Abstract

BACKGROUND: Pacemapping is used to localize the exit site of ventricular arrhythmia. Although the relationship between distance and change in QRS morphology is its basis, this relationship has not been systematically quantified. METHODS AND RESULTS: Patients (n=68) undergoing ventricular arrhythmia ablation between March 2012 and July 2013 were recruited. Pacemapping was targeted to areas of voltage >0.5 mV. Linear mixed-effects models were constructed of distance against morphology difference measured by the root mean square error sum across all 12 ECG leads (E12). Forty of 68 (58%) patients had structural heart disease, and 21/40 (53%) patients were ischemic. Nine hundred thirty-five pacing points were collected, generating 6219 pacing site pair combinations (3087 [50%] ventricular bodies, 756 [12%] outflow tract, and 162 [3%] epicardial). In multivariable analysis, increase in E12 was predicted by increasing distance (0.07 per mm; 95% confidence interval 0.07-0.08; P<0.001). Compared with the left ventricle, E12 values were lower in the right ventricle (P=0.037) and left ventricular outflow tract (P<0.001) and higher in left ventricle-right ventricle pairs (P=0.021) and left ventricular epicardium (P=0.08). There was no difference in E12 in the right ventricular outflow tract compared with the right-left ventricular outflow tract (P=0.75) pairs. Structural heart disease or inadvertent pacing in scar was not associated with changes in E12; however, the presence of latency and split potentials were associated with higher and lower E12 values, respectively (P<0.001). CONCLUSIONS: A robust positive relationship exists between distance and QRS morphological change when restricting pacing points to areas of voltage >0.5 mV. Significant differences in the spatial resolution of pacemapping exist within the heart.