Ventricular pacing site separation by cardiac computed tomography: validation for the prediction of clinical response to cardiac resynchronization therapy

Cardiac Resynchronization Therapy (CRT) fails to provide benefit in up to one-third of patients. Maximizing the geographic separation of right and left ventricular pacing lead sites has been suggested as one way to improve response. Cardiac CT provides an opportunity to explore 3-dimensional inter-lead distance (ILD) measures for the prediction of CRT response. The objective of this study was to investigate associations between standardized measures of ILD by cardiac CT and echocardiographic response to CRT. Forty-two consecutive patients undergoing CRT had serial clinical and echocardiographic evaluations performed in addition to a post-procedural cardiac-gated CT with blinded measurement of direct and circumferential (via the myocardium) ILD measures. Clinical response to CRT, the primary clinical outcome, was defined as a ≥15% reduction in LVESV using echocardiography at 6-months. The mean age and ejection fraction was 63.6 ± 8.9 years and 25.2 ± 7.8%, respectively. The primary outcome occurred in 35 of 42 patients (83%). Both direct and circumferential CT-based ILD measures were associated with the primary outcome by univariate analysis. Receiver Operator Characteristic analysis identified Circumferential ILD to have the strongest predictive accuracy (AUC 0.78). Inter- and intra-observer reproducibility of CT-derived ILD measures was excellent. Circumferential ILD measures on cardiac CT are predictive of clinical response to CRT. Incorporation of these measures into the selection of optimal pacing targets, particularly from pre-procedural CT coronary vein imaging may be of therapeutic benefit and warrants further investigation.