Diagnostic accuracy of somatosensory evoked potentials during intracranial aneurysm clipping for perioperative stroke.

Abstract

Somatosensory evoked potentials (SSEPs) are utilized during aneurysm clipping to detect intraoperative ischemia. We assess the diagnostic accuracy of SSEPs in predicting perioperative stroke during aneurysm clipping. A retrospective review was conducted of 429 consecutive patients who underwent surgical clipping for ruptured and unruptured cerebral aneurysms with intraoperative SSEP monitoring from 2006 to 2013. The relationship between perioperative stroke and SSEP changes was analyzed by calculating the sensitivity, specificity, and area under a Receiving Operating Characteristic curve. Sensitivity and specificity were 42% and 90%, respectively. Area under the curve was 0.66 (95% confidence interval, 0.53-0.79). Reclassification of reversible temporary clip changes to correct for paradoxical classification of SSEP false positives raised the sensitivity from 42 to 65% (p = 0.041, Chi squared test). EEG (electroencephalography) changes increased the specificity (98% vs. 90%, p < 0.001, McNemar's test), but not sensitivity (48% vs. 42%, p = 0.621, McNemar's test) of SSEPs for perioperative stroke. A stepwise logistic regression model selected SSEP amplitude loss (p = 0.006, OR = 3.7 [95% CI 1.5-9.2]) and the SSEP change duration (p = 0.034, OR = 1.8 [95% CI 1.1-3.1]) as independent predictors of perioperative stroke. SSEP changes induced by temporary clipping were highly reversible compared to other SSEP changes (94% vs. 60%, p = 0.003, Fisher exact test), and typically responded to clip removal or readjustment. SSEP changes have high specificity and modest sensitivity for perioperative stroke. Stroke risk is a function of both the magnitude of SSEP amplitude loss and the duration of its loss. Given the modest sensitivity, patients may benefit from multimodal monitoring including motor-evoked potentials during cerebral aneurysm surgery.