Differential diagnosis of wide complex tachycardias - Part I

There seems to be a lot of confusion with regard to wide complex tachycardias.

For some reason, health care providers of all ranks and stripes forget the most basic rule.

If it's a wide complex rhythm (fast or slow) it's ventricular until proven otherwise!

Reasonable people can disagree as to what constitutes "proof" but you need to start somewhere, and the place to start is not "it's SVT with aberrancy (or a bundle branch block) until proven otherwise."

That's how patients get killed.

A little bit of knowledge can be a dangerous thing, and I'm not a huge fan of Wellens' Criteria or Brugada's Criteria. In the wrongs hands, these algorithms do more harm than good.

When they "rule in" VT, they're fine. When they (wrongly) suggest SVT, they're dangerous.

There are several reasons these algorithms might wrongly suggest SVT. No algorithm is perfect. In addition, mistakes are made when using the algorithms, and the algorithms are sometimes used in ways they were not designed to be used.

For example, with modified leads MCL-1 and MCL-6.

Wellens' Critera and Brugada's Criteria are not the same, and the former is part of the latter, so I don't mean to oversimplify here. But in my experience, a far more valuable tool is to recognize the typical patterns. In other words, typical RBBB, typical LBBB, typical bifascicular block (RBBB/LAFB), and typical bifascicular block (RBBB/LPFB).

Having said that, just because the ECG shows a typical pattern, that doesn't mean the ECG shows SVT!

VT can mimic virtually all typical patterns on the 12 lead ECG.

I remember when I was a cardiac monitoring technician on the Critical Care Stepdown (CCSD) unit of a medium-sized (about 400 beds) community hospital with 2 cath labs and an open heart program on the west side of Cleveland, Ohio.

Our unit could handle 32 patients, and we were always full. That translates into 4 monitors, each with 8 different heart rhythms, and it was my job to watch them.

We used Zymed's EASI lead system, which allowed us to view a "derived" 12 lead ECG from 4 modified electrodes (3 leads: Channel 1, Channel 2 and Channel 3).

The technology has since been licensed to Philips Healthcare.

The system is ideal for stepdown units (where hooking a patient up to continuous 12 lead ECG monitoring isn't practical), holter monitors (where the morphology of an arrhythmia might help the cardiologist), NASA astronauts, and so on.

Much of what I know about 12 lead ECGs stems from my experience comparing the "derived" 12 lead ECG to the standard 12 lead ECG in the patient's chart, and asking cardiologists lots of questions.

The EASI lead system has many advantages, but it has a couple of drawbacks. It is very dependent upon proper lead placement. I mean no disrespect to my nurse colleagues (they were very competent and I enjoyed working with them), so let's just say their self esteem did not depend on lead placement precision.

The other main drawback (for a stepdown unit) is simple. Where is the surgical wound for an open heart patient? Right down the middle of the sternum. Where do you have to place the electrodes for Channel 1 in the EASI lead system? On the top and bottom of the sternum. Since that's not possible for an open heart patient, you have to 1.) place the electrodes off to the side (which is what we did 99.9% of the time), 2.) use standard lead placement, or 3.) use some type of modified lead placement.

The whole advantage of the EASI lead system is that it gives you the opportunity to view arrhythmias in 12 leads. It's a lot easier to figure out what those FLBs (funny little beats) are when you can look at the limb leads and the precordial leads. So what's the next best thing?

If we had a set of leads that allowed the limb leads plus one precordial lead, that would have been a good alternative. But since we didn't, I decided to place a post-CABG patient in MCL-1 and MCL-6.

I had read (and this was validated by a discussion with a cardiologist) that you could differentiate wide complex tachycardias using these leads, since they mimic leads V1 and V6, and those are the relevant leads for Wellen's Criteria.

Wouldn't you know it? The patient cooperated and experienced a wide complex tachycardia!

The arrhythmia lasted about 12 seconds. The patient was totally asymptomatic.

I was thrilled! To an ECG dork, the opportunity to view a brand new wide complex tachycardia in MCL-1 and MCL-6 is quite a treat.

It didn't take me long to decide that the "right bunny ear" was taller than the "left bunny ear" in lead MCL-1 (surrogate for lead V1). Yup, this was typical RBBB. SVT with aberrancy all the way! The cardiologist agreed, and wrote down in the chart that the patient experienced a short run of atrial fibrillation with "rate activated RBBB".

We were both wrong.

No, the patient didn't die, or experience an adverse outcome. At least, I don't think s/he did. But then again, I don't know for sure. It wasn't until years later that I realized this ECG shows VT.

Let me prove it to you.


Take a good look at the circled QRS complexes. I didn't notice it right away (okay, until years later) but they are wider than the previous QRS complexes. There is also a change in amplitde.

There's something different about these QRS complexes, all right.

They are fusion complexes. In other words, they are hybrid QRS complexes. A "fusion" QRS complex is the result of ventricular depolarization happening through two different mechanisms at the same time. In this case, "normal" ventricular depolarization through the AV node/His/Purkinje system, and spontaneous ventricular depolarization from a PVC (most likely triggering the run of VT that follows).

Still aren't sure?

Look at the PR interval of the underlying rhythm, and the PR interval of the fusion complex. The PR interval of the fusion complex is slightly shorter (thank you Tom G.).

If this evidence isn't compelling enough (and I think it is), look at the following image.


If you march out the P waves, AV dissociation is present.

Point, game and match.

I know what you're thinking. Brugada's Criteria would have caught this run of VT! Sure. If you are skilled enough to identify AV dissociation at the point of patient contact and decision making.

My point is simple. Your default diagnosis should be VT anyway! In my opinion, QRS morphology is insufficient evidence that a wide complex tachycardia is SVT with aberrancy.

That's all I wanted to establish for Part I. :)

See also:

Differential diagnosis of wide complex tachycardias - Part II

Differential diagnosis of wide complex tachycardias - Part III

Differential diagnosis of wide complex tachycardias - Part IV

Differential diagnosis of wide complex tachycardias - Part V

Differential diagnosis of wide complex tachycardias - Part VI

*** Update 03/01/09 ***

I recently discovered this article from the March 2006 issue of Emergency Medical Services. In it, the author states:

With the introduction of new pharmacological interventions that target specific areas of the cardiac conduction system, it has become increasingly important for EMS providers to make an accurate interpretation of an ECG. Though most paramedics have no difficulty distinguishing VT from narrow complex supraventricular tachycardia (SVT), some might fall victim to the "wide + fast = VT" trap when looking at SVT with aberrant conduction. Although VT and SVT with aberrant conduction look similar, they vary greatly in terms of origin, pathophysiology and treatment. Mislabeling dysrhythmias can have severe consequences. Improper identification of VT could place a patient in grave danger by delaying indicated pharmacological and electrical interventions.

Listen carefully.

"Wide and fast = VT" is not a trap! It's a rule of thumb that exists to protect you and your patient!

The author continues:

A common aphorism among advanced practitioners is, "When in doubt whether a WCT is VT or SVT, treat patients as if they are experiencing VTs." This stems from a statistic showing that approximately 80% of all WCTs are VT. Though this aphorism is generally a good rule of thumb, it is also important to acknowledge that one in five WCTs is not VT and therefore requires different treatment regimens. One must possess the proper diagnostic tools and knowledge to decide whether a WCT is VT or SVT with aberrant conduction. EMS providers should be able to differentiate VT and SVT with aberrant conduction with confidence and a high degree of certainty.

They do not require different treatment regimens!

Unstable SVT is treated the same as unstable VT! Anyone disagree?

It's debatable whether or not a stable wide complex tachycardia should be treated in the field at all, but if you do reach for an antiarrhythmic, it better be one that works for SVT and VT (i.e., amiodarone or procainamide).

If you give a calcium channel blocker to a wide complex tachycardia without knowing with 100% certainty that it's SVT with BBB (or aberrancy) you are a fool.

The article also contains outright errors. Here's one of the most disturbing.

SVTs with aberrancy will produce either a right or left axis deviation. If the aberrancy is conducted in a RBBB pattern, right axis deviation will be present. If the aberrancy is conducted in a LBBB pattern, left axis deviation will be present. In almost all VT, the axis will be in the extreme right quadrant.

This is nonsense!

In the first place, RBBB and LBBB aberrancy can both show a normal axis. RBBB aberrancy in particular can show a normal axis, right axis deviation, or left axis deviation (bifascicular patterns).

Most cases of VT present with an other-than-extreme axis.

See what I mean about a little bit of knowledge being dangerous?

*** End update ***

Politics vs. Patient Care

I'm always interested in free, high quality medical podcasts. So hat tip to the Centeral Mass Medics blog for turning me on to the Surgery ICU Rounds podcast.

I enjoyed listening to Jeffrey Guy M.D.'s podcast on Therapeutic Hypothermia for Cardiac Arrest, in which he brings up a recent article in the New York Times.

According to the article, starting January 1, 2009, New York City ambulances will take many cardiac arrest patients only to hospitals capable of inducing therapeutic hypothermia.

The article states:

It amounts to an endorsement by the Bloomberg administration of a labor-intensive, often expensive and still-developing therapy that smaller community hospitals say they lack the staffing and financial wherewithal to provide.

Some hospital officials fear that the policy could be unfair to these smaller hospitals, depriving them of income from emergency-room patients and hurting their reputations with the public.

Dr. Jeffrey Guy is quick to point out that 1.) the idea of therapeutic hypothermia was not Mayor Bloomberg's, 2.) the scientific evidence shows that patients with out-of-hospital VF arrest who receive therapeutic hypothermia in a timely fashion have better neurological outcomes and better survival, and 3.) therapeutic hypothermia is not new! It's been recommended by ILCOR for over 5 years!

Money quote:

"Decisions regarding outcomes and patient therapy ... should be mostly based on patient survival and patient outcomes, and the idea that we're going to be depriving smaller hospitals of potential income and hurting their reputations and that could be 'unfair' is irrelevant."

No kidding!

This is an excellent example of EMS using the power of the transport decision to effect positive change.

The fact that some hospitals are either unwilling or unable to provide the appropriate care to cardiac patients is not the patients' problem.

It's the hospitals' problem.

This is similar to local community hospitals losing revenue when they are bypassed for hospitals capable of performing primary PCI.

I'm sorry they're losing revenue, especially if they are unfairly punished by Medicare, but it's not the patient's problem.

In a perfect world, patient care would always trump politics.

See also:

Treatment Guidelines Lead to Four-fold Increase in Survival Rate for Cardiac Arrest

Merry Christmas!

Here's wishing you and your family a very Merry Christmas!



And for those of you who hate the holidays (I'm talking to you, Dave)....

"Funky Trouble Looking" RBBB with AMI

You might remember from my intro that my inspiration for starting the Prehospital 12 Lead ECG blog was the Capnography for Paramedics blog (which is no longer active but still an outstanding educational resource).

You might also recall from my Superlative Hieroglyphicist Award that author of the Capnography for Paramedics blog Peter Canning keeps an active blog called Street Watch: Notes of a Paramedic.

Of course, you probably didn't need me to tell you that, because he's been around the blogosphere a lot longer than I have! :)

Regardless, Peter posted a most interesting case study today called "Funky Troubling Looking" -- Right Bundle Branch Block and MI that is similar to the case I posted on 11/29/08.

It also dovetails nicely with my recent posts on Sgarbossa's Criteria.

While Sgarbossa's Criteria is generally used to identify AMI in the presence of LBBB or paced rhythm, I also discussed the "rule of appropriate T wave discordance" which can be applied to RBBB if it is well understood.

In other words, if you think in terms of the terminal deflection of the QRS complex, and not the main deflection.

*** UPDATE 12/21/08 ***

Here's an interesting case sent to me by Thomas Bernesser of Mint Hill Fire and Rescue.

50 year old male with an acute onset of sub-sternal chest pain and dyspnea. He states he was at rest when the sharp, tight feeling began in the center of his chest and radiated to his left side. A 10 on 10 with the pain. He also remarked of left arm pain and tingling. He c/o of associated nausea and diaphoresis along with the pain and dyspnea. He had taken two of his own SL NTGs prior to EMS arrival without any relief. He had a past history of multiple MIs with stent placement, the last being in October of this year.

He's pale, diaphoretic, clutching his chest, looks very uncomfortable and anxious. His initial BP was 92/50, a little tachy between 100-110 and a respiratory rate at about 18/minute.

And on to the EKGs, I ran numerous serial EKGs and they were all identical - including the ones done in the ED at the receiving facility. Just based on his presentation alone I was convinced that he was having a significant cardiac event, but the EKG was a little less convincing for me. Each one of the interpretations made reference to ST elevation in the inferior leads and diagnosed an Acute MI. While I do note elevated segments, the EKG just strikes me as odd. I'm not seeing the classic ST elevation above baseline coming off the S wave of the complex. I might be missing something but I found it interesting nonetheless and wanted to share it with you.

Just so you know, he was worked up in the ER for an MI and sent to the cath lab. He was cathed and came back clean, no signs of occlusion. So, definitely a false MI identification for the Philips MRx.

There's no perfect solution for patients with baseline abnormalities on the 12 lead ECG. One interesting point for this case is the absence of a definite TP segment as a baseline for ST segment measurement due to the sinus tachycardia.

Normally, we don't think of RBBB as distorting the ST segment in a discordant direction (which is why LBBB is so problematic). However, both of these cases demonstrate that RBBB can can be challenging when evaluating a chest pain patient.

One final note. Mr. Bernesser advised me that this ECG also fooled the computer at the hospital, which I'm assuming used the GE-Marquette 12SL interpretive algorithm, so it wasn't just the Philips monitor that gave the >>>> ACUTE MI <<<< message.

See also:

The problem of ST segment elevation

False positive cardiac cath lab activations

Identifying AMI in the presence of LBBB - Sgarbossa's Criteria - Part I

Identifying AMI in the presence of LBBB - Sgarbossa's Criteria - Part II

Contiguous and reciprocal lead charts

Here are some charts to help you identify and localize acute STEMI on the 12 lead ECG.

Contiguous leads

What do we mean when we say leads are "contiguous"?

Contiguous leads are "next" to one another anatomically speaking. They view the same general area of the heart (specifically the left ventricle).

For example, these states in the upper-midwest are contiguous, because they are all touching and in the same region of the country.

The "inferior" leads (II, III and aVF) view the inferior wall of the left ventricle. Remember that the inerior leads make up the lower-left corner of the 12 lead ECG.

The "septal" leads (V1 and V2) view the septal wall of the left ventricle. They are sometimes grouped together with the anterior leads.

The "anterior" leads (V3 and V4) view the anterior wall of the left ventricle.

The "lateral" leads (I, aVL, V5 and V6) view the lateral wall of the left ventricle. Leads I and aVL are sometimes referred to as the "high lateral" leads, because their positive electrode is on the left shoulder. Leads V5 and V6 are sometimes referred to as the "low lateral" leads because their positive electrodes are on the lateral left chest.

In addition, any two precordial leads that a next to one another are contiguous. In other words, V4 and V5 are contiguous, even though V4 is an anterior lead and V5 is a lateral lead. This makes sense when you consider that leads V4 and V5 are next to each other on the patient's chest.

It's worth mentioning that the standard 12 lead ECG does a relatively poor job examining the lateral wall of the left ventricle, and does not directly examine the posterior wall of the left ventricle. That's the reason we sometimes miss acute STEMI in the distribution of the circumflex artery.

This image from Rescue One EMS Prehospital Program © 1999 Centric Medical Communications, Inc. illustrates the point nicely. This was from a class sponsored by Centocor (makers of the drug Retavase) that was taught by a Miami-Dade Fire Captain. In case you weren't aware, Miami-Dade was the largest enroller in ER-TIMI-19 which was a clinical trial involving prehospital administration of thrombolytic therapy.

Think of it this way. There are 3 main epicardial coronary arteries, the right coronary artery (RCA), left anterior descending (LAD) and the circumflex (LCX).

It stands to reason that approximately 33% of documented acute STEMIs should occur in the distribution of each of the 3 main arteries. But that's not what we find. Most acute STEMIs are documented in the distribution of the right coronary artery or the left anterior descending.

In other words, the standard 12 lead ECG does a relatively poor job examining the lateral and posterior walls of the left ventricle, so there's a danger of missing STEMI in the distribution of the circumflex artery.

That's the main reason it's so important to carefully analyze the right precordial leads (V1-V3) for reciprocal changes that may indicate posterior STEMI. You can also consider using modified leads V7, V8 and V9 to increase the sensitivity.

Right ventricular infarction is another issue that will have to be addressed another time.

Reciprocal leads


What do we mean when we say that a lead is reciprocal? It means that during an acute STEMI, when ST segment elevation is present in leads that face the acute injury, ST segment depression will often be present in leads that face the "ischemic boundary".

Many theories have been advanced to help explain reciprocal changes. I can't go into all of them here, but consider this diagram modified from A Mechanism for ST Depression Associated with Contiguous Subendocardial Ischemia by Bruce Hopenfeld. Jeroen Stinstra, and Rob MacLeod. J. Cardiovasc. Electrophys, 15(10), 1200--1206, 2004.


Computer modeling has shown that as the ischemic zone extends from the endocardium to the epicardium, it creates a relatively positive area above the ischemic zone, and a relatively negative area at the ischemic boundaries.

This computer model helps explain why reciprocal changes may appear prior to ST segment elevation. Some authors have suggested that the first sign of acute inferior STEMI is a downsloping ST segment in lead aVL, and I have seen this happen many times.

Regardless of why reciprocal changes occur, clinical experience shows that the most important reciprocal changes can be viewed between the high lateral leads (I and aVL) and the inferior leads (II, III and aVF).

Keep in mind that reciprocal changes can be subtle, and may present as nothing more than a flattening of the ST segment in the reciprocal leads.

*** Update 01/15/09 ***

Check out this case at Dr. Smith's ECG blog to see just how subtle reciprocal changes can be! And how they can prevent you from discharging a patient home to experience cardiac arrest!

*** End update ***

You will sometimes notice reciprocal changes in the anterior leads (V1, V2, V3 and V4). These usually represent reciprocal changes associated with injury of the posterior wall of the left ventricle. Since we don't usually view modified chest leads V7, V8 and V9, we most often see these changes associated with acute inferior STEMI, because the posterior descending artery branches off the right coronary artery (RCA), which also supplies the inferior wall of the left ventricle.

With anterior STEMI, the occlusion is often in the left anterior descending artery (LAD) which branches off the left main coronary artery. Depending on the patient's coronary vasculature, the culprit artery, and the location of the occlusion, the blood supply may also effect the lateral wall of the left ventricle, which can create reciprocal changes in the inferior leads (sometimes very subtle depending on the stage of the infarct).

Reciprocal changes may not always be present, but when they are present, it is very strong supporting evidence that the patient is experiencing actue STEMI.

See also:

12 lead ECG - lead placement diagrams

The problem of ST segment elevation

Identifying AMI in the presence of LBBB - Sgarbossa's Criteria - Part II

In Part I, we discussed Sgarbossa's Criteria for identifying AMI in the presence of LBBB. We also talked about the "rule of appropriate T wave discordance" for bundle branch blocks and other forms of abnormal depolarization (like ventricular rhythms or paced rhythms).

You will recall that I drew a distinction between a QRS complex's main deflection and its terminal deflection, even though they are one in the same for LBBB. I explained that it's helpful to think in terms of the terminal deflection, because then you can apply the "rule of appropriate T wave discordance" to RBBB as well as LBBB.

And so you can!

Let's look at an ECG.

This is from one of my "old" 12 lead ECG classes. In those days, I cropped the computer measurements and interpretive statements because I didn't want the students to "cheat". Nowadays, whether it's EMS or firefighting, I've come to believe in reality-based training. In real life, for good or bad, you get an interpretive statement.

But, this ECG is a good example of an important concept.

So let's look at this ECG. It's a sinus rhythm. It has a normal axis. We know that for several reasons.

The QRS complex is smallest in lead aVL, so the perpendicular lead on the hexaxial reference system is lead II. Lead I is almost equiphasic so the perpendicular lead is aVF. The value of lead II is 60 degrees and the value of lead aVF is 90 degrees, so the frontal plane axis is somewhere between 60 and 90 degrees.

Or, to do it the "easy" way, lead I and lead aVF are both positively deflected, so we know we're in the left inferior quadrant.

Or, because leads I, II, and III are all positive, so we know the axis is normal.

It really doesn't matter what method you use. I use all three for every ECG.

The QRS duration is wide. When supraventricular rhythms are wide, we look at lead V1 to see if it shows RBBB or LBBB morphology. This ECG shows a terminal R wave in lead V1, which is RBBB morphology. Next we check lead I and look for a terminal S wave. We find one!

This is a simple RBBB.

You will notice that in many leads, the T wave is deflected the same direction as the QRS complex (II, III, aVR, aVF, V2, V3, V4, V5, and V6). In other leads, the T wave is deflected opposite the main deflection of the QRS complex (aVL, V1). I did not list lead I because the QRS complex is close to equiphasic.

So, how should these T waves be deflected?

The answer is, they should be deflected opposite the terminal deflection of the QRS complex, and so they are!

Look at the following image.


As you can see, when the terminal deflection of the QRS complex is negative, the T wave is positive. When the terminal deflection is positive, the T wave is negative. In other words, even if the main deflection of the QRS complex is positive, as long as the terminal deflection (or last deflection) is negative, the T wave is positive.

That's why I'm encouraging you to always think in terms of the terminal deflection, even though for LBBB, the terminal deflection is also the main deflection.

There is method to this madness!

Although not part of Sgarbossa's Criteria, the "rule of appropriate T wave discordance" can help you pick up on AMI in the setting of RBBB (or bifascicular block) because an inappropriately concordant T wave can tip you off that something is wrong!

In the last lesson, we introduced Sgarbossa's Criteria. Let's take a look at a graphic that shows exactly what we're looking for.


The first example shows > 1 mm of concordant ST segment elevation (and a concordant T wave). Both are abnormal for LBBB. The second example shows > 5 mm of discordant ST segment elevation and a discordant T wave. Discordant ST segment elevation > 5 mm is abnormal for LBBB (with one very important caveat) but a discordant T wave is normal for LBBB! In the last example, there is concordant ST segment depression in the right precordial leads, which is abnormal for LBBB, but a discordant T wave, which is normal for LBBB.

If you have a patient with signs and symptoms consistent with ACS and the ECG shows LBBB with concordant ST segment elevation, then chances are excellent that you are dealing with a STEMI.

Likewise, if you have a patient with signs and symptoms consistent with ACS and the ECG shows LBBB with concordant ST segment depression, especially in the right precordial leads, then chances are excellent that you are dealing with a STEMI.

If you have a patient with signs and symptoms consistent with ACS, and the ECG shows discordant ST segment elevation > 0.25 the depth of the S wave, then chances are excellent you are dealing with a STEMI (thanks to Stephen W. Smith, M.D for this tip).

The original criteria didn't take into account the depth of the S wave, and as we know from other STE-mimics like LVH, the deeper the S wave, the higher the ST segment elevation. So a blanket statement that = or > 5 mm of discordant ST segment elevation indicates an evolving AMI is not helpful in those situations where the S wave is > 50 mm deep (see example below).

Dr. Smith has suggested that a more specific marker is discordant ST segment elevation > 0.25 the depth of the S wave.


If you have any examples of AMI in the presence of LBBB, please share them! I'd be happy to post them to this article and give you the credit.

*** Update 12/29/08 ***

Some of you have suggested it would be nice to see some examples of LBBB with STEMI.

Click here to see an excellent example of LBBB with STEMI from the ECGpedia.

The "old" ECG that shows the baseline without STEMI is here.

Also see Dr. Smith's ECG blog here and scroll down to the second case.

You can also see an example of LBBB with STEMI if you take the D2B's ECG Challenge which is located here.

You can also find examples in the peer reviewed literature.

Electrocardiographic diagnosis of evolving acute myocardial infarction in the presence of left bundle-branch block. GUSTO-1 (Global Utilization of Streptokinase and Tissue Plasminogen Activator for Occluded Coronary Arteries) Investigators. N Eng J Med. 1996 Feb 22;334(8):481-7 PMID: 8559200

Electrocardiographic Diagnosis of Acute Myocardial Infarction During Ventricular Pacing. Circulation 1998; 97:2274-2275 PMID: 9631878

Electrocardiographic ST-segment Elevation: Correct Identification of Acute Myocardial Infarction (AMI) and Non-AMI Syndromes by Emergency Physicians. Acad Emerg Med 2001; 8:349-360 PMID: 11282670

Related posts:

The problem of ST segment elevation

False positive cardiac cath lab activations

Physio-Control - Voices of Life Saving

In August of 2007 I attended the Company Officer Leadership Symposium at the IAFC's Fire Rescue International (FRI) in Atlanta, GA (which was awesome).

Physio-Control had a large section on the conference floor, so I stopped by to say hello, pick up a really cool pen, and see what was new.

I had an interesting meeting with some of Physio's more knowledgeable technical folks, and I gave a short interview in exchange for a rolling book bag! :)

Now I'm a voice of lifesaving! Check it out here.

Identifying AMI in the presence of LBBB - Sgarbossa's Criteria - Part I

There has been a lot of discussion lately about identifying AMI in the presence of LBBB (see Dr. Bearemy's "My Emergency Medicine Blog" here and a recent thread on the EKG Club). I've also been receiving a lot of emails offlist, so I think a full discussion is in order.

In my recent post Who benefits the most from reperfusion therapy? I posted a graph that demonstrates how patients with new bundle branch block benefit the most from reperfusion therapy.

The problem is that in many prehospital 12 lead programs (and regional STEMI systems), patients with LBBB or a QRS duration > 0.12 sec (120 ms) are excluded! In other words, patients with wide QRS are taken to the local community hospital without interventional capability. Or, the cath lab is not activated while EMS is still in the field.

Why would you exclude the very patients who stand to benefit the most from prompt, expertly performed PCI at a cardiac center?

Simple.

It's too difficult to figure out whether or not the BBB is new! The ECG diagnosis of STEMI can be difficult in the setting of BBB.

In False Positive Cardiac Cath Lab Activations I reviewed Larson, Menssen, Sharkey et all, False-Positive" Cardiac Catheterization Laboratory Activation Among Patients With Suspected ST-Segment Elevation Myocardial Infarction, JAMA 2007;298(23):2754-2760.

I quoted:

Patients with new or presumably new left bundle-branch block had an inordinately high prevalence of false positive catheterization laboratory activation (almost half did not have a culprit artery). Patients with a previous myocardial infarction or previous coronary bypass surgery had a significantly higher prevalence of no culprit artery, likely because of abnormal baseline ECG results.

This is obviously a big problem, and subjecting all patients with LBBB and signs and symptoms of ACS to an emergent cath or the risks associated with thrombolytic therapy is not the answer, as some authors have suggested.

If only there was some kind of algorithm that could help distinguish between patients with LBBB and acute STEMI from patients with LBBB who are not experiencing acute STEMI.

But there is such an algorithm! It's been around for over 10 years!

The GUSTO investigators Sgarbossa et al., Electrocardiographic Diagnosis of Evolving Acute Myocardial Infarction in the Presence of Left Bundle-Branch Block. N Eng J Med 1996; 334(8):481-487 published an algorithm which has come to be known as "Sgarbossa's Criteria".

The criteria seems complicated but it's really not. Like anything else, it's a tool. A very important tool for a critical subset of patients.

The original paper contains a flow chart from which the patient receives a score. I'm not going to publish the flow chart, because it's not something you need to memorize.

Here is the criteria. A patient is presumed to be experiencing an evolving AMI if any of the following are present.

1.) ST segment elevation = or > 1 mm that is concordant with the QRS complex.
2.) ST segment depression = or > 1 mm in leads V1, V2, or V3.
3.) ST segment elevation = or > 5 mm that is discordant with the QRS complex.

It is the last criterion that has caused the most controversy and requires qualification.

However, before we address the third criterion, we have to dispose of a common misunderstanding.

What do we mean by concordant and discordant? The short answer is, concordant means "the same direction" and discordant means "the opposite direction".

The rule of appropriate T wave discordance

In the presence of abnormal ventricular depolarization (left bundle branch block, right bundle branch block, paced rhythm, ventricular rhythms) the T wave should be deflected opposite the terminal deflection of the QRS complex (appropriate T wave discordance).

What is the terminal deflection?

The terminal deflection is the last deflection, or wave, of a QRS complex.

Please take the time to learn this! It is extremely important!

Take a look at the following image. Before someone accuses me of stealing it from the Electrocardiogram article in the English Wikipedia, please know that I'm the one who created it and uploaded it to the English Wikipedia in the first place.

You will notice that each of these "QRS complexes" is labeled according to what waves are present. If the wave is large, it gets a capital letter. If the wave is comparatively small, it gets a lowercase letter.

I could talk about this image for a long time, but for now, I just want you to notice that an Rs complex is positively deflected while an rS complex is negatively deflected, even though both of them contain only an R and an S wave. But the terminal deflection of each is negative, because they both end in an S wave!

Why is this important?

When teaching Sgarbossa's Criteria, students always get confused as to whether or not the ST segments and T waves should be deflected opposite the main deflection of the QRS complex or opposite the terminal deflection.

Well, guess what?

With LBBB, the terminal deflection is the main deflection!

So why are we splitting hairs?

Because if you learn to think in terms of the terminal deflection, you can use the rule of appropriate T wave discordance for RBBB, too!

Let's start by looking at a patient with a normal LBBB.


I have no idea why the GE-Marquette 12SL interpretive algorithm is giving the "data quality prohibits interpretation" message for this ECG. There's a little bit of artifact in the inferior leads, but it's not that bad!

As a side note, which electrode would you troubleshoot to correct this problem? The red one!

This is a normal looking LBBB. We know the frontal plane axis is around 0 degrees, because the QRS complex is isoelectric in lead aVF. Therefore, the perpendicular lead in the hexaxial reference system is lead I. Since lead I is positively deflected, we can place the frontal plane axis at 0 degrees. A physiological left axis deviation (0 to -30) is normal for left bundle branch block.

To put it another way, a negative QRS complex in lead III is normal for LBBB, but it should be upright and monomorphic in lead I.

Now, let's look at the QRS complexes and the T waves.

You will notice that in every lead, the T wave is deflected opposite the QRS complex! This is "appropriate T wave discordance" in the presence of left bundle branch block.

Now look at lead V4. You will notice that V4 is showing an rS complex. Since the terminal deflection (the S wave) is negatively deflected, the terminal deflection is negative. The terminal deflection also happens to be the main deflection, which is normal for left bundle branch block.

When there is only one deflection (for example, the monophasic R wave in lead I or the QS complex in lead V1) then the entire complex is positive (lead I) or negative (lead V1). The first deflection is the first and last deflection, so it is the terminal deflection in these leads.

See what I mean? In LBBB, the main deflection and the terminal deflection are one in the same!

To help illustrate this point, consider the following graphic.


The blue arrow shows the direction of the terminal deflection of the QRS complex (which is also the main deflection in the setting of LBBB). The red arrows shows the direction of the ST segment and the T wave.

This is what we mean by "appropriate T wave (and ST segment) discordance" with LBBB. Note that with RBBB, the T wave should be discordant, but the ST segment should remain isoelectric. This is why RBBB is usually not listed as a STE-mimic.

With LBBB, there is also a discordant shift of the ST segment, which is why it's one of the most common STE-mimics! ST segment elevation in the right precordial leads (V1-V3) is a normal finding for LBBB!

In Part II, we'll look at the "rule of appropriate T wave discordance" as it applies to RBBB and talk more about Sgarbossa's Criteria.

Firefighter job stress and heart attacks

"Maxed Out" Part 1


Here's an outstanding report by WTHR-TV Channel 13 news in Indianapolis, IN that looks at firefighter job stress and heart attacks.

Everyone in public safety needs to watch this video!

See also:

Study Targets Firefighters' Heart-Attack Risk

And from Chief Billy Goldfeder's FireFighterCloseCalls.com:

Maxed Out: Firefighters throughout the city are facing new alarms, and it's personal

Prehospital 12 lead ECG programs

I posted this series of questions today at JEMS Connect. Please feel free to respond here or there.

Does your service perform prehospital 12 lead ECGs?

Does the prehospital 12 lead ECG help determine your transport destination?

Is the cath lab activated based on the prehospital 12 lead ECG?

Does your program rely on the paramedic interpretation of the 12 lead ECG? Does your program rely on computerized interpretation of the 12 lead ECG? Does your program rely on off-site interpretation of the 12 lead ECG by a physician? Does your program utilize a combination of these three methods?

If the prehospital 12 lead ECG is transmitted for off-site interpretation by a physician, what technology are you using? What is the failure rate? How is the data quality?

What was your initial education in 12 lead ECG interpretation?

Did it include a strong emphasis on STE-mimics (left bundle branch block, paced rhythm, left ventricular hypertrophy, benign early repolarization, pericarditis, hyperkalemia, etc.)?

Is there anything notable about your QA/QI process (regular multidisciplinary meetings at the hospital, data sharing, every 12 lead ECG reviewed by the Medical Control Physician, paramedics taken out of service and allowed to watch the cath procedure, etc.)?

Is your state involved in regionalizing STEMI care (similiar to trauma)?

How often are you called to the local community hospital for interhospital transfer of STEMI patients for primary PCI?

What has been the biggest barrier to the success of your prehospital 12 lead ECG program?

Who receives the most benefit from reperfusion therapy?

Far and away, patients with new bundle branch block receive the highest benefit from reperfusion therapy, followed by patients with acute anterior STEMI, and then acute inferior STEMI.

When fibrinolytic therapy is administered to patients with ST segment depression only, we kill an additional 14 patients per 1000 treated.

When fibrinolytic therapy is administered to patients with a "normal" admission 12 lead ECG, we kill an additional 7 patients per 1000 treated.

* Note: The Fibrinolytic Therapy Trialists (FTT) Collaborative Group obtained patient data from over 60,000 patients in 22 trials. Of those, 4% had BBB, 68% had ST segment elevation, 7% had ST segment depression, 17% had other abnormalities (e.g., inverted T waves), and 5% had near-normal ECGs.

See also:

Indications for fibrinolytic therapy in suspected acute myocardial infarction: collaborative overview of early mortality and major morbidity results from all randomised trials of more than 1000 patients. Lancet 1994; 343: 311-22

41 yom CC: chest pain

41 year old male complaining of chest discomfort. The patient has had similar episodes before (after exertion), but in the past it always cleared up after use of an asthma inhaler. EMS finds the patient sitting in a chair. He had just taken a shower, with no relief of the chest discomfort. He describes the pain as "heavy" and "unrelenting".

Skin cool, pale, and diaphoretic. Breath sounds clear bilaterally.

PMH: Asthma
Meds: Inhaler (unknown type)

Vital signs:

Resp: 20
Pulse: 60
BP: 112/84
SpO2: 91 on RA

The cardiac monitor is attached.


A 12 lead ECG is obtained.


What's going on here?

Update: This ECG was taken in the back of the ambulance.


Does this help with the diagnosis?

Man Dies at Home After Paramedics Diagnose Acid Reflux

Thanks to Dave at STATter911 for turning me on to this story from the Washington Post.

Apparently a 39 year old male died a few hours after paramedics diagnosed his chest pain and shortness of breath as acid reflux and prescribed him Pepto-Bismol.

Nice!

Here's a quote from the story.

"Lolitha Givens said the firefighters asked her son what was wrong, and the emergency medical technicians who arrived by ambulance checked his vital signs and performed an electrocardiogram, the results of which they said were normal."

Oh really?

While it's possible that the ECG was "normal" I'd love to see it!

That's assuming it was a 12 lead ECG and not a rhythm strip.

Does the absence of acute changes mean the patient is experiencing acid reflux? I must have missed that report in the New England Journal of Medicine.

Maybe it was the same issue that explains how acid reflux causes shortness of breath.

In reality, a minority of patients (1% to 4%) who present to the emergency department with chest pain and a normal ECG are ultimately diagnosed with acute myocardial infarction.

The press release by Fire Chief Dennis Rubin states:

"I was notified today about a specific medical call that is currently under investigation by the DC Fire & EMS Department.

At 11:40 last night, Paramedic Engine 30 and Ambulance 30 were summoned to an address in Northeast for a person having trouble breathing. Personnel responded quickly, in just over four minutes, and patient assessment was promptly started. Personnel provided service on the scene, but the patient, identified as an adult male, was not transported to the hospital.

Authorities responded back to the same location early today and found the same patient deceased.

As per protocol, we are conducting a thorough quality assurance case review and we will determine whether proper care was provided and if the two medical events are related. Until this investigation is complete, we will not be able to make any further comment."

I hope it's a coincidence, too, Chief.

Either way, if the allegations are true, it's disturbing to say the least.

This coming right on the heels of the report that a Captain and wagon driver from Engine 30 became involved in a physical altercation on a medical call.

This is an important reminder for all of us.

The only chest pain patients who should not be transported to the emergency department are those who adamantly refuse care against medical advice.

You know, the hospital.

With the doctors, nurses, x-ray machines, laboratory, and all that stuff.

Update: Here is a video from CNN.com with credit to My Emergency Medicine Blog.

Update: RogueMedic writes about the call here.

Lifepak 15

I was invited to the local hospital today to see a presentation on Physio-Control's new STEMI Management Solution (a topic I will write about at a later date). In the process, I managed to get a look at the Lifepack 15. I was impressed!