Wednesday, January 22, 2020

This middle-aged patient presented with SOB, weakness, and mild pulmonary edema.

This middle-aged patient presented with SOB, weakness, and mild pulmonary edema.

She previously had Atrial fibrillation with LBBB.

Here is her ED ECG:
Does this reveal the etiology of her symptoms?

This shows atrial fibrillation.  There is a regular, slow response.  The fact that the response is regular proves that the atrial fibrillation is NOT conducting.  When atrial fib conducts, the ventricular rate must always be irregular.

This is atrial fibrillation with complete (3rd degree) AV block.

The QRS morphology is RBBB with a vertical axis.  This suggests an LV escape rhythm, possibly from the left anterior fascicle.

She previously had known LBBB, so she is lucky that her left sided escape can make it out of the LV through the left sided conducting system to the right side.  She could have developed an escape that is not able to use the conducting system at all; such an escape would be very wide and bizarre, with uncertain efficacy.  She could even have developed asystole.

There are 3 etiologies I always think of with bradycardia and AV block:
1. Medications -- she had been on metoprolol for 4 years at the same dose, so this is unlikely to be the etiology
2. Hyperkalemia.  Her K was normal
3. Ischemia.  There was no evidence of ischemia.

Thus, this is a sick AV node.

She had a permanent pacemaker implanted.

After pacer AND conversion to sinus rhythm:
Computer diagnosis:
What is missing from this interpretation?


Many computer algorithms do not make any attempt at atrial rhythm diagnosis when the ventricle is paced.  This is puzzling to me, because it is very important!

It is common for physicians to ignore the atrium in this situation and forget the stroke risk.

See this terrible case:

Computer often fails to diagnose atrial fibrillation in ventricular paced rhythm, and that can be catastrophic

MY Comment by KEN GRAUER, MD (1/22/2020):
Our THANKS to Dr. Smith for presenting this extremely interesting case. In the interest of academic discussion — I’ll present another perspective on selected aspects this case.
  • Pertinent information for clinical decision-making is not mentioned among the details we are provided with ... For illustrative purposes — I will “walk through” mthought process on examining the 2 ECGs in this case (Figure-1).

Figure-1: The initial ECG recorded in the ED (See text).

Clinical Points Regarding ECG #1:
We are told that the patient is a middle-aged woman — and that she previously had been in AFib with LBBB.
  • PEARL  Realizing that complete information is not always available (and it may not have been available in this case) — I always look through the old chart for the actual ECG at the time the patient was said to be in AFib with LBBB. I learned the importance of doing so through unfortunate personal experience, in which regardless of the capabilities of the provider who wrote out the ECG interpretation — many times I would catch findings not mentioned (or erroneously interpreted) in the written description. KEY  It’s always BEST for YOU to see the actual ECG before accepting a written interpretation by someone else as valid.
  • This is relevant in this case. IF this patient was previously in LBBB — and now manifests RBBB — then we have alternating BBB, which is an extremely high-risk situation for abrupt development of ventricular standstill. Permanent pacing would then be clearly needed.
  • In ECG #1 — We see a combination of baseline artifact presumed “fib waves”. Atrial activity could easily be lost within all the baseline undulations that we see. Therefore — knowing this patient was truly in AFib before would help me to be more comfortable that we are truly dealing with regularization of AFib” in ECG #1 — and not just a simple junctional escape rhythm without AFib (or even something else ...).
  • As per Dr. Smith — the rhythm in ECG #1 is regular, with the exception of 1 PVC ( = beat #5). Keep in mind that the phenomenon of regularization of AFib” (to this degree of regularity) is not common! By far (!!!) — the most common cause of this phenomenon is Digoxin Toxicity — such that it is worth searching through all of the pills that enter this patient’s mouth (We’ve all seen patients who forget some of the meds they are taking, that may not have accompanied them to the hospital). While I agree that AFib + complete AV block is the most likely rhythm diagnosis — I'd like to see additional monitoring strips to be sure ...

Assuming we are dealing with AFib + regularization of the ventricular response due to complete AV block — I do not think the site of the escape rhythm is in the ventricles ( = my opinion). I say this because: i) QRS morphology is precisely consistent with typical complete RBBB (ie, triphasic rSR’ with taller right rabbit ear and S wave that descends below the baseline in lead V1 + wide terminal S waves in lateral leads I and V6, with triphasic opposite qRS morphology in lead V6). In contrast, fasicular escape tends to manifest less typical “pure” RBBB morphology — and — fascicular escape should manifest resemblance to bifascicular block (either RBBB/LAHB or RBBB/LPHB), which is not present in this case; andii) The rate of the escape pacemaker is exactly 50/minute — which is most consistent with a junctional escape focus.
  • This is relevant — because junctional escape with RBBB (instead of postulating a ventricular site of escape— would confirm alternating BBB, which is strong indication for permanent pacing.
  • It is also relevant — because assessment of ST-T wave morphology is clearly more valid when dealing with a supraventricular QRS morphology (See below).

The question then arises as to whether this patient has SSS ( = Sick Sinus Syndrome) — and — whether sufficient investigation has been completed to confirm this diagnosis.
  • SSS is by far the most common cause of permanent pacing in the U.S. The syndrome is characterized by a long “prodrome” period (often 10-20 years) — during which a series of suggestive rhythm disorders evolve. These include sinus bradycardia & arrhythmia — sinus pauses/sinus arrest — associated “sick AV node” disease — SA block/AV block of varying degrees — slow AFib — slow AV nodal escape rhythms — and often periods of “tachyarrhythmias” (ie, “tachy-brady” syndrome), including fast AFib, ATach, reentry SVT — ultimately, with delayed SA node recovery time (leading to long pauses) after tachyarrhythmias terminate.
  • While the diagnosis of SSS may become readily apparent early on in many patients — clear indication for pacing usually does not become established until there is profound, symptomatic bradycardia — usually requiring pauses of at least ≥3.0 seconds in duration.
  • Although SSS can be seen at any age — it is by far (!!!) most commonly seen in an older population (patients at least 60, and often much older)NOTE  The patient in this case was only “middle-aged” — so SSS would not yet be expected in most patients in this age range ... — which makes one ask IF something else (ie, underlying coronary disease) was going on?
  • Finally — SSS is a diagnosis oexclusion. In addition to ruling out rate-slowing medication — serum electrolyte disorders — and/or ischemia/infarction as potential causes of bradyarrhythmias — one should also rule out hypothyroidism + sleep apnea.
Ischemic Heart Disease can be recent or longstanding. We are not told how “ischemia” has been ruled out in this case ...
  • Has this patient been cathed? Given the heart failure and markedly abnormal ECGs — I would expect cardiac cath at some point to have been performed.
  • In my opinion — there are ischemic changes on ECG #1. These are not acute changes — but I see definite ECG findings not expected in simple RBBB.
  • With typical RBBB — the ST-T wave should be oppositely directed to the last QRS deflection in leads I, V1 and V6. While the T wave in V1 is inverted as expected with RBBB — the ST segment in both V1 and V2 is not at all depressed (RED arrows). This is not what should be expected with simple RBBB ...
  • Typically, the most J-point ST depression with RBBB is seen early (ie, in leads V1, V2). Instead, we see more J-point ST depression in lead V3 (highlighted by the dark BLUE arrows) than we do in leads V1 and V2.
  • J-point ST depression continues in leads V4, V5 and V6 (light BLUE arrows).
  • The ST segment in lead I is uncharacteristically straight (straight RED lines in this lead).
Conclusion  The above are all subtle-but-real abnormal ST-T wave findings that should not be seen with simple RBBB. They clearly do not suggest acute coronary occlusion — but they are consistent with ischemia. I’d want to know more information (ie, prior tracings, more history, serial troponins, follow-up ECGs, cardiac cath results) — before confirming the need for pacing, and before ruling out potentially treatable ischemic heart disease as a cause.

Clinical Points Regarding ECG #2:
We are then shown ECG #2, in which conversion to sinus rhythm has occurred — and a permanent pacemaker has been implanted (Figure-2).

Figure-2: The 2nd ECG in this case, that was obtained some time after conversion to sinus rhythm, and after insertion of a permanent pacemaker (See text).

The following are thoughts to consider regarding the limited information we have been given in this case.
  • WHEN in the process was ECG #2 obtained? A day after ECG #1? or many days later? Clinically — What has happened in the interim?
  • HOW was this patient converted to sinus rhythm? NOTE  Given apparent long-standing AFib in this patient with severe conduction system disease — restoration of sinus rhythm is far from an easy task (ie, the longer a patient has been in AFib — the more difficult it becomes to cardiovert the patient, and then maintain sinus rhythm) = all the more reason to search for additional monitoring strips before the pacer was inserted, to verify that the initial regular rhythm (that was laden with artifact) in ECG #1 was indeed AFib ...

Regarding ECG #2:
I’m not bothered by lack of computer algorithms commenting on the underlying rhythm in pacer tracings. In my experience — the aspect of interpretation that computer algorithms do most poorly in, is interpretation of complex arrhythmias. I hold little hope that computer algorithms will provide useful information given the subtleties of complex pacer tracings.
  • NOTE  Despite conversion to sinus rhythm in ECG #2, there remain significant baseline undulations. This is relevant — because I raised the valid question in ECG #1 as to how much of the undulations we saw at that time were the result of “fib waves” (from presumed AFib with regularization of the ventricular responsevs baseline artifact. Looking at ECG #2 suggests that a large part of the undulations we saw in ECG #1 probably were artifact and not “fib waves”.
  • I believe we see an underlying sinus rhythm in ECG #2 (with spontaneous P waves) — which is followed ~0.24 second later by ventricular pacing spikes. I’ve added dark BLUE arrows at selected places where we clearly see ventricular pacing spikes. But note that we do not see these everywhere, and that the clarity of these pacing spikes sometimes varies for consecutive beats in certain leads. Nevertheless — I believe all QRS complexes (except the PVC = beat #11) are ventricular paced.
  • It is of interest that the QRS of paced beats in the long lead II rhythm strip looks narrow. A look at simultaneously-recorded other leads however, confirms that these paced ventricular complexes are indeed wide.
  • Following the PVC (beat #11) — there is a short pause. The RED arrow suggests that this pause was long enough to initiate atrial pacing of the P wave before beat #12.
  • Although I do not see any ventricular pacing spike in front of beat #12 — the fact that this QRS looks identical to all other paced beats in the long lead II tells us that beat #12 is also ventricular paced.
Conclusion — It appears that there is a DDD pacemaker that is functioning appropriately (ie, both sensing and pacing both atria and ventricles). Details could be confirmed, depending on pacer specifications (ie, rates, intervals) that were set, which we are not privvy to. 

P.S.  There is NO way any computer algorithm would be able to sort out this degree of complexity that we see in ECG #2, in which there is so much confusion generated by the baseline artifact.


  1. This is a dual chamber pacemaker, and afib requires single chamber pacing. That means the rhythm conversion happened before device implantation.

    1. Excellent point Max! This supports my quest for additional information and tracings regarding the time course and sequence of events that occurred BETWEEN the time that ECG #1 and ECG #2 was done. (P.S. I wrote My Comment [above] late yesterday — which may have been after the time that you submitted this question — :)

    2. I'm not sure what the point is (??)

    3. Just trying to piece together the sequence of what happened when ... Atrial pacing is hemodynamically superior when there is sinus rhythm. The only other ECG in this case presumably showed AFib — so a good clinical point to be aware of, is that sinus rhythm should be reestablished prior to atrial pacing.

  2. Thank you very much for this amazing case and explanation.
    I read that there was no evidence of ischemia but I wondered why there were so pronounced q waves in the second ecg in the anterior and inferior leads.
    Is this all due to ventricular pacing?
    Greetings from Germany.

    1. Flo — Vielen Dank für dein Kommentar! (Many thanks for your comment!). I don’t think Q waves in a ventricular paced rhythm are a reliable indicator of infarction (though the marked fragmentation that we do see in ECG #2 could be consistent with underlying “scar”). As I mentioned in My Comment (above) — I thought there were ECG indicators of ischemia in ECG #1 ( = my opinion). In any case — I thought more information than we are given would be needed to rule out ischemia as a potential contributing factor ( = my opinion).

  3. You people don't sleep? Sitting in front of your desk top, putting notes at 2:25 AM, 6:41 AM.... ??

    1. It's called dedication (the thought came to me ... ) — :)

  4. Amazing post, thanks a lot.I would like to know if are there flutter waves in V1-V2, and the pattern S1Q3T3?

    1. Thanks for the kind words. On occasion, AFib may “organize” for a short period of time — and when it does, it may look like the atrial activity that you see in ECG #1 in leads V1 and V2. But this is NOT AFlutter because: i) consistently regular atrial activity is NOT seen; ii) the rate is FASTER than 1 large box in duration (ie, >300/minute) — which is not too fast to completely rule out flutter, but which makes flutter LESS likely; iii) there is a LOT of baseline ARTIFACT throughout the rest of this tracing — so it may well be that these deflections we see in V1 and V2 do not represent atrial activity at all; and iv) the ventricular rate is "off" for what I'd typically expect with AFlutter (usually there is 2:1 or 4:1 atrial activity). Otherwise — there is complete RBBB — which BY DEFINITION gives you an S wave in lead I. It also often gives you a q and T inversion in lead III. So in the absence of anything else in the history or this ECG that suggests pulmonary embolism — I do not ( = my personal preference) even mention an “S1Q3T3” in tracings like this in which there is complete RBBB — :)

  5. very cool thoughts. thank you Steve and Ken


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