Open Access
CC BY 4.0 · Journal of Coloproctology 2025; 45(03): s00451811261
DOI: 10.1055/s-0045-1811261
Original Article

Pelvic Floor Physiotherapy with Biofeedback Improves Constipation in Post-Bariatric Surgery Patients

Authors

  • Marcieli Schuster

    1   Graduate Program in Biosciences and Health, Center for Biological and Health Sciences, Western Paraná State University (UNIOESTE), Cascavel, PR, Brazil
  • Caroline Dresch Sabadin

    2   Western Paraná State University (UNIOESTE), Cascavel, PR, Brazil
  • Cláudia Silveira Viera

    3   Center for Biological and Health Sciences, Western Paraná State University (UNIOESTE), Cascavel, PR, Brazil
  • Maria Lúcia Bonfleur

    3   Center for Biological and Health Sciences, Western Paraná State University (UNIOESTE), Cascavel, PR, Brazil

Funding This study forms part of the PhD Thesis of Marciele Schuster and was supported by grants from Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES, PROAP, Process N.: 888881.690222/2022-01).
 

Abstract

Bariatric surgery has proven effective in managing severe obesity but may lead to complications involving pelvic floor function and anorectal health. This retrospective cohort study examined the effect of pelvic floor physical therapy (PFPT) with surface electromyography (EMG) biofeedback on anorectal alterations in individuals following surgery. The analysis included patients who underwent either Roux-en-Y gastric bypass (RYGB) or sleeve gastrectomy (SG) from 2015 to 2019 and subsequently experienced proctological complaints. Data on clinical profiles, demographics, and pelvic floor functionality were extracted from medical records. The PFPT protocol consisted of personalized weekly sessions using surface EMG feedback, with pre- and post-treatment EMG data collected for comparison. Out of 1,639 patients who underwent bariatric procedures during the period, 477 (29%) developed proctological symptoms, with constipation being the most frequent (42%). Nevertheless, only 64 (13.42%) were referred to as PFPT. Anorectal manometry assessments showed that most of these individuals had reduced resting and contraction pressures, generally without signs of anismus. Notably, differences in muscle function patterns were observed between surgical techniques. After undergoing PFPT, participants presented increased EMG activity and substantial relief from constipation. These results highlight the frequent occurrence of anorectal dysfunction after bariatric surgery and support PFPT with EMG biofeedback as a valuable non-invasive intervention for symptom relief and functional recovery.


Introduction

The global rise in obesity has reached critical levels, representing a major public health challenge and exerting a substantial negative impact on individuals' well-being.[1] Alongside widely recognized metabolic and cardiovascular comorbidities—such as type 2 diabetes, dyslipidemia, hypertension, and joint degeneration—there is growing awareness of the musculoskeletal effects of obesity, including disturbances in pelvic floor muscle (PFM) function.[2] These muscles are fundamental for maintaining continence, supporting pelvic organs, and facilitating sexual health, yet their dysfunctions in the context of obesity have been relatively underexplored.[3]

Pelvic floor disorders (PFDs) encompass a broad spectrum of clinical conditions, including urinary incontinence (UI), pelvic organ prolapse (POP), fecal incontinence (FI), anal incontinence (AI), constipation, and sexual dysfunction. The association between obesity and urinary dysfunction is well established; however, data regarding the relationship between obesity and anorectal disorders, such as AI, FI, and constipation, remain limited and often contradictory.[4] [5]

The pathophysiological link between obesity and PFM dysfunction is largely attributed to the chronic increase in intra-abdominal pressure, which is positively correlated with body mass index (BMI).[6] This persistent elevation in intra-abdominal pressure, due to excessive abdominal wall mass, places continuous strain on the PFMs, potentially leading to fatigue, weakening of the musculature, and subsequent functional impairments.[7] As a result, weight reduction is expected to contribute not only to metabolic recovery but also to improvement in pelvic floor performance.

Bariatric surgery has become a widely adopted strategy for long-term weight management in cases of severe obesity, with Roux-en-Y gastric bypass (RYGB) and sleeve gastrectomy (SG) being the most common techniques.[1] [8] While these procedures offer considerable benefits, including substantial weight loss and comorbidity resolution, their influence on anorectal physiology is not fully understood. Reports have noted the emergence or worsening of bowel dysfunctions—such as fecal incontinence and constipation—even after successful surgery, suggesting a complex interplay between anatomical changes, dietary habits, and pelvic floor adaptation.[9] [10]

In this context, pelvic floor physical therapy (PFPT), particularly using surface electromyographic (EMG) biofeedback, has shown promise as a non-invasive approach to restore PFM function. This technique has been associated with improvements in sphincter control, muscle strength, and overall quality of life in patients with PFD.[11] [12] There is currently a complete lack of research specifically addressing the impact of PFPT on anorectal outcomes following bariatric surgery, highlighting the need for investigation in this area. Therefore, the present study aimed to evaluate the effects of PFPT on anorectal dysfunction in post-bariatric patients, utilizing EMG biofeedback and validated scoring tools to measure treatment outcomes.


Methods

This retrospective study included patients who underwent bariatric procedures, specifically RYGB or SG, between 2015 and 2019. Individuals were eligible if they presented proctological symptoms following surgery. Cases with incomplete records or with symptoms limited to the urinary tract were excluded from the analysis. Data collection was conducted via the SISCLINICA® electronic medical record system, encompassing demographic details, clinical information, and physical therapy follow-up.

Study variables were categorized into three main domains: sociodemographic characteristics, clinical data, and PFM functionality parameters. The same physical therapist evaluated all patients to ensure consistency in assessment procedures. During the initial consultation and at the end of the physiotherapy intervention, Wexner/Cleveland Clinic Fecal Incontinence Score (CCFIS) and the Constipation Severity Score (CSS) were applied.[13] For analytical clarity, the results from both instruments were categorized into clinically relevant severity levels. CSS scores were classified as follows: 0–no constipation; 1–10–mild constipation; 11–20–moderate constipation; and 21–30–severe constipation. CCFIS scores were categorized as: 0–perfect continence; 1–7–mild incontinence; 8–14–moderate incontinence; and 15–20–severe incontinence.

PFM activity was quantified using intracavitary EMG assessment with the Miotool® (four-channel) device, following the Glazer protocol. EMG outcomes included baseline muscle tone and maximal voluntary contraction, both recorded before and after the therapeutic intervention. The rehabilitation protocol comprised weekly 40-minute individualized sessions, during which EMG biofeedback was used to guide muscle training, tailored to the specific functional impairments of each patient.

For statistical analysis, sample characterization included variables such as sex, marital status, education, occupation, religion, type of surgery, alcohol consumption, current or past smoking, physical activity, and presence of comorbidities. These variables were summarized using descriptive statistics, including absolute and relative frequencies, presented in tables. Continuous variables such as age and body mass index (BMI) were described as means and standard deviations. The chi-square goodness-of-fit test was used to examine the distribution of categorical characteristics. Changes in proportions before and after physiotherapy were analyzed using the Z-test for proportions. Categorical data were presented as absolute numbers (n) and percentages (%). Outliers were identified using the interquartile range (IQR) method, with atypical values defined as those below Q1 − 1.5 × IQR or above Q3 + 1.5 × IQR. Normality of distributions was tested using the Shapiro-Wilk test for both the overall sample and subgroups. EMG outcomes pre- and post-intervention were analyzed using repeated-measures analysis of variance (ANOVA) within a linear mixed-effects model. Parametric data were reported as mean ± standard error, and non-parametric data as median ± IQR. A significance level of p < 0.05 was adopted for all analyses.

The study was approved by the Human Research Ethics Committee of the State University of Western Paraná (UNIOESTE; approval number 5.516.757) and by the Ethics Committee of the Innovation and Science in Technology Center (NICT) at Gastroclínica Cascavel (approval number 001/21). All procedures complied with ethical standards and ensured data confidentiality and participant safety.


Results

Sample Characterization

Between 2015 and 2019, 1,639 patients underwent bariatric surgery (either RYGB or SG) at our service. Of these, 477 (29%) reported proctological symptoms postoperatively and were included in our sample. Of the 477 patients, 75.47% were female and 24.53% were male ([Table 1]). The mean age of the total sample was 37.3 ± 11.1 years, and the mean BMI was 36.69 ± 4.60 kg/m2. The most frequently performed surgical procedure was RYGB, accounting for 70.44% of the cases. Most of the sample consisted of Brazilian nationals (98.95%) and married individuals (67.09%), with 44.23% having completed high school and 40.25% holding a university degree. Regarding religion, 296 participants self-identified as Catholic or Christian. Most individuals reported alcohol consumption, while 94.13% were non-smokers, and 76.52% had never smoked. Additionally, 77.78% reported not engaging in regular physical activity, and 64.57% denied having any comorbidities. Among those who reported comorbid conditions, the most prevalent was hypertension (22.43%).

Table 1

Sociodemographic and clinical profile of the study sample (n = 477)

Variable

n (%) or Mean ± SD

Sex

 - Female

360 (75.47%)

 - Male

117 (24.53%)

Age years

37.31 ± 11.14

BMI

39.69 ± 4.60

Surgical procedure

 - SG

141 (29.56%)

 - RYGB

336 (70.44%)

Nationality

 - Brazilian

472 (98.95%)

 - Foreigner

5 (1.05%)

Marital status

 - Married

320 (67.09%)

 - Divorced

29 (6.08%)

 - Single

122 (25.58%)

 - Widowed

6 (1.26%)

Education level

 - Elementary school

27 (5.66%)

 - High school

211 (44.23%)

 - Undergraduate degree

192 (40.25%)

 - Postgraduate degree

47 (9.85%)

Religion

 - Atheist, agnostic, none, or not reported

25 (5.24%)

 - Catholic or Christian

348 (72.96%)

 - Evangelical, Adventist, Baptist or Congregational

86 (18.03%)

 - Spiritist (Kardecist)

14 (2.94%)

 - Islamic

1 (0.21%)

 - Jehovah's witness or David's witness

1 (0.21%)

 - Umbanda

1 (0.21%)

 - Wicca

1 (0.21%)

Alcohol Consumption

 - Yes

390 (81.76%)

 - No

87 (18.24%)

Smoking history

 - Yes

28 (5.87%)

 - No

449 (94.13%)

Physical activity

 - Yes

106 (22.22%)

 - No

371 (77.78%)

Comorbidities

 - None reported

308 (64.57%)

 - Sleep apnea

3 (0.63%)

 - Anxiety

2 (0.42%)

 - Asthma

2 (0.42%)

 - Hypertension

107 (22.43%)

 - Depression

1 (0.21%)

 - Diabetes

11 (2.31%)

 - PCOS

1 (0.21%)

 - Arthropathy

20 (4.19%)

 - Two or more comorbidities

22 (4.61%)

Abbreviations: BMI, Body Mass Index; PCOS, Polycystic Ovary Syndrome; RYGB, Roux-en-Y Gastric Bypass; SV, Sleeve Gastrectomy.



Prevalence of Proctological Symptoms in the Total Sample

As shown in [Table 2], constipation was the most frequently reported symptom among post-bariatric patients (42%), followed by anal fissures (22%) and hemorrhoids (13%). When stratified by type of surgical procedure, constipation was significantly more prevalent in patients who underwent RYGB compared with those who had SG ([Table 2]). The prevalence of other proctological symptoms did not differ significantly between surgical procedures.

Table 2

Incidence of proctological symptoms in the total sample and according to the type of bariatric surgery

Proctological symptoms

Total

n = 477

n (%)

SG

n = 141

n (%)

RYGB

n = 336

n (%)

P value

Constipation

200 (42%)

47 (33%)

153 (46%)

0.0137*

Fecal incontinence

19 (4%)

6 (4%)

13 (4%)

0.8439

Fecal urgency

5 (1%)

2 (1%)

3 (1%)

0.6070

Anal fissure

111 (23%)

41 (29%)

70 (21%)

0.0518

Hemorrhoids

61 (13%)

17 (12%)

44 (13%)

0.7566

Diarrhea

45 (9%)

18 (13%)

27 (8%)

0.1068

Two or more symptoms

36 (8%)

10 (7%)

26 (8%)

0.8075

Abbreviations: RYGB, Roux-en-Y Gastric Bypass; SG, Sleeve Gastrectomy.


*P-value < 0.05, Chi-square test for k proportions.



Clinical Profile and Prevalence of Proctological Symptoms Among Patients Referred to PFPT

Of the 477 patients presenting with proctological symptoms, only 64 (13.42%) were referred to PFPT. Among these, 87.5% were female and 12.5% were male, with a mean age of 41 ± 11.52 years. The mean BMI was 39.56 ± 4.26 kg/m2, and RYGB was again the most common procedure (62.5%) than SG (37.5%). Regardless of the surgical technique, constipation was the most prevalent symptom in this subgroup (40.6%, p < 0.0001; [Table 3]), compared with other symptoms (Anal Fissure 25%; FI 18.8%; Two or more symptoms 10.94%; Fecal urgency 3.1% and Hemorrhoids 1.56%).

Table 3

Anorectal manometry profile of 54 patients referred to the physiotherapy service, according to the type of bariatric surgery performed

Variable

Total

N = 54

SG

N = 18

RYGB

N = 35

z

P- value

n (%)

n (%)

n (%)

Resting pressure

- Normotonic

21(38.89%)

9 (47.37%)

12 (34.29%)

1.8817

0.0600

- Hypotonic

30 (55.56%)

8 (42.11%)

22 (62.86%)

−2.9381

0.0033*

- Hypertonic

3 (5.56%)

2 (10.53%)

1 (2.86%)

2.1699

0.0300*

Squeeze pressure

- Normotonic

18 (33.33%)

9 (47.37%)

9 (25.71%)

3.1806

0.0015*

- Hypotonic

35 (64.81%)

10 (52.63%)

25 (71.43%)

−2.7391

0.0062*

- Hypertonic

1 (1.85%)

0 (0,0%)

1 (2.86%)

−1.7033

0.0885

Anismus

- Present

42 (77.78%)

14 (73.68%)

28 (80.0%)

−1.0593

0.2894

- Absent

12 (22.22%)

5 (26.32%)

7 (20.0%)

1.0593

0.2894

P-value 0.05, Z-test for two proportions.



Anorectal Manometry (ARM) Profile of Patients Referred to PFPT

Of the 64 patients referred to PFPT, 54 (84.38%) underwent ARM before the intervention. As shown in [Table 3], most patients exhibited hypotonic resting and squeeze pressures, along with the presence of anismus (77.78%). When stratified by surgical procedure, hypotonic resting and squeeze pressures were significantly more frequent in RYGB patients (62.86% and 71.43%, respectively) than in SG patients (p = 0.0033 and p = 0.0062, respectively). Conversely, SG patients are more frequently presented with normotonic or hypertonic values for both pressures compared with RYGB patients. No significant differences were observed between surgical groups regarding the presence or absence of anismus.


Effects of PFPT on Constipation and FI in Post-Bariatric Patients

The distributions of constipation and FI categories differed significantly both before (χ2 = 38.38, p < 0.0001; χ2 = 45.38, p < 0.0001) and after (χ2 = 74.13, p < 0.0001; χ2 = 33.06, p < 0.0001) PFPT, indicating that certain categories were more prevalent within the group ([Table 4]). Before intervention, moderate constipation was the most common (57.8%), followed by severe (18.8%) and mild constipation (15.6%). Regarding FI, most patients had perfect continence (59.4%), followed by moderate incontinence (25%) and both severe and mild incontinence (7.8% each).

Table 4

Effects of pelvic floor physiotherapy on constipation and fecal incontinence as assessed by the Constipation Severity Score (CSS) and the Cleveland Clinic Fecal Incontinence Score (CCFIS), respectively

Pre-PFPT

χ2

P value

Post-PFPT

χ2

P value

P value

pre x post

Constipation (CSS)

 No constipation

5 (7.8%)

38.38

<0.0001

23 (35.9%)

74.13

<0.0001

<0.0001

 Mild constipation

10 (15.6%)

41 (64.1%)

<0.0001

 Moderate constipation

37 (57.8%)

0 (0%)

<0.0001

 Severe constipation

12 (18.8%)

0 (0%)

<0.0005

Fecal Incontinence (CCFIS)

 Perfect continence

38 (59.4%)

45.38

<0.0001

55 (85.9%)

130.1

< 0.0001

0.0008

 Mild continence

5 (7.8%)

9 (14.1%)

0.2538

 Moderate continence

16 (25%)

0 (0%)

<0.0001

 Severe continence

5 (7.8%)

0 (0%)

0.0226

Abbreviation: PFPT, Pelvic Floor Physical Therapy.


P-valor 0.05, Z-tests for two proportions.


To evaluate treatment effects, proportions for each category were compared before and after the intervention using z-tests for two proportions. For the “no constipation” category, the proportion increased from 7.8% pre-treatment to 35.9% post-treatment (z = 4.808, p < 0.0001). Mild constipation increased from 15.6% to 64.1% (z = 7.005, p < 0.0001), while moderate and severe constipation significantly decreased (moderate: from 57.8% to 0%, z = 9.016, p < 0.0001; severe: from 18.8% to 0%, z = 4.555, p < 0.0005).

In terms of FI, perfect continence increased from 59.4% to 85.9% (z = 3.363, p = 0.0008). Although mild incontinence rose slightly (7.8% to 14.1%), this difference was not statistically significant (z = 1.141, p = 0.254). Moderate and severe incontinence both decreased to 0% post-treatment (moderate: from 25%, z = 4.276, p < 0.0001; severe: from 7.8%, z = 2.279, p = 0.023; [Table 4]).


Effects of Physical Therapy on Pelvic Muscle Activity

Pelvic muscle activity was assessed using electromyography (EMG) before and after PFPT. A significant effect of time on EMG values was observed (F (1.63) = 455.922, p < 0.0001; [Table 5]), with post-treatment values significantly higher than pre-treatment values. On average, EMG values were 25.725 units lower before therapy compared with after, regardless of surgical technique (p < 0.0001). The surgical procedure type did not significantly influence EMG outcomes (interaction coefficient = –1.88, p = 0.584). However, the interaction between time and surgical technique was significant (interaction coefficient = 4.93), indicating that the increase in EMG values from pre- to post-treatment was greater among RYGB patients compared with SG patients (p = 0.028; [Table 5]).

Table 5

Effects of pelvic floor physiotherapy on pelvic floor muscle activity assessed by intracavitary electromyography (EMG)

EMG

Pre-PFPT

Post-PFPT

P value

P value

31.22 ± 1.73

55.09 ± 1.58

< 0.001

SG

33.12 ± 2.80

53.92 ± 2.56

< 0.001

0.028

RYGB

30.07 ± 2.21

55.80 ± 2.04

< 0.001

Abbreviation: PFPT, Pelvic Floor Physical Therapy.


P value 0.05, Wilcoxon Signed-Rank Test (within-group); Mann–Whitney U Test (between-group).




Discussion

Our findings indicate that the prevalence of proctological symptoms following BS is substantial, affecting 29% of patients within the evaluated period, with constipation representing the most frequently reported anorectal dysfunction. Despite the clinical relevance of these symptoms, referral to PFPT programs remains low; only 13.42% of patients in this study were directed to such interventions. Notably, our data demonstrates that physiotherapeutic treatment is effective in significantly alleviating symptoms, particularly by reducing constipation.

Constipation is a well-documented and clinically significant adverse effect associated with surgical interventions for obesity.[10] [14] [15] In our present study, constipation was the most prevalent proctological complaint, reported by 42% of patients after bariatric surgery. This outcome is likely multifactorial: caloric restriction (a cornerstone of these procedures) diminishes the physiological stimulation of colonic motility. Additionally, the marked reduction in dietary fiber intake commonly observed in intensive weight-loss regimens contributes to reduced stool bulk and compromised bowel function.[16] [17] Notably, we also observed a higher prevalence of constipation in patients who underwent RYGB compared with those who received SG, likely due to alterations in gastrointestinal transit and fluid absorption associated with the RYGB procedure.[10]

Among the various therapeutic approaches for managing constipation and other anorectal symptoms,[18] [19] [20] PFPT, particularly when employing EMG as a diagnostic and therapeutic tool,[21] [22] stands out. Remarkably, in our study, only 64 patients (13.42%) were referred to PFPT, clearly reflecting a lack of awareness among physicians regarding the effectiveness of pelvic floor physiotherapy in addressing these dysfunctions. Furthermore, the scientific literature on anorectal disorders following BS remains scarce,[14] [15] and no previous studies have been identified that evaluate the efficacy of PFPT in this specific clinical context.

Of the patients referred to PFPT, 54 underwent MAR, a common diagnostic method for anorectal and PFD, widely used in both clinical and scientific practice to provide an objective assessment of anal sphincter function, anorectal reflexes, and sensitivity in this region.[23] For the first time, we observed that, from this population of post-surgical patients referred to PFPT, the majority exhibited hypotonic resting pressure and contraction pressure, with no evidence of anismus ([Table 3]). This suggests that post-surgical patients referred for PFPT presented with internal anal sphincter weakness, external anal sphincter and/or PFMD, but without evacuation dyssynergy. Weakness of the anal sphincter can compromise the effectiveness of evacuation, leading to constipation symptoms. Furthermore, reduced rectal sensitivity and altered compliance may also contribute to evacuation difficulties.[24] Also, the MAR results reveal statistically significant differences in anorectal pressures between the surgical groups analyzed ([Table 3]). BS induces changes in intestinal habits, with characteristics varying according to the type of procedure performed,[10] corroborating the findings obtained in our study. These results emphasize the importance of a personalized pelvic floor functional approach in the post-operative follow-up of patients who have undergone BS, particularly in those with persistent proctological symptoms.

To our knowledge, this is the first study to demonstrate the effectiveness of PFPT using EMG biofeedback in the management of proctological symptoms following BS. Patients presenting with moderate to severe constipation showed a significant reduction in symptoms after completing PFPT, underscoring the clinical value of this intervention. These findings are consistent with a previous study reporting that PFM training enhances anal sphincter strength and control.[13] EMG biofeedback plays a key role in optimizing the coordination between external anal sphincter contraction and rectal sensory responses, thereby enhancing both defecatory function and fecal continence.[25] Its efficacy has also been supported in the treatment of constipation,[26] reinforcing its value as a core rehabilitation strategy for patients with pelvic floor dysfunction, regardless of the bariatric technique employed.

In addition, our results suggest that although PFPT appears broadly beneficial in the postoperative rehabilitation of bariatric patients, the underlying mechanisms driving variability in treatment response across surgical types remain poorly understood. This gap represents a promising avenue for future research. Investigating how different bariatric techniques influence pelvic floor neuromuscular function, particularly through EMG-based metrics, may provide critical insights for tailoring postoperative rehabilitation and optimizing functional outcomes.


Conclusion

This study provides novel evidence supporting the effectiveness of PFPT, particularly with EMG biofeedback, in alleviating proctological symptoms, primarily constipation, after BS. Despite the high prevalence of anorectal dysfunction in this population, referral rates to PFPT remain disappointingly low. Our findings reinforce the need for greater clinical awareness and routine incorporation of individualized pelvic floor rehabilitation into postoperative care protocols. Future studies should further explore the differential impact of bariatric procedures on pelvic floor function and optimize therapeutic strategies based on objective functional assessments.



Conflict of Interest

The authors declare that they have no conflict of interest.


Address for correspondence

Maria Lúcia Bonfleur, PhD
Center for Biological and Health Sciences, Western Paraná State University (UNIOESTE)
Cascavel, PR, Brazil CEP: 85819-110

Publikationsverlauf

Eingereicht: 09. April 2025

Angenommen: 21. Juli 2025

Artikel online veröffentlicht:
25. September 2025

© 2025. The Author(s). This is an open access article published by Thieme under the terms of the Creative Commons Attribution 4.0 International License, permitting copying and reproduction so long as the original work is given appropriate credit (https://creativecommons.org/licenses/by/4.0/)

Thieme Revinter Publicações Ltda.
Rua Rego Freitas, 175, loja 1, República, São Paulo, SP, CEP 01220-010, Brazil

Bibliographical Record
Marcieli Schuster, Caroline Dresch Sabadin, Cláudia Silveira Viera, Maria Lúcia Bonfleur. Pelvic Floor Physiotherapy with Biofeedback Improves Constipation in Post-Bariatric Surgery Patients. Journal of Coloproctology 2025; 45: s00451811261.
DOI: 10.1055/s-0045-1811261