Multiple Reflex Pathways Contribute to Bladder Activation by Intraurethral Stimulation in Persons with Spinal Cord Injury

Abstract

Objectives

To measure the urodynamic effects of electrical co-stimulation of two individual sites in the proximal and distal urethra in persons with spinal cord injury (SCI). This work was motivated by preclinical findings that selective co-stimulation of the cranial urethral sensory nerve and the dorsal genital nerve, which innervate the proximal and distal portions of the urethra, respectively, increased reflex bladder activation and voiding efficiency.

Methods

Electrical co-stimulation of urethral afferents was conducted in persons with chronic SCI during urodynamics. The effects of different frequencies of intraurethral stimulation at multiple urethral locations on bladder pressure and pelvic floor electromyographic (EMG) activity were measured.

Results

EMG activity indicated that multiple reflex pathways were recruited through stimulation that contributed to bladder activation. The size of reflex bladder contractions evoked by stimulation was dependent on stimulation location or reflex activated and stimulation frequency.

Conclusions

Pudendal nerve afferents are a promising target to restore lost bladder control, as stimulation with different frequencies may be used to treat urinary incontinence and increase continent volumes or to generate stimulation-evoked bladder contractions for on-demand voiding. This work identified that co-stimulation of multiple afferent reflex pathways can enhance activation of spinal circuits and may enable improved bladder emptying in SCI when stimulation of a single pathway is not sufficient.

Introduction

Lower urinary tract dysfunction, including neurogenic detrusor overactivity, urinary incontinence, chronic retention of urine, and detrusor-sphincter dyssynergia, can be caused by spinal cord injury (SCI) [1,2]. Bladder management in persons with SCI typically includes intravesical botulinum toxin together with anticholinergic medications and intermittent catheterization. However, these treatments often involve side effects and frequent urinary tract infections, further impacting quality of life [2]. Electrical stimulation is an alternative approach for restoration of bladder function that avoids these side effects [3].

Several different stimulation locations have been investigated to restore bladder control in SCI. Sacral anterior root stimulation (SARS) is effective for bladder control following SCI, but not widely accepted due to the requirement for dorsal rhizotomy [4,5], which impairs residual sensation, reflex erection and defecation. Sacral nerve stimulation (SNS) with Interstim® is less invasive, but is generally ineffective in SCI [6]. Pudendal nerve stimulation (PNS) can produce bladder inhibition to maintain continence or bladder activation to produce voiding in persons with SCI [7–13] and thus represents a strong potential alternative treatment.

Although promising, stimulation of the compound pudendal nerve in humans with SCI has not produced sufficiently efficient voiding. Selective co-stimulation of two distinct branches of the PN (dorsal genital, DGN, and cranial urethral sensory nerve, CSN) in the cat increased reflex bladder contractions and produced more efficient voiding than single branch stimulation [14]. These distal branches of the PN innervate the proximal and distal urethra [15], and intraurethral stimulation that selectively targets each of these branches produces bladder responses comparable to those produced by direct nerve stimulation [16,17], studies in patients with SCI demonstrate bladder activation is possible through proximal and distal urethra stimulation [18]. The purpose of this study was to test the hypothesis that selective co-stimulation of distal branches of the pudendal nerve (PN) enhances bladder activation as compared to single-site stimulation.

Materials and Methods

All study procedures were approved by the Duke University Health System Institutional Review Board. Written informed consent was obtained from each subject. Seventeen subjects (13 males, 4 females) at least one year post suprasacral SCI were enrolled in the study (Table 1).

Table 1.

Summary of spinal cord injured participant demographics and responses to intraurethral stimulation

n	Sex	Age	Injury Level	Complete or Incomplete SCI	Years Since Injury	Proximal or Distal Stimulation-Evoked Reflexeas						Δ Pdet with Co-stimulation
						Proximal Electrodes (#)	Latency (ms)	T (mA)	Distal Electrodes (#)	Latency (ms)	T (mA)
1	M	66	C5	Complete	24	3–5*	77	15	13–15*	–	–	N
2	M	28	T6	Incomplete	12	4–5*	91	7	12–13*	44	15	N
3	M	53	T12	Complete	33	3–4*	–	–	9–10*	74	20	Y
4	M	67	T2	Incomplete	12	1–2*	–	–	10–11*	–	–	N
5	M	63	T8	Complete	26	1–2*	68	45	8–9*	41	45	N
6	F	22	T1/T5	Incomplete	2	n/a	n/a	n/a	n/a	n/a	n/a	n/a
7	M	43	T7	Incomplete	24	3–4*	–	–	12–13*	–	–	Y
8	M	71	T2	Incomplete	8	n/a	n/a	n/a	n/a	n/a	n/a	n/a
9	M	39	C5–C6	Incomplete	12	3–4	73	15	8–9	42	25	N
10	F	65	T6–T10	Incomplete	9	1–2	–	–	3–4	–	–	Y
11	F	55	T8–T12/L3	Incomplete	3	1-R	80	20	2-R	85	27	Y
12	M	58	T12	Incomplete	4	3–4	48	10	7–8	58	12	N
13	M	45	C5–C6	Complete	5	n/a	n/a	n/a	n/a	n/a	n/a	n/a
14	M	36	T12	Incomplete	9	1–2	80	8	5–7	31	14	Y
15	F	42	T5	Incomplete	5	1–2	75	32	3–4	77	42	Y
16	M	57	C4–C5	Incomplete	22	1–2	83	13	7–8	–	–	Y
17	M	35	T9	Complete	7	2–3	–	–	8–9	–	–	N

Seventeen individuals with either complete or incomplete SCI above the sacral level participated in the study. The mean number of years since injury was 12.8 years. Stimulation-evoked reflex responses from either proximal or distal intraurethral stimulation were demonstrated in 10/17 subjects. Electrode location, reflex latency, and stimulation threshold amplitude necessary to evoke reflex response are listed for each location. Asterisk (*) indicates experiments that were conducted with a 15-electrode catheter and reflect different anatomical locations than the same electrode on a 12-electrode catheter. R indicates that an electrode patch on the leg was used as the stimulation return electrode, and this is not expected to have impacted the results. If the subject terminated study prior to completion of testing, n/a is indicated. Changes in bladder pressure (P_det) with intraurethral stimulation were observed in 7 subjects. Injury level and completeness of injury were reported by patients and have not been confirmed by MR imaging.

Subjects were instrumented with a 4 Fr catheter to adjust bladder volume and measure vesical pressure (P_ves), and a 9 Fr balloon rectal catheter to measure abdominal pressure (P_abd). Detrusor pressure (P_det) was calculated as P_det = P_ves − P_abd. Electromyographic (EMG) activity was measured with percutaneous stainless steel wires inserted into the perineum (external urethral sphincter, EUS) and external anal sphincter (EAS), amplified (1k – 5k), filtered (10 Hz – 10 kHz), and recorded. A custom 12 Fr Foley stimulating catheter made of silicone with cylindrical, stainless steel electrode contacts was inserted into the urethra such that the balloon lay in the bladder neck and the electrodes targeted locations in the proximal and distal urethral shown to evoke bladder contractions in humans with SCI [18]. Because the length of the urethra varies between male and female participants, multiple electrodes along the length of the catheters enabled the stimulation location to be customized for each participant.

The particular stimulating catheter electrode contacts used for each subject were selected that minimized the threshold (T) to evoke reflex EMG activity in either the EAS or EUS with 2 Hz stimulation in the proximal or distal urethra when the bladder was empty. A control cystometrogram was conducted at the beginning of each experiment to determine the volume threshold for distension-evoked reflex contractions (DECs) of the bladder. Battery powered electrical stimulators (Empi 300PV, St. Paul, MN) delivered trains of charge-balanced, asymmetric biphasic current pulses (pulse width = 0.2 ms, duration = 20 s). Blocks of sixteen randomized combinations of single-site and two-site co-stimulation at proximal and distal electrodes at different frequencies (2, 10, 20, and 40 Hz) were presented at amplitudes of 2T and 4T under isovolumetric conditions (80% of DEC threshold volume).

Mean P_det was calculated as the average bladder pressure during stimulation minus the average pressure during the 5 s preceding each trial of stimulation. Maximum P_det was the maximum pressure evoked during stimulation, minus the baseline average. For all subjects who had stimulation-evoked contractions, the mean and maximum P_det were compared in an ANOVA and post hoc paired comparisons using Dunn tests with Bonferroni correction. Normalized, rectified and integrated (RI) EMGs, after stimulation artifact subtraction, were compared in an ANOVA and post hoc paired comparisons with Bonferroni correction.

Results

Reflex responses evoked by intraurethral stimulation

The average electrode locations for proximal and distal stimulation were 3.9 ± 0.4 cm and 16.5 ± 1.4 cm from the bladder neck, respectively (Figure 1). Evoked reflex EMG activity confirmed neural activation and reflex responses were evoked in 10/17 subjects. In seven of those 10 subjects, reflex responses were evoked by stimulation in both the proximal and distal urethra, while in the three remaining subjects, reflex responses were evoked from either only the distal site (n=1) or only the proximal site (n=2).

Figure 1. Anatomy of the lower urinary tract and intraurethral stimulation locations.

Anatomy of the male lower urinary tract, including expected urethral innervation by the pudendal and pelvic nerves. The pudendal and pelvic nerves originate from the sacral level (S2–S4 in the human) and innervate the urethra and external genitalia, and bladder and bladder neck, respectively. The urethral meatus and distal portion of the urethra is innervated by a distal branch of the pudendal nerve, the dorsal genital nerve (DGN). The pudendal nerve also innervates the external urethral sphincter (EUS) and the external anal sphincter (EAS). The proximal urethra may be innervated by both the pelvic and pudendal nerves. The catheter was positioned with electrodes (2.75 mm width) targeting the proximal urethra (5 electrodes with 0.9 cm spacing or 6 electrodes with 0.5 cm spacing), and distal urethra (7 electrodes with 1.85 cm spacing or 9 electrodes with 1.4 cm spacing). Two different stimulating catheter designs were employed in this study due to changes in manufacturing that occurred during the study; however, both catheter designs provided the ability to customize location of stimulation for each participant. In female participants, only the proximal electrodes were used due to the shorter urethral length.

The latency of EMG responses evoked by proximal intraurethral stimulation (Figure 2) was 76 ± 3 ms, while the latency of responses evoked by distal stimulation was 48 ± 6 ms (p=0.0006, ANOVA, n=10). Four of 10 subjects exhibited both long latency responses to proximal intraurethral stimulation and short latency responses to distal stimulation (Table 1). In the two female subjects who had reflex responses to stimulation, only long latency responses were evoked with intraurethral stimulation of the proximal or distal locations in the urethra. There were no differences between reflex thresholds for proximal (13 ± 4 mA, n=11) or distal stimulation (12 ± 5 mA, n=6; p=0.934 ANOVA), or between thresholds to evoke short (20 ± 5 mA, n=11) or long latency (22 ± 4 mA, n=6; p=0.764 ANOVA) responses.

Figure 2. Reflex EMG responses to proximal and distal intraurethral stimulation.

A) Examples of reflex EMG activity evoked in the EUS by 2 Hz stimulation of the proximal or distal urethra. Light gray traces are individual responses to stimulation and black trace is the average. B) The latency of reflex response evoked by proximal stimulation was significantly longer than distal stimulation (*p=0.0006, ANOVA, n=10). C) Reflex latencies are plotted against electrode contact distance from the bladder neck, where “x” denotes a distal and “o” denotes proximal a electrode (Fig. 1). Electrodes positioned closer to the bladder neck evoked reflex responses with longer latencies than stimulation at electrodes farther from the bladder neck.

Effects of co-stimulation on bladder pressure and EMG activity

Intraurethral stimulation evoked increases in P_det in seven subjects. In ten subjects, several factors limited the ability to evaluate intraurethral stimulation, including absence of distension-evoked bladder contractions (n=4), stimulation amplitude limited by uncomfortable sensations (n=3), inability to insert the stimulation catheter (n=1), and voluntary subject withdrawal (n=2).

Figure 3A shows examples of bladder contractions evoked by single-site stimulation or two-site co-stimulation. Stimulation-evoked mean P_det was dependent on the stimulation frequency applied in the proximal and/or distal urethra (Figure 3C; p<0.0001, ANOVA, n=6). Post hoc comparisons with Bonferroni correction (p<0.0001) revealed combinations of proximal and distal intraurethral stimulation that produced significantly different effects on the mean P_det. For example, 10 Hz proximal stimulation produced excitation, while 10 Hz distal stimulation produced inhibition. Changing one of the frequencies in co-stimulation had a significant effect on mean P_det. For example, 40P-20D (40 Hz proximal, 20 Hz distal) was significantly different than 20D alone, 10P-20D, 40P-40D, and 20P-20D.

Figure 3. Bladder contractions evoked by intraurethral stimulation.

A) Co-stimulation of proximal and distal urethra produced changes in detrusor pressure that were dependent on stimulation frequency. Traces represent detrusor pressure in response to individual stimulation of either proximal or distal urethra or two-site co-stimulation in one subject. The first column of plots represent responses to only proximal stimulation at 2, 10, 20 and 40 Hz, while the first row of plots represents only distal stimulation. Each of the plots of co-stimulation was produced by stimulation at the frequencies shown by the row of proximal stimulation and column of distal stimulation. Bars indicate when stimulation was applied. B) Proportion of subjects who exhibited at least a 10 cmH₂O increase in maximum P_det with stimulation. Some frequency combinations were less effective than others. 10 Hz distal stimulation failed to evoke a robust contraction in any subject (0%), while 2P-2D and 40P-40D evoked robust contractions in 4/7 subjects (57%). C) Mean changes in detrusor pressure (P_det) evoked by stimulation, averaged across 6 subjects. There was a significant effect of stimulation frequency on mean P_det (p<0.001, ANOVA, n=6). Post hoc comparisons with Bonferroni correction revealed stimulation frequency combinations that were significantly different from each other (* p < 0.001).

To explore further the effectiveness of co-stimulation, we quantified the percent of subjects where stimulation produced maximum P_det greater than 10 cmH₂O for each frequency pair (Figure 3B). This revealed patterns of co-stimulation that were more effective (2P-2D and 40P-40D: robust contractions in 57% of subjects) or less effective (10D: robust contractions in 0% of subjects).

There was a significant difference in the normalized EMG activity measured across subjects (p<0.0001, ANOVA, n=6) and between periods with stimulation-on compared to stimulation-off (p<0.0001, ANOVA, n=6). Further, a significant interaction between normalized EMG and frequency of stimulation and stimulation-on periods (p=0.0057) indicated that the changes in RI EMG activity were dependent on the stimulation frequency. Although no frequencies of stimulation were significantly different from each other (post hoc comparisons with Bonferroni correction), higher proximal stimulation frequencies tended to produce greater normalized RI EMG activity.

Comment

This study found that while there were no significant differences in stimulation thresholds required to produce reflexes, the latency of reflex responses to stimulation in the proximal and distal urethra varied. The ~30 ms difference between the proximal and distal reflex latencies could not be caused by differences in neural conduction times from urethral stimulation sites, because a difference of 13 cm between electrode locations would only result in a latency difference of 4.5 ms (assuming 29 m/s conduction velocity [19]), which is much smaller than the observed difference.

Therefore, the different latencies of reflex responses evoked by proximal and distal stimulation are likely the result of activation of separate neural pathways. This is corroborated by a previous study that identified proximal stimulation reflex responses in the bulbocavernosus muscle at 70 ms, approximately double the latency of distal stimulation of the DGN (30–40 ms) [20]. Further, short latency responses (36 ms) were reported to stimulation of the glans penis in normal individuals and patients with neurogenic bladder from spinal cord disease [21]. In addition, there was no significant difference in reflex thresholds, indicating that large diameter fibers were likely activated at both locations. Thus, variations in reflex latencies are not due to differences in location along the same nerve, neural fiber activation, or pathological changes from SCI, but rather differences in the complement of sensory fibers activated and the reflex that was subsequently engaged. In the female subjects in whom reflex responses were evoked, an absence of short latency reflex responses suggests that pelvic and pudendal innervation of the urethra differs from the male.

In the cat, pudendal afferents innervate the distal urethra and both pelvic and pudendal nerves have been identified in the prostatic urethra [22,23]. Therefore, long latency reflex responses reported here may be mediated by proximal intraurethral stimulation of pelvic afferents rather than pudendal afferents. Distal stimulation-evoked EMG, which evoked reflex responses at short latencies [18], was conversely mediated by stimulation of pudendal afferents. In two subjects, stimulation at the most proximal electrode locations produced both a short latency reflex response and long latency response, suggesting that in some humans the proximal urethra may include both pudendal and pelvic afferent innervation.

Co-stimulation of the proximal and distal urethra

Intraurethral co-stimulation produced larger bladder contractions than stimulation of individual sites alone, similar to results in the cat where co-stimulation produced synergistic bladder activation and improvements in voiding efficiency compared to individual stimulation [14]. There was a significant effect of stimulation frequency on mean P_det evoked by co-stimulation, consistent with other work where the size of stimulation-evoked bladder contractions was dependent on stimulation frequency and location [16,24,17,25,26]. However, various combinations of frequencies evoked robust bladder contractions, and, while some patterns were more or less likely to evoke increases in bladder pressure across the population (Figure 3), there was no optimal combination of frequencies that was effective in every subject. The changes in EMG activity during stimulation varied with frequency combinations. A clear correlation between increases in EMG with co-stimulation and the reflex that was engaged (response latency) was not observed. A previous study of intraurethral stimulation showed that proximal stimulation produced more EMG activity than distal stimulation [18], and high frequency pudendal afferent stimulation in the cat did not increase EMG activity [14,26], whereas pelvic nerve stimulation produced concomitant activation of EUS and bladder in dogs [27]. Thus, the prevalence of co-stimulation patterns that increased EMG activity may be due to the activation of pelvic afferents by stimulation in the proximal urethra.

Conclusions

This work demonstrates the importance of stimulation frequency and stimulation location, or reflex pathway activated, on the effects of electrical stimulation on the bladder and muscles of the pelvic floor. Stimulation frequency and location affected the size of bladder contractions evoked by stimulation, similar to the results from the cat [24].

Electrical stimulation of proximal and distal urethral afferents produced bladder activation through two separate pathways and selective co-stimulation of these pathways can produce enhanced bladder activation, as was demonstrated previously in the cat [14]. In the human, unlike the cat where these pathways were both mediated by pudendal afferents, it appears that these two pathways are mediated by pelvic and pudendal afferents, with different latencies responses to intraurethral stimulation. Future work needed includes additional animal studies with controlled SCI lesions and nerve transections to confirm the origin of reflex contributions, as well as additional clinical testing of these nerve targets in humans with SCI.

Importantly, these results support that the reflex circuits mediating the effects of afferent stimulation are present in the lumbosacral spinal cord and preserved after suprasacral SCI. Furthermore, for other applications of stimulation of afferent pathways in the pelvic region, reflex latency may be used as an indicator of which neural pathways was activated. Continued development of afferent stimulation for restoration of bladder function should include co-stimulation to improve bladder control when individual site stimulation is ineffective.