Literature Review:             Which patients with suspected myocardial ischemia and left bundle branch block should receive thrombolytic agents?

1.      Gallagher EJ:  Which patients with suspected myocardial ischemia and left bundle branch block should receive thrombolytic agents?  Ann Emerg Med.  2001;37:439-444.

2.      Kontos MC et al:  Can myocardial infarction be rapidly  identified in emergency department patients who have left bundle branch block?  2001;37:431-438.

3.      Li SF et al:  Electrocardiographic diagnosis of myocardial infarction in patients with left bundle branch block.  Ann Emerg Med.  2000;36:561-565

Background

These articles (Gallaghers in particular) are extremely important as they address the question of what to do with patients c/o chest pain who have a LBBB.

Definition of LBBB

QRS e 120ms; no secondary Rwave in lead V1; no Qwaves in lateral leads (I,aVL,V5-6); ST  changes opposite (discordant) to the dominant QRS component & Tw changes in the same direction as the ST segments.  Also, ST elevation in leads V1-4, & Tw inversion I & aVL.

Major Points

·        Should we be concerned about patients c/o chest pain with RBBB?  Normal conduction traverses the AV node and depolarizes the septum from left to right, thus activation the left ventricle first.  ECGs identify ischemia or infarction most readily when it occurs in those areas of the heart depolarized early.  Hence, an RBBB, which does not interfere with either the direction of septal depolarization or the order of ventricular depolarization, will not ordinarily obscure the ST-segment elevation of AMI (acute MI) because it does not greatly alter initial electrical forces.  In contrast to LBBB, conduction traverses the septum in the opposite direction (right to left) & depolarizes the ventricles in the reverse order through propagation of electrical front spreading radially across the left ventricle from the termination of the right bundle.  This obscures early vectors that ordinarily inscribe the characteristic ECG signature of AMI, rendering it illegible in the vast majority of patients.

·        More than 50 ECG signs have been proposed over the past 50 yrs for the detection of ischemia in LBBB.  Of these, only 2 of the 3 criteria proposed by Sgarbossa et al have sufficiently powerful likelihood ratios to be helpful at the bedside (a& b):

·        Sgarbossa criteria:      a)  ST Ý e 1mm in the same direction as the QRS(concordant);

b) ST ß d 1mm in leads V1-V3;

c) ST Ý  e 5mm in the opposite direction as the QRS (discordant)

·        Unfortunately, the Sgarbossa criteria are too insensitive to be used as a screening (rule out) test to determine which patients with an LBBB don not have an MI.  The 1^st two criteria were found in only 3% of 372 patients with LBBB & suspected ischemia.  Among the subset of 49 patients with LBBB & confirmed AMI, at least one of the criteria was present in only 20% of patients.  Although the Sgarbossa criteria are insensitive, they are, however, highly specific and can be used reliable as a confirmatory test to rule in AMI in patients with LBBB.

·        In terms of likelihood ratios (LR), the positive LR (presence of either concordant criterion) is 22; whereas the negative LR is 0.8.  Note that a LR of 1 indicates a 50% odds of the test being negative or positive.  Consequently, the further a LR is away from 1, the more relevant it is.  Therefore a positive LR increases your post test odds by a factor of 22; whereas the negative LR only decreases your post test odds by a factor of 0.8.

·        Based on these findings, Kontos et al argue that the low sensitivity (poor negative LR) of the Sgarbossa criteria would result in too few people with AMI would be treated if these criteria were used; and that the low prevalence of AMI among patients with LBBB would unnecessarily expose too many people to the risks of thrombolytic therapy.

·        Gallagher states that the latter statement requires closer examination & proposes the use of decision analysis as a means of explicitly weighing the quantitative implications of competing therapeutic recommendations.

·        The prevalence of AMI in LBBB is ~13%. 

·        Decision analysis divided patients into 2 groups for comparison:  a) combined criteria of either concordant Sgarbossa ECG criteria OR a new/indeterminate LBBB as a positive test result indicating a need for thrombolysis; b)  administering thrombolysis to all patients with LBBB & suspected AMI.

·        Probability estimates for decision analysis

·        Outcome  of decision analysis comparing the 2 strategies in 1,000 patients w/LBBB & suspected AMI

Interestingly, the outcomes of the 2 strategies are remarkably similar.  Therefore, an algorithm incorporating the concordant Sgarbossa criteria combined with the age of the LBBB performs no better than the much simpler strategy of thrombolysis for all suspected AMIs in patients with LBBB.

Conclusion

Administer thrombolytics to all patients with LBBB & suspected AMI.  Not only is this evidence-based, but this position is also recommended by the American College of Cardiology & the American Heart Association.