High level of cerebral microembolization in patients supported with the DeBakey left ventricular assist device

Microembolic signals detected by transcranial Doppler ultrasonography have been demonstrated to be clinically relevant in patients supported with pulsatile left ventricular assist devices. We prospectively investigated the quantity of microembolic signals in patients supported with the continuous-flow DeBakey left ventricular assist device (MicroMed DeBakey VAD; MicroMed Technology, Inc, Houston, Tex) including the refined Carmeda BioActive Surface system (Carmeda AB, Stockholm, Sweden).

Twenty-three patients (20 male) aged 14 to 62 years supported with DeBakey left ventricular assist devices (n = 6 with Carmeda) were enrolled in this study. Microembolic signal monitorings were performed twice weekly by insonating the middle cerebral artery for 20 minutes without and 20 minutes with oronasal application of oxygen (6 L/min). Evidence of clinically manifest thromboembolic events was based on regular questionnaires, clinical examinations, and results of diagnostic procedures.

Despite a low incidence of thromboembolic complications (0.24 per 100 left ventricular assist device days), 20 patients (87%) showed circulating microemboli. Overall, microembolic signals were found in 175 of 499 transcranial Doppler ultrasonographic examinations (35.1%), with mean counts of 81.2 +/- 443 (range 0-5042 signals/h). Both microembolic signal prevalence (25% vs 34%, P = .01) and absolute signal counts (46.5 vs 104, P < .01) significantly declined with oxygen delivery. There was no significant correlation between the individual microembolic signal activity and the incidence of clinical thromboembolism or the intensity of antihemostatic treatment. Patients supported with the Carmeda device did not show reduced rates of clinical thromboembolization or cerebral microemboli.

In patients with DeBakey left ventricular assist devices, a high load of clinically silent microemboli can be detected within the cerebral arteries despite a low incidence of embolic complications. It needs to be investigated whether such continuous, presumably gaseous microembolization causes cognitive or neuropsychologic deficits.