Differentiation between true microembolic signals and artefacts using an arbitrary sample volume

Abstract

We evaluated a new discrimination technique between microemboli (MES) and artefact signals. Monitoring was performed over the middle cerebral artery (55 mm) and the brain parenchyma (29 mm). Intensity changes were expressed as percent of change compared to the value measured in the proximal depth. The cut-off value providing the highest sensitivity and specificity in the differentiation was evaluated using 250 MES and 250 artefact signals, and subsequently analysed in the first part of the study. Intensity values derived from the distal depth were subsequently evaluated in 10 patients undergoing cardiac surgery and 45 patients with potential arterial or cardioembolic source. Intensity changes of 87% (84%–90%) and −58% (−71%–−48%) were measured in the initial 500 signals for MES and artefact signals, respectively. The best intensity cut-off value was calculated at 27%. This value was subsequently applied to a total of 1858 MES and 1958 artefacts, resulting to sensitivity and specificity of 96% and 98%, respectively. The proposed technique provided adequate results, warranting further evaluation.