Value of the ECG in suspected acute myocardial infarction with left bundle branch block.

Uncomplicated left bundle branch block (LBBB) is characterized by true ST-segment shifts resulting from delayed repolarization in the left ventricle with respect to the right ventricle. When acute coronary occlusions develop in the setting of previous or new LBBB, 12-lead eCG manifestations of injury may also appear. They consist of a more pronounced ST-segment elevation, of ST-segment deviations opposite to those of uncomplicated LBBB, or both. We have reported that the only 3 independent ECG signs of acute MI during LBBB among patients with chest pain or history of coronary disease are: ST elevation > or = 1 mm in leads with a positive QRS, ST-depression > or = 1 mm in V1 to V3, and ST elevation > or = 5 mm in leads with a negative QRS. In our study, the clinical prediction rule score values of these signs were 5; 3; and 2, respectively. A score > or = 3 made a diagnosis of MI with a 90% specificity and a score of 2 with > 80%, specificity. Recent validation studies have confirmed that the presence of any of these ECG signs is associated with a sensitivity of 44 to 79% and a specificity of 93 to 100%. Sensitivity increases if serial or previous ECGs are available for comparison. Interobserver agreement is very high. While current practice guidelines recommend thrombolysis for all patients with chest pain and LBBB, concern among physicians about hemorrhagic stroke prevents many of these patients from receiving timely treatment. In a population with LBBB and chest pain where our proposed ECG criteria were not ascertained, only 73% of eligible patients received thrombolysis; on the other hand, 48% of patients with no biochemical evidence of MI were thrombolyzed. For the latter group, the clinical prediction rule had a score of 0. Instead, 79% of patients with confirmed acute MI had a prediction rule score > or =2. Similar values applied to a subgroup of patients with serial ECGs. We propose that thrombolysis among patients with chest pain and LBBB be decided on the basis of a systematic ECG review to "rule patients in". This provision may result in both a significant reduction in the number of patients without infarction who receive thrombolysis and in timely treatment of those who do have MI.