Unilateral vs. bilateral ultrasound in the monitoring of cerebral microemboli

We used bilateral transcranial Doppler to monitor the number of microembolic events (ME) in the left and right middle cerebral arteries of 29 patients during cardiac surgery that required extracorporeal circulation. Based on a previously published study, we hypothesized that the commonly used method of doubling unilateral ME counts to obtain an estimated bihemispheric load would result in significant errors of estimation. In our sample, estimated bihemispheric counts were inaccurate by an average of 18% (range 0--80%). Despite this large range of error, calculation of Cronbach's alpha revealed that actual error due to unreliability (4%) was small relative to the large variation in ME counts across subjects in this patient series. These findings suggest that unilateral monitoring is sufficient when the goal is to characterize a given subject's ME load within the context of the other subjects in the sample. However, when precise ME counts are required, bilateral monitoring is essential.