Search

Introduction and hypothesis: Female stress urinary incontinence is highly prevalent, and synthetic midurethral sling placement is the most common type of anti-incontinence surgery performed in the USA. We aimed to identify risk factors associated with surgery used to treated vaginal mesh exposure after midurethral sling placement for stress urinary incontinence. Methods: We identified women who underwent anti-incontinence procedures from January 2002 through December 2012. Patients with vaginal mesh exposure undergoing surgical repair after midurethral sling placement were compared with a control group without mesh exposure in a 1:3 ratio. Patients with ObTape sling placement (Mentor Corporation) were excluded. Logistic regression models were used to evaluate associations between clinical risk factors and vaginal mesh exposure. Results: Overall, 2,123 patients underwent primary sling placement, with 27 (1.3 %) having vaginal mesh exposure necessitating surgical repair. Patients with mesh exposure were more likely to have undergone previous bariatric surgery ( P = 0.008), hemoglobin <13 g/dL ( P = 0.006), premenopausal status ( P = 0.008), age <50 years ( P = 0.001), and the retropubic approach to sling placement ( P = 0.03). Multivariate analysis identified these risk factors: previous bariatric surgery (odds ratio [OR], 7.0; 95 % CI, 1.1-61.4), retropubic approach (OR, 5.7; 95 % CI, 1.1-107.0), preoperative hemoglobin <13 g/dL (OR, 2.8; 95 % CI, 1.1-7.5), and premenopausal status (OR, 2.6; 95 % CI, 1.0-7.3). Among postmenopausal patients, those with mesh exposure were significantly more likely to receive preoperative estrogen therapy (OR, 12.4; 95 % CI, 2.7-57.8). Conclusions: Previous bariatric surgery, retropubic approach, premenopausal status, and lower preoperative hemoglobin were associated with a significantly increased risk of surgery for vaginal mesh exposure after midurethral sling placement. Recognizing these factors can improve preoperative patient counseling. [ABSTRACT FROM AUTHOR]

The purpose of the study was to determine the efficacy and safety of nonantimuscarinic treatments for overactive bladder. Medline, Cochrane, and other databases (inception to April 2, 2014) were used. We included any study design in which there were 2 arms and an n > 100, if at least 1 of the arms was a nonantimuscarinic therapy or any comparative trial, regardless of number, if at least 2 arms were nonantimuscarinic therapies for overactive bladder. Eleven reviewers double-screened citations and extracted eligible studies for study: population, intervention, outcome, effects on outcome categories, and quality. The body of evidence for categories of interventions were summarized and assessed for strength. Ninety-nine comparative studies met inclusion criteria. Interventions effective to improve subjective overactive bladder symptoms include exercise with heat and steam generating sheets (1 study), diaphragmatic (1 study), deep abdominal (1 study), and pelvic floor muscle training exercises (2 studies). Pelvic floor exercises are more effective in subjective and objective outcomes with biofeedback or verbal feedback. Weight loss with diet and exercise, caffeine reduction, 25-50% reduction in fluid intake, and pelvic floor muscle exercises with verbal instruction and or biofeedback were all efficacious. Botulinum toxin A improves urge incontinence episodes, urgency, frequency, quality of life, nocturia, and urodynamic testing parameters. Acupuncture improves quality of life and urodynamic testing parameters. Extracorporeal magnetic stimulation improves urodynamic parameters. Mirabegron improves daily incontinence episodes, nocturia, number of daily voids, and urine volume per void, whereas solabegron improves daily incontinence episodes. Short-term posterior tibial nerve stimulation is more efficacious than pelvic floor muscle training exercises and behavioral therapy for improving: urgency, urinary incontinence episodes, daily voids, volume per void, and overall quality of life. Sacral neuromodulation is more efficacious than antimuscarinic treatment for subjective improvement of overactive bladder and quality of life. Transvaginal electrical stimulation demonstrates subjective improvement in overactive bladder symptoms and urodynamic parameters. Multiple therapies, including physical therapy, behavioral therapy, botulinum toxin A, acupuncture, magnetic stimulation, mirabegron, posterior tibial nerve stimulation, sacral neuromodulation, and transvaginal electrical stimulation, are efficacious in the treatment of overactive bladder. [ABSTRACT FROM AUTHOR]

Objectives: To establish a prediction model for fetal congenital heart disease (CHD) that could facilitate selective fetal echocardiography screening for diabetic women, and to provide cost analysis of selective versus global screening of these women.Study Design: A historical cohort study included women with pre-gestational diabetes who received perinatal care at Mayo Clinic, Minnesota from 2007 through 2013. Anatomical scans and fetal echocardiograms were reviewed and charts for postnatal outcomes were abstracted. Logistic regression models were utilized to identify predictors of CHD. Cost of global versus selective screening was estimated using cost per case detected.Results: A cohort of 152 women was included. Abnormal anatomy scan was reported in five (3.3%). Among 51 (33.6%) who had fetal echocardiography, eight (15.7%) had abnormal findings. Postnatal echocardiography was conducted in 36 neonates, 12 had abnormalities. Only first trimester HbA1c was independent predictor with observed incidence of CHD of 1.8, 14.3, and 40% and HbA1c of <8, 8-10, and >10% respectively. Global screening strategy missed 22% of fetal cardiac abnormalities, while selective screening missed 33%.Conclusions: According to our probability model, selective screening based on HbA1c and/or anatomic surveys seems to be an accurate and cost effective strategy for prenatal diagnosis of CHD among diabetic pregnancies. [ABSTRACT FROM AUTHOR]

Study Objective: To identify intraoperative predictors of radiofrequency ablation (RFA) failure after adjusting for clinical risk factors.Design: A cohort study (Canadian Task Force II-2).Setting: An academic institution in the Upper Midwest.Patients: Data were retrospectively collected from medical records of women who underwent RFA and who had a postprocedure gynecologic assessment between April 1998 and December 2011.Interventions: RFA.Measurements and Main Results: The primary outcome was RFA failure, which was defined as hysterectomy, repeat ablation, synechiolysis, or treatment with gonadotropin-releasing hormone analogue for postablation pain or bleeding. Cox proportional hazards regression was used to test the predictability of intraoperative variables on RFA failure with adjustment for baseline predictors. We created an RFA index to capture the procedure duration divided by the uterine surface area. One thousand one hundred seventy-eight women were eligible. The median age at ablation was 44 years (interquartile range, 40-48 years), and the median parity was 2 (interquartile range, 2-3). Dysmenorrhea and prior tubal ligation were reported in 37.1% and 37.2% of women, respectively. After adjustment for baseline characteristics, intraoperative predictors of failure were uterine sounding length >10.5 cm (adjusted hazard ratio [HR] = 2.58; 95% confidence interval [CI], 1.31-5.05), uterine cavity length >6 cm (adjusted HR = 2.06; 95% CI, 1.30-3.27), uterine width >4.5 cm (adjusted HR = 2.06; 95% CI, 1.29-3.28), surface area >25 cm(2) (adjusted HR = 2.02; 95% CI, 1.26-3.23), procedure time <93 seconds (adjusted HR = 2.61; 95% CI, 1.25-5.47), and RFA index <3.6 (adjusted HR = 3.14; 95% CI, 1.70-5.77).Conclusion: Intraoperative parameters are predictive of long-term adverse outcomes of RFA independent of patient clinical characteristics. Uterine length, procedure duration, and RFA index are associated with unfavorable outcomes and thus could be used to optimize postprocedure patient counseling. [ABSTRACT FROM AUTHOR]

The contribution of genetic factors to the presence of an overactive bladder is recognized. This study aimed to (1) assemble and synthesize available data from studies assessing differential gene expression in patients with overactive bladder vs controls without overactive bladder and (2) determine possible correlations and functional pathways between genes. We searched PubMed, Ovid or Medline, and Wiley Cochrane Central Register of Controlled Trials databases between January 1, 2000, and December 15, 2021. Studies were included if gene expression was detected and quantified using molecular approaches performed on human bladder tissue specimens directly and excluded if the gene expression analysis was carried out from blood and urine specimens alone. A systematic review was completed to identify publications that reported differently expressed gene candidates among patients with overactive bladder vs healthy individuals. Gene networking connections and pathway analysis were performed employing Metascape software, where inputs were identified from our systematic review of differentially expressed genes in overactive bladder. A total of 9 studies were included in the final analysis and 11 genes were identified as being up-regulated (purinergic receptor P2X 2 [P2RX2], smoothelin [SMTN], growth-associated protein 43 [GAP43], transient receptor potential cation channel subfamily M member 8 [TRPM8], cadherin 11 [CDH1], gap junction protein gamma 1 [GJC1], cholinergic receptor muscarinic 2 [CHRM2], cholinergic receptor muscarinic 3 [CHRM3], and transient receptor potential cation channel subfamily V member 4 [TRPV4]) or down-regulated (purinergic receptor P2X 2 [P2RX3] and purinergic receptor P2X 5 [P2RX5]) in patients with overactive bladder. Gene network analysis showed that genes are involved in chemical synaptic transmission, smooth muscle contraction, blood circulation, and response to temperature stimulus. Network analysis demonstrated a significant genetic interaction between TRPV4, TRPM8, P2RX3, and PR2X2 genes. Outcomes of this systematic review highlighted potential biomarkers for treatment efficacy and have laid the groundwork for developing future gene therapies for overactive bladder in clinical settings. [ABSTRACT FROM AUTHOR]

Study Objective: To evaluate the feasibility and impact of levonorgestrel intrauterine system (LNG-IUS) on treatment failure after endometrial ablation (EA) in women with heavy menstrual bleeding (HMB) and dysmenorrhea at 4 years.Design: Cohort study (Canadian Task Force II-2).Setting: An academic institution in the upper Midwest.Patients: All women with HMB and dysmenorrhea who underwent EA with combined placement of LNG-IUS (EA/LNG-IUS cohort, 23 women) after 2005 and an historic reference group from women who had EA alone (EA cohort, 65 women) from 1998 through the end of 2005.Intervention: Radiofrequency EA, thermal balloon ablation, and LNG-IUS.Measurements and Main Results: The primary outcome was treatment failure defined as persistent pain, bleeding, and hysterectomy after EA at 4 years. The combined treatment failure outcome was documented in 2 patients (8.7%) in the EA/LNG-IUS group and 19 patients (29.2%) in the EA group with an unadjusted OR of .23 (95% CI, .05-1.08). After adjusting for known risk factors of failure, the adjusted OR was .19 (95% CI, .26-.88). None of the women who underwent EA/LNG-IUS had hysterectomy for treatment failure compared with 16 (24%) in the EA group (p = .009); postablation pelvic pain was documented in 1 woman (4.3%) in the EA/LNG-IUS group compared with 8 women (12.3%) in the EA group (p = .24). One woman in the EA/LNG-IUS group (4.3%) presented with persistent bleeding compared with 15 (23.1%) in the EA group (p = .059). Office removal of the intrauterine device was performed in 4 women with no complications.Conclusion: LNG-IUS insertion at the time of EA is feasible and can provide added benefit after EA in women with dysmenorrhea and HMB. [ABSTRACT FROM AUTHOR]

Study Objective: To evaluate the efficacy and safety of endometrial ablation (EA) for the treatment of abnormal uterine bleeding (AUB) associated with ovulatory dysfunction.Design: A retrospective cohort study (Canadian Task Force classification II-2).Setting: An academic medical center.Patients: Women with AUB who underwent EA during an 8-year period.Interventions: EA by radiofrequency or thermal balloon ablation techniques.Measurements and Main Results: Women with AUB were divided into 2 groups: irregular bleeding with ovulatory dysfunction (AUB-O) or regular heavy bleeding related to a primary endometrial disorder (AUB-E). Outcome measures included rates of amenorrhea and treatment failure (ie, need for reablation or hysterectomy). Outcomes were compared between groups using survival analyses and chi-square tests. Known confounders were adjusted for using Cox and logistic regression models. Five-year cumulative treatment failure rates were 11.7% (95% confidence interval [CI], 6.5%-16.9%) for AUB-O and 12.3% (95% CI, 8.4%-16.2%) for AUB-E (p = .62). The unadjusted hazard ratio for treatment failure was 0.87 (95% CI, 0.72-1.05, p = .16). After adjusting for known risk factors for failure, the hazard ratio was 1.48 (95% CI, 0.82-2.65, p = .19). The rates of amenorrhea were 11.8% for AUB-O and 13.8% for AUB-E with an unadjusted odds ratio of 0.84 (95% CI, 0.48-1.48, p = .55). After adjusting for factors for amenorrhea after EA, the odds ratio was 1.08 (95% CI, 0.62-1.84, p =.78). No pregnancies or endometrial cancers occurred after EA.Conclusion: EA is effective in women with AUB-O and can be used as an alternative to hysterectomy or in patients with contraindications to medical management of AUB-O. [ABSTRACT FROM AUTHOR]

Background and Aims: To report on and review the outcome after laparoscopic cervicoisthmic cerclage (LCC) and robotic-assisted laparoscopic cervicoisthmic cerclage. Methods: We reported on 4 cases of LCC and conducted a systematic review of the literature up to May 2012 to identify obstetric outcomes after LCC and robotic-assisted LCC. Results: The median age of our series was 35 years (range: 31-41) with median previous pregnancies 3.5 (2-5). All 4 women had successful obstetric outcomes with a median gestational age at delivery of 37 weeks (range: 36-38). The systematic review identified 25 studies (162 women underwent LCC and 3 had robotic-assisted LCC). In the interval LLC studies, the median age was 33 years (range: 22-42); with a median gestational age at delivery of 37 weeks (range: 34-38). For prophylactic LCC, the median age was 31 years (range: 27-41); with a gestational age at delivery of 37 weeks (range: 19-39). Two of the three robotic-assisted LCC procedures were done prophylactically. The median age was 27 years (range: 23-37) with a median gestational age at delivery of 37 weeks (range: 35-38). Conclusion: LCC is feasible during and in between pregnancies as well as in congenitally malformed uteri. Current evidence suggests that LCC might be of benefit in selected cases of cervical insufficiency with short cervices. Copyright © 2012 S. Karger AG, Basel [ABSTRACT FROM AUTHOR]

Background: Postoperative urinary retention is common after female pelvic reconstructive surgery. Alpha receptor antagonists can improve dysfunctional voiding by relaxing the bladder outlet and may be effective in reducing the risk of postoperative urinary retention.Objective: This study aimed to determine whether tamsulosin is effective in preventing postoperative urinary retention in women undergoing surgery for pelvic organ prolapse.Study Design: This was a multicenter, double-blind, randomized controlled trial between August 2018 and June 2020, including women undergoing surgery for pelvic organ prolapse. Patients were excluded from recruitment if they had elevated preoperative postvoid residual volume, history of postoperative urinary retention, or a contraindication to tamsulosin. Those who experienced cystotomy were excluded from analysis. Participants were randomized to a 10-day perioperative course of tamsulosin 0.4 mg vs placebo, beginning 3 days before surgery. A standardized voiding trial was performed on postoperative day 1. The primary outcome was the development of postoperative urinary retention, as defined by the failure of the voiding trial or subsequent need for catheterization to empty the bladder. Secondary outcomes included the rate of urinary tract infection and the impact on lower urinary tract symptoms as measured by the American Urological Association Symptom Index.Results: Of 119 patients, 57 received tamsulosin and 62 received placebo. Groups were similar in regard to demographics, preoperative prolapse and voiding characteristics, and surgical details. Tamsulosin was associated with a lower rate of postoperative urinary retention than placebo (5 patients [8.8%] vs 16 patients [25.8%]; odds ratio, 0.28; 95% confidence interval, 0.09-81; P=.02). The number needed to treat to prevent 1 case of postoperative urinary retention was 5.9 patients. The rate of urinary tract infection did not differ between groups. American Urological Association Symptom Index scores significantly improved after surgery in both groups (median total score, 14 vs 7; P<.01). Scores related to urinary stream improved more in the tamsulosin group than in placebo (P=.03).Conclusion: In this placebo-controlled trial, tamsulosin use was associated with a reduced risk of postoperative urinary retention in women undergoing surgery for pelvic organ prolapse. [ABSTRACT FROM AUTHOR]

Objective: To compare the reoperation rates for recurrent stress urinary incontinence (SUI) after retropubic and transobturator sling procedures.Methods: We conducted a retrospective cohort study of all women who underwent midurethral sling procedures at a single institution for primary SUI between 2002 and 2012. To minimize bias, women in the two groups were matched on age, body mass index, isolated compared with combined procedure, and preoperative diagnosis. The primary outcome was defined as reoperation for recurrent SUI. Secondary outcomes included intraoperative complications and mesh-related complications requiring reoperation after the index sling procedure.Results: We identified 1,881 women who underwent a sling procedure for primary SUI-1,551 retropubic and 330 transobturator. There was no difference between groups in any of the evaluated baseline variables in the covariate-matched cohort of 570 with retropubic slings and 317 with transobturator slings; results herein are based on the covariate-matched cohort. Women undergoing a transobturator sling procedure had an increased risk of reoperation for recurrent SUI compared with women undergoing a retropubic sling procedure (hazard ratio 2.42, 95% CI 1.37-4.29). The cumulative incidence of reoperation for recurrent SUI by 8 years was 5.2% (95% CI 3.0-7.4%) in the retropubic group and 11.2% (95% CI 6.4-15.8%) in the transobturator group. Women in the retropubic group had a significantly higher rate of intraoperative complications compared with women in the transobturator group (13.7% [78/570] vs 4.7% [15/317]; difference=9.0%, 95% CI for difference 5.3-12.6%); the majority of this difference was due to bladder perforation (7.0% [40/570] vs 0.6% [2/317]; difference=6.4%, 95% CI for difference 4.1-8.7%). The cumulative incidence of sling revision for urinary retention plateaued at 3.2% and 0.4% by 5 years in the two groups.Conclusion: Women with primary SUI treated with a retropubic sling procedure have significantly lower cumulative incidence of reoperation for recurrent SUI compared with women who were treated with a transobturator sling procedure. Retropubic slings were associated with a higher risk of sling revision for urinary retention. [ABSTRACT FROM AUTHOR]

Background: Postoperative pain control after urogynecological surgery has traditionally been opioid centered with frequent narcotic administration. Few studies have addressed optimal pain control strategies for vaginal pelvic reconstructive surgery that limit opioid use.Objective: The objective of the study was to determine whether, ice packs, Tylenol, and Toradol, a novel opioid-sparing multimodal postoperative pain regimen has improved pain control compared with the standard postoperative pain regimen in patients undergoing inpatient vaginal pelvic reconstructive surgery.Study Design: This was a multicenter randomized controlled trial of women undergoing vaginal pelvic reconstructive surgery. Patients were randomized to the ice packs, Tylenol, and Toradol postoperative pain regimen or the standard regimen. The ice packs, Tylenol, and Toradol regimen consists of around-the-clock ice packs, around-the-clock oral acetaminophen, around-the-clock intravenous ketorolac, and intravenous hydromorphone for breakthrough pain. The standard regimen consists of as-needed ibuprofen, as-needed acetaminophen/oxycodone, and intravenous hydromorphone for breakthrough pain. The primary outcome was postoperative day 1 pain evaluated the morning after surgery using a visual analog scale. Secondary outcomes included the validated Quality of Recovery Questionnaire, satisfaction scores, inpatient narcotic consumption, outpatient pain medication consumption, and visual analog scale scores at other time intervals. In all, 27 patients in each arm were required to detect a mean difference of 25 mm on a 100 mm visual analog scale (90% power).Results: Thirty patients were randomized to ice packs, Tylenol, and Toradol and 33 to the standard therapy. Patient and surgical demographics were similar. The median morning visual analog scale pain score was lower in the ice packs, Tylenol, and Toradol group (20 mm vs 40 mm, P = .03). Numerical median pain scores were lower at the 96 hour phone call in the ice packs, Tylenol, and Toradol group (2 vs 3, P = .04). Patients randomized to the ICE-T regimen received fewer narcotics (expressed in oral morphine equivalents) from the postanesthesia care unit exit to discharge (2.9 vs 20.4, P < .001) and received fewer narcotics during the entire hospitalization (55.7 vs 91.2, P < .001). At 96 hour follow up, patients in the ice packs, Tylenol, and Toradol group used 4.9 ketorolac tablets compared with 4.6 oxycodone/acetaminophen tablets in the standard group (P = .81); however, ice packs, Tylenol, and Toradol patients required more acetaminophen than ibuprofen by patients in the standard arm (10.7 vs 6.2 tablets, P = .012). There were no differences in Quality of Recovery Questionnaire or satisfaction scores either in the morning after surgery or at 96 hour follow up.Conclusion: The ice packs, Tylenol, and Toradol multimodal pain regimen offers improved pain control the morning after surgery and 96 hours postoperatively compared with the standard regimen with no differences in patient satisfaction and quality of recovery. Ice packs, Tylenol, and Toradol can significantly limit postoperative inpatient narcotic use and eliminate outpatient narcotic use in patients undergoing vaginal pelvic reconstructive surgery. [ABSTRACT FROM AUTHOR]

Introduction and Hypothesis: The objective was to develop a prediction model for urinary tract infection (UTI) after pelvic surgery.We utilized data from three tertiary care centers of women undergoing pelvic surgery. The primary outcome was a UTI within 8 weeks of surgery. Additional variables collected included procedural data, severity of prolapse, use of mesh, anti-incontinence surgery, EBL, diabetes, steroid use, estrogen use, postoperative catheter use, PVR, history of recurrent UTI, operative time, comorbidities, and postoperative morbidity including venous thromboembolism, surgical site infection. Two datasets were used for internal validation, whereas a third dataset was used for external validation. Algorithms that tested included the following: multivariable logistic regression, decision trees (DTs), naive Bayes (NB), random forest (RF), gradient boosting (GB), and multilayer perceptron (MP).For the training dataset, containing both University of British Columbia and Mayo Clinic Rochester data, there were 1,657 patients, with 172 (10.4%) UTIs; whereas for the University of Calgary external validation data, there were a total of 392 patients with a UTI rate of 16.1% (n = 63). All models performed well; however, the GB, DT, and RF models all had an area under the curve (AUC) > 0.97. With external validation the model retained high discriminatory ability, DT: AUC = 0.88, RF: AUC = 0.88, and GB: AUC = 0.90.A model with high discriminatory ability can predict UTI within 8 weeks of pelvic surgery. Future studies should focus on prospective validation and application of randomized trial models to test the utility of this model in the prevention of postoperative UTI.Methods: The objective was to develop a prediction model for urinary tract infection (UTI) after pelvic surgery.We utilized data from three tertiary care centers of women undergoing pelvic surgery. The primary outcome was a UTI within 8 weeks of surgery. Additional variables collected included procedural data, severity of prolapse, use of mesh, anti-incontinence surgery, EBL, diabetes, steroid use, estrogen use, postoperative catheter use, PVR, history of recurrent UTI, operative time, comorbidities, and postoperative morbidity including venous thromboembolism, surgical site infection. Two datasets were used for internal validation, whereas a third dataset was used for external validation. Algorithms that tested included the following: multivariable logistic regression, decision trees (DTs), naive Bayes (NB), random forest (RF), gradient boosting (GB), and multilayer perceptron (MP).For the training dataset, containing both University of British Columbia and Mayo Clinic Rochester data, there were 1,657 patients, with 172 (10.4%) UTIs; whereas for the University of Calgary external validation data, there were a total of 392 patients with a UTI rate of 16.1% (n = 63). All models performed well; however, the GB, DT, and RF models all had an area under the curve (AUC) > 0.97. With external validation the model retained high discriminatory ability, DT: AUC = 0.88, RF: AUC = 0.88, and GB: AUC = 0.90.A model with high discriminatory ability can predict UTI within 8 weeks of pelvic surgery. Future studies should focus on prospective validation and application of randomized trial models to test the utility of this model in the prevention of postoperative UTI.Results: The objective was to develop a prediction model for urinary tract infection (UTI) after pelvic surgery.We utilized data from three tertiary care centers of women undergoing pelvic surgery. The primary outcome was a UTI within 8 weeks of surgery. Additional variables collected included procedural data, severity of prolapse, use of mesh, anti-incontinence surgery, EBL, diabetes, steroid use, estrogen use, postoperative catheter use, PVR, history of recurrent UTI, operative time, comorbidities, and postoperative morbidity including venous thromboembolism, surgical site infection. Two datasets were used for internal validation, whereas a third dataset was used for external validation. Algorithms that tested included the following: multivariable logistic regression, decision trees (DTs), naive Bayes (NB), random forest (RF), gradient boosting (GB), and multilayer perceptron (MP).For the training dataset, containing both University of British Columbia and Mayo Clinic Rochester data, there were 1,657 patients, with 172 (10.4%) UTIs; whereas for the University of Calgary external validation data, there were a total of 392 patients with a UTI rate of 16.1% (n = 63). All models performed well; however, the GB, DT, and RF models all had an area under the curve (AUC) > 0.97. With external validation the model retained high discriminatory ability, DT: AUC = 0.88, RF: AUC = 0.88, and GB: AUC = 0.90.A model with high discriminatory ability can predict UTI within 8 weeks of pelvic surgery. Future studies should focus on prospective validation and application of randomized trial models to test the utility of this model in the prevention of postoperative UTI.Conclusions: The objective was to develop a prediction model for urinary tract infection (UTI) after pelvic surgery.We utilized data from three tertiary care centers of women undergoing pelvic surgery. The primary outcome was a UTI within 8 weeks of surgery. Additional variables collected included procedural data, severity of prolapse, use of mesh, anti-incontinence surgery, EBL, diabetes, steroid use, estrogen use, postoperative catheter use, PVR, history of recurrent UTI, operative time, comorbidities, and postoperative morbidity including venous thromboembolism, surgical site infection. Two datasets were used for internal validation, whereas a third dataset was used for external validation. Algorithms that tested included the following: multivariable logistic regression, decision trees (DTs), naive Bayes (NB), random forest (RF), gradient boosting (GB), and multilayer perceptron (MP).For the training dataset, containing both University of British Columbia and Mayo Clinic Rochester data, there were 1,657 patients, with 172 (10.4%) UTIs; whereas for the University of Calgary external validation data, there were a total of 392 patients with a UTI rate of 16.1% (n = 63). All models performed well; however, the GB, DT, and RF models all had an area under the curve (AUC) > 0.97. With external validation the model retained high discriminatory ability, DT: AUC = 0.88, RF: AUC = 0.88, and GB: AUC = 0.90.A model with high discriminatory ability can predict UTI within 8 weeks of pelvic surgery. Future studies should focus on prospective validation and application of randomized trial models to test the utility of this model in the prevention of postoperative UTI. [ABSTRACT FROM AUTHOR]