A young woman in her early 20s with syncope

Written by Pendell Meyers


A 20 year old female with an episode of syncope was triage to my low acuity zone one morning. Her vitals were within normal limits except for her heart rate of 109 bpm.

I immediately went to evaluate her, without looking in the chart first. I found a well appearing young lady in the room with her parents who witnessed the event. She stated that she was sitting on a shallow ledge in a pool when she became lightheaded, so she got up out of the pool and then briefly syncopized next to the pool in front of her parents, who were able to catch her preventing any trauma. She returned to normal within 30 seconds to a minute. She said she had never had any episode like this before. She denied having any symptoms before or after the event, and she was asymptomatic on my initial exam.

After getting the story I asked about any known medical problems. They informed me that she had just been hospitalized 10 days ago for "some fluid around the heart" and was discharged after one day without incident.

So I immediately left the room to get an ultrasound machine.

As I rolled the machine back in the room, she suddenly became ashen and lethargic, then slumped back on the bed and was unresponsive but making some nonpurposeful movements and groaning. The cardiac monitor showed sinus rhythm but the automatic blood pressure cuff was not reading.

While calling for some help and arranging to have her transported to our critical care zone, I got this quick ultrasound which confirmed my suspicion:

This quick view was all I was able to obtain in the circumstances. It shows a moderate circumferential pericardial effusion, with the heart swinging back and forth during the cardiac cycle. More specific information such as definitive RV diastolic collapse was not indicated or available at this time given the obvious clinical context. 

We arrived in the resuscitation bay and recorded a heart rate of 115 bpm and blood pressure of 50/30 mm Hg. We obtained access and monitoring, but she showed no signs of improvement, and we judged that an intervention must be done in the ED without delay.

We performed a pericardiocentesis using a triple lumen central line kit under ultrasound guidance in a transthoracic position. Approximately 500 mL of serous fluid was withdrawn from the catheter with immediate normalization of all vital signs. 

A CT was obtained later and showed appropriate positioning of the catheter:

She was admitted to the ICU and the catheter was used several times to withdraw more fluid. Ultimately, she spent several days in the hospital and no further fluid collected. The ED catheter was removed at that point and she did not require any further procedures. She was discharged home in good condition.

Why did she go from stable to shock in seconds?  Why would she have such varying blood pressure?  Here is one possible explanation: She was sitting upright and started laughing.  Laughing, especially while upright, may have caused increased intrathoracic pressure and decreased venous return (patients in any tamponade are extremely vulnerable to changes in intrathoracic pressure and  venous return -- this is the basis of pulsus paradoxus in tamponade).

Looking Back to previous hospitalization

Only after the case was I able to go back and dig through her prior hospitalization. She had presented initially with 6 days of recurrent chest, back, and shoulder pain and was worked up with a reportedly normal ECG, CXR, and d-dimer. 

Here is her ECG at that first visit:

There is the slightest STE in lead II and III. There is awkward ST segment morphology in lead III. No prior was available. Overall I do not think I would have considered this ECG concerning on that first visit given the history provided. Smith: there are abnormal Q waves in II, III, and aVF, V4-V6, and abnormal RSR' in V1.

She improved with toradol and pepcid, so she was discharged home with motrin. She returned several days later with recurrent pain, but also nausea, diarrhea, and chills. Another ECG was performed, and this time was noted to be markedly abnormal. An echo was performed which showed a "trivial circumferential pericardial effusion." She was diagnosed with pericarditis and spent one day in the hospital without events. 

Below you will see serial ECGs from that hospitalization.

In addition to STE in V3-V6 and II, III, and aVF, there is bizarre morphology occurring in the end of the T wave, most bizarre in leads I, III, and aVF. I have not seen this particular morphology before, and it certainly does not match the known patterns of hypokalemia or reperfusion.

Much more classic findings of pericarditis. Note the diffuse ST elevation, and especially the lack of reciprocal ST depression and/or T wave inversion in aVL despite STE in III. Of course, some OMIs can produce diffuse STE such as a type III LAD, however the interpretation is greatly aided by the clinical scenario. Ultimately, in many cases there is no sure way to distinguish peri/myocarditis from OMI on ECG alone.

1 week later (about 1 week prior to the tamponade visit) she had a follow up outpatient visit and this ECG was recorded:

Appears to show resolving findings.

Here are her ECGs during the tamponade visit over several days:

First ECG collected during her visit with tamponade, performed soon after decompensation in the resuscitation room, but before pericardiocentesis. There is widespread STE returning slightly compared to the prior.

One day later after pericardiocentesis, again showing relative resolution of findings.

Learning Points:

Pericardial effusion is a key piece of information for the diagnosis and prognosis of pericarditis. It would be interesting to know how the case would have played out had another ultrasound been done at that follow up visit one week prior to her visit with tamponade. Maybe an increase in pericardial fluid could have been observed and helped to prevent her acute decompensation one week later.

This paper by Bischof and Smith compared inferior MI to pericarditis and found that of 154 patients with inferior STEMI, 17% of whom had less than 1 mm of STE in any inferior lead, all 154 had at least 0.25 mm STE depression in aVL. Among 49 patients with pericarditis who had inferior STE, zero of 49 had STD in aVL (though there are always rare exceptions such as this case). Interestingly this study also looked at 54 "subtle" inferior MIs, and of these 49 had some STD in aVL. Moreover, T wave inversion in aVL was also found to be 100% sensitive and 86% specific for inferior STEMI.

===================================

MY Comment by KEN GRAUER, MD (7/12/2020):

===================================

Tremendously insightful case by Dr. Meyers, regarding this 20yo woman who would not have survived if not for presenting to the ED precisely when she did + lifesaving pericardiocentesis by Dr. Meyers and the emergency care team.

My THOUGHTS: This 20yo woman had been diagnosed as having acute pericarditis + pericardial effusion several weeks earlier. I’d add the following general thoughts to the above discussion by Dr. Meyers:

• The stat Echo picture from the current ED visit tells all = a large, circumferential pericardial effusion with a “swinging heart” sign. (CLICK HERE — For those wanting an excellent review on “When does an Effusion become Pericardial Tamponade? — with dynamic echo illustrations by Dr. Stephen Alerhand).
• The BEST clinical evidence in support that this patient truly had Pericardial Tamponade — was immediate restoration of consciousness and normal vital signs following emergency pericardiocentesis of 500 mL of fluid from this markedly hypotensive and unresponsive patient.
• For an excellent review of the pathophysiological explanation of signs and symptoms associated with Pericardial Tamponade — SEE this Review by Jensen et al in the e-Journal Card. Prac 15(17), 2017. Among the clinical PEARLS in this Review are: i) As a result of increasing intrapericardial pressure resulting from a large pericardial effusion — diastolic filling is markedly reduced — which leads to sinus tachycardiaas an early ECG sign (in an attempt to maintain cardiac output); and, ii) The symptoms of tamponade depend greatly on the rapidity with which pericardial fluid accumulates. With rapid accumulation of pericardial fluid — there is far less time for pericardial elasticity to adapt to increasing intrapericardial pressures — therefore, far greater chance of abrupt hemodynamic decompensation.

Application to Today’s Case: The apparent relative lack of symptoms until this patient’s syncope, followed not long after by acute decompensation in the ED — suggest sudden rapid accumulation of pericardial fluid (for whatever reason) that given this patient’s history, probably occurred well after 1-2 weeks following her initial hospitalization.

• One has to wonder about the etiology of acute pericarditis in this previously healthy 20-year old woman! Work-up during her previous hospitalization was initially “normal” for ECG, CxR and d-dimer (although as discussed by Dr. Meyers and myself below — ECGs from that prior visit were not “normal”).
• By far, the most common etiology of acute pericarditis in a young adult is “idiopathic” (Khandaker et al: Mayo Clin Proc 85(6): 572-593, 2010) — although many (most) of these “idiopathic” cases are probably viral. Yet there is no mention from the brief history available for the 1st ED visit in this case of any viral prodrome (Nausea, diarrhea and chills were only noted when she returned to the ED, several days after her initial ED visit). Given this patient’s demographics (ie, a young adult female) — one has to consider the possibility that pericardial effusion could have been a 1st presentation of systemic autoimmune disease (ie, RA, SLE, etc.). This is especially true in this case given the presumed rapidity of pericardial fluid accumulation after seeming recovery from her initial hospitalization — and, the frank severity of her symptoms with this recurrence (ie, acute decompensation from pericardial tamponade). Follow-up (if not already done on this patient) should include serologic testing to rule out the possibility of an autoimmune disorder that would be likely to result in additional recurrence.

Regarding the ECG with Pericarditis & Pericardial Tamponade: Acute pericarditis is often a difficult clinical entity to detect — especially when it presents in a previously healthy young adult with no preceding viral illness and no pericardial friction rub on exam.

• There are too-numerous-to-count cases on Dr. Smith’s ECG Blog in which emergency providers diagnosed acute pericarditis that was really acute coronary syndrome. This is why Dr. Smith insightfully emphasizes, “You diagnose acute pericarditis at your peril”. Clearly — acute pericarditis is a most overdiagnosed entity!
• That said — there are occasions when you can definitively diagnosis acute pericarditis. This may be possible on physical exam! (See Pearl #2 in next bullet). Although most of the time with acute idiopathic (or acute viral) pericarditis, the Echo will be normal — the presence of pericardial effusion on Echo can be helpful in diagnosis, as it was in today’s case.
• PEARL #2 — Often overlooked is the diagnostic utility of hearing a pericardial friction rub. Characteristics of a Pericardial Friction Rub — are that it is a scratchy, superficial sound (like walking on snow). There may be only one or several components to the rub. It may wax and wane in loudness — often becoming louder during inspiration. The rub is often transient, and may be intermittent — so it is worthwhile auscultating the patient on at least several occasions! BOTTOM Line: IF a pericardial friction rub is heard — then the diagnosis of acute pericarditis is made! However — not hearing a rub does not rule out pericarditis. Clinical Reality: Despite a cited incidence in the literature of well over 50% of patients with acute pericarditis having a rub at some point during their course — the diagnosis of many (if not most) cases of acute pericarditis is made in the absence of hearing a rub. In My Teaching Experience: Over my nearly 4 decades of teaching, both during direct supervision rounds, as well as through internet ECG interpretation — a majority of providers fail to even mention that they listened for a rub (let alone listened carefully on more than one occasion). DON’T be embarrassed by missing (by not listening for) the rub that the cardiologist who sees the patient after you hears.
• PEARL #3 — In those cases in which the ECG is helpful in the diagnosis of acute pericarditis — one should be aware of the usual sequence of 4 Stages of ECG Evolution. Of these — only Stage I is potentially diagnostic (in which there is diffuse ST elevation — often with generalized PR segment depression + PR elevation in aVR). This may be followed in hours-to days-to weeks later by Stages II and III — in which elevated ST segments return to baseline (ie, “pseudo-normalization”) — and then go on to evolve to diffuse T wave inversion. (CLICK HERE and HERE— for “My Take” on the ECG diagnosis of acute pericarditis).
• PEARL #4 — The ECG is not at all sensitive or specific for detecting a large pericardial effusion with pericardial tamponade. There may or may not be Stage I diffuse ST elevation — which even if present, reveals nothing about the likelihood of pericardial tamponade. One looks for sinus tachycardia and diffuse low voltage — but many conditions produce these nonspecific findings. The one ECG finding that is strongly suggestive of tamponade if seen under the right clinical circumstances = Electrical Alternans! Physiologically — the ECG finding of electrical alternans correlates with the presence of a “swinging heart” within the pericardial sac on Echo. As a result — electrical alternans is unlikely to be seen with smaller (or even moderate) pericardial effusions — and, even larger effusions (as was seen in today’s case) do not necessary result in electrical alternans! But IF you do see electrical alternans — it is a helpful ECG finding (CLICK HERE — for my detailed discussion on ECG recognition and clinical implications of electrical alternans).

Regarding ECGs in Today’s Case: A total of 7 tracings are shown above in today’s case. I comment on 5 of these 7 tracings — which for clarity, I have numbered in the sequence in which they are shown above in discussion by Dr. Meyers (Figure-1).

Figure-1: The 1st, 3rd, 4th, 5th and 6th ECGs shown above in this case (See Text).

My thoughts regarding the 5 ECGs shown in Figure-1 are the following:

• ECG #1 — Did YOU notice that P wave morphology is different in ECG #5, compared to P wave morphology is each of the other 4 tracings? Of note — ECG #5 is the only tracing obtained at a time when this patient’s clinical condition had improved. I suspect that there was an ectopic atrial rhythm in ECGs #1, 3, 4 and 6 because: i) the P wave is either not clearly positive in lead II — or (as is seen in ECG #6) the P wave in lead II is smaller than the P wave in lead I, and is associated with negative P waves in leads III and aVF; and, ii) the PR interval is abnormally short in each of the 5 tracings with an ectopic atrial rhythm. In contrast, in ECG #5 (done when the patient felt well) — normal sinus rhythm is seen, because the P wave is clearly upright in lead II, and the PR interval is normal.
• PERHAPS (at least in this patient) — it may be that an ectopic atrial rhythm is a marker of increased intrapericardial pressure (which physiologically is transmitted to the atria with larger effusions). I found it remarkable that at least in this patient — the only ECG showing sinus rhythm was the one obtained when the patient’s clinical condition was at its best.
• NOTE: Among the ECG findings we look for when assessing a tracing for the possibility of acute pericarditis is PR depression in a number of leads with PR elevation in lead aVR. The fact that an ectopic atrial rhythm with a short PR interval is present in each of the 5 ECGs in this case in which the patient was symptomatic is relevant — because this finding may negate being able to recognize PR depression or elevation.
• There are a number of descriptive findings in ECG #1 that should be noted. I’ll emphasize that none of these are necessarily indicative of any pathology in an otherwise healthy young adult — but given the eventual diagnosis in this patient, it should be noted in ECG #1 that: i) There are narrow Q waves in multiple leads — and in some leads, these Q waves are fairly deep. This is relevant in this case — because Q waves may be a marker of acute myocarditis (and, in an older patient — Q waves might be a marker for infarction); ii) There is incomplete RBBB in ECG #1 — as seen from the narrow, but multiphasic QRS complex in lead V1, with terminal S waves in leads I and V6. (Note that the QRS complex in lead V1 is unusual in having 4 components = an rSR’s’ complex); and, iii) There is an abnormally rightward frontal plane axis in ECG #1 (the QRS in lead I is clearly more negative than positive). Under different clinical circumstances — this could be indicative of LPHB in association with incomplete RBBB. Finally = iv) There is subtle elevation of the J-point in a number of leads — with the shape (ie, ST straightening) and amount of ST elevation in leads I, II and V2 in ECG #1 being more than I would normally expect in an otherwise healthy young adult. I would usually ignore findings like this if the history was unremarkable — but in a clinical setting in which I was assessing the patient for possible acute pericarditis — I would wonder if these unusual but nondiagnostic ST-T wave changes might be a harbinger of what was to come.

ECG #3: As per Dr. Meyers, this tracing obtained during the patient’s 1st hospitalization does show a number of ECG findings consistent with Stage I of Acute Pericarditis. These include:

• Diffuse ST segment elevation that is either upsloping or straight in shape (but not coved) is clearly evident in at least 9 of the 12 leads. That this is a real finding is firmly established by comparison of ST-T wave appearance with what we saw in ECG #1. This also supports my initial impression of ECG #1 — in which I thought the elevated and straightened ST segment appearance might be a “harbinger” of what was to come.
• Typically with Stage I pericarditis — it is the right-sided leads (ie, leads aVR and V1) that are among those leads that do not show ST elevation. This is the case here in ECG #3.
• PEARL #5: Typically with Stage I pericarditis — the ST segment appearance of lead II resembles that seen in lead I (as we see here in ECG #3). This differs from acute MI — in which the ST segment appearance of lead II resembles lead III (instead of lead I).
• PEARL #6: 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 — SEE My Comment at the bottom of the December 13, 2019 post). ECG #3 satisfies abnormal criteria for this ratio — as there is at least 1 mm of ST elevation in lead V6 — with a T wave amplitude = 4 mm in this lead — for an abnormal ratio of at least 0.25!
• NOTE #1: Q waves (as seen here in ECG #3 in multiple leads) are not part of the ECG findings suggestive of acute pericarditis. If new, in a 20yo patient — such Q waves could be a marker for myocarditis. In this patient — they most probably are a longterm normal finding for her.
• NOTE #2: The fact that a non-sinus rhythm with a short PR interval is present in ECG #3 negates the absence of PR depression as a factor against pericarditis.

ECG #4: Compared to ECG #3 — an even greater amount of diffuse ST elevation is seen in ECG #4.

• NOTE: Did YOU notice that QRS morphology has changed a bit in ECG #4 compared to ECG #3. Note there is no longer an rSR’ complex in lead V1 — and the R wave in lead V2 is markedly taller in respect to the S wave in this lead, than it was for lead V2 in ECG #3. In addition — the ST-T wave is now positive in lead V1 of ECG #4. I suspect the reason for these changes is likely to be slight shift in precordial lead placement. That said — clinically, our impression is the same (ie, ECG findings in ECG #4 are again consistent with Stage I acute pericarditis).

ECG #5: As noted earlier — sinus rhythm with a normal PR interval is seen in ECG #5. The 4-component (rSR’s’) incomplete RBBB complex is again seen in lead V1.

• As per Dr. Meyers — diffuse ST elevation is significantly less than it was in ECGs #3 and 4 — corresponding to this patient’s improved clinical condition.

ECG #6: The bottom tracing in Figure-1 was obtained in the ED, soon after hemodynamic decompensation, but before 500 mL of fluid were withdrawn with emergency pericardiocentesis.

• Other than return of the low atrial rhythm (very short PR interval; now again with negative P waves in leads III and aVF) — ECG #6 otherwise shows no ECG signs of acute pericardial tamponade. That is: i) There is no tachycardia; ii) There is no low voltage; and, iii) Despite the “swinging heart” on Echo (shown above in discussion by Dr. Meyers) — there is no Electrical Alternans in ECG #6 (CLICK HERE — for much more on Electrical Alternans).

Our THANKS to Dr. Meyers for this superb clinical case!