HTLV-I associated myelopathy in Brazil: Clinical and epidemiological features

Abstract

Periprocedural stroke after carotid endarterectomy increases long-term mortality. Intraoperative monitoring with electroencephalography (EEG) and somatosensory-evoked potentials (SSEPs) helps predict perioperative stroke risk. However, the sensitivity of each technique when used independently still remains low. The aim of this study is to determine whether multimodal monitoring leads to an increase in sensitivity and diagnostic accuracy.Relevant literature was obtained through a search of Embase, PubMed, and Web of Science databases and data were extracted. Data from the University of Pittsburgh Medical Center hospital records for the 2000-2012 period were included. Pooled estimates of sensitivity, specificity, and diagnostic odds ratio were obtained for single and multimodality neurophysiologic monitoring. A McNemar test was used to evaluate for any statistically significant differences in the sensitivities and false-positive rates.The diagnostic odds ratio of dual modality monitoring was found to be 17.4. The specificity of concurrent EEG and SSEP changes in predicting perioperative strokes was calculated to be 96.8% (95% confidence interval 94.1%–98.3%). The sensitivity of combined monitoring with a change in either modality designated as significant was 58.9% (95% confidence interval 41.2%–74.7%). Multimodality monitoring with a change in either EEG or SSEP as the alarm criteria was 1.32 times more sensitive than EEG alone and 1.26 times more sensitive than SSEP alone.The odds of having a change in either EEG or SSEP are 17 times more in patients with perioperative strokes. Dual modality monitoring is more sensitive at predicting perioperative deficits than EEG or SSEP used independently.