEART pathway, then I'll use it. I don't have the time, nor inclination, to read all the specific studies regarding chest pain workups in the ED.

You know how you get these patients who have numerous risk factors for having an MI (previous stents, CABG, hyperlipidetes, the entire thing).....and you get two troponins over 3-4 hours and they are <0.015 x2. And you are just thinking to yourself...well they ruled out. And you really want to send him home because he ain't having a heart attack. he probably scores a 5 just based on his age, EKG non-specific repolarization, and RF. Just wanted to know if there is evidence to discharge this kind of patient.
 
thegenius said:
I basically follow the herd on this one. If everybody starts espousing something, like the HEART pathway, then I'll use it. I don't have the time, nor inclination, to read all the specific studies regarding chest pain workups in the ED.

You know how you get these patients who have numerous risk factors for having an MI (previous stents, CABG, hyperlipidetes, the entire thing).....and you get two troponins over 3-4 hours and they are <0.015 x2. And you are just thinking to yourself...well they ruled out. And you really want to send him home because he ain't having a heart attack. he probably scores a 5 just based on his age, EKG non-specific repolarization, and RF. Just wanted to know if there is evidence to discharge this kind of patient.
Click to expand...

Yes, the Kaiser Permanente study looked at this. They routinely sent home HEART scores up to 5.

www.ncbi.nlm.nih.gov

The HEART Score for Suspected Acute Coronary Syndrome in U.S. Emergency Departments

www.ncbi.nlm.nih.gov www.ncbi.nlm.nih.gov
 
  • Like
Reactions: 1 user
  • Like
Reactions: 1 user
Fox800 said:
Yes, the Kaiser Permanente study looked at this. They routinely sent home HEART scores up to 5.

www.ncbi.nlm.nih.gov

The HEART Score for Suspected Acute Coronary Syndrome in U.S. Emergency Departments

www.ncbi.nlm.nih.gov www.ncbi.nlm.nih.gov
Click to expand...

This is probably closer to the truth and where we need to be. The problem with the early data on stress testing (and coronary CTA) from the ED is that they were observational studies that enrolled a lot of low and no-risk patients. These were patients that would have had HEART scores in the 1-3 range. They unleashed tests with poor specificity on populations where the prevalence of disease was close to zero with predictable results. Truth be told, we really don’t know how nuc med studies, stress echo, and CTA perform just on these truly moderate risk patients with HEART scores greater than 4 or 5. If I had to hazard a guess, the most cost effective approach would look something like this:
1) HEAR 0 and 1 - don’t even get a troponin unless some extenuating circumstances (cocaine) as these are the truly zero-risk patients
2) HEAR 2-3 - 2 trops over 3 hrs and D/C if negative
3) HEAR 4-5 - 2 trops over 6 hours and D/C if negative
4) HEAR 6-7 - Nuc Med, CTA, or CMR depending on the patient and score
5) HEAR > 7 Cards consult for provocative vs. invasive strategy.

In other words, there are a lot of HEART score 4 & 5 patients who get a stress that they do not need.
 
Last edited:
  • Like
Reactions: 1 user
You must log in or register to reply here.
Next unread thread

Similar threads

Mount Asclepius
Chest Pain Free
Replies
39
Views
2K
Zebra Hunter
Z
SpacemanSpifff
Using high sensitivity troponin in chest pain dispo
Replies
43
Views
3K
Mount Asclepius
Mount Asclepius
Momoffuturemedstudent
Heart Expelled Through Chest Wall, Sternotomy Disaster
2
Replies
59
Views
4K
ShockIndex
ShockIndex
A
Back pain management
Replies
37
Views
2K
dozitgetchahi
D
Tenk
Me ordering a d-dimer or lactate lately
Replies
14
Views
2K
turkeyjerky
T
Top