Thursday, May 26, 2022

Does a Single Troponin below the 99th percentile URL Rule out Acute MI if the Chest pain is very prolonged?

66 y.o. male who presented for chest pain that started this AM when he woke up, and has  persisted throughout the day prompting him to call 911. He says the pain is dull in nature and located across the chest, does not radiate, that it is worse with exhalation. He denies worsening with activity or positioning.  He endorses SOB and requested to sit up. He says this has not happened to him before. He endorses cough productive of yellow sputum.  He denies any edema. Denies history of venous thromboembolism.  He endorses a 50 pack year history of smoking. He denies recent illness or recent sick contacts. He denies fevers, sweats, chills, headache, dizziness, lightheadedness.

Here is his ED ECG:

Artifact.  Probable LVH.  No clear evidence of OMI or ischemia or reperfusion.
If this patient had brief pain that resolved, one might interpret the T-wave inversion in inferior leads as reperfusion.
However, when the pain is constant for 12 hours, and persistent, one must interpret them as entirely non-diagnostic.
And if the troponin is not elevated, it would prove them to not be due to ischemia.

I saw this patient and on my history, his pain had been present and constant for 12-36 hours.

We did an ED bedside cardiac ultrasound, which was normal.

After a single troponin returned "negative" at 8 ng/L (URL = 34 ng/L; Limit of Detection = 4 ng/L), I believed we had ruled out MI.

Here is the Assessment

  1. CBC, Chem, and dimer were all unremarkable 
  2. EKG was non ischemic
  3. Cardiac US was within normal limits 
  4. CXR showing posterior costophrenic angle interstitial opacities with atelectasis or infiltrates.  No distinct osseous abnormality.
  5. Given albuterol nebulization, with improvement. 
Impression: Bronchitis and Pneumonia
Treatment for asthma and pneumonia

Discharge home after a single "negative" troponin?
Is there data to support sending a patient home after prolonged constant chest pain, a non-ischemic ECG, and a single Abbott Architect troponin I that is ≥ 5 ng/L but below the URL (34 ng/L for men and 16 ng/L for women), which is the level required for a diagnosis of MI?

There is only one study of which I am aware that addressed this, and I published it 18 years ago (see below).  There are indeed studies of discharge after a single troponin and a low HEART score, but they do not reach a sensitivity of less than 1% for myocardial infarction.  

--This patient's HEART score was 3, so he would be eligible for discharge by such protocols. 
--His EDACS score is 14, with a non-ischemic ECG, so he would be eligible for discharge with 2 troponins below the URL. 

With constant pain lasting 12 hours prior to presentation, a single troponin should suffice, but other than my 18 year old study (again, see below), there is no data that I am aware of. 

On the other hand, we and others have shown that, if a patient presents with acute chest pain and has a troponin sample that is drawn at 6 hours (and especially 12 hours) after presentation and is below the 99% URL, that acute MI is ruled out with high sensitivity (see slide below).  

Because this is well known, we often extrapolate this data and apply it to patients who present with prolonged pain, and we frequently discharge such patients if a single initial troponin is below the URL and the ECG is non-ischemic.  We don't always get a 2nd troponin.  Of course this depends on the patient's symptom quality and risk factors as well.

Discharge with a low risk score and 2 troponins below the URL is definitely safe (NPV ≥ 99% for both acute MI and 30 day adverse events), by this study and by many others:

Smith SW. Tibbles C.  Apple FS.  Outcome of low risk patients discharged home after a normal cardiac troponin I.  Published 18 years ago based on a very insensitive troponin, but there were very few adverse events.

Case Continued

The next day:

The patient returned in respiratory failure and cardiogenic shock!

I interpreted this as subacute LAD Occlusion.
In other words:
It appeared as if I had sent a patient home who had unstable angina and it had progressed to acute anterior STEMI/OMI!!

He required intubation.
The cath lab was activated.
We did a bedside ultrasound: The left ventricular ejection fraction appears: severely reduced in function when compared to LV function yesterday on ED POCUS

The first troponin returned at 4000 ng/L, consistent with subacute OMI.

I was distraught.



Diagnosis: Takotsubo, probably due to pneumonia.

So it was not acute coronary syndrome at all.  I had not sent ACS home.  Phew!

Formal echo day 3:


Mildly enlarged left ventricle with severely reduced systolic function. Estimated ejection fraction 21%.

Severe hypokinesis to akinesis of the mid and apical left ventricle in all segments with best preserved function in the basal inferolateral wall.

Day 4 ECG:

Now with T-wave inversions

Day 6

Echo recovery with EF 45%

Learning Points:

1. Takotsubo mimics LAD Occlusion

2. Otherwise, I'm not certain!  Except to say that this case is not a counterexample for sending a patient home who has had pre-arrival chest pain for 12 hours and a single troponin that is measurable but below the URL.

Comment by KEN GRAUER, MD (5/27/2022):
Interesting case — with the important clinical question posed as to how effective a single troponin value can be for ruling out infarction when chest pain is prolonged. In the interest of academic discussion — I focus my comment on a few specific aspects regarding the ECGs in today's case.
  • For clarity in Figure-1 — I've reproduced the first 2 tracings in today's case.

Regarding ECG #1:
The initial ECG in the ED clearly does not show evidence of acute OMI (Occlusion-based MI). That said, in the context of a 66-year old man with a history of chest pain persistent for the previous 12-to-36 hours — I thought there are some findings worthy of note on ECG #1:
  • There is a large amount of artifact, especially in the limb leads.
  • The rhythm in ECG #1 is sinus. The axis is leftward, consistent with LAHB (Left Anterior HemiBlock).
  • QRS amplitude in a number of chest leads is significantly increased — consistent with voltage for LVH (based on the very deep S in V3 >25 mm — and the tall R in V5).
  • Although difficult to assess because of limb lead artifact — T waves appear to be negative in all inferior leads (RED outlining in these leads). Whether this is in response to predominant negativity of the QRS in these leads — or whether it reflects some element of ischemia is uncertain from this artifact-marred picture.
  • QS complex is seen in leads V1 and V2 — and no more than a tiny r wave is seen in lead V3. Whether this poor R wave progression is the result of LAHB (which results in unopposed posterior forces initially) — or — the result of LVH (which may produce prominent posterior forces with loss of anterior r wave amplitude) — or — anterior infarction at some point in time — is uncertain from this single tracing.
  • I found it a bit unusual for the ST-T waves in leads V1,V2,V3 to be nearly flat — especially in view of probable LVH that typically produces a definitely upright (if not slightly elevated) ST-T wave in anterior leads as a manifestation of LV "strain" when there are deep anterior S waves. Could this be pseudo-normalization?

  • To EMPHASIZE: This initial ECG #1 is not suggestive of acute OMI. But in a 66-year old man with risk factors and a concerning history — I thought the above ECG findings opened some questions. In view of the importance of this tracing in the decision-making process — and in view of all the limb lead artifact — I would have repeated the tracing.

Take Another LOOK at ECG #2:
  • Why might the frontal plane axis in ECG #2 be so different than it was the day before in ECG #1?

Figure-1: The first 2 tracings in today's case.

The reason the frontal plane axis in ECG #2 is so different from the axis in ECG #1 — is that there is LA-LLead Reversal in ECG #2! For clarity in Figure-2 — I show what the limb leads in ECG #2 should look like if the extremity electrode leads were correctly placed.
  • The tipoff to LA-LL Reversal in ECG #2 — is that the P wave in lead I is clearly larger than the P wave in lead II (and that is distinctly unusual when there is sinus rhythm).
  • The other "tipoff" that something is amiss — is the marked right axis in ECG #2, that of itself is unusual outside of the context of RVH. Lead-to-lead comparison of ECG #1 with ECG #2 shows a dramatic shift from a leftward axis — to a rightward axis over the course of 1 day. That rarely happens.

  • P.S. — In the July 28, 2020 post in Dr. Smith’s ECG Blog — I cited my favorite on-line “Quick GO-TO” reference for the most common types of lead misplacement, which comes from LITFL ( = Life-In-The-Fast-Lane). Simply put in, “LITFL Lead Reversal” into the Search bar — and the link comes up instantly!

And IF We Account for Correction of Lead Placement in ECG #2:
Note ECG #2a in Figure-2 — now shows: i) Return of the leftward axis (LAHB); ii) A sinus P wave in lead II that is taller than the P wave in lead I (as should be the case with sinus rhythm); andiii) ST coving and T wave inversion in the inferior (instead of high lateral) leads.

Figure-2: Showing the effects of LA-LL Lead Reversal (See text).

Final Comparison — Of ECG #2a and ECG #3:
Use the corrected version of ECG #2 ( = ECG #2a) — it now makes more sense when we compare the last 2 tracings in today's case (Figure-3):
  • There is no significant difference in heart rate or frontal plane axis between ECG #2a and ECG #3.
  • There may be slight change in chest electrode lead placement (ie, the S wave in lead V2 is much deeper in ECG #2a compared to ECG #3). That said — this difference is unlikely to alter our overall assessment of these 2 tracings.
  • The overall "shape" of ST-T waves is similar in most leads — with the main difference being some improvement in ECG #3, which was done 2 days after ECG #2a.

As per Dr. Smith — the final diagnosis that accounts for the dramatic ECG changes seen beginning with the 2nd ECG in today's case was Takotsubo Cardiomyopathy. Findings consistent with this diagnosis (seen in Figure-3) include sinus tachycardia — the long QT — diffuse ST-T wave abnormalities without anatomic localization.
  • For review of the ECG findings seen with Takotsubo Cardiomyopathy — Please check out My Comment at the bottom of the page in the March 25, 2020 post in Dr. Smiths Blog.

Figure-3: Comparison of the 2nd and 3rd tracings in today's case (accounting for what ECG #2 would look like if leads were correctly placed).

No comments:

Post a Comment

DEAR READER: I have loved receiving your comments, but I am no longer able to moderate them. Since the vast majority are SPAM, I need to moderate them all. Therefore, comments will rarely be published any more. So Sorry.

Recommended Resources