How do you guys deal with the feeling of "missing something"?

[wamcp]

Young ED Heart Attack

When things like this happen...and you are seeing 2-3 patients per hour, sending folks home whom you just barely met

---

Members don't see this ad.

 

[southerndoc]

In this case, the commentary is spot on: repeat visits are high risk and should be carefully evaluated, try to be mindful of getting tunnel vision from prior diagnoses, and always ask yourself if you would do the same thing if it were a family member.

A few things come to mind:

1. All chest pain patients -- except traumatic -- get a troponin. Some might say that one troponin begets a second troponin, but this is not always the case. A low HEART score and a single troponin with pain >3 hours can be used sometimes. Everyone seems to center on the fact that a troponin is to assess cardiac ischemia. Remember, it can also help with myocarditis and myopericarditis. Both of these are important conditions to catch and they primarily affect the young. So yes, an 18-year-old will get an i-STAT troponin if he/she presents with chest pain.

2. Having done a countless number of reviews (peer review for state medical board, peer review for CQI at a very large health system, and expert witness review for the defense), I see one theme that has been frequent and consistent in missed MI's: they frequently present with atypical characteristics and often are relieved with GI cocktails. Perhaps it's the doc trying to make a chart fit his or her diagnosis, or perhaps the patient thinks it helped from a placebo effect. Perhaps its transmigration of lidocaine across the esophagus and affecting the vagal nerve. Whatever the physiology, just keep in mind that pain relieved with a GI cocktail does not rule out cardiac cause just as much as pain relieved with nitroglycerin does not rule in cardiac cause.

3. Young individuals should always get extra scrutiny. They have long lives ahead of them. That means no matter what your state caps are, you could be sued for lost wages and loss of consortium.

4. When a good plaintiff attorney secures even a semi-decent expert witness, he or she will scrutinize your care over the tiniest thing that even you will second guess your care. Subtle EKG changes that are only apparent with a magnifying glass suddenly become "obvious" and you are "grossly negligent, doctor, for not recognizing this." Your time for recognizing labs and performing actions will be subject to analysis. If the lab reports a troponin ten times normal and you don't note it until 30 minutes later, you should document that you were tied up with another critical patient, that the department was in a high saturation state, etc. The jury and even plaintiff expert witnesses will sometimes -- not always -- have sympathy on this.

5. Repeat visits should receive more of your attention than typical patients. Repeat after me: Repeat visits are high risk. Repeat visits are high risk. Repeat visits are high risk. Blow off a repeat visit at your own peril. Plenty of highly qualified physicians have stumbled into this hole, and they have not emerged unscathed. No, it doesn't mean you need to pan scan or do a million dollar workup on every repeat visit, but you should put more thought into things and ask yourself "Am I missing something? What if this was presented in front of a jury?"

6. A patient's numerous other visits for anything are likely inadmissable in most courts. A patient presented 200 times over a 2 year period with severe alcohol intoxication but the last visit resulted in his death from a missed subdural? A good plaintiff attorney will prevent this information from being presented. You'll likely have to defend your case as if it was the patient's sentinel visit.

7. Use proven risk stratification tools (HEART score, Canadian Head CT rule, etc.). It shows that you put thought into a patient's care instead of blowing them off.

8. Most successful litigation happens because the patient feels as if he or she wasn't take seriously or there was a breakdown in communication. Had the third visit included a more extensive workup and the patient been discharged with a negative workup, there likely would not have been a lawsuit. Discharging a patient with a serious complaint like non-traumatic chest pain within 5 minutes looks flagrant and grossly negligent to any jury. Furthermore, this hospital could have been placed on a fast track by CMS for an EMTALA violation since he was not properly screened and was discharged with an unstable medical condition. The thing to remember is that a patient treated by a physician they view as caring and compassionate is unlikely to be sued no matter how bad the outcome.

 

[startupquick]

Fantastic response. Thanks

In this case, the commentary is spot on: repeat visits are high risk and should be carefully evaluated, try to be mindful of getting tunnel vision from prior diagnoses, and always ask yourself if you would do the same thing if it were a family member.

A few things come to mind:

1. All chest pain patients -- except traumatic -- get a troponin. Some might say that one troponin begets a second troponin, but this is not always the case. A low HEART score and a single troponin with pain >3 hours can be used sometimes. Everyone seems to center on the fact that a troponin is to assess cardiac ischemia. Remember, it can also help with myocarditis and myopericarditis. Both of these are important conditions to catch and they primarily affect the young. So yes, an 18-year-old will get an i-STAT troponin if he/she presents with chest pain.

2. Having done a countless number of reviews (peer review for state medical board, peer review for CQI at a very large health system, and expert witness review for the defense), I see one theme that has been frequent and consistent in missed MI's: they frequently present with atypical characteristics and often are relieved with GI cocktails. Perhaps it's the doc trying to make a chart fit his or her diagnosis, or perhaps the patient thinks it helped from a placebo effect. Perhaps its transmigration of lidocaine across the esophagus and affecting the vagal nerve. Whatever the physiology, just keep in mind that pain relieved with a GI cocktail does not rule out cardiac cause just as much as pain relieved with nitroglycerin does not rule in cardiac cause.

3. Young individuals should always get extra scrutiny. They have long lives ahead of them. That means no matter what your state caps are, you could be sued for lost wages and loss of consortium.

4. When a good plaintiff attorney secures even a semi-decent expert witness, he or she will scrutinize your care over the tiniest thing that even you will second guess your care. Subtle EKG changes that are only apparent with a magnifying glass suddenly become "obvious" and you are "grossly negligent, doctor, for not recognizing this." Your time for recognizing labs and performing actions will be subject to analysis. If the lab reports a troponin ten times normal and you don't note it until 30 minutes later, you should document that you were tied up with another critical patient, that the department was in a high saturation state, etc. The jury and even plaintiff expert witnesses will sometimes -- not always -- have sympathy on this.

5. Repeat visits should receive more of your attention than typical patients. Repeat after me: Repeat visits are high risk. Repeat visits are high risk. Repeat visits are high risk. Blow off a repeat visit at your own peril. Plenty of highly qualified physicians have stumbled into this hole, and they have not emerged unscathed. No, it doesn't mean you need to pan scan or do a million dollar workup on every repeat visit, but you should put more thought into things and ask yourself "Am I missing something? What if this was presented in front of a jury?"

6. A patient's numerous other visits for anything are likely inadmissable in most courts. A patient presented 200 times over a 2 year period with severe alcohol intoxication but the last visit resulted in his death from a missed subdural? A good plaintiff attorney will prevent this information from being presented. You'll likely have to defend your case as if it was the patient's sentinel visit.

7. Use proven risk stratification tools (HEART score, Canadian Head CT rule, etc.). It shows that you put thought into a patient's care instead of blowing them off.

8. Most successful litigation happens because the patient feels as if he or she wasn't take seriously or there was a breakdown in communication. Had the third visit included a more extensive workup and the patient been discharged with a negative workup, there likely would not have been a lawsuit. Discharging a patient with a serious complaint like non-traumatic chest pain within 5 minutes looks flagrant and grossly negligent to any jury. Furthermore, this hospital could have been placed on a fast track by CMS for an EMTALA violation since he was not properly screened and was discharged with an unstable medical condition. The thing to remember is that a patient treated by a physician they view as caring and compassionate is unlikely to be sued no matter how bad the outcome.

 

[Angry Birds]

Young ED Heart Attack

When things like this happen...and you are seeing 2-3 patients per hour, sending folks home whom you just barely met

sh1t happens

 

[GeneralVeers]

Uncommon things happen uncommonly. Unfortunately it's not possible to tailor workups to find the zebras. I agree with risk stratification tools like HEART etc. I also like the ACEP chest pain > 8 hours guideline. Documenting chest wall tenderness also is a highly negative predictor for cardiac origin of chest pain. Most of us are going to miss uncommon presentations of uncommon things.

 

[gamerEMdoc]

Negative ecg + neg trop and a heart <3 = 2% risk regardless of timing. The second troponin take that risk down to below 1%. I would argue the second trop doesn’t really add that much, although it’s probably more medical legally protective than actually that important in low heart score chest pain.

In the end this wasn’t a case of needing to practice fearful medicine, It was just bad medical care. The first time the patient came, and EKG only was probably appropriate assuming they did a decent history. Had they asked about cocaine use, their plan may have changed. The second visit, the patient had new ST segment changes and didn’t get a trop. The third visit, he got chastised and didn’t get an ECG and was DC home in 5 minutes on his 3rd visit without even getting an ECG on his 3rd visit to the ED for chest pain.

Over the top care would be admitting a 26 year old with a normal EKG and normal troponin for chest pain. But on the second visit, he definitely should’ve had a troponin and the third, an ECG and troponin.

One thing that drives me nuts is when a resident tells me that a patient has a low risk HEART score and they don’t want to do a troponin. My response is always the same... the troponin is part of the heart score.

In terms of the originial question, how do you get comfortable missing things, the answer is easy. You practice good medicine, and you learn to understand acceptable miss rate can’t be 0% or people will get hurt. A heart catheterization can cause a heart attack and can kill people. People can die from iatrogenic problems in the hospital. There’s just a point that doing more is worse for the patient, and its better to have real discussions with the patient about risk and the limitations of testing and good return precautions.

There is a good study on what happens to CP patients admitted to the hospital after a nonischemic ECG and negative troponin by Scott Weingart that was in the NEJM (if I remember correctly). They looked at thousands of patients, and the patients were much higher risk than a standard low HEART score. They looked solely and patient oriented endpoints, basically risk of MI or death. The risk of these two end points was extraordinarily low and all of the deaths were either completely unrelated to why they came in for CP and most were iatrogenic (ie gi bleed after heparin, MI from a cath, etc).

Take home point, its ok to stop after a thorough history and solid reasonable ED workup. But its not ok to not even take a history (no one asked about cocaine in 3 visits) nor to do an incomplete evaluation, especially when someone is bouncing back.

 

[gman33]

Visit 1 maybe ekg only.
Visit 2 ekg and 2 sets enzymes +/- observation.
Visit 3. The unit clerk admits them as soon as they check in.

 

[Arcan57]

Agree with above, dissections can also present like this. I think it's tough sometimes to see these cases when you're in them because there are all sorts of cues that don't make it into the chart. Due to outcome and description of case we're all picturing a citizen that's genuinely concerned about what's going on and is participating in the history and physical to the best of their abilities. That's certainly the picture the plaintiff's attorney is going to paint. But what if the patient is on their phone while you're trying to get history, called the ambulance because they didn't want to wait in the lobby, and decided it was cheaper to come to the ED then buy a bottle of Comfort Liquid at Walgreens? Now all you're instincts are screaming "F%# this guy", and you start turning into a PCP trying to encourage medication compliance and appropriate venues of care instead of being an ED doc.

That's why cognitive forcing strategies like Southerndoc pointed out are important. The HEART score doesn't care if your patient has 3x upper limit of normal squirrel titers. PERC doesn't factor in a patient's Axis I or II disorders. You can certainly pull in data from other sources (- CT abd/pel yesterday means it's highly unlikely pt just developed a ruptured AAA, etc) but you're never excused from the thought process of ruling out the emergent/life threatening causes of the patient's presentation during THIS visit.

 

[PlutoBoy]

Visit 1 maybe ekg only.
Visit 2 ekg and 2 sets enzymes +/- observation.
Visit 3. The unit clerk admits them as soon as they check in.

I would have admitted him on second visit. It's just easier and safer to do a full work up than it is to accept the liability of missing something with lethal consequences.

Sometimes our pride takes us to dangerous places. It sounds like the physicians involved in that case didn't want to consider that they were wrong and the patient was right.

I no longer work in the hospital but I am faced with similar issues in the office. A common one is abdominal pain. Sure, I'm not going to scan everyone or have everyone scoped but my conversation and documentation usually goes like this;

"Mr./Ms. X, your history and physical exam suggests a benign process. Now, I cannot tell you with 100% certainty what is going on but in my experience I think it is okay to monitor this. I would hate to put you through unnecessary procedures or expose you through radiation. If your abdominal pain persists for longer than 24-48 hours, worsens, you develop a fever, or you bleed, then we definitely should scan you.

The alternative is to just do the test today. Full disclosure: my physical exam may not be as sensitive as a CT scan. I'm not looking at what is going on inside of your body. What do you want us to do?"

Usually people want to wait it out and I document that conversation verbatim.

Never dismiss patient complaints. It's dangerous for your patients and for yourself. In our legal environment it is best to be aggressive.

I read an article not long ago that showed that physicians that did more tests were less likely to get sued. I'm not judging if this is bad or good but it is what it is.

 

[BoardingDoc]

1. All chest pain patients -- except traumatic -- get a troponin.

ALL of them? The 18 yr old guy with pleuritic pain that's PERC negative gets a trop? The 22F who has been coughing for a week and now complaining of CP? Etc etc etc? I'm not saying you shouldn't work these people up for badness, I'm just saying that I personally think it's entirely justifiable to send very low risk patients home with an EKG and return instructions. I completely agree that if they come back, they get the million dollar workup. That said, I sure as hell don't trop every single person who mentions the words "chest pain."

 

[RoyBasch]

If you ever read Mike Weinstock's "Bouncebacks" series on risk mitigation in medicine; he frequently emphasizes that the "bounceback" should put you on high alert and is an inherently higher risk patient. I find when teaching residents, this is a concept many seem very resistant to. Repeat visits seem to convince most residents that the patient is malingering or otherwise unconcerning. For me a relatively simple algorithm/heuristic I apply for repeat visits especially with complaints related to neuro/cardiovascular/GI/respiratory is:

First visit: labs, plainfilms, etc.
second visit: at this point advanced imaging (particularly CT) needs to be strongly considered and is nearly mandatory
third visit: almost always results in admission, and coordination of care for an expensive, painful, invasive test whether its cardiac catheterization, lumbar puncture, endoscopy, etc.

 

[GTP]

If you ever read Mike Weinstock's "Bouncebacks" series on risk mitigation in medicine; he frequently emphasizes that the "bounceback" should put you on high alert and is an inherently higher risk patient. I find when teaching residents, this is a concept many seem very resistant to. Repeat visits seem to convince most residents that the patient is malingering or otherwise unconcerning. For me a relatively simple algorithm/heuristic I apply for repeat visits especially with complaints related to neuro/cardiovascular/GI/respiratory is:

First visit: labs, plainfilms, etc.
second visit: at this point advanced imaging (particularly CT) needs to be strongly considered and is nearly mandatory
third visit: almost always results in admission, and coordination of care for an expensive, painful, invasive test whether its cardiac catheterization, lumbar puncture, endoscopy, etc.

This is how I've been taught too and I'm glad I practice this way. I sleep better at night. The exception is of course the people who have had abdominal pain since 1997 and get monthly negative cts.

 

[WilcoWorld]

Agree with above, dissections can also present like this. I think it's tough sometimes to see these cases when you're in them because there are all sorts of cues that don't make it into the chart. Due to outcome and description of case we're all picturing a citizen that's genuinely concerned about what's going on and is participating in the history and physical to the best of their abilities. That's certainly the picture the plaintiff's attorney is going to paint. But what if the patient is on their phone while you're trying to get history, called the ambulance because they didn't want to wait in the lobby, and decided it was cheaper to come to the ED then buy a bottle of Comfort Liquid at Walgreens? Now all you're instincts are screaming "F%# this guy", and you start turning into a PCP trying to encourage medication compliance and appropriate venues of care instead of being an ED doc.

That's why cognitive forcing strategies like Southerndoc pointed out are important. The HEART score doesn't care if your patient has 3x upper limit of normal squirrel titers. PERC doesn't factor in a patient's Axis I or II disorders. You can certainly pull in data from other sources (- CT abd/pel yesterday means it's highly unlikely pt just developed a ruptured AAA, etc) but you're never excused from the thought process of ruling out the emergent/life threatening causes of the patient's presentation during THIS visit.

This is how I've been taught too and I'm glad I practice this way. I sleep better at night. The exception is of course the people who have had abdominal pain since 1997 and get monthly negative cts.

Yep. As per Arcan's post, it's OK to incorporate the data from earlier visits (cath results, CT scans, culture results). In fact - that's good medicine. The mistakes are made when we assume that the earlier visits' impressions apply to today's visit.

 

[PlutoBoy]

This is how I've been taught too and I'm glad I practice this way. I sleep better at night. The exception is of course the people who have had abdominal pain since 1997 and get monthly negative cts.

Well if that's the case then another CT can't hurt

 

[southerndoc]

ALL of them? The 18 yr old guy with pleuritic pain that's PERC negative gets a trop? The 22F who has been coughing for a week and now complaining of CP? Etc etc etc? I'm not saying you shouldn't work these people up for badness, I'm just saying that I personally think it's entirely justifiable to send very low risk patients home with an EKG and return instructions. I completely agree that if they come back, they get the million dollar workup. That said, I sure as hell don't trop every single person who mentions the words "chest pain."

If I'm drawing labs, it's quick and easy to get an i-STAT troponin. Takes less than 10 minutes and the cartridge costs $17 each with our discount. It's a non-invasive test where I have seen numerous physicians squirm (one even becoming diaphoretic) attempting to defend why they didn't order a troponin on someone with "atypical" chest pain who ultimately had a bad outcome.

Patients will often downplay their symptoms and omit significant details -- either intentionally or by thinking it is irrelevant. A well-seasoned litigator, however, will make their omissions your fault because you were incompetent by not obtaining a thorough history. "Doctor, you had a wanton disregard for your patient's safety and were grossly negligent by not checking a test that cost less than $25. Doctor, how much did you bill my client's family for the grossly negligent care you provided to my client?" (This from an actual case for a 33-year-old male with a "non-specific EKG," non-productive cough, pleuritic chest pain; negative D-dimer, negative EKG, negative CXR, no other labs. Discharged home with his 100% occluded LAD to only die 93 minutes after being discharged from the ER. Nobody enquired about his family history. Father died of an MI at 45 and his brother had 2 stents at the age of 35. His "negative EKG" as documented by the ER doc had some flattening of T waves in inferolateral leads (likely start of becoming inverted) and beginning of hyperacute T waves in anterior leads.)

Perhaps I'm jaded by seeing so many cases where I can put myself in their shoes and see it not being remotely serious based on their presentation and the patient had a very unfortunate and untimely bad outcome. It does change your practice of medicine.

 

[BAM!]

Young ED Heart Attack

When things like this happen...and you are seeing 2-3 patients per hour, sending folks home whom you just barely met

This article rubs me the wrong way. A retired internist playing monday morning quarterback shouldn't be the one to berate an Emergency Doc.

 

[southerndoc]

This article rubs me the wrong way. A retired internist playing monday morning quarterback shouldn't be the one to berate an Emergency Doc.

She's well qualified to comment on this. She is not commenting as an internist but as her position as a risk management consultant for CRICO (Controlled Risk Insurance Company, a Harvard-based insurance that covers their hospitals and clinics). Most malpractice carriers employ or contract with physicians, and nearly all of them are FP's or internists.

On another note, there are quite a few states that do not require or make it customary to have someone within your own specialty sign off as your expert witness to file a malpractice suit. Nothing like a neurosurgeon to critique your work as an emergency physician.

 

[alpinism]

This thread is a perfect example of why emergency medicine is doomed in the United States.

The lessons learned from this case should be to ask about drug history and broaden your workup for return visits not to get a troponin for every chest pain.

 

[BoardingDoc]

If I'm drawing labs,

Right. I'm not drawing labs on these people. So you draw blood on every single person who comes through the ED mentioning chest pain? I'm not saying you're wrong if you do, I'm saying that I don't do that and I think it's entirely defensible.

 

[deleted859535]

This thread is a perfect example of why emergency medicine is doomed in the United States.

The lessons learned from this case should be to ask about drug history and broaden your workup for return visits not to get a troponin for every chest pain.

I also am quick to get troponins in chest pain. Seventeen year old with clearly MSK chest wall discomfort, okay, fine, if I don't think it's actually a myocarditis. But 25 year old with CP? 30 year old with CP? Sure. Labs include a troponin in a majority. Dimer if appropriate.

I'm not here to be a 100% good steward and warm the hearts of the hospital administration because I order the absolute minimum. I'm here to look for bad things and, crucially, not get sued over atypical badness (or typical badness). So I err towards the thorough/conservative side and document MDMs accordingly. I'm not ridiculous. I don't MRI and CTA everyone. But I find it better to get more reassurance, not less, if that inside voice nags me enough to do it.

When someone can promise me that the medicolegal climate in this country will improve, maybe it'll be different. And that's coming from a middle-of-the-road state for such things.

 

[PlutoBoy]

This article rubs me the wrong way. A retired internist playing monday morning quarterback shouldn't be the one to berate an Emergency Doc.

ACS is well within the scope of practice of an internist.

 

[PlutoBoy]

I also am quick to get troponins in chest pain. Seventeen year old with clearly MSK chest wall discomfort, okay, fine, if I don't think it's actually a myocarditis. But 25 year old with CP? 30 year old with CP? Sure. Labs include a troponin in a majority. Dimer if appropriate.

I'm not here to be a 100% good steward and warm the hearts of the hospital administration because I order the absolute minimum. I'm here to look for bad things and, crucially, not get sued over atypical badness (or typical badness). So I err towards the thorough/conservative side and document MDMs accordingly. I'm not ridiculous. I don't MRI and CTA everyone. But I find it better to get more reassurance, not less, if that inside voice nags me enough to do it.

When someone can promise me that the medicolegal climate in this country will improve, maybe it'll be different. And that's coming from a middle-of-the-road state for such things.

Exactly. Hindsight is 20/20. I always wonder how a case would play out in front of a jury if things turn sour while also considering EBM.

The fact of the matter is this: ACS is about the easiest thing to rule out in EM/IM. Just do the damn test, cover your ass, and reassure the patient.

Medicolegally, the outcome would have been different if the patient was admitted, got serial troponins, +/-cardiology consult , +/- stress test/cath.

I really can't grasp the rationale of a doctor that does not do this when the patient presented for a THIRD time.

All young people with CP should be asked about drug use. I also used to drug screen them.

 

[gamerEMdoc]

Right. I'm not drawing labs on these people. So you draw blood on every single person who comes through the ED mentioning chest pain? I'm not saying you're wrong if you do, I'm saying that I don't do that and I think it's entirely defensible.

I think you are right in the correct setting. His first visit, I think it is entirely defensible not to get a troponin. However, on the second visit, they said the ED doc noted that there were EKG St segment changes compared to the EKG on the first visit. I think if you have dynamic EKG changes, its really hard to argue against getting a troponin. After all, what’s the point of getting EKG if you aren’t going to do anything if it’s abnormal?

 

[Perrotfish]

ALL of them? The 18 yr old guy with pleuritic pain that's PERC negative gets a trop? The 22F who has been coughing for a week and now complaining of CP? Etc etc etc? I'm not saying you shouldn't work these people up for badness, I'm just saying that I personally think it's entirely justifiable to send very low risk patients home with an EKG and return instructions. I completely agree that if they come back, they get the million dollar workup. That said, I sure as hell don't trop every single person who mentions the words "chest pain."

I wonder if the standard of care is changing just because the test is becoming cheaper and easier to perform.

I've noticed that there has been a similar shift with meningitis patients. Before viral PCRs became common the clear standard of care was to send someone home if their spinal fluid and presentation was typical of viral meningitis, even knowing that a very small percentage (<2%) would have an early bacterial meningitis that just happened to have a normal glucose and an only a slightly elevated white count. Now that viral PCRs are more common, it seems like the standard of care is shifting towards treating everyone with an even slightly elevated white count on their spinal fluid with abx pending a PCR confirming a viral meningitis. A cheap iStat troponin might have the same effect on shifting the standard of care in chest pain patients.

 

[Perrotfish]

If you ever read Mike Weinstock's "Bouncebacks" series on risk mitigation in medicine; he frequently emphasizes that the "bounceback" should put you on high alert and is an inherently higher risk patient.

My favorite Weinstock quote: "The third time the pizza guy comes to my house I admit him".

 

[littlejuan]

If I'm drawing labs, it's quick and easy to get an i-STAT troponin. Takes less than 10 minutes and the cartridge costs $17 each with our discount. It's a non-invasive test where I have seen numerous physicians squirm (one even becoming diaphoretic) attempting to defend why they didn't order a troponin on someone with "atypical" chest pain who ultimately had a bad outcome.

Patients will often downplay their symptoms and omit significant details -- either intentionally or by thinking it is irrelevant. A well-seasoned litigator, however, will make their omissions your fault because you were incompetent by not obtaining a thorough history. "Doctor, you had a wanton disregard for your patient's safety and were grossly negligent by not checking a test that cost less than $25. Doctor, how much did you bill my client's family for the grossly negligent care you provided to my client?" (This from an actual case for a 33-year-old male with a "non-specific EKG," non-productive cough, pleuritic chest pain; negative D-dimer, negative EKG, negative CXR, no other labs. Discharged home with his 100% occluded LAD to only die 93 minutes after being discharged from the ER. Nobody enquired about his family history. Father died of an MI at 45 and his brother had 2 stents at the age of 35. His "negative EKG" as documented by the ER doc had some flattening of T waves in inferolateral leads (likely start of becoming inverted) and beginning of hyperacute T waves in anterior leads.)

Perhaps I'm jaded by seeing so many cases where I can put myself in their shoes and see it not being remotely serious based on their presentation and the patient had a very unfortunate and untimely bad outcome. It does change your practice of medicine.

I wish I had istat trops at my place. They take anywhere from 30-60 min to get back. It’s terrible. That being said. I still agree with you. I caught an NSTEMI in a young (30ish) healthy, fit female with history of GERD complaining of chest pain radiating up that didn’t quite feel like her previous GERD symptoms. The trop was crazy high too. No EKG changes. My colleagues working with me said they probably wouldn’t have even gotten the trop.

I’ve resigned to the fact that I am probably pretty conservative with my management, I may not be the fastest guy but I am not the slowest either and at least I sleep well at night.

 

[southerndoc]

Right. I'm not drawing labs on these people. So you draw blood on every single person who comes through the ED mentioning chest pain? I'm not saying you're wrong if you do, I'm saying that I don't do that and I think it's entirely defensible.

Not entirely defensible...

 

[deleted859535]

I wish I had istat trops at my place. They take anywhere from 30-60 min to get back. It’s terrible. That being said. I still agree with you. I caught an NSTEMI in a young (30ish) healthy, fit female with history of GERD complaining of chest pain radiating up that didn’t quite feel like her previous GERD symptoms. The trop was crazy high too. No EKG changes. My colleagues working with me said they probably wouldn’t have even gotten the trop.

I’ve resigned to the fact that I am probably pretty conservative with my management, I may not be the fastest guy but I am not the slowest either and at least I sleep well at night.

That "sleep well at night" part being key. I and my patients live with the consequences of what I do and don't do -- nobody else.

 

[Angry Birds]

I wish I had istat trops at my place. They take anywhere from 30-60 min to get back. It’s terrible. That being said. I still agree with you. I caught an NSTEMI in a young (30ish) healthy, fit female with history of GERD complaining of chest pain radiating up that didn’t quite feel like her previous GERD symptoms. The trop was crazy high too. No EKG changes. My colleagues working with me said they probably wouldn’t have even gotten the trop.

I’ve resigned to the fact that I am probably pretty conservative with my management, I may not be the fastest guy but I am not the slowest either and at least I sleep well at night.

I think i-stats are overrated. I had them back in residency, and it was not an uncommon experience that the regular labs were back before the i-stats.

This of course depends on a number of factors, but I think if you order stuff when patients come through the door (after reading the triage note instead of actually talking to them), and you time how you see patients and in what order, then waiting for the labs usually isn't an issue.

 

[BAM!]

She's well qualified to comment on this. She is not commenting as an internist but as her position as a risk management consultant for CRICO (Controlled Risk Insurance Company, a Harvard-based insurance that covers their hospitals and clinics). Most malpractice carriers employ or contract with physicians, and nearly all of them are FP's or internists.

On another note, there are quite a few states that do not require or make it customary to have someone within your own specialty sign off as your expert witness to file a malpractice suit. Nothing like a neurosurgeon to critique your work as an emergency physician.

I guess that's what I was referring to. I must have gotten used to living in a state where only Emergency Medicine specialists are able act as expert witnesses.

 

[RoyBasch]

The lessons learned from this case should be to ask about drug history and broaden your workup for return visits not to get a troponin for every chest pain.

I suppose my post above is evidence to the fact that I am quite conservative in my practice; however, I do consider a troponin mandatory for the workup of every atraumatic chest pain. I have diagnosed myocarditis, myopericarditis, and other cardiac abnormalities because an abnormal troponin was detected. I do draw troponins even in children/teenagers with atraumatic chest pain.

I think you are right in the correct setting. His first visit, I think it is entirely defensible not to get a troponin. However, on the second visit, they said the ED doc noted that there were EKG St segment changes compared to the EKG on the first visit. I think if you have dynamic EKG changes, its really hard to argue against getting a troponin. After all, what’s the point of getting EKG if you aren’t going to do anything if it’s abnormal?

A colleague of mind who is a medical malpractice guru explained that from a medicolegal perspective, it you get an EKG, if is very hard to defend not getting a troponin. It is actually MORE defensible to get neither (e.g. young patient, suspected MSK or traumatic chest wall pain) because getting neither an EKG nor troponin implies there was NO reason to suspect cardiac etiology whatsoever. Your chart shows an internal consistency with this approach. However, once you get the EKG, you are now demonstrating a somewhat haphazard workup and inconsistent (to an attorney) train of thought. "You were concerned enough about a cardiac problem that you got an EKG, then why didn't you get a troponin?"

 

[southerndoc]

A colleague of mind who is a medical malpractice guru explained that from a medicolegal perspective, it you get an EKG, if is very hard to defend not getting a troponin. It is actually MORE defensible to get neither (e.g. young patient, suspected MSK or traumatic chest wall pain) because getting neither an EKG nor troponin implies there was NO reason to suspect cardiac etiology whatsoever. Your chart shows an internal consistency with this approach. However, once you get the EKG, you are now demonstrating a somewhat haphazard workup and inconsistent (to an attorney) train of thought. "You were concerned enough about a cardiac problem that you got an EKG, then why didn't you get a troponin?"

This is generally how the malpractice world views it. Unless your EKG is to determine an arrhythmia or something else, an EKG in a chest pain patient equates to a troponin in nearly every case. Failure to do so it at your own peril. We're not talking about cathing someone here. It's a freaking troponin. No radiation exposure, etc.

During residency I remember every <30 year old with chest pain got an EKG and a chest x-ray. I now realize just how risky that was.

 

[GeneralVeers]

This is generally how the malpractice world views it. Unless your EKG is to determine an arrhythmia or something else, an EKG in a chest pain patient equates to a troponin in nearly every case. Failure to do so it at your own peril. We're not talking about cathing someone here. It's a freaking troponin. No radiation exposure, etc.

During residency I remember every <30 year old with chest pain got an EKG and a chest x-ray. I now realize just how risky that was.

The problem with this logic, is that the EKGs are ordered at one hospital I work at for every epigastric abdominal pain, asthma attack, shoulder pain, arm pain, neck pain, "dizziness" or high blood pressure. There is no rational reason to get a troponin on all these people.

 

[RoyBasch]

The problem with this logic, is that the EKGs are ordered at one hospital I work at for every epigastric abdominal pain, asthma attack, shoulder pain, arm pain, neck pain, "dizziness" or high blood pressure. There is no rational reason to get a troponin on all these people.

Sure, but we're not talking about EKGs to workup syncope, we're talking about chest pain and the EKG is being used to assist in the diagnosis or rule out of an acute coronary syndrome. I would argue that if some of those other vague complaints such as neck pain, epigastric pain, are being worked up as atypical anginal symptoms, then the troponin should be ordered in addition to the EKG.

I agree though if some of these EKGs are being ordered by the triage RN in accordance with a protocol and after your physician evaluation you deem there to be no suspicion of an acute coronary syndrome and you don't want to get at troponin, you're kind of in a bind.

 

[Brigade4Radiant]

So know we are getting more lab tests just because of the medical legal climate. Also please don't think that one troponin can rule out ACS it is the trend of troponin that you need to be on the look out for. Also it's just cheaper to do a history of drugs (smoking and cocaine or family history)

Then again an young person can have myocarditis and just have a bad outcome regardless of what you do.

 

[lucid_leprechaun]

So know we are getting more lab tests just because of the medical legal climate. Also please don't think that one troponin can rule out ACS it is the trend of troponin that you need to be on the look out for. Also it's just cheaper to do a history of drugs (smoking and cocaine or family history)

Then again an young person can have myocarditis and just have a bad outcome regardless of what you do.

So a single Troponin is only 60% per Rosen’s, and initial ECG negative in 50%of STEMI. Obviously if concerned for ACS can pursue, however one is typically reasonable in low risk cases. There are of course always atypical cases. If the story is the least concerning, trend em. Long stay vs. a dead 21 y/o w/ congenital risk factors.

 

[Interpolfanclub]

Wow, I thought I was conservative until I read this thread. I almost never get a troponin on people less than 30 with chest pain unless the story is very concerning. EKG and CXR sure. I get a dimer 10x more than I get a trop on a young person with chest pain.

 

[shoal007]

Wow, I thought I was conservative until I read this thread. I almost never get a troponin on people less than 30 with chest pain unless the story is very concerning. EKG and CXR sure. I get a dimer 10x more than I get a trop on a young person with chest pain.

I'm with you. I think it's kinda crazy to get a trop on every healthy less than 30 yr old.

Thankfully I practice in a reasonable medmal climate.

Sent from my Pixel 2 using Tapatalk

 

[TooMuchResearch]

Sure, but we're not talking about EKGs to workup syncope, we're talking about chest pain and the EKG is being used to assist in the diagnosis or rule out of an acute coronary syndrome. I would argue that if some of those other vague complaints such as neck pain, epigastric pain, are being worked up as atypical anginal symptoms, then the troponin should be ordered in addition to the EKG.

I agree though if some of these EKGs are being ordered by the triage RN in accordance with a protocol and after your physician evaluation you deem there to be no suspicion of an acute coronary syndrome and you don't want to get at troponin, you're kind of in a bind.

18 yo, cough x1 day with URI symptoms? These sometimes get a ECG before they come back. Sometimes a troponin is resulted on this patient before I see them. "The good news is your cold isn't a heart attack. The bad news is this visit is going to veeeeeeeerrrrrrrrrry expensive. Do you want some ibuprofen before you go?"

 

[ballinsncbirth]

Its interesting seeing the debate regarding getting troponins, vs not depending on the "description " of the pain . Heres the thing. I do not even care about how anyone describes their pain anymore. It really makes no difference in my decision making.Maybe its because I am 5 years out of residency, or that I have seen to many missed MI 's. .... but the reality is we have to understand several things when approaching these patients

1. WE ARE NOT PRACTICING MEDICINE ANYMORE
The reality is we stopped practicing evidence based medicine long ago , in America we are practicing "DO NOT MISS ANYTHING EVER " medicine. I don't care if you see 1000 low risk chest pain patients that end up being nothing, and 1 dies. guess what your were a 99.9 % "good"doctor , here's your lawsuit. You will not be able to use the defense of how "unlikely " this event is based on presentation etc ever. no one will care. You will be dragged through the coals. Once you realize this then you can proceed to step 2 . So take your HEART SCOR E, PORT SCORE , TIMI SCORE and shove them up your **s, as this will not save you .

2.HISTORY DOES NOT RULE OUT MI ...
I don't care whether your pain is "burning" "pressure like" sharp " etc.. IT DOES NOT MATTER , as soon as there is chest pain on the chart your job is to RULE OUT emergency . Description , character, location, duration, is will not rule out MI . Also if you have spent anytime talking to the populace you will realize PEOPLE CANNOT ACURETLY DESCRIBE ANYTHING . This along with the subjective variation of how people experience pain , etc . mean that there is literally nothing anyone can tell me that will matter as to ruling out emergency . History is useful in RULING IN chest pathology but you cant count on it for ruling out MI , PE or any of the serious chest pathologies.

3. YOU WILL NEVER REGRET ORDERING A TROPONIN
I never understood why some people will literally argue with themselves on whether to get a troponin or not. Even to the point where people will order other labs on a chest pain patient and purposefully not get a troponin . We were taught this in residence (ie troponin means you have to get another one, admit, rule out etc. ) That is bull. I guarantee you no one was ever sued for NOT getting a troponin. even a negative troponin x1 will help in that the jury and lawyers are NOT Medically people. they will not understand ruling out , trop length , etc etc. What they will understand is that there was a test that can tell whether you are having a heart attack and you didn't order it . THe reality is most of your chest pains will be getting ekg, chest xray . they will be in the ER for several hours(at least in my shop) in that time you could have easily gotten a troponin, helped your liability AND increased the billing of the chart. Also troponin doesnt just rule out MI , but most importantly is the difference between a mild pericarditis, and dangerous myocarditis, heart strain in PE , pericardial effusion, all can present with elevated troponin in an otherwise young person.


4. DONT DIAGNOSE NONEMERGENCIES
I don't care if a person has, gerd, pleurisy , gastritis, . I will not be diagnosing them with any of these. I may tell the patient, and treat them . but on the chart I will only diagnose CHEST PAIN. etc. Noone will ever come back and tell you how great a doctor you were for diagnosing GERD costochrondritis. The only time you will ever here about these type of patients again are when you missed a diagnosis, chart review or something went wrong. Having the diagnosis of GERD , after the patient died of a heart attack will lend you in a lot of trouble. Normal people and lawyers dont understand that your there are not test to identify GERD , or pleurisy and that just you best guess on what the person had. . Patients also take this a then decide they dont need to come back when things get worse or change because "the doctor told me what I had"

5. Repeat visits for the same thing get upgrade workups.
yes I know its annoying, but stop thinking about it , stop trying to rationalize it . If a person comes in multiple times, whatever workup they got last time, increase it. 1st time chest pain ek, 2nd tme labs, 3rd time ct chest and admission. Same thing with abd pain . I dont care whether the patient belly is completely perfect, they come in a second time and weren't scanned the first , they surely will get a scan the second time. Not only will it make the patient think you "care" which is a big factor in lawsuits. but it shows the jury and lawyers that you took their complaints seriously and thought of other things.

So just understand that when your are making your decision .I am not saying get a trop in every patient, yet understand you will be judged on it if you miss something.
You will miss an MI at some point. You will only miss an MI in a low risk patient. You will likely be sued at some point in your career. Think about what you would like your chart to say when it happens, and how you would like it do defend you.


PS sorry for the misspelling and grammar errors, I am lazy, on shift, and don't really care to proofread

 

[Brigade4Radiant]

If you have a bad outcome and send someone home and they are young you'll get sued and maybe lose even if you have a negative troponin because the lawyers who know that you need a troponin will critize you for not admitting the patient or doing a second troponin. A jury of your less educated peers will decide on a verdict and mostly it will be out of sympathy. If you order a troponin on every young patient you would be best served by practicing in a state with better mal-practice protection.

You get sued due to bad outcomes

Keep in mind women can present with nausea, fatigue, diarrhea and hip pain and it still be an MI. Diabetics can have a painless MI. If you are ordering troponins for every chest pain you might as well order a troponin every time you get blood work.

But seriously though order the troponin I mean how often do we get xrays of extremites when we know damn well it's not broken.

 

[deleted859535]

I love it when residents argue with me about getting a troponin in cases like those above. I'm not terribly far from residency myself, but nothing makes you think more about how you practice than being the one ultimately responsible for how you practice.

Yep, we all will miss things. But the goal is to make those misses as few as possible, as far apart as possible, and with as atypical of situations as possible.

 

[GonnaBeADoc2222]

So as a new attending, this discussion has been very very interesting. I definitely label myself as conservative (pretty sure I have a high CT and admit rate) but was definitely not getting troponins (ALWAYS get EKG) on young people (<30) with chest pain without a multitude of risk factors or an abnormal EKG. I'm wondering how those who are only getting 1 troponin on chest pain < 30 years old are documenting this in the chart. I feel like everything I've been exposed to during training, etc. has taught me that unless the pain has been for > 4 hrs, you need a 2nd troponin.

 

[PlutoBoy]

Its interesting seeing the debate regarding getting troponins, vs not depending on the "description " of the pain . Heres the thing. I do not even care about how anyone describes their pain anymore. It really makes no difference in my decision making.Maybe its because I am 5 years out of residency, or that I have seen to many missed MI 's. .... but the reality is we have to understand several things when approaching these patients

1. WE ARE NOT PRACTICING MEDICINE ANYMORE
The reality is we stopped practicing evidence based medicine long ago , in America we are practicing "DO NOT MISS ANYTHING EVER " medicine. I don't care if you see 1000 low risk chest pain patients that end up being nothing, and 1 dies. guess what your were a 99.9 % "good"doctor , here's your lawsuit. You will not be able to use the defense of how "unlikely " this event is based on presentation etc ever. no one will care. You will be dragged through the coals. Once you realize this then you can proceed to step 2 . So take your HEART SCOR E, PORT SCORE , TIMI SCORE and shove them up your **s, as this will not save you .

2.HISTORY DOES NOT RULE OUT MI ...
I don't care whether your pain is "burning" "pressure like" sharp " etc.. IT DOES NOT MATTER , as soon as there is chest pain on the chart your job is to RULE OUT emergency . Description , character, location, duration, is will not rule out MI . Also if you have spent anytime talking to the populace you will realize PEOPLE CANNOT ACURETLY DESCRIBE ANYTHING . This along with the subjective variation of how people experience pain , etc . mean that there is literally nothing anyone can tell me that will matter as to ruling out emergency . History is useful in RULING IN chest pathology but you cant count on it for ruling out MI , PE or any of the serious chest pathologies.

3. YOU WILL NEVER REGRET ORDERING A TROPONIN
I never understood why some people will literally argue with themselves on whether to get a troponin or not. Even to the point where people will order other labs on a chest pain patient and purposefully not get a troponin . We were taught this in residence (ie troponin means you have to get another one, admit, rule out etc. ) That is bull. I guarantee you no one was ever sued for NOT getting a troponin. even a negative troponin x1 will help in that the jury and lawyers are NOT Medically people. they will not understand ruling out , trop length , etc etc. What they will understand is that there was a test that can tell whether you are having a heart attack and you didn't order it . THe reality is most of your chest pains will be getting ekg, chest xray . they will be in the ER for several hours(at least in my shop) in that time you could have easily gotten a troponin, helped your liability AND increased the billing of the chart. Also troponin doesnt just rule out MI , but most importantly is the difference between a mild pericarditis, and dangerous myocarditis, heart strain in PE , pericardial effusion, all can present with elevated troponin in an otherwise young person.


4. DONT DIAGNOSE NONEMERGENCIES
I don't care if a person has, gerd, pleurisy , gastritis, . I will not be diagnosing them with any of these. I may tell the patient, and treat them . but on the chart I will only diagnose CHEST PAIN. etc. Noone will ever come back and tell you how great a doctor you were for diagnosing GERD costochrondritis. The only time you will ever here about these type of patients again are when you missed a diagnosis, chart review or something went wrong. Having the diagnosis of GERD , after the patient died of a heart attack will lend you in a lot of trouble. Normal people and lawyers dont understand that your there are not test to identify GERD , or pleurisy and that just you best guess on what the person had. . Patients also take this a then decide they dont need to come back when things get worse or change because "the doctor told me what I had"

5. Repeat visits for the same thing get upgrade workups.
yes I know its annoying, but stop thinking about it , stop trying to rationalize it . If a person comes in multiple times, whatever workup they got last time, increase it. 1st time chest pain ek, 2nd tme labs, 3rd time ct chest and admission. Same thing with abd pain . I dont care whether the patient belly is completely perfect, they come in a second time and weren't scanned the first , they surely will get a scan the second time. Not only will it make the patient think you "care" which is a big factor in lawsuits. but it shows the jury and lawyers that you took their complaints seriously and thought of other things.

So just understand that when your are making your decision .I am not saying get a trop in every patient, yet understand you will be judged on it if you miss something.
You will miss an MI at some point. You will only miss an MI in a low risk patient. You will likely be sued at some point in your career. Think about what you would like your chart to say when it happens, and how you would like it do defend you.


PS sorry for the misspelling and grammar errors, I am lazy, on shift, and don't really care to proofread

Exactly. I just don't understand why people resist ordering the test. I mean, ultimately I don't mind because it's not my ass on the line but there's very little to lose by doing an appropriate work up. A CC of CP and you didn't rule out MI? You can say whatever you want to a jury but no one will understand why you didn't do it. All they would see is that you are an emergency doctor that didn't diagnose an emergency. That's it.

I loved admitting chest pain. These are the easiest admissions ever and they often did have ACS.

I am very conservative as well. I want my charts to be tight from a medicolegal perspective. For example, I always get an ECG in middle aged and older adults if the heart rate is >105-110 even if I know from a physical exam that the pulse is regular. If that guy has a stroke that evening and someone notices that my patient was tachycardic someone may question if my patient was actually in atrial fibrillation.

How long does it take me to do an ECG and interpret it? Most of the time about five minutes. It is an appropriate test for tachycardic patients and no one can ever blame me for working things up with a simple test but they can definitely blame you if you don't.

Furthermore, sometimes you are surprised and you actually find pathology that needs immediate treatment.

 

[PlutoBoy]

I guess that another thing that helps is having colleagues that work with you as a team. You want a hospitalist that understands the position you are in and our medicolegal environment, not someone that is there to block your admissions.

I fostered a very good relationship with my ER colleagues. If they wanted to admit CP I just admitted every single one of them without question. It's just easy to do and everybody sleeps well. OTOH, they understood when I told them that the septic 25 year old IVDA with no clear source was better served at another hospital because he probably has endocarditis and we had no ID, CC, or cardiac surg backup. It worked.

 

[joeDO2]

A lot of fixation on taking a good history and the importance of asking about cocaine use in young chest pain. I find it almost useless. My personal experience thus far is that asking about cocaine and not performing a UDS is like the 17 year old with belly pain telling me that there is no possibility of pregnancy and skipping the HCG. Trust, but verify.

 

[WilcoWorld]

A lot of fixation on taking a good history and the importance of asking about cocaine use in young chest pain. I find it almost useless. My personal experience thus far is that asking about cocaine and not performing a UDS is like the 17 year old with belly pain telling me that there is no possibility of pregnancy and skipping the HCG. Trust, but verify.

I see your point, but I don't fully agree. If the MD had asked about cocaine, documented that the patient denied it, and then the patient tested (+) it might have significantly helped the defense.*

Also, cocaine metabolites only remain positive for a day or so, whereas the accelerated atherosclerosis lasts a lifetime.

*disclaimer - I am not a lawyer

 

[Angry Birds]

I see your point, but I don't fully agree. If the MD had asked about cocaine, documented that the patient denied it, and then the patient tested (+) it might have significantly helped the defense.*

*disclaimer - I am not a lawyer

In fact, even the patient doing cocaine helps the defense, as the plaintiff is looked on less favorably.

 

[RoyBasch]

Keep in mind women can present with nausea, fatigue, diarrhea and hip pain and it still be an MI. Diabetics can have a painless MI. If you are ordering troponins for every chest pain you might as well order a troponin every time you get blood work.

Yes but once again we are not talking about "silent MIs" or atypical presentations; we are talking about people with chest pain who have an MI. That is considered typical.

I think in the eyes of the laity, "chest pain" + "emergency" = "heart attack." And you are the emergency doctor, and they have chest pain, so your job is to investigate for "heart attack."

You can certainly quibble that should you miss an MI in chest pain because EKGs aren't 100% sensitive and how many troponins you need to get, how far apart, at what time intervals, and who should be admitted, and who should have gotten a cardiology consult, you might still find yourself in a situation that is questionable. However, I think at least showing some basic (and negative) investigation into an acute coronary syndrome etiology of the chest pain with an EKG and troponin at least bolsters your defense.

I think if you miss an MI in a 25 year old woman with no risk factors whose only complaint was nausea, that would be considered a fairly forgivable miss because the situation is so "atypical." Even the best hired gun expert witnesses would be on the back foot to convince anyone otherwise (not to say it is impossible and they never could). But I'd say their job is a heck of a lot harder than if the patient has chest pain.

I'm wondering how those who are only getting 1 troponin on chest pain < 30 years old are documenting this in the chart. I feel like everything I've been exposed to during training, etc. has taught me that unless the pain has been for > 4 hrs, you need a 2nd troponin.

The number of troponins in low risk patients for me (and I think many others) has to do with the time onset of the pain. If the patient has had pain for >3-4 hours I think one troponin is enough as the sensitivity of the troponin assay in the setting of a real MI increases with time (troponin level peaks at like 8-10 hours or something). If the pain started immediately prior to arrival then a delta trop over 3-4 hours of obs in the ER (with a repeat EKG as well to look for dynamic changes) is probably necessary even if the patient is low risk. Another factor is kind of "stuttering angina" where the patient may have had intermittent pain over a longer time period but it got "much worse" say "1 hour ago," in that case--again--a delta trop 3-4 hours after the time of exacerbation of the pain is probably necessary.

For patients who are high risk, even >10 hours of constant pain still requires multiple troponins like 3-4 q6h. This is one of the main impetus for admission in "high risk" chest pain. At this point I trust non-invasive stress tests about as far as I can throw them, and I think the main utility (besides demonstrating to an attorney that the patient was taken "seriously" and to diffuse the liability to the internist and/or cardiologist) of a chest pain admit is serial troponins to detect an ACS.

 

[gamerEMdoc]

Re: one troponin.

It's about statistics and shared decision making. The HEART score is based off one troponin and the timing of that troponin isn't factored in. If you have a patient with a HEART score of 3 or less, that persons 30 day risk of MI/death is 2%. If you get a second troponin, that risk goes under 1%. Now obviously, this is only for low risk CP (defined by a HEART of 3 or less). People with an abnormal ECG, lots of risk factors, great story, etc... for those people the risk with one troponin, or even more, is greater and therefore generally an observation for provocative testing is recommended.

Here's what I do:

HEART > 3: Observation. If the patient doesn't want to, I explain the higher risk, develop an outpatient followup plan, make sure they take an aspirin a day, and tell them they can come back if they change their mind. And I have them leave AMA. Our hospital policy recommends obs for HEART > 3, so them leaving is technically always AMA based on hospital standard of care.
HEART < 3: I use shared decision making. I explain the risk (2%) based on the single trop. I explain that risk is lower (under 1%) with a second negative trop and recommend a 2 hr delta troponin, but if the patient is totally cool with 2% and doesn't want to stay (personally, I wouldn't stay), they go home with outpt followup at this point with good return precautions. If the patient stays for the delta trop and if there is any rise, they get observation. If unchanged and normal, they go home.