The brain stem plays an essential role in the processing of visceroceptive information. In particular, the periaqueductal gray (PAG) and pontine micturition (Barrington’s nucleus) and storage centers are brainstem areas indicated to play an essential role in lower urinary tract (LUT) control. Post-mortem human and animal studies have indicated that the PAG is symmetrically organized in functionally and anatomically distinct columns which are involved in sympathetic or parasympathetic control of the LUT. Pontine regions involved in LUT control have been identified in humans using neuroimaging approaches and animal studies has indicated direct interactions between the PAG and pontine nuclei involved in LUT control (1,2). In earlier work, it has been shown that in vivo parcellation of the PAG into symmetrical clusters can be accomplished using 7T resting-state fMRI (3). In this study, we aim to investigate changes in functional connectivity patterns between PAG clusters and voxels in the pons associated with changes in bladder sensations.

For this analysis, we evaluated data from 8 female participants. During a bladder filling protocol, we ran a baseline resting-state fMRI scan while participants had an empty bladder and experienced no desire to void, followed by a full bladder resting-state fMRI while participants experienced a strong desire to void. Data were preprocessed using BrainVoyager and normalized to MNI space. PAG and
pons voxels were selected based on two separate masks generated in MNI space and covering the PAG and pons. A voxel-by-voxel correlation matrix of the PAG was created and parcellated using the Louvain module detection algorithm. For each resulting cluster, functional connectivity was assessed with each voxel in the pons for both fMRI scans. Based on these data we computed the difference in PAG-pons functional connectivity between empty and full bladder scans for each separate PAG cluster per participant. Next, we computed the difference in PAG-pons functional connectivity patterns between empty and full bladder scans for 1000 randomly generated PAG clusters per participant to obtain a distribution under the null hypothesis. We then statistically compared our observed values to what would be expected based on chance under the null distribution.

For each participant, we observed several pontine regions of voxels exhibiting large differences in functional connectivity patterns between empty and full bladder states for different PAG clusters. Interestingly, different PAG clusters appear to have opposite interaction patterns with the pons. The anti-correlation values (used to determine the extent to which interaction patterns are each other's opposite) between the difference maps (empty vs. full bladder) for different PAG clusters and the pons were significantly higher for at least one cluster comparison per participant than could be expected based on chance (p ≤ 0.05, after multiple comparison corrections). This indicates that different PAG clusters show significant opposite changes in functional connectivity with the pons during bladder filling.

The approach introduced and outlined here can be used to assess interaction between brain stem nuclei associated with visceroceptive processing at unprecedented resolution. Neuroanatomical and electrophysiological studies of the PAG indicate that different PAG columns are associated with sympathetic or parasympathetic functions. The opposing interactions between different PAG clusters resulting from our parcellation procedure with the pons indicates that the observed pontine clusters may be involved in opposing functional processes associated with changes in bladder sensations, such as the pontine micturition and storage center. Furthermore, it may be stipulated these functional connectivity patterns indicate that parcellations of the PAG using the Louvain module detection algorithm results in meaningful clustering of the PAG into clusters that show distinct functional connectivity patterns with other brain areas.

The results outlined here indicate that ultra-high field (7 Tesla) MRI can be utilized to study interaction patterns between brain stem nuclei during bladder filling. The PAG was subdivided into different functional clusters which showed opposite interaction patterns with the pons during bladder filling. These results enable assessment of brain stem activity associated with bladder sensations and will help to study mechanisms of motor control of the lower urinary tract in unprecedented detail in human volunteers. Since brain stem regions have been indicated to be involved in several types of LUT dysfunction, and mechanisms underlying therapeutic approaches such as sacral neuromodulation are suggested to involve these nuclei, a better understanding of these pathways is essential to improve patient treatment.

Michels, L., Blok, B. F., Gregorini, F., Kurz, M., Schurch, B., Kessler, T. M., ... & Mehnert, U. (2015). Supraspinal control of urine storage and micturition in men—an fMRI study. Cerebral cortex, 25(10), 3369-3380.
Verstegen, A. M., Klymko, N., Zhu, L., Mathai, J. C., Kobayashi, R., Venner, A., ... & Zeidel, M. L. (2019). Non-Crh glutamatergic neurons in Barrington’s nucleus control micturition via glutamatergic afferents from the midbrain and hypothalamus. Current Biology, 29(17), 2775-2789.
de Rijk, M. M., van den Hurk, J., Rahnama'i, M. S., & van Koeveringe, G. A. (2021). Parcellation of human periaqueductal gray at 7-T fMRI in full and empty bladder state: The foundation to study dynamic connectivity changes related to lower urinary tract functioning. Neurourology and Urodynamics, 40(2), 616-623.

Continence 7S1 (2023) 100758
DOI: 10.1016/j.cont.2023.100758