Friday, December 13, 2019

"Pericarditis" strikes again

Written by Pendell Meyers

A man in his late 40s with several ACS risk factors presented with a chief complaint of chest pain.  Several hours prior to presentation, while driving his truck, he started experiencing new central chest pain, without radiation, aggravating/alleviating factors, or other associated symptoms. On review of systems the patient reported back pain for approximately 1 week which he was treating with NSAIDs with minimal relief. On exam the patient was well appearing, with normal vitals signs other than BP 155/82, no murmurs or rubs, normal pulses, no reproducible chest pain.

Here is his triage ECG:

What do you think?

Sinus rhythm
Normal QRS morphology
Diffuse STE, including leads V2-V6, I, II, III, aVF (with obligatory reciprocal STD in aVR)
Perhaps a tiny bit of PR depression (up to 0.8 mm has been described in normal subjects)

Overall impression: In my opinion and experience, this ECG most likely represents a normal baseline ECG, but with a small chance of pericarditis instead. I do not believe there is any finding here suggestive of OMI.

I texted this to Dr. Smith without any information, and this was his reply: 
"This could be pericarditis but probably is normal variant."

The treating clinicians were worried about the ST elevations and called cardiology for emergent PCI consult. The cardiologists felt that the ECG did not represent ACS, and thought it was more likely pericarditis, so they did not take him to the cath lab.

His first troponin T then resulted elevated at 0.19 ng/mL.

A repeat ECG was performed and cardiology was re-consulted:
Roughly unchanged.

With the troponin elevated and ongoing pain, cardiology now decided to take him to the lab.

They found non-obstructive CAD, with only a 20% stenosis of OM2 and 10% RCA. No acute culprit.

He was admitted to cardiology.

Here is a quote from his initial cardiology admission note (after cath was done showing no acute culprit):

"...chest pain, non-radiating, pleuritic in nature, relieved by sitting forward"

"Plan: likely myocarditis"

Colchicine was ordered and the patient received the first dose in the afternoon.

Troponins gradually trended down from 0.19 ng/mL.

The next morning the patient went for his routine echocardiogram, where the operator noticed a dilated aortic root at 5.47 cm with severe aortic insufficiency. The team was notified and they ordered a stat aortagram which showed type A aortic dissection from the aortic valve to the iliacs.

Aortic Root Dissection

The aorta was emergently repaired, and the patient had a complex course (including a saddle PE and subsequent GI bleed!) in the ICU but survived with excellent function.

Not surprisingly the cardiology HPI changed yet again in the next note following diagnosis of the aortic dissection:

"...chest pain that began acutely while was attempting to park his truck, described as dull pain which radiates to the back, without exacerbating or alleviating factors..."

Here is his ECG several days after diagnosis:
Again, normal variant STE
Pericarditis would have its own typical evolution, as would acute MI.

The patient returned one month later for an unrelated problem:

Learning Points:

It seems likely to me that the notion of "pericarditis" delayed or completely prevented diagnosis of aortic dissection in this young man. As we have described multiple times on this blog, false positive "pericarditis" kills by distracting the clinician from actual emergencies including OMI, dissection, PE, and others.

The vast majority of cases with chest pain diffuse ST Elevation are due to Normal Variant ST Elevation, NOT to pericarditis.

You diagnose pericarditis at your peril!  Acute MI is frequently misdiagnosed as pericarditis.  Patients with pulmonary embolism or aortic dissection who have normal variant ST elevation are at high risk of being diagnosed with pericarditis when what they have is far more serious!!  Pericarditis is a diagnosis of EXCLUSION.

On the ECG of pericarditis:
--There should be marked PR depression
--There should be flat ST elevation with low T-wave to ST segment ratio.
--There should be no reciprocal ST depression anywhere except aVR.
--And there should be confirmatory evidence, and evidence of EXCLUSION of other serious pathology.

See this case, also written by Pendell Meyers when he was a medical student:

31 Year Old Male with RUQ Pain and a History of Pericarditis. Submitted by a Med Student, with Great Commentary on Bias!  (normal variant misdiagosed as pericarditis)

Other cases:

Palpitations and Chest Tightness: Should You Activate the Cath Lab (or Give Thrombolytics)? (normal variant, not pericarditis)

A Young Man with Sharp Chest pain
 (normal variant, not pericarditis)

You Diagnose Pericarditis at your Peril (at the Patient's Peril!)  --Acute MI misdiagnosed as pericarditis.

MY Comment by KEN GRAUER, MD (12/13/2019):
I LOVE this post — as it explores the important decision-making process associated with the ED diagnosis of acute Pericarditis. I completely agree with many of the KEY findings conveyed by Dr. Meyers.
  • That said — I offer another perspective on some aspects of this case.
Before discussing my thoughts on interpretation of ECG in this case — it may be helpful to review some of the ECG criteria for acute pericarditis. I’ve excerpted in Figure-1, relevant paragraphs from a 2017 ESC (European Society of Cardiologyarticle on this subject by Xanthopoulos & Skoularigis (ESC: Vol. 15, No.15-9/6/2017).

Figure-1: Excerpt from ESC review on acute pericarditis (See text).

Among the KEY points regarding ECG diagnosis of acute pericarditis from Figure-1 (to which I’ve added some of my own points) — are the following:
  • Typical ECG findings of acute pericarditis are not always present! The ECG is far from a perfect diagnostic tool.
  • 4 STAGES are described in the ECG evolution of Acute Pericarditis findings. These can be summarized as: i) Stage I — in which there is generalized ST elevation in most leads (except perhaps in the “right-sided” leads = leads III, aVR and V1)ii) Stage II — in which this generalized ST elevation returns to baseline (ie, “pseudo-normalization” phase); iiiStage III — in which there is diffuse T wave inversionandiv) Stage IV — in which the overall tracing normalizes.
  • Of these 4 stages — only Stage I is readily recognizable as potentially due to acute pericarditis. You would never guess “pericarditis” if given an isolated tracing from Stage II, III or IV.
  • The time course of these 4 ECG Stages (in those pericarditis cases in which they occur) — is highly variable. The ESC summary in Figure-1 states that Stage I may last “between hours to days” — before evolving into Stage II.
  • The shape of the ST elevation is typically concave-up (ie, a “smiley” configuration).
  • ECG signs usually associated with acute MI (ie, abnormally large or wide Q waves — reciprocal ST depression) are absent!
  • PEARL #1 — The appearance of the ST-T wave in lead II tends to resemble that in lead I with acute pericarditis. In contrast, with acute MI — ST-T wave appearance in lead II resembles lead III (and not lead I).
  • There may be PR depression in many leads — with “reciprocal” PR elevation in lead aVR.
  • The RATIO of the amount of ST elevation to T wave amplitude in lead V6 should be less than 0.25 (ie, height of the ST elevation, as measured from the end of the PR segment to the J-point — should be less than 1/4 of the height of the T wave in lead V6).
  • Editorial NOTE: Much attention has focused on this Ratio criterion in recent years. I have not been an advocate of this criterion — because I feel ( = my opinion) that it is often difficult attain consistency in measurements among practicing clinicians. All it takes is a very small error in how the amount of ST elevation is measured — to throw off your calculation of whether this ST segment height is over or under 25% of the height of the T wave in this lead. That said — I illustrate HOW this RATIO is arrived at in Figure-2, which I have adapted from the 3/16/2019 post in Life-In-The-Fast-Lane.

 Illustration of how the ST segment to T wave ratio is calculated — adapted from Life-In-The-Fast-Lane (See text).

  • These words from Drs. Meyers and Smith should be at the forefront of every emergency provider, You diagnose acute pericarditis at YOUR peril! The reason for this is clear — Statistically (if you take all comers who present to the ED) — acute coronary syndromes are much more common than acute pericarditis as a cause of new-onset chest pain with ST elevation on ECG.
  • That said — this statistical prevalence in favor of acute coronary syndromes would be significantly reduced IF— the patient in question was a younger adult with a viral prodrome (ie, of fever, myalgias, pleuritic chest pain) — as a viral cause is the most common etiology of acute pericarditis.
  • IF, on the other hand — the patient with new chest pain is older and lacks predisposing viral symptoms — then acute pericarditis becomes a rare diagnosis in an ED setting.

Regarding the ECGs in THIS Case: For clarity — I put 3 of the 4 ECGs in this case together in Figure-3:

Figure-3: Three of the 4 ECGs in this case (See text).

My Impression of ECG #1: As per Dr. Meyers — there is diffuse ST elevation in ECG #1. There is no more than minimal PR segment depressionMDifferent Perspective on ECG #1: The above said — IF the history would have been that this patient was a young adult with a recent acute viral illness, who now presented with pleuritic and positional chest pain (exacerbated by lying supine, and reduced by sitting and leaning forward) — I would have interpreted ECG #1 as clearly consistent with acute Pericarditis. My reasoning is as follows:
  • ECG #1 — shows more than a little ST segment elevation of upward concavity shape in multiple leads (ie, leads I, II, III; aVF; and V2-thru-V6). I thought the diffuseness and amount of ST elevation in ECG #1 is clearly more than I’m used to seeing with repolarization variants. This is not to say that ECG #1 might not reflect a repolarization variant — but only to say that IF the clinical history and physical exam were consistent with what one might expect for acute pericarditis — that ECG #1 is clearly consistent with this diagnosis ( = my opinion).
  • Although with this amount of generalized ST elevation, I would have expected more PR depression than we see in ECG #1 — the lack of PR depression (in my opinion) does not rule out the possibility of acute pericarditis. In my experience — generalized ST elevation is a much more important ECG sign than PR depression (See My Comment at the bottom of the December 10, 2019 post in Dr. Smith’s ECG Blog).
  • The ST-T wave appearance of lead II in ECG #1 much more closely resembles the ST-T appearance in lead I than in lead III (See Pearl #1 above).
  • There is no reciprocal ST depression in ECG #1 (other than the expected ST depression in lead aVR). The q waves that are present are small and narrow, and consistent with normal “septal” q waves.
  • The ST segment to T wave amplitude RATIO in ECG #1 actually is consistent (to my measurement) with an ECG diagnosis of acute pericarditis (as per the insert to the right of ECG #1 — this ratio is >0.27, and therefore is positive!).
  • PEARL #3: In my experience of reviewing literally hundreds of internet cases in which the diagnosis of acute pericarditis was suspected — the overwhelming majority of clinicians FAIL to ever mention that they listened for a pericardial friction rub. If not mentioned as a positive or negative finding — this usually means that the clinician did not listen for a rub. Failure to listen for a rub is a critical oversight — because IF you hear a pericardial friction rub — then you have made the diagnosis of acute pericarditis! (Dr. Meyers emphasized that no rub was heard in this case). BOTTOM LINE — DONT FORGET to carefully listen for a RUB whenever considering acute pericarditis as a diagnostic possibility!

NOTE: In this particular case — errors were made in assessing the history. As per Dr. Meyers — inappropriate focus on the diagnosis of pericarditis appeared to delay (and almost prevented) the correct diagnosis of acute aortic dissection.
  • As per Dr. Meyers — Acute pericarditis must be a diagnosis of exclusion!
  • That said — acute pericarditis does occur on occasion ... so it is important to be saavy on the ECG signs of this diagnosis

Therefore, realizing that the patient in this case did not have acute pericarditis — I wanted to highlight some points about the 3rd and 4th ECGs that were done in this case.

QUESTION #1: Imagine the patient with ECG #1 did have acute pericarditis.
  • In this case — Would ECG #3, done several days later be consistent with the ECG evolution expected for true acute pericarditis?

ANSWER: There clearly is less ST elevation in ECG #3 in virtually all leads compared to ECG #1. As per Figure-1 — this is consistent with the evolution of Stage I-to-Stage II pericarditis, that is expected to be seen on serial ECGs over ensuing hours-to-days.
  • NOTE — I realize the patient in this case did not have acute pericarditis. I am merely HONING your ECG interpretation skills by highlighting that what we see in ECG #3 could be perfectly consistent with ongoing ECG evolution in a patient who did have pericarditis.

QUESTION #2: ECG #4 was obtained from the patient in this case about 1 month later, when he returned to the ED for an unrelated problem. Does ECG #4 serve to further our understanding? (HINT: Consider ECG #4 as the follow-up tracing to ECGs #1 and #3).

ANSWER: Unfortunately, ECG #4 does not further our understanding about this case. Compared to ECG #3 — there is additional ST-T wave flattening, and beginning T wave inversion in a number of leads, not unlike what might be seen as one evolved into Stage III pericarditis. That said — the heart rate is significantly faster than it was for the prior 2 tracings (ECGs #1 and #3) — so there is really no way to distinguish what might represent ST-T wave changes due to tachycardia vs evolution of the patient’s underlying disorder.

Our THANKS to Dr. Meyers for presenting this informative case!


  1. Great Post!!! But i have two questions regarding firt ecg. Firt, i have seen that there is not terminal QRS distortion in this tracing, can this rule out OMI? Second, the R wave in V5 is more than 26 mm, besides SV2 + RV4 is more than 35 mm, is there LVH?

    1. @ Unknown — THANKS for your comment. Although the SPECIFICITY of T-QRS-D (Terminal-QRS-Distortion) is excellent for ruling in acute OMI, the SENSITIVITY of this sign is poor. This means that most of the time when there is acute OMI, you will probably not see T-QRS-D — and that the LACK of this sign in no way rules out the possibility of acute OMI. Otherwise — YES, voltage criteria for LVH are definitely met in ECG #1. It’s interesting to note that QRS amplitude change on subsequent tracings (I suspect that may be due to changes in lead placement …). THANKS again for your interest! — :)

  2. Great analysis, Ken...

    I especially enjoyed your pearls (actually, I'm jotting them down right now)! But I was wondering, in the case of Pearl #1, I understand the similarity of Leads I and II during phase 1 pericarditis, but shouldn't Lead II look more like aVF than Lead III during an ACS? I have seen a lot of acute epicardial ischemias in which only 2 out of 3 inferior leads were affected, but it has never been limited to Leads II and III since they are on the opposite sides of aVF in the hexaxial reference grid.

    I certainly agree 100% with your point regarding auscultation in the diagnosis of acute pericarditis. In my day, auscultation of the heart was just part of the physical exam done for anyone with chest complaints. It is unfortunate that that part of the exam is frequently omitted nowadays and in too many instances all the patient receives is the ubiquitous "stethoscope tap" - placement of the diaphragm on the chest for considerably less than 0.5 second. And a pericardial friction rub is usually NOT a subtle sound - it's really very easy to hear.

    1. THANKS so much (as always!) Jerry for your Comments! I picked up the point on resemblance of leads I & II (rather than II & III) in acute pericarditis from my mentor, Barney Marriott. I completely agree with you that lead aVF lies “closer” to lead II than does lead II to lead III, on the hexaxial reference grid. That said — I’ve always thought it easier to remember as a “Pearl” to think of the 3 standard limb leads — and visually, I’ve always felt it easiest to compare the ST-T wave appearance of leads I & II — vs that of leads II & III. Otherwise — I’m amazed at how many cases get posted in the EKG Club (and in other ECG forums) in which the presumed diagnosis is acute pericarditis, yet mention of listening for a pericardial friction rub is totally missing … THANKS again Jerry! — :)

  3. Do the j waves in V4/V5 (ecg 1)confirm early repol and go against pericarditis?

    1. Prominent J-point notching in association with upward-concavity (ie, “smiley”-configuration) ST segment elevation in a number of leads is a common finding with repolarization variants. But, these findings by themselves are not definitive. I’d add that patients who start out with a repolarization variant (perhaps with this upward-concavity-type of ST elevation with prominent J-point notching) — may THEN develop acute pericarditis or acute coronary syndrome — in which case that patient might manifest residual J-point notching that was there BEFORE the superimposed new condition developed. Bottom Line: Nothing is 100% predictive as a single factor. ST segment elevation shape and presence of J-point notching may favor a repolarization variant — but the entire case must be assessed. As I emphasized in, “My Impression of ECG #1” (above) — IF the history in today’s case would have been that this patient was a young adut with a recent acute viral llness, who now presented with pleuritic and positional chest pain — despite that J-point notching, ECG #1 could have been perfectly consistent with a diagnosis of acute pericarditis.

  4. "BP 155/82, no murmurs or rubs" features are not convenient for severe aortic regurgitation. We may see wide pulse pressure and diastolic murmur?

    1. @ Dilsuhte — While true that the pulse pressure for a BP of 155/82 is not overly "wide" — this BP does not in any way rule out AI (Aortic Insufficiency). The diastolic murmur of AI is EASY to miss unless you are specifically listening for it. We also need to keep in mind that at least 1 day passed between the time this BP and the lack of a diastolic murmur was reported — and the time the Echo and aortagram that showed severe AI were done. So given this aortic dissection was acute — there is no way to know if severe AI was present the day before … While GOOD to astutely note physical exam findings like these — I find it always difficult when we “weren’t there” to know what it means when physical findings we would have carefully looked for weren’t noted.


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