Missed MI

[Hamhock]

Advertisement - Members don't see this ad

I constantly hear about missed MI.

"We miss more than 2% of MIs!"

"We are struggling to get our miss rate (in reference to MI) below 3%."

Does anyone have any evidence to support this?

Can someone please direct me to some reading that will explain this "truism"?

...even better would be some reading that can describe the outcomes of these "missed MIs".

xaelia?

HH

 

[southerndoc]

I haven't conducted a formal study, but since we opened our ED observation unit, I've admitted 2 patients in their early 30's who had no risk factors who ended up with MI's. Had we not had our obs unit, I would've sent these patients home because cardiology doesn't like to fool with a 32 year old with no risk factors. One ended up STEMI'ing out a few hours later, and the other had an NSTEMI with a first troponin that was <0.01, a second that was 0.6, and a third that was 3.

 

[EctopicFetus]

I constantly hear about missed MI.

"We miss more than 2% of MIs!"

"We are struggling to get our miss rate (in reference to MI) below 3%."

Does anyone have any evidence to support this?

Can someone please direct me to some reading that will explain this "truism"?

...even better would be some reading that can describe the outcomes of these "missed MIs".

xaelia?

HH

I believe the number comes from this article. If I recall there were other presented in residency as well.

http://www.nejm.org/doi/full/10.1056/NEJM200004203421603

Among the 889 patients with acute myocardial infarction, 19 (2.1 percent) were mistakenly discharged from the emergency department (95 percent confidence interval, 1.1 to 3.1 percent); among the 966 patients with unstable angina, 22 (2.3 percent) were mistakenly discharged (95 percent confidence interval, 1.3 to 3.2 percent).

 

[Birdstrike]

.

 

[shoal]

I presented a paper on this subject at the regional SAEM meeting earlier this year. We reviewed 1121 ED visits with complaint of chest pain. followed up with chart review for 30 days to see if any patients were sent home who later came back with a "missed" cardiac event.

We had 2 cases total.

~ 0.2%. Although we have numerous hospitals in the area that the pt may have chosen to go to instead for follow-up or return visits. There are other articles in the literature that report the 2-5% number, but I would argue that the number is significantly less nowdays secondary to observation units and fear of missing a MI.

 

[EctopicFetus]

I presented a paper on this subject at the regional SAEM meeting earlier this year. We reviewed 1121 ED visits with complaint of chest pain. followed up with chart review for 30 days to see if any patients were sent home who later came back with a "missed" cardiac event.

We had 2 cases total.

~ 0.2%. Although we have numerous hospitals in the area that the pt may have chosen to go to instead for follow-up or return visits. There are other articles in the literature that report the 2-5% number, but I would argue that the number is significantly less nowdays secondary to observation units and fear of missing a MI.

What about those who 1) died 2) went to another hospital?

Hard to make a perfect study but those are just a few of the obvious problems in the study you mentioned.

 

[EM6771]

I presented a paper on this subject at the regional SAEM meeting earlier this year. We reviewed 1121 ED visits with complaint of chest pain. followed up with chart review for 30 days to see if any patients were sent home who later came back with a "missed" cardiac event.

We had 2 cases total.

~ 0.2%. Although we have numerous hospitals in the area that the pt may have chosen to go to instead for follow-up or return visits. There are other articles in the literature that report the 2-5% number, but I would argue that the number is significantly less nowdays secondary to observation units and fear of missing a MI.

Your data is pretty much what I would expect in this day of admit all CP for rule out. Where I think that the misses come from are the very atypical patients who present with nausea, vomiting, dental pain, etc. Simply reviewing pts who presented with chest pain would miss all the patients that were initially missed.

The data is pretty clear that even in the most technologically advanced centers we miss 2% of MIs. If you ever get the opportunity to listen to Amal Mattu speak on low risk chest pain he always cites this and it is a very informative talk.

 

[shoal]

What about those who 1) died 2) went to another hospital?

Hard to make a perfect study but those are just a few of the obvious problems in the study you mentioned.

most certainly there are issues with the study,and has a bias towards missing cases. Patients who died, and were in our system were included as a cardiac event.

The next poster mentions atypical chest pain, and I agree -- these are much more likely to be missed. One of the missed cases came back with jaw pain, and was discharged to only come back the following day in v fib. The pt's first ED visit in this dataset left AMA, with a long standing drug abuse hx and multiple stents in the past.

The flip side of this issue is pt presenting with chest pain are being admitted far too often. 12 providers in the study hospital admitted ~150 patients without a single event (MI, UA/NSTEMI, PCI, CAB, death) for all of their admissions.

 

[Aesculapius]

SMART EM has an excellent podcast about this subject: http://www.smartem.org/smartem.org/Blog/Entries/2011/4/5_Chest_Pain_Risk.html

 

[EctopicFetus]

most certainly there are issues with the study,and has a bias towards missing cases. Patients who died, and were in our system were included as a cardiac event.

The next poster mentions atypical chest pain, and I agree -- these are much more likely to be missed. One of the missed cases came back with jaw pain, and was discharged to only come back the following day in v fib. The pt's first ED visit in this dataset left AMA, with a long standing drug abuse hx and multiple stents in the past.

The flip side of this issue is pt presenting with chest pain are being admitted far too often. 12 providers in the study hospital admitted ~150 patients without a single event (MI, UA/NSTEMI, PCI, CAB, death) for all of their admissions.

Of those 150 patients you cite how many benefited from medical management? Had something else found to explain their chest pain. Pneumonia, PE etc.

Now, its not that I believe we DONT admit too many people. Why would I take on the risk of being sued? The atypical complaints that are MIs are whats hard. We all have stories but a few I love... 1) Headache 2) Nausea/Vomiting 3) Hiccups.

In the end the lawyers are why we do this. Its not hard to figure out that the dude with 3 stents clutching his chest and sweating needs an admission.

Perhaps that guy doesnt rule in and doesnt have an event. Would you say that dude doesnt need an admission?

 

[EctopicFetus]

BTW, re Amal Mattu, I have heard him talk a number of times. His take home message.... Admit em all. Unless something has changed this is what i take away from every one of his lectures.

 

[shoal]

Of those 150 patients you cite how many benefited from medical management? Had something else found to explain their chest pain. Pneumonia, PE etc.

Now, its not that I believe we DONT admit too many people. Why would I take on the risk of being sued? The atypical complaints that are MIs are whats hard. We all have stories but a few I love... 1) Headache 2) Nausea/Vomiting 3) Hiccups.

In the end the lawyers are why we do this. Its not hard to figure out that the dude with 3 stents clutching his chest and sweating needs an admission.

Perhaps that guy doesnt rule in and doesnt have an event. Would you say that dude doesnt need an admission?

we looked for PEs as well -- since these may present as chest pain. Although, workup was not completed for PE on a good % of the patients -- no PEs were found in this group of 150 patients who were admitted with chest pain. As for medical management, we did not look at that -- good point.

 

[xaelia]

There are several studies that give the same sort of 2-3% number of missed myocardial infarctions, though they arrive at it in different ways.

In addition to that NEJM article above, there are a couple other ones from the EM literature:
http://www.ncbi.nlm.nih.gov/pubmed/8442548
is the classic one from 1993, which looked at 5,000 patients presenting the ED with chest pain and followed them up for missed MIs. 1000 patients received a diagnosis of AMI, 20 of whom were sent home, 5 of which died. That's the first 2% number, and where the 25% of missed MIs die number first comes from.

http://www.ncbi.nlm.nih.gov/pubmed/17112926
is more recent, in 2006, looks at patients who were admitted with AMI - and then looked back to see if they'd been to an Emergency Department within 7 days with symptoms consistent with cardiac disease. They also come up with that number of 2.1% miss rate.

In reality, these numbers probably minimally underestimate the number of missed MIs in the study populations - the limitations of the diagnosis of AMI in follow-up.

If you look at all the studies of trying to apply the TIMI score to designate a low-risk population in the ED, you also see a 30-day MACE event rate of ~2% - although, I have problems with the combined MACE endpoint used in a lot of the cardiovascular studies.

So, how many are we really missing? I would probably say we're missing fewer - as a result of increased awareness of these historical numbers as well as medicolegal concerns leading us to rule out more people. But, it is precisely those atypical presentations that will simply always be atypical that lead us to discharge patients who go on to have an AMI. I don't think the numbers from 1993 are reliable indicators of how many patients have poor outcomes after a miss - considering a lot of their patients actually received the diagnosis of "ischemic heart disease" on discharge - but it's still clearly a bad thing.

I tend try to talk the low-yield patients out of admission in the interests of cost and resource savings, when feasible. I tell folks they're not zero-risk, but low-risk, and low-risk isn't zero, and, if possible, give them some numbers based on the literature. If they want to stay, they get obs; if they don't, I document the conversation and let them go.

 

[Venko]

If the root of your question comes from wanting to make sure you dont miss MI's unnecessarily while still being realistic with how many people you keep in the hospital or observation unit, check out this article from September in Annals:

EP Hess, RJ Brison, et al. Development of a Clinical Prediction Rule for 30 day Cardiac Events in Emergency Department Patients with Chest Pain and Possible Acute Coronary Syndrome. Annals of EM. September 2011

Its a great study that we've been doing here for multiple years and I think will revolutionize practice in the long run. Going to the point of admit them all, we are working hard to try and narrow down some patients who can safely be discharged.

Its the BEGINNING of a shift...

TL

 

[deleted6669]

since we opened our ED observation unit, I've admitted 2 patients in their early 30's who had no risk factors who ended up with MI's. Had we not had our obs unit, I would've sent these patients home because cardiology doesn't like to fool with a 32 year old with no risk factors. One ended up STEMI'ing out a few hours later, and the other had an NSTEMI with a first troponin that was <0.01, a second that was 0.6, and a third that was 3.

we have had similar experiences with our obs unit which we opened around 5 yrs ago. that place has saved our butts (actually our pts butts)on numerous occasions.
not only mi but also recurrent arrhythmias, tia progressing to cva, rebleeding after tonsilectomy, etc, etc
we really use it as a place to admit folks we know need to be in house and the specialists disagree...great to call them back a few hrs later with the "remember that guy you wanted me to send home" story about why they need to emergently take them to the cath lab, back to the o.r., etc

 

[tkim]

38 y/o M 12 hours of SSCP, pain-free on presentation. Top EKG made me worry - called interventionalist at tertiary who basically shit all over me. Faxed it to my in-house cards who called it 'early repol' and also shit all over me.

Forty minutes later, chest pain returns, second EKG.

Call back the interventionalist - "he's now infarcting. sending him to you now." "make sure you load him on plavix" "he was loaded on plavix ... forty minutes ago."

I then faxed the second EKG to the in-house cards office. A couple of times.

I called the in-house cards later on a different case, and he tells me that the interventionalist called and told him patient had a 95% occluded CFX and now owns a shiny new DES. Gee, thanks assholes, for keeping me in the loop.

 

[Rendar5]

wtf? how could a cardiologist call that early repol...the STE's aren't even in the prechordial leads.

 

[Old_Mil]

38 y/o M 12 hours of SSCP, pain-free on presentation. Top EKG made me worry - called interventionalist at tertiary who basically shit all over me. Faxed it to my in-house cards who called it 'early repol' and also shit all over me.

Forty minutes later, chest pain returns, second EKG.

Call back the interventionalist - "he's now infarcting. sending him to you now." "make sure you load him on plavix" "he was loaded on plavix ... forty minutes ago."

I then faxed the second EKG to the in-house cards office. A couple of times.

I called the in-house cards later on a different case, and he tells me that the interventionalist called and told him patient had a 95% occluded CFX and now owns a shiny new DES. Gee, thanks assholes, for keeping me in the loop.

The unwillingness of cardiology to take people to the cath lab after hours is remarkable. I've had a patient code because they wouldn't.

 

[tkim]

The unwillingness of cardiology to take people to the cath lab after hours is remarkable. I've had a patient code because they wouldn't.

This was during weekday business hours. There is no explanation besides laziness and a distrust of the ER physician's assessment.

 

[xaelia]

The unwillingness of cardiology to take people to the cath lab after hours is remarkable. I've had a patient code because they wouldn't.

Lytics.

I have it ready for pre-hospital STEMI notifications off-hours in the event the cath lab is slow to respond. If the patient looks bad/EKG shows a lot of myocardial involvement, I give it. I've seen them code waiting for the cath lab, too.

 

[Tiger26]

Great call--looks like greater ST elevation in lead III compared to II in the initial one, yeah?

 

[EM2BE]

How often do you do EKGs in those you are concerned about? As in pt looks horrible and the first EKG is not a STEMI? I found I will do q15 min EKGs for about an hour after the first if I'm concerned. The nurses / techs I work with do not complain about this - they will often encourage it if they are worried too. Curious to see other practices in the situation. I have seen the EKG progress when doing this method in the past. Went from cardio saying it's just GERD to an immediate cath with 99% blockage.

 

[Aesculapius]

Anyone doing bedside echo to add another data point/possibly identify MIs sooner? Emergency Ultrasound Podcast recently had an interesting episode about identification of wall motion abnormalities.

 

[WilcoWorld]

a) That 1st ECG meets 1 of 5 criteria for BER (concavity). Good on ya for calling BS on the Cardiologist.

b) If I'm worried by story or an initially bad but inconclusive ECG I'll oder q15 min ECG's x3. If no change by 45 min then I'll ease off.

c) Lytics would be a bold, but good, call in this case. Less bold but "more good" if the patient was looking bad, which leads me to:

c2) If you've been seeing patients for a few years, and a patient looks really bad to you, then you should be more willing to act. This is a great example of "not looking right" but not precisely meeting criteria. Unless I could just smell the troponinemia I probably wouldn't have had the guts to push tPA based on ECG #1.

 

[EM2BE]

Another thought / question for you all. If you do not have an obs unit and you have a healthy-appearing person complaining of chest pain with a few risk factors do you:

a - get a 6 hour trop
b - 1 trop and go home
c - no trop
d - admit for further testing

If the pt appears at hour 6-8 of chest pain?
If the pt appears at day 3 of chest pain?

 

[The White Coat Investor]

Another thought / question for you all. If you do not have an obs unit and you have a healthy-appearing person complaining of chest pain with a few risk factors do you:

a - get a 6 hour trop
b - 1 trop and go home
c - no trop
d - admit for further testing

If the pt appears at hour 6-8 of chest pain?
If the pt appears at day 3 of chest pain?

I tend to repeat EKG and troponin at some point 6 hours after onset of pain, then schedule for outpatient stress test in the next 1-2 days. But, I offer them all admission and suggest they are low-risk, but not no-risk, but that if I were them I'd go home. Most do.

 

[tkim]

Follow up ...

Yep, some early repol.

 

[tkim]

http://journals.lww.com/em-news/Ful...iagnosis__Benign_Early_Repolarization.21.aspx

"The degree of STE related to BER is usually greatest in the mid- to left precordial leads (leads V2 to V5). The ST segments of the remaining electrocardiographic leads are less often elevated to the extent observed in leads V2 through V5. The limb leads (I, II, III, aVl, and aVf) are less often observed to demonstrate STE. One large series reported that the limb leads revealed STE in only 45 percent of cases of BER. Lead aVr does not demonstrate STE due to BER.11 Isolated BER in the limb leads, i.e., no precordial STE, is a very rare finding. Such isolated STE in the inferior (II, III, and aVf) or lateral (I and aVl) leads should prompt consideration of another explanation for the observed ST segment abnormality."

 

[TrumpetDoc]

I def. would not ever call that BER, particularly with symptoms.
The first had subtle limb changes/recip change. Good case.


Sent from my iPhone using Tapatalk

 

[southerndoc]

My interventionalists take early repols to the cath lab emergently if theyre old, have risk factors, or have a good story. A negative cath is better than a low EF in their opinion. I'm thankful I don't have to argue with them for cath lab activations. In fact, we call the cath team and it's OPTIONAL that we talk to the cardiologist.