Differentiation of paroxysmal narrow QRS complex tachycardias using the 12-lead electrocardiogram

OBJECTIVES The purpose of this study was to evaluate the utility of the 12-lead electrocardiogram (ECG) for differentiating paroxysmal narrow QRS complex tachycardias. BACKGROUND Previous studies evaluating the utility of the 12-lead ECG for differentiating paroxysmal supraventricular tachycardia types have shown conflicting results on the usefulness of some ECG criteria, and some criteria that are considered to be useful have never been formally evaluated. METHODS Two hundred forty-two ECGs demonstrating paroxysmal narrow QRS complex (< 0.11 ms) tachycardia (rate > or = 120 beats/min) were analyzed. All ECGs were analyzed by an observer who had no knowledge of the mechanism of the tachycardia. RESULTS There were 137 atrioventricular (AV) reciprocating tachycardias, 93 AV node reentrant tachycardias and 12 atrial tachycardias. Six criteria were found to be significantly different between tachycardia types by univariate analysis. A P wave separate from the QRS complex was observed more frequently in AV reciprocating tachycardia (68%) and atrial tachycardias (75%). A pseudo r' deflection in lead V1 and a pseudo S wave in the inferior leads were more common in AV node reentrant tachycardia (58% and 14%, respectively); QRS alternans was present more often during AV reciprocating tachycardia (27%). When a P wave was present, an RP/PR interval ratio > or = 1 was more common in atrial tachycardias (89%). During sinus rhythm, manifest pre-excitation was observed more often in patients with AV reciprocating tachycardia (45%). By multivariate analysis, the presence of a P wave separate from the QRS complex, pseudo r' deflection in lead V1, QRS alternans during tachycardia and the presence of pre-excitation during sinus rhythm were independent predictors of tachycardia type. These criteria correctly identified 86% of AV node reentrant tachycardias, 81% of AV reciprocating tachycardias and incorrectly assigned the tachycardia type in 19% of cases. CONCLUSIONS Several features on the ECG are useful for differentiating supraventricular tachycardia type. However, approximately 20% of tachycardias may be incorrectly classified on the basis of analysis of the ECG; therefore, the ECG should not serve as the sole means for determining tachycardia mechanism.