Analysis of the degree of QRS fusion necessary for its visual detection: importance for the recognition of transient entrainment.

Abstract

Fixed fusion is the hallmark for the demonstration of transient entrainment. However, the degree of accuracy of its recognition on the surface ECG is unknown. The purpose of the present study was to evaluate the ability to detect fusion in the QRS complex. While pacing the ventricles at a fixed rate, a model of ventricular fusion was created by introducing late extra stimuli at a second site. In this model, the presence and degree of fusion are known. Pacing sites were the RV apex, outflow tract, and left ventricle in various configurations. We analyzed 433 QRS complexes with different degrees of fusion (or no fusion) in 21 patients. Each QRS was "read" by three investigators blinded to intracardiac recordings but having a reference QRS with no fusion. There was a statistically significant correlation between the degree of fusion and its recognition. Fusion was detected with a sensitivity of 75% and a specificity of 87%. Fusion was accurately detected in all configurations only when >22% of the QRS was fused. In patients with organic left ventricular disease, fusion was better recognized when the driving pacing site was the left ventricle than when it was a right ventricular site. The interobserver agreement was moderate between two pairs of observers and only fair between the remaining pair. Our results suggest that an accurate detection of ventricular fusion can only be accomplished when fusion occurs during a significant proportion of the QRS duration. The potential lack of recognition of minor degrees of fusion may produce underdetection of transient entrainment.