Transcutaneous neuromodulation versus oxybutynin for neurogenic detrusor overactivity in persons with spinal cord injury: A randomized, investigator blinded, parallel group, non-inferiority controlled trial

Abstract

Study design

This study is a randomized, investigator-blinded, controlled trial with a non-inferiority design.

Objective

To investigate the effectiveness of neuromodulation by transcutaneous electrical stimulation of the somatic afferent nerves of the foot in neurogenic detrusor overactivity (NDO) in persons with spinal cord injury (SCI) and compare its effectiveness with oral oxybutynin.

Setting

The study was conducted in a rehabilitation in-patient ward of a tertiary care hospital.

Methods

Twenty-nine persons with SCI with NDO, either sex, aged 18 years and above were randomized into two groups, one group receiving oral oxybutynin (5 mg thrice a day for two weeks) and the other transcutaneous electrical stimulation (5 Hz, 200 µs pulse, biphasic, amplitude up to 60 mA, 30 min/day for two weeks). Bladder capacity was evaluated by clinical bladder evaluation (i.e. bladder capacity measured by adding leak volume, voiding volume if any, and post-void residue using a catheter) and cystometric bladder capacity by one-channel cystometry. Maximum cystometric pressure was evaluated by one-channel water cystometry. Data were analyzed with Fisher's Exact, t-test, and Wilcoxon rank sum tests.

Results

Bladder capacity improved significantly in the oxybutynin and neuromodulation groups as measured by one-channel water cystometry (136 ml vs. 120.57 ml) and clinical evaluation (138.93 ml vs. 112 ml). The increase in the neuromodulation group achieved the pre-decided non-inferiority margin of 30 ml over the oxybutynin group when measured by one-channel water cystometry but not by clinical evaluation. Maximum cystometric pressure did not significantly improve in either group when compared with the baseline.

Conclusion

Transcutaneous neuromodulation and oxybutynin effectively increased bladder capacity in persons with SCI with NDO. Neuromodulation by once-a-day transcutaneous electrical stimulation was non-inferior to thrice-a-day oxybutynin when evaluated by one-channel water cystometry.

Trial registration: Clinical Trials Registry India identifier: CTRI/2018/05/013735.

Introduction

Neurogenic bladder is a source of considerable morbidity and a hindrance to community integration in persons with spinal cord injury (SCI) (1). The current mainstay of neurogenic detrusor overactivity (NDO) management, i.e. antimuscarinics and clean intermittent self-catheterization (CISC), may not be adequately efficacious besides having adverse effects (2–4). Further research regarding neurogenic bladder is one of the topmost felt needs (5, 6). Several studies have shown promising findings on the role of sacral neuromodulation in NDO (7–9). Neuromodulation has been achieved by stimulating the sacral nerve roots using an implanted electrode, percutaneous needle stimulation of nerves, and transcutaneous stimulation of nerves (7). Despite encouraging evidence and even FDA approval (10–13), these modalities have yet to gain popularity as they are invasive (10–13) or require frequent clinic visits (14).

Studies on animal models (15, 16) and healthy humans (11) demonstrated that transcutaneous electrical stimulation of somatic afferent nerves in the foot using skin surface electrodes can inhibit reflex micturition and significantly increase bladder capacity. In a single-arm study, foot stimulation could significantly increase bladder capacity in NDO patients with SCI (17). Foot stimulation is non-invasive and can be self-administered at home daily. The objective of this study was to investigate the effectiveness of neuromodulation by transcutaneous electrical foot stimulation in NDO in persons with SCI and compare its effectiveness with oral oxybutynin in increasing bladder capacity (primary objective) measured by clinical bladder evaluation (i.e. bladder capacity measured by adding leak volume, voiding volume if any and post-void residue using a catheter) and increasing cystometric bladder capacity and decreasing maximum detrusor pressure measured by one-channel water cystometry (secondary objectives).

Methodology

A prospective parallel group investigator blinded randomized controlled trial with a non-inferiority type of design was conducted at a rehabilitation in-patient ward in a tertiary care hospital from May 2018 to November 2019. The study is registered in the clinical trial registry of India (CTRI/2018/05/013735) and was commenced after approval from the institutional review board and the institutional ethical committee. It was conducted as per standard ethical principles for medical research involving human subjects.

Persons with SCI undergoing in-patient rehabilitation during the study period, fulfilling the inclusion and exclusion criteria, were recruited after obtaining written informed consent. Persons with SCI (traumatic and non-traumatic) aged 18 years or more, either sex, and with a diagnosis of NDO with urinary incontinence, were included. Exclusion criteria were pregnancy, urinary tract infection as defined by the American paraplegia society (18), urinary calculus, tumor, cardiac pacemaker, epilepsy, local pathology of the foot like ulcer or infection, and history of botulinum toxin injection to detrusor in the past one year. Detailed clinical examination was recorded, including baseline demographic data per international standards for neurological classification of SCI and international SCI data sets. Neurogenic bladder evaluation, urinalysis, urine culture, renal ultrasound including ureter and bladder, and one-channel water cystometry were done.

Madersbacher et al. (19) reported that mean bladder capacity increased from 187.8 ml to 350.9 ml with oxybutynin, and Chen et al. (17) reported that mean bladder capacity increased from 279.4 ± 11.7 ml to 361 ± 18 ml with neuromodulation. With an absolute difference of 10 ± 18 ml and a two-sided α risk of 0.05, the trial with two groups of 15 patients each had a 90% power with a pre-specified non-inferiority margin of 30 ml. Using a computer-based system, the participants were block randomized to two arms in a 1:1 ratio. The randomization sequence was concealed using sealed envelopes.

Arm A received oral oxybutynin (5 mg thrice a day) for two weeks, while arm B received transcutaneous electrical foot stimulation for two weeks. Transcutaneous electrical foot stimulation was administered with a TENS apparatus (T250, Johari Digital, India) with two self-adhesive electrodes (Fig. 1) placed on the sole. One electrode was placed at anterior one third of the sole, while the other electrode was placed at the posterior one third of the sole. Since, a biphasic pattern of stimulation was adopted, interchange of position of cathode and anode did not affect the efficacy of the stimulation. The electrical stimulation was set at a frequency of 5 Hz and pulse duration of 200 microseconds in a continuous biphasic pattern with intensity to cause toe twitch (up to a maximum of 60 mA). Stimulation was given daily but on alternate feet for 30 min for two weeks.

Figure 1.

Set-up for transcutaneous electrical stimulation of the somatic afferent nerve of the foot.

Bladder capacity was measured by clinical bladder evaluation (primary objective) and by one-channel water cystometry (secondary objective). Clinical bladder evaluation is a bedside procedure to evaluate bladder capacity and a direct measure of bladder capacity. The bladder was first emptied by urinary catheterization. A pre-weighed gauze pad was applied as a sanitary pad. Participants were instructed to consume water as usual. The participant or caregiver inspected the applied gauze pad frequently (every five to 10 min). The evaluation was continued till the first significant leak (at least 5 ml). Any prior leak of a few drops was ignored. Leak volume (V1) was calculated by weighing the gauze pad (1 g = 1 ml) with a weighing machine that had the ability to detect a difference of one gram. Participants were then instructed to attempt to void, and self-voiding volume (V2) (if any) was noted. Suprapubic or perineal stimulation was done to initiate reflex voiding, and reflex voiding volume (if any) was noted (V3). Leak volume (V1) and all two voiding volumes (V2 and V3) were checked quickly in the shortest possible time to minimize the effect of further urine collection in the urinary bladder. Finally, post-void residue (PVR) was measured after catheterization. Total bladder capacity was calculated using the formula V1 + V2 + V3 + PVR. Measurement of PVR with catheterization is considered the gold standard method of measuring PVR (20). Thus, carefully measured leak volume (V1) and voiding volumes (V2 and V3) yield an accurate and reliable measurement of bladder capacity. Water intake by the participants during the time of evaluation was kept uniform. They were allowed to drink water sip by sip as per need. Any large volume of water intake was not allowed.

Maximum detrusor pressure and cystometric bladder capacity were measured by one-channel water cystometry following the Wyndaele protocol (21). One-channel water cystometry is an alternative to multi-channel cystometry. It has been shown to have a 93.33% positive predictive value and 100% negative predictive value for neurogenic bladder diagnosis in persons with SCI (22). All measurements were recorded with the participants resting in the supine position. The one-channel water cystometry system was set up as shown in Fig. 2, and zero of the pressure column was positioned at the probable bladder level. A collecting device (urinal bed pan) was used to collect voiding or leakage. Bladder filling was 20–50 ml/min, and pressures were noted every 30 s. A low gauge (8F) tube was used for the filling of the bladder. This thin tube allowed peri-catheter leak when leak point pressure is reached. The filling was stopped when leakage occurred, or at a pressure rise over 50–60 cm water, or at 500 ml when no pressure rise was detected. When there was no significant rise in pressure even after filling 500 ml, it was assumed that the bladder had detrusor hyporeflexia. Any abrupt pressure rise above 50–60 cm of water was noted as NDO. Bladder pressure fluctuations due to abdominal pressure were determined and excluded by close patient observation (21). All evaluations were done between 10 AM and 12 noon. Oxybutynin was given a thrice daily dose starting at 8 AM. The time lapse between the oxybutynin dose and evaluation was two to four hours. Electrical stimulation of the foot was given once daily in the evening (between 6 and 8 PM). The time lapse between stimulation of the foot and evaluation was around 12–14 h. Although this difference may not be ideal, we attempted to keep the oxybutynin dosing schedule according to usual practice and the schedule for stimulation of the foot according to the convenience of the participants. This design was adopted to provide additional insight regarding the residual effect of neuromodulation throughout the day.

Figure 2.

Schematic set-up for one-channel water cystometry (created with BioRender).

A statistician performed randomization. Recruitment and allocation were executed by a consultant physiatrist with 16 years of experience. Evaluations were done by a resident physiatrist who was blinded to group allocation. The resident was guided by a senior consultant with 39 years of experience in case of any difficulty in evaluation. A consultant physiatrist with ten years of experience performed data compilation and analysis with the help of a statistician. Participants could not be blinded.

Data were tested for normal distribution and summarized as number (%) and mean ± SD as appropriate. Demographic characteristics were compared using a two-sample t-test with equal variances, Chi² exact test, and a two-sample Wilcoxon rank-sum test. Outcome measures were analyzed within and between the groups using paired t-tests and two-sample t-tests with equal variances, respectively. The primary outcome, change in bladder capacity by clinical evaluation, has been reported as a difference in volume between the groups (95% confidence interval). Neuromodulation was hypothesized to be inferior to oxybutynin if the mean difference between the groups by clinical bladder evaluation at two weeks exceeded 30 ml. Data were analyzed with Stata 14.0 software.

Results

Twenty-nine patients were enrolled in the study (Fig. 3). The mean duration since injury in the oxybutynin group was 18.8 ± 16.12 months, and in the neuromodulation group was 44.42 ± 39.04 months. Demographic data are presented in Table 1.

Figure 3.

CONSORT 2010 flow diagram.

Table 1.

 Demographic data (N = 29).

Characteristics	Oxybutynin Group (n = 15)	Neuromodulation Group (n = 14)	p value
Age in years (Mean ± SD)	28.2 ± 8.18	31.85 ± 9.41	0.27321
Sex [N (%)]
Male	11 (73.33)	10 (71.43)	0.9992
Female	4 (26.67)	4 (28.57)
AIS [N (%)]
A	6 (40)	8 (57.14)	0.6402
B	2 (13.33)	2 (14.29)
C	4 (26.67)	1 (7.14)
Non-traumatic	3 (20)	3 (21.43)
Neurological level [N (%)]
Cervical	1 (6.67)	1 (7.14)	0.8412
D1–D6	2 (13.33)	4 (28.57)
D7–D12	11 (73.33)	9 (64.28)
Lumbar	1 (6.67)	0

(AIS: American Spinal Injury Association Impairment Scale; D: Dorsal/thoracic vertebra)

¹Two-sample t test with equal variances, ²Chi² exact test.

Bladder capacity, measured by clinical evaluation, significantly increased at two weeks compared to baseline in both groups. However, there was no significant difference between the two groups at baseline or two weeks (Table 2). The mean difference between the two groups at two weeks was 30.42 ml, failing to meet the pre-decided non-inferiority margin of 30 ml by a minimal margin of 0.42 ml.

Table 2.

 Bladder capacity by clinical bladder evaluation (N = 29).

Group	Bladder Capacity in ml (Mean ± SD)		p value within the groups
	Baseline	Two weeks
Oxybutynin Group (n = 15)	148.20 ± 50.60	287.13 ± 51.43	<0.0011
Neuromodulation Group (n = 14)	144.00 ± 39.96	256.71 ± 53.85	<0.0011
p value between the groups	0.80692	0.13112

¹Paired t test, ²Two-sample t test with equal variances.

Cystometric bladder capacity, measured by one-channel water cystometry, significantly increased at two weeks compared to baseline in both groups. There was no significant difference between the two groups at baseline or two weeks, and the non-inferiority margin of 30 ml of the neuromodulation group over the oxybutynin group was achieved (Table 3).

Table 3.

 Cystometric bladder capacity by one-channel water cystometry (N = 29).

Group	Bladder Capacity in ml (Mean ± SD)		p value within the groups
	Baseline	Two weeks
Oxybutynin Group (n = 15)	143.33 ± 51.63	279.33 ± 63.29	<0.0011
Neuromodulation Group (n = 14)	140.14 ± 43.31	260.71 ± 50.60	<0.0011
p value between the groups	0.84702	0.39152

¹Paired t test, ²Two-sample t test with equal variances.

Maximum detrusor pressure, measured by one-channel water cystometry, did not change significantly at two weeks compared to baseline in both groups. Also, there was no significant difference between the two groups either at baseline or at two weeks (Table 4).

Table 4.

 Maximum detrusor pressure by one channel cystometry (N = 29).

Group	Maximum detrusor pressure in cm of water (Mean ± SD)		p value within the groups
	Baseline	Two weeks
Oxybutynin Group (n = 15)	67.86 ± 10.67	67.73 ± 11.11	0.85881
Neuromodulation Group (n = 14)	62.42 ± 9.83	62.71 ± 10.83	0.76461
p value between the groups	0.16602	0.22922

¹Paired t test, ²Two-sample t test with equal variances.

Discussion

This study investigated the effectiveness of neuromodulation by transcutaneous electrical stimulation in NDO in persons with SCI and compared its effectiveness with oxybutynin with the primary objective of increase in bladder capacity assessed by clinical evaluation, in a non-inferiority design, with a predefined non-inferiority margin of 30 ml.

In a single-arm study by Chen G et al. in persons with SCI with overactive neurogenic bladder, tibial nerve stimulation increased the mean bladder capacity to 361.1 ± 18.1 ml, much higher than our study (17). However, unlike our study, Chen G et al. measured bladder capacity by catheterization immediately after stimulation. In our study, the transcutaneous electrical stimulations of the somatic afferent nerves were administered in the evening, and clinical bladder evaluations, at baseline and two weeks, were done the subsequent morning. This methodology provided additional insight regarding the carry-over effect. In a study on patients with post-operative bladder spasms, transcutaneous electrical stimulations of the somatic afferent nerves relieved the discomfort as measured by a visual analog scale (23).

Oxybutynin is considered the first choice for NDO symptoms, and literature indicates that it can significantly increase bladder capacity by relaxing the detrusor muscle (24–26). While oxybutynin is efficacious, it has several adverse effects that may complicate the overall management of a person with spinal cord injury (24). Dry mouth as one potential adverse effect has been focussed as the prime tolerability endpoint (27). Management of neurogenic bladder usually involves fluid restriction, while dry mouth may make it more difficult for the person with spinal cord injury to cope with restricted fluid intake. Constipation caused by antimuscarinics, including oxybutynin, makes the management of the neurogenic bowel difficult (28). Dizziness and somnolence may pose a significant challenge in gait and balance training (29).

No study comparing antimuscarinics with transcutaneous electrical stimulation of the foot's somatic afferent nerves was available to compare our findings. Regarding the mechanism of action, the closest modality resembling our study is posterior tibial nerve stimulation. Both techniques target the S2 segment of the spinal cord to achieve sacral neuromodulation. Two studies comparing posterior tibial nerve stimulation with solifenacin in women with non-SCI overactive bladder found both effective. Moreover, posterior tibial nerve stimulation was superior to solifenacin (30, 31).

In a study among persons with SCI with NDO, the effectiveness of posterior tibial nerve stimulation was equivalent to solifenacin (14). In our study, the dose of oral oxybutynin was five mg thrice a day, as per the recommended practice facilitating its action throughout the day. The dosing of electrical stimulation (frequency per day) for sacral neuromodulation has, however, never been more than once daily as they were either invasive or required multiple visits to the hospital. Literature suggests that the peak effect of sacral neuromodulation, post-procedure, lasts up to two hours with some residual effect after that (32–34). Increasing the frequency of application to more than once a day may have a better outcome. As proposed in our study, the stimulation technique is easy to self-administer at home or work and of a reasonable duration. So, it is feasible to use it multiple times a day. Moreover, neuromodulation acts on the spinal cord. In contrast, antimuscarinics act at the bladder and thus make a potential candidate for combination to achieve synergistic or additive effects with different mechanisms of action.

Maximum detrusor pressure, measured by one-channel water cystometry, did not change significantly in both the intervention groups at two weeks (Table 4). This result contradicts the existing literature, and studies have reported that antimuscarinics reduce maximum detrusor pressure (35, 36). A study from Switzerland reported that the mean change in detrusor pressure after 13.1 months of oral solifenacin therapy was only seven cm of water (36). In our study, the intervention period was only for two weeks, which might have resulted in a minor change that might not have been detected in one-channel water cystometry (21).

Transcutaneous electrical stimulation to the insensate or hypo-sensate area may result in burn injury. It may also result in autonomic dysreflexia in susceptible individuals. Transcutaneous neuromodulation may also require increased intensity of current having to travel through the skin from a TENS external pad. However, the maximum amplitude used in this study was 60 mA, and within that limit, no adverse effects, as discussed above, were observed. Patients and family members were able to apply electrodes themselves and switch on the TENS machine. Thus, transcutaneous neuromodulation in the form of electrical stimulation of the foot may find good acceptability among the persons with NDO, as this is non-invasive and can be self-administered at home or workplace without help of any health care worker. Additionally, the use of a physical modality may lower the medication burden on persons with spinal cord injuries who tend to take many medicines already.

Our study had some limitations. The duration of the study was only two weeks. A more extended duration study will be required to determine the long-term efficacy. Also, multi-channel cystometry is the gold standard for evaluating neurogenic bladder and can provide objective and accurate results. It was substituted with the simple one-channel water cystometry in our study. The mean duration since injury was different between the two groups. This resulted due to a few outliers in the stimulation group. They were not excluded from the study as other baseline characteristics were comparable even after including those outliers. However, this difference may have reduced the efficacy of the intervention in the stimulation group. This is because a more extended duration since injury may result in more fibrosis in bladder musculature, making it less responsive to any therapy. The beneficial effect of applying stimulation two or three times a day as against once per day protocol needs further investigation. The synergistic or additive effect of combining stimulation with antimuscarinics can be explored.

Conclusion

To conclude, transcutaneous neuromodulation and oxybutynin effectively increased bladder capacity in persons with SCI with NDO. Neuromodulation by once-a-day transcutaneous electrical stimulation was non-inferior to thrice-a-day oxybutynin when evaluated by one-channel water cystometry.

Disclaimer statements

Author contribution statement SS and SV conceived the idea. SV recruited and allocated subjects. SS carried out all the measurements. US guided SS in case of any difficulty. ARC compiled and analyzed the data. SS and SV prepared the manuscript with inputs from US and ARC. All the authors discussed the results and commented on the manuscript.

Conflicts of interest None of the authors declared any conflict of interest.

Data availability The datasets generated and analyzed during the current study are available from the corresponding author upon reasonable request.

Ethical approval We certify that all applicable institutional and governmental regulations concerning the ethical use of human volunteers were followed during this research. Approval was taken from Institute Ethics Committee for Post Graduate Research, All India Institute of Medical Sciences, New Delhi, India. (Ref. No. IECPG-105/21.03.2018).

Financial support No financial support was received for work reported in this paper.