Brain activity during intraoperative general anesthesia using resting-state functional magnetic resonance imaging ~Feasibility study~

doi:10.21203/rs.3.rs-4699067/v1

Download PDF

Article

Brain activity during intraoperative general anesthesia using resting-state functional magnetic resonance imaging ~Feasibility study~

https://doi.org/10.21203/rs.3.rs-4699067/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Electroencephalogram (EEG) monitors are widely used to evaluate depth of anesthesia. However, EEG monitors have problems such as differences in evaluation depending on the type of anesthetic and influences from drugs other than anesthetic agents. Monitoring of brain activity using intraoperative resting-state functional MRI (rs-fMRI) would be expected to improve evaluation of the depth of anesthesia in EEG monitoring. We examined the feasibility of evaluating the effects of general anesthesia with sevoflurane using rs-fMRI during neurosurgery. We enrolled 20 adult patients scheduled for transsphenoidal surgery. We compared differences between before and during general anesthesia in terms of brain functional connectivity of the thalamus by seed-to-voxel correlation analysis and local neural activity using fractional amplitude of low-frequency fluctuations (fALFF) analysis. An exclusion mask was applied to exclude brain areas showing intraoperative spatial artifacts and correct for differences in the magnitude of intra- and preoperative head movements. We analyzed 16 patients. Functional connectivity of the thalamus to the contralateral thalamus, bilateral caudate nuclei and globi pallidi was significantly decreased during anesthesia. The precuneus and posterior cingulate cortex showed significantly decreased fALFF values during anesthesia. These findings were consistent with previous studies and indicate the feasibility of intraoperative rs-fMRI during deep anesthesia.

Biological sciences/Neuroscience/Neural circuit

Biological sciences/Neuroscience/Neuronal physiology

functional connectivity

functional MRI

general anesthesia

intraoperative neurophysiological monitoring

MRI

During general anesthesia, the depth of anesthesia needs to be assessed to prevent unexpected body movements and intraoperative awakening. The electroencephalogram (EEG) monitor is commonly used to evaluate depth of anesthesia,¹ but has critical issues such as the fact that evaluations can differ depending on the type of general anesthetic.^{2, 3} In addition, since general anesthesia during surgery is administered not only with general anesthetics, but also with narcotics and muscle relaxants, the effects of these drugs must be taken into consideration for accurate assessment of the depth of anesthesia using an EEG monitor.^4,5

In recent years, the effects of general anesthetics on the brain have been widely studied at the sedation level using resting-state functional magnetic resonance imaging (rs-fMRI).^6–9 Most anesthesia protocols use a single agent, and changes in spontaneous brain activity are examined to show the characteristics of each anesthetic agent. However, no studies have used rs-fMRI to evaluate the effects of anesthesia during actual surgery.

Evaluation of spontaneous brain activity during general anesthesia using intraoperative rs-fMRI may be very useful in exploring the physiological basis of EEG monitoring from the perspective of clarifying brain networking. Thus, we examined the feasibility of evaluating the effects of general anesthesia (mainly from sevoflurane) on the brain, and the thalamus in particular, using rs-fMRI during neurosurgery.

Design and ethics

This study used a single-center, prospective, observational design. This study was approved by the Kobe University Hospital Ethics Committee (approval no. 1787), and conducted in accordance with the Declaration of Helsinki. Written informed consent was obtained from all participants prior to enrolment.

Setting and participants

We screened 20 adult patients scheduled for elective transsphenoidal surgery in the study site from December 2015 to April 2019. None of these 20 patients had any disturbance of consciousness preoperatively as of the date of admission (1–3 days before surgery). Among these, we enrolled those patients who were scheduled for intraoperative MRI to evaluate the progress of surgery. Patients with valid scans less than 90% of all scans, maximum body motion values greater than 3 mm, and average body motion values greater than 1 mm were excluded from the analysis of this study.

Anesthetic management during operation and fMRI

All patients were anesthetized with propofol, midazolam, sevoflurane, fentanyl, remifentanil, and rocuronium before intraoperative MRI. Each patient was intubated and received an arterial line. Anesthesia with sevoflurane, fentanyl, and rocuronium was induced using an MRI-compatible anesthetic machine (Aestiva/5 MRI; GE HealthCare) during the MRI session. We collected data from the electronic medical records, including type and dose of anesthetics, and timing of administrations before and during the intraoperative MRI session.

Acquisition of rs-fMRI data

We performed rs-fMRI preoperatively (control) and intraoperatively. Data acquisition was performed using a 3-T MRI scanner (MAGNETOM Skyra 3-Tesla; Siemens Healthineers, Erlangen, Germany). A 20-channel head/neck coil was used for preoperative MRI, and a 8-channel head/holder coil (FLEXIBILITY OR Head Holder & 8-Ch Coil; NORAS MRI products, Hoechberg, Germany) was used for intraoperative MRI. Prior to performing preoperative rs-fMRI, each patient was given the following instructions: “Close your eyes, relax, and do not think of anything in particular. Do not fall asleep.” Intraoperative imaging was performed after completion of tumor removal under general anesthesia. Functional images were acquired using a gradient-echo echo-planar imaging sequence with two slightly different parameters (in this study, two different data sets were used for analysis). For Patients 1–9, the following parameters were applied: repetition time, 2530 ms; echo time, 30 ms; flip angle, 80°; field of view, 240×240 mm; matrix size, 64×64; number of axial slices, 40; slice thickness, 3.0 mm with 0.6-mm gap; images, 150. For Patients 10–16, the parameters applied were: repetition time, 2000 ms; echo time, 30 ms; flip angle, 90°; field of view, 240×240 mm; matrix size, 64×64; number of axial slices, 36; slice thickness, 3.0 mm with 0.6-mm gap; images, 300 (the first 150 images were used for analysis). Anatomical images were acquired using a T1-weighted three-dimensional (3D) magnetization-prepared rapid acquisition with gradient echo sequence. For the preoperative scan, the following parameters were used: repetition time, 2300 ms; echo time, 2.3 ms; flip angle, 8°; sagittal slices, 192; voxels, 1×1×1 mm. For the intraoperative scan, the following parameters were used: repetition time, 2300 ms; echo time, 2.3 ms; flip angle, 8°; sagittal slices, 512; voxels, 0.5×1.0×1.0 mm.

Analysis of rs-fMRI data

Preprocessing procedures were performed using the CONN functional connectivity toolbox (version 17.f; www.nitrc.org/projects/conn) and SPM12 (version 7771; www.fl.ion.ucl.ac.uk/spm/). Briefly, functional magnetic resonance (fMR) images of each patient were first realigned (motion correction) and corrected for slice acquisition timing (slice timing correction). Functional and structural images were normalized to Montreal Neurological Institute templates of echo planar images and 3D T1-weighted images followed by nonlinear transformation, respectively. Finally, images were smoothed with a Gaussian kernel of 8 mm at full width at half maximum. In individual analyses, we performed seed-to-voxel correlation analysis. Before functional connectivity (FC) and fractional amplitude of low-frequency fluctuations (fALFF) analysis, two denoising processes were performed. One of the denoising processes was removal of the effect of low-frequency drift and high-frequency noise by band-pass filtering (band-pass frequency, 0.008–0.09 Hz). The other was removal of non-neural activity effects from ventricular regions, white matter, and head motions by multiple regression. CONN implements an anatomical component-based noise correction method (aCompCor)¹⁰ for physiological noise component extraction. Individual FC maps were generated in CONN by extracting the mean BOLD time course from each seed region and calculating correlation coefficients with the BOLD time course of each voxel throughout the whole brain. The resulting coefficients were converted to normally distributed scores using Fisher’s transformation to give maps of voxel-wise FC for each seed for each subject. The value of each voxel throughout the whole brain represents the relative degree of FC with each seed. These maps were subsequently used for second-level analysis of relative FC using a two-sided paired t-test, implemented in CONN, to investigate differences in seed-to-voxel connectivity between pre- and intraoperative conditions.

The seeds used for analysis were the left and right thalamus, as previous reports have noted that the loss of consciousness resulting from induction of anesthesia was related to declines in FC between the thalamus and cortex,¹¹ and declines in thalamic activity.¹²

We also performed fALFF analysis to investigate differences in the degree of local neural activity between pre- and intraoperative conditions.¹³ The fALFF is defined as the ratio of the amplitude of the low-frequency band (0.01–0.1 Hz) to that of the detectable frequency range. The amplitude of the low-frequency band itself is known as ALFF. Although both the ALFF and fALFF reflect the intensity of regional spontaneous brain activity, the ALFF is more sensitive to physiological noise. Therefore, the fALFF was used as a value with improved sensitivity and specificity for detecting spontaneous brain activities. The temporal series of BOLD signals was transformed to the frequency domain and the power spectrum was then obtained. Since the power of a given frequency is the square of the amplitude of that frequency component, the square root was calculated at each frequency of the power spectrum and the averaged square root was obtained across the range of 0.01–0.1 Hz and the detectable frequency range at each voxel, respectively. Then, the ratio of those two values was calculated at each voxel.

Spatial artifacts on intraoperative fMRI and masking

The following three types of spatial artifacts were observed on intraoperative fMRI (Fig. 1). The first type was head fixation pin artifacts, representing artifacts around the pins used to fixate the head during surgery. The second type was air artifacts, representing artifacts around the air that enters the skull. The third type was off-center artifacts (field inhomogeneity artifacts or susceptibility artifacts caused by an off-center position on the gantry), representing artifacts caused by the head fixation position being off the center of the gantry due to the surgical procedure.

Signal loss on fMRI due to these artifacts covers a large number of regions, and masking was performed to exclude these regions from analysis (Fig. 1).

SPM12 was used to create the exclusion mask by superimposing extracted images of the areas showing signal loss in the functional images for each patient, and excluding those areas found to show signal loss in more than four patients (Supplementary Fig. 1).

The present study compared FC strength and fALFF values taken preoperatively in a resting, eyes closed condition with those taken intraoperatively under deep anesthesia. Preoperative rs-fMRI data with eyes closed at rest were affected by head movements, such as swallowing, that are seen in an awake state, but intraoperative data under deep anesthesia were unaffected by such head movements. We therefore corrected for differences in head motion to cancel head movement effects on FC strength and fALFF values.

To evaluate the effects of head movements, we compared pre- and intraoperative head movements based on framewise displacement (FD).¹⁴ We then calculated the difference between pre- and intraoperative FD values for each subject by averaging the FD values for each subject detected by CONN. The difference in FD values between the pre- and intraoperative rs-fMRI data was used for paired t-tests as a covariate (or a nuisance regressor), correcting for the influence of head motion.

Statistical analysis

Paired t-testing was performed to investigate differences in brain FC and fALFF between pre- and intraoperative conditions. In all comparisons of FC strength between pre- and intraoperative conditions, the significance level was p < 0.001 (uncorrected for multiple comparisons) at the voxel level and p < 0.05 (family-wise error [FWE]-corrected) at the cluster level. Moreover, Bonferroni correction for multiple comparisons was applied to the cluster-level threshold because the left and right thalamus were seeds. Thus, an FWE-corrected value of p < 0.025 (0.05/2) was required for a statistically significant result. In the comparison of fALFF between pre- and intraoperative conditions, the significance level was p < 0.001 (uncorrected for multiple comparisons) at the voxel level and p < 0.05 (FWE-corrected) at the cluster level. We performed these statistical analyses using SPM12.

During the study period, 20 adult patients underwent intraoperative rs-fMRI during elective transsphenoidal surgery. We excluded 1 patient who was lacking an accurate anesthesia record during intraoperative fMRI and 3 patients showing motion artifacts in MRI signals. Eventually, we analyzed rs-fMRI data from 16 patients.

Characteristics of patients

Analyzed data were acquired from 8 women and 8 men with a mean age of 49 years (range, 21–76 years). The underlying pathology in these patients was pituitary adenoma in 9 patients, craniopharyngioma in 2 patients, and Rathke’s cleft cyst, chondrosarcoma, sebaceous carcinoma, fibrous dysplasia and suspected hypophysitis in 1 patient each (Table 1).

Management of anesthesia

Table 1 shows the anesthetic agents used before and during intraoperative MRI. Four patients (Patients 2, 8, 9, and 15) were anesthetized using propofol (target-controlled infusion) and all others were anesthetized using sevoflurane before the intraoperative MRI session. The effect-site concentration of propofol, the inspired sevoflurane concentration, and the dose of remifentanil administered just before intraoperative MRI were 2.3–3.6 µg ml^− 1 (mean, 2.93 µg ml^− 1), 1–3% (mean, 1.83%), and 0.05–0.3 µg kg^− 1 min^− 1 (mean, 0.177 µg kg^− 1 min^− 1), respectively. The total dose of fentanyl administered before the intraoperative MRI session was 1.96–10.57 µg kg^− 1 (mean, 6.11 µg kg^− 1). Only Patient 2 was administered 2 mg (0.031 mg kg^− 1) of midazolam intravenously.

All patients were anesthetized with sevoflurane during the intraoperative MRI session. Half of the patients (Patients 1, 2, 4, 6, 7, 12, 13, and 16) were administered fentanyl during the intraoperative MRI session, but the others were not. The inspired sevoflurane concentration and total dose of fentanyl administered during intraoperative MRI were 1.5–3% (mean, 2.05%) and 0.83–2.41 µg kg^− 1 (mean, 1.66 µg kg^− 1), respectively.

Table 1

Characteristics of patients and anesthetic data
					Anesthetic data
	Characteristics				Before intraoperative MRI session					During intraoperative MRI session
ID	Sex	Age (years)	ASA-PS	Target disease	Propofol TCI (µg ml^− 1)	Inspired sevo concentration (%)	Remifentanil dose (µg kg^− 1 min^− 1)	Fentanyl dose (µg kg^− 1)	BIS value	Inspired sevo concentration (%)	Fentanyl dose (µg kg^− 1)
1	M	58	2	Rathke's cleft cyst	N/A	1.4	0.15	3.83	N/A	2	1.53
2	F	49	1	pituitary adenoma	2.3	2.5	0.3	7.58	N/A	3	1.52
3	F	42	2	pituitary adenoma	N/A	1.5	0.17	1.96	N/A	1.5	N/A
4	M	21	2	craniopharyngioma	N/A	1.5	0.3	10.46	N/A	2	2.2
5	M	63	2	craniopharyngioma	N/A	2	0.1	3.46	N/A	3	N/A
6	M	35	1	fibrous dysplasia	N/A	2.5	0.15	5.17	N/A	3	1.29
7	M	68	2	pituitary adenoma	N/A	1	0.15	9.64	N/A	2	2.41
8	F	57	2	pituitary adenoma	2.8	2	0.2	10.57	N/A	2	N/A
9	F	28	2	pituitary adenoma	3	3	0.25	7.1	N/A	2.5	N/A
10	M	61	2	s/o hypophysitis	N/A	1.6	0.05	6.62	48	1.6	N/A
11	F	31	1	pituitary adenoma	N/A	1.6	0.15	4.9	41	1.6	N/A
12	F	40	1	pituitary adenoma	N/A	1.8	0.16	4.16	49	1.8	0.83
13	M	76	2	sebaceous carcinoma	N/A	1.5	0.1	5.22	44	1.5	1.49
14	M	58	1	chondrosarcoma	N/A	1.5	0.25	3.81	62	1.5	N/A
15	F	72	2	pituitary adenoma	3.6	1.8	0.25	N/A	53	1.8	N/A
16	F	27	1	pituitary adenoma	N/A	2	0.1	7.14	48	2	2.04

*ASA-PS, American Society of Anesthesiologists physical status; TCI, target-controlled infusion; sevo, sevoflurane; BIS, bispectral index; F, female; M, male; N/A, not available; s/o, suggestive of.

Comparisons of FC involving the thalamus between pre- and intraoperative conditions, and effects of FD correction and spatial artifact masking

FC strengths of the right thalamus with the contralateral thalamus, bilateral caudate nuclei, and globi pallidi were significantly decreased during anesthesia without FD correction (Fig. 2A, Table 2). In contrast, FC strengths of the right thalamus with the operculum, insula, parahippocampal gyrus, hippocampus, and fusiform gyrus were significantly increased during anesthesia without FD correction (Fig. 2B, Table 2). However, after FD correction, differences in FC strength between the two conditions decreased. As a result, the region in which FC decreased during anesthesia reduced (Fig. 2C, Table 2), and clusters with increased FC strength during anesthesia no longer showed significant differences (Fig. 2D, Table 2). In addition, application of the exclusion mask partially excluded the region in which FC decreased during anesthesia (Fig. 2E, Table 2). Almost no differences in results of FC of the thalamus were seen between left and right sides (Supplementary Fig. 2).

Table 2

Results of functional connectivity (FC) analysis
Comparison	Seed	FD	Mask	Clusters (x, y, z)	Size (voxels)	Peak t value	Brain regions
pre > intra	right thalamus	-	-	+ 10, -12, +6	4105	14.87	left thalamus, right thalamus, left caudate, right caudate, right pallidum, left pallidum
		+	-	+ 10, -12, +6	2368	9.87	left thalamus, right thalamus, right pallidum, right caudate, left caudate
		+	+	+ 10, -12, +6	1586	9.87	left thalamus, right thalamus, right pallidum, right caudate, left caudate
	left thalamus	-	-	+ 12, -24, +8	3706	14.2	right thalamus, left thalamus, left pallidum, left caudate, right caudate, left putamen, right pallidum
		+	-	-8, -12, +6	1926	10.68	right thalamus, left thalamus, right caudate, left pallidum
		+	+	-8, -12, +6	1362	10.68	right thalamus, left thalamus, left caudate, right pallidum
intra > pre	right thalamus	-	-	-30, + 10, -40	1092	8.91	left TP, left Cereb45, left posterior TFusC, left posterolateral ITG, left posterior PaHC, left anterior STG, left anterior TFusC, left anterior PaHC, left hippocampus, left PP, left Cereb6, left anterolateral ITG, left amygdala, left anterior MTG, brainstem
				-44, + 2, +8	187	5.17	left IC, left CO, left PreCG, left opercular IFG, left FO, left putamen
				0, -70, -38	151	5.67	Ver8, Ver7, right Cereb2, right Cereb7, left Cereb2, right Cereb8, left Cereb8
				+ 46, 0, + 12	146	8.16	right CO, right IC
		+	-	no suprathreshold clusters
		+	+	no suprathreshold clusters
	left thalamus	-	-	+ 26, -18, -32	350	5.74	right anterior PaHC, right anterior TFusC, right amygdala, right hippocampus, right posterior PaHC, right TP, right FOrb, brainstem, right posterior TFusC
				+ 42, + 2, +10	307	6.91	right IC, right CO, right FO, right opercular IFG
				-34, + 14, +6	223	5.88	left IC, left FO l, left CO
				-32, -38, -24	172	5.56	left posterior TFusC, left Cereb45, brainstem, left Cereb6
				-6, -62, -20	136	5.73	Ver8, left Cereb6, Ver6, right Cereb2, Ver7, left Cereb45, left Cereb8, right Cereb7, right Cereb8
		+	-	no suprathreshold clusters
		+	+	no suprathreshold clusters

*PaHC, parahippocampal gyrus; TFusC, temporal fusiform cortex; TP, temporal pole; FOrb, frontal orbital cortex; IC, insular cortex; CO, central opercular cortex; FO, frontal opercular cortex; IFG, inferior frontal gyrus; ITG, inferior temporal gyrus; MTG, middle temporal gyrus; STG, superior temporal gyrus; PP, planum polare; PreCG, precentral gyrus; Cereb, cerebellum; Ver, vermis; pre > intra, brain regions that showed significant FC decrease during anesthesia; intra > pre, brain regions that showed significant FC increase during anesthesia; FD, framewise displacement.

Comparisons of fALFF between pre- and intraoperative conditions, and effects of FD correction and masking

The thalamus, precuneus, posterior cingulate cortex (PCC), supramarginal gyrus, lingual gyrus, occipital pole, lateral occipital cortex, and postcentral gyrus showed significantly decreased fALFF values during anesthesia (Fig. 3A, Table 3). In contrast, fALFF values were significantly increased in the temporal pole, middle temporal gyrus, inferior temporal gyrus, fusiform gyrus, frontal pole, several parts of the cerebellum, and left orbitofrontal cortex during anesthesia (Fig. 3B, Table 3). However, after FD correction, the extent of significant clusters was markedly decreased (Fig. 3C, 3D, Table 3). Conversely, no differences were seen in the results of fALFF analysis, regardless of the application of the exclusion mask (Fig. 3E, 3F, Table 3).

Table 3

Results of fractional amplitude of low-frequency fluctuations (fALFF) analysis
Comparison	FD	Mask	Clusters (x, y, z)	Size	Peak t value	Brain regions
pre > intra	-	-	+ 12, -56, + 30	226	6.91	precuneus, PC, right cuneus
			+ 58, -20, + 28	92	5.25	right anterior SMG, right postCG
			-46, -26, + 30	77	7.14	left anterior SMG, left postCG, left PO
			-16, -08, + 12	76	6.93	left thalamus, left pallidum, left putamen
			-08, -26, + 36	71	7.1	PC
			-20, + 14, +52	53	6.53	left SFG, left midFG
			-40, -12, + 30	53	5.51	left postCG
			+ 40, -80, + 08	47	5.44	right inferior LOC
			+ 40, -56, + 22	42	5.74	right AG
			-14, -106, + 04	42	5.16	left OP
			+ 22, -96, + 18	40	5.94	right OP
			-10, -64, + 38	34	5.38	precuneus
			+ 18, + 14, +50	32	6.68	right SFG
			+ 38, -66, + 02	30	6.33	right inferior LOC
			+ 16, -50, + 02	30	4.36	right LG, right Cereb45
	+	-	+ 40, -82, + 06	64	6.66	right inferior LOC
			+ 10, -58, + 36	60	11.91	precuneus, PC
	+	+	+ 40, -82, + 06	64	6.66	right inferior LOC
			+ 10, -58, + 36	60	11.91	precuneus, PC
intra > pre	-	-	+ 60, + 12, -12	251	8.28	right TP, right anterior MTG, right anterior STG
			+ 46, -44, -20	251	8.5	right temporooccipital ITG, right posterior ITG, right Cereb1, right TOFusC, right posterior MTG, right Cereb2
			-46, -68, -22	199	9.99	left OFusG, left Cereb1, left inferior LOC, left temporooccipital ITG, left TOFusC, left Cereb6
			-08, + 22, -24	136	6.24	SubCalC, left FOrb, MedFC
			-64, -08, -22	127	6.53	left anterior MTG, left posterior MTG, left anterior STG
			-04, + 64, -18	97	5.34	left FP, right FP, MedFC
			-48, + 08, -32	95	7.69	left TP
			+ 54, -22, -38	67	7.18	right posterior ITG, right posterior TFusC
			-40, -46, -52	61	7.05	left Cereb8
	+	-	+ 48, -42, -14	102	7.16	right posterior ITG, right temporooccipital ITG, right Cereb1, right TOFusC
			-42, -70, -16	73	6.94	left inferior LOC, left OFusG, left Cereb1
	+	+	+ 48, -42, -14	102	7.16	right posterior ITG, right temporooccipital ITG, right Cereb1, right TOFusC
			-42, -70, -16	73	6.94	left inferior LOC, left OFusG, left Cereb1
*PC, posterior part of cingulate gyrus; LOC, lateral occipital cortex; SMG, supramarginal gyrus; OP, occipital pole; SFG, superior frontal gyrus; PostCG, postcentral gyrus; LG, lingual gyrus; Cereb, cerebellum; MidFG, middle frontal gyrus; AG, angular gyrus; PO, parietal opercular cortex; TP, temporal pole; ITG, inferior temporal gyrus; SubCalC, subcallosal cortex; OFusG, occipital fusiform gyrus; MTG, middle temporal gyrus; FP, frontal pole; FOrb, frontal orbital cortex; TOFusC, temporal occipital fusiform cortex; MedFC, frontal medial cortex; STG, superior temporal gyrus; TFusC, temporal fusiform cortex; Cereb, cerebellum; pre > intra, brain regions showing significant decrease in fALFF value during anesthesia; intra > pre, brain regions showing significant increase in fALFF value during anesthesia; FD, framewise displacement

Since the purpose of this study was to compare images taken while awake and during surgery and a state of general anesthesia results in virtually no body movement, the effects of the magnitude of body movement needed to be excluded. Corrections were therefore made based on the difference in mean frame displacement for each subject. FD represents the sum of the magnitude of linear motions of the head in the x, y, and z directions and rotational motion in the three directions, providing an indicator of overall head movements.

Previous reports have evaluated the effects of head movements in neonates¹⁵ and the elderly¹⁶ on rs-fMRI using FD. With this method, we were able to exclude the effects of body motion in the FC and fALFF analyses. In the present study, spatial artifacts caused by the intraoperative condition were observed, resulting in signal loss from functional brain images.

To minimize the effects of artifacts while maintaining detection power, a mask was created and adapted to exclude areas in which artifacts were seen in four or more subjects.

EEG monitoring is currently used in clinical practice to monitor the depth of intraoperative anesthesia. However, many problems remain with the widely used EEG monitors, such as the fact that evaluations differ depending on the type of general anesthetic.³ In addition, since general anesthesia in actual surgery is performed not only with general anesthetics, but also with narcotics and muscle relaxants, the effects of those additional drugs must also be taken into consideration. In particular, the EEG monitor uses an EEG database, and the widely used bispectral index (BIS) is not actually measured, but is instead estimated.²

The EEG represents a summation of local postsynaptic potentials of pyramidal cells in the cerebral cortex,¹⁷ and the thalamus plays a major role in the formation of the electroencephalographic rhythm. In particular, the rhythms of sleep spindle waves with a frequency of around 10 Hz and delta waves with a lower frequency, which are predominantly observed during anesthesia, are formed in the thalamic ciliary nucleus.^18,19 Elucidation of the changes in thalamic activity during anesthesia is thus important to understand the background of the BIS score.

The present study found a decrease in fALFF values in the thalamus and a decrease in FC within and between the thalamus during anesthesia. These results indicate that sevoflurane anesthesia reduces both neural activity in the thalamus and interthalamic synchrony. Previous studies on the thalamus and anesthesia have reported that loss of consciousness during anesthesia is associated with hypometabolism in the thalamus, decreased interthalamic FC or decreased thalamocortical interactions.^{11, 12,20,21} The results of the present study demonstrate that those findings from previous studies of reduced activity and FC of the thalamus itself in association with anesthesia-induced loss of consciousness are also observed during balanced anesthesia with sevoflurane.

The present study found that fALFF values in the precuneus and PCC decreased during anesthesia. The PCC and precuneus are components of the DMN, which is thought to play an important role in the formation of consciousness. Previous studies on the DMN and anesthesia have shown that anesthesia is related to impaired FC within the DMN,^22–24 decreases local synchronicity in the DMN²⁵ and decreases neural activity in the DMN.^26,27 The results of the present study and the findings from those previous studies support the association of reduced activity and FC of the DMN with anesthesia-induced loss of consciousness in humans under anesthesia with sevoflurane.

Decreased FC between the thalamus, caudate nucleus and globus pallidus was observed during anesthesia. Previous studies on anesthesia and thalamic-basal ganglia connectivity have reported that sevoflurane anesthesia decreases FC between the thalamus and caudate nucleus,⁹ that isoflurane decreases glucose metabolism in the thalamus and basal ganglia,²⁸ and that propofol decreases neural activity of the basal ganglia.²⁹ However, no previous studies have mentioned the significance of changes in neural activity in the thalamus and basal ganglia during anesthesia. The present results suggest that the cortico-basal ganglia loop through the thalamus may be inhibited under balanced anesthesia with sevoflurane.

Despite the same eye closure condition, decreases in fALFF values were seen in the lingual gyrus, occipital pole, and lateral occipital cortex during anesthesia. Previous studies on anesthesia and activity in visual areas suggest that isoflurane decreases neural activity in the visual cortex,²⁶ that desflurane interferes with visual sensory processing,³⁰ and that volatile anesthetics suppress visual evoked potentials.^31–33

As with findings from previous studies, the present results suggest that anesthesia with sevoflurane results in decreased neural activity in the visual area.

Various limitations to this study need to be kept in mind. Although we chose transsphenoidal surgery, which is relatively less invasive to brain parenchyma, we could not exclude the effects of surgical invasion. We therefore cannot conclude that the results were exclusive to the effects of anesthesia. In particular, we could not perform network analysis using the whole brain, because we excluded areas in which spatial susceptibility artifacts were observed. In addition, the results may vary depending on the dose of inhaled anesthetics and narcotics used, and the generalizability of results from this study need to be verified in the future.

In this study, we performed rs-fMRI during neurosurgery and showed results consistent with previous findings indicating the feasibility of intraoperative rs-fMRI during deep anesthesia. We believe that further findings will help to clarify the characteristics of changes in brain function during intraoperative anesthesia in the context of EEG monitoring.

Data Availability

The datasets generated during the current study are available from the corresponding author on reasonable request.

Competing interests:
The authors declare no conflicts of interest.

Funding

This study was funded by a Grant-in-Aid for Scientific Research from the Japan Society for the Promotion of Science (19K18295).

Author Contribution

Research design: J.W., Y.O., S.K., M.K., M.T., N.O., M.O., E.K., T.S., S.M.Conduct of research: J.W., Y.O., M.K.Patient recruitment: J.W., M.K., M.T.Data collection: J.W., M.K.Analytical tools: Y.O., S.K.Data analysis: J.W., Y.O., S.K.Writing of paper: J.W., Y.O., S.K., N.O., M.O., S.M.

Acknowledgement

The authors are grateful to Yuichiro Somiya, R.T. (Department of Radiology, Kobe University Hospital, Kobe, Hyogo, Japan) for technical assistance with the MRI experiments, and Moritoki Egi, M.D., Ph.D. (Department of Anesthesia, Kyoto University Hospital, Kyoto, Kyoto, Japan) for useful discussions. We also thank FORTE Science Communications (https://www.forte-science.co.jp/) for English language editing. This work is supported by Grant-in-Aid for Scientific Research from the Japan Society for the Promotion of Science (19K18295).

Data Availability

The datasets generated during the current study are available from the corresponding author on reasonable request.

Chaudhuri, S., Banerjee, S., Chattopadhyay, U. & Hussain, S. S. Comparison of recovery times by using bispectral index monitoring versus end-tidal agent concentration monitoring in patients undergoing inhalational general anaesthesia. Indian J Anaesth 66, S161-S168, doi:10.4103/ija.ija_716_21 (2022).
Glass, P. et al. Bispectral analysis measures sedation and memory effect of propofol, midazolam, isoflurane, and alfentanil in healthy volunteers. Anesthesiology 86, 836–847, doi:10.1097/00000542-199704000-00014 (1997).
Andra, E. I., Julie, K. T. & Evan, D. K. Bispectral index monitoring during sedation with sevoflurane, midazolam, and propofol. Anesthesiology 95, 1151–1159, doi:10.1097/00000542-200111000-00019 (2001).
Messner, M., Beese, U., Romstock, J., Dinkel, M. & Tschaikowsky, K. The bispectral index declines during neuromuscular block in fully awake persons. Anesth Analg 97, 488–491, doi:10.1213/01.ANE.0000072741.78244.C0 (2003).
Barr, G., Anderson, R. E., Owall, A. & Jakobsson, J. G. Effects on the bispectral index during medium-high dose fentanyl induction with or without propofol supplement. Acta Anaesthesiol Scand 44, 807–811, doi:10.1034/j.1399-6576.2000.440707.x (2000).
Ramani, R., Qiu, M. & Constable, R. T. Sevoflurane 0.25 MAC preferentially affects higher order association areas: a functional magnetic resonance imaging study in volunteers. Anesth Analg 105, 648–655, doi:10.1213/01.ane.0000277496.12747.29 (2007).
Daniel, G. S., K. L.; Audrey, V.; Alain, P.; Melanie, B.; Carol, D. P.; Andreas, R.; Gerhard, S.; Steven, L.; Denis, J.; Vincent, B.; Rüdiger, I. Change in Whole Brain Dynamics and Connectivity Patterns during Sevoflurane- and Propofol-induced Unconsciousness Identified by Functional Magnetic Resonance Imaging. Anesthesiology 130, 898–911, doi:10.1097/ALN.0000000000002704 (2019).
Palanca, B. J. A., Avidan, M. S. & Mashour, G. A. Human neural correlates of sevoflurane-induced unconsciousness. Br J Anaesth 119, 573–582, doi:10.1093/bja/aex244 (2017).
Martuzzi, R., Ramani, R., Qiu, M., Rajeevan, N. & Constable, R. T. Functional connectivity and alterations in baseline brain state in humans. Neuroimage 49, 823–834, doi:10.1016/j.neuroimage.2009.07.028 (2010).
Yashar, B., Khaled, R., Joy, L. & Thomas, T. L. A component based noise correction method (CompCor) for BOLD and perfusion based fMRI. Neuroimage 37, 90–101, doi:10.1016/j.neuroimage.2007.04.042 (2007).
Ranft, A. et al. Neural Correlates of Sevoflurane-induced Unconsciousness Identified by Simultaneous Functional Magnetic Resonance Imaging and Electroencephalography. Anesthesiology 125, 861–872, doi:10.1097/ALN.0000000000001322 (2016).
Alkire, M. T., Haier, R. J. & Fallon, J. H. Toward a unified theory of narcosis: brain imaging evidence for a thalamocortical switch as the neurophysiologic basis of anesthetic-induced unconsciousness. Conscious Cogn 9, 370–386, doi:10.1006/ccog.1999.0423 (2000).
Zou, Q. H. et al. An improved approach to detection of amplitude of low-frequency fluctuation (ALFF) for resting-state fMRI: fractional ALFF. J Neurosci Methods 172, 137–141, doi:10.1016/j.jneumeth.2008.04.012 (2008).
Power, J. D., Barnes, K. A., Snyder, A. Z., Schlaggar, B. L. & Petersen, S. E. Spurious but systematic correlations in functional connectivity MRI networks arise from subject motion. Neuroimage 59, 2142–2154, doi:10.1016/j.neuroimage.2011.10.018 (2012).
Kim, J. H., De Asis-Cruz, J., Kapse, K. & Limperopoulos, C. Systematic evaluation of head motion on resting-state functional connectivity MRI in the neonate. Hum Brain Mapp 44, 1934–1948, doi:10.1002/hbm.26183 (2023).
Kato, S. et al. Effects of Head Motion on the Evaluation of Age-related Brain Network Changes Using Resting State Functional MRI. Magn Reson Med Sci 20, 338–346, doi:10.2463/mrms.mp.2020-0081 (2021).
Olejniczak, P. Neurophysiologic basis of EEG. J Clin Neurophysiol 23, 186–189, doi:10.1097/01.wnp.0000220079.61973.6c (2006).
Steriade, M., Contreras, D., Curro Dossi, R. & Nunez, A. The slow (< 1 Hz) oscillation in reticular thalamic and thalamocortical neurons: scenario of sleep rhythm generation in interacting thalamic and neocortical networks. The Journal of Neuroscience 13, 3284–3299, doi:10.1523/JNEUROSCI.13-08-03284.1993 (1993).
Steriade, M., Nuiiez, A. & Amzica, F. Intracellular analysis of relations between the slow (< I Hz) neocortical oscillation and other sleep rhythms of the electroencephalogram. The Journal of Neuroscience 13, 3266–3283, doi:10.1523/JNEUROSCI.13-08-03266.1993 (1993).
White, N. S. & Alkire, M. T. Impaired thalamocortical connectivity in humans during general-anesthetic-induced unconsciousness. Neuroimage 19, 402–411, doi:10.1016/s1053-8119(03)00103-4 (2003).
Palanca, B. J. et al. Resting-state Functional Magnetic Resonance Imaging Correlates of Sevoflurane-induced Unconsciousness. Anesthesiology 123, 346–356, doi:10.1097/ALN.0000000000000731 (2015).
Boveroux, P. et al. Breakdown of within- and between-network resting state functional magnetic resonance imaging connectivity during propofol-induced loss of consciousness. Anesthesiology 113, 1038–1053, doi:10.1097/ALN.0b013e3181f697f5 (2010).
Schrouff, J. et al. Brain functional integration decreases during propofol-induced loss of consciousness. Neuroimage 57, 198–205, doi:10.1016/j.neuroimage.2011.04.020 (2011).
Greicius, M. D. et al. Persistent default-mode network connectivity during light sedation. Hum Brain Mapp 29, 839–847, doi:10.1002/hbm.20537 (2008).
Huang, Z. et al. Altered temporal variance and neural synchronization of spontaneous brain activity in anesthesia. Hum Brain Mapp 35, 5368–5378, doi:10.1002/hbm.22556 (2014).
Lv, P. et al. Dose-dependent effects of isoflurane on regional activity and neural network function: A resting-state fMRI study of 14 rhesus monkeys: An observational study. Neurosci Lett 611, 116–122, doi:10.1016/j.neulet.2015.11.037 (2016).
Kaike, K. et al. Effect of sevoflurane, propofol, and adjunct nitrous oxide on regional cerebral blood flow, oxygen consumption, and blood volume in humans. Anesthesiology 99, 603–613, doi:10.1097/00000542-200309000-00015 (2003).
Park, T. Y. et al. Effects of isoflurane anesthesia and intravenous morphine self-administration on regional glucose metabolism ([(18) F]FDG-PET) of male Sprague-Dawley rats. Eur J Neurosci 45, 922–931, doi:10.1111/ejn.13542 (2017).
Martinez-Simon, A. et al. Effect of Dexmedetomidine and Propofol on Basal Ganglia Activity in Parkinson Disease: A Controlled Clinical Trial. Anesthesiology 126, 1033–1042, doi:10.1097/ALN.0000000000001620 (2017).
Lee, H., Tanabe, S., Wang, S. & Hudetz, A. G. Differential Effect of Anesthesia on Visual Cortex Neurons with Diverse Population Coupling. Neuroscience 458, 108–119, doi:10.1016/j.neuroscience.2020.11.043 (2021).
Hayashi, H. & Kawaguchi, M. Intraoperative monitoring of flash visual evoked potential under general anesthesia. Korean J Anesthesiol 70, 127–135, doi:10.4097/kjae.2017.70.2.127 (2017).
Kumar, A., Bhattacharya, A. & Makhija, N. Evoked potential monitoring in anaesthesia and analgesia. Anaesthesia 55, 225–241, doi:10.1046/j.1365-2044.2000.01120.x (2000).
Ma, J. et al. Effects of Sevoflurane-Propofol-Balanced Anesthesia on Flash Visual Evoked Potential Monitoring in Spine Surgery: A Randomized Noninferiority Trial. Anesth Analg 134, 1054–1061, doi:10.1213/ANE.0000000000005742 (2022).

No competing interests reported.

sevofMRI20240416supple.docx

Download PDF

Version 1

posted

You are reading this latest preprint version

Brain activity during intraoperative general anesthesia using resting-state functional magnetic resonance imaging ~Feasibility study~

Status:

Version 1

Abstract

Figures

Introduction

Methods

Design and ethics

Setting and participants

Anesthetic management during operation and fMRI

Acquisition of rs-fMRI data

Analysis of rs-fMRI data

Spatial artifacts on intraoperative fMRI and masking

Statistical analysis

Results

Characteristics of patients

Management of anesthesia

Comparisons of fALFF between pre- and intraoperative conditions, and effects of FD correction and masking

Discussion

Data Availability

Declarations

Competing interests:
The authors declare no conflicts of interest.

Funding

Author Contribution

Acknowledgement

Data Availability

References

Additional Declarations

Supplementary Files

Status:

Version 1

Brain activity during intraoperative general anesthesia using resting-state functional magnetic resonance imaging ~Feasibility study~

Status:

Version 1

Abstract

Figures

Introduction

Methods

Design and ethics

Setting and participants

Anesthetic management during operation and fMRI

Acquisition of rs-fMRI data

Analysis of rs-fMRI data

Spatial artifacts on intraoperative fMRI and masking

Statistical analysis

Results

Characteristics of patients

Management of anesthesia

Comparisons of fALFF between pre- and intraoperative conditions, and effects of FD correction and masking

Discussion

Data Availability

Declarations

Competing interests: The authors declare no conflicts of interest.

Funding

Author Contribution

Acknowledgement

Data Availability

References

Additional Declarations

Supplementary Files

Status:

Version 1

Competing interests:
The authors declare no conflicts of interest.