Send Orders for Reprints to
[email protected] 6524
Current Pharmaceutical Design, 2017, 23, 6524-6532
REVIEW ARTICLE
Intraoperative Seizures: Anesthetic and Antiepileptic Drugs Alberto Uribe1,*, Alix Zuleta-Alarcon1, Mahmoud Kassem1, Gurneet S. Sandhu1 and Sergio D. Bergese1,2 1
Department of Anesthesiology, The Ohio State University Wexner Medical Center, Columbus, OH, USA; 2Department of Neurological Surgery, The Ohio State University Medical Center, Columbus, OH, USA Abstract: Background: Epilepsy is a common condition with up to 1% prevalence in the general population. In the perioperative course of neurologic surgery patients, the use of prophylactic and therapeutic antiepileptic drugs is a common practice. Nonetheless, there is limited evidence supporting the use of prophylactic antiepileptics to prevent perioperative seizures and there are no guidelines for which anesthetic technique is preferred. Objective: To discuss the seizurogenic potential of anesthetic drugs and to discuss intraoperative seizures in neurosurgical patients. ARTICLE HISTORY Received: September 28, 2017 Accepted: October 20, 2017 DOI: 10.2174/1381612823666171024154026
Method: We performed a search of the literature available in PubMed and Ovid MEDLINE. We also included articles identified in the review of the references of these articles. Results: The incidence of seizures is heterogenic among neurosurgical patients. Seizure prophylaxis is widely administered despite limited available evidence of its effectiveness. In epileptic patients, the recommendation is to continue antiepileptic drugs in the perioperative setting. In these patients, anesthesiologists may also limit the use of medications that alter the seizure threshold and avoid medications that pose significant pharmacological interaction with antiepileptic drugs. Conclusion: In conclusion, a knowledgeable multidisciplinary perioperative team is essential to avoid, identify and treat intraoperative seizures competently. In patients with a history of epilepsy it is recommended to continue antiepileptic therapy. Therefore, clinical judgment should guide the decision of administering seizure prophylaxis in neurosurgery patients according to an individual assessment of potential risk for seizures. Furthermore, there is a need for randomized controlled trials that support new protocols and/or guidelines for anesthetic and perioperative regimens to prevent and treat intraoperative seizures.
Keywords: Intraoperative, seizure, epilepsy, antiepileptic drugs, anesthetics, neurosurgery. 1. INTRODUCTION Seizures are episodes of neuronal activity with abnormally enhanced synchrony [1]. Intraoperative seizures (IOS) are a significant cause of morbidity and mortality as they can prematurely interrupt tumor resection, interfere with functional monitoring, cause status epilepticus, postoperative temporary or prolonged neurological deficit, and prolong length of hospital stay [1, 2]. In addition to the aforementioned, intraoperative seizures can also increase cerebral metabolism, oxygen consumption, blood flow and intracranial pressure while also generating brain edema [3, 4]. IOS could be a devastating and life-threatening adverse event during surgery; especially during awake craniotomy (AC) procedures where functional intraoperative electrical cortical and subcortical mapping plays an important role in enhancing the benefit-risk ratio of the surgical procedure through the extent of resection [1, 2, 5, 6]. The incidence of IOS ranges from 0% to 54% among different medical institutions [1, 2, 5]. This wide range could be related to different variables, including, but not limited to, multiple definitions for IOS, varying techniques to detect IOS, anesthesia regimen used, election of AC versus asleep surgery (ASL), electrical stimulation used, surgeon’s experience, and the inherent epileptogenic activity of the tumor [1]. Furthermore, a meta-analysis with fortyfour heterogeneous studies showed an incidence of 8% of IOS in patients undergoing craniotomies, with most of the cases related to cortical stimulation [5]. Seemingly, there is an important *Address correspondence to this author at the Department of Anesthesiology, Department of Anesthesiology, The Ohio State University, Wexner Medical Center Doan Hall N 411, 410 W 10th Ave., Columbus, OH. 43210; Tel: +1(614) 293-8487; Fax: +1(614) 293-8153; E-mail:
[email protected] 1381-6128/17 $58.00+.00
link between mastery of mapping techniques and the incidence of seizures [1, 2]. A study showed that focal and generalized seizures are more common (53.3%) than focal (33.3%) or generalized (13.3%) alone [1]. The incidence of IOS also differs according to the anesthetic protocols; this study reported a higher incidence of IOS in AC than ASL (18.6% versus 8.8%) [1]. Ulkatan et al. reported an incidence of 0.8% during procedures that required motor evoked potential monitoring; however, this monitoring technique may have other benefits and should not be avoided for this reason [7]. Seizure detection in the operating room is challenging. The use of intraoperative electroencephalogram (EEG) for central nervous system monitoring in all patients is not practical and requires the presence of a neurophysiology team. On the other hand, bispectral index (BIS) is widely used for sedative hypnotic-state monitoring. In multiple cases, seizure activity has been evidenced by a sudden reduction in the BIS index value and appearance of low-frequency and high amplitude waves [8]. Some other cases conversely report electrophysiological silence and periodic high-amplitude spikes associated to abnormally high BIS index values or abnormal intraoperative fluctuation [9-11]. In neurosurgical patients, BIS is indicated to monitor the level of consciousness; and although its software is not intended to detect epileptogenic activity (abnormal EEG patterns) [12], sudden abnormal changes in waves and BIS index may indicate abnormal epileptiform activity and should be interpreted carefully. EEG monitoring in high risk patients is an alternative, however, it is important to consider that EEG monitoring might not be sensitive to hypersynchrony in subcortical structures, or spatially limited or brief epileptiform episodes that are better detected by electrocorticography [13-15].
© 2017 Bentham Science Publishers
Intraoperative Seizures: Anesthetic Drugs and Antiepileptics
Table 1.
Current Pharmaceutical Design, 2017, Vol. 23, No. 42
6525
Risk factors for the onset of IOS. Variables
1.
Younger age
2.
Tumor location: frontal and parietal lobes
Source Conte, Carrabba et al. 2015 [2] Conte, Carrabba et al. 2015 [2]
Description Young age upturns the risk of IOS by 4% per year [2]. Frontal and parietal lobe tumor location, especially the supplementary motor area.
Boetto, Bertram et al. 2015 3.
Anesthetic protocol: General Anesthesia
Conte, Carrabba et al. 2015 Stevanovic, Rossaint et al. 2016 Yuan, Peizhi et al. 2016
General Anesthesia maintained with low doses of propofol during brain mapping/tumor removal has been associated with increased frequency of background ECoG activity in epileptic patients [2].
[2, 5, 6, 16] 4.
Hypocapnia: Non-physiologic levels of arterial CO2
Conte, Carrabba et al. 2015 [2]
Conte, Carrabba et al. 2015 5.
Tumor histology
6.
History of recent preoperative seizures
7.
Duration and mastery of brain mapping technique
[2]
Spena G. et al. 2016 [1] Conte, Carrabba et al. 2015 [2] Spena G. et al. 2016 [1] Conte, Carrabba et al. 2015 [2]
2. RISK FACTORS FOR THE ONSET OF INTRAOPERATIVE SEIZURES There are several conditions that the literature describes as perioperative risk factors related to the development of IOS [2, 5, 6, 16]. These factors are described in Table 1. Moreover, Conte, Carrabba et al. identified that younger age, tumors involving frontal and parietal lobes, general anesthesia and hypocapnia were independent risk factors for the onset of IOS [2]. 3. EFFECT OF ANESTHETIC AGENTS ON PERIOPERATIVE SEIZURES 3.1. Anesthetic Agents One of the anticipated mechanisms for the development of anesthetic drug-induced seizures is amplified neuronal synchrony in the thalamo-cortical neurons during anesthesia and sleep. These events occur more commonly during induction and emergence from anesthesia [13, 15]. Several studies demonstrated that without EEG monitoring, myoclonus, dystonic reactions, extreme shivering and other non-epileptic seizure related movements could mimic seizures. Notably, during the induction of anesthesia, unusual movements are a poor indicator of seizure activity evidenced on EEG [19-22]. Many anesthetic drugs have pro-convulsant properties and can induce epileptiform changes on EEG leading to the development of seizure activity. To decrease the incidence of seizures in high-risk patients, the anesthetics with pro-convulsant properties must be identified and avoided [13, 19, 23].
Hypocapnia increases risk for seizures [2]. The risk of seizures in patients with brain tumors is related to the type of the tumor; as examples, oligodendroglioma (81 percent), astrocytoma (66 percent), ependymoma (50 percent), glioblastoma (42 percent), meningioma (40 percent), and metastatic disease (19 percent) [17, 18]. The occurrence of seizures within one month prior to surgery can significantly increase the threat of IOS, regardless of the use of AEDs regimens [1, 2]. Cortical or subcortical stimulus performed between the start of brain mapping until the conclusion of tumor removal could trigger the presence of IOS during craniotomies [2].
Modica et al. evaluated the pro- and anticonvulsant properties of anesthetic and analgesic medications. Multiple anesthetics and analgesics have described pro-convulsant properties; such as inhalational anesthetics, induction anesthetics, and local anesthetics.[24, 25]. 3.1.1. Inhalational Anesthetics • Enflurane: There is imprecise evidences that enflurane induces seizure, but the literature reports EEG spiking activity when is associated with low concentrations of partial pressure of arterial carbon dioxide (PaCO2) [26, 27]. At a normal level of PaCO2, higher amplitude of spiking is noticed at 2-3% of inhaled enflurane; at higher PaCO2 spiking activity is reduced [28]. Lebowitz et al. and Neigh et al. demonstrated that increasing levels of enflurane anesthesia in both normal and epileptic patients stimulates the appearance of high voltage spikes [26, 27]. • Nitrous oxide (N2O): In animal studies, N2O provokes seizures [29]. Artru et al. demonstrated seizure activity after N2O withdrawal confirmed by EEG monitoring [29]. Hornbein et al. observed myoclonus activity in volunteers exposed to hyperbaric N2O [30]. There are non-human studies to confirm that N2O induces seizures. • Sevoflurane: Several studies explored sevoflurane-induced seizures and demonstrated EEG activity and presence of cortical hyper-excitability in both epileptic and non-epileptic patients at low minimal alveolar concentrations (MAC). Spike discharge may be shown at levels greater than 1.5 MAC. They recommend that the level of sevoflorane be maintained below 1.5 MAC to avoid the epileptiform activity in high-risk
6526 Current Pharmaceutical Design, 2017, Vol. 23, No. 42
patients [19, 31-33]. Fentanyl may prevent sevofluraneinduced seizures [34]. There are no reports of halothane or isoflurane related seizures. However, there are a few studies showing epileptiform activity and EEG changes during halothane and sevoflurane anesthesia when combined with N2O [35-38]. 3.1.2. Intravenous Anesthetics • Propofol: The rapid fluctuation of propofol levels in the brain at the onset or emergence of anesthesia may predispose patients to seizures and EEG changes, even at a low level of propofol [39]. Wang et al. noticed that pro- and anticonvulsant mechanisms altered by the dose of propofol are more sensitive to inhibition than excitation on the central nervous system [40]. The propofol-induced seizure activity may be associated with its constant depressant action through potentiation of ᵧ-aminobutyric acid-mediated pre- and post-synaptic inhibition by augmenting inward ᵧ -aminobutyric acid chloride currents and reducing the release of glutamate and aspartate [40, 41]. • Etomidate: A retrospective study done by Hoyer et al. demonstrated that etomidate is responsible for longer seizure duration in patients undergoing electroconvulsive therapy. They considered etomidate, the drug of choice in these patients in order to induce high value spikes on EEG [42]. • Ketamine: Seizure activity or EEG changes may be caused by either, intravenous or intramuscular ketamine injections specifically in epileptic patients in whom ketamine induces subcortical activity [43, 44]. Factors such as nutrition, environment (smoking, alcohol consumption, diet) and drug interaction may contribute to the pro-convulsant effect of ketamine [45]. Ferrer-Allado et al. confirmed that ketamine may induce subcortical EEG activities and is commonly associated with unusual movements and posturing[46]. • Methohexital: The pro-convulsant properties of methohexital may be associated with its temporary and dose-dependent stimulatory effects [24]. Hufnagel et al. and Folkerts H. showed that low doses of methohexital might potentiate epileptiform activities in epileptic patients and occasionally induce seizures in a patient with no previous history [47, 48]. • Opioids: Effective mu opioid receptor agonists can stimulate EEG activity [49] and seizures in epileptic and non-epileptic patients [50-52]. Kofke et al. and Siggins et al. demonstrated that opioid drugs may stimulate the epileptiform activity and EEG changes in animals and humans [53, 54]. Remifentanil, alfentanil, and sufentanil (i.e., short-acting opioids) may be used to induce epileptiform activity during EEG monitoring during focal cortical resection [55-57]. Yillar et al. observed a strong relationship between noticeable myoclonic and clonic seizures and meperidine administration during anesthesia [58]. The etiology of opioid-induced limbic system seizures is still unclear; suggested mechanisms include: selective triggering of limbic opioid receptors, magnified release of stimulatory amino acids (e.g., glutamate), acceleration of coupling between excitatory postsynaptic potentials and somatic spikeactivating sites, or suppression of inhibitory interneurons [25, 53, 54, 59, 60]. Management and prevention of opioidinduced seizures include; enhancement of drug clearance, benzodiazepine use and opioid discontinuation [61]. 3.1.3. Local Anesthetics High systemic concentrations of local anesthetic, which may result from injection into highly vascular areas or inadvertent intravascular injection, may be followed by generalized seizures due to its ability to cross the blood brain barrier [62-64]. Local anesthetic potency is correlated to intravenous central nervous system toxicity (CNS) [65]. CNS toxicity usually occurs before cardiovascular toxicity, although cardiovascular toxicity can occur in the
Uribe et al.
absence of obvious CNS toxicity (patients under general anesthesia or sedated) [66]. In order to decrease local anesthetic systemic toxicity, adequate patient monitoring during anesthesia, usage of low drug concentrations, and adjustment of the dose before drug administration are essential [67, 68]. Patients with local anesthetic systemic toxicity should be ventilated with 100% oxygen and undergo prompt airway management. Benzodiazepines are the preferred drug for seizure suppression. Propofol and thiopental are acceptable options; however, propofol should be avoided in patients with cardiovascular instability. In cases of persistent seizure activity, succinylcholine should be considered to decrease hypoxemia and acidosis. Lipid rescue therapy is an effective antidote to local anesthetic toxicity. An intravenous bolus of 20% lipid emulsion (1.5 ml/kg of lean body mass) should be administered over 1 minute with continuous infusion of 0.25 ml/kg/min. If cardiovascular collapse persists, the bolus can be repeated once or twice, and the infusion can be increased to 0.5 ml/kg/min. After attaining circulatory stability, lipid emulsion should be continued for at least 10 minutes (10 ml/kg is the recommended upper limit for lipid emulsion administration over the first 30 minutes). Cardiac arrhythmias should be managed according to Basic and Advanced Cardiac Life Support with some medication adjustments; individual epinephrine doses should be reduced to 1 and p values
