Pelvic floor muscle exercises plus biofeedback versus pelvic floor muscle exercises for patients with stress urinary incontinence: A systematic review and meta-analysis of randomized controlled trials

versus


INTRODUCTION
Stress urinary incontinence (SUI) is a kind of urinary incontinence that happens when physical exertion or movement increases pressure on the bladder, producing urine leakage [1][2][3]. This may develop as a result of weak pelvic floor muscles, which support the bladder and urethra [2][3][4][5]. Coughing, sneezing, laughing, running, or hard lifting might cause urine to flow from the bladder due to the pressure involved [4][5][6][7]. SUI is a widespread illness that mostly affects women, particularly those who have recently given birth or gone through menopause [8,9]. Obesity, nerve injury, and pelvic surgery are among the additional risk factors for SUI [8][9][10].
Conservative measures such as pelvic floor muscle exercises (PFME), weight loss, bladder training, biofeedback, and the use of absorbent items or devices are the first line of SUI therapy [5,6,11,12]. Surgery could be required in some circumstances to maintain or restore the pelvic floor muscles [13,14]. PFME is a frequent conservative therapy for SUI [6,[15][16][17][18]. Strengthening the pelvic floor muscles as part of this therapy aids in improving bladder control and lessening the consequences of incontinence [17]. Nevertheless, some women can struggle to complete the PFME accurately, which might result in suboptimal treatment results [19]. The use of biofeedback has been suggested as a complement to PFME for SUI or even as an alternative for PFME and other conventional therapies [20][21][22][23]. This approach makes use of tools that provide women with immediate feedback on the activity of their pelvic floor muscles, enabling them to improve their workout methods and get better results [22].
The usefulness of PFME alone or in conjunction with biofeedback for the treatment of SUI has been examined in a number of research to date [22,24]. The outcomes of this research, however, are often contradictory, making it challenging to choose the best strategy. To solve this problem, a meta-analysis is required, which synthesizes the findings of much research to provide a more solid and statistically significant conclusion.
The purpose of this meta-analysis is to compare the effectiveness of PFME plus biofeedback to PFME alone in treating SUI in female patients. Changes in incontinence symptoms, quality of life, and pelvic floor muscle strength are the main outcome measures of interest. The findings of this research will have significant ramifications for the treatment of this widespread and incapacitating disorder and will aid in directing clinical judgement for medical professionals and SUI patients.

METHODS
We followed the principles outlined in the PRISMA declaration while publishing this systematic review and metaanalysis [25]. Cochrane handbook of systematic reviews and meta-analyses of interventions was strictly followed throughout every step in this article [26].

Eligibility Criteria
Studies were considered for our analysis if they met the following requirements: Studies provided as abstracts only or theses, studies for which complete full-texts were not readily accessible, research using animals or in vitro, observational studies, review articles, case reports, and case series, as well as studies not written in English were all excluded.

Information Sources and Search Strategy
Three databases (PubMed, Scopus, and Web of Science) have been searched using MeSH terms ((stress urinary incontinence) AND (biofeedback OR feedback OR myofeedback) AND (women OR female OR girl)) to identify articles for review. The search was conducted until February 7, 2022. Language, publishing time, gender, race, or country are all unrestricted. Further, the references of the included studies were manually searched for any potentially eligible studies.

Selection Process
Using Endnote (Clarivate Analytics, PA, USA), duplicates were eliminated, and the recovered references were evaluated in two stages: the first stage included screening the titles and abstracts of all identified papers independently by two authors to determine their relevancy to this meta-analysis, and the second step involved screening the full-text versions of the identified abstracts to determine final qualification to metaanalysis. Rayyan website was used for the selecting process [27].

Data Collection Process and Data Items
Data from the included records was extracted by two impartial reviewers in a preformatted Excel spreadsheet. In addition to outcome indicators, this data will include baseline characteristics of the included studies and the study population. Disagreements will be resolved by evidence-based discussions.

Assessing Risk of Bias in Individual Studies
Using the Cochrane tool for assessment of the risk of bias-2 (RoB2), two independent reviewers will assess the quality of the included studies [28]. Risk of bias assessment included the following domains: bias arising from the randomization process, bias due to deviations from intended interventions, bias due to missing outcome data, bias in the measurement of the outcome, bias in the selection of the reported result, and other biases. The authors' judgments are categorized as "low risk," "high risk," or "some concerns" of bias. Any disagreement will be resolved with debate till a consensus is reached. If authors could not agree, a senior author will be consulted.

Synthesis Methods
For categorical variables, odds ratio/risk ratio with a 95% confidence interval (CI) were calculated to estimate the effect size and compare between intervention and control groups. For continuous variables, mean difference or standardized mean difference (SMD) with 95% CI were calculated to estimate the effect size to assess the difference in outcome measures between intervention and control groups.

Choice of Meta-Analysis Model
If there is no significant heterogeneity, study-specific estimates were pooled using a fixed-effects model; otherwise, a random-effects model was used.

Assessment of Heterogeneity
Chi-square test was used to assess the statistical heterogeneity between the studies (Cochrane Q test). The Isquared was then calculated using Chi-square statistic, Cochrane Q, using the following formula: Significant heterogeneity was defined as a chi-square P value of <0.1. I-square values above 50% were a sign of significant heterogeneity.

Reporting Bias Assessment
We created funnel plots to show the link between effect size and standard error in order to investigate the publication bias across research. Evidence of publication bias was evaluated using two methods: 1. Begg and Mazumdar rank correlation test (Kendall's tau) [29] and 2. Egger's regression test [30].

Certainty Assessment
We performed a sensitivity analysis to conduct a certainty assessment in order to examine the validity of the evidence (also called a leave-one-out meta-analysis). We conducted sensitivity analyses for each outcome in the meta-analysis, eliminating one study from each scenario to ensure that the total impact size was independent of any particular research.

Literature Search Results
Our search for literature turned up 1,074 results. 80 articles were qualified for full-text screening after being subjected to title and abstract screening. The meta-analysis comprised 15 of these investigations. No further papers were included after manually searching the references of the listed studies.

Characteristics of Included Studies
The meta-analysis includes 15 trials with a combined total of 788 SUI patients. Patients were randomly randomized to either receive PFME plus BF or PFME alone in all trials. Table 1 shows the summary of the studies included in this systematic review and meta-analysis.   Table 2 provide an overview of the characteristics of the included articles. In Table 1, we summarized the studies included in this review. Similarly, there are 17 studies in Table  2, where we scrutinized the included studies.
We briefly explain the baseline characteristics of the studies included in this systematic review and meta-analysis. Overall, risk of bias in the included studies varied from a high risk to a low risk according to the RoB2 checklists.

Quality of Life
Overall effect estimate between PFME+BF and PFME alone groups in improving the quality of life does not favor any of the two groups (SMD=-0.22, 95% CI [-0.44 to 0.00], p=0.05) ( Figure  5). The pooled studies were homogenous (p=0.84; I 2 =0%).

Social Activity
Overall effect estimate between PFME+BF and PFME alone groups in improving the social activity does not favor any of the      (Figure 5). The pooled studies were not homogenous (p=0.01; I 2 =77%).

Publication Bias Assessment
The funnel plot shows a relatively symmetrical distribution of studies on both sides of the plot, indicating a low risk of publication bias Figure 6. This suggests that the results of the meta-analysis are less likely to be influenced by small or unpublished studies. However, it is important to note that the precision of the effect estimates may be limited by the small sample sizes of some studies. Future research should aim to increase the sample size of studies and reduce heterogeneity to provide more precise estimates of the effect size. Egger's test slope coefficient of 1.597 suggests a positive relationship between the effect size and its precision in the meta-analysis, but the p-value of 0.110 indicates that there is no significant evidence of publication bias.

DISCUSSION
PFMT is a treatment program for urinary incontinence based on a regular contraction of the pelvic floor muscles in a fashion taught by a healthcare professional. PFMT is often the first choice of treatment for patients with SUI who are seeking conservative management options [6,46,48]. Biofeedback is a technique used in conjunction with pelvic floor muscle training PFMT to enhance the effectiveness of the treatment for SUI [47,Figure 5. Forest plots comparing PFME+BF versus PFME alone in quality of life, leakage, pad weight test, cure rate, & social activity (Source: Authors' own elaboration) 49]. The combination of biofeedback mechanisms with PFMT increases its efficacy, providing patients with better control over their muscles and improving the effectiveness of their PFMT [34,50]. The evidence of the superiority of the effect of combination therapy over PFMT alone is still unclear [48].
This study showed that the addition of biofeedback to PFMT is associated with a statistically significant improvement in several outcomes related to the management of SUI. The results showed that PFMT with biofeedback was more effective than PFMT alone in increasing the strength of the pelvic floor muscles (p<0.0009). However, these promising results regarding the PFM strength and cure rate, both the effect on improving quality of life, promoting social activities, and decreasing frequency and amount of leakage, showed a nonstatistically significant difference between the two arms of the analysis. Taking together the add-on effect of BF to the BFMT plays a great role in the most important outcome of management of SUI despite its effect on other outcomes, but it could be explained by the subsequent causes.
Biofeedback may improve the cure rate and strength of PFME but not affect the frequency and amount of leakage or quality of life because improvement in the strength of PFME strength does not mean by default improvement of the other outcomes because it is expected to still experience leakage with a considerable amount and frequency as leakage does not directly relate to the power of PFME but also other factors as bladder capacity, individual habits and smooth muscle tone, which not related to biofeedback mechanism [48,51,52]. Also, it is expected that the benefits of biofeedback on these outcomes to take a longer time to be effective [53][54][55]. Also, the non-significant superiority of the effect of adding BF to PFME over PFME alone regarding the amount of leakage rate could be attributed to the evaluation tool. Most of the included studies depended on pad tests to reflect the amount and frequency of leakage, but this may cause some limitations on the significance of the superiority of PFME with BF over PFME alone as it is a subjective assessment, which depends on individual reporting, which is variable across individuals and across all included study participants therefore, pad test may not reflect the actual improvement of leakage improvement [56][57][58]. The non-significant results regarding the cure rate are attributed to the very small number of studies assessing this outcome (only three) and the difference between the studies in the definition of the cure. Moreover, when removing ong2015 [41] study to resolve the heterogeneity, the results become significant, favoring the PFME+BF group.
Non-significant superiority of BF and PFME over the PFME group alone on the quality of life of SUI patients could be explained by the concept of quality of life itself, which is variable and depends on multiple variables ranging from physical, psychological and even social domains, which are multifactorial and not relay of PFM strength, which is the primary outcome from adding BF to PFME, which is not necessary to extend its effect to other aspects of life as patient satisfaction, social interaction and quality of life improvement [59][60][61]. Also, the effect of SUI on the patient's quality of life is variable across patients depending on the severity of the disease, other associated diseases and even other personal expectations from treatment [62][63][64][65][66]. Some patients consider the reduction of the frequency and amount of leakage as a primary goal for a better quality of life, while others need additive assistance to enhance their quality of life. Finally, it is important to mention that included studies are variable in the quality-of-life assessment scale, and even the results of the effect of included studies on the quality of life are variable. This may maybe be explained by different designs of measuring this outcome according to the variable population, which also reflex the difficulty of assessing the quality of life with a standardized assessment tool. All these points may explain the non-significance difference between our two arms.
Improving PFM strength helps patients to control incontinence over time, which can increase the cure rate and improvement of SUI symptoms BF and PFME compared to the PFME group alone. It is also expected that increasing the PFM strength over time decreases the frequency of leakage, depending on the role of BF to motivate the patients to adhere to the PFME via visual and auditory information [50,67]. Also, biofeedback could help patients in a way that is appropriate for each of them by learning how to contract the muscle in a manner suitable for each case. But we must take into consideration the type of biofeedback device, which could affect the adherence of individuals [68]. Many women do not favor the vaginal probe defendant devices due to uncomfortable maneuvers for application, which may affect adherence to exercise and also affect the outcomes of improvement. Therefore, we enhance health care providers to choose the best maneuver for each group of patients to increase the adherence rate. Our meta-analysis included a total of 15 randomized controlled trials with a total of 788 participants, comparing PFME alone to PFME with the addition of biofeedback. The evidence from this study strongly supports the additive effect of BF to PFME on the improvement of PFME strength in the management of SUI in women. The inclusion of a large number of clinical trials adds a bigger statistical power, which has a favorable effect on the generalizability of results. Also, still, the sample size of participants needed to increase in future studies to avoid sample size limitation points. We recommend future researchers study the effect of using different biofeedback devices on the cure rate, PFME strength, adherence, and patient satisfaction to find and highlight the most accepted form to conduct the biofeedback. Further research is important to study most effective treatment protocols for SUI patients.

CONCLUSIONS
Drawing upon our discussion, we will be able to conclude that BF addition to PRME is able to improve the cure rate and the strength of PFME without a significant impact on the amount and frequency of leakage, and quality of life. The healthcare provider must consider patients' safety and comfortability while selecting the used BF device with PFME. BF is a valuable tool in improving the outcomes of controlling SUI and is recommended for its management plans. Future research should investigate the use of different leakage measurement techniques and concentrate on creating complete therapies that address both the physical and psychological components of urine incontinence. Also, when offering urinary incontinence treatment choices to patients, healthcare professionals should consider their specific requirements and preferences.
Author contributions: All authors have sufficiently contributed to the study and agreed with the results and conclusions.

Funding:
No funding source is reported for this study.