Pediatr Surg Int DOI 10.1007/s00383-015-3748-2
ORIGINAL ARTICLE
Factors associated with age at pyeloplasty in children with ureteropelvic junction obstruction Vijaya M. Vemulakonda1 • Duncan T. Wilcox1 • Timothy M. Crombleholme2 Michael Bronsert3 • Allison Kempe3
•
Accepted: 24 June 2015 Ó Springer-Verlag Berlin Heidelberg 2015
Abstract Objectives Our objectives were to: (1) delineate factors associated with surgery at \1 year and (2) determine if early intervention was associated with increased risk of readmission. Methods We conducted a retrospective review of children age 0–18 years with a diagnosis of UPJ obstruction who underwent pyeloplasty from 1/1/1999 to 9/1/2009 using the PHIS database. Data collected included patient factors (race, gender, insurance type, APR-DRG severity of illness) and hospital factors (annual case volume, census region, academic status). Outcomes assessed were: age at surgery and hospital readmission within 1 year of surgery. Data were analyzed using logistic regression and Cox PH for multivariate analyses. Results 4499 children met study criteria. Minority race (OR 1.55), male gender (OR 1.49), public insurance (OR 1.37), high severity of illness (OR 3.60), Southern region (OR 1.44), and low hospital volume (OR 1.37) were significant predictors of early surgery (p \ 0.05). Only early surgery (HR 2.42; 95 % CI 1.67–3.49 2.42) was associated with increased risk of readmission. Conclusions In children with UPJ obstruction, age at surgery is associated with patient demographic and hospital
& Vijaya M. Vemulakonda [email protected] 1
Department of Pediatric Urology, Children’s Hospital Colorado, 13123 E. 16th Ave, Campus Box B-463, Aurora, CO, USA
2
Center for Children’s Surgery, Children’s Hospital Colorado, 13123 E. 16th Ave, Aurora, CO, USA
3
Children’s Outcomes Research Program, Children’s Hospital Colorado, 13199 E. Montview Blvd., Aurora, CO, USA
factors. Early surgery is associated with higher rates of readmission, suggesting that variations in age at surgery may be associated with significant differences in outcomes. Keywords Health disparities Ureteropelvic junction obstruction Practice variability Congenital hydronephrosis Pyeloplasty Abbreviations APR-DRG All patient refined diagnosis related groups PHIS Pediatric health information system UPJ Ureteropelvic junction UTI Urinary tract infection
Introduction Ureteropelvic junction (UPJ) obstruction is the most common etiology of congenital hydronephrosis, affecting 1 in 1000 live births [1]. Although traditionally surgery was reserved for symptomatic children, antenatal ultrasound has led to earlier diagnosis and treatment of the asymptomatic patient. The benefits of early diagnosis and treatment, however, may not be equally distributed among children of different racial backgrounds. Unlike appendectomy, where nonwhite children are less likely to undergo timely intervention [2], nonwhite children have been observed to be more likely to undergo early pyeloplasty than their white, non-Hispanic counterparts [3, 4]. The etiology of racial/ethnic variation in timing of surgery is unclear and its effect on outcomes is unknown. However, this finding raises concerns about possible overtreatment of nonwhite and Hispanic children in light of
123
Pediatr Surg Int
studies suggesting that infants with suspected UPJ obstruction may be safely observed [5, 6]. The purpose of this study was to delineate differences in demographic, clinical, and hospital factors between children who undergo early intervention and those who undergo delayed intervention and to determine if differences in age at surgery are associated with differences in risk of hospital readmission. We hypothesized that variations in age at pyeloplasty would be associated with variations in demographic and hospital factors.
Methods Data source Data for this study were obtained from the pediatric health information system (PHIS), an administrative database that contains inpatient, emergency department, ambulatory surgery and observation encounter-level data from over 45 not-for-profit, tertiary care pediatric hospitals in the United States. These hospitals are affiliated with the Children’s Hospital Association (Overland Park, KS). Data quality and reliability are assured through a joint effort between the Children’s Hospital Association and participating hospitals. Portions of the data submission and data quality processes for the PHIS database are managed by Truven Health Analytics (Ann Arbor, MI). For the purposes of external benchmarking, participating hospitals provide discharge/encounter data including demographics, diagnoses, and procedures. Nearly all of these hospitals also submit resource utilization data (e.g., pharmaceuticals, imaging, and laboratory) into PHIS. Data are de-identified at the time of data submission and are subjected to a number of reliability and validity checks before being included in the database. Informed consent was not obtained due to the de-identified nature of this data. Inclusion and exclusion criteria After institutional review board approval was granted, data were obtained for children 0–18 years admitted for pyeloplasty (ICD-9 code 55.87) between January 1, 1999 and September 30, 2009 to hospitals with complete data for the study period (n = 42 hospitals). Patients with a primary ICD-9 diagnosis code for congenital UPJ obstruction (753.20, 753.21 or 753.29) were included in the study population. Patients who had additional diagnoses of other renal or ureteral anomalies, lower urinary tract anomalies, or congenital anomalies associated with hydronephrosis were excluded.
123
Variable definitions Data obtained included patient demographics (race/ethnicity, gender, insurance status), clinical characteristics (primary diagnosis, severity of disease), hospital factors (hospital census region, hospital volume, academic status), and clinical outcomes (age at time of initial pyeloplasty, hospital readmissions associated with surgical complications). Severity of disease was assessed for patients with a primary diagnosis of UPJ obstruction using the all patient refined diagnosis related groups (APR-DRG) severity scoring system. For purposes of this study, the severity of illness subclass score, which classifies the risk of potential organ system loss of function, was used to assess clinical severity of disease [7, 8]. Hospital volume was calculated as the total number of pyeloplasties performed annually and classified as low volume (\10) or high volume (C10). Academic status was defined by the presence of an affiliated pediatric urology fellowship program. Hospital readmissions associated with surgical complication were identified by a surgical complication flag as defined by the PHIS code list and either (1) a primary diagnosis associated with surgical complication, including urinary tract infection (UTI) (ICD-9 590, 595), UPJ obstruction (753.20, 753.21, 753.29), or renal or ureteral calculus (590), or (2) a procedure associated with surgical complication, including ureteral stent placement (59.8), percutaneous nephrostomy placement (55.03), endopyelotomy (55.1), pyeloplasty (55.87), or nephrectomy (55.4, 55.5). Because individual identifiers are not linked between different PHIS hospitals, all readmissions were limited the hospital where initial surgery was performed after discharge for the principal procedure and within the 12 months following initial surgery [9]. Data analysis Primary outcomes analyzed were: age at surgery and risk of hospital readmission within 1 year of initial surgery. Age at surgery was analyzed as both a continuous and a dichotomous variable. In the dichotomous model, patients \1 year old at time of pyeloplasty were classified as having early surgery. All others were classified as having delayed surgery. Bivariate analyses were performed using Pearson’s v2 for dichotomous variables. Logistic regression was performed to further delineate the association of demographic, clinical, and hospital factors with age at surgery and risk of readmission. For multivariable analysis, variables that were significant on bivariate analysis (p \ 0.05) were utilized to create the logistic regression model. All analyses were performed using SASÒ software version 9.3 (Cary, NC).
Pediatr Surg Int
Analyses were initially conducted with all patients meeting study criteria included regardless of missing data. These results were then compared to the results of analyses with patients with missing data excluded. After confirming that exclusion of children with missing data did not significantly impact study results, the final subset of children for analysis was limited to children with complete data for analysis.
Results
severity of illness score of minor (64.9 %) or moderate (30.6 %) at the time of initial surgery. The highest proportion of patients was from the South (31.5 %) and the lowest proportion of patients from the Northeast (17.8 %). A majority of patients were treated in hospitals with academic urology departments (51.3 %). Median hospital volume was 8.0 pyeloplasties per year and differed between hospitals with and without a urology fellowship (13.5 vs. 7.0; p = 0.04). Children undergoing early versus later surgery differed by all sociodemographic factors, disease severity and hospital characteristics examined (Table 1).
Study population characteristics (Table 1) 4566 children were identified who had a primary diagnosis of UPJ obstruction, underwent pyeloplasty during the study period, and had complete data for analysis. 223 (4.7 %) were excluded due to missing data. Median age at time of initial surgery was 10.0 months and did not significantly vary by year (annual median age 7–17 months). The majority of children was white, non-Hispanic (64.6 %), male (71.5 %), had private insurance (66.00 %) and had a
Factors associated with early surgery vs. late surgery (Table 2) On bivariate analysis, we found that sociodemographic factors, disease severity, and hospital characteristics were significantly associated with age at surgery (p \ 0.01). To determine whether racial variations in age at surgery were due to underlying socioeconomic or clinical differences, we assessed racial differences in disease severity
Table 1 Patient and hospital characteristics Category
Total population (n = 4566) N (%)
White, non-Hispanic
Surgery \1 year (n = 2396) N (%)
Surgery C1 year (n = 2170) N (%)
p value*
Patient demographics Race/ethnicity
Gender Insurance
2948 (64.6)
1383 (57.7)
1565 (72.1)
White, Hispanic
678 (14.9)
430 (18.0)
248 (11.4)
Other
440 (9.6)
266 (11.1)
174 (8.0)
Black
395 (8.7)
255 (10.6)
140 (6.5)
Asian
93 (2.0)
55 (2.3)
38 (1.8)
American Indian
12 (0.3)
7 (0.3)
5 (0.2)
Male
3264 (71.5)
1801 (75.2)
1463 (67.4)
Female
1302 (28.5)
595 (24.8)
707 (32.6)
Private
3013 (66.0)
1453 (60.6)
1560 (71.9)
Public
1553 (34.0)
943 (39.4)
610 (28.1)
Minor
2961 (64.9)
1523 (63.6)
1438 (66.3)
Moderate
1399 (30.6)
704 (29.4)
695 (32.0)
206 (4.5)
169 (7.1)
37 (1.7)
South
1436 (31.5)
894 (37.3)
542 (25.0)
North central
1281 (28.1)
616 (25.7)
665 (30.7)
West
1035 (22.7)
573 (23.9)
462 (21.3)
\0.0001
\0.0001 \0.0001
Clinical presentation Disease severity
Major/extreme
\0.0001
Hospital characteristics Census region
Northeast Urology fellowship Hospital volume
814 (17.8)
313 (13.1)
501 (23.1)
Yes
2343 (51.3)
1,1172 (48.9)
1171 (54.0)
No
2223 (48.7)
1224 (51.1)
999 (46.0)
C10/year
2967 (65.0)
1404 (58.6)
1563 (72.0)
1599 (35.0)
992 (41.4)
607 (28.0)
\10/year * Comparisons across age at surgery were by Pearson v
\0.0001
\0.001 \0.0001
2
123
Pediatr Surg Int Table 2 Bivariable and multivariable association of characteristics with early (age \1 year) versus late (age C1 year) surgery (n = 4566) Category
Unadjusted
Adjusted*
OR (95 % CI)
p value
OR (95 % CI)
p value
Patient demographics Race/ethnicity
Non white and Hispanic vs. white, non-Hispanic
1.90 (1.67–2.15)
\0.0001
1.58 (1.38–1.81)
\0.0001
Gender
Male vs. female
1.46 (1.29–1.66)
\0.0001
1.49 (1.31–1.71)
\0.0001
Insurance
Public vs. private
1.66 (1.47–1.88)
\0.0001
1.35 (1.18–1.55)
\0.0001
Major/extreme vs. minor
4.31 (3.00–6.20)
\0.0001
3.79 (2.61–5.48)
\0.0001
Moderate vs. minor
0.96 (0.84–1.09)
Clinical presentation Disease severity
0.94 (0.82–1.07)
Hospital characteristics Census region
North central vs. west
0.75 (0.63–0.88)
Northeast vs. west
0.50 (0.42–0.61)
South vs. west
1.33 (1.13–1.57)
0.97 (0.81–1.15) \0.0001
0.70 (0.57–0.85)
\0.0001
1.36 (1.14–1.62)
Urology fellowship
Yes vs. no
0.82 (0.73–0.92)
0.001
1.07 (0.93–1.22)
0.38
Hospital volume
\10/year vs. C10/year
1.82 (1.61–2.06)
\0.0001
1.49 (1.29–1.72)
\0.0001
* Adjusted for all co-variates found to be significant (p \ 0.05) on bivariate analysis
and insurance status. Major/Extreme severity scores were significantly associated with nonwhite race (OR 1.8; 95 % CI 1.4–2.4). Nonwhite children were also less likely to have private insurance than white, non-Hispanic children (OR 0.28; 95 % CI 0.24–0.31). On multivariable analysis, nonwhite race/Hispanic ethnicity (OR 1.58), male gender (OR 1.49), and public insurance (OR 1.35) remained significant predictors of early surgery. Major/extreme severity scores (OR: 3.79) and low hospital volume (OR 1.49) were also associated with early surgery. Hospital volume remained a predictor of early surgery when assessed as a continuous variable (p \ 0.001). Hospitals in the Southern census region (OR 1.33) were more likely to intervene early while hospitals in the Northeast region were less likely to intervene early (OR 0.70). Academic status was not a significant predictor of early intervention. Associations remained significant when age at surgery was evaluated as a continuous variable (p \ 0.001 for all variables). Risk of hospital readmission (Table 3) 223 patients (4.9 %) were readmitted with a post-operative complication within the 12 months following initial surgery. Of these, 154 (69.1 %) underwent a second operative procedure and 22 (9.9 %) had multiple readmissions. Average length of stay was 3.1 days (median 2 days; IQR 1–3 days). On univariate and multivariate analysis, early intervention (OR 2.29), and major (OR 4.47) or moderate (OR 1.51) disease severity were significantly associated with risk of readmission. Race, gender, insurance type, hospital volume, and academic status were not associated with risk of readmission. Similar to our finding that age
123
was associated with increased disease severity in our initial cohort (p \ 0.001; Table 2), moderate or higher disease severity was also associated with age in children who were readmitted (OR 2.66; 95 % CI 1.16–6.13). Disease severity and age remained significant predictors of readmission when extreme severity patients (n = 52) were excluded from the analysis. Median time to readmission was not significantly different between early and late surgery groups (58 vs. 50 days; p = 0.77).
Discussion The purpose of our study was to examine the role of demographic, clinical and hospital factors associated with age at surgery in children undergoing pyeloplasty for congenital UPJ obstruction. We found that non-clinical patient factors were associated with timing of surgery in children with UPJ obstruction independent of severity of disease. Our study confirms prior studies that have shown a racial difference in age at surgery [4]. Unlike prior studies where the median age at surgery was 3 years of age or older, we found that the median age at surgery in our sample was less than 1 year of age. This may in part be due to our use of both diagnosis codes and procedure codes to identify patients with congenital UPJ obstruction undergoing pyeloplasty rather than use of procedure codes alone [3, 4]. We also found that public insurance is associated with early surgery independent of race. Prior studies have shown that white children are more likely to be privately insured than nonwhite children [10] and that children with public insurance are less likely to receive urologic care than
Pediatr Surg Int Table 3 Bivariable and multivariable association of characteristics with readmission after initial surgery (n = 4566) Category
Unadjusted OR (95 % CI)
Adjusted* p value
OR (95 % CI)
p value
Patient demographics Race/ethnicity
Non white and Hispanic vs. white, non-Hispanic
1.15 (0.87–1.52)
0.32
1.04 (0.77–1.42)
0.79
Gender
Male vs. female
1.06 (0.79–1.44)
0.70
1.01 (0.74–1.38)
0.94
Insurance
Public vs. private
1.09 (0.82–1.44)
0.55
0.92 (0.68–1.25)
0.60
\0.0001
3.77 (2.43–5.83)
\0.0001
\0.0001
1.54 (1.15–2.08) 2.18 (1.61–2.97)
\0.0001
0.02
0.91 (0.57–1.46)
Clinical presentation Disease severity
Major/extreme vs. minor
4.47 (2.92–6.83)
Age at surgery
Moderate vs. minor \1 vs. C1 year
1.51 (1.12–2.03) 2.29 (1.70–3.07)
Hospital characteristics Census region
North central vs. west
1.03 (0.72–1.45)
Northeast vs. west
0.69 (0.44–1.06)
1.12 (0.77–1.62)
South vs. west
0.64 (0.44–0.93)
Urology fellowship
Yes vs. no
0.76 (0.58–1.00)
0.05
0.84 (0.62–1.13)
0.24
Hospital volume
\10/year vs. C10/year
1.15 (0.87–1.52)
0.32
1.04 (0.76–1.42)
0.82
0.02
0.61 (0.41–0.91)
* Adjusted for all co-variates found to be significant (p \ 0.05) on bivariate analysis
children with private insurance [11, 12]. In our study, we found that, although nonwhite children were more likely to have public insurance, this did not explain the difference between white and nonwhite children, suggesting that socioeconomic status alone does not explain racial differences in timing of surgery and raising the possibility that differences in surgical timing may be associated with physician concerns about loss to follow-up in minority groups [13]. The majority of patients in our sample were male [14]. We found that male gender is associated with an increased risk of early surgery. Although males were more likely to undergo early surgery than females, this was not associated with a higher severity score, suggesting that this difference may be due to differences in the timing of diagnosis rather than differences in disease severity. This finding mirrors prior studies showing a higher incidence of males with prenatal hydronephrosis concerning for possible UPJ obstruction [15]. To assess the role of clinical factors on age at intervention, we used the APR-DRG severity of illness score. The APR-DRG pediatric severity scale was developed with the National Association of Children’s Hospitals and Related Institutions to aid in reimbursements. This system incorporates subclass scores for disease specific severity of illness and risk of mortality, adjusting for differences in patient age, secondary diagnoses, and associated procedures. As a result, this system constitutes the most comprehensive system for assessing disease severity in children [16]. This score is calculated based on provider documentation in the health record and is therefore subject to the accuracy of provider coding. For the purpose of this
study, we utilized the disease severity subclass associated with a primary diagnosis of ureteropelvic junction obstruction. The APR-DRG severity of illness score is a measure of the ‘‘associated risk of physiologic decompensation or organ system loss of function’’ [16]. We found that the majority (95.5 %) of children had mild to moderate disease severity, suggesting that the perceived risk of associated loss of kidney function in most cases is low. However, those children with more severe scores were more likely to undergo early surgery, suggesting that higher perceived risk of renal injury may be an impetus for earlier intervention [1]. We also found that nonwhite children were more likely to have a high severity score than white children, suggesting that there may be a difference in severity at initial presentation. While this difference may be due to previously unreported differences in the natural history of the disease, these findings may also be due to differences in referral and treatment patterns, especially given that hospital census region was an independent predictor of age at surgery. We found that Midwestern and Southern practices were more likely to pursue early intervention compared to Western and Northeastern practices. Our study also suggests that early surgery is less likely at higher volume centers. These findings underscore the presence of divergent approaches to congenital hydronephrosis and mirrors prior studies suggesting regional variations in pediatric surgical and urologic intervention [17, 18]. This variation also suggests that the optimal time for surgical intervention has not yet been established and that differences in parental or provider preferences may be driving observed regional differences in treatment.
123
Pediatr Surg Int
While we could not assess the impact of early surgery on renal preservation, we did assess risk of hospital readmission associated with post-operative complication. We found that our overall readmission rate was 4.9 %, which is similar to the re-operative rates found in prior single center case series [19–21]. We found that children with higher disease severity were more likely to also have a post-operative readmission. Unlike prior studies [22, 23], we also found that younger age at pyeloplasty was independently associated with increased risk of hospital readmission, suggesting that complication rates may be higher in younger children. However, due to absence of data regarding surgical approach and use of perioperative stents or drains, it is unclear whether this finding is an independent predictor of complication or is due to differences in surgical technique [21, 22]. There are several limitations to our study. First, our study was limited to tertiary care children’s hospitals. Consequently, our sample does not account for procedures performed at community hospitals and may not be nationally representative. We are, therefore, unable to assess for potential underlying demographic or socioeconomic skewing of the data in our sample compared to a more nationally representative sample. However, given the shift of pediatric pyeloplasties towards academic centers [3], and given that the majority of these centers are included in the PHIS database, we believe that our study provides a reasonable representation of the majority of pediatric urologic practice. Second, due to our reliance on ICD-9 coding and PHIS complication flags, we are limited by the accuracy and completeness of coded data related to inpatient admissions. We are, therefore, unable to assess the role of outpatient radiographic evaluation on timing of surgery. We also did not capture complications that were not identified by a complication flag, may have been addressed in the outpatient setting, or were treated at a hospital other than that where surgery was initially performed, thereby potentially underestimating the risk of surgical complication in our series. Furthermore, we are unable to identify differences in surgical technique that may account for differences in readmission rates. Finally, since we only assessed children undergoing surgery, we cannot assess the potential effect that not seeking or obtaining care might have on the differences we report with respect to timing of surgery. Despite these limitations, our study has several significant findings confirming prior literature, including the presence of regional and demographic variations in timing of surgery. Additionally, our study raises questions regarding the potential association of early age with increased risk of post-operative hospitalization and re-operation. Further studies are needed to more clearly understand the role of non-clinical factors on the incidence and timing of surgery and risk of post-operative complication.
123
Conclusions In children with congenital UPJ obstruction, age at surgery is associated with patient demographic and geographic factors even when controlling for clinical severity. Early surgery is associated with higher risk of readmission, suggesting that demographic and geographic variations in age at surgery may be associated with differences in surgical outcomes. Further studies are needed to better understand the relationship between demographic, clinical, and geographic factors and risk of surgical intervention as well as the impact of timing of surgery on renal outcomes in these patients. Conflict of interest
None.
References 1. Kitchens DM, Herndon CD (2009) Antenatal hydronephrosis. Current Urol Rep 10(2):126–133 2. Ponsky TA, Huang ZJ, Kittle K, Eichelberger MR, Gilbert JC, Brody F, Newman KD (2004) Hospital- and patient-level characteristics and the risk of appendiceal rupture and negative appendectomy in children. JAMA 292(16):1977–1982. doi:10. 1001/jama.292.16.1977 3. Nelson CP (2007) Evidence of variation by race in the timing of surgery for correction of pediatric ureteropelvic junction obstruction. J Urol 178 (4 Pt 1):1463–1468; discussion 1468. doi:10.1016/j.juro.2007.05.167 4. Nelson CP, Park JM, Dunn RL, Wei JT (2005) Contemporary trends in surgical correction of pediatric ureteropelvic junction obstruction: data from the nationwide inpatient sample. J Urol 173(1):232–236. doi:10.1097/01.ju.0000148439.22885.b4 5. Onen A, Jayanthi VR, Koff SA (2002) Long-term followup of prenatally detected severe bilateral newborn hydronephrosis initially managed nonoperatively. J Urol 168(3):1118–1120. doi:10. 1097/01.ju.0000024449.19337.8d 6. Ulman I, Jayanthi VR, Koff SA (2000) The long-term followup of newborns with severe unilateral hydronephrosis initially treated nonoperatively. J Urol 164(3 Pt 2):1101–1105 7. Muldoon JH (1999) Structure and performance of different DRG classification systems for neonatal medicine. Pediatrics 103 (1 Suppl E):302–318 8. Sedman AB, Bahl V, Bunting E, Bandy K, Jones S, Nasr SZ, Schulz K, Campbell DA (2004) Clinical redesign using all patient refined diagnosis related groups. Pediatrics 114(4):965–969. doi:10.1542/peds.2004-0650 9. Berry JG, Hall DE, Kuo DZ, Cohen E, Agrawal R, Feudtner C, Hall M, Kueser J, Kaplan W, Neff J (2011) Hospital utilization and characteristics of patients experiencing recurrent readmissions within children’s hospitals. JAMA 305(7):682–690. doi:10. 1001/jama.2011.122 10. Shone LP, Dick AW, Brach C, Kimminau KS, LaClair BJ, Shenkman EA, Col JF, Schaffer VA, Mulvihill F, Szilagyi PG, Klein JD, VanLandeghem K, Bronstein J (2003) The role of race and ethnicity in the State Children’s Health Insurance Program (SCHIP) in four states: are there baseline disparities, and what do they mean for SCHIP? Pediatrics 112(6 Pt 2):e521 11. Hwang AH, Hwang MM, Xie HW, Hardy BE, Skaggs DL (2005) Access to urologic care for children in California: medicaid versus private insurance. Urology 66(1):170–173. doi:10.1016/j. urology.2005.01.065
Pediatr Surg Int 12. Skinner AC, Mayer ML (2007) Effects of insurance status on children’s access to specialty care: a systematic review of the literature. BMC Health Serv Res 7:194. doi:10.1186/1472-69637-194 13. Langford AT, Resnicow K, Davis RE, Alexander GL, Calvi J, Weise C, Tolsma D (2010) Ethnic Identity predicts loss-to-follow-up in a health promotion trial. Contemp Clin Trials 31(5):414–418. doi:10.1016/j.cct.2010.06.006 14. Longpre M, Nguan A, Macneily AE, Afshar K (2012) Prediction of the outcome of antenatally diagnosed hydronephrosis: a multivariable analysis. J Pediatr Urol 8(2):135–139. doi:10.1016/j. jpurol.2011.05.013 15. Liang CC, Cheng PJ, Lin CJ, Chen HW, Chao AS, Chang SD (2002) Outcome of prenatally diagnosed fetal hydronephrosis. J Reprod Med 47(1):27–32 16. Averill RF, Goldfield N, Hughes JS, Bonazelli J, McCullough EC, Steinbeck BA, Mullin R, Tang AM, Muldoon J, Turner L, Gay J (2003) All Patient Refined Diagnosis Related Groups Version 20.0 Methodology Review. http://www.hcup-us.ahrq. gov/db/nation.nis/APR-DRGsV20MethodologyOverviewandBiblio graphy.pdf. Accessed 17 Dec 2014 17. Somme S, Bronsert M, Morrato E, Ziegler M (2013) Frequency and variety of inpatient pediatric surgical procedures in the United States. Pediatrics 132(6):e1466–e1472. doi:10.1542/peds. 2013-1243 18. Chu DI, Lloyd JC, Balsara ZR, Wiener JS, Ross SS, Routh JC (2014) Variation in use of nephron-sparing surgery among
19.
20.
21.
22.
23.
children with renal tumors. J Pediatr Urol 10(4):724–729. doi:10. 1016/j.jpurol.2013.12.019 Minnillo BJ, Cruz JA, Sayao RH, Passerotti CC, Houck CS, Meier PM, Borer JG, Diamond DA, Retik AB, Nguyen HT (2011) Long-term experience and outcomes of robotic assisted laparoscopic pyeloplasty in children and young adults. J Urol 185(4):1455–1460. doi:10.1016/j.juro.2010.11.056 Braga LH, Lorenzo AJ, Farhat WA, Bagli DJ, Khoury AE, Pippi Salle JL (2008) Outcome analysis and cost comparison between externalized pyeloureteral and standard stents in 470 consecutive open pyeloplasties. The Journal of urology 180 (4 Suppl): 1693–1698; discussion 1698–1699. doi:10.1016/j.juro.2008.05. 084 Bayne AP, Lee KA, Nelson ED, Cisek LJ, Gonzales ET Jr, Roth DR (2011) The impact of surgical approach and urinary diversion on patient outcomes in pediatric pyeloplasty. J Urol 186(4 Suppl):1693–1698. doi:10.1016/j.juro.2011.03.103 Braga LH, Lorenzo AJ, Bagli DJ, Keays M, Farhat WA, Khoury AE, Salle JL (2008) Risk factors for recurrent ureteropelvic junction obstruction after open pyeloplasty in a large pediatric cohort. The Journal of urology 180 (4 Suppl):1684–1687; discussion 1687–1688. doi:10.1016/j.juro.2008.03.086 Sutherland RW, Chung SK, Roth DR, Gonzales ET (1997) Pediatric pyeloplasty: outcome analysis based on patient age and surgical technique. Urology 50(6):963–966. doi:10.1016/s00904295(97)00397-x
123
ORIGINAL ARTICLE
Factors associated with age at pyeloplasty in children with ureteropelvic junction obstruction Vijaya M. Vemulakonda1 • Duncan T. Wilcox1 • Timothy M. Crombleholme2 Michael Bronsert3 • Allison Kempe3
•
Accepted: 24 June 2015 Ó Springer-Verlag Berlin Heidelberg 2015
Abstract Objectives Our objectives were to: (1) delineate factors associated with surgery at \1 year and (2) determine if early intervention was associated with increased risk of readmission. Methods We conducted a retrospective review of children age 0–18 years with a diagnosis of UPJ obstruction who underwent pyeloplasty from 1/1/1999 to 9/1/2009 using the PHIS database. Data collected included patient factors (race, gender, insurance type, APR-DRG severity of illness) and hospital factors (annual case volume, census region, academic status). Outcomes assessed were: age at surgery and hospital readmission within 1 year of surgery. Data were analyzed using logistic regression and Cox PH for multivariate analyses. Results 4499 children met study criteria. Minority race (OR 1.55), male gender (OR 1.49), public insurance (OR 1.37), high severity of illness (OR 3.60), Southern region (OR 1.44), and low hospital volume (OR 1.37) were significant predictors of early surgery (p \ 0.05). Only early surgery (HR 2.42; 95 % CI 1.67–3.49 2.42) was associated with increased risk of readmission. Conclusions In children with UPJ obstruction, age at surgery is associated with patient demographic and hospital
& Vijaya M. Vemulakonda [email protected] 1
Department of Pediatric Urology, Children’s Hospital Colorado, 13123 E. 16th Ave, Campus Box B-463, Aurora, CO, USA
2
Center for Children’s Surgery, Children’s Hospital Colorado, 13123 E. 16th Ave, Aurora, CO, USA
3
Children’s Outcomes Research Program, Children’s Hospital Colorado, 13199 E. Montview Blvd., Aurora, CO, USA
factors. Early surgery is associated with higher rates of readmission, suggesting that variations in age at surgery may be associated with significant differences in outcomes. Keywords Health disparities Ureteropelvic junction obstruction Practice variability Congenital hydronephrosis Pyeloplasty Abbreviations APR-DRG All patient refined diagnosis related groups PHIS Pediatric health information system UPJ Ureteropelvic junction UTI Urinary tract infection
Introduction Ureteropelvic junction (UPJ) obstruction is the most common etiology of congenital hydronephrosis, affecting 1 in 1000 live births [1]. Although traditionally surgery was reserved for symptomatic children, antenatal ultrasound has led to earlier diagnosis and treatment of the asymptomatic patient. The benefits of early diagnosis and treatment, however, may not be equally distributed among children of different racial backgrounds. Unlike appendectomy, where nonwhite children are less likely to undergo timely intervention [2], nonwhite children have been observed to be more likely to undergo early pyeloplasty than their white, non-Hispanic counterparts [3, 4]. The etiology of racial/ethnic variation in timing of surgery is unclear and its effect on outcomes is unknown. However, this finding raises concerns about possible overtreatment of nonwhite and Hispanic children in light of
123
Pediatr Surg Int
studies suggesting that infants with suspected UPJ obstruction may be safely observed [5, 6]. The purpose of this study was to delineate differences in demographic, clinical, and hospital factors between children who undergo early intervention and those who undergo delayed intervention and to determine if differences in age at surgery are associated with differences in risk of hospital readmission. We hypothesized that variations in age at pyeloplasty would be associated with variations in demographic and hospital factors.
Methods Data source Data for this study were obtained from the pediatric health information system (PHIS), an administrative database that contains inpatient, emergency department, ambulatory surgery and observation encounter-level data from over 45 not-for-profit, tertiary care pediatric hospitals in the United States. These hospitals are affiliated with the Children’s Hospital Association (Overland Park, KS). Data quality and reliability are assured through a joint effort between the Children’s Hospital Association and participating hospitals. Portions of the data submission and data quality processes for the PHIS database are managed by Truven Health Analytics (Ann Arbor, MI). For the purposes of external benchmarking, participating hospitals provide discharge/encounter data including demographics, diagnoses, and procedures. Nearly all of these hospitals also submit resource utilization data (e.g., pharmaceuticals, imaging, and laboratory) into PHIS. Data are de-identified at the time of data submission and are subjected to a number of reliability and validity checks before being included in the database. Informed consent was not obtained due to the de-identified nature of this data. Inclusion and exclusion criteria After institutional review board approval was granted, data were obtained for children 0–18 years admitted for pyeloplasty (ICD-9 code 55.87) between January 1, 1999 and September 30, 2009 to hospitals with complete data for the study period (n = 42 hospitals). Patients with a primary ICD-9 diagnosis code for congenital UPJ obstruction (753.20, 753.21 or 753.29) were included in the study population. Patients who had additional diagnoses of other renal or ureteral anomalies, lower urinary tract anomalies, or congenital anomalies associated with hydronephrosis were excluded.
123
Variable definitions Data obtained included patient demographics (race/ethnicity, gender, insurance status), clinical characteristics (primary diagnosis, severity of disease), hospital factors (hospital census region, hospital volume, academic status), and clinical outcomes (age at time of initial pyeloplasty, hospital readmissions associated with surgical complications). Severity of disease was assessed for patients with a primary diagnosis of UPJ obstruction using the all patient refined diagnosis related groups (APR-DRG) severity scoring system. For purposes of this study, the severity of illness subclass score, which classifies the risk of potential organ system loss of function, was used to assess clinical severity of disease [7, 8]. Hospital volume was calculated as the total number of pyeloplasties performed annually and classified as low volume (\10) or high volume (C10). Academic status was defined by the presence of an affiliated pediatric urology fellowship program. Hospital readmissions associated with surgical complication were identified by a surgical complication flag as defined by the PHIS code list and either (1) a primary diagnosis associated with surgical complication, including urinary tract infection (UTI) (ICD-9 590, 595), UPJ obstruction (753.20, 753.21, 753.29), or renal or ureteral calculus (590), or (2) a procedure associated with surgical complication, including ureteral stent placement (59.8), percutaneous nephrostomy placement (55.03), endopyelotomy (55.1), pyeloplasty (55.87), or nephrectomy (55.4, 55.5). Because individual identifiers are not linked between different PHIS hospitals, all readmissions were limited the hospital where initial surgery was performed after discharge for the principal procedure and within the 12 months following initial surgery [9]. Data analysis Primary outcomes analyzed were: age at surgery and risk of hospital readmission within 1 year of initial surgery. Age at surgery was analyzed as both a continuous and a dichotomous variable. In the dichotomous model, patients \1 year old at time of pyeloplasty were classified as having early surgery. All others were classified as having delayed surgery. Bivariate analyses were performed using Pearson’s v2 for dichotomous variables. Logistic regression was performed to further delineate the association of demographic, clinical, and hospital factors with age at surgery and risk of readmission. For multivariable analysis, variables that were significant on bivariate analysis (p \ 0.05) were utilized to create the logistic regression model. All analyses were performed using SASÒ software version 9.3 (Cary, NC).
Pediatr Surg Int
Analyses were initially conducted with all patients meeting study criteria included regardless of missing data. These results were then compared to the results of analyses with patients with missing data excluded. After confirming that exclusion of children with missing data did not significantly impact study results, the final subset of children for analysis was limited to children with complete data for analysis.
Results
severity of illness score of minor (64.9 %) or moderate (30.6 %) at the time of initial surgery. The highest proportion of patients was from the South (31.5 %) and the lowest proportion of patients from the Northeast (17.8 %). A majority of patients were treated in hospitals with academic urology departments (51.3 %). Median hospital volume was 8.0 pyeloplasties per year and differed between hospitals with and without a urology fellowship (13.5 vs. 7.0; p = 0.04). Children undergoing early versus later surgery differed by all sociodemographic factors, disease severity and hospital characteristics examined (Table 1).
Study population characteristics (Table 1) 4566 children were identified who had a primary diagnosis of UPJ obstruction, underwent pyeloplasty during the study period, and had complete data for analysis. 223 (4.7 %) were excluded due to missing data. Median age at time of initial surgery was 10.0 months and did not significantly vary by year (annual median age 7–17 months). The majority of children was white, non-Hispanic (64.6 %), male (71.5 %), had private insurance (66.00 %) and had a
Factors associated with early surgery vs. late surgery (Table 2) On bivariate analysis, we found that sociodemographic factors, disease severity, and hospital characteristics were significantly associated with age at surgery (p \ 0.01). To determine whether racial variations in age at surgery were due to underlying socioeconomic or clinical differences, we assessed racial differences in disease severity
Table 1 Patient and hospital characteristics Category
Total population (n = 4566) N (%)
White, non-Hispanic
Surgery \1 year (n = 2396) N (%)
Surgery C1 year (n = 2170) N (%)
p value*
Patient demographics Race/ethnicity
Gender Insurance
2948 (64.6)
1383 (57.7)
1565 (72.1)
White, Hispanic
678 (14.9)
430 (18.0)
248 (11.4)
Other
440 (9.6)
266 (11.1)
174 (8.0)
Black
395 (8.7)
255 (10.6)
140 (6.5)
Asian
93 (2.0)
55 (2.3)
38 (1.8)
American Indian
12 (0.3)
7 (0.3)
5 (0.2)
Male
3264 (71.5)
1801 (75.2)
1463 (67.4)
Female
1302 (28.5)
595 (24.8)
707 (32.6)
Private
3013 (66.0)
1453 (60.6)
1560 (71.9)
Public
1553 (34.0)
943 (39.4)
610 (28.1)
Minor
2961 (64.9)
1523 (63.6)
1438 (66.3)
Moderate
1399 (30.6)
704 (29.4)
695 (32.0)
206 (4.5)
169 (7.1)
37 (1.7)
South
1436 (31.5)
894 (37.3)
542 (25.0)
North central
1281 (28.1)
616 (25.7)
665 (30.7)
West
1035 (22.7)
573 (23.9)
462 (21.3)
\0.0001
\0.0001 \0.0001
Clinical presentation Disease severity
Major/extreme
\0.0001
Hospital characteristics Census region
Northeast Urology fellowship Hospital volume
814 (17.8)
313 (13.1)
501 (23.1)
Yes
2343 (51.3)
1,1172 (48.9)
1171 (54.0)
No
2223 (48.7)
1224 (51.1)
999 (46.0)
C10/year
2967 (65.0)
1404 (58.6)
1563 (72.0)
1599 (35.0)
992 (41.4)
607 (28.0)
\10/year * Comparisons across age at surgery were by Pearson v
\0.0001
\0.001 \0.0001
2
123
Pediatr Surg Int Table 2 Bivariable and multivariable association of characteristics with early (age \1 year) versus late (age C1 year) surgery (n = 4566) Category
Unadjusted
Adjusted*
OR (95 % CI)
p value
OR (95 % CI)
p value
Patient demographics Race/ethnicity
Non white and Hispanic vs. white, non-Hispanic
1.90 (1.67–2.15)
\0.0001
1.58 (1.38–1.81)
\0.0001
Gender
Male vs. female
1.46 (1.29–1.66)
\0.0001
1.49 (1.31–1.71)
\0.0001
Insurance
Public vs. private
1.66 (1.47–1.88)
\0.0001
1.35 (1.18–1.55)
\0.0001
Major/extreme vs. minor
4.31 (3.00–6.20)
\0.0001
3.79 (2.61–5.48)
\0.0001
Moderate vs. minor
0.96 (0.84–1.09)
Clinical presentation Disease severity
0.94 (0.82–1.07)
Hospital characteristics Census region
North central vs. west
0.75 (0.63–0.88)
Northeast vs. west
0.50 (0.42–0.61)
South vs. west
1.33 (1.13–1.57)
0.97 (0.81–1.15) \0.0001
0.70 (0.57–0.85)
\0.0001
1.36 (1.14–1.62)
Urology fellowship
Yes vs. no
0.82 (0.73–0.92)
0.001
1.07 (0.93–1.22)
0.38
Hospital volume
\10/year vs. C10/year
1.82 (1.61–2.06)
\0.0001
1.49 (1.29–1.72)
\0.0001
* Adjusted for all co-variates found to be significant (p \ 0.05) on bivariate analysis
and insurance status. Major/Extreme severity scores were significantly associated with nonwhite race (OR 1.8; 95 % CI 1.4–2.4). Nonwhite children were also less likely to have private insurance than white, non-Hispanic children (OR 0.28; 95 % CI 0.24–0.31). On multivariable analysis, nonwhite race/Hispanic ethnicity (OR 1.58), male gender (OR 1.49), and public insurance (OR 1.35) remained significant predictors of early surgery. Major/extreme severity scores (OR: 3.79) and low hospital volume (OR 1.49) were also associated with early surgery. Hospital volume remained a predictor of early surgery when assessed as a continuous variable (p \ 0.001). Hospitals in the Southern census region (OR 1.33) were more likely to intervene early while hospitals in the Northeast region were less likely to intervene early (OR 0.70). Academic status was not a significant predictor of early intervention. Associations remained significant when age at surgery was evaluated as a continuous variable (p \ 0.001 for all variables). Risk of hospital readmission (Table 3) 223 patients (4.9 %) were readmitted with a post-operative complication within the 12 months following initial surgery. Of these, 154 (69.1 %) underwent a second operative procedure and 22 (9.9 %) had multiple readmissions. Average length of stay was 3.1 days (median 2 days; IQR 1–3 days). On univariate and multivariate analysis, early intervention (OR 2.29), and major (OR 4.47) or moderate (OR 1.51) disease severity were significantly associated with risk of readmission. Race, gender, insurance type, hospital volume, and academic status were not associated with risk of readmission. Similar to our finding that age
123
was associated with increased disease severity in our initial cohort (p \ 0.001; Table 2), moderate or higher disease severity was also associated with age in children who were readmitted (OR 2.66; 95 % CI 1.16–6.13). Disease severity and age remained significant predictors of readmission when extreme severity patients (n = 52) were excluded from the analysis. Median time to readmission was not significantly different between early and late surgery groups (58 vs. 50 days; p = 0.77).
Discussion The purpose of our study was to examine the role of demographic, clinical and hospital factors associated with age at surgery in children undergoing pyeloplasty for congenital UPJ obstruction. We found that non-clinical patient factors were associated with timing of surgery in children with UPJ obstruction independent of severity of disease. Our study confirms prior studies that have shown a racial difference in age at surgery [4]. Unlike prior studies where the median age at surgery was 3 years of age or older, we found that the median age at surgery in our sample was less than 1 year of age. This may in part be due to our use of both diagnosis codes and procedure codes to identify patients with congenital UPJ obstruction undergoing pyeloplasty rather than use of procedure codes alone [3, 4]. We also found that public insurance is associated with early surgery independent of race. Prior studies have shown that white children are more likely to be privately insured than nonwhite children [10] and that children with public insurance are less likely to receive urologic care than
Pediatr Surg Int Table 3 Bivariable and multivariable association of characteristics with readmission after initial surgery (n = 4566) Category
Unadjusted OR (95 % CI)
Adjusted* p value
OR (95 % CI)
p value
Patient demographics Race/ethnicity
Non white and Hispanic vs. white, non-Hispanic
1.15 (0.87–1.52)
0.32
1.04 (0.77–1.42)
0.79
Gender
Male vs. female
1.06 (0.79–1.44)
0.70
1.01 (0.74–1.38)
0.94
Insurance
Public vs. private
1.09 (0.82–1.44)
0.55
0.92 (0.68–1.25)
0.60
\0.0001
3.77 (2.43–5.83)
\0.0001
\0.0001
1.54 (1.15–2.08) 2.18 (1.61–2.97)
\0.0001
0.02
0.91 (0.57–1.46)
Clinical presentation Disease severity
Major/extreme vs. minor
4.47 (2.92–6.83)
Age at surgery
Moderate vs. minor \1 vs. C1 year
1.51 (1.12–2.03) 2.29 (1.70–3.07)
Hospital characteristics Census region
North central vs. west
1.03 (0.72–1.45)
Northeast vs. west
0.69 (0.44–1.06)
1.12 (0.77–1.62)
South vs. west
0.64 (0.44–0.93)
Urology fellowship
Yes vs. no
0.76 (0.58–1.00)
0.05
0.84 (0.62–1.13)
0.24
Hospital volume
\10/year vs. C10/year
1.15 (0.87–1.52)
0.32
1.04 (0.76–1.42)
0.82
0.02
0.61 (0.41–0.91)
* Adjusted for all co-variates found to be significant (p \ 0.05) on bivariate analysis
children with private insurance [11, 12]. In our study, we found that, although nonwhite children were more likely to have public insurance, this did not explain the difference between white and nonwhite children, suggesting that socioeconomic status alone does not explain racial differences in timing of surgery and raising the possibility that differences in surgical timing may be associated with physician concerns about loss to follow-up in minority groups [13]. The majority of patients in our sample were male [14]. We found that male gender is associated with an increased risk of early surgery. Although males were more likely to undergo early surgery than females, this was not associated with a higher severity score, suggesting that this difference may be due to differences in the timing of diagnosis rather than differences in disease severity. This finding mirrors prior studies showing a higher incidence of males with prenatal hydronephrosis concerning for possible UPJ obstruction [15]. To assess the role of clinical factors on age at intervention, we used the APR-DRG severity of illness score. The APR-DRG pediatric severity scale was developed with the National Association of Children’s Hospitals and Related Institutions to aid in reimbursements. This system incorporates subclass scores for disease specific severity of illness and risk of mortality, adjusting for differences in patient age, secondary diagnoses, and associated procedures. As a result, this system constitutes the most comprehensive system for assessing disease severity in children [16]. This score is calculated based on provider documentation in the health record and is therefore subject to the accuracy of provider coding. For the purpose of this
study, we utilized the disease severity subclass associated with a primary diagnosis of ureteropelvic junction obstruction. The APR-DRG severity of illness score is a measure of the ‘‘associated risk of physiologic decompensation or organ system loss of function’’ [16]. We found that the majority (95.5 %) of children had mild to moderate disease severity, suggesting that the perceived risk of associated loss of kidney function in most cases is low. However, those children with more severe scores were more likely to undergo early surgery, suggesting that higher perceived risk of renal injury may be an impetus for earlier intervention [1]. We also found that nonwhite children were more likely to have a high severity score than white children, suggesting that there may be a difference in severity at initial presentation. While this difference may be due to previously unreported differences in the natural history of the disease, these findings may also be due to differences in referral and treatment patterns, especially given that hospital census region was an independent predictor of age at surgery. We found that Midwestern and Southern practices were more likely to pursue early intervention compared to Western and Northeastern practices. Our study also suggests that early surgery is less likely at higher volume centers. These findings underscore the presence of divergent approaches to congenital hydronephrosis and mirrors prior studies suggesting regional variations in pediatric surgical and urologic intervention [17, 18]. This variation also suggests that the optimal time for surgical intervention has not yet been established and that differences in parental or provider preferences may be driving observed regional differences in treatment.
123
Pediatr Surg Int
While we could not assess the impact of early surgery on renal preservation, we did assess risk of hospital readmission associated with post-operative complication. We found that our overall readmission rate was 4.9 %, which is similar to the re-operative rates found in prior single center case series [19–21]. We found that children with higher disease severity were more likely to also have a post-operative readmission. Unlike prior studies [22, 23], we also found that younger age at pyeloplasty was independently associated with increased risk of hospital readmission, suggesting that complication rates may be higher in younger children. However, due to absence of data regarding surgical approach and use of perioperative stents or drains, it is unclear whether this finding is an independent predictor of complication or is due to differences in surgical technique [21, 22]. There are several limitations to our study. First, our study was limited to tertiary care children’s hospitals. Consequently, our sample does not account for procedures performed at community hospitals and may not be nationally representative. We are, therefore, unable to assess for potential underlying demographic or socioeconomic skewing of the data in our sample compared to a more nationally representative sample. However, given the shift of pediatric pyeloplasties towards academic centers [3], and given that the majority of these centers are included in the PHIS database, we believe that our study provides a reasonable representation of the majority of pediatric urologic practice. Second, due to our reliance on ICD-9 coding and PHIS complication flags, we are limited by the accuracy and completeness of coded data related to inpatient admissions. We are, therefore, unable to assess the role of outpatient radiographic evaluation on timing of surgery. We also did not capture complications that were not identified by a complication flag, may have been addressed in the outpatient setting, or were treated at a hospital other than that where surgery was initially performed, thereby potentially underestimating the risk of surgical complication in our series. Furthermore, we are unable to identify differences in surgical technique that may account for differences in readmission rates. Finally, since we only assessed children undergoing surgery, we cannot assess the potential effect that not seeking or obtaining care might have on the differences we report with respect to timing of surgery. Despite these limitations, our study has several significant findings confirming prior literature, including the presence of regional and demographic variations in timing of surgery. Additionally, our study raises questions regarding the potential association of early age with increased risk of post-operative hospitalization and re-operation. Further studies are needed to more clearly understand the role of non-clinical factors on the incidence and timing of surgery and risk of post-operative complication.
123
Conclusions In children with congenital UPJ obstruction, age at surgery is associated with patient demographic and geographic factors even when controlling for clinical severity. Early surgery is associated with higher risk of readmission, suggesting that demographic and geographic variations in age at surgery may be associated with differences in surgical outcomes. Further studies are needed to better understand the relationship between demographic, clinical, and geographic factors and risk of surgical intervention as well as the impact of timing of surgery on renal outcomes in these patients. Conflict of interest
None.
References 1. Kitchens DM, Herndon CD (2009) Antenatal hydronephrosis. Current Urol Rep 10(2):126–133 2. Ponsky TA, Huang ZJ, Kittle K, Eichelberger MR, Gilbert JC, Brody F, Newman KD (2004) Hospital- and patient-level characteristics and the risk of appendiceal rupture and negative appendectomy in children. JAMA 292(16):1977–1982. doi:10. 1001/jama.292.16.1977 3. Nelson CP (2007) Evidence of variation by race in the timing of surgery for correction of pediatric ureteropelvic junction obstruction. J Urol 178 (4 Pt 1):1463–1468; discussion 1468. doi:10.1016/j.juro.2007.05.167 4. Nelson CP, Park JM, Dunn RL, Wei JT (2005) Contemporary trends in surgical correction of pediatric ureteropelvic junction obstruction: data from the nationwide inpatient sample. J Urol 173(1):232–236. doi:10.1097/01.ju.0000148439.22885.b4 5. Onen A, Jayanthi VR, Koff SA (2002) Long-term followup of prenatally detected severe bilateral newborn hydronephrosis initially managed nonoperatively. J Urol 168(3):1118–1120. doi:10. 1097/01.ju.0000024449.19337.8d 6. Ulman I, Jayanthi VR, Koff SA (2000) The long-term followup of newborns with severe unilateral hydronephrosis initially treated nonoperatively. J Urol 164(3 Pt 2):1101–1105 7. Muldoon JH (1999) Structure and performance of different DRG classification systems for neonatal medicine. Pediatrics 103 (1 Suppl E):302–318 8. Sedman AB, Bahl V, Bunting E, Bandy K, Jones S, Nasr SZ, Schulz K, Campbell DA (2004) Clinical redesign using all patient refined diagnosis related groups. Pediatrics 114(4):965–969. doi:10.1542/peds.2004-0650 9. Berry JG, Hall DE, Kuo DZ, Cohen E, Agrawal R, Feudtner C, Hall M, Kueser J, Kaplan W, Neff J (2011) Hospital utilization and characteristics of patients experiencing recurrent readmissions within children’s hospitals. JAMA 305(7):682–690. doi:10. 1001/jama.2011.122 10. Shone LP, Dick AW, Brach C, Kimminau KS, LaClair BJ, Shenkman EA, Col JF, Schaffer VA, Mulvihill F, Szilagyi PG, Klein JD, VanLandeghem K, Bronstein J (2003) The role of race and ethnicity in the State Children’s Health Insurance Program (SCHIP) in four states: are there baseline disparities, and what do they mean for SCHIP? Pediatrics 112(6 Pt 2):e521 11. Hwang AH, Hwang MM, Xie HW, Hardy BE, Skaggs DL (2005) Access to urologic care for children in California: medicaid versus private insurance. Urology 66(1):170–173. doi:10.1016/j. urology.2005.01.065
Pediatr Surg Int 12. Skinner AC, Mayer ML (2007) Effects of insurance status on children’s access to specialty care: a systematic review of the literature. BMC Health Serv Res 7:194. doi:10.1186/1472-69637-194 13. Langford AT, Resnicow K, Davis RE, Alexander GL, Calvi J, Weise C, Tolsma D (2010) Ethnic Identity predicts loss-to-follow-up in a health promotion trial. Contemp Clin Trials 31(5):414–418. doi:10.1016/j.cct.2010.06.006 14. Longpre M, Nguan A, Macneily AE, Afshar K (2012) Prediction of the outcome of antenatally diagnosed hydronephrosis: a multivariable analysis. J Pediatr Urol 8(2):135–139. doi:10.1016/j. jpurol.2011.05.013 15. Liang CC, Cheng PJ, Lin CJ, Chen HW, Chao AS, Chang SD (2002) Outcome of prenatally diagnosed fetal hydronephrosis. J Reprod Med 47(1):27–32 16. Averill RF, Goldfield N, Hughes JS, Bonazelli J, McCullough EC, Steinbeck BA, Mullin R, Tang AM, Muldoon J, Turner L, Gay J (2003) All Patient Refined Diagnosis Related Groups Version 20.0 Methodology Review. http://www.hcup-us.ahrq. gov/db/nation.nis/APR-DRGsV20MethodologyOverviewandBiblio graphy.pdf. Accessed 17 Dec 2014 17. Somme S, Bronsert M, Morrato E, Ziegler M (2013) Frequency and variety of inpatient pediatric surgical procedures in the United States. Pediatrics 132(6):e1466–e1472. doi:10.1542/peds. 2013-1243 18. Chu DI, Lloyd JC, Balsara ZR, Wiener JS, Ross SS, Routh JC (2014) Variation in use of nephron-sparing surgery among
19.
20.
21.
22.
23.
children with renal tumors. J Pediatr Urol 10(4):724–729. doi:10. 1016/j.jpurol.2013.12.019 Minnillo BJ, Cruz JA, Sayao RH, Passerotti CC, Houck CS, Meier PM, Borer JG, Diamond DA, Retik AB, Nguyen HT (2011) Long-term experience and outcomes of robotic assisted laparoscopic pyeloplasty in children and young adults. J Urol 185(4):1455–1460. doi:10.1016/j.juro.2010.11.056 Braga LH, Lorenzo AJ, Farhat WA, Bagli DJ, Khoury AE, Pippi Salle JL (2008) Outcome analysis and cost comparison between externalized pyeloureteral and standard stents in 470 consecutive open pyeloplasties. The Journal of urology 180 (4 Suppl): 1693–1698; discussion 1698–1699. doi:10.1016/j.juro.2008.05. 084 Bayne AP, Lee KA, Nelson ED, Cisek LJ, Gonzales ET Jr, Roth DR (2011) The impact of surgical approach and urinary diversion on patient outcomes in pediatric pyeloplasty. J Urol 186(4 Suppl):1693–1698. doi:10.1016/j.juro.2011.03.103 Braga LH, Lorenzo AJ, Bagli DJ, Keays M, Farhat WA, Khoury AE, Salle JL (2008) Risk factors for recurrent ureteropelvic junction obstruction after open pyeloplasty in a large pediatric cohort. The Journal of urology 180 (4 Suppl):1684–1687; discussion 1687–1688. doi:10.1016/j.juro.2008.03.086 Sutherland RW, Chung SK, Roth DR, Gonzales ET (1997) Pediatric pyeloplasty: outcome analysis based on patient age and surgical technique. Urology 50(6):963–966. doi:10.1016/s00904295(97)00397-x
123
