Transcranial Doppler Monitoring in Carotid Endarterectomy: A Systematic Review and Meta‐analysis

Objectives To evaluate the efficacy of intraoperative transcranial Doppler monitoring in predicting perioperative strokes after carotid endarterectomy (CEA). Methods An electronic search of PubMed, Embase, and Web of Science databases was conducted for studies on transcranial Doppler monitoring in CEA published from January 1970 through September 2015. All titles and abstracts were independently screened on the basis of predetermined inclusion criteria, which included randomized clinical trials and prospective or retrospective cohort reviews, patients who underwent CEA with intraoperative transcranial Doppler monitoring (either middle cerebral artery velocity [MCAV] or cerebral microembolic signals [MES]) and postoperative neurologic assessments up to 30 days after the surgery, and studies including an abstract, published in English on adult humans 18 years and older with a sample size of 50 or greater. Results A total of 25articles with a sample population of 4705 patients were analyzed. Among the study patients, 189 developed perioperative strokes. Transcranial Doppler monitoring (either MCAV or MES) showed specificity of 72.7% (95% confidence interval [CI], 61.2%–81.8%) and sensitivity of 56.1% (95% CI, 46.8%–65.0%) for predicting perioperative strokes. Intraoperative MCAV changes during CEA showed strong specificity of 84.1% (95% CI, 74.4%–90.6) and sensitivity of 49.7% (95% CI, 40.6%–58.8) for predicting perioperative strokes. Conclusions Patients with perioperative strokes are 4 times more likely to have had transcranial Doppler changes (either MCAV or MES) during CEA compared to patients without strokes. Simultaneous MCAV and MES monitoring by transcranial Doppler sonography and combined intraoperative monitoring of transcranial Doppler sonography with somatosensory evoked potentials and electroencephalography during CEA to predict perioperative stroke could not be evaluated because of a lack of clinical studies combining these measures.