In this original study, we evaluated the feasibility of using 3 Tesla functional magnetic resonance imaging (fMRI) to measure changes in activity in the sacral spinal cord during voluntary contraction and relaxation of the pelvic floor. Voluntary control of the lower urinary tract (LUT), therefore the maintenance of continence, is highly dependent on efficient bi-directional communication between the sacral spinal cord and the brain. Measuring activity in the sacral spinal cord is largely underexplored due to the difficult accessibility using conventional fMRI data analysis techniques. However, evaluation of physiological and pathophysiological functional states in this area is important to improve our understanding of non-neurogenic LUT conditions such as dysfunctional voiding, non-obstructive retention or Fowler’s syndrome. Additionally,  neurogenic patients with conditions such as multiple sclerosis, Parkinson’s disease, or multiple systems atrophy,  which commonly present with LUT dysfunction, may present with maladaptive activity patterns in the brain or spinal cord. Gaining access to evaluate functional and structural integrity of the sacral spinal cord is essential to improve our understanding of the contribution of (mal)adaptive changes in this region associated with LUT dysfunction and will help to improve the diagnostic and therapeutic options for these patients. In the current study, we designed a protocol and data-analysis pipeline to measure changes in blood-oxygen-level-dependent (BOLD) signal in the sacral spinal cord during voluntary contraction and relaxation of the pelvic floor.

For this proof of concept we collected data from 1 healthy female volunteer on two separate recording days. The participant was trained by a pelvic floor physiotherapist to perform isolated contractions of the pelvic floor. MRI data were obtained using a 3T MAGNETOM Prisma Fit (Siemens) scanner and Spine Matrix Coil (Siemens). The participant was instructed to lie down on the scanner bed in supine position, with bent knees and calves supported by a foam cushion. First we obtained anatomical data of the sacral spinal cord with our main region-of-interest centered around the S1 segment of the spinal cord. Next the participant was instructed to perform repeated sustained contraction and relaxation of the pelvic floor in blocks of 8 seconds each (i.e., 8 seconds on, 8 seconds off), with 24 repetitions. Data was preprocessed (slice-scan-time and motion corrected, temporally filtered) and differences in activity between contraction and relaxation of the pelvic floor was statistically assessed.

On both recording days we observed voxels that statistically change in their activity between contraction and relaxation of the pelvic floor (p ≤ 0.05). We found significant changes in activity in voxels in both the ventral horn (at the approximate location of Onuf’s nucleus), as well as the dorsal horn (figure 1).

Our results indicate that assessment of sacral spinal cord activity using 3T fMRI is feasible. We show that the ventral horn of the S1 segment changes significantly in activity between contraction and relaxation of the pelvic floor. In addition to the ventral horn, we also observe a change in activity in the dorsal horn which indicates that sensory processes also changes consistently with our task (most likely these are proprioceptive processes associated with pelvic floor contractions). This proof of concept suggests that functional neuroimaging using fMRI of the spinal cord is a tool that can be utilized to study LUT control in this region of the central nervous system.

In conclusion, our study successfully demonstrates the feasibility of utilizing 3T fMRI to investigate changes in sacral spinal cord activity during pelvic floor contractions and relaxations. The further refinement of this novel approach holds promise for enhancing diagnostic and therapeutic strategies for individuals with a spectrum of both neurogenic and non-neurogenic LUT conditions.