Early-in-life isoflurane exposure alters resting-state functional connectivity in juvenile non-human primates

Abstract

Background: Clinical studies suggest that anaesthesia exposure early in life affects neurobehavioural development. We designed a non-human primate (NHP) study to evaluate cognitive, behavioural, and brain functional and structural alterations after isoflurane exposure during infancy. These NHPs displayed decreased close social behaviour and increased astrogliosis in specific brain regions, most notably in the amygdala. Here we hypothesise that resting-state functional connectivity MRI can detect alterations in connectivity of brain areas that relate to these social behaviours and astrogliosis.

Methods: Imaging was performed in 2-yr-old NHPs under light anaesthesia, after early-in-life (postnatal days 6-12) exposure to 5 h of isoflurane either one or three times, or to room air. Brain images were segmented into 82 regions of interest; the amygdala and the posterior cingulate cortex were chosen for a seed-based resting-state functional connectivity MRI analysis.

Results: We found differences between groups in resting-state functional connectivity of the amygdala and the auditory cortices, medial premotor cortex, and posterior cingulate cortex. There were also alterations in resting-state functional connectivity between the posterior cingulate cortex and secondary auditory, polar prefrontal, and temporal cortices, and the anterior insula. Relationships were identified between resting-state functional connectivity alterations and the decrease in close social behaviour and increased astrogliosis.

Conclusions: Early-in-life anaesthesia exposure in NHPs is associated with resting-state functional connectivity alterations of the amygdala and the posterior cingulate cortex with other brain regions, evident at the juvenile age of 2 yr. These changes in resting-state functional connectivity correlate with the decrease in close social behaviour and increased astrogliosis. Using resting-state functional connectivity MRI to study the neuronal underpinnings of early-in-life anaesthesia-induced behavioural alterations could facilitate development of a biomarker for anaesthesia-induced developmental neurotoxicity.

Keywords: amygdala; anaesthesia; brain development; connectivity; neuroimaging; neurotoxicity; non-human primates; posterior cingulate cortex.

Fig 1

Alterations in functional connectivity of the amygdala with the primary auditory cortex after early-in-life anaesthesia exposure. Functional connectivity (Fisher's z transformed values, no unit) of (a) the left amygdala and the right primary auditory cortex, and (b) the right amygdala and the left primary auditory cortex in juvenile non-human primates aged 2 yr after single (1X), three times (3X), or no exposure (Ctr) to isoflurane during infancy. Data shown are mean (sd) with post hoc adjusted P-values (Tukey's multiple comparison); Ctr: n=6, 1X: n=5, 3X: n=6.

Fig 2

Alterations in functional connectivity of the amygdala with the medial premotor cortex and the posterior cingulate cortex (PCC) after early-in-life anaesthesia exposure. Functional connectivity (Fisher's z transformed values, no unit) of left (a) amygdala with the left medial premotor cortex, and (b) the right amygdala with the left PCC in juvenile non-human primates aged 2 yr after single (1X), three times (3X), or no exposure (Ctr) to isoflurane during infancy. Data shown are mean (sd), with post hoc adjusted P-values (Tukey's multiple comparison); Ctr: n=6, 1X: n=5, 3X: n=6.

Fig 3

Alterations in functional connectivity of the posterior cingulate cortex (PCC) with other brain areas after early-in-life anaesthesia exposure. Functional connectivity (Fisher's z transformed values, no unit) of left PCC with (a) the secondary auditory cortex, and (b) the left polar prefrontal cortex (PFCpol), as well as of the right PCC with (c) the left polar temporal cortex (TCpol) and (d) the left anterior insula cortex in juvenile non-human primates aged 2 yr after single (1X), three times (3X), or no exposure (Ctr) to isoflurane during infancy. Data shown are mean (sd), with post hoc adjusted P-values (Tukey's multiple comparison); Ctr: n=6, 1X: n=5, 3X: n=6.

Fig 4

Correlation between close social behaviour and alterations in functional connectivity. Correlations between close social behaviour ratios (% of time) and functional connectivities (Fisher's z transformed values) of brain area pairs (indicated above each plot) that showed significant resting-state functional connectivity alterations between groups. Data are plotted for each non-human primate with the exposure group indicated in colour. Correlations and linear regressions were analysed either (a) separately for each exposure group or (b–h) regardless of the exposure group to account for a treatment (isoflurane exposure) effect in a (see Results). Linear regression equations are displayed along with correlation coefficients r (95% confidence intervals) and P-values of the correlation analyses; Ctr: n=6, 1X: n=5, 3X: n=6.

Fig 5

Correlation between astrogliosis and alterations in functional connectivity of the right amygdala. Astrogliosis (glial fibrillary acidic protein [GFAP] immunoreactivity) and functional connectivity (Fisher's z transformed values) between the right amygdala and the brain areas (indicated above each plot) that showed significant resting-state functional connectivity alterations between groups. Data are plotted for each non-human primate with the exposure group indicated in colour. Linear regression equations are shown along with the correlation coefficients r (95% confidence intervals) and the P-values of the correlation analyses; Ctr: n=6, 1X: n=3, 3X: n=4. PCC, posterior cingulate cortex.