- Email: [email protected]
Prospective evaluation of urinary incontinence in severely obese adolescents presenting for weight loss surgery
Surgery for Obesity and Related Diseases ] (2017) 00–00
Original article
Prospective evaluation of urinary incontinence in severely obese adolescents presenting for weight loss surgery W. Robert DeFoor Jr, M.D., M.P.H.a,*, Thomas H. Inge, M.D.f, Todd M. Jenkins, Ph.D.a, Elizabeth Jackson, M.D.a, Anita Courcoulas, M.D.c, Marc Michalsky, M.D.d, Mary Brandt, M.D.e, Linda Kollar, C.N.P., C.B.N.a, Changchun Xie, Ph.D.b a
Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio b University of Cincinnati, Cincinnati, Ohio c University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania d Nationwide Children’s Hospital, Columbus, Ohio e Texas Children’s Hospital, Houston, Texas f University of Colorado Denver, Aurora, Colorado Received July 21, 2017; accepted September 7, 2017
Abstract
Background: Obesity has been associated with daytime urinary incontinence (UI), likely due to increased intra-abdominal pressure. Objectives: To assess incontinence symptoms in severely obese adolescents before and 3 years after bariatric surgery. Setting: Tertiary care pediatric hospitals in the United States. Methods: The Teen-Longitudinal Assessment of Bariatric Surgery is a prospective, multicenter study designed to evaluate efficacy and safety of bariatric surgery in adolescents. Patients o19 years of age undergoing bariatric surgery at 5 centers between 2007 and 2012 were enrolled. Trained study staff collected baseline and postoperative anthropometric and clinical data. Presence and severity of UI were determined by standardized interview. Results: A total of 242 patients (76% female) were evaluated at baseline. The mean age was 17.1 years at baseline, and 72% were of white race. The preoperative median body mass index was 50.5 kg/m2. At baseline, 18% of females and 7% of males reported UI. Prediction analysis at baseline indicated that females, white race, and increasing body mass index had greater odds for UI. UI prevalence in females and males decreased to 7% and 0%, respectively, at 6 months after surgery (P o .01) and remained stable out to 36 months postoperatively. Furthermore, older patients were less likely to achieve 3-year UI remission or improvement. Conclusions: In adolescents undergoing bariatric surgery, UI was more common in females than in males. Incontinence status significantly improved by 6 months and was durable to 3 years after surgery, suggesting that bariatric surgery favorably affects anatomic or physiologic mechanisms of bladder control in both males and females. (Surg Obes Relat Dis 2017;]:00–00.) r 2017 American Society for Metabolic and Bariatric Surgery. All rights reserved.
Keywords:
Incontinence; Obesity; Weight loss surgery
Paper presented at the Pediatric Urology Fall Congress, September 2016. * Correspondence: W. Robert DeFoor Jr, M.D., M.P.H., Cincinnati Children’s Hospital Medical Center, 3333 Burnet Avenue MLC 5037, Cincinnati, OH 45229-3039. E-mail: [email protected]
Obesity has long been associated with multiple adverse health outcomes, including diabetes and cardiovascular disease. Its increasing prevalence in the United States and throughout the world has been well documented [1,2].
http://dx.doi.org/10.1016/j.soard.2017.09.510 1550-7289/r 2017 American Society for Metabolic and Bariatric Surgery. All rights reserved.
2
W. R. DeFoor et al. / Surgery for Obesity and Related Diseases ] (2017) 00–00
Daytime urinary incontinence (UI) is well known to be associated with obesity and can be a significant source of distress for patients. Childhood obesity is becoming more of a public health concern, and population studies showing rates of UI in obese children have recently been reported [3,4]. Bariatric surgery in adults has been well established as a safe and effective method of establishing significant and durable weight loss. It is generally considered to be a newer treatment modality in the pediatric population; however, favorable data have been emerging from several recent and ongoing studies [5–7]. Previous investigations in the bariatric literature have reported a substantially reduced UI rate 3 years after weight loss surgery in adults [8]. To date, there are limited data available that characterize UI in adolescents undergoing surgically induced weight loss. The purpose of the present study was to assess baseline and 3-year UI symptoms in severely obese adolescents undergoing bariatric surgery to determine changes in prevalence and potential predictor variables for change. Materials and Methods After obtaining institutional review board approval from all participating Teen-Longitudinal Assessment of Bariatric Surgery (LABS) clinical centers, a prospective cohort study was performed to assess health benefits and risks in severely obese adolescents undergoing weight loss surgery. Data were collected at 5 pediatric institutions participating in the Teen-LABS (U01 DK072493; Principal investigator: Thomas Inge, M.D.), a National Institute of Diabetes and Digestive and Kidney Diseases–funded consortium. Details of participant recruitment have been previously reported [5,9]. After enrollment, a general medical history and physical was performed in the clinic. Blood and urine were collected for metabolic analyses and to assess renal function. Baseline anthropometric and clinical data were collected by trained study staff and have been previously reported for the entire study cohort [5]. Prevalence and severity of UI were determined by a standardized interview questionnaire. Daytime UI was defined as involuntary urinary leakage occurring at least once weekly. Remission was defined as no incontinence, and improvement was defined as UI occurring less frequently than at baseline (less than once weekly). Further questions were asked to delineate urge versus stress UI. No renal imaging or urodynamic studies were performed for research purposes. Study patients underwent a Roux-en-Y gastric bypass, vertical sleeve gastrectomy, or laparoscopic adjustable gastric banding. The surgical details for these procedures have been well described in the bariatric literature. Postoperative research visits were performed at scheduled intervals and included a follow-up questionnaire performed by the same trained study personnel at 6, 12, 24, and 36 months. Baseline and postoperative data from all 5 clinical centers were collected and maintained by a central
Data Coordinating Center at Cincinnati Children’s Hospital Medical Center. Standard descriptive statistics were calculated to summarize participant characteristics. Frequencies and percentages were reported for categorical measures. Means and standard deviations or medians and interquartile ranges were calculated for continuous variables. Logistic regression was used to evaluate predictors of baseline UI. Model-based remission and incidence estimates and 95% confidence intervals were calculated using generalized linear mixed-modeling. Independent variables evaluated included sex, age at surgery, race, diabetes, body mass index (BMI), pannus grade, and 2-way interactions with all variables and sex. The final adjusted model included all variables for which P ≤ .10. These included sex, diabetes, and gender × diabetes interaction. To evaluate predictors of 3-year remission and/or improvement of UI, risk ratios and 95% confidence intervals were calculated using modified Poisson regression models with robust error estimates. Due to the small number of male participants with UI at baseline (n ¼ 4), modeled remission, improvement, and incidence analyses were restricted to females. Independent variables evaluated in the models included age at surgery, race, percent weight loss, presence of depression, and surgical procedure. Race was not included in the remission and improvement models due to lack of variation (only 1 nonwhite subject with baseline UI). Baseline analyses included all patients who had incontinence data. For the remission/improvement analyses, male and laparoscopic adjustable gastric banding patients were not included due to very small numbers. At the 3-year study visit, 21% of the UI outcomes values were missing. Multivariate imputation by fully conditional specification was performed to address these missing data. A total of 25 imputed data sets were created for use in multivariable modeling analyses. All reported P values were 2-sided and considered statistically significant at o.05. Statistical analysis was performed using SAS version 9.3 (SAS Institute, Cary, NC). Results A total of 242 patients (76% female) were evaluated at baseline (Table 1). Of the study participants, 161 underwent Roux-en-Y gastric bypass, 67 had vertical sleeve gastrectomy, and 14 had laparoscopic adjustable gastric banding. Through the 3-year postoperative time point, 99% of patients remained as active study participants and 88% of postoperative visits were completed. The mean age at baseline was 17.1 years, and most (72%) patients were of white race. The preoperative median BMI was 50.5 kg/m2. Type 2 diabetes was present in 14% (n ¼ 33) of participants at baseline. A pannus grade of 3 or more (panniculus reaches the upper thigh or further) was observed in 23% of
Urinary Incontinence and Weight Loss Surgery / Surgery for Obesity and Related Diseases ] (2017) 00–00 Table 1 Univariate analysis of incontinence risk factors at study baseline Total Total, n Female, n (%) Age (yr), mean White race, n (%) BMI (kg/m2), median (IQR) Diabetes, n (%) Pannus grade 3þ, n (%) Toilet use difficulty, n (%) Current smoker, n (%) Presence of depression, n (%)
Incontinence No P value* incontinence
238 37 180 (76) 33 (89) 16.6 16.9 171 (72) 29 (78) 50.6 (45, 58) 49.4 (45, 60)
201 147 (73) 16.6 142 (71) 50.9 (45, 58)
.038 .337 .428 .832
29 (12) 55 (23)
7 (19) 7 (19)
22 (11) 48 (24)
.182 .672
10 (4)
3 (9)
7 (4)
.177
4 (2)
1 (3)
3 (2)
.490
32 (14)
10 (28)
22 (12)
.017
BMI ¼ body mass index; IQR ¼ interquartile range. P values indicate results of significance testing for differences in proportions. *
the study cohort. A small number of participants (5%) reported difficulty using the toilet due to their size and/or lack of mobility. At baseline, more females (n ¼ 33, 18%) reported UI than males (n ¼ 4, 7%). Logistic regression analysis at baseline indicated that males had significantly lower odds than females (odds ratio ¼ 0.12, P o .01) and nonwhites had lower odds than whites (odds ratio ¼ 0.10, P o .01) for having UI. Increasing BMI also served as an independent predictor of UI (odds ratio ¼ 1.06, P ¼ .05). Fig. 1 shows incontinence prevalence rates by sex and study time point. The incontinence prevalence in females and males decreased from 18% and 7% at baseline to 7% and 0% at 6 months after surgery, respectively (P o .01) and remained unchanged up to 36 months. Table 2 presents modeled UI remission and incidence estimates among
Figure 1. Prevalence of urinary incontinence by gender and time point.
3
female patients. Remission is largely achieved by 6 months and is maintained through 3 years. Few incident cases (new incontinence) arose after surgery (Table 2, bottom row). Table 3 lists predictors of UI remission and improvement 3 years after surgery. Only subject age at the time of bariatric surgery was found to be associated with 3-year remission. In other words, increasing age was found to be inversely associated with the likelihood of achieving remission (relative risk ¼ .86, P ¼ .02) and/or improvement (relative risk ¼ .89, P o .01). Percent weight change from baseline was not associated with UI remission (P ¼ .69) or improvement (P ¼ .17). The mean weight change in the female patients with remission was −24.9%, whereas the change in those with persistent UI was similar at −23.6%. Discussion The rising epidemic of obesity and its many related co-morbid diseases has been well documented [1,10–12]. Significant complications, such as cardiovascular disease and diabetes, can increase the risk of mortality [13,14]. In addition to the often debilitating co-morbidities, UI can be a distressing condition for severely obese children and adults. The prevalence of incontinence in obese adolescent girls (average BMI 40 kg/m2) has been reported in a study by Schwartz et al. [3] to be 13%. Similarly, 18% of the females in our cohort (average BMI 50.5 kg/m2) reported leakage before undergoing bariatric surgery. In 1 case series of patients presenting to a voiding dysfunction clinic, 51% and 31% of patients with daytime incontinence were mildly and severely obese, respectively [15]. The present study found significant and apparently durable improvement of UI rates in severely obese children undergoing weight loss surgery. To our knowledge, this is the first study documenting the long-term remission and incidence in obese children after surgical intervention. A few single institution reports in the adult literature have shown similar findings [16,17]. The multi-institutional LABS consortium reported substantially reduced UI in adults over the time frame of the 3-year study period. In women undergoing weight loss surgery, incontinence decreased from 49% at baseline to 18% after 1 year. Obese men also showed improvement with a rate of 22% decreasing to 10% in the same time period [8]. Some relapse occurred after 3 years, but the incontinence prevalence was still significantly lower than baseline. When considering the adult data in light of our observations in adolescents, it is interesting to speculate that UI may represent an obesity-related problem with a cumulative effect and may be more readily reversible when addressed earlier in life. However, with mostly older adolescents in our cohort, it is unclear if proceeding with surgical intervention a year or 2 earlier would lead to better continence outcomes. Stress incontinence from increased intra-abdominal pressure makes intuitive sense as the subtype of incontinence affecting
W. R. DeFoor et al. / Surgery for Obesity and Related Diseases ] (2017) 00–00
4
Table 2 Modeled urinary incontinence remission and incidence after surgery, female participants
Urinary incontinence remission Complete remission Any type Stress type Urinary incontinence incidence (new incontinence) Any type
6 mo
1 yr
2 yr
3 yr
% (95% CI)
% (95% CI)
% (95% CI)
% (95% CI)
79.9 (59.5–91.5) 79.9 (59.5–91.5) 80.3 (55.9–92.9)
89.2 (68.6–96.9) 89.2 (68.6–96.9) 91.5 (66.7–98.3)
78.9 (57.6–91.1) 78.9 (57.6–91.1) 84.0 (59.7––94.9)
74.2 (51.7–88.6) 74.2 (51.7–88.6) 78.9 (52.6–92.6)
3.3 (1.0–10.3)
3.3 (1.0––10.4)
1.6 (.3–7.6)
1.7 (.3–8.1)
P value
.6 .6 .7 .8
CI ¼ confidence interval.
those with obesity. However, urge incontinence and overactive bladder symptoms have also been reported to be higher in obese children than in nonobese controls in 1 communitybased study from Taiwan using a validated dysfunctional voiding symptom survey [4], suggesting a complex interplay between obesity and physical and physiologic mechanisms controlling continence of the urinary bladder. Further urologic evaluation of patients having persistent UI after otherwise successful weight loss surgery may be helpful in making a more precise diagnosis and treatment plan. In the current analysis, females had higher rates of incontinence than males at baseline, as expected. However, the low numbers of male patients in the study did not allow for more detailed modeling of predictors for remission and improvement. The male patients with UI at baseline all had complete remission after surgery, which seemed durable. It is not clear why the resolution occurred in the first 6 months after surgery when there is typically still ongoing weight loss. Limitations of the present study include possible recall bias present in a study based on questionnaire data. In addition, although o2% of the patients were lost to followup, approximately one fifth of the data points were missing at the 3-year evaluation, which could contribute to a possible over-representation of remission and improvement rates. Moreover, a detailed urologic evaluation was not performed at the Teen-LABS research visits in the incontinent patients. Thus, a more thorough evaluation of bladder storage and emptying was not available.
Conclusions In adolescents undergoing bariatric surgery, UI was more common in females than in males. Continence status significantly improved by 6 months and was durable to 3 years postoperatively, suggesting that bariatric surgery favorably affects anatomic or physiologic mechanisms of bladder control in both males and females. Resolution of UI may be related to a younger age at time of surgery. Further studies with detailed urologic evaluations may be helpful to more precisely categorize UI in severely obese children. Disclosure Statement The authors have no commercial associations that might be a conflict of interest in relation to this article. Acknowledgments The Teen-LABS consortium was funded by cooperative agreements with the National Institute of Diabetes and Digestive and Kidney Diseases through grants U01 DK072493, UM1 DK072493, and UM1 DK095710 (University of Cincinnati). The study was also supported by grants UL1 TR000077-04 (Cincinnati Children’s Hospital Medical Center), UL1 RR025755 (Nationwide Children’s Hospital), M01-RR00188 (Texas Children’s Hospital/ Baylor College of Medicine), UL1 RR024153 and UL1 TR000005 (University of Pittsburgh), and UL1 TR000165 (University of Alabama, Birmingham). References
Table 3 Predictors of urinary incontinence remission and improvement after bariatric surgery, female participants Remission
Improvement
RR 95% CI P value RR Age at surgery Percent weight loss from baseline Depression Diabetes Pannus grade ≥3
95% CI
P value
.86 .76–.98 .02 1.00 .98–1.02 .69
.89 .82–.97 o.01 1.01 1.00–1.02 .17
.95 .61–1.49 .83 1.05 .75–1.48 .76 1.38 .91–2.09 .13
1.08 .90 1.23
RR ¼ relative risk; CI ¼ confidence interval.
.89–1.31 .73–1.12 .88–1.73
.44 .35 .23
[1] Shields M, Carroll MD, Ogden CL. Adult obesity prevalence in Canada and the United States. NCHS Data Brief 2011;(56):1–8. [2] Perez Rodrigo C. Current mapping of obesity. Nutr Hosp 2013;28 (Suppl 5):21–31. [3] Schwartz B, Wyman JF, Thomas W, Schwarzenberg SJ. Urinary incontinence in obese adolescent girls. J Pediatr Urol 2009;5(6):445–50. [4] Chang SJ, Chiang IN, Lin CD, Hsieh CH, Yang SS. Obese children at higher risk for having overactive bladder symptoms: a communitybased study. Neurourol Urodyn 2015;34(2):123–7. [5] Inge TH, Zeller MH, Jenkins TM, et al. Perioperative outcomes of adolescents undergoing bariatric surgery: the Teen-Longitudinal Assessment of Bariatric Surgery (Teen-LABS) study. JAMA Pediatr 2014;168(1):47–53.
Urinary Incontinence and Weight Loss Surgery / Surgery for Obesity and Related Diseases ] (2017) 00–00 [6] Inge TH, Jenkins TM, Xanthakos SA, et al. Long-term outcomes of bariatric surgery in adolescents with severe obesity (FABS-5þ): a prospective follow-up analysis. Lancet Diabetes Endocrinol 2017;5 (3):165–73. [7] Olbers T, Beamish AJ, Gronowitz E, et al. Laparoscopic Roux-en-Y gastric bypass in adolescents with severe obesity (AMOS): a prospective, 5-year, Swedish nationwide study. Lancet Diabetes Endocrinol 2017;5(3):174–83. [8] Subak LL, King WC, Belle SH, et al. Urinary incontinence before and after bariatric surgery. JAMA Intern Med 2015;175(8):1378–87. [9] Inge TH, Courcoulas AP, Xanthakos SA. Weight loss and health status after bariatric surgery in adolescents. N Engl J Med 2016;374 (20):1989–90. [10] Pender JR, Pories WJ. Epidemiology of obesity in the United States. Gastroenterol Clin North Am 2005;34(1):1–7. [11] Flegal KM, Carroll MD, Ogden CL, Curtin LR. Prevalence and trends in obesity among US adults, 1999-2008. JAMA 2010;303(3):235–41.
5
[12] Ogden CL, Yanovski SZ, Carroll MD, Flegal KM. The epidemiology of obesity. Gastroenterology 2007;132(6):2087–102. [13] Twig G, Tirosh A, Leiba A, et al. BMI at age 17 years and diabetes mortality in midlife: a nationwide cohort of 2.3 million adolescents. Diabetes Care 2016;39(11):1996–2003. [14] Twig G, Yaniv G, Levine H, et al. Body-mass index in 2.3 million adolescents and cardiovascular death in adulthood. N Engl J Med 2016;374(25):2430–40. [15] Erdem E, Lin A, Kogan BA, Feustel PJ. Association of elimination dysfunction and body mass index. J Pediatr Urol 2006;2 (4):364–7. [16] O'Boyle CJ, O'Sullivan OE, Shabana H, Boyce M, O'Reilly BA. The effect of bariatric surgery on urinary incontinence in women. Obes Surg 2016;26(7):1471–8. [17] Knepfler T, Valero E, Triki E, Chilintseva N, Koensgen S, Rohr S. Bariatric surgery improves female pelvic floor disorders. J Visc Surg 2016;153(2):95–9.
Original article
Prospective evaluation of urinary incontinence in severely obese adolescents presenting for weight loss surgery W. Robert DeFoor Jr, M.D., M.P.H.a,*, Thomas H. Inge, M.D.f, Todd M. Jenkins, Ph.D.a, Elizabeth Jackson, M.D.a, Anita Courcoulas, M.D.c, Marc Michalsky, M.D.d, Mary Brandt, M.D.e, Linda Kollar, C.N.P., C.B.N.a, Changchun Xie, Ph.D.b a
Cincinnati Children’s Hospital Medical Center, Cincinnati, Ohio b University of Cincinnati, Cincinnati, Ohio c University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania d Nationwide Children’s Hospital, Columbus, Ohio e Texas Children’s Hospital, Houston, Texas f University of Colorado Denver, Aurora, Colorado Received July 21, 2017; accepted September 7, 2017
Abstract
Background: Obesity has been associated with daytime urinary incontinence (UI), likely due to increased intra-abdominal pressure. Objectives: To assess incontinence symptoms in severely obese adolescents before and 3 years after bariatric surgery. Setting: Tertiary care pediatric hospitals in the United States. Methods: The Teen-Longitudinal Assessment of Bariatric Surgery is a prospective, multicenter study designed to evaluate efficacy and safety of bariatric surgery in adolescents. Patients o19 years of age undergoing bariatric surgery at 5 centers between 2007 and 2012 were enrolled. Trained study staff collected baseline and postoperative anthropometric and clinical data. Presence and severity of UI were determined by standardized interview. Results: A total of 242 patients (76% female) were evaluated at baseline. The mean age was 17.1 years at baseline, and 72% were of white race. The preoperative median body mass index was 50.5 kg/m2. At baseline, 18% of females and 7% of males reported UI. Prediction analysis at baseline indicated that females, white race, and increasing body mass index had greater odds for UI. UI prevalence in females and males decreased to 7% and 0%, respectively, at 6 months after surgery (P o .01) and remained stable out to 36 months postoperatively. Furthermore, older patients were less likely to achieve 3-year UI remission or improvement. Conclusions: In adolescents undergoing bariatric surgery, UI was more common in females than in males. Incontinence status significantly improved by 6 months and was durable to 3 years after surgery, suggesting that bariatric surgery favorably affects anatomic or physiologic mechanisms of bladder control in both males and females. (Surg Obes Relat Dis 2017;]:00–00.) r 2017 American Society for Metabolic and Bariatric Surgery. All rights reserved.
Keywords:
Incontinence; Obesity; Weight loss surgery
Paper presented at the Pediatric Urology Fall Congress, September 2016. * Correspondence: W. Robert DeFoor Jr, M.D., M.P.H., Cincinnati Children’s Hospital Medical Center, 3333 Burnet Avenue MLC 5037, Cincinnati, OH 45229-3039. E-mail: [email protected]
Obesity has long been associated with multiple adverse health outcomes, including diabetes and cardiovascular disease. Its increasing prevalence in the United States and throughout the world has been well documented [1,2].
http://dx.doi.org/10.1016/j.soard.2017.09.510 1550-7289/r 2017 American Society for Metabolic and Bariatric Surgery. All rights reserved.
2
W. R. DeFoor et al. / Surgery for Obesity and Related Diseases ] (2017) 00–00
Daytime urinary incontinence (UI) is well known to be associated with obesity and can be a significant source of distress for patients. Childhood obesity is becoming more of a public health concern, and population studies showing rates of UI in obese children have recently been reported [3,4]. Bariatric surgery in adults has been well established as a safe and effective method of establishing significant and durable weight loss. It is generally considered to be a newer treatment modality in the pediatric population; however, favorable data have been emerging from several recent and ongoing studies [5–7]. Previous investigations in the bariatric literature have reported a substantially reduced UI rate 3 years after weight loss surgery in adults [8]. To date, there are limited data available that characterize UI in adolescents undergoing surgically induced weight loss. The purpose of the present study was to assess baseline and 3-year UI symptoms in severely obese adolescents undergoing bariatric surgery to determine changes in prevalence and potential predictor variables for change. Materials and Methods After obtaining institutional review board approval from all participating Teen-Longitudinal Assessment of Bariatric Surgery (LABS) clinical centers, a prospective cohort study was performed to assess health benefits and risks in severely obese adolescents undergoing weight loss surgery. Data were collected at 5 pediatric institutions participating in the Teen-LABS (U01 DK072493; Principal investigator: Thomas Inge, M.D.), a National Institute of Diabetes and Digestive and Kidney Diseases–funded consortium. Details of participant recruitment have been previously reported [5,9]. After enrollment, a general medical history and physical was performed in the clinic. Blood and urine were collected for metabolic analyses and to assess renal function. Baseline anthropometric and clinical data were collected by trained study staff and have been previously reported for the entire study cohort [5]. Prevalence and severity of UI were determined by a standardized interview questionnaire. Daytime UI was defined as involuntary urinary leakage occurring at least once weekly. Remission was defined as no incontinence, and improvement was defined as UI occurring less frequently than at baseline (less than once weekly). Further questions were asked to delineate urge versus stress UI. No renal imaging or urodynamic studies were performed for research purposes. Study patients underwent a Roux-en-Y gastric bypass, vertical sleeve gastrectomy, or laparoscopic adjustable gastric banding. The surgical details for these procedures have been well described in the bariatric literature. Postoperative research visits were performed at scheduled intervals and included a follow-up questionnaire performed by the same trained study personnel at 6, 12, 24, and 36 months. Baseline and postoperative data from all 5 clinical centers were collected and maintained by a central
Data Coordinating Center at Cincinnati Children’s Hospital Medical Center. Standard descriptive statistics were calculated to summarize participant characteristics. Frequencies and percentages were reported for categorical measures. Means and standard deviations or medians and interquartile ranges were calculated for continuous variables. Logistic regression was used to evaluate predictors of baseline UI. Model-based remission and incidence estimates and 95% confidence intervals were calculated using generalized linear mixed-modeling. Independent variables evaluated included sex, age at surgery, race, diabetes, body mass index (BMI), pannus grade, and 2-way interactions with all variables and sex. The final adjusted model included all variables for which P ≤ .10. These included sex, diabetes, and gender × diabetes interaction. To evaluate predictors of 3-year remission and/or improvement of UI, risk ratios and 95% confidence intervals were calculated using modified Poisson regression models with robust error estimates. Due to the small number of male participants with UI at baseline (n ¼ 4), modeled remission, improvement, and incidence analyses were restricted to females. Independent variables evaluated in the models included age at surgery, race, percent weight loss, presence of depression, and surgical procedure. Race was not included in the remission and improvement models due to lack of variation (only 1 nonwhite subject with baseline UI). Baseline analyses included all patients who had incontinence data. For the remission/improvement analyses, male and laparoscopic adjustable gastric banding patients were not included due to very small numbers. At the 3-year study visit, 21% of the UI outcomes values were missing. Multivariate imputation by fully conditional specification was performed to address these missing data. A total of 25 imputed data sets were created for use in multivariable modeling analyses. All reported P values were 2-sided and considered statistically significant at o.05. Statistical analysis was performed using SAS version 9.3 (SAS Institute, Cary, NC). Results A total of 242 patients (76% female) were evaluated at baseline (Table 1). Of the study participants, 161 underwent Roux-en-Y gastric bypass, 67 had vertical sleeve gastrectomy, and 14 had laparoscopic adjustable gastric banding. Through the 3-year postoperative time point, 99% of patients remained as active study participants and 88% of postoperative visits were completed. The mean age at baseline was 17.1 years, and most (72%) patients were of white race. The preoperative median BMI was 50.5 kg/m2. Type 2 diabetes was present in 14% (n ¼ 33) of participants at baseline. A pannus grade of 3 or more (panniculus reaches the upper thigh or further) was observed in 23% of
Urinary Incontinence and Weight Loss Surgery / Surgery for Obesity and Related Diseases ] (2017) 00–00 Table 1 Univariate analysis of incontinence risk factors at study baseline Total Total, n Female, n (%) Age (yr), mean White race, n (%) BMI (kg/m2), median (IQR) Diabetes, n (%) Pannus grade 3þ, n (%) Toilet use difficulty, n (%) Current smoker, n (%) Presence of depression, n (%)
Incontinence No P value* incontinence
238 37 180 (76) 33 (89) 16.6 16.9 171 (72) 29 (78) 50.6 (45, 58) 49.4 (45, 60)
201 147 (73) 16.6 142 (71) 50.9 (45, 58)
.038 .337 .428 .832
29 (12) 55 (23)
7 (19) 7 (19)
22 (11) 48 (24)
.182 .672
10 (4)
3 (9)
7 (4)
.177
4 (2)
1 (3)
3 (2)
.490
32 (14)
10 (28)
22 (12)
.017
BMI ¼ body mass index; IQR ¼ interquartile range. P values indicate results of significance testing for differences in proportions. *
the study cohort. A small number of participants (5%) reported difficulty using the toilet due to their size and/or lack of mobility. At baseline, more females (n ¼ 33, 18%) reported UI than males (n ¼ 4, 7%). Logistic regression analysis at baseline indicated that males had significantly lower odds than females (odds ratio ¼ 0.12, P o .01) and nonwhites had lower odds than whites (odds ratio ¼ 0.10, P o .01) for having UI. Increasing BMI also served as an independent predictor of UI (odds ratio ¼ 1.06, P ¼ .05). Fig. 1 shows incontinence prevalence rates by sex and study time point. The incontinence prevalence in females and males decreased from 18% and 7% at baseline to 7% and 0% at 6 months after surgery, respectively (P o .01) and remained unchanged up to 36 months. Table 2 presents modeled UI remission and incidence estimates among
Figure 1. Prevalence of urinary incontinence by gender and time point.
3
female patients. Remission is largely achieved by 6 months and is maintained through 3 years. Few incident cases (new incontinence) arose after surgery (Table 2, bottom row). Table 3 lists predictors of UI remission and improvement 3 years after surgery. Only subject age at the time of bariatric surgery was found to be associated with 3-year remission. In other words, increasing age was found to be inversely associated with the likelihood of achieving remission (relative risk ¼ .86, P ¼ .02) and/or improvement (relative risk ¼ .89, P o .01). Percent weight change from baseline was not associated with UI remission (P ¼ .69) or improvement (P ¼ .17). The mean weight change in the female patients with remission was −24.9%, whereas the change in those with persistent UI was similar at −23.6%. Discussion The rising epidemic of obesity and its many related co-morbid diseases has been well documented [1,10–12]. Significant complications, such as cardiovascular disease and diabetes, can increase the risk of mortality [13,14]. In addition to the often debilitating co-morbidities, UI can be a distressing condition for severely obese children and adults. The prevalence of incontinence in obese adolescent girls (average BMI 40 kg/m2) has been reported in a study by Schwartz et al. [3] to be 13%. Similarly, 18% of the females in our cohort (average BMI 50.5 kg/m2) reported leakage before undergoing bariatric surgery. In 1 case series of patients presenting to a voiding dysfunction clinic, 51% and 31% of patients with daytime incontinence were mildly and severely obese, respectively [15]. The present study found significant and apparently durable improvement of UI rates in severely obese children undergoing weight loss surgery. To our knowledge, this is the first study documenting the long-term remission and incidence in obese children after surgical intervention. A few single institution reports in the adult literature have shown similar findings [16,17]. The multi-institutional LABS consortium reported substantially reduced UI in adults over the time frame of the 3-year study period. In women undergoing weight loss surgery, incontinence decreased from 49% at baseline to 18% after 1 year. Obese men also showed improvement with a rate of 22% decreasing to 10% in the same time period [8]. Some relapse occurred after 3 years, but the incontinence prevalence was still significantly lower than baseline. When considering the adult data in light of our observations in adolescents, it is interesting to speculate that UI may represent an obesity-related problem with a cumulative effect and may be more readily reversible when addressed earlier in life. However, with mostly older adolescents in our cohort, it is unclear if proceeding with surgical intervention a year or 2 earlier would lead to better continence outcomes. Stress incontinence from increased intra-abdominal pressure makes intuitive sense as the subtype of incontinence affecting
W. R. DeFoor et al. / Surgery for Obesity and Related Diseases ] (2017) 00–00
4
Table 2 Modeled urinary incontinence remission and incidence after surgery, female participants
Urinary incontinence remission Complete remission Any type Stress type Urinary incontinence incidence (new incontinence) Any type
6 mo
1 yr
2 yr
3 yr
% (95% CI)
% (95% CI)
% (95% CI)
% (95% CI)
79.9 (59.5–91.5) 79.9 (59.5–91.5) 80.3 (55.9–92.9)
89.2 (68.6–96.9) 89.2 (68.6–96.9) 91.5 (66.7–98.3)
78.9 (57.6–91.1) 78.9 (57.6–91.1) 84.0 (59.7––94.9)
74.2 (51.7–88.6) 74.2 (51.7–88.6) 78.9 (52.6–92.6)
3.3 (1.0–10.3)
3.3 (1.0––10.4)
1.6 (.3–7.6)
1.7 (.3–8.1)
P value
.6 .6 .7 .8
CI ¼ confidence interval.
those with obesity. However, urge incontinence and overactive bladder symptoms have also been reported to be higher in obese children than in nonobese controls in 1 communitybased study from Taiwan using a validated dysfunctional voiding symptom survey [4], suggesting a complex interplay between obesity and physical and physiologic mechanisms controlling continence of the urinary bladder. Further urologic evaluation of patients having persistent UI after otherwise successful weight loss surgery may be helpful in making a more precise diagnosis and treatment plan. In the current analysis, females had higher rates of incontinence than males at baseline, as expected. However, the low numbers of male patients in the study did not allow for more detailed modeling of predictors for remission and improvement. The male patients with UI at baseline all had complete remission after surgery, which seemed durable. It is not clear why the resolution occurred in the first 6 months after surgery when there is typically still ongoing weight loss. Limitations of the present study include possible recall bias present in a study based on questionnaire data. In addition, although o2% of the patients were lost to followup, approximately one fifth of the data points were missing at the 3-year evaluation, which could contribute to a possible over-representation of remission and improvement rates. Moreover, a detailed urologic evaluation was not performed at the Teen-LABS research visits in the incontinent patients. Thus, a more thorough evaluation of bladder storage and emptying was not available.
Conclusions In adolescents undergoing bariatric surgery, UI was more common in females than in males. Continence status significantly improved by 6 months and was durable to 3 years postoperatively, suggesting that bariatric surgery favorably affects anatomic or physiologic mechanisms of bladder control in both males and females. Resolution of UI may be related to a younger age at time of surgery. Further studies with detailed urologic evaluations may be helpful to more precisely categorize UI in severely obese children. Disclosure Statement The authors have no commercial associations that might be a conflict of interest in relation to this article. Acknowledgments The Teen-LABS consortium was funded by cooperative agreements with the National Institute of Diabetes and Digestive and Kidney Diseases through grants U01 DK072493, UM1 DK072493, and UM1 DK095710 (University of Cincinnati). The study was also supported by grants UL1 TR000077-04 (Cincinnati Children’s Hospital Medical Center), UL1 RR025755 (Nationwide Children’s Hospital), M01-RR00188 (Texas Children’s Hospital/ Baylor College of Medicine), UL1 RR024153 and UL1 TR000005 (University of Pittsburgh), and UL1 TR000165 (University of Alabama, Birmingham). References
Table 3 Predictors of urinary incontinence remission and improvement after bariatric surgery, female participants Remission
Improvement
RR 95% CI P value RR Age at surgery Percent weight loss from baseline Depression Diabetes Pannus grade ≥3
95% CI
P value
.86 .76–.98 .02 1.00 .98–1.02 .69
.89 .82–.97 o.01 1.01 1.00–1.02 .17
.95 .61–1.49 .83 1.05 .75–1.48 .76 1.38 .91–2.09 .13
1.08 .90 1.23
RR ¼ relative risk; CI ¼ confidence interval.
.89–1.31 .73–1.12 .88–1.73
.44 .35 .23
[1] Shields M, Carroll MD, Ogden CL. Adult obesity prevalence in Canada and the United States. NCHS Data Brief 2011;(56):1–8. [2] Perez Rodrigo C. Current mapping of obesity. Nutr Hosp 2013;28 (Suppl 5):21–31. [3] Schwartz B, Wyman JF, Thomas W, Schwarzenberg SJ. Urinary incontinence in obese adolescent girls. J Pediatr Urol 2009;5(6):445–50. [4] Chang SJ, Chiang IN, Lin CD, Hsieh CH, Yang SS. Obese children at higher risk for having overactive bladder symptoms: a communitybased study. Neurourol Urodyn 2015;34(2):123–7. [5] Inge TH, Zeller MH, Jenkins TM, et al. Perioperative outcomes of adolescents undergoing bariatric surgery: the Teen-Longitudinal Assessment of Bariatric Surgery (Teen-LABS) study. JAMA Pediatr 2014;168(1):47–53.
Urinary Incontinence and Weight Loss Surgery / Surgery for Obesity and Related Diseases ] (2017) 00–00 [6] Inge TH, Jenkins TM, Xanthakos SA, et al. Long-term outcomes of bariatric surgery in adolescents with severe obesity (FABS-5þ): a prospective follow-up analysis. Lancet Diabetes Endocrinol 2017;5 (3):165–73. [7] Olbers T, Beamish AJ, Gronowitz E, et al. Laparoscopic Roux-en-Y gastric bypass in adolescents with severe obesity (AMOS): a prospective, 5-year, Swedish nationwide study. Lancet Diabetes Endocrinol 2017;5(3):174–83. [8] Subak LL, King WC, Belle SH, et al. Urinary incontinence before and after bariatric surgery. JAMA Intern Med 2015;175(8):1378–87. [9] Inge TH, Courcoulas AP, Xanthakos SA. Weight loss and health status after bariatric surgery in adolescents. N Engl J Med 2016;374 (20):1989–90. [10] Pender JR, Pories WJ. Epidemiology of obesity in the United States. Gastroenterol Clin North Am 2005;34(1):1–7. [11] Flegal KM, Carroll MD, Ogden CL, Curtin LR. Prevalence and trends in obesity among US adults, 1999-2008. JAMA 2010;303(3):235–41.
5
[12] Ogden CL, Yanovski SZ, Carroll MD, Flegal KM. The epidemiology of obesity. Gastroenterology 2007;132(6):2087–102. [13] Twig G, Tirosh A, Leiba A, et al. BMI at age 17 years and diabetes mortality in midlife: a nationwide cohort of 2.3 million adolescents. Diabetes Care 2016;39(11):1996–2003. [14] Twig G, Yaniv G, Levine H, et al. Body-mass index in 2.3 million adolescents and cardiovascular death in adulthood. N Engl J Med 2016;374(25):2430–40. [15] Erdem E, Lin A, Kogan BA, Feustel PJ. Association of elimination dysfunction and body mass index. J Pediatr Urol 2006;2 (4):364–7. [16] O'Boyle CJ, O'Sullivan OE, Shabana H, Boyce M, O'Reilly BA. The effect of bariatric surgery on urinary incontinence in women. Obes Surg 2016;26(7):1471–8. [17] Knepfler T, Valero E, Triki E, Chilintseva N, Koensgen S, Rohr S. Bariatric surgery improves female pelvic floor disorders. J Visc Surg 2016;153(2):95–9.