Abstract 551: Ratio of Discordant ST Segment Elevation or Depression to QRS Complex Amplitude is an Accurate Diagnostic Criterion of Acute Myocardial Infarction in the Presence of Left Bundle Branch Block
Introduction: Left bundle branch block (LBBB) is notorious for obscuring the electrocardiographic (ECG) diagnosis of acute myocardial infarction and, therefore, the decision to undergo reperfusion therapy. Current diagnosis guidelines recommend the three Sgarbossa criteria: 1) discordant ST segment elevation (STE) of >/= 5 mm, 2) concordant STE of </= 1 mm, and 3) concordant ST depression (STD) >/= 1 mm in any of leads V1 to V3. However, these criteria used CK-MB enzyme levels, not acute coronary occlusion, as the diagnostic standard, nor did they account for the importance of proportionality on the ECG. We hypothesized that a ratio of discordant STE or STD to QRS amplitude (ST/QRS) > 0.25 in any one lead would be superior to the Sgarbossa criteria.
Methods: Records of consecutive patients with LBBB and chest pain and/or dyspnea who presented to the emergency department at one of three hospitals were reviewed while reviewers were blinded to the ECGs. Study group patients had a cardiac catheterization report supporting acute coronary occlusion at the time of the ECG. Control group patients had no evidence of recent acute coronary occlusion; non STE myocardial infarction patients were included as controls if acute coronary occlusion could be excluded. Baseline ECGs of all patients, recorded at the time of symptoms, were blindly compared. Measurements included ST segment at the J-point and QRS amplitude relative to the TP segment in all 12 leads. The highest discordant ST/QRS ratio of all leads was selected for each ECG. Statistics were by Fisher’s exact test and Student’s t-test.
Results: The study and control groups consisted of 34 and 130 patients, respectively. Sensitivity, specificity, and accuracy of ST/QRS ratio > 0.25 vs. at least one of the three Sgarbossa criteria were 97% vs. 74% (p < 0.01), 96% vs. 93% (p = NS), and 96% vs. 89% (p < 0.01); the mean highest ST/QRS ratio (+/−CI) was 0.70 +/− 0.16 vs. 0.15 +/− 0.02 (p < 0.001). The presence or absence of concordant STE did not improve accuracy or sensitivity.
Conclusions: When acute coronary occlusion occurs with LBBB, a discordant ST/QRS ratio > 0.25 is significantly more sensitive and accurate for the diagnosis of acute coronary occlusion than the Sgarbossa criteria, and may guide reperfusion decisions.