Spontaneous Neural Activity Changes after Bariatric Surgery: a resting-state fMRI study

Abstract

Background: Metabolic disorders associated with obesity could lead to alterations in brain structure and function. Whether these changes can be reversed after weight loss is unclear. Bariatric surgery provides a unique opportunity to address these questions because it induces marked weight loss and metabolic improvements which in turn may impact the brain in a longitudinal fashion. Previous studies found widespread changes in grey matter (GM) and white matter (WM) after bariatric surgery. However, findings regarding changes in spontaneous neural activity following surgery, as assessed with the fractional amplitude of low frequency fluctuations (fALFF) and regional homogeneity of neural activity (ReHo), are scarce and heterogenous. In this study, we used a longitudinal design to examine the changes in spontaneous neural activity after bariatric surgery (comparing pre- to post-surgery), and to determine whether these changes are related to cardiometabolic variables. Methods: The study included 57 participants with severe obesity who underwent sleeve gastrectomy (SG), biliopancreatic diversion with duodenal switch (BPD), or Roux-en-Y gastric bypass (RYGB), scanned prior to bariatric surgery and at follow-up visits of 4 months (N=36), 12 months (N=29), and 24 months (N=14) after surgery. We examined fALFF and ReHo measures across 1022 cortical and subcortical regions (based on combined Schaeffer-Xiao parcellations) using a linear mixed effect model. Voxel-based morphometry (VBM) based on T1-weighted images was also used to measure GM density in the same regions. We also used an independent sample from the Human Connectome Project (HCP) to assess regional differences between individuals who had normal-weight (N=46) or severe obesity (N=46). Results: We found a global increase in the fALFF signal with greater increase within dorsolateral prefrontal cortex, precuneus, inferior temporal gyrus, and visual cortex. This effect was more significant 4 months after surgery. The increase within dorsolateral prefrontal cortex, temporal gyrus, and visual cortex was more limited after 12 months and only present in the visual cortex after 24 months. These increases in neural activity measured by fALFF were also significantly associated with the increase in GM density following surgery. Furthermore, the increase in neural activity was significantly related to post-surgery weight loss and improvement in cardiometabolic variables, such as insulin resistance index and blood pressure. In the independent HCP sample, normal-weight participants had higher global and regional fALFF signals, mainly in dorsolateral/medial frontal cortex, precuneus and middle/inferior temporal gyrus compared to the obese participants. These BMI-related differences in fALFF were associated with the increase in fALFF 4 months post-surgery especially in regions involved in control, default mode and dorsal attention networks. Conclusions: Bariatric surgery-induced weight loss and improvement in metabolic factors are associated with widespread global and regional increases in neural activity, as measured by fALFF signal. These findings alongside the higher fALFF signal in normal-weight participants compared to participants with severe obesity in an independent dataset suggest an early recovery in the neural activity signal level after the surgery.

Intermittent energy restriction changes the regional homogeneity of the obese human brain

“…Taken together, these findings revealed a possible underlying pathophysiology between restrained eating and the risk of weight gain (Dong et al, 2015). In addition, the ReHo method was used to study the effect of bariatric surgery-induced weight loss on brain activity (Zeighami et al, 2021). Zeighami et al (2021) observed a significant increase in ReHo in the visual cortex, medial temporal gyrus, and DLPFC at 4 months after surgery.…”

“…Although neuroimaging studies on obesity have identified obesity-related abnormalities in a wide range of brain areas, these abnormalities can be reshaped after weight loss by CER involving reward cognitive control and sensory processing (Carnell et al, 2012;Zhang et al, 2015;Belaïch et al, 2017;Zeighami et al, 2021). Amanda et al used the method of caloric restriction for weight loss in obese patients for 3 months and found that prefrontal cortical activation influences health behavior changes in the context of an intervention to produce weight loss and manage obesity (Szabo-Reed et al, 2020).…”

“…Recently, resting-state fMRI (rsfMRI) has been used to investigate functional alterations in obesity, and this method has unique advantages in clinical research (García-García et al, 2015;Zhang et al, 2015;Ding et al, 2020;Zhang et al, 2020;Parsons et al, 2022). ReHo is a data-driven method for rsfMRI that reflects spontaneous neuronal activity from different perspectives and demonstrates excellent performance in depicting clinical traits (Van Opstal et al, 2019;Zeighami et al, 2021;Li et al, 2022). It measures the similarity or synchronicity of the time series of the nearest neighboring voxels and can reflect the strength of local spontaneous neural activity in the brain (Zang et al, 2004;Li et al, 2022).…”

“…In addition, the ReHo method was used to study the effect of bariatric surgery-induced weight loss on brain activity (Zeighami et al, 2021). Zeighami et al (2021) observed a significant increase in ReHo in the visual cortex, medial temporal gyrus, and DLPFC at 4 months after surgery. Therefore, ReHo analysis can be used to gain insights into the neural basis underlying IER for weight loss.…”

Neuroanatomical correlates of genetic risk for obesity in children

“…Brain age for each participant in the ABCD dataset was estimated based on imaging features and the model described above. For each participant, we calculated a difference between chronological age and brain age – delta age [62, 63]. Finally, we correlated delta age with BMI-PRS (corrected for first 20 genetic principal components).…”