Keywords
bispectral index - burst suppression - depth of anesthesia - electroencephalogram
- Intraoperative seizure
Introduction
Seizure is a common manifestation of brain tumors with a reported incidence as high
as 30 to 100%.[1] Around 20 to 50% of patients with supratentorial meningioma present with seizure
as the only symptom before being diagnosed radiologically.[2] In addition, patients may present with seizures for the first time during the intraoperative
period. Various risk factors for intraoperative seizures include previous history
of seizure, low grade lesions, frontal or temporal lobe lesions, and transcranial
motor evoked potential monitoring.[3]
[4] Though the exact incidence of intraoperative seizure under general anesthesia is
not known, the reported incidence lies between 1.8 and 2.3% among patients undergoing
motor evoked potential monitoring or elective supratentorial craniotomy.[4]
[5] Intraoperative seizures can lead to brain bulge, increased cerebral metabolic oxygen
demand, hypoxic brain damage, and hemodynamic instability. Moreover, if undetected,
they can delay the emergence from anesthesia. Therefore, early detection and prompt
termination are crucial. Conventional electroencephalogram (EEG) has limitations in
the intraoperative period, which hinders the operative field. Bispectral index (BIS)
can add a new dimension to intraoperative neuromonitoring in patients at risk of seizure.
Apart from indices for the depth of anesthesia monitoring, it also displays the real-time
raw EEG waveform, thus contributing to the diagnosis of intraoperative seizure, and
guiding the therapeutic achievement of burst suppression. Its placement on the forehead
also does not interfere with the operative field. Our case series illustrates how
BIS, along with its raw EEG display, can facilitate early detection and management
of intraoperative seizures.
Anesthesia Protocol
To facilitate the intraoperative evoked potential monitoring, total intravenous anesthesia
with propofol target-controlled infusion (TCI) (3 mcg/mL) and fentanyl (3–4 mcg/kg)
was used for induction and propofol (TCI titrated to target BIS value) with dexmedetomidine
(0.5 mcg/kg/hr) and ketamine (0.25 mg/kg/hr) infusion was used for maintenance of
anesthesia. Rocuronium (1 mg/kg) was used as a single bolus to aid the endotracheal
intubation. A scalp block with 15 mL of 0.2% ropivacaine and 4 mg dexamethasone was
performed prior to the Mayfield head frame application. Along with standard intraoperative
monitoring, a frontal BIS electrode (BIS Quatro Sensor, Medtronic, India) was applied
on the contralateral forehead to monitor the depth of anesthesia with a target BIS
value of 40 to 60.
Case 1
A 66-year-old female of American Society of Anesthesiology class 2 with right suprasellar
and left temporal meningioma was scheduled for elective craniotomy and excision of
both lesions under evoked potential monitoring. She did not have any previous seizure
history. During surgical wound closure, prior to any reduction in depth of anesthesia
(TCI effect site concentration 2.7 mcg/mL) there was a sudden spike in BIS value from
40 to 45 to 85 to 95 ([Fig. 1A]) along with appearance of high-amplitude waves followed by generalized tonic-clonic
seizure (GTCS), which was terminated using midazolam 4 mg and propofol bolus, leading
to electrographic supersession on BIS. The seizure activity was concurrent with a
sharp rise in EtCO2 (> 50), blood pressure ([Fig. 1A]), and changes in arterial blood gas (ABG) (raised PaCO2 and lactate levels with normal blood sugar level). Any increase in the BIS electrode
impedance was also ruled out. The patient also received an additional dosage of levetiracetam
1.5 g. Subsequently, no seizure activity was noted in the postoperative period, and
the patient was extubated on the next day in the intensive care unit (ICU) with uneventful
further recovery.
Fig. 1 Image showing intraoperative vitals trends of cases 1 and 2. (A) Case 1 with black arrow showing rise in EtCO2, white arrow showing rise in bispectral index (BIS) value on vital trends, and red
arrow showing suppression of electrographic activity on raw electroencephalogram (EEG).
(B) Case 2 with red arrow showing high-amplitude, low-frequency waveform pattern on
raw EEG during intraoperative seizure.
Case 2
A 51-year-old female presented for resection of left parieto-occipital parasagittal
meningioma without any history of previous seizures. Following craniotomy and prior
to dura opening without any reduction in the depth of anesthesia (propofol TCI effect
site concentration 3 mcg/mL), the patient developed GTCS, which was associated with
a rise in BIS value followed by the appearance of high-amplitude, low-frequency waveform
on the raw EEG ([Fig. 1B]). It was terminated using midazolam 4 mg and burst suppression on BIS using propofol
bolus. The rest of the intraoperative period was uneventful, and the patient was extubated
in the evening on the same day when fully awake.
Case 3
A 70-year-old male with renal cell carcinoma and secondary right parietal brain metastasis
was scheduled for craniotomy and resection of the lesion under evoked potential monitoring.
Intraoperatively, during resection of the lesion, there was the appearance of high-amplitude
waves with a sudden spike in the BIS value ([Fig. 2A] and [B]), followed by clinically evident GTCS and associated brain bulge, hypertension,
tachycardia, and hypercarbia ([Fig. 2B]). Any decrease in the depth of anesthesia (TCI effect site concentration 3.0 mcg/mL)
or increase in the BIS electrode impedance was also ruled out. The seizure was terminated
immediately using midazolam 4 mg and propofol bolus, followed by an additional dose
of levetiracetam. Brain bulge was also settled following head-end elevation, hyperventilation,
and electrographic suppression of BIS. ABG showed a normal metabolic profile with
low PaCO2 secondary to hyperventilation. The subsequent perioperative period was uneventful,
and the patient was extubated on the same day in the ICU upon regaining complete consciousness.
Fig. 2 Image showing intraoperative vital with trend in case 3. (A) Red arrow shows rise in bispectral index (BIS) value and white arrow shows electrographic
activity in raw electroencephalogram (EEG). (B) Black arrow showing rise in EtCO2 and red arrow showing rise in BIS trend, and blue arrow show suppression of electrographic
activity.
Discussion
Conventional methods to detect seizures have relied mainly upon continuous EEG monitoring.[6] However, they have limitations of bulky setup, low temporal or spatial resolution,
power line interference, and most importantly, hindrance in the operative field.[6] Therefore, intraoperative seizure has been an underdetected or underreported event
in literature.
BIS is an EEG-based algorithm to monitor the depth of anesthesia. Apart from other
indices, it also displays the real-time raw EEG waveform.[7] Although the EEG is reflective of a selected part of the hemicranium, it can be
valuable in the diagnosis of an intraoperative seizure using various waveform patterns
such as spikes, spike-wave, and poly-spike patterns, as well as to guide the therapeutic
achievement of burst suppression.[8] An increase in the BIS value during intraoperative seizure has also been reported,
which may result from an increase in the high amplitude and frequency component of
EEG.[9]
[10] This increase in BIS could also be a result of increased electromyographic (EMG)
activity; however, increased BIS value in the presence of neuromuscular blockade has
also been reported previously.[11]
[12]
[13] In addition, EMG activity can reduce the probability of encountering classical patterns
of epileptogenic waveform. Iturri Clavero et al and Srinivas et al reported high-amplitude
waves during intraoperative seizures similar to our case, which disappeared after
antiepileptic drug administration.[9]
[10] A higher BIS/SR (suppression ratio) has been used to differentiate epileptic and
normal bursts in patients with refractory status epilepticus under propofol sedation.[14]
Various anesthetic agents have variable effects on seizure threshold and BIS value
at different doses. Ketamine is considered proepileptic, but when given along with
propofol, it has antiepileptic properties. Ketamine bolus may also increase BIS value,
which occurs several minutes after administration. However, this effect on BIS is
not seen when ketamine is administered as a slow infusion under propofol TCI anesthesia.[15] Dexmedetomidine has also been shown to reduce seizure threshold in rat models, but
this effect has not been noted in humans.[16]
Although BIS monitor is not Food and Drug Administration-approved as a diagnostic
tool for seizures, BIS value, along with the understanding of raw EEG, can prove useful
during the intraoperative period for the diagnosis of nonconvulsive status epilepticus
or seizures in the presence of neuromuscular blocking agents (NMBAs), especially when
used with bilateral montage.[13]
[17]
[18]
[19] In addition, concurrent disturbances in hemodynamics and metabolic profile may aid
in the diagnosis. Our patients did not experience significant hemodynamic disturbances,
probably due to the dexmedetomidine infusion. But, in the absence of a change in depth
of anesthesia, changes in raw EEG and metabolic profile, along with the reversal of
waveform with antiepileptic medications, favor the diagnosis of intraoperative seizure.
The high-amplitude and low-frequency EEG patterns seen during the intraoperative seizure
in our cases have not been widely reported. Our cases emphasize that while these findings
are preliminary, they add evidence to the role of BIS beyond anesthetic depth and
could improve intraoperative safety by facilitating prompt detection and treatment
of seizures. However, BIS has certain limitations, such as limited spatial resolution
compared with multilead EEG and false positive increase (due to EMG interference or
anesthetic depth fluctuations). In the absence of a large data or case reports, larger
multicenter studies focusing on EEG changes in the presence or absence of NMBA, along
with the effect of various anesthetic agents on intraoperative EEG and the development
of BIS software optimized for seizure detection, are required to validate the findings.
Conclusion
Intraoperative seizure is an unwanted event with possible deleterious consequences.
Due to the inability to use conventional EEG during neurosurgical procedures, BIS,
along with raw EEG display, can facilitate early detection and management of intraoperative
seizures.
