Objectives: transcranial Doppler (TCD) measurement of middle cerebral artery velocity (MCAV) is an indirect method of assessing cerebral blood flow and therefore predicting patients at risk of stroke during carotid endarterectomy (CEA), and may be used to determine the need for shunting. This study evaluates the accuracy of three accepted TCD criteria in predicting the need for a shunt.
Design: prospective study.
Methods: one hundred and twenty consecutive CEA were performed under loco/regional anaesthesia. Patients monitored by TCD and Awake neurological examination were included. Shunts were inserted if there was neurological deterioration. Awake patient monitoring was compared with the three TCD criteria.
Results: inadequate TCD recordings were obtained in 16 operations (13%). In the remainder (104 cases), 12 developed symptoms of cerebral ischaemia and required a shunt (12%). Comparisons with the three accepted criteria were as follows: (1) m MCAV <30 cm/s had a sensitivity, specificity, PPV and NPV of 92%, 49%, 19%, and 98%, respectively; (2) clamp/pre-clamp ratio <0.6 had a sensitivity, specificity, PPV and NPV of 92%, 75%, 33% and 99%, respectively; (3) greater than 50% reduction in m MCAV had a sensitivity, specificity, PPV and NPV valves of 83%, 77%, 32% and 97%, respectively.
Conclusions: TCD flow velocities are not a reliable method for detecting cerebral ischaemia and therefore determining the need for a shunt in CEA.
Copyright 2001 Harcourt Publishers Limited.