Baseline wander. But what else do you see?

Written by Magnus Nossen

The below ECG was obtained from a 65 year old man with ongoing chest pain. He has a history of hypertension and tobacco use. He is otherwise healthy. There is no prior ECG for comparison. What do you think?

ECG #1

Interpretation: The quality of the above ECG is not optimal. There is significant baseline wander that does make it more challenging to interpret ischemic changes. The ECG shows sinus rhythm at a rate of 90 bpm. There is a slightly wide QRS (just about 120ms) consistent with a nonspecific intraventricular conduction delay (IVCD). The PR interval is on the short side, but there are no clear delta waves.

Despite the baseline wander there is appreciable concordant ST segment elevation in lead III. There is likely some very slight concordant ST elevation also in lead aVF. The lateral leads I and aVL show reciprocal ST depression, with lead aVL showing concordant ST depression. The S-wave in lead V2 measures 17mm and the ST segment in lead V2 is isoelectric. With the slightly prolonged QRS and significant S-wave one would expect some discordant ST elevation in lead V2. In other words there is inappropriate isoelectric ST segment in V2.

Putting all the above findings together, the ECG is very concerning if not diagnostic for inferior and posterior OMI. These findings were not appreciated by providers. However, due to the nature of the symptoms, the patient was treated for ACS and admitted to the hospital. The below ECG was obtained 45 minutes after the first one with the patient being free of chest pain at the time of recording of ECG #2. 

What do you think now?

ECG #2

Again there is sinus rhythm, short PR and IVCD. The ECG now shows resolution of the concordant ST segments in the inferior leads. There is now a very slight ST elevation in V2, as one would expect appropriate discordant ST Elevation, discordant (in the opposite direction) from the preceeding majority negative QRS (S-wave). 

The artery has re-perfused. 

There was no comment by the providers on the dynamic ECG changes on the repeat ECG.

The patient in today's case was lucky to re-perfuse. He remained chest pain free and underwent coronary angiography the following day. The CAG showed a 99% thrombotic lesion of the proximal RCA, which was stented. Due to rapid spontaneous reperfusion, high sensitivity troponin I peaked at just under 5000 ng/L. 

This OMI went unrecognized and, had the artery not re-perfused by itself, the patient could have suffered a very large myocardial infarction. He was diagnosed with NSTEMI. 

What does the Queen of Hearts AI model say about the ECGs in today's case?

Below: Queen of Hearts interpretation ECG #1

Below: Queen of hearts interpretation ECG #2 

Discussion: OMI findings can be subtle. Whenever there is a wide QRS complex and baseline wander the interpretation is more difficult. The Queen of Hearts AI model confidently identifies the first ECG in today's case as occlusion myocardial infarction (OMI HIGH confidence). The repeat ECG, after reperfusion is equally confidently recognized as not OMI. 

The outcome in today's case was good but that was not due to expert ECG interpretation but rather a bit of luck. The OMI could have been recognized if the QoH had been used. 

Smith comment:

Approximately 33% of cases that everyone would call STEMI reperfuse spontaneously (TIMI-1, 2, 3 flow) before they undergo emergent angiogram (usually under 90 minute door to balloon time)  

20% have TIMI-3 flow.  

Of all OMI, approximately 43% present with diagnostic ST Elevation.  57% do not and are called "NSTEMI."   But many of these OMI will spontaneously reperfuse by the time of next day angiogram so that, at next day angiogram, 25% of NSTEMI have TIMI-0 flow and 34% have TIMI 0-1 flow without collateral flow. 

You don't know which will reperfuse at the time they present with OMI.  So all must be treated emergently.  Many subtle OMI do not get diagnosed until their next-day angiogram, at which time much myocardium is lost.  They have much higher mortality and worse LV function than the NSTEMI with open artery next day. 

Therefore, you must find a way to recognize these subtle OMI at arrival in order to get them rapid intervention.  Expert or AI ECG interpretation is the only way to do so.

A significant proportion of patients with OMI do not re-perfuse on their own and the use of the QoH can provide more timely treatment and a better outcome for these patients.  

Click here to sign up for Queen of Hearts Access.

Learning points: 

• With training one can learn to identify OMI despite baseline wander and a wide QRS complex. 
  
  
• The Queen of Hearts already does this very well and can improve the management of ACS patients 
  
  
• We need to better identify patients with ongoing coronary occlusion as we cannot count on patients being as lucky at the patient in today's case

• Whenever the ECG has poor quality obtaining a repeat tracing is advisable. Sometimes a poor quality tracing is all you have to make your decision.
  
  

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

MY Comment, by KEN GRAUER, MD (11/8/2024):  

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

Today’s case by Dr. Nossen brings out a series of important points:

• As per comments by Drs. Nossen and Smith — the initial ECG is absolutely diagnostic of acute infero-postero OMI.
• As per Dr. Nossen — the presence of both QRS widening and baseline wander + significant artifact make assessment of acute OMI more challenging. That said — this 65-year old man with risk factors (smoking, hypertension) — was admitted to the hospital for “ongoing CP” that apparently was severe enough, such that this chest pain (and not the initial ECG) was the reason that cardiac cath was ultimately performed (albeit cardiac cath was not done until the next day). The point is that today's patient was high-risk for having an acute event.

KEY Point: Although it is understandable that providers may have missed the subtle-but-real ST-T wave changes of acute OMI on the initial ECG — given the high-risk likelihood for an acute event — the technical concerns regarding the initial ECG should have immediately prompted repeating the ECG:

• Consider Figure-1 — in which I have labeled key findings in the limb leads. The question that arises is which of the 8 beats in the limb leads of the initial ECG accurately convey ST-T wave appearance?
• For example — IF the only 3 beats seen in the limb leads were beats #5,6,7 (within the dotted RED rectangles in Figure-1) — it would have been much clearer that there is hyperacute ST elevation in lead III, and to a lesser extent, also in lead aVF (RED arrows in these leads) — with reciprocal ST depression in lead aVL (PINK arrows).
• On the other hand — I would not have been at all certain about ST elevation in leads III and aVF (nor of reciprocal ST depression in lead aVL) — if the only beats seen in the limb leads looked like beats #3,4.
• And, as per Dr. Nossen — in the presence of acute inferior OMI — loss of the usual normal slight, gently upsloping ST elevation in lead V3 indicates associated posterior OMI (BLUE arrows in this lead showing slight J-point ST depression).

NOTE: It may help to recognize that the reason for so much artifact in the limb leads — is the result of some disturbance or movement (ie, tremor? fidgeting? scratching?) in the LA electrode.

• As per My Comment in the February 18, 2024 post in Dr. Smith’s ECG Blog — the LA (Left Arm) can be quickly identified as the “culprit” extremity because baseline artifact is maximal in leads I, III and aVL — present but less in leads aVR and aVF — and not present at all in lead II. 
• Being able to identify within seconds the “culprit” extremity responsible for the artifact may suggest a “quick fix” (ie, Asking the patient to stop scratching or moving the "culprit" extremity) — and then repeating the ECG to clarify the true nature of ST-T wave appearance.

Side NOTE: Did you notice how much narrower the QRS in lead V4 looks compared to other leads? This is because the subtle, terminal part of the QRS in V4 goes undetected because it almost entirely lies on the baseline (YELLOW arrows in this lead).

• This finding illustrates the importance of always using more than a single monitoring lead when assessing QRS width in a tachyarrhythmia (since you otherwise might mistake a wide tachycardia for an SVT — if part of the QRS lies on the baseline in the single lead of your 1-lead rhythm strip).

Figure-1: I've labeled the initial ECG in today's case. (To improve visualization — I've digitized the original ECG using PMcardio).

How Not to Miss "Dynamic" ST-T Wave Changes:

Not only were the acute ST-T wave findings missed in ECG #1 — but the new limb lead ST-T wave changes in ECG #2 (compared to ECG #1) were also missed. Awareness of the following points can avoid overlooking these telltale "dynamic" ST-T wave changes:

• We need to be looking for "dynamic" ST-T wave changes in serial tracings — because when found (especially if the nature of these changes correlates with the clinical situation) — they are diagnostic of an acute event!
• In today's case — this patient with "ongoing CP" at the time of admission — no longer had CP 45 minutes later, at the time ECG #2 was obtained. We need to be aware that if ST elevation resolves in association with reduced (or relieved) CP, with replacement of ST elevation by reperfusion T waves (ST depression and/or T wave inversion) — that these "dynamic" ST-T wave changes tell us that the "culprit" artery has spontaneously reperfused!

PEARL: Unless serial ECGs are compared side-by-side — subtle-but-important ST-T wave changes will be missed. These ST-T wave changes between ECG #1 and ECG #2 were not seen in today's case.

• LOOK at Figure-2. 
• Isn't it much easier to recognize reperfusion ST-T wave changes now that I have placed the limb leads from ECG #1 and ECG #2 next to each other? (Comparison between RED/PINK vs BLUE arrows in Figure-2).
• 
  
• BOTTOM Line: Even if the acute findings in ECG #1 went unrecognized — the occurrence of reperfusion ST-T waves in association with resolution of CP at the time ECG #2 was obtained — would have confirmed acute OMI and the need for prompt cath (instead of cath being delayed until the next day).

Figure-2: Comparison of ST-T wave changes in the limb leads in ECG #1 vs ECG #2. (To improve visualization — I've digitized the original ECG using PMcardio).