Neuromodulation of Pelvic Nerves Increases Nerve Regeneration After Injury

Introduction: Perineal and pelvic floor nerves experience excessive stretching during pregnancy and childbirth, and consequent nerve injury contributes to stress urinary incontinence (SUI). This condition is characterized by involuntary urine leakage during sudden increase in intra-abdominal pressure, resulting from failure in urethral closure by pelvic muscles which normally serve as secondary urethral sphincters. Female rabbits are an accepted model for SUI, and aging and multiparous animals show damaged perineal and pelvic nerves, weakening pelvic floor muscles and reduced bladder efficiency. We have reported that electrical stimulation (ES) of the bulbospongiosus nerve (BsN) in these animals reversed SUI-like deficits. While the mechanism of action is not fully understood, it is known that ES to somatic injured nerves accelerates their intrinsic regenerative capacity. To confirm that neuromodulation to pelvic nerves can also have a regenerative effect, we tested the effect of daily stimulation for 3 weeks post injury, using a uniform nerve damage model (i.e., hemi-crush). The effect of this treatment was evaluated specifically in the early stages of nerve repair including a) Wallerian degeneration (WD), b) Myelin compaction, c) Axon growth, d) Radial axonal sorting, and e) Re-myelination.

Materials and

Methods: A total of 8 young nulliparous female rabbits were included and divided into three groups: 1) healthy (n=2), hemi-crush injury (n=3), and hemi-crush injury with an implanted NeuroClip electrode receiving ES for 3 weeks (n=3). The hemi-crush injury was inflicted unilaterally. Under gas anesthesia, the perineal skin was cut and BsN identified and released. A No. 5 Dummond Forceps with a frozen tip was used to crush half the nerve for 3 seconds. Injury was verified visually as a transparent area due to myelinated axon transection. In group 3, a wireless NeuroClip electrode was implanted in the injured BsN and the wound was closed. The animals were allowed to recover for 2 weeks. ES was applied wirelessly using 4 cycles of 30 second pulse trains (0.9 mAmp) at 30 Hz (200usec pulse width) and 150 seconds of resting period between cycles. ES was applied 3 times per week for a total of for 3 weeks. At the end of the study, the animals were euthanized, and BsN nerves collected for histology. Total nerve samples: 5 healthy (3 from group 1 and 2 from contralateral side of group 2), 3 injured and 3 injured with ES. Nerve samples were prepared for toluidine-blue staining and Transmission Electron Microscopy (TEM) imaging. Toluidine-blue images were used to quantify the total nerve diameter and area, and total number of myelinated axons using ImageJ. The TEM images were analyzed to quantify ovoid formations, WD and Schwann cells (SCs). Statistical analysis was performed using GraphPad Prism.

Results, Conclusions, and Discussions: Normal WD was observed at 1.32 ± 1.32 per 10,000 µm2. Injury increased this to 5.00 ± 0.98 per 10,000 µm2, p< 0.05, and was comparable to ES (4.36 ± 0.41 per 10,000 µm2). Normal ovoid formation was 0.86 ± 0.59 per 10,000 µm2 and 5.06 ± 0.94 per 10,000 µm2 in injured animals. ES reduced that significantly to 1.73 ± 0.44 per 10,000 µm2 p< 0.05), suggesting ES accelerated WD and myelin clearance.

Cryo-crush decreased Remak bundle size by half (healthy: 40.75 ± 9.33 µm2 vs injured: 16.32 ± 1.99 µm2), while UA number per bundle remained comparable (healthy: 7.63 ± 1.27 UA/bundle vs injured: 6.130 ± 0.62 UA/bundle). Following ES, size and UA number per bundle increased significantly (71.71 ± 11.55 µm2 and 16.36 ± 2.06 UA/bundle, p< 0.0001 vs injured group), indicating increased UA proliferation post-regeneration, and resulting in significantly increased bundled UA density compared to healthy (treatment: 14.31 ± 2.32 per 1000 µm2 vs healthy: 2.44 ± 0.88 per 1000 µm2 p< 0.05, injured: 3.84 ± 0.50 per 1000 µm2). Healthy group showed highest % of UA in Remak bundles undergoing proper radial sorting (78.69%). Injury and treatment groups showed 13.48% and 47.19% UA with proper radial sorting respectively.

SC density was 6.73 ± 2.04 cells/10,000 µm2 in normal nerves and reduced in hemi-crush injury to 1.71 ± 0.39 cells/10,000 µm2, p< 0.05). Neuromodulation significantly increased SC proliferation 7.6-fold compared to injured (12.98 ± 2.10 cells/10,000 µm2, p< 0.005). Total myelinated axons (MA) in healthy group were 87.74 ± 17.77. Injury reduced MA by 44% (38.33 ± 7.80). Neuromodulation increased that by 18% compared to injured (56.67 ± 6.09). Axonal density reduced from 7.02 ± 1.30 axons/1000 µm2 to 2.10 ± 0.02 axons/1000 µm2 in injured animals. ES increased this by 31% (2.77 ± 0.46 axons/1000 µm2) p< 0.05) compared to injury.

Neuromodulation promotes nerve regeneration post-injury in perineal nerves by accelerating WD, increasing SCs and radial sorting of UA formation. We demonstrated perineal nerve ES is neuro-regenerative, explaining the reported increase in functional recovery of bladder efficiency in aging and multiparous rabbits, offering a plausible mechanism of action for this therapy.

Acknowledgements (Optional):