- Email: [email protected]
Effect of baseline obesity and postoperative weight gain on the risk of channel revision following continent catheterizable urinary channel surgery
Journal of Pediatric Urology (2016) 12, 249.e1e249.e7
Effect of baseline obesity and postoperative weight gain on the risk of channel revision following continent catheterizable urinary channel surgery a
Division of Pediatric Urology, Riley Hospital for Children at Indiana University Health, Indiana University School of Medicine, Indianapolis, IN, USA
Katherine H. Chan a,b, Konrad M. Szymanski a, Xiaochun Li c, Susan Ofner c, Chandra Flack a, Benjamin Judge a, Benjamin Whittam a, Rosalia Misseri a, Martin Kaefer a, Richard C. Rink a, Mark P. Cain a
b
Department of Biostatistics, Indiana University-Purdue University, Indianapolis, IN, USA
c
Center for Pediatric and Adolescent Comparative Effectiveness Research, Arkanasas, USA Correspondence to: K.H. Chan, 705 Riley Hospital Drive, Division of Pediatric Urology, Riley Outpatient Center, 4230 Indianapolis, IN 46202, USA [email protected] (K.H. Chan) Keywords Obesity; Urinary bladder; Neurogenic; Postoperative complications Received 30 December 2015 Accepted 22 May 2016 Available online 11 June 2016
Summary
Objective Prior studies suggest that obese patients are at increased risk for complications following continent catheterizable urinary (CCU) channel surgery. We hypothesized that postoperative weight gain increases the risk of channel angulation, difficulty catheterizing, and possible channel perforation requiring subfascial revision. The purpose of this study was to evaluate whether baseline obesity or becoming overweight/obese postoperatively was associated with a greater risk of subfascial revision.
Method We reviewed retrospectively an institutional database of patients who underwent CCU channel surgery between the ages of 2 and <20 years from January 1990 to May 2013, excluding those with continent urinary reservoirs, continent vesicostomies, and those without body mass index (BMI) data. We collected data on patient/procedure characteristics, baseline/most recent BMI, and subfascial revision(s). We used Cox proportional hazard multivariable regression to assess the association of being overweight/obese at baseline (85% BMI) with time to first subfascial revision, and Fisher’s exact test to compare rates of subfascial revision between those who became overweight/obese and those who did not.
Results Of the patients, 328/501 (65.5%) had baseline and post-baseline BMI data available: 53.4% male, 90.6% white, median age 7.4 years; median follow-up 76.4 months. Of the 328 patients, 38 (11.6%) had subfascial revisions. Baseline BMI data were available for 378 patients, and, of these, 130 (34.4%) were overweight/obese at baseline. Overweight/obese patients were more likely to undergo umbilical Monti (10% vs. 8.1%), non-umbilical spiral Monti (33.8% vs. 13.7%), and spiral umbilical Monti channels (13.8% vs. 7.3%) versus normal/underweight patients (p < 0.0001). From a multivariable Cox proportional hazard model controlling for age, BMI category, diagnosis, and ambulatory status, the hazard of subfascial revision for spiral umbilical Monti channels was 2.1 that of other channels (hazard ratio (HR) 2.1 [95% CI 1.2e3.8], p Z 0.01). Fifty-one out of 328 patients (15.6%) became overweight/obese postoperatively, with 7.8% having a subfascial revision vs. 12.3% of those whose weight category decreased or remained stable (p Z 0.3) (Table 1). Conclusions Patients who were overweight/obese at baseline were more likely to have channels constructed that are at the highest risk of subfascial revision. Patients who became overweight/obese postoperatively were not at greater risk of subfascial revision. Limitations include potential bias because of differential follow-up and inaccuracy of BMI percentile as a measurement of obesity.
http://dx.doi.org/10.1016/j.jpurol.2016.05.021 1477-5131/ª 2016 Published by Elsevier Ltd on behalf of Journal of Pediatric Urology Company.
249.e2
Introduction A number of postoperative complications have been reported after creation of continent catheterizable urinary (CCU) channels, and these can be categorized into either stomal complications (e.g., stenosis or prolapse) or subfascial complications (e.g., channel angulation, diverticulum, or incontinence) [1e4]. We previously reported that the increased risk of subfascial revision in Monti channels, particularly the longer spiral umbilical Monti channels, is believed to be secondary to a longer, unsupported extravesical channel [5]. Prior studies have suggested that baseline obesity may increase the risk of postoperative complications following lower urinary tract reconstruction [6,7]. In a study of myelodysplasia patients undergoing lower urinary tract reconstruction, Donovan et al. noted a significant association between baseline obesity and the presence of any medical or surgical postoperative complication [7]. Clark et al. noted an association between percentile weight and stomal stenosis rate following CCU channel procedures, although the differences did not reach statistical significance [6]. To our knowledge there are no prior studies on the effects of postoperative weight gain in children undergoing lower urinary tract reconstruction. We hypothesized that patients with CCU channels who gain weight in the postoperative period may increase the distance from the bladder to the skin (particularly with non-umbilical stomas) resulting in channel angulation, difficulty catheterizing, and possible channel perforation. This type of suprafascial issue may require channel revision including subfascial repair. The purpose of this study was to examine whether patients who are obese at the time of surgery or who become overweight or develop obesity postoperatively, have an increased risk of subfascial channel revision.
Materials and methods We included all patients 2 years of age and <20 years of age from our retrospective institutional database who underwent CCU channels at our institution between January 1990 and May 2013. The database was constructed from a combination of billing data, review of electronic medical records, and paper charts. Patients with continent urinary reservoirs and channels constructed of tissue other than appendix or ileum (e.g., colon or ureter or continent vesicostomies) were excluded from the database. We also excluded those with missing height or weight data as their body mass index (BMI) could not be calculated. Patients <2 years of age and 20 years of age at the time of surgery were excluded from the study as it is not possible to calculate BMI in these age groups using the Centers for Disease Control (CDC) pediatric criteria [8]. The primary outcome was the need for at least one subfascial revision, defined as a need for laparotomy to correct difficult catheterization because of issues such as channel angulation, perforation, diverticulum, false passage, or stenosis. Demographics, clinical and CCU channel
K.H. Chan et al. characteristics (e.g., stomal location), and follow-up duration were abstracted from the institutional database and patient charts. Height and weight data were identified from retrospective review of hospital records. Preoperative or “baseline” BMI was defined as a patient’s BMI measured at the closest time point within 1 year of the date of CCU channel surgery. Postoperative BMI was defined as the BMI at the most recent postoperative visit or within 1 year of subfascial revision or channel excision. The 1-year perioperative timeframe was chosen to maximize the number of patients with available BMI data, while minimizing inaccuracies of BMI data that are not representative of true baseline or postoperative BMI. BMI was calculated according to the standard formula, BMI Z kg/m2. Weight categories were defined for pediatric patients 2 years and <20 years based on an age- and genderspecific BMI percentile according to CDC criteria [8]. BMI percentile, equivalent to the BMI z-score, was used because children’s body composition varies by age and between boys and girls [9]. Patients were categorized as low/normal weight (<85th BMI percentile) and overweight/obese (85th BMI percentile) [8]. A categorical variable was created for all patients in the cohort using the following categories: a) became overweight/obese, b) maintained stable low/normal BMI, c) maintained overweight/obese BMI, and d) became low/normal (overweight/obese at baseline). We used Fisher’s exact test to examine the association between baseline BMI percentile category and channel type.
Data analysis of baseline obesity Demographic and clinical data were compared between those with and without a first revision by use of Fisher’s exact and chi-square tests, to elucidate risk factors for subfascial revisions. Continuous characteristics were analyzed using an analysis of variance model or Wilcoxon rank sum test as appropriate based on data characteristics. We examined the association between BMI percentile category and channel type using Fisher’s exact test. We also compared demographic and clinical data between different BMI percentile categories including: a) missing BMI percentile vs. underweight/healthy BMI percentile, and b) overweight/obese BMI percentile vs. underweight/ healthy BMI percentile. We assessed the association of the baseline BMI percentile category with time to the first subfascial revision, our primary outcome, controlling for demographic and clinical characteristics, using Cox proportional hazards regression. Patient characteristics of age, gender, race, BMI percentile category, presence or absence of myelomeningocele, ambulatory status, channel type, and stomal location were thought to be clinically important. These variables were selected, a priori, for inclusion in the model of time to first subfascial revision. The proportional hazards assumption was assessed for each covariate by graphical means. Variables (gender and race) for which the assumption did not hold were removed from the multivariable model. The final Cox model included age, BMI percentile category, spiral Monti channel at the umbilicus, myelomeningocele diagnosis, and ambulatory status.
Weight gain after urinary channel surgery
Data analysis of postoperative obesity We also compared rates of subfascial revision respectively between those who became overweight/obese (BMI 85%) versus those who did not, using Fisher’s exact test. All analyses were performed using SAS version 9.3 (SAS Institute Inc., Cary, NC, USA). A p-values less than 0.05 was considered to be significant.
Ethical approval The institutional review board approved this study.
Results Patient population Of the 510 patients in the database, three were excluded as they were of age <2 years and six were of age 20. Baseline BMI data were available for 378/501 patients (75.4%). Baseline and post-baseline BMI data were available for 328/ 501 patients (65.5%): 53.4% male, 90.6% white, median age 7.4 years. An appendicovesicostomy (APV) was present in 40.9%, 10.9% had a spiral umbilical Monti, and 49.0% had a standard Monti. A primary or secondary diagnosis of myelomeningocele had been given to 59.8%. Wheelchair-bound patients comprised 23.8%. Those with missing BMI data were significantly older (mean age 10.2 vs. 9.0 years, p Z 0.0001), were more likely to have myelomeningocele (64.2% vs. 59.5%, p < 0.0001), and were more likely to be wheelchair bound (43.9% vs. 27.5%, p < 0.0001) than those with BMI data available. Median follow-up was 76.4 months. At least one subfascial revision was present in 38/328 patients (11.6%).
Baseline obesity Of those with baseline BMI data available, 130/378 (34.4%) were overweight or obese at the time of CCU channel surgery. Overweight/obese patients were significantly more likely to undergo umbilical Monti (10% vs. 8.1%), nonumbilical spiral Monti (33.8% vs. 13.7%), and spiral umbilical Monti channels (13.8% vs. 7.3%) compared with low/ normal weight patients (p < 0.0001) (Table 2). The median BMI percentile in patients with spiral umbilical Monti channels was significantly higher than in the APV group (84.7% vs. 56.6%, p Z 0.002). Overweight/obese patients were also significantly older (median age 8.7 vs. 7.3 years, p Z 0.01), more likely to have myelomeningocele (76.9% vs. 50.4%, p < 0.0001), and more likely to be wheelchairbound (35.5% vs. 23.3%, p < 0.0001) than low/normal weight patients (Table 2). Of the patients, 64/501 (12.8%) had a subfascial revision. There was no association between either baseline weight category or median BMI percentile and subfascial revision on bivariate analysis (Table 3). Confirming our previous work, patients with spiral umbilical Monti channels were significantly more likely to have a subfascial revision than other types of channels (28.6% vs. 10.8%, p Z 0.004) (Table 3). From a multivariable Cox proportional hazard model
249.e3 Table 1 Development of overweight/obesity in postoperative period is not associated with subfascial revisions. Subfascial revision n (row %) Overweight/ obese remained overweight/obese Low/normal became overweight/obese Low/normal remained Low/normal Overweight/obese became low/normal Total
Total n (%)
9 (10.3)
87 (26.5)
4 (7.8)
51 (15.5)
19 (11.6)
164 (50.0)
6 (23.1) 38 (11.6)
26 (7.9) 328
controlling for age, baseline BMI percentile category, myelomeningocele diagnosis, and ambulatory status, the hazard of subfascial revision for spiral umbilical Monti channels was 2.1 times that of other channels (hazard ratio (HR) 2.1 [95% CI 1.2e3.8], p Z 0.01). No other variable, including baseline BMI percentile category, was significantly associated with time to subfascial revision in the Cox model (overweight/obese HR 1.2 [95% CI 0.6e2.3], p Z 0.6) (Fig. 1). We performed a sub-analysis of patients with baseline BMI percentile data available (n Z 378). From a multivariable Cox proportional hazard model controlling for the same factors as above, the hazard of subfascial revision for spiral umbilical Monti channels was 2.2 times that of other channels (HR 2.2 [95% CI 1.1e4.6], p Z 0.03). No other variable, including baseline BMI percentile category, was significantly associated with time to subfascial revision in the Cox model (overweight/obese HR 1.2 [95% CI 0.6e2.2], p Z 0.7).
Postoperative obesity Of those with baseline and post-baseline BMI data available, 113/328 (34.5%) were overweight or obese at baseline. Fifty-one of the 328 (15.6%) became overweight or obese postoperatively (Table 1). Four of the 51 (7.8%) of those who became overweight or obese had a subfascial revision versus 34/277 (12.3%) of those whose weight category decreased or remained stable (p Z 0.3).
Discussion This study demonstrated that, contrary to our hypothesis, baseline obesity or becoming overweight or obese in the postoperative period, as measured by BMI percentile, is not associated with subfascial revision following CCU channel surgery in children. The majority of our patients had stable weight at their most recent postoperative visit, with only 16% becoming overweight or obese during postoperative follow-up. It is also interesting that patients who lost weight after surgery were not at increased risk for channel revision compared with the remainder of the cohort. Any development of redundancy in the channel because of shortening of the distance between the bladder and the surface did not translate into an increased risk of channel
249.e4 Table 2
K.H. Chan et al. Comparison of patients by BMI categories.
Age
Gender Race Primary diagnosis
Myelomeningocelea Channel type
Ambulatory status
N Mean SD Median (min, max) Female Male Non-Caucasian Caucasian Bladder exstrophy/isolated epispadias Cloacal exstrophy Cloacal anomaly Posterior urethral valves Imperforate anus Non-neurogenic neurogenic bladder Other spinal abnormality Oncologic Myelomeningocele Sacral agenesis Tethered cord Caudal regression syndrome Cerebral palsy Other bladder anomaly Trauma Other No Yes Non-umbilical APV Umbilical APV Non-umbilical Monti Umbilical Monti Non-umbilical spiral Monti Spiral umbilical Monti Missing Wheelchair bound Ambulates with assist (braces/crutches) Indep. ambulation
BMI low/healthy
BMI overweight/obese
248 8.4 3.9 7.3 (2.4, 19.6) 111 (44.8) 137 (55.2) 25 (10.1) 223 (89.9) 24 (9.7) 4 (1.6) 11 (4.4) 16 (6.5) 9 (3.6) 8 (3.2) 0 (0) 4 (1.6) 125 (50.4) 7 (2.8) 7 (2.8) 1 (0.4) 7 (2.8) 12 (4.8) 9 (3.6) 4 (1.6) 123 (49.6) 125 (50.4) 77 (31) 44 (17.7) 55 (22.2) 20 (8.1) 34 (13.7) 18 (7.3) 16 54 (23.3) 49 (21.1) 129 (55.6)
130 9.5 4.1 8.7 (2.8, 19.6) 65 (50) 65 (50) 9 (6.9) 121 (93.1) 5 (3.8) 2 (1.5) 3 (2.3) 2 (1.5) 3 (2.3) 2 (1.5) 2 (1.5) 1 (0.8) 99 (76.2) 2 (1.5) 0 (0) 0 (0) 1 (0.8) 0 (0) 5 (3.8) 3 (2.3) 30 (23.1) 100 (76.9) 16 (12.3) 20 (15.4) 19 (14.6) 13 (10) 44 (33.8) 18 (13.8) 9 43 (35.5) 44 (36.4) 34 (28.1)
p-value 0.01
0.3 0.3
<0.0001 <0.0001
<0.0001
APV Z appendicovesicostomy. a Spina bifida includes myelomeningocele and lipomeningocele.
revision. We also found that surgeons at our institution were more likely to select standard or spiral Monti channels than APVs for overweight/obese patients. This was likely a result of the technical challenge of creating a sufficiently long channel in patients with a greater distance from the bladder to the skin level because of a thicker abdominal wall or the thickened and poorly mobile mesentery to the appendix. Prior studies about the effects of postoperative weight gain have focused primarily on the development of medical complications [10,11]. There is conflicting evidence in the literature about the association between baseline obesity and postoperative complications following CCU channel surgery in children. In a study of myelodysplasia patients undergoing urinary or fecal reconstructive procedures, Donovan et al. noted a significant association between the presence of any medical or surgical postoperative complication and baseline obesity [7]. Despite demonstrating a potential association, they did not identify which type of complication (i.e., medical versus surgical) is most strongly
associated with baseline obesity. Other authors have found no significant association between baseline weight and channel-related complications [6,12]. Our study has several limitations. There may be comorbidities related to the postoperative weight gain that we did not measure. For example, many children who are overweight or obese develop hypertension, dyslipidemia, fatty liver disease, diabetes, polycystic ovary syndrome, sleep apnea, and other comorbidities [13]. We were also unable to provide any further details about factors that may affect postoperative weight gain, such as diet history and/or nutritional status. Our study highlights the importance of screening for obesityrelated comorbidities given that 42% of our surgical cohort was overweight or obese at the most recent postoperative visit. Second, there is potential for inaccuracy in the measurement of BMI, particularly in wheelchair-bound patients who may not be measured in a standing position. Furthermore, BMI may not be the ideal marker of obesity in spina bifida patients given the reduced bone and muscle mass in this population [3]. Despite these limitations, we chose to use BMI
Comparison of patients with and without a subfascial revision.
Age
Gender Race Primary diagnosis category
Myelomeningocelea Channel type and location
BMI category
Median BMI percentile (min, max) Ambulatory status
Overall
N Mean SD Median (min, max) Female Male Non-Caucasian Caucasian Bladder exstrophy/isolated epispadias Cloacal exstrophy Cloacal anomaly/urogenital sinus Posterior urethral valves Imperforate anus Non-neurogenic neurogenic bladder Other spinal abnormalities Oncologic Myelomeningocele Sacral agenesis Tethered cord Caudal regression syndrome Cerebral palsy Other bladder anomaly Trauma Other No Yes Non-umbilical APV Umbilical APV Non-umbilical Monti Umbilical Monti Non-umbilical spiral Monti Spiral umbilical Monti Missing Underweight/healthy Overweight/obese
501 9.1 4.2 7.9 (2.1, 19.9) 235 (46.9) 266 (53.1) 45 (9) 456 (91) 42 (8.4) 7 (1.4) 15 (3) 23 (4.6) 14 (2.8) 12 (2.4) 3 (0.6) 5 (1) 303 (60.5) 14 (2.8) 7 (1.4) 2 (0.4) 9 (1.8) 13 (2.6) 24 (4.8) 8 (1.6) 197 (39.3) 304 (60.7) 114 (22.8) 96 (19.2) 95 (19) 48 (9.6) 92 (18.4) 56 (11.2) 123 248 (65.6) 130 (34.4) 68.3 (0.0, 99.9) 34 147 (31.5) 114 (24.4) 206 (44.1)
Missing Wheelchair-bound Ambulates with assistance (braces, crutches) Ambulates independently
APV Z appendicovesicostomy. a Primary or secondary diagnosis of myelomeningocele.
No subfascial revision
Subfascial revision
n (column %)
n (column %)
437 9.2 4.2 8.0 (2.1, 19.9) 205 (46.9) 232 (53.1) 40 (9.2) 397 (90.8) 38 (8.7) 6 (1.4) 14 (3.2) 22 (5) 13 (3) 11 (2.5) 3 (0.7) 5 (1.1) 258 (59) 11 (2.5) 7 (1.6) 1 (0.2) 8 (1.8) 11 (2.5) 22 (5) 7 (1.6) 178 (40.7) 259 (59.3) 106 (24.3) 84 (19.2) 81 (18.5) 43 (9.8) 83 (19) 40 (9.2) 103 222 (66.5) 112 (33.5) 68.3 (0.0, 99.9) 32 122 (30.1) 98 (24.2) 185 (45.7)
64 8.5 4.0 7.5 (2.6, 19.7) 30 (46.9) 34 (53.1) 5 (7.8) 59 (92.2) 4 (6.3) 1 (1.6) 1 (1.6) 1 (1.6) 1 (1.6) 1 (1.6) 0 (0) 0 (0) 45 (70.3) 3 (4.7) 0 (0) 1 (1.6) 1 (1.6) 2 (3.1) 2 (3.1) 1 (1.6) 19 (29.7) 45 (70.3) 8 (12.5) 12 (18.8) 14 (21.9) 5 (7.8) 9 (14.1) 16 (25) 20 26 (59.1) 18 (40.9) 74.1 (1.0, 99.7) 2 25 (40.3) 16 (25.8) 21 (33.9)
p-value
0.21 0.99 0.73
0.09 0.004
0.33
0.3 0.17
249.e5
Response category/statistic
Weight gain after urinary channel surgery
Table 3
249.e6
K.H. Chan et al.
Figure 1
Probability of revision-free survival by channel type.
percentile as a measure of obesity because it was readily available for data extraction in a retrospective manner. Finally, we measured BMI at only two time points with a variable interval between baseline and postoperative measurements. We could not account for fluctuations in weight nor were we able to control for bias resulting from differential follow-up times using censoring methods in a survival analysis. To our knowledge, this is the first study to analyze the risks of channel-related complications associated with becoming overweight or obese in the postoperative period in a large pediatric cohort undergoing lower urinary tract reconstruction. In our current practice, we preferentially utilize the appendix, when possible, for CCU channels in obese patients in light of the findings presented in this study. Future studies are needed to further explore the effects of obesity on surgical outcomes for children, particularly in the myelodysplasia population.
Conclusion Overweight/obese patients are more likely to require types of channels that have a significantly increased risk of subfascial revision compared with appendicovesicostomies. Patients who became overweight or obese postoperatively, however, were not at increased risk of subfascial revision.
Conflict of interest None.
Funding None.
References [1] Cain MP. Reconstruction for urinary and fecal incontinence in the pediatric patient: what we can accomplish today and a look to the future. Medscape. http://www.medscape.com/ viewarticle/499029. [2] Leslie JA, Cain MP, Kaefer M, Meldrum KK, Dussinger AM, Rink RC, et al. A comparison of the Monti and Casale (spiral Monti) procedures. J Urol 2007;178(4 Pt 2):1623e7. discussion 1627. [3] Thomas JC, Dietrich MS, Trusler L, DeMarco RT, Pope 4th JC, Brock 3rd JW, et al. Continent catheterizable channels and the timing of their complications. J Urol 2006;176(4 Pt 2): 1816e20. discussion 1820. [4] Leslie B, Lorenzo AJ, Moore K, Farhat WA, Ba ¨gli DJ, Pippi Salle JL. Long-term followup and time to event outcome analysis of continent catheterizable channels. J Urol 2011; 185(6):2298e302. [5] Szymanksi KM, Whittam B, Misseri R, Flack CK, Hubert KC, Kaefer M, et al. Long-term outcomes of catheterizable continent urinary channels: what do you use, where do you put it and does it matter? J Pediatr Urol 2015;11(4). 210.e1e7. [6] Clark T, Pope 4th JC, Adams mC, Wells N, Brock 3rd JW. Factors that influence outcomes of the Mitrofanoff and Malone antegrade continence enema reconstructive procedures in children. J Urol 2002;168(4 Pt 1):1537e40. discussion 1540. [7] Donovan BO, Boci M, Kropp BP, Bright BC, Roth CC, Confer SD, et al. Body mass index as a predictive value for complications associated with reconstructive surgery in patients with myelodysplasia. J Urol 2009;181(5):2272e5. discussion 2276. [8] Centers for Disease Control. Defining childhood obesity. http://www.cdc.gov/obesity/childhood/basics.html. [9] Must A, Anderson SE. Body mass index in children and adolescents: considerations for population-based applications. Int J Obes (Lond) 2006;30(4):590e4. [10] Grady KL, Naftel D, Pamboukian SV, Frazier OH, Hauptman P, Herre J, et al. Post-operative obesity and cachexia are risk factors for morbidity and mortality after heart transplant: multi-institutional study of post-operative weight change. J Heart Lung Transpl 2005;24(9):1424e30.
Weight gain after urinary channel surgery [11] Kugler C, Einhorn I, Gottlieb J, Warnecke G, Schwarz A, BargHock H, et al. Postoperative weight gain during the first year after kidney, liver, heart, and lung transplant: a prospective study. Prog Transpl 2015;25(1):49e55. [12] Barqawi A, de Valdenebro M, Furness 3rd PD, Koyle MA. Lessons learned from stomal complications in children with
249.e7 cutaneous catheterizable continent stomas. BJU Int 2004; 94(9):1344e7. [13] Estrada E, Eneli I, Hampl S, Mietus-Snyder M, Mirza N, Rhodes E, et al. Children’s Hospital Association consensus statements for comorbidities of childhood obesity. Child Obes 2014;10(4):304e17.
Effect of baseline obesity and postoperative weight gain on the risk of channel revision following continent catheterizable urinary channel surgery a
Division of Pediatric Urology, Riley Hospital for Children at Indiana University Health, Indiana University School of Medicine, Indianapolis, IN, USA
Katherine H. Chan a,b, Konrad M. Szymanski a, Xiaochun Li c, Susan Ofner c, Chandra Flack a, Benjamin Judge a, Benjamin Whittam a, Rosalia Misseri a, Martin Kaefer a, Richard C. Rink a, Mark P. Cain a
b
Department of Biostatistics, Indiana University-Purdue University, Indianapolis, IN, USA
c
Center for Pediatric and Adolescent Comparative Effectiveness Research, Arkanasas, USA Correspondence to: K.H. Chan, 705 Riley Hospital Drive, Division of Pediatric Urology, Riley Outpatient Center, 4230 Indianapolis, IN 46202, USA [email protected] (K.H. Chan) Keywords Obesity; Urinary bladder; Neurogenic; Postoperative complications Received 30 December 2015 Accepted 22 May 2016 Available online 11 June 2016
Summary
Objective Prior studies suggest that obese patients are at increased risk for complications following continent catheterizable urinary (CCU) channel surgery. We hypothesized that postoperative weight gain increases the risk of channel angulation, difficulty catheterizing, and possible channel perforation requiring subfascial revision. The purpose of this study was to evaluate whether baseline obesity or becoming overweight/obese postoperatively was associated with a greater risk of subfascial revision.
Method We reviewed retrospectively an institutional database of patients who underwent CCU channel surgery between the ages of 2 and <20 years from January 1990 to May 2013, excluding those with continent urinary reservoirs, continent vesicostomies, and those without body mass index (BMI) data. We collected data on patient/procedure characteristics, baseline/most recent BMI, and subfascial revision(s). We used Cox proportional hazard multivariable regression to assess the association of being overweight/obese at baseline (85% BMI) with time to first subfascial revision, and Fisher’s exact test to compare rates of subfascial revision between those who became overweight/obese and those who did not.
Results Of the patients, 328/501 (65.5%) had baseline and post-baseline BMI data available: 53.4% male, 90.6% white, median age 7.4 years; median follow-up 76.4 months. Of the 328 patients, 38 (11.6%) had subfascial revisions. Baseline BMI data were available for 378 patients, and, of these, 130 (34.4%) were overweight/obese at baseline. Overweight/obese patients were more likely to undergo umbilical Monti (10% vs. 8.1%), non-umbilical spiral Monti (33.8% vs. 13.7%), and spiral umbilical Monti channels (13.8% vs. 7.3%) versus normal/underweight patients (p < 0.0001). From a multivariable Cox proportional hazard model controlling for age, BMI category, diagnosis, and ambulatory status, the hazard of subfascial revision for spiral umbilical Monti channels was 2.1 that of other channels (hazard ratio (HR) 2.1 [95% CI 1.2e3.8], p Z 0.01). Fifty-one out of 328 patients (15.6%) became overweight/obese postoperatively, with 7.8% having a subfascial revision vs. 12.3% of those whose weight category decreased or remained stable (p Z 0.3) (Table 1). Conclusions Patients who were overweight/obese at baseline were more likely to have channels constructed that are at the highest risk of subfascial revision. Patients who became overweight/obese postoperatively were not at greater risk of subfascial revision. Limitations include potential bias because of differential follow-up and inaccuracy of BMI percentile as a measurement of obesity.
http://dx.doi.org/10.1016/j.jpurol.2016.05.021 1477-5131/ª 2016 Published by Elsevier Ltd on behalf of Journal of Pediatric Urology Company.
249.e2
Introduction A number of postoperative complications have been reported after creation of continent catheterizable urinary (CCU) channels, and these can be categorized into either stomal complications (e.g., stenosis or prolapse) or subfascial complications (e.g., channel angulation, diverticulum, or incontinence) [1e4]. We previously reported that the increased risk of subfascial revision in Monti channels, particularly the longer spiral umbilical Monti channels, is believed to be secondary to a longer, unsupported extravesical channel [5]. Prior studies have suggested that baseline obesity may increase the risk of postoperative complications following lower urinary tract reconstruction [6,7]. In a study of myelodysplasia patients undergoing lower urinary tract reconstruction, Donovan et al. noted a significant association between baseline obesity and the presence of any medical or surgical postoperative complication [7]. Clark et al. noted an association between percentile weight and stomal stenosis rate following CCU channel procedures, although the differences did not reach statistical significance [6]. To our knowledge there are no prior studies on the effects of postoperative weight gain in children undergoing lower urinary tract reconstruction. We hypothesized that patients with CCU channels who gain weight in the postoperative period may increase the distance from the bladder to the skin (particularly with non-umbilical stomas) resulting in channel angulation, difficulty catheterizing, and possible channel perforation. This type of suprafascial issue may require channel revision including subfascial repair. The purpose of this study was to examine whether patients who are obese at the time of surgery or who become overweight or develop obesity postoperatively, have an increased risk of subfascial channel revision.
Materials and methods We included all patients 2 years of age and <20 years of age from our retrospective institutional database who underwent CCU channels at our institution between January 1990 and May 2013. The database was constructed from a combination of billing data, review of electronic medical records, and paper charts. Patients with continent urinary reservoirs and channels constructed of tissue other than appendix or ileum (e.g., colon or ureter or continent vesicostomies) were excluded from the database. We also excluded those with missing height or weight data as their body mass index (BMI) could not be calculated. Patients <2 years of age and 20 years of age at the time of surgery were excluded from the study as it is not possible to calculate BMI in these age groups using the Centers for Disease Control (CDC) pediatric criteria [8]. The primary outcome was the need for at least one subfascial revision, defined as a need for laparotomy to correct difficult catheterization because of issues such as channel angulation, perforation, diverticulum, false passage, or stenosis. Demographics, clinical and CCU channel
K.H. Chan et al. characteristics (e.g., stomal location), and follow-up duration were abstracted from the institutional database and patient charts. Height and weight data were identified from retrospective review of hospital records. Preoperative or “baseline” BMI was defined as a patient’s BMI measured at the closest time point within 1 year of the date of CCU channel surgery. Postoperative BMI was defined as the BMI at the most recent postoperative visit or within 1 year of subfascial revision or channel excision. The 1-year perioperative timeframe was chosen to maximize the number of patients with available BMI data, while minimizing inaccuracies of BMI data that are not representative of true baseline or postoperative BMI. BMI was calculated according to the standard formula, BMI Z kg/m2. Weight categories were defined for pediatric patients 2 years and <20 years based on an age- and genderspecific BMI percentile according to CDC criteria [8]. BMI percentile, equivalent to the BMI z-score, was used because children’s body composition varies by age and between boys and girls [9]. Patients were categorized as low/normal weight (<85th BMI percentile) and overweight/obese (85th BMI percentile) [8]. A categorical variable was created for all patients in the cohort using the following categories: a) became overweight/obese, b) maintained stable low/normal BMI, c) maintained overweight/obese BMI, and d) became low/normal (overweight/obese at baseline). We used Fisher’s exact test to examine the association between baseline BMI percentile category and channel type.
Data analysis of baseline obesity Demographic and clinical data were compared between those with and without a first revision by use of Fisher’s exact and chi-square tests, to elucidate risk factors for subfascial revisions. Continuous characteristics were analyzed using an analysis of variance model or Wilcoxon rank sum test as appropriate based on data characteristics. We examined the association between BMI percentile category and channel type using Fisher’s exact test. We also compared demographic and clinical data between different BMI percentile categories including: a) missing BMI percentile vs. underweight/healthy BMI percentile, and b) overweight/obese BMI percentile vs. underweight/ healthy BMI percentile. We assessed the association of the baseline BMI percentile category with time to the first subfascial revision, our primary outcome, controlling for demographic and clinical characteristics, using Cox proportional hazards regression. Patient characteristics of age, gender, race, BMI percentile category, presence or absence of myelomeningocele, ambulatory status, channel type, and stomal location were thought to be clinically important. These variables were selected, a priori, for inclusion in the model of time to first subfascial revision. The proportional hazards assumption was assessed for each covariate by graphical means. Variables (gender and race) for which the assumption did not hold were removed from the multivariable model. The final Cox model included age, BMI percentile category, spiral Monti channel at the umbilicus, myelomeningocele diagnosis, and ambulatory status.
Weight gain after urinary channel surgery
Data analysis of postoperative obesity We also compared rates of subfascial revision respectively between those who became overweight/obese (BMI 85%) versus those who did not, using Fisher’s exact test. All analyses were performed using SAS version 9.3 (SAS Institute Inc., Cary, NC, USA). A p-values less than 0.05 was considered to be significant.
Ethical approval The institutional review board approved this study.
Results Patient population Of the 510 patients in the database, three were excluded as they were of age <2 years and six were of age 20. Baseline BMI data were available for 378/501 patients (75.4%). Baseline and post-baseline BMI data were available for 328/ 501 patients (65.5%): 53.4% male, 90.6% white, median age 7.4 years. An appendicovesicostomy (APV) was present in 40.9%, 10.9% had a spiral umbilical Monti, and 49.0% had a standard Monti. A primary or secondary diagnosis of myelomeningocele had been given to 59.8%. Wheelchair-bound patients comprised 23.8%. Those with missing BMI data were significantly older (mean age 10.2 vs. 9.0 years, p Z 0.0001), were more likely to have myelomeningocele (64.2% vs. 59.5%, p < 0.0001), and were more likely to be wheelchair bound (43.9% vs. 27.5%, p < 0.0001) than those with BMI data available. Median follow-up was 76.4 months. At least one subfascial revision was present in 38/328 patients (11.6%).
Baseline obesity Of those with baseline BMI data available, 130/378 (34.4%) were overweight or obese at the time of CCU channel surgery. Overweight/obese patients were significantly more likely to undergo umbilical Monti (10% vs. 8.1%), nonumbilical spiral Monti (33.8% vs. 13.7%), and spiral umbilical Monti channels (13.8% vs. 7.3%) compared with low/ normal weight patients (p < 0.0001) (Table 2). The median BMI percentile in patients with spiral umbilical Monti channels was significantly higher than in the APV group (84.7% vs. 56.6%, p Z 0.002). Overweight/obese patients were also significantly older (median age 8.7 vs. 7.3 years, p Z 0.01), more likely to have myelomeningocele (76.9% vs. 50.4%, p < 0.0001), and more likely to be wheelchairbound (35.5% vs. 23.3%, p < 0.0001) than low/normal weight patients (Table 2). Of the patients, 64/501 (12.8%) had a subfascial revision. There was no association between either baseline weight category or median BMI percentile and subfascial revision on bivariate analysis (Table 3). Confirming our previous work, patients with spiral umbilical Monti channels were significantly more likely to have a subfascial revision than other types of channels (28.6% vs. 10.8%, p Z 0.004) (Table 3). From a multivariable Cox proportional hazard model
249.e3 Table 1 Development of overweight/obesity in postoperative period is not associated with subfascial revisions. Subfascial revision n (row %) Overweight/ obese remained overweight/obese Low/normal became overweight/obese Low/normal remained Low/normal Overweight/obese became low/normal Total
Total n (%)
9 (10.3)
87 (26.5)
4 (7.8)
51 (15.5)
19 (11.6)
164 (50.0)
6 (23.1) 38 (11.6)
26 (7.9) 328
controlling for age, baseline BMI percentile category, myelomeningocele diagnosis, and ambulatory status, the hazard of subfascial revision for spiral umbilical Monti channels was 2.1 times that of other channels (hazard ratio (HR) 2.1 [95% CI 1.2e3.8], p Z 0.01). No other variable, including baseline BMI percentile category, was significantly associated with time to subfascial revision in the Cox model (overweight/obese HR 1.2 [95% CI 0.6e2.3], p Z 0.6) (Fig. 1). We performed a sub-analysis of patients with baseline BMI percentile data available (n Z 378). From a multivariable Cox proportional hazard model controlling for the same factors as above, the hazard of subfascial revision for spiral umbilical Monti channels was 2.2 times that of other channels (HR 2.2 [95% CI 1.1e4.6], p Z 0.03). No other variable, including baseline BMI percentile category, was significantly associated with time to subfascial revision in the Cox model (overweight/obese HR 1.2 [95% CI 0.6e2.2], p Z 0.7).
Postoperative obesity Of those with baseline and post-baseline BMI data available, 113/328 (34.5%) were overweight or obese at baseline. Fifty-one of the 328 (15.6%) became overweight or obese postoperatively (Table 1). Four of the 51 (7.8%) of those who became overweight or obese had a subfascial revision versus 34/277 (12.3%) of those whose weight category decreased or remained stable (p Z 0.3).
Discussion This study demonstrated that, contrary to our hypothesis, baseline obesity or becoming overweight or obese in the postoperative period, as measured by BMI percentile, is not associated with subfascial revision following CCU channel surgery in children. The majority of our patients had stable weight at their most recent postoperative visit, with only 16% becoming overweight or obese during postoperative follow-up. It is also interesting that patients who lost weight after surgery were not at increased risk for channel revision compared with the remainder of the cohort. Any development of redundancy in the channel because of shortening of the distance between the bladder and the surface did not translate into an increased risk of channel
249.e4 Table 2
K.H. Chan et al. Comparison of patients by BMI categories.
Age
Gender Race Primary diagnosis
Myelomeningocelea Channel type
Ambulatory status
N Mean SD Median (min, max) Female Male Non-Caucasian Caucasian Bladder exstrophy/isolated epispadias Cloacal exstrophy Cloacal anomaly Posterior urethral valves Imperforate anus Non-neurogenic neurogenic bladder Other spinal abnormality Oncologic Myelomeningocele Sacral agenesis Tethered cord Caudal regression syndrome Cerebral palsy Other bladder anomaly Trauma Other No Yes Non-umbilical APV Umbilical APV Non-umbilical Monti Umbilical Monti Non-umbilical spiral Monti Spiral umbilical Monti Missing Wheelchair bound Ambulates with assist (braces/crutches) Indep. ambulation
BMI low/healthy
BMI overweight/obese
248 8.4 3.9 7.3 (2.4, 19.6) 111 (44.8) 137 (55.2) 25 (10.1) 223 (89.9) 24 (9.7) 4 (1.6) 11 (4.4) 16 (6.5) 9 (3.6) 8 (3.2) 0 (0) 4 (1.6) 125 (50.4) 7 (2.8) 7 (2.8) 1 (0.4) 7 (2.8) 12 (4.8) 9 (3.6) 4 (1.6) 123 (49.6) 125 (50.4) 77 (31) 44 (17.7) 55 (22.2) 20 (8.1) 34 (13.7) 18 (7.3) 16 54 (23.3) 49 (21.1) 129 (55.6)
130 9.5 4.1 8.7 (2.8, 19.6) 65 (50) 65 (50) 9 (6.9) 121 (93.1) 5 (3.8) 2 (1.5) 3 (2.3) 2 (1.5) 3 (2.3) 2 (1.5) 2 (1.5) 1 (0.8) 99 (76.2) 2 (1.5) 0 (0) 0 (0) 1 (0.8) 0 (0) 5 (3.8) 3 (2.3) 30 (23.1) 100 (76.9) 16 (12.3) 20 (15.4) 19 (14.6) 13 (10) 44 (33.8) 18 (13.8) 9 43 (35.5) 44 (36.4) 34 (28.1)
p-value 0.01
0.3 0.3
<0.0001 <0.0001
<0.0001
APV Z appendicovesicostomy. a Spina bifida includes myelomeningocele and lipomeningocele.
revision. We also found that surgeons at our institution were more likely to select standard or spiral Monti channels than APVs for overweight/obese patients. This was likely a result of the technical challenge of creating a sufficiently long channel in patients with a greater distance from the bladder to the skin level because of a thicker abdominal wall or the thickened and poorly mobile mesentery to the appendix. Prior studies about the effects of postoperative weight gain have focused primarily on the development of medical complications [10,11]. There is conflicting evidence in the literature about the association between baseline obesity and postoperative complications following CCU channel surgery in children. In a study of myelodysplasia patients undergoing urinary or fecal reconstructive procedures, Donovan et al. noted a significant association between the presence of any medical or surgical postoperative complication and baseline obesity [7]. Despite demonstrating a potential association, they did not identify which type of complication (i.e., medical versus surgical) is most strongly
associated with baseline obesity. Other authors have found no significant association between baseline weight and channel-related complications [6,12]. Our study has several limitations. There may be comorbidities related to the postoperative weight gain that we did not measure. For example, many children who are overweight or obese develop hypertension, dyslipidemia, fatty liver disease, diabetes, polycystic ovary syndrome, sleep apnea, and other comorbidities [13]. We were also unable to provide any further details about factors that may affect postoperative weight gain, such as diet history and/or nutritional status. Our study highlights the importance of screening for obesityrelated comorbidities given that 42% of our surgical cohort was overweight or obese at the most recent postoperative visit. Second, there is potential for inaccuracy in the measurement of BMI, particularly in wheelchair-bound patients who may not be measured in a standing position. Furthermore, BMI may not be the ideal marker of obesity in spina bifida patients given the reduced bone and muscle mass in this population [3]. Despite these limitations, we chose to use BMI
Comparison of patients with and without a subfascial revision.
Age
Gender Race Primary diagnosis category
Myelomeningocelea Channel type and location
BMI category
Median BMI percentile (min, max) Ambulatory status
Overall
N Mean SD Median (min, max) Female Male Non-Caucasian Caucasian Bladder exstrophy/isolated epispadias Cloacal exstrophy Cloacal anomaly/urogenital sinus Posterior urethral valves Imperforate anus Non-neurogenic neurogenic bladder Other spinal abnormalities Oncologic Myelomeningocele Sacral agenesis Tethered cord Caudal regression syndrome Cerebral palsy Other bladder anomaly Trauma Other No Yes Non-umbilical APV Umbilical APV Non-umbilical Monti Umbilical Monti Non-umbilical spiral Monti Spiral umbilical Monti Missing Underweight/healthy Overweight/obese
501 9.1 4.2 7.9 (2.1, 19.9) 235 (46.9) 266 (53.1) 45 (9) 456 (91) 42 (8.4) 7 (1.4) 15 (3) 23 (4.6) 14 (2.8) 12 (2.4) 3 (0.6) 5 (1) 303 (60.5) 14 (2.8) 7 (1.4) 2 (0.4) 9 (1.8) 13 (2.6) 24 (4.8) 8 (1.6) 197 (39.3) 304 (60.7) 114 (22.8) 96 (19.2) 95 (19) 48 (9.6) 92 (18.4) 56 (11.2) 123 248 (65.6) 130 (34.4) 68.3 (0.0, 99.9) 34 147 (31.5) 114 (24.4) 206 (44.1)
Missing Wheelchair-bound Ambulates with assistance (braces, crutches) Ambulates independently
APV Z appendicovesicostomy. a Primary or secondary diagnosis of myelomeningocele.
No subfascial revision
Subfascial revision
n (column %)
n (column %)
437 9.2 4.2 8.0 (2.1, 19.9) 205 (46.9) 232 (53.1) 40 (9.2) 397 (90.8) 38 (8.7) 6 (1.4) 14 (3.2) 22 (5) 13 (3) 11 (2.5) 3 (0.7) 5 (1.1) 258 (59) 11 (2.5) 7 (1.6) 1 (0.2) 8 (1.8) 11 (2.5) 22 (5) 7 (1.6) 178 (40.7) 259 (59.3) 106 (24.3) 84 (19.2) 81 (18.5) 43 (9.8) 83 (19) 40 (9.2) 103 222 (66.5) 112 (33.5) 68.3 (0.0, 99.9) 32 122 (30.1) 98 (24.2) 185 (45.7)
64 8.5 4.0 7.5 (2.6, 19.7) 30 (46.9) 34 (53.1) 5 (7.8) 59 (92.2) 4 (6.3) 1 (1.6) 1 (1.6) 1 (1.6) 1 (1.6) 1 (1.6) 0 (0) 0 (0) 45 (70.3) 3 (4.7) 0 (0) 1 (1.6) 1 (1.6) 2 (3.1) 2 (3.1) 1 (1.6) 19 (29.7) 45 (70.3) 8 (12.5) 12 (18.8) 14 (21.9) 5 (7.8) 9 (14.1) 16 (25) 20 26 (59.1) 18 (40.9) 74.1 (1.0, 99.7) 2 25 (40.3) 16 (25.8) 21 (33.9)
p-value
0.21 0.99 0.73
0.09 0.004
0.33
0.3 0.17
249.e5
Response category/statistic
Weight gain after urinary channel surgery
Table 3
249.e6
K.H. Chan et al.
Figure 1
Probability of revision-free survival by channel type.
percentile as a measure of obesity because it was readily available for data extraction in a retrospective manner. Finally, we measured BMI at only two time points with a variable interval between baseline and postoperative measurements. We could not account for fluctuations in weight nor were we able to control for bias resulting from differential follow-up times using censoring methods in a survival analysis. To our knowledge, this is the first study to analyze the risks of channel-related complications associated with becoming overweight or obese in the postoperative period in a large pediatric cohort undergoing lower urinary tract reconstruction. In our current practice, we preferentially utilize the appendix, when possible, for CCU channels in obese patients in light of the findings presented in this study. Future studies are needed to further explore the effects of obesity on surgical outcomes for children, particularly in the myelodysplasia population.
Conclusion Overweight/obese patients are more likely to require types of channels that have a significantly increased risk of subfascial revision compared with appendicovesicostomies. Patients who became overweight or obese postoperatively, however, were not at increased risk of subfascial revision.
Conflict of interest None.
Funding None.
References [1] Cain MP. Reconstruction for urinary and fecal incontinence in the pediatric patient: what we can accomplish today and a look to the future. Medscape. http://www.medscape.com/ viewarticle/499029. [2] Leslie JA, Cain MP, Kaefer M, Meldrum KK, Dussinger AM, Rink RC, et al. A comparison of the Monti and Casale (spiral Monti) procedures. J Urol 2007;178(4 Pt 2):1623e7. discussion 1627. [3] Thomas JC, Dietrich MS, Trusler L, DeMarco RT, Pope 4th JC, Brock 3rd JW, et al. Continent catheterizable channels and the timing of their complications. J Urol 2006;176(4 Pt 2): 1816e20. discussion 1820. [4] Leslie B, Lorenzo AJ, Moore K, Farhat WA, Ba ¨gli DJ, Pippi Salle JL. Long-term followup and time to event outcome analysis of continent catheterizable channels. J Urol 2011; 185(6):2298e302. [5] Szymanksi KM, Whittam B, Misseri R, Flack CK, Hubert KC, Kaefer M, et al. Long-term outcomes of catheterizable continent urinary channels: what do you use, where do you put it and does it matter? J Pediatr Urol 2015;11(4). 210.e1e7. [6] Clark T, Pope 4th JC, Adams mC, Wells N, Brock 3rd JW. Factors that influence outcomes of the Mitrofanoff and Malone antegrade continence enema reconstructive procedures in children. J Urol 2002;168(4 Pt 1):1537e40. discussion 1540. [7] Donovan BO, Boci M, Kropp BP, Bright BC, Roth CC, Confer SD, et al. Body mass index as a predictive value for complications associated with reconstructive surgery in patients with myelodysplasia. J Urol 2009;181(5):2272e5. discussion 2276. [8] Centers for Disease Control. Defining childhood obesity. http://www.cdc.gov/obesity/childhood/basics.html. [9] Must A, Anderson SE. Body mass index in children and adolescents: considerations for population-based applications. Int J Obes (Lond) 2006;30(4):590e4. [10] Grady KL, Naftel D, Pamboukian SV, Frazier OH, Hauptman P, Herre J, et al. Post-operative obesity and cachexia are risk factors for morbidity and mortality after heart transplant: multi-institutional study of post-operative weight change. J Heart Lung Transpl 2005;24(9):1424e30.
Weight gain after urinary channel surgery [11] Kugler C, Einhorn I, Gottlieb J, Warnecke G, Schwarz A, BargHock H, et al. Postoperative weight gain during the first year after kidney, liver, heart, and lung transplant: a prospective study. Prog Transpl 2015;25(1):49e55. [12] Barqawi A, de Valdenebro M, Furness 3rd PD, Koyle MA. Lessons learned from stomal complications in children with
249.e7 cutaneous catheterizable continent stomas. BJU Int 2004; 94(9):1344e7. [13] Estrada E, Eneli I, Hampl S, Mietus-Snyder M, Mirza N, Rhodes E, et al. Children’s Hospital Association consensus statements for comorbidities of childhood obesity. Child Obes 2014;10(4):304e17.