Abstract
Objectives
To investigate and analyze the factors affecting postoperative urinary retention (POUR) after pelvic floor reconstruction, and to construct and validate a risk prediction model.
Methods
This retrospective cohort study included 258 pelvic floor reconstruction patients (2023–2024) from a Southwest China tertiary hospital. Patients were classified into POUR and non-POUR groups and split 7:3 into training and internal validation cohorts. Predictors were identified through univariate analysis and multivariate logistic regression analysis to construct a Nomogram model. Receiver Operating Characteristic (ROC) curves, calibration curves, and the Hosmer–Lemeshow test evaluated the model's differentiation, calibration, goodness-of-fit, and predictive performance.
Results
Independent POUR risk factors were: urinary retention history (OR = 10.008, 95% CI 1.368–73.195, P = 0.023), heart disease (OR = 14.416, 95% CI 2.872–72.376, P = 0.001), number of vaginal deliveries (OR = 1.569, 95% CI 1.076–2.289, P = 0.019), and maximal urinary flow rate (OR = 0.845, 95% CI 0.76–0.94, P = 0.002). The AUC values of the training cohort and internal validation cohort were 0.812 (95% CI 0.726–0.899) and 0.822 (95% CI 0.703–0.941), respectively. Calibration curves indicated good agreement between predicted and observed values, and the Hosmer–Lemeshow test demonstrated high predictive accuracy (P > 0.05).
Conclusions
The nomogram model effectively predicts POUR risk, aiding early perioperative identification of high-risk patients.
Supplementary Information
The online version contains supplementary material available at 10.1186/s40001-026-03853-8.
Keywords: Urinary retention, Pelvic floor reconstruction, Postoperative complications, Risk prediction, Nomogram
Introduction
Urinary retention following pelvic floor reconstruction has been the focus of many researchers, with reported rates ranging from 7 to 30% among postoperative patients, as determined by residual urine volume after bladder emptying before hospital discharge [1]. Improperly managed urinary retention can lead to renal insufficiency from ureteral reflux or necessitate emergency admissions postoperatively. This condition can cause psychological distress and emotional trauma for women requiring indwelling or intermittent catheterization, it carries a degree of tube burden. [2] Urinary retention is associated with urinary tract infections, prolonged hospital stays, and increased healthcare costs [3].
The International Continence Society and the International Association of Gynecologic Urology define urinary retention as “the complaint of inability to pass urine despite persistent effort” [4]. It’s a common issue in clinical practice. Previous studies have generally categorized the assessment of urinary retention following pelvic floor reconstruction into two methods: measuring residual urine after self-voiding and conducting retrograde voiding tests. Chapman et al. [5] concluded that a residual urine volume exceeding 100 ml after self-voiding is indicative of urinary retention, while other studies have used criteria of greater than 1/2 to 1/3 of normal bladder capacity or retrograde filling volume [6].
Nonetheless, researchers are still actively pursuing several explorations in the field of urinary retention in an attempt to mitigate the adverse effects of this common complication on patients. Recent findings indicate that urinary retention following pelvic floor reconstruction may be associated with factors, such as age, BMI, type of anesthesia, type of surgery, and others [7]. To facilitate early identification of urinary retention following pelvic floor reconstruction, researchers have explored risk prediction models [8–10]. Anglim et al. [8] developed an online risk calculator for POUR following pelvic floor surgery, using a retrospective cohort study to analyze perioperative characteristics that increase risk. Zhang et al. [9] created a risk prediction model for early and late urinary retention following pelvic floor reconstruction by reviewing perioperative data from patients undergoing gynecologic urologic surgery over 6 years at a single center. Li et al. [10] developed a risk prediction nomogram for POUR by investigating potential preoperative and intraoperative risk factors. However, these tools have their advantages and limitations. For instance, Anglim et al.’s [8] risk calculator is user-friendly and facilitates early communication about POUR among physicians. However, this online tool requires internet access and is limited in its applicability, specifically for patients undergoing vaginal surgery; its relevance for laparoscopic pelvic floor surgery patients remains uncertain. The applicability of the tool to patients undergoing laparoscopic pelvic floor surgery remains unknown. Li et al.’s [10] nomogram visually represents the risk associated with each factor, making it user-friendly in clinical scenarios. However, it does not account for the varying degrees of risk related to urodynamic outcomes among the factors, as noted in Zhang et al.'s [9] study.
Urinary retention in these women following pelvic floor reconstruction may be potentially preventable. Cao et al. [11] applied low-frequency electrical stimulation therapy to the bladder area and sacral region in the early postoperative period. Results indicated that this approach improved urinary flow rates and reduced the incidence of urinary retention following pelvic floor reconstruction. Chapman et al. [5] investigated the use of tamsulosin, an alpha-adrenergic receptor blocker, during the perioperative period to relax the bladder outlet and effectively prevent POUR. Nonetheless, these investigations have predominantly concentrated on postoperative interventions and have not integrated the screening of high-risk populations into clinical practice. Furthermore, there is a notable lack of recommendations and effective tools for managing common complications following pelvic floor reconstruction. Therefore, this study aimed to develop and validate an easy-to-use assessment tool based on common risk predictors to identify patients at high risk of urinary retention following pelvic floor reconstruction and to inform early prevention of the development of POUR.
Materials and methods
Design and sample
This study is a single-center, retrospective cohort analysis of patients undergoing pelvic floor reconstruction for pelvic organ prolapse at a tertiary hospital in Southwest China from February 2023 to February 2024.
Inclusion criteria: (a) age ≥ 18 years; (b) confirmed diagnosis of pelvic organ prolapse according to the Chinese Guidelines for the Diagnosis and Treatment of Pelvic Organ Prolapse (2020 Edition) [12]; (c) severity of prolapse classified as II degree or higher in the main preoperative diagnosis.
Exclusion criteria: (a) previous history of pelvic floor reconstruction surgery; (b) repair without mesh-augmented; (c) incomplete/inaccessible medical records.
The final number of variables entered into the multivariable logistic regression model was nine, and the sample size was calculated using the logistic independent variable event number method. Based on the findings of Deffieux et al. [1], the incidence of urinary retention following pelvic floor reconstruction was 30%. Considering a 10% attrition rate, the minimum required sample size was calculated to be 165. Our final analytic cohort comprised 190 patients, which exceeded this a priori estimate. Participants were allocated to the training cohort (134 cases) and the internal validation cohort (56 cases) in a 7:3 ratio. A waiver of informed consent was obtained from the Ethics Committee as the study did not involve direct patient contact. The study protocol was approved by the Ethical Committee for Clinical Research and Animal Trials at the Chongqing Health Center for Women and Children (2024-030 on May 31, 2024).
Data collection and definitions
The data collectors are research team member, with a bachelor's degree and over 2 years of experience in gynecological pelvic floor urology was trained and tested before data collection. Team members communicated openly to resolve issues quickly. One researcher extracted patient, clinical, and urodynamic data from electronic records, while another verified and corrected any discrepancies to ensure accuracy.
In the study, nurses removed urinary catheters on the second postoperative day, monitored spontaneous urination, and measured post-void residual (PVR) with a portable bladder scanner within 10 min. Urinary retention was defined as a PVR over 100 ml, a threshold aligned with clinical practice and previous literature [13].
Selection of predictors
This study identified 14 predictors of urinary retention following pelvic floor reconstruction using interview methods, literature reviews, and expert consultations, leading to the development of a risk factor questionnaire. The questionnaire can be found in Appendix S1.
A researcher conducted semi-structured interviews with six experienced doctors and nurses in the pelvic floor urology ward to identify influencing factors until reaching information saturation. A literature review was also conducted to gather relevant studies, which, along with the interview results, helped create an initial questionnaire. Seven experts from various regions in China, meeting specific experience and title criteria, participated in two rounds of Delphi consultations to finalize the questionnaire on risk factors for urinary retention after pelvic floor reconstruction.
The questionnaire comprised two sections: (i) patient demographics such as age, BMI, pregnancy and childbirth history, and menstruation and (ii) clinical information and urodynamic test results, including comorbidities (e.g., hypertension, diabetes, and heart disease), major surgery categories, bladder pressure and volume, and uroflowmetry.
Statistical methods
Data processing and graphing were performed using SPSS (Version 26.0; IBM, Armonk, NY, USA) and R software (Version 4.3.2; R Foundation for Statistical Computing, Vienna, Austria). Data completeness was assessed prior to analysis. Among the 190 patients included, sporadic missing values were present in a few continuous variables. Given that the missing proportion for each variable was below 5%, we employed mean imputation to handle these missing values, replacing them with the arithmetic mean of the available cases for that variable. Normally distributed continuous measures were reported as mean ± standard deviation. Non-normally distributed measures were reported as medians and quartiles, while counts data were expressed as frequencies and percentages. Group comparisons were conducted using the independent samples t-test (for normally distributed data), the Wilcoxon–Mann–Whitney test (for non-normally distributed data) or the χ2 test (for categorical data). Logistic regression analysis was conducted to identify influencing factors, while the ROCR and rms statistical packages in R were employed for Nomogram modeling and evaluation. The area under the curve (AUC) of the model's performance characteristics evaluates its discrimination ability; an AUC closer to 1 indicates stronger discrimination. The calibration curve and the Hosmer–Lemeshow (H–L) test assess the model's calibration, with better prediction reflected by greater alignment between the predicted and ideal curves. An H–L test result with P > 0.05 indicates good model fit. The difference was considered statistically significant at P < 0.05.
The data set was randomly split into a training cohort (70%, n = 134) for model development and an internal validation cohort (30%, n = 56). To assess the stability and precision of the model's performance estimate in an independent but similarly sourced sample, we performed a bootstrap resampling procedure (with 1000 replicates) specifically on this internal validation cohort. This provides a robust confidence interval for the performance metrics obtained from the validation set, reflecting how the performance might vary across different samples drawn from the same underlying population.
Results
A total of 258 female patients underwent pelvic floor reconstruction at this institution for prolapse classified as II degree or greater. Based on the pre-defined criteria (Sect. “Design and sample”), 68 patients were excluded: nine had a history of previous pelvic floor reconstruction, 48 had incomplete key clinical information (primarily missing the preoperative urodynamic report), and 11 did not undergo mesh-augmented repair. Consequently, 190 patients were included in the final analysis (Fig. 1). Among these 190 patients, sporadic missing values were observed in a few continuous variables (all with missingness < 5%), including height, urinary flow rate parameters, and storage/voiding-phase parameters. As per the statistical plan (Sect. “Statistical methods”), these were addressed using mean imputation. All participants were female, with a mean age of 58.60 ± 11.84 years.
Fig. 1.
Flow chart of the constructing of the cohort for developing a postoperative urinary retention nomogram
Thirty-eight patients developed POUR and were assigned to the case cohort, while 152 patients without urinary retention were assigned to the control cohort. These patients were then randomly split into a training cohort and an internal validation cohort in a 7:3 ratio, ensuring both case and control patients were proportionally represented in each. Thus, the training cohort comprised 134 subjects, and the internal validation cohort 56 subjects. The baseline characteristics of both cohorts are shown in Table 1.
Table 1.
Basic characteristics of the study cohort
| Variables | Training set (N = 134) | Validation set (N = 56) | p value |
|---|---|---|---|
| Age, years (%) | 0.817 | ||
| < 60 | 79 (59.0) | 32 (57.1) | |
| ≥ 60 | 55 (41.0) | 24 (42.9) | |
| Pregnancy, n (%) | 0.081 | ||
| 1 | 6 (4.5) | 2 (3.6) | |
| 2 | 30 (22.4) | 5 (8.9) | |
| ≥ 3 | 98 (73.1) | 49 (87.5) | |
| Vaginal delivery, n (%) | 0.520 | ||
| 0 | 1 (0.7) | 0 (0.0) | |
| 1 | 50 (37.3) | 16 (28.6) | |
| 2 | 49 (36.6) | 26 (46.4) | |
| ≥ 3 | 34 (25.4) | 14 (25.0) | |
| Menstrual status, n (%) | 0.413 | ||
| Normal | 33 (24.6) | 17(30.4) | |
| Menopause | 101 (75.4) | 39 (69.6) | |
| History of urinary retention, n (%) | 0.606 | ||
| Yes | 7 (5.2) | 4 (7.1) | |
| No | 127 (94.8) | 52 (92.9) | |
| Hypertension, n (%) | 0.530 | ||
| Yes | 37 (27.6) | 18 (32.1) | |
| No | 97 (72.4) | 38 (67.9) | |
| Diabetes, n (%) | 0.945 | ||
| Yes | 21 (15.7) | 9 (16.1) | |
| No | 113 (84.3) | 47 (83.9) | |
| Heart disease, n (%) | 0.397 | ||
| Yes | 9 (6.7) | 2 (3.6) | |
| No | 125 (93.3) | 54 (96.4) | |
| Hysterectomy, n (%) | 0.191 | ||
| Yes | 9 (6.7) | 7 (12.5) | |
| No | 125 (93.3) | 49 (87.5) | |
| Urinary control surgery or pelvic floor reconstruction surgery, n (%) | 0.128 | ||
| Yes | 126 (94.0) | 7 (12.5) | |
| No | 8 (6.0) | 49 (87.5) |
Analysis of influential factors and construction of nomograms
Univariate analysis and multifactorial logistic regression analysis were conducted on the questionnaire variables assessing risk factors for urinary retention following pelvic floor reconstruction. Univariate analysis revealed significant differences between the two groups regarding history of urinary retention, heart disease, post-void residual urine, mean urinary flow rate, maximum urinary flow rate, maximum pressure of the detrusor muscle, age, and number of deliveries (P < 0.05) (Table 2). The assignment of independent variables to the above factors (Appendix S2) and multifactorial logistic regression analysis showed that there was a significant difference between the two groups in terms of history of urinary retention, heart disease, number of vaginal deliveries, and maximal urinary flow rate (P < 0.05). These four variables were identified as independent predictors of POUR in female patients following pelvic floor reconstruction (Table 3). A nomogram was constructed based on these four variables (Fig. 2).
Table 2.
Univariate analysis of urinary retention in patients after pelvic floor reconstruction (N = 134)
| Variables | Non-POUR (N = 107) | POUR (N = 27) | Statistical results | p value |
|---|---|---|---|---|
| Age, years (%) | 4.6361 | 0.031 | ||
| < 60 | 68 (63.6) | 11 (40.7) | ||
| ≥ 60 | 39 (36.4) | 16 (59.3) | ||
| BMI [kg/m2, ] | 24.72 ± 2.59 | 24.51 ± 3.65 | 0.3412 | 0.734 |
| Pregnancy [M (P25, P75)] | 3(2, 5) | 4(2, 5) | − 0.1053 | 0.917 |
| Vaginal delivery [M (P25, P75)] | 2(1, 2) | 2(2, 4) | − 2.3693 | 0.033 |
| Menstrual status, n (%) | 1.1541 | 0.283 | ||
| Normal | 29 (27.1) | 4 (14.8) | ||
| Menopause | 78 (72.9) | 23 (85.2) | ||
| History of urinary retention, n (%) | 8.9421 | 0.003 | ||
| Yes | 2 (1.9) | 5 (18.5) | ||
| No | 105 (98.1) | 22 (81.5) | ||
| Hypertension, n (%) | 1.5031 | 0.220 | ||
| Yes | 27 (25.2) | 10 (37.0) | ||
| No | 80 (74.8) | 17 (63.0) | ||
| Diabetes, n (%) | 1.0981 | 0.295 | ||
| Yes | 15 (14.0) | 6 (22.2) | ||
| No | 92 (86.0) | 21 (77.8) | ||
| Heart disease, n (%) | 5.3431 | 0.021 | ||
| Yes | 4 (3.7) | 5 (18.5) | ||
| No | 103 (96.3) | 22 (81.5) | ||
| History of hysterectomy, n (%) | 0.3491 | 0.555 | ||
| Yes | 6 (5.6) | 3(11.1) | ||
| No | 101 (94.4) | 24 (88.9) | ||
| History of urinary control surgery or pelvic floor reconstruction surgery, n (%) | 0.011 | 0.919 | ||
| Yes | 7(6.5) | 1 (3.7) | ||
| No | 100 (93.5) | 26 (96.3) | ||
| Uterine prolapse, n (%) | 3.5201 | 0.318 | ||
| None/Degree I | 35 (32.7) | 12 (44.4) | ||
| Degree II | 32 (29.9) | 5 (18.5) | ||
| Degree III | 35 (32.7) | 7 (25.9) | ||
| Degree IV | 5 (4.7) | 3 (11.1) | ||
| Anterior wall prolapse, n (%) | 3.3261 | 0.344 | ||
| None/Degree I | 15 (14.0) | 2 (7.4) | ||
| Degree II | 32 (29.9) | 5 (18.5) | ||
| Degree III | 55 (51.4) | 19 (70.4) | ||
| Degree IV | 5 (4.7) | 1 (3.7) | ||
| Posterior wall prolapse, n (%) | 6.7271 | 0.081 | ||
| None/Degree I | 87 (81.3) | 16 (59.3) | ||
| Degree II | 11 (10.3) | 5 (18.5) | ||
| Degree III | 8 (7.5) | 4 (14.8) | ||
| Degree IV | 1 (0.9) | 2 (7.4) | ||
| Urinary incontinence, n (%) | 0.0011 | 0.971 | ||
| Yes | 44 (41.1) | 11 (40.7) | ||
| No | 63 (58.9) | 16 (59.3) | ||
| Operation of urinary incontinence, n (%) | 2.4921 | 0.114 | ||
| Implementation | 25 (23.4) | 2 (7.4) | ||
| Not implemented | 82 (76.6) | 25 (92.6) | ||
| Anterior pelvic reconstruction, n (%) | 1.5701 | 0.210 | ||
| Yes | 57 (53.3) | 18 (66.7) | ||
| No | 50 (46.7) | 9 (33.3) | ||
| Middle pelvic reconstruction, n (%) | 0.8171 | 0.366 | ||
| Yes | 54 (50.5) | 11 (40.7) | ||
| No | 53 (49.5) | 16 (59.3) | ||
| Posterior pelvic reconstruction, n (%) | 3.0011 | 0.083 | ||
| Yes | 8 (7.5) | 5 (18.5) | ||
| No | 99 (92.5) | 22 (81.5) | ||
| Total pelvic reconstruction, n (%) | 0.0021 | 0.961 | ||
| Yes | 13 (12.1) | 4 (14.8) | ||
| No | 94 (87.9) | 23 (85.2) | ||
| TVT-E, n (%) | 3.5101 | 0.061 | ||
| Yes | 23 (21.5) | 1 (3.7) | ||
| No | 84 (78.5) | 26 (96.3) | ||
| TVT-O, n (%) | 01 | 1.000 | ||
| Yes | 2 (1.9) | 1 (3.7) | ||
| No | 105 (98.1) | 26 (96.3) | ||
| Unexcised uterus, n (%) | 01 | 1.000 | ||
| Yes | 90 (84.1) | 23 (85.2) | ||
| No | 17 (15.9) | 4 (14.8) | ||
| Nature—maximum urinary flow rate [ml/s, M (P25, P75)] | 20 (14,26) | 17 (11,24) | − 1.5713 | 0.116 |
| Nature—volume of urine [ml, M (P25, P75)] | 237 (165,329) | 201 (168,283) | − 0.8243 | 0.410 |
| Nature—mean urinary flow rate [ml/s, M (P25, P75)] | 10 (6,13) | 7 (5,10) | − 2.0183 | 0.044 |
| Postvoid residual [ml, M (P25, P75)] | 0 (0,0) | 0 (0,20) | − 2.9083 | 0.004 |
| Time to maximum urinary flow rate [s, M (P25, P75)] | 6.26 (4.8,9.2) | 5.7 (3.3,8) | − 1.7283 | 0.084 |
| Urine flow time [s, M (P25, P75)] | 25.35 (17,37.1) | 30.25 (21.2,39) | − 1.1093 | 0.267 |
| Time to urinate [s, M (P25, P75)] | 26.05 (17.05,40.05) | 31.25 (21.2,40.2) | − 1.3533 | 0.176 |
| Maximum bladder capacity [ml, M (P25, P75)] | 312.5 (303.6,334.6) | 313.2 (305.6,325.5) | − 0.3473 | 0.729 |
| Phase of voiding—maximum urinary flow rate [ml/s, M (P25, P75)] | 16 (12,20) | 14 (11,15) | − 2.9663 | 0.003 |
| Phase of voiding—mean urinary flow rate [ml/s, ] | 7.81 ± 4.41 | 6.07 ± 2.91 | 2.4722 | 0.016 |
| Phase of voiding—volume of urine [ml, M (P25, P75)] | 327 (263,373) | 302 (197,339) | − 1.6613 | 0.097 |
| Bladder capacity for initial urination [ml, ] | 170.55 ± 42.25 | 170.06 ± 30.32 | 0.0562 | 0.955 |
| Initial bladder pressure during urination [cmH2O, M (P25, P75)] | 23.82 (13.7,39.02) | 31.57 (17.23,52) | − 1.4283 | 0.153 |
| Maximum detrusor pressure [cmH2O,M (P25, P75)] | 23.93 (16.59,32.78) | 27.03 (21.86,44.37) | − 2.1193 | 0.034 |
| Bladder pressure at maximum urinary flow rate [cmH2O, M (P25, P75)] | 26.41 (15.84,47.55) | 30.2 (19.01,43.09) | − 0.7903 | 0.429 |
| Detrusor pressure at maximum urinary flow rate [cmH2O, M (P25, P75)] | 18.55 (11.26,25.52) | 18.93 (14.61,33.18) | − 1.2593 | 0.208 |
| Abdominal leak point pressures, n (%) | 0.0011 | 0.980 | ||
| > 60 cmH2O | 79 (73.8) | 20 (74.1) | ||
| < 60 cmH2O | 28 (26.2) | 7 (25.9) |
POUR: postoperative urinary retention; TVT-E: transobturator vaginal tape exact; TVT-O: tension-free vaginal tape obturator
1χ2 value
2t value
3Z value
Table 3.
Multivariable logistic regression analysis of POUR predictors
| Variables | p value | OR | 95% CI | |
|---|---|---|---|---|
| Lower limit | Upper limit | |||
| Constant | 0.633 | 0.669 | ||
| History of urinary retention | 0.023 | 10.008 | 1.368 | 73.195 |
| Heart disease | 0.001 | 14.416 | 2.872 | 72.376 |
| Phase of voiding—maximum urinary flow rate | 0.002 | 0.845 | 0.760 | 0.940 |
| Vaginal delivery | 0.019 | 1.569 | 1.076 | 2.289 |
Fig. 2.
Nomogram models to predict the risk of urinary retention in patients after pelvic floor reconstruction
Performance of nomogram for POUR in women following pelvic floor reconstruction
The predictive performance of the nomogram for urinary retention following pelvic floor reconstruction is illustrated in Fig. 3. The results indicated AUC of 0.812 (95% CI 0.726–0.899) for the training cohort (Fig. 3a). For the internal validation cohort, the AUC was 0.822 (95% CI 0.703–0.941) derived from 1000 bootstrap resamples of this cohort (Fig. 3b). The calibration curves of the predictive model demonstrated good agreement between predicted and measured values (Appendix S3). The Hosmer–Lemeshow test results indicated a χ2 of 9.921 (P = 0.271) for the training cohort and χ2 of 2.274 (P = 0.971) for the internal validation cohort, suggesting that the predictive probability of this model for POUR has a high degree of calibration with a predicted probability similar to the actual probability.
Fig. 3.
The receiver operating characteristic curve (ROC) performance of the nomogram.a. In the training cohort, the AUC was 0.812. b In the internal validation cohort, the AUC was 0.822.
Discussion
This study used a retrospective cohort design to create and validate a simple, effective predictive tool. It identified urinary retention history, heart disease, number of vaginal deliveries, and maximum urine flow rate as key factors affecting POUR in women after pelvic floor reconstruction. The model showed strong accuracy and reliability in predicting urinary retention post-surgery.
Previous studies have linked a history of urinary retention to an increased risk of POUR in various surgical settings, such as hernia repair and joint replacement [14, 15]. In addition, a meta-analysis confirmed that preoperative PVR predicts POUR in patients undergoing surgery for pelvic organ prolapse [16]. Consistent with these reports and supported by our findings, a history of urinary retention is also associated with an increased risk of POUR following pelvic floor reconstruction. In this study, a history of urinary retention—encompassing self-reported symptoms, postpartum retention, or ultrasound-detected retention—was confirmed as an independent predictor of POUR after reconstruction. Patients with such a history had an odds ratio of 10.008 for developing POUR compared to those without. While the observed association is strong, the precision of this estimate is limited, as reflected in the wide confidence interval. This association suggests that a prior history of retention may serve as a marker for underlying bladder dysfunction or other risk factors. Therefore, thorough preoperative assessment of urinary retention history—including when feasible, PVR measurement via bladder ultrasound—is recommended. Such evaluation can enhance patient counseling and support targeted preventive strategies to reduce postoperative complications.
The risk of urinary retention after pelvic floor reconstruction is 14.416 times higher in heart disease patients than in the general population. This finding aligns with the established understanding that comorbidities influence postoperative outcomes. Ripperda et al. [17] reported the Charlson Comorbidity Index predicts—which specifically includes a history of myocardial infarction and congestive heart failure—POUR after urogynecology surgery, with a 41% increase in the risk of voiding test failure when the index rises from 0 to 1. This association is also supported by Roadman et al., [3] who observed a significant link between congestive heart failure and POUR in patients undergoing laparoscopic sleeve gastrectomy. The underlying mechanism may involve several pathways. Cardiac conditions such as coronary artery disease, heart valve disease, and arrhythmias often lead to heart failure, with patients using diuretics for edema, which can influence urinary dynamics. Furthermore, the use of cholinesterase inhibitors (e.g., Neostigmine) to facilitate bladder emptying is typically avoided in cardiac patients after pelvic floor surgery due to potential cardiovascular risks. It is crucial to emphasize that these findings demonstrate observational associations, not causality. Therefore, while suggestive, the role of medication profiles or cardiac pathophysiology requires confirmation through targeted research.
This study confirmed that the number of vaginal deliveries is an independent predictor of POUR following pelvic floor reconstruction. Vaginal delivery itself is a recognized event that causes acute injury to the pelvic floor structures, often due to factors, such as fetal compression and instrumental assistance, and is a common cause of postpartum urinary retention [18]. Multiple deliveries may subject the pelvic floor to repeated, cumulative damage. This may be linked to the participants' average age of 58.6 years and their childbirths occurring 30–40 years ago (in the 1970s–1980s), a time when postpartum pelvic floor rehabilitation was less emphasized. For instance, standardized management promoted by associations such as the European Biofeedback Association began only about 30 years ago, while systematic promotion of similar rehabilitation protocols in China was not implemented until after 2008 [19]. Under these socioeconomic circumstances, many Chinese women, who often served as primary household laborers, resumed heavy physical work shortly after childbirth without access to scientific rehabilitation guidance. Consequently, the cumulative damage from multiple vaginal deliveries may not have been adequately repaired, potentially leading to long-term functional weakening of the pelvic floor support system. Thus, a history of multiple births can be viewed as a marker of reduced pelvic floor reserve. When combined with the trauma of reconstruction, this pre-existing vulnerability is likely associated with an increased risk of POUR.
This study identified that a low preoperative maximum urinary flow rate (Qmax) as a predictor of urinary retention following pelvic floor reconstruction. This finding aligns with existing evidence. Zhang et al. [20] followed patients who had pelvic organ prolapse repaired using a mesh for a mean of 4.2 years and found that a low mean preoperative urinary flow rate predicted postoperative voiding difficulties, similar to the results of this study. The predictive utility of a low Qmax has also been confirmed in other surgical contexts, such as following prostate enucleation [21] and among mid-urethral sling populations [22]. Notably, this is the first investigation to establish this predictive association specifically within a pelvic organ prolapse cohort, thereby extending the potential clinical applicability of this urodynamic parameter. The clinical significance is further underscored by health-economic considerations: managing postoperative urinary retention with catheterization imposes a substantial burden, with estimated costs ranging from $79 to $185 per episode, encompassing medical visits, treatment of urinary tract infections, transportation, nursing, education, and supplies [23]. Consequently, the preoperative identification of patients with a low Qmax could enable targeted preventive strategies, potentially reducing the need for postoperative catheterization and alleviating the associated clinical and economic burdens.
This study has several limitations that should be considered when interpreting the results. First, as a single-center, retrospective study, the findings and the prediction model may not be fully generalizable to other populations or healthcare settings. External validation in prospective, multi-center cohorts is required before widespread clinical application. Second, the model’s reliance on an urodynamic parameter presents a dual constraint. While this is a standard preoperative assessment in our tertiary center, it may not be routinely available in all institutions, limiting practical utility. Moreover, the exclusion of 46 patients due primarily to missing this key data may have introduced selection bias. Although a post-hoc comparison showed no significant differences in available baseline characteristics between excluded and included patients, the potential for bias cannot be ruled out. Finally, the sample size, particularly the number of POUR events (n = 38), is modest. This may affect the model’s stability and the precision of the effect estimates for strong predictors, as indicated by the wide confidence intervals.
Conclusion
In this study, we created and validated a prediction model, visualized with a nomogram, to identify urinary retention risk in women after pelvic floor reconstruction. The model uses easily accessible clinical indicators and effectively predicts high-risk patients, aiding in early care planning and symptom management during the perioperative period.
Supplementary Information
Author contributions
S H: writing—original draft, methodology, formal analysis, data curation, funding acquisition, software. LL Z: writing—review and editing, project administration, funding acquisition. CY S: Methodology, investigation, resources, supervision. YX G: data curation, methodology, visualisation, software. YJ H: investigation, resources. LB L: conceptualisation, formal analysis, resources, supervision. L D: conceptualisation, funding acquisition, resources, supervision, validation, project administration. All authors read and approved the manuscript.
Funding
This research was supported by the Science-Health Joint Medical Scientific Research Project of Chongqing (grant/award number: 2023MSXM055).
Data availability
The data that support the findings of this study are not openly available and are available from the corresponding author upon reasonable request.
Declarations
Ethics approval and consent to participate
The study protocol was approved by the Ethical Committee for Clinical Research and Animal Trials of the Chongqing Health Center for Women and Children (2024-030 on May 31, 2024).
Competing interests
The authors declare no competing interests.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Deffieux X, Perrouin-Verbe MA, Campagne-Loiseau S, et al. Diagnosis and management of complications following pelvic organ prolapse surgery using a synthetic mesh: French national guidelines for clinical practice. Eur J Obstet Gynecol Reprod Biol. 2024;294:170–9. [DOI] [PubMed] [Google Scholar]
- 2.Dieter AA, Wu JM, Gage JL, Feliciano KM, Willis-Gray MG. Catheter burden following urogynecologic surgery. Am J Obstet Gynecol. 2019;221(5):501–7. [DOI] [PubMed] [Google Scholar]
- 3.Roadman D, Helm M, Goldblatt MI, Kindel TL, Gould JC, Higgins RM. Postoperative urinary retention after bariatric surgery: an institutional analysis. J Surg Res. 2019;243:83–9. [DOI] [PubMed] [Google Scholar]
- 4.Haylen BT, Maher CF, Barber MD, et al. An international urogynecological association (IUGA) / international continence society (ICS) joint report on the terminology for female pelvic organ prolapse (POP). Int Urogynecol J. 2016;27(2):165–94. [DOI] [PubMed] [Google Scholar]
- 5.Chapman GC, Sheyn D, Slopnick EA, et al. Tamsulosin vs placebo to prevent postoperative urinary retention following female pelvic reconstructive surgery: a multicenter randomized controlled trial. Am J Obstet Gynecol. 2021;225(3):271–4. [DOI] [PubMed] [Google Scholar]
- 6.Sappenfield EC, Scutari T, O’Sullivan DM, Tulikangas PK. Predictors of delayed postoperative urinary retention after female pelvic reconstructive surgery. Int Urogynecol J. 2021;32(3):603–8. [DOI] [PubMed] [Google Scholar]
- 7.Anglim BC, Ramage K, Sandwith E, Brennand EA. Postoperative urinary retention after pelvic organ prolapse surgery: influence of peri-operative factors and trial of void protocol. BMC Womens Health. 2021;21(1):195. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Anglim BC, Tomlinson G, Paquette J, McDermott CD. A risk calculator for postoperative urinary retention (POUR) following vaginal pelvic floor surgery: multivariable prediction modelling. BJOG. 2022;129(13):2203–13. [DOI] [PubMed] [Google Scholar]
- 9.Zhang BY, Wong J, A KN, Lee T, Geoffrion R. Risk factors for urinary retention after urogynecologic surgery: a retrospective cohort study and prediction model. Neurourol Urodyn. 2021;40(5):1182–91. [DOI] [PubMed] [Google Scholar]
- 10.Li A, Zajichek A, Kattan MW, Ji XK, Lo KA, Lee PE. Nomogram to predict risk of postoperative urinary retention in women undergoing pelvic reconstructive surgery. J Obstet Gynaecol Can. 2020;42(10):1203–10. [DOI] [PubMed] [Google Scholar]
- 11.Cao TT, SunSun HXXL. The efficacy of transcutaneous electrical stimulation on urinary retention in women with pelvic reconstructive surgery. Chin J Clin Obstet Gynecol. 2019;20(02):104–7. [Google Scholar]
- 12.Urogynecology Subgroup CSOO. Chinese guideline for the diagnosis and management of pelvic orang prolapse (2020 version). Chin J Obstet Gynecol. 2020;55(5):300–6. [DOI] [PubMed] [Google Scholar]
- 13.Barbier H, Carberry CL, Karjalainen PK, et al. International urogynecology consultation chapter 2 committee 3: the clinical evaluation of pelvic organ prolapse including investigations into associated morbidity/pelvic floor dysfunction. Int Urogynecol J. 2023;34(11):2657–88. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Croghan SM, Mohan HM, Breen KJ, et al. Global incidence and risk factors associated with postoperative urinary retention following elective inguinal hernia repair: the retention of urine after inguinal hernia elective repair (RETAINER I) study. JAMA Surg. 2023;158(8):865–73. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Cha YH, Lee YK, Won SH, Park JW, Ha YC, Koo KH. Urinary retention after total joint arthroplasty of hip and knee: systematic review. J Orthop Surg. 2020;28(1):2309499020905134. [DOI] [PubMed] [Google Scholar]
- 16.Zhou L, Dai L, Hu S, et al. Risk factors for postoperative urinary retention following pelvic organ prolapse surgery: a meta-analysis. Sci Rep. 2025;15(1):41715. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Ripperda CM, Kowalski JT, Chaudhry ZQ, et al. Predictors of early postoperative voiding dysfunction and other complications following a midurethral sling. Am J Obstet Gynecol. 2016;215(5):651–6. [DOI] [PubMed] [Google Scholar]
- 18.Nutaitis AC, Meckes NA, Madsen AM, et al. Postpartum urinary retention: an expert review. Am J Obstet Gynecol. 2023;228(1):14–21. [DOI] [PubMed] [Google Scholar]
- 19.Zhu Lan LW. Current status of Chinese women’s pelvic floor rehabilitation. Chin J Fam Plan Gynecotokol. 2020;12(10):3–4. [Google Scholar]
- 20.Zhang L, Zhu L, Xu T, Liang S, Lang J. Postoperative voiding difficulty and mesh-related complications after Total Prolift System surgical repair for pelvic organ prolapse and predisposing factors. Menopause. 2015;22(8):885–92. [DOI] [PubMed] [Google Scholar]
- 21.Hsu YH, Hou CP, Weng SC, et al. Analysis of urinary retention after endoscopic prostate enucleation and its subsequent impact on surgical outcomes. World J Urol. 2024;42(1):305. [DOI] [PubMed] [Google Scholar]
- 22.Wheeler TN, Richter HE, Greer WJ, Bowling CB, Redden DT, Varner RE. Predictors of success with postoperative voiding trials after a mid urethral sling procedure. J Urol. 2008;179(2):600–4. [DOI] [PubMed] [Google Scholar]
- 23.Wang R, Tunitsky-Bitton E. Short-term catheter management options for urinary retention following pelvic surgery: a cost analysis. Am J Obstet Gynecol. 2022;226(1):101–2. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data that support the findings of this study are not openly available and are available from the corresponding author upon reasonable request.