Objective
Although the application of evoked potential (EP) in intracranial aneurysm surgery has been well demonstrated, the relationship between the duration of EP deterioration and postoperative neurological deficits (PNDs) is still not clear. The objectives of this study were (1) to evaluate the sensitivity, specificity, and predictive value of intraoperative somatosensory evoked potential (SSEP) and motor evoked potential (MEP) monitoring for predicting postoperative stroke after intracranial aneurysm surgery; (2) to detect the relationship between the EP deterioration duration and PND.
Methods
We reviewed the results of intraoperative SSEP and MEP monitoring in 728 patients who underwent intracranial aneurysm neck clipping between June 2013 and December 2018. Postoperative stroke was diagnosed using a medical record about the motor and sensory deficit, aphasia and mental status.
Results
The mean age of the 728 patients was 59.7 years and consisted of 485 women (66.6%). Subarachnoid hemorrhage were 42 cases and non-ruptured aneurysm were 686 cases. The change of SSEP occurred in 9 out of 728 cases (1.24%), and the change of MEP in 14 out of 728 cases (1.92%). The sensitivity, specificity, PPV and NPV were 38.5%, 99.4%, 55.6% and 98.9%, respectively when there was only SEP change, and 46.2%, 98.9%, 42.9%, 99.0%, respectively when there was only MEP change.
We analyzed data from 23 patients with EP deterioration. Four patients presented irreversible change and 19 patients presented reversible change. In the patients with deteriorated SSEP, 5 patients had PNDs and the mean SSEP deterioration duration was 48 ± 43 min (mean standard deviation, here and elsewhere). Four patients did not have PNDs, and mean SSEP deterioration duration was 6 ± 4 min. In the patients with deteriorated MEP, there were 6 patients with PNDs and the mean duration of MEP deterioration was 63 ± 39 min. Eight patients had no PNDs, and mean MEP deterioration duration was 24 ± 30 min. There was no statistically significant difference between the patients with and without PND (p-value=0.30), but the MEP deterioration duration showed statistical difference between patients with and without PND (p-value=0.05). The optimal cut-off value of MEP change duration avoiding PND was 14 min (area under the curve = 0.867, sensitivity 100%, specificity 66.7%).
Conclusion
Intraoperative MEP monitoring can provide higher sensitivity for postoperative stroke than SSEP monitoring. We can conclude that the duration of MEP deterioration is important to postoperative neurological function, and in order to avoid PND, the MEP deterioration duration must not exceed 14 min.
|
