Effects of Anesthesia on Cerebral Blood Flow and Functional Connectivity of Nonhuman Primates

Abstract

Nonhuman primates (NHPs) are the closest living relatives of humans and play a critical and unique role in neuroscience research and pharmaceutical development. General anesthesia is usually required in neuroimaging studies of NHPs to keep the animal from stress and motion. However, the adverse effects of anesthesia on cerebral physiology and neural activity are pronounced and can compromise the data collection and interpretation. Functional connectivity is frequently examined using resting-state functional MRI (rsfMRI) to assess the functional abnormality in the animal brain under anesthesia. The fMRI signal can be dramatically suppressed by most anesthetics in a dose-dependent manner. In addition, rsfMRI studies may be further compromised by inter-subject variations when the sample size is small (as seen in most neuroscience studies of NHPs). Therefore, proper use of anesthesia is strongly demanded to ensure steady and consistent physiology maintained during rsfMRI data collection of each subject. The aim of this review is to summarize typical anesthesia used in rsfMRI scans of NHPs and the effects of anesthetics on cerebral physiology and functional connectivity. Moreover, the protocols with optimal rsfMRI data acquisition and anesthesia procedures for functional connectivity study of macaque monkeys are introduced.

Keywords: alfaxalone; functional connectivity; isoflurane; ketamine; monkey; rsfMRI.

Figure 1

Demonstration of the multiband MRI technique to improve detection sensitivity of BOLD signal in anesthetized monkeys on a Siemens TIM Trio 3T scanner with an 8-channel array volume coil. (A) Normalized correlation maps of the default mode network (DMN) of anesthetized (1% isoflurane) rhesus monkeys scanned by using the multiband echo planar imaging (multiband EPI) sequence with the parameters: TR = 1059 ms/TE = 25 ms, MB factor = 4, voxel size=1.5 mm × 1.5 mm × 1.5 mm, time of acquisition = 10 min), and conventional single-shot EPI acquisition (conventional EPI) with TR = 2190 ms/TE=25 ms, voxel size=1.5 mm × 1.5 mm × 1.5 mm, time of acquisition = 10 min. The z-score threshold: p < 8 × 10⁻¹⁸ plus cluster threshold 169 mm³/overall, mean ± stdev. n = 4. *: p = 0.08, paired t-test. (B) Comparison of normalized voxel numbers (normalized by average voxel numbers across DMN from the two EPI acquisitions (multiband EPI vs. single EPI). MB, multiband.