Physical and Rehabilitation Therapy for Overactive Bladder in Women: A Systematic Review and Meta-Analysis

Objective To compare the effects of different physical and rehabilitation therapies on women with overactive bladder (OAB). Design Network meta-analysis. Data source: The Embase, Scopus, and PubMed databases were systematically searched from their inception to June 22, 2022. We included only RCTs, with no language restrictions. Articles in the reference lists and related studies were thoroughly reviewed. Review Methods. This network meta-analysis included related studies on different physical and rehabilitation therapies for OAB. Data were extracted independently from the included randomized controlled trials by two authors, and they used the Cochrane Collaboration's tool to evaluate the risk of bias. We used RevMan to assess the risk assessment of research bias. This network meta-analysis was performed using the Stata software. We completed the review in accordance with the PRISMA items for systematic reviews and meta-analyses statement. Results Twelve randomized controlled trials involving 637 patients were included in this meta-analysis. All physical and rehabilitation therapies improved daytime micturition frequency and nocturia frequency in OAB patients. Percutaneous tibial nerve stimulation (PTNS), BT + ES, and BT + BF + ES are better interventions for OAB treatment. There were no significant differences in PTNS, BT + ES, and BT + BF + ES. Conclusion All physical and rehabilitation therapies can improve daytime micturition and nocturia frequency in OAB. PTNS, BT + ES, and BT + BF + ES were the priority therapies.


Introduction
Overactive bladder (OAB) is defned by the International Continental Society (ICS) as urgency, with or without urge incontinence, usually with frequency and nocturia, which is a symptom syndrome [1]. Diseases with defnite pathological factors or those associated with infection must be excluded from the diagnosis [1]. Te reported incidence among females in diferent regions varies, with approximately 6% incidence in China [2], 16.9% in the United States [3], and 16.6% in Europe [4], and an overall trend that increases with age [5,6]. Moreover, OAB has a signifcant negative impact on the quality of live, also with increasing healthcare costs [7][8][9][10].
Te American Urology Association and Society of Urodynamics, Female Pelvic Medicine, and Urogenital Reconstruction guidelines recommend four levels of management for OAB: conservative therapy, drug therapy, intradetrusor injection of onabotulinum toxin, sacral neuromodulation, and peripheral tibial nerve stimulation, and surgical therapy [11]. Among the therapeutic techniques for OAB, drug therapy is the simplest and most convenient, which can efectively relieve the symptoms of OAB [12]. However, due to side efects such as dry mouth, constipation, and vision irregularity, the sustainability of drug therapy is seriously afected [13,14]. Surgical treatment is often used after failure of drugs, physical therapy, and other therapies [15,16].
Physical and rehabilitation therapies provide more options for female patients with OAB. However, there is a lack of relevant comparative research. Terefore, we searched for relevant randomized controlled trials (RCTs) to compare the efects of diferent physical and rehabilitation therapies in women with OAB.

Search Strategy.
We performed a network meta-analysis of all studies according to the PRISMA 2020 statement [17]. Te Embase, Scopus, and PubMed databases were systematically searched from their inception to June 22, 2022. We searched for titles, abstracts, and keywords by combining the following terms: (overactive bladder) OR (overactive detrusor)) AND (bladder training) OR (pelvic foor muscle training) OR (biofeedback) OR (electric * ) OR (Kinesio taping) OR (pelvic foor muscle exercise) OR (PTNS) OR (percutaneous tibial nerve stimulation) OR (TTNS) OR (transcutaneous tibial nerve stimulation) OR (tibial nerve stimulation) OR (sacral neuromodulation) AND (random * ). Tere were no language restrictions. Articles in the reference lists and related studies were thoroughly reviewed. We did not register for this study.

Inclusion and Exclusion
Criteria. Trough a systematic review of all relevant literature, the following studies that met the criteria were included: (1) randomized controlled trials of physical and rehabilitation therapy in women with OAB and (2) women over the age of 18 years. Studies that met the following criteria were excluded: (1) studies of drug therapy or studies without drug washout; (2) studies with OAB surgery; (3) infectious factors, such as urinary tract infection or other obvious pathological changes, such as neurological diseases; and (4) other literature, such as conference abstracts, systematic reviews or meta-analyses, case reports, letters to the editor, or comments.

Data Extraction.
All data were extracted by two authors from randomized controlled trials that met the inclusion criteria. Te year of publication, specifc randomization method, sample size of patients, interventions, and outcomes after the intervention were gathered from each study. We entered the extracted information into Microsoft Excel, and any confict between the two authors was resolved through full discussion with the third author.

Risk of Bias Assessment.
Te risk of bias assessment for the included trials was performed independently by two authors using the Cochrane Collaboration's tool [18]. Factors included in the evaluation were sequence generation, allocation concealment, blinding, incomplete outcome data, selective outcome reporting, and other issues. Any disagreement during the evaluation was resolved through a full discussion with the third author.

Statistical Analysis.
Review Manager Software (RevMan 5.3; Cochrane Corporation) was used to assess the risk of research bias [19]. Te pairwise meta-analysis of direct comparisons was performed using STATA-17.0 (StataCorp, Texas). Te mean diference (MD) and 95% confdence interval (CIs) were used to describe the results for continuous data. Statistical signifcance was set at P < 0.05. Statistical heterogeneity among studies was assessed using the I 2 statistic. If I 2 ≤ 50%, a fxed-efects model was used for synthesis, otherwise, the random efects model was used. Subgroup and sensitivity analyses were performed when necessary. We used STATA-17.0's "network map," "ifPlot," "netfunnel," and "sucra" programs to generate the resulting statistical graphs.

Identifcation of Relevant Studies.
We identifed 1318 studies by searching the Embase, Scopus, and PubMed databases. Of these, 334 duplicate studies were deleted, and 984 studies were retrieved, with a total of 266 randomized controlled trials. Among them, nine were conference abstracts, and 232 were deleted according to the exclusion criteria. Finally, 25 studies were included according to the inclusion criteria. Te risk of literature bias was assessed based on the results reported by the studies, and data from only 12 studies were combined for this network metaanalysis. A research fowchart related to this literature review is shown in Figure 1.

Characteristics of the Included Studies and Risk of Bias
Assessment. For the 25 RCTs included, the risk of bias was assessed using RevMan 5.3. Te generated risk of bias graph is shown in Figure 2, and the risk of bias summary is shown in Figure 3. In terms of random sequence generation, 92% had a low risk of bias and 8% had an unclear risk of bias. In terms of allocation concealment, 80% had a low risk of bias and 20% had an unclear risk of bias. In terms of blinding of participants and personnel, 60% had a low risk of bias, 20% had a high risk of bias, and 20% had an unclear risk of bias. In terms of blinding of outcome assessment, 48% had a low risk of bias, 16% had a high risk of bias, and 36% had an unclear risk of bias. In terms of incomplete outcome data, 92% had a low risk of bias, 4% had a high risk of bias, and 4% had an unclear risk of bias. In terms of selective reporting, 96% had a low risk of bias and 4% had an unclear risk of bias. In terms of other biases, 96% had a low risk of bias, and 4% had an unclear risk of bias.
Among the 25 studies, only 12 studies collected data on daytime micturition frequency or nocturia frequency; therefore, 12 studies were included in the meta-analysis. A total of 637 women completed the 12 randomized control trials in this network meta-analysis. Twelve physical and rehabilitation therapies for OAB were included. Table 1 summarizes the basic characteristics of the included studies. Tere were eight two-arm trials, two three-arm trials, and one four-arm trial. A total of 112 patients were treated with percutaneous tibial nerve stimulation  Figure 4 shows the "network map" of daytime micturition and nocturia frequencies.  International Journal of Clinical Practice

Inconsistency Test.
Te comparison-corrected funnel plots for daytime micturition and nocturia frequencies are shown in Figure 5. All the studies were symmetrically distributed on both sides of the midline in the middle and upper parts, suggesting a low risk of publication bias. Inconsistencies were tested for daytime micturition frequency and nocturia frequency, with P values of 0.9680 and 0.3628, respectively; hence, there were no inconsistencies observed. As shown in Figure 6, the ring inconsistency detection shows that the ring inconsistency is not signifcant, and the direct comparison results are basically consistent with the indirect comparison results; therefore, it cannot be considered that there is great heterogeneity.

Daytime Micturition Frequency.
A Bayesian mesh metaanalysis of the daytime micturition frequency was conducted for the included studies. Te SUCRA values are shown in Figure 7, and the probability ranking results show that the advantages and disadvantages of diferent interventions are ranked in the following order: Tere were 78 pairwise comparisons in the network meta-analysis, of which, four were statistically signifcant (P < 0.05). BT + ES, BT + BF + ES, and PTNS were superior to BT alone. BT + ES were superior to BT + MStim, and there were no signifcant diferences among the other interventions.

Nocturia Frequency.
A Bayesian mesh meta-analysis of nocturia frequency was conducted on the included studies.
Te SUCRA values are shown in Figure 8, and the probability ranking results show that the advantages and disadvantages of diferent interventions were ranked in the following order: BT + BF + ES > PTNS > PFMT + ES > BT + ES > BT + BF > SNS > BT + MStim > PFMT + TTNS > TTNS > PFMT > placebo > acupuncture > BT, with 78 pairwise comparisons in the network meta-analysis, and 18 of which were statistically signifcant (P < 0.05). BT + BF + ES were superior to 11 treatments, except PFMT + ES; PTNS was superior to TTNS, placebo, acupuncture, and BT; PFMT + ES, BT + ES, and BT + BF were superior to BT, and there were no signifcant diferences among the other interventions.

Discussion
Physical and rehabilitation therapies for OAB are either single or combined. According to this network metaanalysis, all therapies can reduce the daytime micturition frequency or nocturia frequency of OAB to a certain extent. Combined with the improvement in daytime micturition frequency and nocturia frequency, PTNS in single therapy and BT + ES and BT + BF + ES in combination therapy may be efective therapies for OAB treatment. Tere was no signifcant diference in PTNS, BT + ES, and BT + BF + ES.
Bladder training (BT), as the frst-line recommended therapy for female incontinence can reduce the frequency of urination [20], which is consistent with the results of this network meta-analysis. However, only BT can improve the frequency of urination to a limited extent; therefore, Firinci et al., Rizvi et al., and Phillips et al. tried to fnd a better combination therapy based on BT [21][22][23]. Te results of this network meta-analysis showed that BT + ES and BT + BF + ES showed better therapeutic efects in controlling daytime micturition frequency and nocturia frequency, and that aging leads to a reduction in bladder and urethral function [6,24,25]. RCT studies included in this network meta-analysis used transvaginal electrodes for electrical stimulation therapy, which can efectively reduce pudendal nerve aferents or inhibit detrusor activity through current stimulation [26,27], thus improving OAB symptoms. In these two RCT studies, ES was treated three days a week with a single treatment time of 20 minutes for 24 times, i.e., the total treatment time was 8 weeks. ES therapy settings were as   follows: 10 HZ stimulation parameters, work-rest cycle 5-10 s, pulse width 100 ms. Symmetrical biphasic pulses were adjusted between 1 and 100 mA depending on the patient's discomfort level [4,21]. Based on the improvement in daytime micturition frequency and nocturia frequency, this setting seems reasonable. BT + ES can efectively reduce the daytime micturition frequency and nocturia frequency, patients with ES therapy have good tolerance, and adverse reactions include discomfort and vaginal infection [4,28]. Whether there is a better setting method for electrical   International Journal of Clinical Practice stimulation and therapy course needs further attention in future studies, as it can guide patients to correct contraction control of pelvic foor muscle activity, thus making pelvic foor muscle training more efective [29]. As seen from the RANK plot, BT + BF + ES is the best therapy for improving nocturia frequency and one of the best interventions to improve daytime micturition frequency. In practice, biofeedback can be used at the same time as ES, and the treatment time needs to be extended each time, with good acceptance by patients [21]. Percutaneous tibial nerve stimulation (PTNS), which is a minimally invasive treatment, uses a fne needle inserted 5 cm cephalad from the medial malleolus and posterior to the edge of the tibia [30,31]. Te sacral plexus (S2-S4) is stimulated to reach the sacral urination center by stimulating the aferents of the tibial nerve (which contains the parasympathetic plexus of the bladder), and detrusor relaxation is induced by the inhibition of parasympathetic motor neurons [30,32]. Because PTNS is a minimally invasive treatment, some scholars have proposed a noninvasive treatment for TTNS that replaces fne needles with surface electrodes. In the study by Sonmez et al., Ramírez-García et al., and Zonić-Imamović et al., PTNS and TTNS had similar clinical efects [31,33,34], but the results of this network meta-analysis showed that there was no statistically signifcant diference between PTNS and TTNS in the improvement of the daytime micturition frequency, but the probability ranking of RANK was better for PTNS than TTNS. PTNS was better than TTNS in improving nocturia frequency, possibly because fne needle aspiration enables electrical stimulation to act directly on the nerve and the efect is more precise; however, PTNS has more adverse reactions than TTNS [31,32].
Te therapies included in this network meta-analysis all improved symptoms of OAB, with therapy durations ranging from 8 to 12 weeks; however, according to existing studies, most of these improvements were time-limited, and  International Journal of Clinical Practice symptoms of OAB gradually worsened as treatment was stopped for longer [25,32,35,36]. Te efective control of OAB in the long term is worthy of further study, including physical and rehabilitation therapies, such as BT, behavioral therapy, PFMT, and TTNS, which can be independently treated at home [31,35,37], but PTNS, BF, and ES, etc., need to be carried out by professionals in medical institutions with professional equipment [4,21,30,33]. PTNS, BT + ES, and BT + BF + ES can be the priority therapy when circumstances permit. Te strengths of our network meta-analysis include the following: (1) minimizing publication bias, we searched all domains using keyword combinations; (2) faced with different research methods, we only included randomized controlled trials; (3) the results of the study combined direct and indirect comparisons; (4) Bayesian network metaanalysis was conducted [38] to compare the treatment effects of diferent physical and rehabilitation therapies on OAB; and (5) posterior probabilities and SUCRA were used to distinguish diferences between the 12 physiotherapy and rehabilitation therapies. However, these are some limitations to this network meta-analysis: (1) the number of included studies and patients was small; (2) characteristics vary among diferent studies, which may have a potential impact on the results of the research; and (3) many symptom evaluation indicators and evaluation methods of OAB lack a standardized and unifed evaluation system; (4) only two representative indicators of daytime micturition frequency and nocturia frequency were selected; (5) the duration of the intervention included in the studies was short; and (6) there was a lack of assessment of the long-term efect of improvement in OAB symptoms. Terefore, larger randomized controlled trials that include assessments of long-term improvement in OAB symptoms are needed to enrich the conclusions of this network meta-analysis.

Conclusions
Diferent physical and rehabilitation therapies can improve the symptoms of OAB. Each institution should choose physical and rehabilitation therapies according to its own setup and personnel base. PTNS, BT + ES, and BT + BF + ES can be considered priority therapies when the situation permits.

Data Availability
All data and codes are available from the corresponding author upon request.

Conflicts of Interest
Te authors declare that there are no conficts of interest in publishing this article.

Authors' Contributions
Jingwen Bai and Ping Wang participated in the study design and in writing the manuscript. Jingwen Bai, Yilan Tian, and Yiran Wang participated in the data collection. Jingwen Bai and Yilan Tian analyzed the data. Jingwen Bai and Xiaofang Zhang interpreted the data and critically reviewed the paper under the supervision of Ping Wang. All authors read and approved the fnal manuscript.