Abstracts

Rationale: Intracranial EEG monitoring presents definite risks in addition to those usually encountered in the epilepsy monitoring unit (EMU), including unintentional electrode removal and intracranial hemorrhage. Expert consensus statements recommend the continuous observation of intracranial EEG patients, but this practice is neither standardized nor specific as to the type of observation. We sought to compare seizure response and safety outcomes for intracranial stereotactic EEG (SEEG) patients before and after the adoption of continuous bedside staff monitoring. Methods: We retrospectively studied 25 adult patients who underwent SEEG placement at the Columbia University Medical Center between 5/2014 and 3/2017. In addition to nursing staff monitoring all EMU video recordings at a centralized location, we augmented intracranial patient supervision with a full-time bedside sitter in 5/2016. We divided patients into two groups: those monitored before (pre-sitter; n=13), and after (post-sitter; n=12) this change. We then analyzed nursing response times for recorded clinico-electrographic seizures and characterized adverse outcomes for both groups. Results: We reviewed 106 seizures (60 pre-sitter; 46 post-sitter). 18/60 seizures (16 focal impaired aware, 2 focal to bilateral-tonic clonic) were unrecognized by nursing staff in the pre-sitter group, compared to 8/46 seizures post-sitter (all focal impaired aware), the rates of which did not differ between groups (p=0.13). Electrographic seizure onset to nursing response was 78.4 seconds pre-sitter versus 59.5 seconds post-sitter, also not significantly different (p=0.10). Two patients in the pre-sitter group pulled out their electrodes during periods of peri-ictal confusion, one requiring reimplantation in the operating room. There was one asymptomatic intracranial hemorrhage after initial electrode placement in the post-sitter group, with no other hemorrhages noted in either group. Conclusions: The rates of unrecognized seizures and time to response after electrographic onset were not significantly different between the pre-sitter and post-sitter group, likely related to a small sample size as well as the heterogeneity of clinical seizure manifestations. However, in the pre-sitter group, we noted two unrecognized focal to bilateral tonic-clonic seizures, known to carry a risk of cardiorespiratory depression, as well as two instances of accidental electrode removal, with one patient requiring reimplantation. By contrast, no such events occurred in the post-sitter group. Preventing avoidable adverse events therefore justifies the cost of a full-time bedside staff sitter for intracranial SEEG patients. Funding: None