Lumbar to sacral root rerouting to restore bladder function in a feline spinal cord injury model: Urodynamic and retrograde nerve tracing results from a pilot study

Abstract

Aims: Lumbar to sacral rerouting surgery can potentially allow voiding via a skin-central nervous system-bladder reflex pathway. Here, we assessed if this surgery was effective in treating neurogenic bladder dysfunction/sphincter in felines.

Methods: Eight cats underwent spinal cord transection (SCT) at thoracic level 10/11. Unilateral L7 to S1 ventral root anastomosis was performed 1 month later in six cats. Two cats served as transection-only controls. Electrical and manual stimulation of L6-S1 dermatomes, and urodynamics were performed at 3, 5, 7, and 9/10 months post transection. At 9/10 months, cats were also evaluated by direct electrophysiological testing of anastomosed roots with urodynamics, then tissue collection and examination of the root anastomosis site and lumbosacral cord ventral horns for cells retrogradely labeled from tracer dye injected 2 weeks earlier into the bladder wall.

Results: At 9/10 months, four of six rerouted cats exhibited increased detrusor pressure provoked by cutaneous stimulation, one cat bilaterally. Two cats presented with a voiding stream after ipsilateral cutaneous stimulation at 7 and 9 months. All six rerouted animals showed regrowth of axons from the L7 ventral horn to the bladder, although some aberrant axonal regrowth was also observed.

Conclusion: L7 to S1 ventral root rerouting below the level of SCT showed successful axonal regrowth to the bladder from the L7 spinal cord segment in all rerouted animals, and induced increased detrusor pressure response to cutaneous stimulation in a subset. This feasibility study paves the way for future animal studies for bladder reinnervation.

Keywords: cat; neurogenic bladder/sphincter dysfunction; rerouting surgery; spinal cord injury; spinal cord transection; voiding dysfunction.

Figure 1

Design and a rerouted root example. (A) Spinal cord surgery methods. Spinal cords were transected at T10 or T11. One month post spinal transection, the proximal ventral root of L7 was cut and transferred to the distal ventral root of S1, and anastomosed end-on-end (termed rerouted), intradurally. The remaining dorsal and ventral roots of L6, S1 and below were transected proximal to the dorsal root ganglia, thereafter. The L7 dorsal root ganglion (DRG) was left intact. (B) An example (Cat 6) showing a left L7 ventral root transfer with anastomosis to a left S1 ventral root. The transferred L7 root was cut proximal to the anastomosis site during terminal electrophysiological testing. The intact L7 dorsal root and its DRG is shown at left, the caudal end of the cord is shown at right and is marked blue with a histological tissue marker.

Figure 2

Maximum detrusor pressure (cm H₂O) occurring in response to stimulation of anterior thigh dermatomes L6-S1 at 7 and 9/10 months post transection surgery. Cat 1 did not undergo scratching or electrical stimulation at 9 months. (A) Scratching stimulation of these dermatomes. (B) Electrical stimulation of these dermatomes. Results for each cat is shown individually, ipsilaterally and contralaterally. Incidence of a triggered voiding stream (TSV) in response to stimulation is noted on occurrence. Results of two-way ANOVAs and post hoc tests are shown. (C) Cysto manometry coupled with electromyography of the urethral sphincter showing increased detrusor pressure associated with detrusor sphincter synergy and a triggered voiding stream (TSV; also indicated as a “Leak” on the figure) that was triggered by scratching stimulation of the left L7 dermatome in Cat 6.

Figure 3

Maximum detrusor pressure (cm H₂O) occurring in response to direct electrical stimulation of L6 to S1 ventral roots at 9/10 months post transection surgery. Testing was performed both extradurally and intradurally following a laminectomy. (A) Results for each cat shown individually for ventral roots, bilaterally, except for Cat 3 for which only ipsilateral data is presented here, as described in the text. (B) Mean results for direct electrical stimulation of intact roots are shown, prior to terminal root transection. There were no significant (n.s.) differences between groups in using the factors, transection group, side or segment. (C) The roots were transected and retested via direct electrical stimulation of their proximal end (not all animals or roots were assayed post root transection). Previously increased bladder pressure in these same roots is now mostly absent, although a low level of firing in the proximal end of the left L6 root was preserved in Cat 1.

Figure 4

Representative microscope images showing the root anastomosis site in three different L7-to-S1 rerouted animals. (A) A representative image from Cat 1 showing regrowth of axons across the ventral root anastomosis site in a H&E stained longitudinal section of the roots. The arrows indicate sutures. (B) Higher power photograph of the sutures indicated in panel A. (C) An image from Cat 2 showing a similar scenario as Cat 1 of clear regrowth across the anastomosis site. The sutures auto fluoresce in the ultraviolet wavelength allowing further confirmation of the surgery site. (D) An image from another cat (Cat 5) showing a kink at the nerve anastomosis site, yet regrowth of nerve axons to the right of suture site (a blue suture was used, arrow). (E) An image of a contralateral transected roots showing scar tissue between the cut ends, although close proximity and a large number of disordered axons. (F) Representative low power figure of longitudinal sections of the lumbosacral spinal cord showing an intact undamaged ventral horn region from the L7 segment through the upper part of S1 and the presence of many hematoxylin-stained neuronal cell bodies. (G) Representative figure of Fluorogold labeled neuronal cell bodies in the L7 spinal cord segment. Inset shows a higher power image of Fluorogold labeled neuronal cell bodies.

Figure 5

Fluorogold labeled neurons in ventral horn regions of L6-S2 spinal cord segments at 9/10 months post transection. (A) Estimated number of Fluorogold labeled neuronal cell bodies is shown for each cat and segment analyzed on the side ipsilateral to the anastomosed roots. Bilateral results from the transected control animal in which the Fluorogold labeling was successful (Cat 7) is shown as mean ± SEM. Data from the cat in which both right and left ventral roots were annealed to the right S1 by the time of tissue collection (Cat 3) is shown with the ipsilateral data, as mean ± SEM. (B) Contralateral data. (C) Same data as in panels A and B reorganized by segment and side. Results of two-way ANOVA and posthoc results are shown in panels A and C. *p<0.05, compared to the transected control animal.