Diagnosing acute myocardial infarction in the setting of left bundle branch block: prevalence and observer variability from a large community study.

BACKGROUND: Despite the known benefit of thrombolysis it remains under-utilized among eligible patients with acute myocardial infarction (AMI) and left bundle branch block (LBBB). We sought to determine the test characteristics and observer reliability of well-known criteria for the diagnosis of AMI when LBBB is present on the electrocardiogram (ECG).

METHODS: Four hundred and fourteen ECGs with LBBB from a large cohort of AMI patients (7.4% of the total) and 85 ECGs with LBBB not in the setting of acute coronary syndromes were interpreted for the presence of the Sgarbossa criteria.

RESULTS: Agreement for the various Sgarbossa criteria ranged from only fair to moderate. The three-way comparison kappa values were significantly better for ST depression than for both discordant (P<0.001) and concordant (P=0.001) ST-segment elevation. Concordant ST-segment elevation [6.3%, 95% confidence interval (CI) 4.3-9.1%] and depression (3.1%, 95% CI 1.8-5.4%) were infrequently seen in the setting of AMI and rarely seen otherwise. Discordant ST-segment elevation was seen more frequently (19.0%, 95% CI 15.5-23.1%). Concordant ST elevation and ST depression in V1-V3 were highly specific, but insensitive, for the diagnosis of AMI. The presence of discordant ST elevation was neither sensitive nor specific.

CONCLUSION: The low prevalence, poor sensitivity and marked observer variability make the Sgarbossa criteria for AMI in the setting of LBBB less than adequate. Although use of these criteria would be an advance over contemporary practice, it would still fall short among this high-risk subset.