Transcranial Doppler detected cerebral microembolism following carotid endarterectomy

We were interested to read the prospective study by Levi et al. (1991) describing the transcranial Doppler (TCD) detection of emboli in the early postoperative phase of carotid endarterectomy. Their finding that embolization was significantly associated with the development of postoperative strokes strongly supports the findings in our own prospective study of perioperative embolization which the authors kindly referred to in their article (Gaunt et al., 1994). The results from their well-conducted study provide further confirmation of the importance of the early postoperative phase in the occurrence of perioperative strokes. However, it should not be ignored that TCD can detect and help prevent strokes occurring during all the phases of the operation, especially embolization of unstable plaque during the initial dissection of the carotid artery and haemodynamic stroke resulting from either inadequate cerebral blood supply at the time of clamping or shunt malfunction (Gaunt e al., 1994). There are two essential features of an effective monitoring method: (i) it can detect the majority of abnormalities; and (ii) abnormalities are detected soon enough to enable corrective action to prevent permanent neurological damage (Gauntet al., 1997). TCD detects embolization in the postoperative phase before the development of neurological deficits, and in our original study we advocated immediate reoperation to remove the thrombus from the carotid artery and prevent further embolization (Gaunt et al., 1994). However, the possibility that this may represent an overtreatment in some cases, and occasional practical difficulties in returning a patient to theatre, prompted us to investigate the alternative therapy of dextran 40 infusion. Our centre recently completed a prospective study of 100 consecutive patients undergoing carotid endarterectomy investigating the use of dextran 40 to arrest postoperative embolization detected by TCD monitoring. Overall, 48% of patients experienced emboli in the 6-h postoperative monitoring period; however, only 8% had .50 and 5%.100 emboli. An incremental dextran infusion was instituted in all patients with.25 emboli detected in any 10-min period of monitoring and this resulted in cessation of embolization in