Assessment of 3D motion increases the applicability of accelerometers for monitoring left ventricular function†

OBJECTIVES: Miniaturized accelerometers attached to the epicardium have been shown to provide useful clinical information. However, attachment of such a sensor has been cumbersome due to requirement of aligning the three sensor axes with the cardiac coordinate axes, limiting clinical utility. We propose a new method to process the three-dimensional (3D) accelerometer signal that does not require such alignment. METHODS: In 20 open-chest pigs, miniaturized 3D accelerometers were fixated on the epicardium in apical and basal regions of left ventricle. Accelerations in circumferential, longitudinal and radial directions were measured and a 3D velocity vector was calculated. Systolic velocity along the 3D vector and velocities in circumferential, longitudinal and radial directions were compared with the positive time derivate of left ventricular pressure during changes in global left ventricular function (epinephrine, esmolol and fluid loading) and to strain echocardiography during left anterior descending artery occlusion. RESULTS: Distinct changes in all accelerometer velocities were observed during alterations on global and regional left ventricular function. Accelerometer 3D and circumferential systolic velocities in apical region best reflected left ventricular function during interventions on global function by correlating significantly with the positive time derivate of left ventricular pressure, r = 0.83 and r = 0.86, respectively. The accelerometer 3D velocity also demonstrated equally good capacity as circumferential velocity in discriminating coronary occlusion from interventions on global left ventricular function with sensitivity/specificity of 0.90/0.83 and 0.90/0.86, respectively. CONCLUSIONS: Accelerometer 3D systolic velocity showed very good correspondence to changes in global and regional left ventricular function. Our results demonstrate that by the use of the accelerometer 3D motion vector, no alignment of the sensor with the cardiac coordinate axes was required. This increases potential clinical applicability of the accelerometer in cardiac surgery.