P544Evaluation of three-dimensional trajectory of pacing cathode pole in coronary sinus to predict long-term response to cardiac resynchronization therapy

Abstract

Abstract Background An automated method for 3D reconstruction of coronary sinus (CS) lead’s pacing cathode trajectory (3DTJ) was proposed to acutely predict long term left ventricular (LV) mechanic response to cardiac resynchronization therapy (CRT). Preliminary data showed that 3DTJ at biventricular pacing (BIV) start changed in CRT responders (R) to be, becoming less eccentric and more multi-directional, as described by the ratio between its two major axes (S1/S2). Purpose The TRAJECTORIES study (Trajectory Changes Of Coronary Sinus Lead Tip And Cardiac Resynchronization Therapy Outcome, NCT02340546) is an observational study by seven Italian centers about the prediction of CRT-induced LV reverse remodeling by means of the acute 3DTJ changes at CRT implant. Methods In CRT implants with standard indications, stable CHF and regular ventricular rhythm, a fluoroscopic sequence in two standard X-rays views of a few seconds was acquired immediately before (T-1) and after the start of BIV (T0). 3DTJ of CS lead cathode pole throughout the cardiac cycle at T-1 and T0 were reconstructed and analyzed. Changes of the ratio between its two major axes (S1/S2) between T-1 and T0 (ΔS1/S2), were compared with the volumetric response at six-month f.u: the percent negative variation of S1/S2 (ΔS1/S2 < 0), marking a more multi-directional shape of 3DTJ, was assumed to predict the response to CRT. Volumetric response was adjudicated by a core-lab using a cut-off reduction ≥ 15% in echocardiographic LV end-systolic volume at f.u.. Results Out of 119 patients enrolled in 42 months, 74 pts ended f.u. (55 m; age 69 ± 10) and 30 dropped–out. Patients baseline features were: ischemic heart disease (IHD) 34 /74 pts; sinus rhythm 64/74 pts; upgrade from PM/ICD 13/74 pts; QRS morphology with LBBB 57/74, intraventricular aspecific delay 6 and RV pace 11 pts; LV ejection fraction (EF) 30 ± 9%; QRS duration 162 ± 25 ms. At f.u., volumetric R were 45/74 (60%). Concordance between ΔS1/S2 (as either ΔS1/S2 < 0 or ΔS1/S2 > 0) and volumetric response was 77% overall (57/74), 82% in R (37/45), 69% in non-R (20/29). Non-concordant patients were mostly non-R: 52% vs 35% of non-R in concordant group, but no other differences were found. The proposed 3DTJ metric showed sensitivity = 72%, specificity =80%; positive predictive value = 69%, negative predictive value = 82%. Conclusions Metrics of 3DTJ can be useful to acutely predict CS pacing site-specific response to CRT in long-term, above all in R. 3DTJ assessment might highlight aspects of CRT effects on LV mechanics.