Wednesday, April 15, 2020

60-something with wide complex tachycardia: from where does the rhythm originate?



An elderly woman with history of coronary disease presented with CP and SOB and hypotension by EMS.  EMS had attempted adenosine x 2 without success.

Here is her ED ECG:

Here is the ED physician's interpretation:
IMPRESSION
UNCERTAIN REGULAR RHYTHM, wide complex tachycardia, likely p-waves.
LEFT BUNDLE BRANCH BLOCK  [120+ ms QRS DURATION, 80+ ms Q/S IN V1/V2, 85+ ms R IN I/aVL/V5/V6]
Comparison Summary: LBBB and tachycardia are new.

What do you think?













Smith:  This is indeed a regular wide complex tachycardia.  I do not see P-waves.  Retrograde P-waves after the R-wave are possible (see lead II across the bottom).  This is clearly ventricular tachycardia.  It has a left bundle branch block morphology (small r-wave in V1 with deep S-wave), but the r-wave is wide and the RS interval is almost 100 ms.  LBBB should have a narrow r-wave and the RS interval should be no longer than 60-70 ms, unless there is also some other entity which widens the QRS, such as hyperkalemia or Na channel blocking medications.  


That the VT is "LBBB-type" indicates that it originates in the right ventricle.  Since all inferior QRS axes are negative (going away from the inferior wall) and positive in I and aVL, (going towards I and aVL), the point of origin can be predicted to be in the inferior part of the right ventricle or the inferior and right side of the septum.  

Course

The medics had already tried adenosine x 2, which is safe and not contraindicated.  But, predictably, it did not have any effect.

The patient was given a small dose of etomidate and electrically cardioverted in the ED.

Here is the post-cardioversion ECG:

There is sinus with RBBB
There are inferior Q-waves suggesting old inferior MI.  

Inferior MI results in scar tissue which is a likely source of a re-entrant ventricular dysrhythmia.
This would be the likely source of the VT.



Angiogram:

Severe coronary artery calcification
Moderate to severe distal small vessel disease mainly seen in RPL1, 2
Otherwise, Mild plaque, no angiographically significant obstructive coronary artery disease.


Note of electrophysiologist

PRESENTING EKG: Wide complex tachycardia with atypical left bundle branch block, slurred deflection in AVR and V6 at onset of QRS to the deflection of 100 ms, fusion beats noted in aVR on EKG reading station, consistent with ventricular tachycardia with inferoseptal origin.

Tele Monitor:  Normal sinus rhythm throughout, no ectopic atrial or ventricular beats.

Echocardiogram:  
Estimated left ventricular ejection fraction, lower limits of normal; 45-50%.
Regional wall motion abnormality-inferior/inferoseptum: akinetic.
Regional wall motion abnormality-inferolateral.
Regional wall motion abnormality-apex, probable.
Evidence for dense scar involving the inferior and inferoseptal walls

Impression
Wide complex tachycardia in the setting of structural heart disease, known inferior and inferoseptal dense scar from previous MI, with symptoms of palpitation and diaphoresis, morphology based on Brugada and Vereckei for aVR most consistent with ventricular tachycardia origin inferoseptal region. 

However, due to LBBB with some typical feature, difficult to rule out SVT completely. 

EP study:
--SVT could not be induced
--LBBB aberrancy could not be induced
--Therefore, it is ventricular tachycardia
--ICD placed




===================================
MY Comment by KEN GRAUER, MD (4/15/2020):
===================================
The patient in this case presented with a regular WCT (Wide-Complex Tachycardia). As always — it’s nice when we have “the Answer”, here in the form of an EP study documenting the absence of any SVT — with confirmation that the rhythm is VT. That said, because of uncertainty in the initial ED diagnosis (with apparent strong consideration of an SVT etiology + thinking that P waves were present) — I’ll add to the points brought out by Dr. Smith’s comments, and from data in the EP cardiologist’s report. 
  • For clarity — I’ve reproduced and labeled the initial ECG with the post-conversion tracing below it in Figure-1.

Figure-1: The 2 ECGs in this case (See text).



MTHOUGHTS: I have previously reviewed my Systematic Approach to Rhythm Interpretation (See My Comment in the October 16, 2019 post).
  • There are KEY parameters to consider in systematic assessment of any arrhythmia. The 1st (and most important) parameter — is whether or not the patient is hemodynamically stable. Although not expressly stated in the presentation here — I’ll assume that despite palpitations that awakened this nursing home patient diaphoresis and some shortness of breath — that the patient was still hemodynamically stable — since Adenosine was given by EMS.
  • While specific criteria to assess hemodynamic stability include the presence (or absence) of arrhythmia-related hypotension, chest pain, shortness of breath, and/or mental status changes — an overall “Gestalt” assessment from “being there” (ie, at the patient’s bedside) needs to be consciously made. This is because IF the patient is hemodynamically unstable as a direct result of the tachyarrhythmia — then immediate cardioversion will be indicated regardless of whether the rhythm is an SVT or VT.
  • Once you ensure that the patient is hemodynamically stable — the other 5 parameters to assess can be easily remembered by the saying, “Watch Your Ps and Qs, and the 3 Rs” (ie, presence of P waves — QRS width — and the Rate + Regularity of the rhythm + determination if P waves are Related to neighboring QRS complexes).
  • It does not matter in what sequence these 5 parameters are looked at — as long as each of the 5 are looked at! This is relevant to this case — because there is no mention above as to what the rate of this WCT is — and no mention in the initial ED interpretation of the possible relation that suspected P waves might have to neighboring QRS complexes.
NOTE: We define the rhythm in ECG #1 as a regular WCT — because the rate is rapid, and QRS complexes are both regular and wide.
  • Looking closer — the QRS complex is very wide. Measurement of any interval should be made in that lead where you clearly see onset and offset of the interval — and, in which the interval is longest. When measuring — I look for a QRS complex that either begins or ends on a heavy line. In ECG #1 — I chose to use lead V1. Vertical PINK lines show the last QRS complex in this lead to begin on a heavy line — and end on the next heavy line. Thus, the QRS complex in ECG #1 measures 0.20 second (which turns out to be significantly wider than the wide QRS in the post-conversion tracing). A “soft” criterion in favor of VT — is the presence of an extremely wide QRS complex, as we see in ECG #1.
  • The QRS complex in the long lead II rhythm strip of ECG #1 is precisely regular. Knowing the rhythm is precisely regular rules out any possibility of atrial fibrillation — and facilitates calculation of heart rate.
  • When the rhythm is fast and regular — the Every-Other (or Every-Third) Beat Method allows rapid and accurate rate estimation. Find a QRS complex that begins on a heavy line. In ECG #1 — I chose the 2nd beat in lead II (See the 1st vertical RED line under this beat). Using the Every-Third Beat Method — the amount of time that it takes to record 3 beats (RED numbers in lead II) is just under 5 large boxes (BLUE numbers in this Figure). Therefore — ONE THIRD the rate is a little faster than 300/5 ~60-65/minute.
  • The actual rate for the rhythm in ECG #1 is therefore ~60-65 X 3 ~185-190/minute. Knowing the actual rate of the WCT in ECG #1 is relevant to this case — since while not impossible, sinus tachycardia is extremely unlikely at this fast of a rate. (In my experience — sinus tach rarely exceeds 170/minute in a non-exercising adult patient.)
  • CLICK HERE if interested in brief video review of this Every-Other-Beat Method.
NOTE: Mention was made in the initial ED interpretation of ECG #1 of “likely P waves”. Neither myself nor Dr.Smith thought P waves were present in ECG #1 — but it is often difficult to determine with certainty if atrial activity is present in a wide-complex rhythm as fast as we see here.
  • I suspect the ED provider thought the rounded deflections at the onset of the ST segment in each of the inferior leads represented atrial activity (RED arrowheads). These positive, rounded deflections correspond temporally to the negative deflection we see just after the QRS complex in lead aVL (highlighted by the thin purple double-arrow). But IF these deflections were P waves — then they are not retrograde P waves, because they are upright (instead of negative) in the inferior leads.
  • IF instead, these rounded, positive deflections in the inferior leads are forward-conducting P waves — this would make the rhythm sinus tachycardia. But that is highly unlikely because: i) Sinus tach rarely occurs at 185-190/minute! — andii) These deflections are huge in leads III and aVF. Therefore — I didn’t think P waves were present in ECG #1. Instead, I thought the rounded inferior lead deflections and the negative deflection in lead aVL represented the onset of the ST segment after the wide QRS.
In summary — this leaves us with the abrupt onset of a regular WCT rhythm at ~185-190/minute (that woke this nursing home patient up from a sound sleep) — without any clear sign of atrial activity.
  • PEARL #1 — Remember statistical likelihoods. The literature suggests that over 80% of new-onset regular WCT rhythms without clear sign of atrial activity are VT. Even before you look at the ECG — this statistic increases to over 90% if the patient in question is older and has underlying heart disease. Therefore, our mindset always needs to be to Assume VT until proven otherwise — and treat the patient accordingly.
  • The principal differential diagnosis of a regular WCT rhythm without clear sign of atrial activity is VT vs SVT. Remember that SVT rhythms may present with a wide QRS complex not only because of aberrant conduction — but also because of preexisting bundle branch block. Less commonly, other causes of a regular WCT rhythm might include: i) AVRT, with anterograde conduction down the AP (accessory pathway) producing a regular wide SVT rhythm; and/orii) Some toxicity (ie, tricyclic overdose or other sodium channel blocker toxicity, hyperkalemia, etc.).
  • Regarding the regular WCT rhythm in ECG #1 — QRS morphology superficially resembles LBBB (Left Bundle Branch Block) — because the QRS complex manifests a monophasic (all upright) QRS complex in left-sided leads I and V6 — and a predominantly negative QRS in the anterior leads.

PEARL #2: Despite the above noted superficial resemblance to LBBB — QRS morphology in ECG #1 is not “typical” for LBBB!
  • With typical LBBB morphology — there is either no initial positive deflection (r wave) in the anterior leads — or — the initial positive deflection in anterior leads is very narrow. Instead, BLUE arrows in leads V1, V2 and V3 of ECG #1 highlight a surprisingly wide initial r wave deflection! Starting with a statistical ~90% likelihood that the regular WCT rhythm in ECG #1 of this nursing home patient is VT — the finding of atypical QRS morphology (ie, that does not resemble any form of BBB or hemiblock) increases the likelihood that this rhythm is VT to over 95%.
  • NOTE — an over 95% likelihood of VT is not 100%. Exceptions are always possible (ie, the patient may have had an extremely abnormal baseline ECG — which we could not know unless we had access to a prior tracing). That said, there is more than enough convincing evidence to assume that the rhythm in ECG #1 is VT (and to treat accordinglyuntil we prove otherwise.
  • CLICK HERE if interested in more on “My Take” regarding ECG assessment of the regular WCT.

This patient was electrically cardioverted in the ED. The post-conversion tracing is shown in Figure-1 as ECG #2.
  • How would you interpret the 12-lead tracing shown in ECG #2?
  • Does ECG #2 support or refute our conclusion that the regular WCT in ECG #1 was VT?



ANSWER: Normal sinus rhythm has been restored in ECG #2. The QRS complex is wide — though as noted earlier, not as wide as it was during the WCT.
  • QRS morphology in ECG #2 is consistent with RBBB because: i) The QRS complex is wide (I measure a QRS duration ~0.14 second); ii) There is an rsR’ complex in right-sided lead V1; and, iii) There are terminal S waves in left-sided leads I and V6.
  • There also appears to be LPHB (Left Posterior HemiBlock) — as the straight downward deflection of the S wave in lead I is predominantly negative + there are qR complexes in the inferior leads. The presence of RBBB/LPHB qualifies as bifascicular block — and is a marker of significant underlying heart disease.
  • There are inferior Q waves, that especially in leads III and aVF are deeper and wider than normally expected. Considering the modest QRS amplitude of the complexes in leads V5 and V6 — the Q waves in these leads are also larger-than-expected. These findings are consistent with prior infero-lateral infarction.
  • There is marked fragmentation in multiple leads. This is best seen in leads V1 (the double notch in the R’ wave) — V2 (marked notching at the nadir of the S wave) — and in leads V3 and V4 (notching on the upstroke and downstroke of the R wave, respectively). The presence of marked fragmentation in ECG #2, in association with bifascicular block + infarction Q waves further solidifies the conclusion that this patient has severe underlying heart disease. Knowing this, and seeing how different QRS morphology is in ECG #2 (compared to what it was during the regular WCT rhythm in ECG #1) — adds virtual certainty that the WCT in ECG #1 was VT.
  • IF interested in more on the topic of fragmentation — See My Comment in the January 31, 2020 post.





22 comments:

  1. Great case as usual!

    One question and one observation:

    -In the first ECG, isn't J point a little bit too "high" in the inferior leads and too "low" in the lateral leads, even in the presence of LBBB? Could this represent demand ischemia due to a very fast heart rate?
    -In the second ECG, of course I agree about QRS fragmentation, but honestly Q waves didn't look that abnormal to me on first sight. I agree about duration which is over 40 msec, but they don't look too deep compared to QRS size.. I would have easily mistaken them for "appropriate" Q waves in the setting of LPFH.

    Thanks for sharing!

    ReplyDelete
    Replies
    1. @ Anonymous — Thanks for your comment. Astute observation you made (!) about the “too high” J-point in ECG #1 (Please see insightful Comment by K.Wang below + My reply to his comment). Regarding ECG #2 — in my opinion, the Q waves in leads II and V5 are of questionable duration — but the Q waves in leads III, aVF and definitely in lead V6 are wider-than-I-think-they-shoud-be in a patient who has not had previous infarction. Of course, Q wave size and width is not a perfect criterion for assessing the presence of acute or previous infarction — and some of this assessment is indeed subjective. But to my measurement — Q wave width in lead III of ECG #2 is at least 1 little box in duration (ie, ≥0.04 second), which is “too” wide for me — and proportionately, there is NO way in my opinion that a “normal septal Q wave” in lead V6 should be as wide as this one is given the small amplitude of the R wave in this lead. THANKS again for your excellent comments! — :)

      Delete
  2. Nice case Dr Smith and comments Dr Grauer...
    It hits indeed a number of points in Brugada algorithm...
    Also if one only scrutinised aVR lead, then there is initial negative QRS deflection, i.e. QW which is >40ms (it's close to 80ms) as well as notching of the initial downstroke of negative QRS...
    Amal Mattu in one of his talks mentioned think of max HR... its 220 minus your age if it is more than that, then it is not SR
    And then of course age with Hx of CAD...
    Interesting to read EP's interpretation too...
    Good innings...
    OR

    ReplyDelete
    Replies
    1. @ Anonymous — Thanks for your comment. I know there are excellent clinicians who depend entirely on lead aVR for distinguishing VT vs SVT in a regular WCT. I always look at lead aVR in regular WCT rhythms — because if the QRS complex is ENTIRELY positive (no negativity at all!) — then specificity for VT approaches 100% — because this means the electrical impulse must be arising from the apex, and SVT’s don’t do that. That said, in the absence of an entirely positive QRS complex in lead aVR — I’ve not found this lead overly helpful in differentiating WCT rhythms because of difficulty I often have in confidently determining precise QRS morphology (ie, is there a QS complex in ECG #1 in this case — or are we seeing a tiny initial r wave in this lead?). I realize others may disagree with my opinion about the utility of lead aVR with WCT rhythms — and that’s fine — :)

      Otherwise — the “220 minus Age” rule is derived from exercise testing (ETT), and provides a quick estimate as to the expected maximal heart rate to be attained during FULL EXERCISE. When doing ETT in the office — I always tried (if possible) to attain at least 85% of a patient’s maximal predicted heart rate — since not achieving this heart rate goal meant reduced sensitivity and specificity for your results. That said — the “220 minus Age” does NOT apply to usual rate limits for a “horizontal patient” (ie, a patient who has not just performed maximal exercise). In my experience — it is unusual for sinus tachycardia to attain heart rates >170/minute for a “horizontal adult”. Please NOTE — I did NOT say impossible, as we all have seen occasional adult patients with sinus tachycardia at heart rates greater than this — but that’s not the usual. Please also NOTE that these limits do not hold true for pediatric patients — for whom sinus tach on occasion. may attain rate up to 200/minute or more. BOTTOM LINE — The above explanation is why I said the heart rate of ~185-190/minute in ECG #1 of this case was extremely unlikely to be sinus tach. THANKS again for your comment! — :)

      Delete
  3. The infero-lateral STEMI could have been told already from the QRSs in leads II, III, aVF and V5 during the VT (note the red arrow-heads by Dr. Ken Grauer in the ECG #1 above). Yes, QRSs originating from the ventricle, such as QRSs of PVC, accelerated idioventricular rhythm, VT or ventricularly paced rhythm, can reveal STEMI very effectively. So we should not say "it's no use looking at the ECG" because the patient is in VT".
    K. Wang.

    ReplyDelete
    Replies
    1. SSmith Blog
      ==========

      THANKS for your comment K! Excellent point you make (and which we have made a number of times in previous blog posts in Dr. Smith’s ECG Blog). It apparently turned out in this case that there was no acute occlusion on cath — and there were no acute ECG changes on the post-conversion 12-lead … Although marked tachycardia sometimes results in ST elevation that resolves after rhythm conversion — I would not have expected THIS MUCH elevation of ST segments (with this much reciprocal ST depression in leads I and aVL) without there being acute occlusion — but as best I can tell from the information provided, there was no acute OMI in this case. But I should have mentioned this in my comment — so THANK YOU for pointing it out! I will pass on your comment to Dr. Smith to see if he has anything more to add — :)

      Delete
    2. Agree this looks like OMI in VT, but it was not.

      Delete
  4. Thanks Dr S & KG..why not its Atrial Flutter with underlying RBBB?

    ReplyDelete
    Replies
    1. @ ECG Life — I assume you are asking about ECG #1, which is the initial tracing in the ED, which shows tachycardia. The ventricular rate of ~185-190/minute is too fast for 2:1 AFlutter (would mean the atria would have to be ~370-380/minute, which is much faster than AFlutter typically is). If you are thinking those humps (under my RED arrows) are extra atrial waves — they are not — because if you use calipers, these do not march out 2:1 with any other deflection in the inferior leads. This is not AFlutter. QRS morphology (as I describe above in My Comment) superficially resembles LBBB and not RBBB — but as I mention, QRS morphology actually has very atypical features for lbbb. Therefore, this is VT (and that diagnosis was supported by EP study).

      Delete
  5. Gracias por compartir su conocimiento. Excelente Blog.

    ReplyDelete
    Replies
    1. Gracias a ti for tu interĂ©s! (Thanks to you for your interest — :)

      Delete
  6. Was possible to predict that arrhythmia?

    EF 50% was good !

    ReplyDelete
    Replies
    1. @ Felipe — 50% is a surprisingly good EF for this patient — but that in no way rules out the possibility of VT. As per My Comment above — there are many indicators here of severe underlying heart disease in this elderly woman — so statistical odds that a regular WCT rhythm will turn out to be VT in this patient (even BEFORE you look at the actual ECG!) are >90% (as I describe above). THANKS for your comment — :)

      Delete
  7. Thank you for the excellent description of the method to differentiate VT from other entities. Very clear and instructive treatise of a topic usually considered somehow "esoteric" by cardiac surgeons like me!

    Aldo

    ReplyDelete
  8. Great comments, Ken. Just by looking at V1 and V2, I would say that there is way too much daylight between the ascending and descending limbs of those r waves. That is just too much width for a LBBB which always begins normally.

    ReplyDelete
  9. Hello sir, How can I send a ecg for clarification of findings? Your email id please.

    ReplyDelete
    Replies
    1. @ Ahammed — For questions or comments regarding the ECG(s) in this case — simply write your comment/question here. For submission to Dr. Smith of ECGs of potential interest for a new case on our Blog — You'll note in the RIGHT column, near the top — the "Contact for Interesting ECGs" — which when clicked sends an email to Dr. Smith. Please BE SURE any tracings you send provide pertinent clinical details, with ECGs that are upright and of high resolution. Thank you — :)

      Delete
    2. I'm sorry I have been so busy I have not been able to answer these emails in a long time.

      Delete

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.