Changes in myocardial perfusion according to left ventricular pacing site in cardiac resynchronization therapy

FUNDING ACKNOWLEDGEMENTS: Type of funding sources: Foundation. Main funding source(s): 1. The Danish Heart Foundation 2. The Health Research Fund of the Central Denmark Region BACKGROUND/INTRODUCTION: Changes in left ventricular (LV) myocardial perfusion according to LV pacing site during cardiac resynchronization therapy (CRT) has only been sparsely described. PURPOSE: To assess changes in LV global and regional myocardial perfusion according to LV pacing site after six months CRT. METHODS: Data from patients enrolled in the randomized controlled trial "Electrically vs. Imaging-guided Left Ventricular Lead Placement in Cardiac Resynchronization Therapy" was evaluated. Global and segmental absolute myocardial perfusion in ml/g/min was assessed by 82Rb-PET the day before CRT implantation and at six-months follow-up (6MFU). The exact LV pacing site was determined by post-implant cardiac CT. Perfusion and pacing site were evaluated using the standard 17’-segment LV model. Echocardiographic assessment of LV ejection fraction (EF) was performed the day before CRT implantation and at 6MFU. RESULTS: 93 patients (age 71±9 years; ischemic heart disease 48%; female 25%; median QRS 166±20 ms; mean LVEF 30±7%; NYHA class I/II/III/IV (%) 0/63/33/3) were included. Distribution of LV pacing sites is shown in Figure 1. No change in LV global perfusion was observed from baseline to 6MFU (0.79 to 0.81 ml/g/min, P=0.41; respectively). LV EF increased by 9±10%. Baseline global perfusion was higher in the 39 patients in whom LV EF had increased ≥10% at 6MFU (0.86 vs. 0.74 ml/g/min, P=0.02; respectively). However, no difference in perfusion-change was observed between patients with ≥10% increase in LV EF at 6MFU and patients with <10% increase (0.01 vs. 0.02 ml/g/min, P=0.76; respectively). Baseline perfusion was higher in the segments chosen for LV-pacing than in the non-paced segments (0.96 vs. 0.78 ml/g/min, P=0.00; respectively). Perfusion in the paced segments dropped to 0.86 ml/g/min (P=0.00) at 6MFU. Perfusion in the neighbouring segments also dropped during six months (from 0.91 to 0.86 ml/g/min, P=0.03), while perfusion in the non-paced and non-neighbouring segments increased (from 0.75 to 0.80 ml/g/min, P=0.01). CONCLUSION: Despite a mean increase in LV EF of 9%, global myocardial perfusion remained unchanged. CRT resulted in a homogenous distribution of myocardial perfusion with a decreased perfusion in segments concordant and adjacent to LV pacing site and a corresponding increased perfusion in remote segments. [Figure: see text] [Figure: see text]