All patients (both seizure and control groups) received total IV anesthesia. The dose of propofol and remifentanil at the time of seizure was 159±54.5 and 0.19±0.09 mcg/kg/min, respectively. The majority of patients (82%) were normocapnic (ETCO2=30 to 40 mm Hg) at the time of seizure, whereas 16% were hypocapnic (ETCO2<30 mm Hg) and 3% were hypercapnic (ETCO2>40 mm Hg). Fifty-six percent of the patients who seized were electively kept intubated at the conclusion of surgery at the discretion of the operative team.
All patients underwent intraoperative evoked potential monitoring consisting of SSEPs, MEPs, and EEG. The timing of seizures in relation to MEP runs was variable, (median 57.5 [0 to 142] min), as was the median number of MEP runs before seizure (6 [0 to 72]). Highest MEP stimulation intensity before seizure was also variable, with a median train of 6 (range, 3 to 8), median electrical potential of 365 (range, 160 to 560) V, and median current of 771.5 (range, 332 to 1312) mA. Of note, no seizures were noted in patients undergoing spine surgery with similar intraoperative evoked potential monitoring and TIVA anesthetics during the same period. Figure 4 demonstrates the EEG from a representative case of intraoperative seizure.
To match the 45 patients with intraoperative seizures, 270 control patients without intraoperative seizure were selected using a random number generator. The mean age of the control group was 49.5±14.4 years old (P=0.34, compared with the case group.) The univariate analyses were composed of 4 potential contributing factors, including patient characteristics, preoperative radiologic findings, surgical diagnosis, and approach, and prophylactic anticonvulsant use. The estimates are summarized in Tables 1–3. A history of preoperative seizure was significantly associated with intraoperative seizure (odds ratio [OR], 2.18; confidence interval [CI], 1.07-4.46; P=0.03), whereas age, sex, other investigated medical conditions, and preoperative radiologic findings (lesion size, mass effect, midline shift, hydrocephalus, and cerebral edema) were not associated with intraoperative seizures. Brain tumor diagnosis was associated with higher likelihood of intraoperative seizure compared with arteriovenous malformation, aneurysm, and vascular occlusion. The difference was significant between brain tumor and other diagnoses altogether (surgical diagnosis analysis 2 in Table 2: OR, 2.41; 95% CI, 1.16-4.19; P=0.02). Likewise, temporal craniotomy were associated with the highest likelihood of intraoperative seizure and the difference was significant between temporal and other approaches altogether (surgical approach analysis 2 in Table 2: OR, 5.18; 95% CI, 2.03-13.25; P=0.001). There was no significant difference in intraoperative seizures either in patients taking anticonvulsants preoperatively compared with those who did not (P=0.45), or in patients receiving local anesthetics before incision compared with those who did not (P=0.55). However, prophylactic intraoperative use of phenytoin or fosphenytoin was associated with a significant reduction in intraoperative seizures compared with levetiracetam (OR, 0.31; 95% CI, 0.10-0.99; P=0.048) or no use of intraoperative prophylaxis (OR, 0.12; 95% CI, 0.04-0.35; P<0.001). Among the patients with tumors (24 cases and 93 controls), there was no statistically significant difference in the histopathologic diagnosis of the tumor.
Multivariate logistic regression (Table 4) demonstrated that after controlling for the other significant risk factors identified in the univarite analyses, seizure history (OR, 2.76; 95% CI, 1.22-6.23; P=0.02), diagnosis of brain tumor (OR, 2.15; 95% CI, 1.04-4.46; P=0.04), and temporal craniotomy approach (OR, 6.57; 95% CI, 2.17-19.85; P=0.001) remained significant. In addition, prophylactic intraoperative use of phenytoin/fosphenytoin and levetiracetam were found to be protective against intraoperative seizure compared with no use of prophylactic anticonvulsants (phenytoin/fosphenytoin: OR, 0.09; 95% CI, 0.03-0.29; P<0.001 and levetiracetam: OR, 0.34; 95% CI, 0.14-0.87; P=0.03). When compared with each other, phenytoin/fosphenytoin was significantly more protective than levetiracetam (OR, 0.28; 95% CI, 0.08-0.98; P=0.046). The result is robust to the sensitivity test using P-value of 0.1 as the criterion of forward selection.
In summary, the overall incidence of intraoperative seizures during elective supratentorial craniotomy with evoked potential monitoring during the study period was 2.3%. A history of seizures, diagnosis of intracranial tumor, and temporal craniotomy approach were associated with higher risk of intraoperative seizure, whereas intraoperative prophylactic anticonvulsant use was protective; although phenytoin/fosphenytoin provided better protection than levetiracetam or no prophylaxis. To the best of our knowledge, this study provides the first estimates of this intraoperative complication and identifies the risk factors for the same.
Intraoperative seizures during cranial surgery may be detrimental. In a previous report, Howe et al,5 noted a 0.5% incidence of intraoperative seizures in a cohort of 400 patients with EEG monitoring. Although specific anesthetic and monitoring details were not provided, the seizures in their series were identified on EEG and the patients were not reported to receive evoked potential monitoring. Electrical stimulation of the brain has been reported to result in a clinical seizure.11,12 However, the seizure-inducing current amplitude thresholds are typically 2 to 3 orders of magnitude above the maximum tcMEPs and evoked potential monitoring is generally considered safe.11,12 In fact, tcMEPs during spinal deformity surgery do not appear to trigger intraoperative seizures in patients suffering seizures preoperatively.13 It has been demonstrated that a craniotomy affects current spread from a MEP while this is not seen in spinal surgery.14 However, in a recent study, Ulkatan et al,15 investigated the incidence of seizures during the intraoperative monitoring of MEPs in a wide spectrum of surgeries including orthopedic spine, spinal cord, and peripheral nerves, interventional radiology procedures, and craniotomies. In this mixed surgical population cohort of 4179 patients, 32 (0.8%) had intraoperative seizures and the incidence of seizures in cranial procedures was 1.8%.15 None of the patients who underwent surgery for conditions of the spinal cord, neck, or peripheral nerves or who underwent cranial or noncranial interventional radiology procedures had intraoperative seizures. Although the incidence of intraoperative seizures was low, the study did indicate a relatively higher risk during craniotomy.15 Our findings add to the existing literature by specifically providing estimates of seizure during supratentorial craniotomy with evoked potential monitoring and identifying risk factors for the same.
It is not surprising that a history of seizure put patients at increased risk for intraoperative seizures, especially during intracranial procedures. It has been previously demonstrated that patients with seizure disorders are at increased risk of perioperative seizures even when intracranial procedures are excluded from analysis.16 Factors that may play a role in this increased risk of intraoperative and perioperative seizures include missed doses of antiepileptic medications, medication interactions, administration of medications that lower the seizure threshold, altered gastrointestinal absorption, and sleep deprivation.16 Although a history of seizure was an independent risk factor for intraoperative seizures, the majority of patients in our study group did not have a history of seizures. Anesthesiologists should remain vigilant about the possibility of intraoperative seizure during craniotomy with MEP monitoring even in patients without a seizure history.
Wong et al2,3 reported the incidence of perioperative seizures following craniotomy for a variety of pathologies, based on literature review. They found that early postoperative seizures occurred in 1% to 12% of patients with intracranial tumors,2 while 4% to 42% of patients undergoing craniotomy for aneurysm experienced a perioperative seizure.3 Others have observed perioperative seizures in 15% to 50% of patients undergoing brain tumor suergery.4 It is hypothesized that tumor-related seizures are due to structural and metabolic derangements related to the mass itself, as well as the subsequent craniotomy that is often necessary for diagnosis and treatment.4 We found that patients with intracranial tumors had a higher risk of seizures during surgery (OR, 3.04; 95% CI, 1.44-6.44; P=0.004). This may have implications for administration of prophylactic intraoperative anticonvulsants. Practice guidelines published in 2000 by the American Academy of Neurology advised that antiepileptic medications are not effective in preventing first seizures in patients with newly diagnosed brain tumors and those patients who are started on antiepileptics in the perioperative period should be quickly tapered off of these medications.7 These recommendations were based on a meta-analysis that found limited or no benefit to the use of prophylactic anticonvulsants in the postcraniotomy setting, particularly in those patients without preexisting seizures.4,7 However, our finding may warrant consideration for prophylactic anticonvulsants to prevent intraoperative seizures during temporal craniotomy with evoked potential monitoring in patients with supratentorial tumors who have a history of seizures.
We also found temporal craniotomy to be an independent risk factor for intraoperative seizure. The surgical approach may, in fact, be a surrogate for the location of the intracranial pathology. The association of frontal or temporal lobe tumors with increased risk of perioperative seizures appears to corroborate our observation.17–19 The majority of intraoperative seizures identified in the current study occurred before brain or lesion manipulation, suggesting direct cortical stimulation was not the cause of the seizure. Rather, the seizure threshold may have been lowered due to factors related to the lesion itself, such as the production of inflammatory biomarkers, changes in neurotransmitter activity, or regional metabolic changes resulting from intracranial pathology.17
Levetiracetam has been shown to have a more favorable side effect profile than phenytoin,20,21 which has led to a reduction in the use of phenytoin for seizure prophylaxis in the perioperative period. Specifically, phenytoin is capable of cytochrome P450 hepatic enzyme induction and can thus interfere with the efficacy and metabolism of many other medications, including corticosteroids and chemotherapeutic agents often used in the treatment of brain tumors.4,21,22 Levetiracetam undergoes primarily renal metabolism and thus, has limited impact on other coadministered medications.4 Studies evaluating the efficacy of levetiracetam versus phenytoin in the prevention of postoperative seizures following craniotomy have had mixed results, with some studies finding no significant difference in efficacy20,22 and others showing levetiracetam to be more effective.21 However, we found prophylactic phenytoin or fosphenytoin to be more effective than levetiracetam in preventing intraoperative seizures. Although levetiracetam may be less effective in this setting due to possible underdosing (fixed dose administration as opposed to weight based) or renal clearance, the reason for this is unclear and warrants further investigation. Of note, levetiracetam has been reported to be associated with a paradoxical increase (>25%) in seizure frequency in patients with refractory epilepsy.23
There are several limitations of this study. First, it is a single institution, retrospective study and patients were not randomized into treatment groups. Data collection is subject to potential documentation bias. Perioperative and intraoperative environments were not standardized. The administration of prophylactic anticonvulsants was dependent on the providers involved in the case and plasma levels of anticonvulsant agents were not checked intraoperatively. Therefore, underdosing of anticonvulsant agents cannot be ruled out. Confounding factors that were not measured may have affected the results of this study. In addition, the incidence of the outcome of interest (intraoperative seizures) is small and the limited EEG montage does not allow definitely excluding movement artefact. A larger study population would provide better statistical power for a more robust analysis of potential risk factors.
The incidence of intraoperative seizures during elective craniotomy with evoked potential monitoring is low. Patients are at increased risk of intraoperative seizure if they have a history of seizure, a tumor diagnosis, or are undergoing a temporal craniotomy. The risk of intraoperative seizure may be reduced with the administration of prophylactic intraoperative phenytoin or fosphenytoin compared with no prophylaxis or levetiracetam.
1. Klimek M, Dammers R. Antiepileptic drug therapy in the perioperative course of neurosurgical patients. Curr Opin Anaesthesiol. 2010;23:564–567.
2. Wong JM, Panchmatio JR, Ziewacz JE, et al. Patterns in neurosurgical adverse events: intracranial neoplasm surgery. Neurosurg Focus. 2012;33:E16.
3. Wong JM, Ziewacz JE, Ho AL, et al. Patterns in neurosurgical adverse events: cerebrovascular neurosurgery. Neurosurg Focus. 2012;33:E15.
4. Sayegh ET, Fakurnejad S, Oh T, et al. Anticonvulsant prophylaxis for brain tumor surgery: determining the current best available evidence. J Neurosurg. 2014;121:1139–1147.
5. Howe J, Lu X, Thompson Z, et al. Intraoperative
seizures during craniotomy
under general anesthesia
6. Diringer MN, Bleck TP, Hemphill JC, et al. Critical care management of patients following subarachnoid hemorrhage: recommendations from the Neurocritical Care Society’s multidisciplinary consensus conference. Neurocrit Care. 2011;15:211–240.
7. Glantz JM, Cole BF, Forsyth PA, et al. Practice parameter: anticonvulsant prophylaxis in patients with newly diagnosed brain tumors. Report of the quality standards subcommittee for the American Academy of Neurology. Neurology. 2000;54:1886–1893.
8. Mikkelsen T, Paleologos NA, Robinson PD, et al. The role of prophylactic anticonvulsants in the management of brain metastases: a systematic review and evidence-based clinical practice guideline. J Neurooncol. 2010;96:97–102.
9. Siomin V, Angelov L, Li L, et al. Results of a survey of neurosurgical practice patterns regarding the prophylactic use of anti-epilepsy drugs in patient with brain tumors. J Neurooncol. 2005;74:211–215.
10. Pearce N. Analysis of matched case-control studies. BMJ. 2016;352:i969.
11. Macdonald DB. Intraoperative
motor evoked potential monitoring: overview and update. J Clin Monit Comput. 2006;20:347–377.
12. Sackeim HA, Devanand DP, Prudic J. Stimulus intensity, seizure
threshold, and seizure
duration: impact on the efficacy and safety of electroconvulsive therapy. Psychiatr Clin North Am. 1991;14:803–843.
13. Salem KM, Goodger L, Bowyer K, et al. Does transcranial stimulation for motor evoked potentials (TcMEP) worsen seizures in epileptic patients following spinal deformity surgery? Eur Spine J. 2016;25:3044–3048.
14. Tomio R, Akiyama T, Toda M, et al. The impact of several craniotomies on transcranial motor evoked potential monitoring during neurosurgery. J Neurosurg. 2017;127:543–552.
15. Ulkatan S, Jaramillo AM, Tellez MJ, et al. Incidence of intraoperative
seizures during motor evoked potential monitoring in a large cohort of patients undergoing different surgical procedures. J Neurosurg. 2017;126:1296–1302.
16. Niesen AD, Jacob AK, Aho LE, et al. Perioperative seizures in a patients with a history of seizure
disorder. Anesth Analg. 2010;111:729–735.
17. Gonen T, Grossman R, Sitt R, et al. Tumor location and IDH1 mutation may predict intraoperative
seizures during awake craniotomy
. J Neurosurg. 2014;121:1133–1138.
18. Chang EF, Potts MB, Keles GE, et al. Seizure
characteristics and control following resection in 332 patients with low-grade gliomas. J Neurosurg. 2008;108:227–235.
19. Stockhammer F, Misch M, Helms HJ, et al. IDH1/2 mutations in WHO grade II astrocytomas associated with localization and seizure
as the initial symptom. Seizure
20. Milligan TA, Hurwitz S, Bromfield EB. Efficacy and tolerability of levetiracetam
after supratentorial neurosurgery. Neurology. 2008;71:665–669.
21. Pourzitaki C, Tsaousi G, Apostolidou E, et al. Efficacy and safety of prophylactic levetiracetam
in supratentorial brain tumor surgery: a systematic review and meta-analysis. Br J Clin Pharmacol. 2016;82:315–325.
22. Lim DA, Tarapore P, Chang E, et al. Safety and feasibility of switching from phenytoin
monotherapy for glioma-related seizure
control following craniotomy
: a randomized, phase II pilot study. J Neurooncol. 2009;93:349–354.
23. Nakken KO, Eriksson AS, Lossius R, et al. A paradoxical effect of levetiracetam
may be seen in both children and adults with refractory epilepsy. Seizure
Keywords:Copyright © 2019 Wolters Kluwer Health, Inc. All rights reserved
intraoperative; seizure; craniotomy; anesthesia; phenytoin; fosphenytoin; levetiracetam