1. Diagnosis of Myocardial Infarction/Ischemia with Bundle Branch Blocks
Mark I. Langdorf, MD, MHPE, FACEP, FAAEM, RDMS
Chair and Associate Residency Director
Medical Director
Department of Emergency Medicine
University of California, Irvine

2. Objectives
To understand the interpretation of 12 lead ECG with regard to
Infarction
Ischemia
In the presence of
Right bundle branch block
Left bundle branch block
Ventricular paced rhythms
To understand the utility of decision rules on this topic.

3. Take-home Messages:
You can make the diagnosis of acute myocardial infarction or ischemia in the face of bundle branch blocks or paced rhythm.
Secondary ST-T wave changes are normal, and go in the opposite direction of the last portion of the QRS complex.
Primary ST-T wave changes mean ischemic or infarction, and go in the same direction as the last portion of the QRS complex.

4. Take-home Messages:
Left bundle branch block that is new or not known to be old, in the setting of a clinical picture of MI, likely indicates infarction.
Reperfusion therapy is recommended.
Serial ECGs may clarify the situation.
Immediate angiogram is preferred.

5. Cardiac Conduction System
Bachmann’s bundle
Sinus node
Internodal pathways
Left bundle branch
AV node
Bundle of His
Posterior division
Anterior division
Purkinje fibers
Right bundle branch

6. Truth?
“The diagnosis of myocardial infarction in the presence of left bundle branch block is impossible.”

7. Partially true…
True: Diagnosis of completed Mi in left bundle branch block is difficult
Q waves may be present with LBBB in the precordial leads without anterior infarction.
ST segment elevation can be hidden in the usual repolarization changes.
But, Q waves in two contiguous lateral leads suggest completed MI
R wave regression from V1-V4 suggests transmural necrosis.

8. But…
…”ongoing ischemia and injury can be detected in the presence of LBBB, and may be seen as often as they are in the presence of normal cardiac conduction.”
Comparison with old ECGs helpful
Serial ECGs while in the ED also helpful
Fesmire, Annals EM, 26:69, 1995

9. Dr. Braunwald says:
Some findings are highly specific and predictive (90-100%) for MI with left bundle branch block.”
Q waves in at least two contiguous lateral leads (I, aVL, V5 and V6)
R wave regression from V1 to V4
Primary ST-T wave changes in two or more contiguous leads

10. ST-T Wave Changes with Bundle Branch Blocks:
Changes are the same with both right and left
Also applies to LVH with “strain”
“Secondary means normal, expected
“Primary” means abnormal: ischemia or infarction

11. Terminal portion of QRS
J point
ST segment
Terminal portion of QRS

12. Secondary ST-T Wave Changes
These are normal, expected
Terminal portion of the QRS complex is the key
J point displaced away from the terminal portion of the QRS complex
T wave oriented away from the terminal portion of the QRS complex

13. Primary ST-T Wave Changes
“Primary” = abnormal, not a result of BBB
ST elevation still means injury or infarction
ST depression still means ischemia
Exceptions:
Prinzmetal’s angina: reversible ST elevation
ST depression/T wave inversion can represent infarction: “sub-endocardial” or “non-Q wave.”

14. Primary ST-T Wave Changes
Must be in two contiguous leads
Inferior:
II and aVF
III and aVF
not II and III
Septal: V1 and V2
Anterior: V3 and V4
Lateral: V5, V6, I and aVL (high lateral)

15. Primary ST-T Wave Changes
One major caveat:
Allowed one lead that has concordant terminal QRS complex and T wave
QRS changes from predominately positive deflection to predominately negative
Don’t infer ischemia/infarction if only one lead

16. Concept of Dis/Concordance
Refers to whether the last portion of the QRS complex goes in the same or different direction as the T wave
Discordance=good
Concordance=bad

17. ECG of Evolving MI with Left Bundle Branch Block
Review of 26,003 GUSTO patients (1993)
Derivation set: 131 (0.5%) patients with left bundle branch block
Average time from onset of symptoms to ECG: 120 minutes
Validation set: 45 patients from GUSTO-2A with AMI and LBBB
Sgarbossa et al., NEJM, 334:481, 1996

18. ECG of Evolving MI with Left Bundle Branch Block
Identified three predictive criteria:
ST segment elevation > 1 mm concordant with QRS
ST segment depression >1 mm concordant with QRS
ST segment elevation> discordant with QRS
How did these factors perform on the validation set?

19. ECG of Evolving MI with LBBB
ST elevation > 1 mm concordant with QRS
Sensitivity: 73%
Specificity: 92%
Odds ratio 25.2 (95% CI )
ST depression >1 mm concordant with QRS
Sensitivity: 25%
Odds ratio: 6.0 (95% CI )

20. ECG of Evolving MI with Left Bundle Branch Block
ST elevation > 5 mm discordant with QRS
Sensitivity: 26%
Specificity: 92%
Odds ratio: 4.3 (95% CI )
Decision tree incorporates all three factors in order of predictive power

21. ECG of Evolving MI with Left Bundle Branch Block
Does the T wave go the wrong way up?
Does the T wave go the wrong way down?
Does the T wave/ST segment go the right way, but too far?
Three “yes” answers = 100% MI
Three “no” answers = 16% MI

22. Probability of MI %

23. Right Bundle Branch Block in V1
“up down”
Secondary normal ST-T Wave changes

24. Right Bundle Branch Block in V6
“down up”
Secondary normal ST-T Wave Changes

25. Left Bundle Branch Block V1
“down up”
Secondary normal ST-T Wave changes

26. Left Bundle Branch Block V6
“up down”
Secondary normal ST-T Wave changes

27. Right Bundle Branch Block V1
“up up”
Primary Infarction ST-T Wave changes

28. Right Bundle Branch Block V6
“down down”
Primary Ischemic ST-T Wave Changes

29. Left Bundle Branch Block V1
“down down”
Primary Ischemic ST-T Wave Changes

30. Left Bundle Branch Block V6
“up up”
Secondary Infarction ST-T Wave Changes

31. Left Bundle Branch Block V1
> 5 mm
“too far up”
Primary Infarction ST-T Wave change
Exaggerated ST Segment Elevation

32. Right Bundle Branch Block with Secondary ST-T Wave Changes

33. Left Bundle Branch Block with Secondary ST-T Wave Changes

34. Right Ventricular Paced Rhythm with LBBB Pattern: Secondary ST-T Wave Changes

35. RBBB with Anteroseptal Ischemia

36. RBBB with Anteroseptal Infarction

37. LBBB with Exaggerated ST Elevation: Anteroseptal/lateral Infarction

38. LBBB with Lateral Infarction

39. Left Bundle Branch Block with Anteroseptal Ischemia

40. Left Bundle Branch Block
Primary ST T Wave depression
Primary STT Wave elevation
Exaggerated ST segment elevation

41. Are These Criteria Valid?
Poor performance
190 patients/13% with AMI
Sensitivity: 0-16%
Specificity: %
Treat all LBBB not known to be old as acute MI
Good performance
224 patients/45% with AMI
Sensitivity: 73% (cardiologist) vs. 67% (EP)
Specificity: 98% (both)
Li, et al., Annals EM, 36:561, 2000
Sokolove, et al., Annals EM, 36:566, 2000

42. ECG of Evolving MI with LBBB
414 ECGs with AMI and LBBB, 85 with LBBB without AMI
Prevalence of findings:
Concordant ST-segment elevation: 6.3%
Concordant ST-segment depression: 3.1%
Discordant ST-segment elevation: 19.0%
Concordant ST elevation and ST depression in V1-V3 were highly specific for diagnosis of AMI.
Gula LJ et al., Coron Artery Dis. 14:387-93, 2003

43. ECG of LBBB Without MI 124 patients with LBBB and no MI
Only 1 had primary ST segment depression anteriorly
Only 1 had primary ST segment elevation
9 had exaggerated ST segment elevation > 5 mm
Sgarbossa criteria are sufficiently specific (few false positives)
Madias JE, et al., Clin Cardiol. 24:652-5, 2001

44. ECG of Evolving MI with LBBB
182 patients with LBBB and acute MI
New LBBB: Sens = 46%, Spec = 65%
Concordant ST-segment elevation or depression (Sgarbossa criteria)
Specificity = 100%
Positive predictive values = 100%
Sensitivity for ST elevation = 8%
Sensitivity for ST depression = 17%
Of patients with acute MI and LBBB, only in 46% of them was the LBBB NEW from their MI.
Conversely, only 65% of patients with new LBBB actually had MI.
So, if we treated everyone with new LBBB and chest pain, we’d treat 35% unnecessarily. If we only treated NEW LBBB, we’d miss treating 54% of patients who needed it.
Kontos MC, et al., Ann Emerg Med. 37:431-8, 2001

45. ECG of Evolving MI with LBBB
Of patients with acute MI and LBBB, the LBBB was NEW (from their MI) in only 46%.
If we only treated NEW LBBB, we’d miss treating 54% of patients who needed it.
Conversely, only 65% of patients with NEW LBBB actually had MI.
So, if we treated everyone with new LBBB and chest pain we’d treat 35% unnecessarily.

46. Should We Treat All Patients?
American Heart Association:
Literature: not necessarily
35% with new LBBB and chest pain did not have an MI (Kontos)
48% with LBBB not known to be old did not have an MI (Edhouse)
Sgarbossa criteria are helpful
If present, thrombolytics indicated
If absent, serial ECGs or catheterization
Yes!
Edhouse, et al., J Accid Emerg Med. 16:331-5, 1999.

47. Take-home Messages:
You can make the diagnosis of acute myocardial infarction or ischemia in the face of bundle branch blocks or paced rhythm.
Secondary ST-T wave changes are normal, and go in the opposite direction of the last portion of the QRS complex.
Primary ST-T wave changes mean ischemic or infarction, and go in the same direction as the last portion of the QRS complex.

48. Take-home Messages:
Left bundle branch block that is new or not known to be old, in the setting of a clinical picture of MI, likely indicates infarction.
Reperfusion therapy is recommended.
Serial ECGs may clarify the situation.
Immediate angiogram is preferred.