Keywords
epilepsy surgery - electrocorticography - dexmedetomidine - isoflurane - propofol
- anesthesia
Introduction
Intraoperative electrocorticography (ECoG) guidance during epilepsy surgery is used
for precisely identifying the epileptogenic focus and ascertaining the adequacy of
surgical resection.[1] Interference with the ECoG signals, mainly suppression, can lead to incorrect mapping
and subsequent unsuccessful surgery. Commonly used anesthetic drugs are known to interfere
with ECoG signals by causing excitation or inhibition of epileptiform discharges.[2]
[3]
[4]
[5]
[6] To minimize interference, the anesthetic drugs are temporarily discontinued, or
their doses reduced, before obtaining ECoG recordings[3]
[7]
[8]; however, this action may lead to intraoperative awareness,[9] an unpleasant experience described as “postoperative recall of sensory perception
during anesthesia.”[10] A suitable anesthetic drug for ECoG-guided epilepsy surgery would thus be one that
has no influence, or minimal depressant effect, on intraoperative ECoG signals without
requiring dose reduction.
Dexmedetomidine, an α-2 agonist drug with neuroprotective properties, is commonly
used as a sedative or an anesthetic adjuvant in neurosurgical patients. It is currently
generating interest as a potentially useful drug for epilepsy surgeries due to its
reported absence of influence, or minimally depressant effect, on ECoG signals.[7]
[8]
[11]
[12]
[13] The effect of dexmedetomidine on intraoperative ECoG signals has also been studied.[7]
[8]
[11]
[12] As dexmedetomidine is not used as a sole anesthetic agent, its effect on ECoG has
been evaluated in combination with other anesthetic agents like sevoflurane[11] and propofol.[7]
[8] Evaluation of more such combinations of dexmedetomidine with various commonly used
anesthetic agents will enable identification of a suitable anesthetic regimen for
epilepsy surgeries.
We undertook this prospective, randomized study with the primary objective of evaluating
the effect of dexmedetomidine used in combination with either isoflurane- or propofol-based
anesthetic regimen on intraoperative ECoG in patients undergoing ECoG-guided epilepsy
surgeries. We hypothesized that dexmedetomidine used with either of these anesthetic
regimens would have no influence on intraoperative ECoG signals. The secondary objective
of this study was to determine the safety of using dexmedetomidine in these combination
regimens by assessing its effects on hemodynamic parameters, anesthesia recovery time,
and incidence of intraoperative awareness.
Materials and Methods
After obtaining approval from the Institutional Ethics Committee, consecutive patients
in the age group of 12 to 60 years and requiring ECoG-guided epilepsy surgery for
drug-resistant epilepsy were selected for the study. A single neurologist evaluated
all eligible patients through clinical assessment, magnetic resonance imaging, video
electroencephalography (EEG), or positron emission tomography. The decision to perform
epilepsy surgery was taken collectively by the epilepsy team comprising the neurologist,
neurosurgeon, neuro-radiologist, and neuro-anesthesiologist. Patients with bradyarrhythmias,
hypotension, uncontrolled hypertension, pregnancy, hepatic and renal impairment, cardiac
disease, morbid obesity, and known allergy to anesthetic drugs were excluded. This
study was conceived as a pilot feasibility trial to evaluate the change in ECoG score
with a dexmedetomidine bolus administered with propofol versus the isoflurane anesthetic
agent. A sample size of 15 participants per group was chosen based on anticipated
recruitment feasibility within the study period and to provide preliminary estimates
of variability for future definitive sample size calculations.
Conduct of Study
Using a computer-generated randomization method and sealed opaque envelopes, the recruited
patients were randomized into two groups: (1) Isoflurane group (Group-I), who received
dexmedetomidine-isoflurane anesthesia, and (2) Propofol group (Group-P), who received
dexmedetomidine-propofol anesthesia. All personnel present in the operating room (OR),
except the neuro-anesthesia team, were unaware of the anesthetic regimen allotted
to the study patients. Standard neurosurgical monitoring and additional bispectral
index (BIS) to monitor the anesthetic depth were used for surgery. General anesthesia
(GA) was induced in both groups with an intravenous (IV) bolus of fentanyl 2 µg/kg,
thiopentone 3 to 5 mg/kg, and atracurium 0.5 mg/kg, and maintained with infusions
of fentanyl (0.5 µg/kg/hour) and atracurium (5–7 µg/kg/hour), nitrous oxide (N2O) and oxygen mixture (50:50), and BIS-guided doses of isoflurane and propofol targeting
a BIS value of 40 to 60. Following craniotomy and dural reflection, the BIS was kept
at 60 to 70 with an isoflurane end-tidal concentration of 0.3 to 0.5% in Group-I,
and a propofol dose of 25 to 50 µg/kg/min in Group-P, and ECoG recording was started.
Signals were obtained from brain surface strips or grid, and depth electrodes were
placed at various locations depending on the surgical requirement. ECoG recording
was done for 2 minutes on a 64-channel recording device (Nicolet Viasys, USA version
5.32), and analyzed in real time by the ECoG team, comprising a neurologist and an
ECoG technician, present in the operating room; this team was also blinded to the
anesthetic regimen used. Following the initial recording, dexmedetomidine (Neon, Neon
Laboratories Ltd., Mumbai, India), in a bolus dose of 1 µg/kg, was infused over 5 minutes
in both groups. A second ECoG recording was obtained at each of the prior locations,
5 minutes after dexmedetomidine administration, to align with its rapid distribution
half-life (∼6 minutes) and onset of central effects. The data collected from both
recordings were interpreted collectively by the ECoG team, and in case of disagreement
or uncertainty, the neurologist's opinion prevailed. Dexmedetomidine was thereafter
continued as an infusion of 0.5 µg/kg/h for the remaining surgery in both groups,
along with BIS-titrated doses of isoflurane and propofol to maintain the BIS value
at 40 to 60. Dexmedetomidine, fentanyl, and atracurium infusions were stopped at the
end of dura closure, and N2O, isoflurane, or propofol were stopped at the time of Mayfield head clamp removal.
Neuromuscular blockade was reversed with neostigmine (50–70 µg/kg) and glycopyrrolate
(8–10 µg/kg) when the patient started breathing, and tracheal extubation was done
when the patient obeyed commands. IV paracetamol 15 mg/kg was used for postoperative
analgesia. The Consort Diagram for methodology is depicted in [Fig. 1].
Fig. 1 CONSORT flow diagram.
Data Recorded in Both Groups
-
ECoG signals: recorded just before administering dexmedetomidine bolus (baseline/ pre-dexmedetomidine)
and 5 minutes after the end of the bolus dose (post-dexmedetomidine).
-
Heart rate (HR), mean arterial pressure (MAP), and BIS value: recorded at (1) baseline, (2) every minute during bolus administration, (3) five
minutes after the end of the bolus dose, and (d) every 15 minutes thereafter till
the end of surgery.
-
Time to anesthesia emergence: measured as the duration between the stoppage of N2O till regaining consciousness[14]; this usually correlates with a BIS of 90 or higher.
-
Incidence of intraoperative awareness: considered as per definition in the Modified Brice Questionnaire.[15]
-
Intraoperative complications: seizures and inadvertent electrode migration or malposition during surgery.
Interpretation of Data Collected
-
ECoG scores were graded based on the method described by Mathern et al.[16]
ECoG score 0: presence of regular activity; ECoG score 1: presence of normal background of mixed γ, β, and α frequencies of moderate to low
amplitude (usually < 20–30 mV), and a few observed low-amplitude spikes; ECoG score 2: presence of background of mixed α, β, and δ frequencies but of low to moderate amplitude
with loss of fast (>20 Hz) background frequencies, and often observed repeated but
non-continuous spikes, poly-spikes or paroxysmal fast activity of medium amplitude;
ECoG score 3: presence of mostly 6 to 20 Hz background frequencies with some localized nearly
continuous interictal epileptiform features of moderate amplitude, or persistent repetitive
spiking, and very rarely, capture of electrographic seizures; ECoG score 4: presence of slow (<6 Hz) background frequencies with continuous synchronous features
of moderate to high amplitude, and multiple independent epileptiform abnormalities
like poly-spikes, paroxysmal fast activity, and electrographic seizures which could
be recorded; ECoG score 5: detection of slow rhythmic, usually synchronous background (<4 Hz) and often of
high amplitude, and continuous synchronized or independent high-amplitude epileptiform
abnormalities observed in multiple cortical sites, and rarely recorded ictal discharges
but observed in surrounding cortex. The mean ECoG scores were compared before and
after dexmedetomidine administration in each group. Post-dexmedetomidine scores higher
than pre-dexmedetomidine scores indicated ECoG augmentation, while lower scores indicated
ECoG suppression.
-
Hemodynamic parameters (HR and MAP) changes of ± 20% from baseline were considered
abnormal.
-
A BIS value of 40 to 60 suggested adequate anesthetic depth; values below or above
this range were considered as indicating increased and decreased anesthetic depths,
respectively.
-
Intraoperative awareness was evaluated as per the criteria described in the Modified
Brice Questionnaire.
-
Delayed emergence was considered if the patient was not obeying simple verbal commands
20 to 30 minutes after cessation of anesthesia.[17]
Statistical Analysis
The collected data were transformed into variables, coded, and entered into Microsoft
Excel. Statistical evaluation was performed using SPSS PC 25 version. Normality of
data was tested by the Kolmogorov-Smirnov test. Quantitative data were expressed as
the mean and standard deviation (SD), depending on the distribution's normality. The
difference between two comparable groups was tested by Student's t-test (unpaired) or Mann-Whitney “U”-test. The Wilcoxon signed-rank test was used for comparison between pre-dexmedetomidine
and post-dexmedetomidine data. The qualitative data were expressed in percentages,
and statistical differences between the proportions were tested by the chi-squared
test or Fisher's exact test. A p-value less than 0.05 was considered statistically significant.
Results
Preoperative Data
The preoperative demographic characteristics were comparable between Group-I (15 patients)
and Group-P (15 patients), and no patient had any comorbid condition ([Table 1]). The distribution of epileptogenic lesions was also comparable between the groups
([Table 2]).
Table 1
Comparison of demographic characteristics between Group I and Group P
|
Demographic characteristics
|
Group-I (n = 15)
|
Group-P (n = 15)
|
p-Value
|
|
Male
|
8 (53.3%)
|
6 (40%)
|
0.46
|
|
Female
|
7 (46.67%)
|
9 (60%)
|
|
Mean age in years
|
26.47 ± 7.44
|
24.53 ± 8.28
|
0.69
|
|
Weight in kg
|
60.4 ± 15.53
|
57.07 ± 14.05
|
0.71
|
Table 2
Distribution of types of epileptogenic lesions in Group-I and Group-P
|
Diagnosis
|
Group-I (n = 15)
|
Group-P (n = 15)
|
|
Left MTS
|
4 (26.67%)
|
6 (40%)
|
|
Right MTS
|
7 (46.67%)
|
7 (46.67%)
|
|
Left FCD
|
2 (13.33%)
|
0
|
|
Left hippocampal gliosis
|
1 (6.67%)
|
0
|
|
Left parietooccipital sclerosis
|
1 (6.67%)
|
0
|
|
Left temporal DNET
|
0
|
1 (6.67%)
|
|
Right frontal FCD
|
0
|
1 (6.67%)
|
Abbreviations: DNET, dysembryoplastic neuroepithelial tumor; FCD, focal cortical dysplasia;
MTS, mesial temporal sclerosis.
ECoG Data
The ECoG scores, expressed as mean, are depicted in [Fig. 2]. In Group-I, the pre-dexmedetomidine mean score was 1.8 ± 0.7, and the post-dexmedetomidine
mean score was 2.1 ± 0.9; the increase in ECoG score following dexmedetomidine administration
was statistically significant (p = 0.02). In Group-P, the pre-dexmedetomidine mean score was 2.0 ± 0.7, and the post-dexmedetomidine
mean score was 2.10 ± 0.7 with no significant change observed in ECoG scores (p = 0.16). On intergroup comparison, the pre-dexmedetomidine ECoG scores and the post-dexmedetomidine
scores were comparable between the groups (p = 0.99 and p = 0.36, respectively).
Fig. 2 Comparison of mean ECoG scores before and after dexmedetomidine in Group-I and Group-P.
Hemodynamic Parameters and BIS Values
In Group-I, the pre-dexmedetomidine HR was 68 ± 8.14, and the post-dexmedetomidine
HR was 67 ± 11.1, with no significant change in HR (p = 0.26). The pre-dexmedetomidine MAP was 88.7 ± 9.52, and the post-dexmedetomidine
MAP was 81.1 ± 8.95, showing no significant change in MAP (p = 0.82). Similarly, in Group-P, the pre-dexmedetomidine HR was 71.8 ± 9.81, and the
post-dexmedetomidine HR was 64.9 ± 13.2, with no significant alteration in HR (p = 0.28). The pre-dexmedetomidine MAP was 86.7 ± 4.92, and the post-dexmedetomidine
MAP was 81.7 ± 8.09, with no significant alteration in MAP (p = 0.25). The mean ± SD values of HR and MAP were comparable between the two groups
and remained within normal limits throughout surgery ([Fig. 3]). BIS values were maintained at 60 to 70 just prior to administering dexmedetomidine
and were comparable between the groups (Group-I, 65 ± 4; Group-P, 64.2 ± 2.48; p = 0.08). BIS measured at 1-minute intervals during dexmedetomidine bolus infusion
declined similarly in both groups (mean decline in BIS was 10.3 in Group-I and 9.6
in Group-P). Following dexmedetomidine administration, the BIS values were in the
range of 40 to 60 and were comparable between the two groups. (Group-I, 52.3 ± 6.28;
Group-P, 51.9 ± 7.09; p = 0.0.65). BIS values remained in the range of 40 to 60 at the 15-minute reading,
and during all subsequent readings throughout surgery.
Fig. 3 Comparison of hemodynamic parameters between Group-I and Group-P at different intervals:
(A) heart rate (HR) comparison, (B) mean arterial pressure (MAP) comparison.
Recovery from Anesthesia
The time to emergence was within the normal range in both groups, but on inter-group
comparison, time to emergence was significantly longer in Group-I compared with that
in Group-P (Group-I, 14.4 ± 3.7 minutes; Group-P, 7.33 ± 2.05 minutes; p = 0.03).
Discussion
In this study, we observed that administration of dexmedetomidine did not suppress
the epileptiform activity when used with either isoflurane- or propofol-based anesthetic
regimens, but caused significant augmentation of activity when used in the isoflurane
regimen. The hemodynamic parameters were not altered by the addition of dexmedetomidine
to either of the anesthetic regimens. The time to emergence from anesthesia was within
defined normal limits in both groups, but was significantly longer with isoflurane
than with propofol. A decrease in the BIS value was observed with dexmedetomidine
administration, which was comparable in both groups. There was no incident of intraoperative
awareness in any patient.
No suppression of intraoperative ECoG signals by dexmedetomidine, observed in both
anesthesia regimens, partially supports our hypothesis. This non-inhibitory effect
of dexmedetomidine on interictal epileptiform activity is attributed to its lack of
action on the gamma-aminobutyric acid (GABA) receptor-mediated pathway, which is linked
to epileptiform discharges.[13]
[18] Some earlier authors have also reported no suppression of intraoperative ECoG signals.[7]
[8]
[11]
[12] Dexmedetomidine-propofol combination was used by Souter et al as sedation during
awake craniotomy for seizure area resection in six patients,[7] and by Pacreu et al for GA in one patient undergoing right selective amygdalohippocampectomy[8]; both case reports revealed no ECoG suppression with the use of dexmedetomidine,
thus enabling accurate mapping of epileptic foci. Chaitanya et al studied the effects
of a single bolus of dexmedetomidine used alone during surgery for drug-resistant
epilepsy and found no suppression of ECoG, but a significant increase in the mean
spike rate of ECoG.[12] Oda et al used a single dose infusion of dexmedetomidine and sevoflurane combination
in patients undergoing surgery for temporal lobe epilepsy and found that dexmedetomidine
decreased the median frequency of ECoG without affecting spike activity.[11] We noticed that dexmedetomidine caused no inhibition of the intraoperative ECoG
signals in its normally used dose, thereby suggesting a potential suitability of this
drug in ECoG-guided epilepsy surgery.
Our second observation regarding significant augmentation of ECoG scores by dexmedetomidine
in the isoflurane anesthetic regimen is a serious concern, even though no seizures
were observed; our ECoG score increased up to 3, which is below a defined seizure
threshold score of 4 and above.[16] Dexmedetomidine-induced increased epileptiform activity has been attributed to a
reduction in noradrenaline secretion and modulation of inhibitory neurotransmitters
that potentially lower the seizure threshold, and is the same as that seen with sleep
deprivation.[19]
[20] Some prior studies have also reported ECoG augmentation with dexmedetomidine showing
as a significant increase in mean spike rate of ECoG without seizures,[12] increased spike frequency without seizures,[20] subclinical seizures and an increase in spike and wave activity,[7] and increased interictal epileptiform activity of 4 foci.[13] Prior evidence reveals that clonidine, another α-2 agonist drug, also increases
the epileptic discharges in animals and humans,[21] and it has been implied that α-2 agonists may either have no effect or cause enhancement
of epileptiform discharges in patients with medically intractable seizures.[13] Though dexmedetomidine has shown proconvulsant activities in animals,[22]
[23] there is, so far, sparse evidence of such effects in humans.[24] However, an ECoG-enhancing effect of dexmedetomidine deserves more attention because
it could be its potential adverse effect in terms of inducing epileptic seizures.[20] Whether dexmedetomidine causes selective ECoG augmentation when used with different
anesthetic agents, as was observed in its combination with isoflurane but not with
propofol or sevoflurane in an earlier study,[11] deserves further investigation. In our study, a possible contribution to ECoG enhancement
by isoflurane, known to have neuro-excitatory effects,[25] also needs to be considered and explored further.
Combining dexmedetomidine with other anesthetic agents raises safety concerns regarding
hemodynamic stability and timely recovery from anesthesia. Some studies have reported
episodes of bradycardia and hypotension, though transient and self-corrected, due
to the sympatholytic effects of dexmedetomidine.[11]
[26] We did not observe any significant intraoperative hemodynamic alterations in either
group. Emergence from anesthesia occurred within 20 minutes in all our patients, which
is consistent with acceptable neurosurgical recovery timelines.[17] However, on comparison, a significantly longer emergence time was seen in the isoflurane
group, which is more likely due to isoflurane's known delayed anesthesia washout[27]
[28]
[29] rather than any sedative prolongation from dexmedetomidine.
Development of intraoperative awareness is a possibility due to the commonly followed
protocol of anesthetic dose reduction to minimize interference with ECoG signals.
An 8% incidence of intraoperative awareness has been previously reported when BIS
was maintained at 70 or higher before starting ECoG recording.[9] This side effect may have been obviated in our study by the additional sedation
provided by dexmedetomidine, which decreased the BIS from a value of 60 to 70 to 40
to 60. Use of dexmedetomidine is known to reduce the requirements of the concurrent
anesthetic agents, and we too observed that this lower BIS value could be maintained
throughout the remaining surgery using minimal doses of propofol and isoflurane. However,
analysis of BIS requires caution in the presence of dexmedetomidine, which can have
a confounding effect on BIS interpretation due to its own sedative effect contributing
to the lower BIS. No alternate anesthetic depth monitor was available to us, and we
additionally used hemodynamic monitoring for assessment of anesthetic depth. With
BIS and hemodynamic monitoring, we did not find any evidence suggestive of a lighter
plane of anesthesia in our patients. Furthermore, postoperative assessment using the
Modified Brice Questionnaire for the detection of any episodes of intraoperative awareness
also did not reveal any such incident.
There were various limitations in our study. The sample size was small, based on convenience
sampling, due to the lack of prior data for formal sample size calculation and the
limited availability of eligible patients. The dexmedetomidine bolus was administered
over 5 minutes instead of the standard 10 minutes due to logistical constraints. We
used manual analysis of ECoG recordings, which can potentially introduce an interpretation
bias. Formal inter-rater reliability metrics were not available to us, but a combined
analysis and interpretation of the data by the neurologist and the EEG technician
helped mitigate the interpretation bias. We only had BIS as the anesthetic depth monitor,
which can show misleading values in the presence of dexmedetomidine. These limitations
suggest that while our findings are promising, they should be interpreted with caution
and corroborated with larger, well-powered studies.
Conclusion
This study further adds to the limited existing literature on the effects of dexmedetomidine
on intraoperative ECoG signals and is among the few early studies evaluating dexmedetomidine
as part of combined anesthetic regimens. The results endorse prior findings of a non-suppressant
effect of dexmedetomidine on intraoperative ECoG signals. Due to its sedative effect,
dexmedetomidine additionally helps in minimizing ECoG interference by enabling the
use of concomitant anesthetic drugs in lower doses, while preventing any consequent
intraoperative awareness. However, dexmedetomidine-induced ECoG augmentation, observed
in this study and earlier, needs additional research for validation and to ascertain
if dexmedetomidine has any proconvulsant propensity. Our observation of possible differential
ECoG-enhancing effects of dexmedetomidine when combined with various anesthetic agents
also warrants further evaluation to enable selection of the safest anesthetic combination
regimen. Based on the present study, we feel that the dexmedetomidine-propofol combination
appears to be a suitable anesthetic regimen for routine anesthesia practice in ECoG-guided
epilepsy surgeries.
