Abstract Body (Do not enter title and authors here): Background: Left ventricular (LV) efficiency is a key pathophysiological marker in heart failure (HF). LV pressure curve estimation and three-dimensional (3D) volumes via echocardiography allows for non-invasive pressure-volume (PV) analysis and calculation of an index of efficiency (Figure 1).

Aim: Validate efficiency index by 3D echocardiography by comparing it to invasive measurements and efficiency derived from Positron Emission Tomography (PET) metabolism.

Methods: Two experimental models were utilized - one with canines (n=21) and another with sheep (n=12). In canines, echocardiography provided 3D LV volumes and valve event timings. The pressure curve was estimated by adjusting the time axis of a reference pressure curve to the valve events and its amplitude to measured peak LV pressure, which in clinical practice would be approximated as systolic brachial cuff pressure. Invasive pressure was measured by micromanometer and volume by subendocardial piezoelectric crystals. Non-invasive and invasive efficiency indices, calculated as the ratio of stroke work to total mechanical energy (Figure 1), were compared during baseline, right ventricular pacing, aortic constriction and dobutamine infusion.

Twelve sheep implanted with a pacemaker underwent 8 weeks of rapid dyssynchronous DDD pacing to induce dilated HF. LV efficiency index was calculated as in Figure 1, using measurements from a PV-catheter calibrated to end-diastolic volume by cardiac MRI, and compared to efficiency calculated as stroke work divided by total LV glucose metabolism via PET/CT. Recordings were performed during AAI and DDD pacing, at baseline (n=3) and following 8 weeks pacing-induced HF (n=12).

Results: In canines, non-invasive and invasive efficiency indices showed excellent agreement (r=0.95, p<0.0001; mean difference 0.4%, SD 3.4%) (Figure 2). In sheep, the efficiency index showed similar decline compared to efficiency by PET/CT after induction of HF and after switching from AAI to DDD pacing (p<0.05 for all interventions). Overall correlation between the methods was moderate (r=0.67, p=0.0001) but excellent in animals with longitudinal data (r=0.89, p=0.0001) (Figure 3).

Conclusions: Efficiency index by 3D echocardiography is accurate and reflects changes in efficiency measured by PET/CT. It may thus be of interest for clinical use.