Introduction

Stroke is the second leading cause of death and disability worldwide1,2,3,4. Moreover, stroke patients are susceptible to many complications. Urinary incontinence can affect 40–60% of people admitted to the hospital after a stroke5. The predominant symptoms are urinary frequency, urgency, and urge incontinence6,7,8. Detrusor overactivity (DO) is the most common urodynamic finding in patients with poststroke urinary incontinence (PSI)9,10. It not only tends to cause negative emotions such as shame, helplessness, anxiety, and even depression but also increases the burden on the patient’s family5. On the other hand, long-term urinary incontinence is more likely to lead to complications such as pressure sores11. It is one of the poor predictors of survival and prognosis after stroke12. Management options for urinary incontinence include bladder retraining, prompt toileting, fluid/diet management13, physiotherapy, drug therapy, and botulinum toxin injections in the bladder14. However, behavioral interventions are influenced by cognitive and physical limitations; there is a lack of evidence on whether bladder relaxant intervention for PSI patients would also benefit5; botulinum toxin injection causes high incidences of difficult urination and acute urinary retention15; and transcutaneous electrical nerve stimulation is currently known to fail at a daytime frequency in PSI patients9,16. New approaches are needed to consider the bladder symptoms and challenges faced by patients with PSI.

The brain via a long-loop spinobulbospinal reflex pathway controls bladder activities. These pathways include the midbrain periaqueductal gray (PAG), and the pontine micturition center (PMC), and these nerves, in turn, are controlled by supratentorial neural networks17,18. When patients suffer PSI, they have a loss of connectivity between the brain areas of autonomic and voluntary control, which results in a distinct pattern of brain activity19 and an overactive detrusor (OD)20. Recent studies in animal models have suggested that these M1 layer 5 (L5) neurons travel directly to the pontine micturition center (PMC) to drive urination, which is highly correlated with the onset of urination21. Therefore, by regulating the cortex of micturition, a potential brain intervention site (M1 area) is provided for urinary dysfunction, e.g. PSI.

Previous studies have demonstrated that applying LF-rTMS to the M1 could inhibit detrusor overactivity in Parkinson's patients22 and high-frequency (HF) rTMS applied to the M1 has been reported to ameliorate detrusor contraction23,24 in patients with SCI and MS. Moreover, rTMS is increasingly widely used to induce neuroplasticity and promote the recovery of brain function in stroke patients. We are hence interested in whether LF rTMS on M1 could effectively alleviate the symptoms of poststroke urine incontinence.

In the current trial (RCT), LF rTMS (1 Hz) was applied to the contralesional M1 for bladder function recovery in patients with PSI. We pursued two specific aims in this study: (1) to assess the benefits of LF rTMS on key symptoms of PSI, as assessed via urodynamic examination and questionnaires related to urine incontinence, and (2) to evaluate the effects of LF rTMS on the function of pelvic floor muscles via surface electromyography.

Results

Patient characteristics

As summarized in Fig. 1, this study enrolled 178 participants, 100 of whom met the inclusion criteria and agreed to participate in the trial. Eighty-six patients completed the 4 week intervention (44 in the rTMS group and 42 in the sham-rTMS group), and they participated in the before and after assessments. Some participants were excluded or dropped out, and the reasons are described in (Fig. 1). Moreover, neither the clinical features nor the demographic variables differed between the two groups (all P > 0.05, Table 1).

Figure 1
figure 1

Flow diagram of the selection process in the study.

Full size image
Table 1 Baseline characteristics.
Full size table

Primary outcomes

As shown in Table 2, the MCC improved in both groups following the 4-week intervention (from 258.13 ± 77.82 mL to 381.14 ± 72.05 mL in the rTMS group vs. 259.49 ± 75.39 mL to 307.34 ± 97.25 mL in the sham-rTMS group) (both within groups, P < 0.05). Moreover, the MCC in the rTMS group (381.14 ± 72.05 mL) was greater than that in the sham-rTMS group (307.34 ± 97.25 mL) after the intervention (between groups, P = 0.000). Furthermore, the volume of MCC increased between the two groups. Specifically, the volume increased by 123.02 ± 98.19 ml in the rTMS group and by 47.85 ± 127.74 ml in the sham-rTMS group. Similarly, the residual urine volume decreased in both groups compared with that at baseline(both within groups, P < 0.05). After the intervention, the residual urine volume in the rTMS group was 70.76 ± 29.55 mL, which was smaller than that in the sham-rTMS group (125.26 ± 38.81 mL). Moreover, the decrease in residual urine volume was greater in the rTMS group (80.40 ± 27.55 mL) than in the sham-rTMS group (20.62 ± 40.80 mL) (P = 0.000). Pdet.max refers to the detrusor pressure at maximum bladder storage in our study. After the intervention, the Pdet.max decreased from 53.29 ± 18.23 cmH2O to 41.66 ± 11.79 cmH2O in the rTMS group and from 51.97 ± 19.86 cmH2O to 48.34 ± 15.81 cmH2O in the sham group. However, significant differences only appeared in patients in the rTMS group (within-group effect P = 0.001), and the degree of improvement was 11.63 ± 22.91 cmH2O. In addition, the decrease in the rTMS group after the intervention was significantly lower than that in the sham-rTMS group (P = 0.008).

Table 2 The urodynamics data before and after intervention in the rTMS and sham-rTMS groups.
Full size table

Secondary outcomes

After the intervention, the OABSS total score in the rTMS group was significantly lower than that in the sham group, at 6.12 ± 3.17 and 8.86 ± 2.86 respectively (P = 0.000). Similarly, improvements were also shown in the OABSS items, such as frequency, urgency, and UUI. All these parameters were significantly lower in the rTMS group than in the sham group (P < 0.05) (Table 3). Patients in both groups exhibited significant improvements in quality of life as assessed by the ICIQ-UI SF after 4 weeks of intervention. Moreover, there were significant improvements in the rTMS group compared with the sham-rTMS group after the intervention (P = 0.000) (Table 3). The electromyography of the pelvic floor included resting and contraction values. However, neither resting nor contraction improved after 4 weeks of intervention (Table 4).

Table 3 The questionnaires (OABSS and ICIQ-UI SF) before and after intervention in the rTMS and sham-rTMS groups.
Full size table
Table 4 The pelvic floor surface electromyography (resting values and systolic values) before and after intervention in the rTMS and sham-rTMS groups.
Full size table

Adverse events

Notably, no serious adverse events were reported during or after the interventions in either group. Three patients experienced mild headaches and dizziness during treatment, which were quickly alleviated.

Discussion

The main finding of our study was that 4 weeks of contralesional LF rTMS (1 Hz) intervention effectively improved the maximum cystometric capacity (MCC), maximum detrusor pressure (Pdet.max), residual urine output, overactive bladder score (OABSS) and ICIQ-UI SF compared with those of patients in the sham-rTMS group. However, our data did not reveal any changes in pelvic floor muscle EMG signals in patients with PSI. No differences were observed between the rTMS group and the sham group before or after the intervention. Collectively, our study indicated that LF-rTMS positively affects several aspects of end-filling Pdet, OABSS, and ICIQ-UI SF scores in patients with poststroke urinary incontinence.

Urinary incontinence occurs in 40–60% of stroke survivors, and detrusor overactivity is the most common type of urinary incontinence assessed by urodynamic studies4,7,25. An increase in maximal bladder capacity reduces the frequency of urination and the number of incontinences4. Therefore, the MCC is recognized as one of the mian parameters for diagnosing patients with poststroke incontinence4,26,27. Studies on spinal cord injury showed that an elevated detrusor pressure above 40 cm H2O may lead to upper urinary tract impairment28,29. However, there is a lack of studies to show upper urinary tract dilatation and hydronephrosis in poststroke urinary incontinence. In addition, the decrease in residual urine volume is an interesting phenomenon. It is affected not only by detrusor pressure but also by urethral sphincter contraction or relaxation. Meanwhile, with the progression of the disease, changes in the patient's urination position and improvements in cognitive ability reduce the residual urine volume. Overall, PSI patients experienced significant improvement in bladder function after 4 weeks of LF-rTMS intervention.

There is emerging evidence that rTMS may have therapeutic applications in neurological disease30,31,32. Recently, some studies have used HL (≥ 5 Hz) rTMS for urine retention in multiple sclerosis (MS) patients23,24, which revealed a significant decrease in bladder capacity and improvement in bladder contractility in the rTMS group. One study 22 used LF (≤ 1 Hz) rTMS on lower urinary tract (LUT) dysfunction characterized for 2 weeks in patients with Parkinson's disease (PD), which increased bladder capacity and the first sensation of the filling phase. Their results showed that LF rTMS is a non-invasive, effective treatment for PD patients with urinary disturbances that increase their bladder capacity and the first sensation of the filling phase.

The OABSS is a symptom assessment questionnaire that assesses symptoms such as frequency, nocturia, urgency, and urgency urinary incontinence (UUI). This questionnaire allows for an understanding of the patient’s perceived changes in detrusor function. The patient's bladder filling and desire for micturition come from sensory (afferent) feedback from the bladder and are effectively controlled by the brain18. PSI patients treated with LF-rTMS experienced better control of the detrusor muscle not only because of a significant reduction in the frequency of micturition but also because of an improvement in the incidence of urinary urgency and urge incontinence. The ICIQ-UI SF is a self-report questionnaire that assesses the frequency and degree of urinary incontinence, the impact of urinary incontinence on health-related quality of life, and the perceived cause of urinary incontinence and is a Grade A measure used by the International Continence Society33. Poststroke urinary incontinence can significantly worsen health-related quality of life34 and bladder function. The improvement in bladder function led to a reduction in the frequency of urinary incontinence and a significant improvement in the quality of life of the patient. Improvements in urinary incontinence associated with rTMS treatment may significantly improve questionnaire scores and lead to better ICIQ-UI SF scores.

In previous studies of rTMS for the treatment of lower urinary tract dysfunction, the mechanism may be that rTMS is involved in strengthening the pelvic floor muscles to improve symptoms35,36; therefore we were curious whether LF rTMS can have a positive effect on the pelvic floor muscles in patients with PSI, and previous data suggest that LF rTMS can improve PSI symptoms. Electromyography was used to document the activity of muscles under different conditions in patients with PSIs37,38. The values for contraction sEMG activity describe the activity of the pelvic floor muscles during contraction, whereas the remaining values describe the relaxation state39. However, our data did not support this speculation as the resting and contraction values of pelvic floor surface electromyography did not change after rTMS intervention. The reason might be that the activity of muscles continually changes in response to multiple factors, e.g., emotions and cognitive states37,40.

It was reported that patients who are elderly or have severe dysfunction, the symptoms of urinary incontinence are more severe11. Meanwhile, we found that urinary incontinence improved less when stroke patients were older, had more severe functional impairment, or later in the intervention (data not shown). Therefore, for these patients, other treatments need to be explored.

Limitations should be addressed. First, all participants included in this study were enrolled in a single center. The emergence of the COVID-19 pandemic resulted in a smaller sample size than expected and prevented follow-up on treatment effects. Second, only rTMS has been studied as a single treatment, and combined pelvic floor training can be considered in the future. Third, owing to the excessive effects of rTMS, it is difficult to explain the mechanism of PSI improvement, and fMRI and other studies can be performed in the future to explore the potential mechanism involved.

In summary, our study indicated that rTMS positively affects several aspects of the mean frequency, MCC, and OABSS score. Future clinical experiments aimed at determining the effectiveness of rTMS for PSI patients should be based on the timing and degree of damage, which could result in more effective rTMS programs.

Materials and methods

Study design

This study was designed according to the CONSORT 2010 statement41. This study was a double-blind, parallel-arm randomized controlled trial (RCT) (rTMS and sham groups). The study was approved by the Institutional Review Board and Hospital Research Ethics Committee of Chongqing Medical University (approval NO. 2020066). The study was conducted in accordance with the Declaration of Helsinki. This study was approved by the Chinese Clinical Experiment Registry on 26/01/2021 (identifier: ChiCTR2100042688, trial registry name: Effectiveness of Repetitive Transcranial Magnetic Stimulation in Treating Urinary Disorders after Stroke). All patients signed a written informed consent form before the experiment (Supplementary Material). We published the study protocol, hypotheses, and methods10. Here, we briefly summarize the key elements of this study.

Participants and setting

The stroke patients who participated in this study suffered from urinary incontinence. The patients were rehabilitated at the Second Affiliated Hospital of Chongqing Medical University. The first patient entered the study on January 26, 2021, enrollment was completed on December 15, 2022, and the last patient was assessed on January 15, 2023. Figure 1 shows the process of patient inclusion and participation in this experiment. The inclusion criteria for PSI patients were as follows: (a) first ischemic stroke; (b) urinary incontinence after stroke; (c) onset of stroke within 0.5–3 months of (acute or subacute phase of stroke) enrollment; (d) age over 30 years of any sex; (e) no medication used to treat urinary incontinence; and (f) provision of written informed consent. The exclusion criteria10 were as follows: (a) heamorrhagic stroke; (b) any urinary incontinence dysfunction before stroke; (c) absolute/relative contraindications for rTMS treatment (e.g., metallic implant, cardiac pace, pregnancy, epilepsy, head trauma, history of cranial operation); (d) insufficiency of the heart, lungs, and kidneys, uncontrolled known systemic disease, and/or an unstable medical condition; (e) bilateral brain lesions; and (f) urinary tract infections.

Allocation and blindness

According to our published protocol, the original target sample size was 140 patients10. Owing to the COVID-19 pandemic and funding considerations and because an increase in bladder capacity of more than 50 mL was considered a treatment effect42, the final sample size was 100 patients (Fig. 1). In accordance with the case-control principle, the participants were randomly allocated to the rTMS group or the sham-rTMS group at a 1:1 ratio according to computer-based allocation, and the inclusion criteria included age, sex, duration of stroke onset, stroke hemiplegia site, and hr. Each group comprised 50 patients. Both the statistical analysts and the outcome testers were blinded to the patients’ information. Two professional therapists intervened in all patients but were not involved in the evaluation of outcomes. For evaluation and statistical analysis, uncorrelated codes such as 1 and 2 were used to represent the groups.

Procedures

All randomized subjects underwent 20 training sessions (5/week × 4 weeks). Except for the use of rTMS for those patients who were assigned to the rTMS group, the interventions were similar between the two groups. The same rehabilitation therapists presided over all intervention sessions.

According to the International Federation of Clinical Neurophysiology, single-pulse TMS was used to evaluate motor cortical excitability in the hemisphere10,43. rTMS was delivered across the hotspot of the contralesional primary motor cortex (M1) via a BYA90A TMS machine equipped with figure-of-eight coils. To ensure consistency across sessions, the location of the stimulation site was marked using vertex-centered coordinates as a reference point. The coil was oriented at approximately 0° over the transverse plane with the handle pointing laterally to induce lateral-to-medial current flow in the cortex44. We identified a hotspot, where the largest motor evoked potential (MEP) could be consistently evoked on the motor characterization of the contralateral abductor pollicis brevis muscle43. The resting motor threshold (rMT) was subsequently determined for each stimulator separately, and it was stipulated that TMS consistently evokes the amplitude of motor-evoked potentials (MEPs) with a peak-to-peak amplitude of > 50 μV in at least 5 out of 10 trials45,46. The rTMS intervention was performed based on the methods of A Matsuura, J Du, and Jiang Li et al.47,48,49 at a rate of 1 Hz and with an intensity of 80% of the motor threshold for 1200 pulses per session. The sham rTMS used a sham 8 coil, which reproduced the noise of a 1 Hz stimulus and tactile sensation on the scalp without cortical stimulation10,47,50. The use of regular rehabilitation therapy, necessary medications (to control blood pressure, blood sugar, and cholesterol, but not for bladder dysfunction), and nursing management of the bladder were supported for all patients during the 4-week intervention or sham stimulation.

The primary outcome (urodynamic) and secondary outcome (questionnaire and pelvic floor muscle (PFM) surface electromyography) were assessed in the two groups before and after the intervention, and the questionnaire included the OABSS and ICIQ-UI SF.

Outcome measures

The treatment-blinded assessors evaluated all outcome measures. The primary outcome measure was the urodynamic data, which included the MCC, Pdet.max, and residual urine volume. Pdet.max refers to the maximum detrusor pressure during filling, which occurs during the maximal capacity of the bladder during the storage phase. In accordance with the AUA/SUFU guidelines51, urodynamic experiments were conducted via a multichannel urodynamic test system (Nidoo 970A + , China). During the testing, the room was kept quiet and warm. Patients were informed of goals and procedures to minimize stress. If the subjects felt intense or experienced an undesirable autonomic nervous system response, the test was stopped immediately. The pressure sensor was connected to the instrument, and perfusion was performed on the system to ensure that there was no air in the pressure sensor or pressure transmission tube. Moreover, the pressure sensor and priming pump were calibrated. When the patient was ready, it was instructed to urinate and then placed in a semi-reclining position. A sterile double-lumen bladder measuring tube was used as one chamber for saline infusion and one chamber for bladder pressure, which was gently inserted into the urethra. The pump was then connected, and normal saline was infused into the bladder at 30 mL/min. Moreover, this system includes urodynamic testing equipment and computer automatic analysis software. The computer’s automatic analysis software collects and processes the test data, and issues the test reports. Patients underwent only one urodynamic test before and after the intervention. The tester and the patient were blinded to the grouping of the participants.

The secondary outcome measures included questionnaires and pelvic floor surface electromyography. The questionnaires included the OABSS and ICIQ-UI SF. The OABSS measures the impact of urge incontinence and symptoms associated with overactive bladder (OAB)52. The score ranges from 0 to 15, with a lower score indicating less severe OAB. The data included the mean number of episodes during the day, the nighttime frequency, urgency, and incontinence. The ICIQ-UI SF is a brief questionnaire for evaluating the frequency and severity of UI in patients and its impact on quality of life. The score ranges from 0 to 21, with a lower score indicating less severe urinary leakage. PFM surface electromyography included resting and contraction values used to assess PFM activity53. A neurofunctional reconstruction therapy system (AM1000A, China) was used to detect PFM via electromyography.

Data processing and statistical analysis

SPSS 26.0 (SPSS, Inc., Chicago, IL) was used to analyze the data. Statistical significance was indicated by values of P < 0.05 (two-tails). If participants dropped out during the intervention period, the missing data were replaced by the last observation. Owing to the principle of intention to treat (ITT), these patients were not excluded.

Demographic and clinical characteristics were summarized via the same methods: the means and standard deviations (SDs) were used for continuous variables, and categorical variables were expressed as frequencies (percentages). Before performing the comparisons, we used the Kolmogorov–Smirnov test for normally distributed data. Paired sample t-tests were used to compare the pre- and post-intervention changes in each group. Independent samples t-tests were used to compare the two groups, including differences between groups and differences (Δ). If the data were not normally distributed, the nonparametric test was applied. The chi-square test was used for categorical variables. For all values obtained P < 0.05 was considered to indicate statistical significance.