Wednesday, January 14, 2015

Cardiac arrest, severe acidosis, and a bizarre ECG

A middle aged male had an unwitnessed PEA arrest associated with cocaine use.  Whether there was a shockable rhythm prior to PEA is unknown, but he was never defibrillated.  He received chest compressions with LUCAS and 3 doses of epinephrine, and was intubated by prehospital providers.  He had intermittent pulses.  Here is his initial ECG, with a pH of 6.50:
The rhythm is uncertain: probably an accelerated junctional rhythm with RBBB and PVCs, but it could be an accelerated rhythm initiated in the left bundle, mimicking RBBB.  It is important to ascertain the end of the QRS, which I attempt to do below.
Also present are 6 PVCs: complexes 3, 6, 7, 9, 10 (second PVC morphology), and 13  

With this wide complex, hyperkalemia should be high on the differential diagnosis

Here I try to find the end of the QRS:
Try to find the end of the QRS in any lead.  If you use lead II, it appears as if the blue line is at the end of the QRS. But it is uncertain.  The black line may represent the end of the QRS in leads V5 and V6, but I don't think so.
If the QRS end is represented by the blue line, then there is quite a bit of ST elevation (V1-V4) and ST depression (V5 and V6).

I was very suspicious of hyperkalemia, so we gave 3 g of calcium gluconate.  But the K returned at 4.5 mEq/L.

So is this STEMI?

Some clinical context is important:

1. It was not ventricular fibrillation, but PEA.  This suggests another mechanism other than ischemia.
2. We found that the pH was 6.50 (pCO2 100, HCO3 8).  Severe acidosis can result in very deranged ECGs.
3. Cocaine complicates the clinical picture.  As he had zero neurologic function after resuscitation, we were suspicious of possible cocaine associated head bleed followed by resp arrest then PEA arrest.

I did not think this was STEMI, but rather the consequences of cocaine, acidosis, and cardiac arrest.

The patient was ventilated and another ECG was recorded at 15 minutes, at a pH of 6.70 (pCO2 50, HCO3 5):
 I believe it is easier to find the end of the QRS in this one.  See below.
Also, it now it resembles Brugada Type 1 pattern ECG, which can be induced by cocaine, a sodium channel blocker (like all ***caines, including lidocaine).

Here I draw lines again:
The end of the QRS is indicated by the blue lines.  There still appears to be some ST elevation and depression, but it is less.  There is a Brugada Type 1 morphology in V1-V3.

The patient was in cardiogenic shock, partly due to bradycardia and partly due to presumed low SVR, as the ejection fraction by bedside ultrasound was excellent.  He was put on an epinephrine drip to increase heart rate and Systemic Vascular Resistance.  75 minutes later, the pH was 7.00 and this was the ECG:
Still bizarre, and Brugada-like, but normalizing.

Further information:
It turns out he had had a bradycardia arrest 7 weeks prior, also associated with cocaine, and with apparent status epilepticus.  In that case, there probably was never a shockable rhythm.  This was his first ECG at that time (pH 6.68)
Again, it is a bizarre RBBB pattern with Brugada type 1 morphology.  There is ST elevation but it does not look like the ST elevation of STEMI as it is downsloping with a negative T-wave.

7 minutes later, this ECG was recorded:
It is narrowing a bit but still bizarre with RBBB morphology and downsloping STE.

And yet another had been recorded at 24 minutes after the 1st (17 after the 2nd)

At that visit, he had been cooled. Angiogram had shown no Coronary disease.  He had awoken and been discharged.  His cardiac arrest had been attributed to status epilepticus and polysubstance abuse.

Further Events on this Presentation:

Pupils were still fixed and dilated. Even with the ambient lights off, there was no response to light.  Since PEA arrest with severe neurologic deficits may be due to intracranial bleed, including subarachnoid hemorrhage (and SAH may be caused by cocaine), we obtained a head CT before initiating cooling.  It was negative.

He was cooled and admitted to the ICU.

The next day he was moving all extremities, had a peak troponin of 0.50 ng/mL, and had a normal echocardiogram.

Here was the next day ECG at a temp of 33 degrees:
I am uncertain, but I suspect that he is shivering.  All the bizarre findings are gone.

He awoke fully and neurologically intact (so much for GCS 3 with fixed and dilated pupils!).

We consulted the electrophysiologist, who plans to do a workup for occult Brugada syndrome.  The workup may include family history, provocative testing with Na channel blocking agents, and ECGs of family members.

1. Cocaine-induced myocardial infarction in patients with normal coronary arteries probably involves adrenergically mediated increases in myocardial oxygen consumption, vasoconstriction of large epicardial arteries or small coronary resistance vessels, and coronary thrombosis.  Such ischemia may cause cardiac arrest.

2. But cocaine can also cause cardiac arrest from its sodium channel (local anesthetic) properties. There need not be an underlying sodium channel defect (e.g., Brugada) for cocaine to result in arrhythmic cardiac arrest.

3. Furthermore, these same Na channel blocking effects can unmask underlying Brugada syndrome, just like any Na channel blocker.  Electrophysiologists perform provocative testing with Na channel blockers to help diagnose Brugada syndrome.

Literature on Cocaine and Brugada Pattern ECG is limited to case reports:

1. Here is one case report in which the ECG looks like hyperkalemia, but the author assures me the K was only 4.5 mEq/L: Cardiac arrest from cocaine with Brugada pattern (full text)

2. Here is another full text case: Brugada Pattern ECG and cardiac arrest in cocaine toxicity: reading between the white lines.

3. Aborted Sudden Death, Transient Brugada Pattern, and Wide QRS Dysrrhythmias After Massive Cocaine Ingestion

4Hyperkalemia and cocaine induced dynamic Brugada-type electrocardiogram


Severe Acidosis may also affect the ECG, but this pattern was quite specific and unlikely to be due to acidosis alone.


  1. Nice case with some important concepts. THANKS!

  2. At first instance ecg is telling can cause vt.

  3. Hello Dr Smith , again thank you for such a great blog and extra references. I am a paramedic. This post had me thinking. What are your thoughts on empiric dosing of Sodium Bicarbonate in the resuscitation phase post ROSC to combat suspected Na channel toxicity myocardial depression in cases that may present like this. The initial ECG in both post ROSC ECG demonstrate bradyarrhythmia, wide QRS, terminal rightward axis deviations present in Lead 1 and AVR, and the Brugada patterns in the context of arrest after cocaine . Thanks again Troy

    1. Troy,
      Yes Bicarb is indicated empirically in a case like this.


Recommended Resources