Journal of Pediatric Surgery xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Journal of Pediatric Surgery journal homepage: www.elsevier.com/locate/jpedsurg
VACTERL associations in children undergoing surgery for esophageal atresia and anorectal malformations: Implications for pediatric surgeons☆,☆☆ Timothy B. Lautz a,⁎, Ankur Mandelia a, Jayant Radhakrishnan b a b
Division of Pediatric Surgery, Department of Surgery, Ann & Robert H. Lurie Children's Hospital of Chicago and Northwestern University Feinberg School of Medicine, Chicago, IL, USA University of Illinois College of Medicine, Chicago, IL USA
a r t i c l e
i n f o
Article history: Received 6 August 2014 Received in revised form 8 February 2015 Accepted 14 February 2015 Available online xxxx Key words: Trachea-esophageal fistula Esophageal atresia Anorectal malformation VACTERL
a b s t r a c t Purpose: The aim of this study was to compare the frequency and nature of VACTERL associations between children who underwent surgery for esophageal atresia/tracheoesophageal fistula (EA/TEF) and anorectal malformation (ARM). Methods: We identified all children who underwent surgery for EA/TEF and/or ARM at hospitals participating in the Pediatric Health Information System (PHIS) database between 2004 and 2012. PHIS is an administrative database of free-standing children's hospitals managed by the Child Health Corporation of America (Overland Park, KS) that contains patient-level care data from 43 hospitals. The complete records of patients in this cohort were cross-referenced for diagnoses of vertebral, cardiac, renal and limb anomalies. Results: 2689 children underwent repair of esophageal atresia. Mean gestational age was 36.5 ± 3.2 weeks and mean birth weight was 2536.0 ± 758.7 g. Associated VACTERL diagnoses included vertebral anomaly in 686 (25.5%), ARM in 312 (11.6%), congenital heart disease in 1588 (59.1%), renal disease in 587 (21.8%) and limb defect in 192 (7.1%). 899 (33.4%) had 3 or more anomalies and met criteria for a VACTERL diagnosis. 4962 children underwent repair of ARM. Mean gestational age was 37.4 ± 2.7 weeks and mean birth weight was 2895.2 ± 765.1 g. Associated VACTERL diagnoses included vertebral anomaly in 1562 (31.5%), congenital heart disease in 2007 (40.4%), EA/TEF in 348 (7.0%), renal disease in 1723 (34.7%) and limb defect in 359 (7.2%). 1795 (36.2%) had 3 or more anomalies and met criteria for a VACTERL diagnosis. Conclusion: VACTERL associations are relatively common in children with EA/TEF and ARM and occur in specific clusters, with cardiac disease more common in EA/TEF and spinal and renal/urinary anomalies more common in ARM. © 2015 Elsevier Inc. All rights reserved.
Anomalies of the spine or vertebrae (V), anorectal malformations (A), congenital cardiac anomalies (C), esophageal atresia/tracheoesophageal fistula (TE), renal and urinary abnormalities (R), and limb lesions (L) frequently occur together and are referred to as VACTERL anomalies. This association was first described by Quan and Smith in 1972 and 1973 [1,2], while the cardiac component was incorporated by Temtamy and Miller in 1974. In more recent years, a number of publications from single centers and birth defect registries have further delineated the patterns and frequency of these associations [3–12]. However, the diagnostic
☆ Funding source: There are no sources of funding to report for this study. ☆☆ Financial disclosure: The authors have no financial relationships relevant to this article to disclose. ⁎ Corresponding author at: Division of Pediatric Surgery, Ann & Robert H. Lurie Children's Hospital of Chicago, 225 E. Chicago Ave, Box 63, Chicago, IL 60611. Tel.: +1 312 227 4210. E-mail address: [email protected] (T.B. Lautz).
criteria for VACTERL, the list of specific diagnoses included within this spectrum, and the optimal evaluation of children born with one or more apparent components remain debated [13–15]. In the field of pediatric surgery, VACTERL associations are encountered most frequently in patients who require an operation for esophageal atresia/tracheoesophageal fistula (EA/TEF) or anorectal malformation (ARM). Pediatric surgeons play an important role in counseling parents of children with EA/TEF and ARM in the prenatal and early postnatal periods. A prenatal diagnosis of EA/TEF can now be established in approximately 25% of cases [16]. Prenatal diagnosis of ARM, though still rare, is increasingly possible, especially in high risk patients [17]. Comprehensive, modern, multiinstitutional data on the frequency and significance of VACTERL associations in these patients can better inform these conversations and guide the patients' diagnostic evaluations. To this end, this study was designed to compare the frequency of VACTERL anomalies between patients with EA/TEF and ARM using a national database of large children's hospitals and to determine predictors of mortality.
http://dx.doi.org/10.1016/j.jpedsurg.2015.02.049 0022-3468/© 2015 Elsevier Inc. All rights reserved.
Please cite this article as: Lautz TB, et al, VACTERL associations in children undergoing surgery for esophageal atresia and anorectal malformations: Implications for pediatric surgeons, J Pediatr Surg (2015), http://dx.doi.org/10.1016/j.jpedsurg.2015.02.049
2
T.B. Lautz et al. / Journal of Pediatric Surgery xxx (2015) xxx–xxx
1. Methods We performed a retrospective cohort study to identify all patients who underwent surgery for EA/TEF or ARM at hospitals participating in the Pediatric Health Information System (PHIS) database between 2004 and 2012. PHIS is an administrative database of free-standing children's hospitals managed by the Child Health Corporation of America (Overland Park, KS) that contains patient-level care data. The PHIS hospitals are 43 of the largest children's hospitals in America. The participating hospitals are located in noncompeting markets in 27 states and the District of Columbia. Participating hospitals provide patientlevel data including demographics, diagnoses, and procedures, as well as billing data which includes all medication, diagnostic imaging, laboratory, and supply charges. Data in PHIS are deidentified, but contain an encrypted medical record number that allows tracking of individual patients across multiple inpatient and outpatient encounters. In accordance with PHIS policies, analysis was blinded to the identity of the hospitals. The application of the PHIS database to important issues in the field of pediatric surgery has considerable precedent [18,19]. Patients were identified based on International Classification of Diseases 9th Revision (ICD-9) diagnosis codes for the various VACTERL anomalies shown in Table 1. Inclusion in the study was limited to patients who also had an ICD-9 procedure code indicative of surgery for EA/TEF (31.73 or 42.89) or ARM (48.49, 49.79, 46.11, 46.10, 46.03, 46.13). All hospital encounters for these patients were captured, regardless of whether a VACTERL diagnosis was addressed during that encounter. Data were aggregated across all encounters for a given patient to determine the total number of visits (inpatient, emergency department, ambulatory surgery or outpatient) and the range of time (and age) between those visits. Not all types of encounters were captured for all hospitals in every year.
Table 1 ICD-9 codes used to identify VACTERL associations. Spinal and vertebral anomalies Congenital absence of vertebra Hemivertebra Spina bifida Tethered cord Absent rib Congenital vertebral fusion Other Anal atresia Cardiac anomalies ASD VSD Tetralogy of Fallot Anomalies of the great veins Transposition of the great vessels Congenital anomalies of the aorta Truncus arteriosis Hypoplastic left heart Endocardial cushion defect Common ventricle Other Esophageal atresia Renal anomalies Agenesis/dysgenesis Obstructive uropathy Vesicoureteral reflux Dysplastic kidney Posterior urethral valves Other Limb deformities Reduction deformity upper limb Reduction deformity lower limb Polydactyly Syndactyly Other anomalies Trisomy 21 Single umbilical artery Duodenal atresia
Demographic data and birth information (birth weight and gestational age) were captured from the patient's earliest encounter. Birth weight and gestational age were generally only available for patients whose first admission was in the neonatal period. The median age at first encounter was 0 days (interquartile range 1 day) for those with a birthweight reported compared to 151 days (interquartile range 454 days) for those whose birth weight was not recorded. Race was summarized as white, black, Asian, and other/unknown. The complete records of patients in this cohort were searched for ICD-9 diagnosis codes indicative of the other VACTERL anomalies listed in Table 1. Selection of the exact diagnoses meeting criteria for inclusion as a VACTERL association was based on review of prior literature. The “vertebral” category was broadened to include both bony vertebral anomalies and spinal dysraphisms [7,20]. The “anorectal malformations” category was limited to atresia of the rectum and not expanded to include other intestinal atresias as has been suggested by some [7]. For purposes of comparing to other studies, these more proximal gastrointestinal atresias were reported separately. “Cardiac” defects included major structural anomalies of the heart, but not a patent ductus arteriosus or noncardiac aortic anomalies. The “renal” anomalies category was broadened to include all renal and internal urinary anomalies but not external defects such as hypospadias. Finally the “limb anomalies” included both upper and lower extremity reduction deformities, polydactyly and syndactyly. The study was approved with exempt status by the Institutional Review Board of the Ann & Robert H. Lurie Children's Hospital of Chicago (#2013-15531). Statistical analysis was performed using IBM SPSS Statistics v20. Q-Q plots were examined to determine variation from a normal distribution in this large sample. Variables with a clear nonnormal distribution were reported as a median (interquartile range [IQR]). Otherwise, continuous variables were reported as a mean ± standard deviation. A logistic regression was performed to identify factors associated with mortality. For all analyses, statistical significance was set at a twotailed p b 0.01 because of the large sample size. 2. Results
756.13 756.14 741.X 742.59 756.3 756.15 756.10, 756.19 751.2 745.5 745.4 745.2 747.4X 745.1X 747.2X 745.0 746.7 745.6X 745.3 745.8, 745.9 750.3 753.0 753.2 593.7X 753.15 753.6 753.3, 753.4 755.2X 755.3X 755.0X 755.1X
2.1. Esophageal atresia/tracheoesophageal fistula 2689 children underwent repair of EA/TEF. Demographics are shown in Table 2. The majority of patients (n = 1429, 53.1%) were admitted to their respective PHIS hospital on the day of birth, while 708 (26.3%), 113 (4.2%), and 48 (1.8%) were admitted/transferred on the 1st, 2nd, and 3rd days of life, respectively. For the remaining 391 (14.5%) patients, the first admission was at a later age. Mean gestational age was 36.5 ± 3.2 weeks among the 1649 (61.3%) patients in whom it was Table 2 Patient demographics.
Gender Male Female Race Caucasian African-American Asian Other Unknown Ethnicity Hispanic Not Hispanic Unknown Birth weight (mean ± SD) Gestational age (mean ± SD)
758.0 747.5 751.1
Hospital encounters, median (interquartile range)
EA/TEF
ARM
55.4% 44.6%
54.0% 46.0%
67.9% 9.6% 1.3% 16.5% 4.6%
62.5% 12.0% 3.8% 16.5% 5.1%
15.8% 36.0% 48.2% 2536.0 ± 758.7 g (n = 2287) 36.5 ± 3.2 weeks (n = 1649) 4 (IQR 8)
19.4% 28.9% 51.6% 2895.2 ± 765.1 g (n = 3267) 37.4 ± 2.7 weeks (n = 2443) 4 (IQR 7)
Please cite this article as: Lautz TB, et al, VACTERL associations in children undergoing surgery for esophageal atresia and anorectal malformations: Implications for pediatric surgeons, J Pediatr Surg (2015), http://dx.doi.org/10.1016/j.jpedsurg.2015.02.049
T.B. Lautz et al. / Journal of Pediatric Surgery xxx (2015) xxx–xxx
reported. Gestational age was between 24 and 29 weeks in 57 (3.5%), between 30 and 36 weeks in 632 (38.3%), and between 37 and 42 weeks in 960 (58.2%). Associated VACTERL diagnoses included spinal or vertebral anomaly in 683 (25.4%), ARM in 312 (11.6%), congenital heart disease in 1588 (59.1%), renal or internal urinary disease in 587 (21.8%) and limb defect in 171 (6.4%). Details of these associated diagnoses are shown in Table 3. Associated procedures for these VACTERL diagnoses were performed at the respective PHIS hospital in 218 (8.1%) patients with ARM and 372 (13.8%) with cardiac disease. Eight hundred and ninety-two patients (33.2%) had 3 or more anomalies and met criteria for a VACTERL diagnosis. Overall, 760 (28.3%) had isolated EA/TEF while 1037 (38.6%) had two total anomalies, 528 (19.6%) had three, 235 (8.7%) had four, 102 (3.8%) had five, and 27 (1.0%) had all six categories of VACTERL anomalies. Other conditions included trisomy 21 in 89 (3.3%), single umbilical artery in 78 (2.9%), and duodenal atresia in 127 (4.7%). 2.2. Anorectal malformation 4962 children underwent repair of ARM. The majority of patients (n = 1980, 39.9%) were admitted to their respective PHIS hospital on the day of birth, while 937 (18.9%), 231 (4.7%), and 53 (1.1%) were admitted/transferred on the 1st, 2nd, and 3rd days of life, respectively. For the remaining 1761 (35.5%) patients, the first admission was at an older age. Mean gestational age was 37.4 ± 2.7 weeks among the 2443 (49.2%) patients in whom it was reported. Gestational age was between 23 and 29 weeks in 45 (1.8%), between 30 and 36 weeks in 653 (26.7%), and between 37 and 42 weeks in 1745 (71.4%).
Table 3 VACTERL associations in patients undergoing repair of EA/TEF(n = 2689).
Spinal and vertebral anomalies Congenital absence of vertebra Hemivertebra Spina bifida Tethered cord Absent rib Congenital vertebral fusion Other Anorectal malformation Cardiac disease ASD VSD Tetralogy of Fallot Anomalies of the great veins Transposition of the great vessels Congenital anomalies of the aorta Truncus arteriosis Hypoplastic left heart Endocardial cushion defect Common ventricle Other Renal anomalies Agenesis/dysgenesis Obstructive uropathy Vesicoureteral reflux Dysplastic kidney Posterior urethral valves Other Limb deformities Reduction deformity upper limb Reduction deformity lower limb Polydactyly Syndactyly Other anomalies Trisomy 21 Single umbilical artery Duodenal atresia a
Diagnosis
Diagnosis and surgerya
683 (25.4%) 31 (1.2%) 139 (5.2%) 32 (1.2%) 223 (8.3%) 172 (6.4%) 26 (1.0%) 369 (13.7%) 312 (11.6%) 1588 (59.1%) 1252 (46.6%) 569 (21.2%) 164 (6.1%) 124 (4.6%) 86 (3.2%) 151 (5.6%) 24 (0.9%) 29 (1.1%) 66 (2.5%) 21 (0.8%) 6 (0.2%) 587 (21.8%) 129 (4.8%) 254 (9.4%) 212 (7.9%) 65 (2.4%) 22 (0.8%) 191 (7.1%) 171 (6.4%) 107 (4.0%) 15 (0.6%) 43 (1.6%) 22 (0.8%)
n/a
218 (8.1%) 372 (13.8%) 267 (9.9%) 211 (7.8%) 119 (4.4%) 59 (2.2%) 70 (2.6%) 59 (2.2%) 20 (0.7%) 24 (0.9%) 50 (1.9%) 18 (0.7%) 2 (0.1%) n/a
n/a
n/a 89 (3.3%) 78 (2.9%) 127 (4.7%)
Captures surgery at the same institution only.
3
Associated VACTERL diagnoses included spinal or vertebral anomaly in 1558 (31.4%), congenital heart disease in 2007 (40.4%), EA/TEF in 348 (7.0%), renal or internal urinary disease in 1723 (34.7%) and limb defect in 280 (5.6%). Details of these associated diagnoses are shown in Table 4. Associated procedures for these VACTERL diagnoses were performed at the respective PHIS hospital in 218 (4.4%) patients with EA/TEF and 404 (8.1%) with cardiac disease. 1774 (35.8%) had 3 or more anomalies and met criteria for a VACTERL diagnosis. Overall, 1802 (36.3%) had isolated anal atresia while 1386 (27.9%) had two total anomalies, 1000 (20.2%) had three, 593 (12.0%) had four, 154 (3.1%) had five, and 27 (0.5%) had all six categories of VACTERL anomalies. Other conditions included trisomy 21 in 245 (4.9%), single umbilical artery in 100 (2.0%), and duodenal atresia in 176 (3.5%). 2.3. Comparison of EA/TEF and arm cohorts The frequency of VACTERL associations was compared between the EA/TEF and ARM groups. Cardiac disease, with or without an association operation, was more common in the EA/TEF cohort (p b 0.0001 both for diagnosis and for diagnosis with or without an associated surgery). Spinal or vertebral anomalies in general, and the specific diagnosis of tethered cord, spina bifida, and absent vertebra were all more common in the ARM group (p b 0.001). An absent rib was more common in EA/ TEF (p b 0.001). Renal and internal urinary anomalies as a whole, as well as the specific diagnoses of agenesis/dysgenesis, obstructive uropathy, VUR, dysplastic kidney, and posterior urethral valve were all more common in the ARM group (all p b 0.001). Limb anomalies in general were equally distributed between the two cohorts (p = 0.20), Table 4 VACTERL associations in patients undergoing repair of anorectal malformation (n = 4962).
Spinal and vertebral anomalies Congenital absence of vertebra Hemivertebra Spina bifida Tethered cord Absent rib Congenital vertebral fusion Other Cardiac disease ASD VSD Tetralogy of Fallot Anomalies of the great veins Transposition of the great vessels Congenital anomalies of the aorta Truncus arteriosis Hypoplastic left heart Endocardial cushion defect Common ventricle Other Esophageal atresia/tracheoesophageal fistula Renal anomalies Agenesis/dysgenesis Obstructive uropathy Vesicoureteral reflux Dysplastic kidney Posterior urethral valves Other Limb deformities Reduction deformity upper limb Reduction deformity lower limb Polydactyly Syndactyly Other anomalies Trisomy 21 Single umbilical artery Duodenal atresia a
Diagnosis
Diagnosis and surgerya
1558 (31.4%) 216 (4.4%) 196 (4.0%) 211 (4.3%) 897 (18.1%) 192 (3.9%) 48 (1.0%) 752 (15.2%) 2007 (40.4%) 1674 (33.7%) 675 (13.6%) 148 (3.0%) 152 (3.1%) 86 (1.7%) 105 (2.1%) 28 (0.6%) 35 (0.7%) 69 (1.4%) 18 (0.4%) 4 (0.1%) 348 (7.0%) 1723 (34.7%) 466 (9.4%) 756 (15.2%) 794 (16.0%) 229 (4.6%) 107 (2.2%) 496 (10.0%) 280 (5.6%) 115 (2.3%) 50 (1.0%) 107 (2.2%) 58 (1.2%) 245 (4.9%) 100 (2.0%) 176 (3.5%)
n/a
404 (8.1%) 303 (6.1%) 248 (5.0%) 111 (2.2%) 78 (1.6%) 67 (1.4%) 57 (1.1%) 23 (0.5%) 29 (0.6%) 43 (0.9%) 15 (0.3%) n/a 218 (4.4%)
n/a
n/a
n/a
Captures surgery at the same institution only.
Please cite this article as: Lautz TB, et al, VACTERL associations in children undergoing surgery for esophageal atresia and anorectal malformations: Implications for pediatric surgeons, J Pediatr Surg (2015), http://dx.doi.org/10.1016/j.jpedsurg.2015.02.049
4
T.B. Lautz et al. / Journal of Pediatric Surgery xxx (2015) xxx–xxx
Table 5 Predictors of mortality among patients with EA/TEF or ARM. Odds ratio (95% confidence interval) Gender Male Female Race White Black Asian Other/unknown Ethnicity Hispanic Not-Hispanic Unknown Birthweight b2.0 kg N2.0 kg Unknown First admission to PHIS hospital ≤30 days of life N30 days of life TEFa ARMa Vertebral or spinal anomalyb Cardiac disease None Diagnosis without surgery Diagnosis and surgery Renal/urinary anomalyb Limb anomalyb a b
p 0.27
1.0 0.87 (0.68–1.11) b0.001 1.0 2.04 (1.46–2.86) 1.78 (0.92–3.47) 1.24 (0.92–1.66) 1.0 0.80 (0.55–1.16) 1.20 (0.86–1.68)
0.017 b0.001
3.15 (1.82–5.48) 0.89 (0.52–1.52) 1.0 b0.001 4.50 (2.13–9.51) 1.0 1.39 (0.80–2.42) 1.15 (0.66–2.00) 0.72 (0.54–0.95) 1.0 1.97 (1.46–2.65) 5.77 (4.14–8.03) 1.12 (0.85–1.47) 1.53 (1.02–2.30)
0.25 0.64 0.02 b0.001
0.42 0.04
Diagnosis and procedure. Diagnosis only.
although reduction deformity of the upper limb was more common in the EA/TEF group (p b 0.001). Trisomy 21 was more common in the ARM group (p = 0.001). There was no clinical or statistical difference in the frequency of a VACTERL diagnosis between the EA/TEF (33.2%) and ARM (35.8%) groups (p = 0.024). Inpatient mortality during any encounter was 6.1% among patients with EA/TEF and 3.3% among patients with ARM (p b 0.001). Mortality was driven by comorbid conditions, especially congenital heart disease. Taken together, mortality was 1.3% with isolated EA/TEF or ARM, 5.3% with 2 conditions, 6.0% with 3 conditions, 5.8% with 4 conditions, 6.7% with 5 conditions and 7.9% with all 6 conditions. Most notably, mortality was 3.2% for those who did not undergo cardiac surgery compared to 13.5% for those who had a cardiac operation. Predictors of mortality were assessed in a multivariate logistic regression analysis as shown in Table 5. Independent predictors of mortality included black race, birthweight b 2 kg, congenital heart disease (greater for those who required cardiac surgery than in those with a diagnosis but no operation). Patients whose first encounter at the PHIS hospital was in the first month of life also had a higher mortality, likely because those who were seen at a later age were a self-selected population who came for an elective operation later in life. 3. Discussion The associations between EA/TEF, ARM and other VACTERL diagnoses have important implications not only for the management of
children born with these anomalies, but also for prenatal and postnatal counseling of their parents. It is essential for pediatric surgeons, obstetricians and pediatricians to have current data on the frequency and nature of these associations, derived from a diverse and generalizable patient population, to facilitate optimal management and counseling. We analyzed more than 2500 children with EA/TEF and nearly 5000 with ARM managed at large children's hospitals across the USA. Based on weighted estimates from the Kids' Inpatient Database, there were an estimated 1254 children born with EA/TEF and 1937 with ARM in the year 2009, which corresponds to approximately 10,000 cases of EA/TEF and 15,500 cases of ARM over an 8 year period [21]. Therefore, the current cohort captured a significant fracture of children with these conditions during this period, and these data represent some of the most current and comprehensive information on the frequency of VACTERL associations. Modern descriptions of the frequency of VACTERL associations in children with EA/TEF and ARM have largely come from single institutional reports with differing criteria, making comparisons difficult [5,7,22,23]. For children with EA/TEF, the frequency of associated anomalies identified in the current study was similar to those reported by Quan and Smith in their initial description, by Keckler et al in a review of 112 patients, as well as Brown et al who contacted 579 families of children with EA/TEF from two support groups (Table 6) [2,7,24]. Results are also similar to those reported by Pedersen et al in an analysis of a large European birth defects registry [25]. The higher frequency of renal/urinary conditions in the current study is likely related to the fact that other analyses did not all include nonrenal internal urinary anomalies in their categorization. There is still considerable debate over whether functional conditions such as VUR should be included with structural conditions such as renal agenesis. Likewise, the higher rate of cardiac anomalies in this study may relate to the broad definition used, the long duration of follow-up available beyond the neonatal period for many patients in PHIS, and/or the inclusion of patients with a diagnosis of a small ASD or VSD that didn't require intervention. Similarly, for those children with ARM, the frequency of associated VACTERL anomalies is comparable to those reported by Quan and Smith, as well as Cho et al in an analysis of 103 patients at three medical centers in a single city, de Blaauw et al in a registry analysis of 203 ARM patients, and Rittler et al in a study of 887 children (Table 7) [2,5,8,12]. Differences in the rates of renal/urinary and cardiac anomalies again likely relate to variable inclusion criteria. Internal urinary anomalies were included by Rittler but not de Blaauw, while Cho also included external genitourinary anomalies. Iuchtman et al in 1992 demonstrated that mortality is worse in babies with multiple VACTERL anomalies and often related to their cardiac condition and very low birth weight [26]. However, at the time of their study in the late 1980s, the mortality rate for these patients was much higher (24%). Nonetheless, the trends were reproduced in the current study. Mortality was significantly higher in patients with EA/ TEF compared to ARM, but this was driven by the comorbid conditions, especially congenital heart disease requiring surgery, but also low birth weight and black race. It has been proposed that the VACTERL associations actually occur in specific clusters [6,9]. Kallen et al identified upper and lower groups [6]. The upper group includes a cluster of EA/TEF, cardiac defects, upper
Table 6 Frequency of VACTERL associations in children with TEF.
Spinal and vertebral Anorectal Cardiac Renal and internal urinary Limb
Quan and Smith [2] (n = 19)
Brown et al [22] (n = 579)
Keckler et al [7] (n = 112)
Brosens et al [11] (n = 582)
Pedersen et al [25] (n = 1222)
Current study (n = 2689)
23% 10%
17% 12% 20% 16%
24% 14% 32% 17%
21% 14% 27% 16%
16% 29% 16%
25% 12% 59% 22%
7%
10%
16%
12%
13%
6%
Please cite this article as: Lautz TB, et al, VACTERL associations in children undergoing surgery for esophageal atresia and anorectal malformations: Implications for pediatric surgeons, J Pediatr Surg (2015), http://dx.doi.org/10.1016/j.jpedsurg.2015.02.049
T.B. Lautz et al. / Journal of Pediatric Surgery xxx (2015) xxx–xxx
5
Table 7 Frequency of VACTERL associations in children with ARM.
Spinal and vertebral Cardiac EA/TEF Renal and internal urinary Limb
Quan and Smith [2] (n = 19)
Cho et al [5] (n = 103)
Rittler et al [8] (n = 524)
de Blaauw et al [12] (n = 203)
Current study (n = 4962)
27%
9%
20%
31%
27% 11% 49% 4%
16% 20% 21% 11%
18% spinal 18% vertebral 31% 9% 29% 7% upper 11% lower
7% 5%
costovertebral defects and upper preaxial limb reduction defects. In contrast, the lower group includes ARM and kidney malformations, as well as EA/TEF but not cardiac anomalies or upper limb reduction defects. These groupings were generally supported by our data. Cardiac conditions were diagnosed in 59.1% of neonates with EA/TEF compared to 40.4% of those with ARM, and required operation in 13.8% and 8.1%, respectively. While an echocardiogram is advised for all of these patients in the neonatal period, parents can be advised that the likelihood of significant findings is higher in the EA/TEF group. Likewise, as described above, these cardiac comorbidities drive the higher mortality in the upper group. Conversely, renal and internal urinary anomalies occurred in 34.7% with ARM compared to only 21.8% with EA/TEF. Importantly, among the ARM group, vesicoureteral reflux was the most common renal/urinary anomaly, occurring in 16.0% compared to obstructive uropathy in 15.2% and renal agenesis/dysgenesis in 9.4%. Prior reports have demonstrated previously high frequencies of vesicoureteral reflux among children with ARM. For instance, Nah et al reported a rate of 12% in a cohort of 99 neonates with ARM [20]. While renal ultrasound to screen for structural anomalies is widely performed as a screening tool in any neonate with a VACTERL condition, voiding cystourethrogram (VCUG) to assess for reflux is utilized less frequently. Given the frequency of reflux in neonates with ARM, combined with recent data showing that ultrasound is a poor screening test for this condition [27], VCUG must be considered an important screening tool in this population. Recommendations for screening have been described elsewhere. Based on an analysis of renal and urinary anomalies in 331 children born with ARM, Goossens et al recommended performing ultrasonography in all patients but reserving VCUG for patients with dilated upper urinary tracts, lumbosacral and spinal abnormalities or recurrent urinary tract infections [28]. Similarly, in an older study, Boemers et al recommended reserving VCUG for patients with hydronephrosis and those without a perineal fistula [13]. Tethered cord occurred in 18% of patients in this study with ARM. In a prior review of 99 patients with ARM, Nah et al found that tethered cord was the most common spinal anomaly, occurring in 20% of patients [20]. Tethered cord was seen in patients with all types of ARM. Furthermore, Kim et al found that the rate of tethered cord correlated with the level of the ARM [29]. The frequency of tethered cord was 9% with perineal fistula, 13% with vestibular fistula, 25% with rectouretheral fistula, 27% with rectovesicular fistula, and 47% with cloacal anomaly. In that study, the frequency of tethered cord correlated with the sacral ratio, and a cutoff of b 0.6 was an optimal cutoff for predicting the need for MRI. Lumbar ultrasound is also frequently used as a screening tool for tethered cord and other occult spinal dysraphisms. While abnormal findings on lumbar US have good specificity, the sensitivity of this study is poor [24]. Because ultrasound is an imperfect screening tool for tethered cord, MRI screening should be performed liberally in children with ARM, especially those with sacral ratio b 0.6 and high imperforate anus. While benefiting from the large patient cohort drawn from many institutions, the current study does have several important limitations. Because of the large sample size, even small differences between the EA/TEF and ARM cohorts were statistically significant. This was partially addressed by utilizing a stricter definition of statistical significance (p b 0.01). Nonetheless, differences between the cohorts should be
40% 7% 35% 6%
considered based on both clinically significant and statistically significant differences. Furthermore, our analysis was based on ICD-9 diagnosis codes and thus was limited in specificity. For example these codes could not be used to distinguish the type of fistula in neonates with EA/TEF nor the level of the fistula in ARM. It has been demonstrated that the frequency of associations is different in children with low versus high lesions and these differences may be further defined in the future [5,28]. In addition, we could not distinguish children with VACTERL association from those with the same anomalies owing to an underlying genetic disorder such as CHARGE syndrome or Fanconi anemia. This limitation may have resulted in an overestimation of the frequency of the VACTERL association in these patients. The analysis was also limited by the availability of birthweight and gestational ages in the database. These data were generally only available for those patients whose first encounter at a PHIS hospital was in the first few days of life. Finally, while PHIS allows for tracking of the cohort of patients across multiple encounters at the same hospital, it does not capture diagnoses made or operations performed at other hospitals (including other PHIS hospitals). As such, the frequency of associations, especially those not necessarily made during the index hospitalization (such as vesicoureteral reflux) is expected to be underestimated. In conclusion, the data reported herein represent a comprehensive and generalizable description of the frequencies of VACTERL associations in children with EA/TEF and ARM. They are a valuable resource for guiding the assessment of children born with one of these conditions and for performing prenatal and postnatal counseling of parents. At that present time, comprehensive guidelines for the assessment of children with suspected VACTERL association are available which include parameters for initial evaluation in all cases and selective subsequent investigation [14]. With further insight into the patterns of the VACTERL associations, it may be possible to optimize the initial evaluation of these patients based on the presenting diagnoses.
References [1] Quan L, Smith DW. The VATER association: vertebral defects, anal atresia, tracheoesophageal fistula with esophageal atresia, radial dysplasia. The clinical delineation of birth defects of the cardiovascular system; 1972 75–8. [2] Quan L, Smith DW. The VATER association. Vertebral defects, Anal atresia, T-E fistula with esophageal atresia, Radial and Renal dysplasia: a spectrum of associated defects. J Pediatr 1973;82:104–7. [3] Barnes JC, Smith WL. The VATER association. Radiology 1978;126:445–9. [4] Botto LD, Khoury MJ, Mastroiacovo P, et al. The spectrum of congenital anomalies of the VATER association: an international study. Am J Med Genet 1997;71:8–15. [5] Cho S, Moore SP, Fangman T. One hundred three consecutive patients with anorectal malformations and their associated anomalies. Arch Pediatr Adolesc Med 2001;155: 587–91. [6] Kallen K, Mastroiacovo P, Castilla EE, et al. VATER non-random association of congenital malformations: study based on data from four malformation registers. Am J Med Genet 2001;101:26–32. [7] Keckler SJ, St Peter SD, Valusek PA, et al. VACTERL anomalies in patients with esophageal atresia: an updated delineation of the spectrum and review of the literature. Pediatr Surg Int 2007;23:309–13. [8] Rittler M, Paz JE, Castilla EE. VACTERL association, epidemiologic definition and delineation. Am J Med Genet 1996;63:529–36. [9] Solomon BD, Pineda-Alvarez DE, Raam MS, et al. Analysis of component findings in 79 patients diagnosed with VACTERL association. Am J Med Genet A 2010;152A: 2236–44. [10] Temtamy SA, Miller JD. Extending the scope of the VATER association: definition of the VATER syndrome. J Pediatr 1974;85:345–9.
Please cite this article as: Lautz TB, et al, VACTERL associations in children undergoing surgery for esophageal atresia and anorectal malformations: Implications for pediatric surgeons, J Pediatr Surg (2015), http://dx.doi.org/10.1016/j.jpedsurg.2015.02.049
6
T.B. Lautz et al. / Journal of Pediatric Surgery xxx (2015) xxx–xxx
[11] Brosens E, Ploeg M, van Bever Y, et al. Clinical and etiological heterogeneity in patients with tracheo-esophageal malformations and associated anomalies. Eur J Med Genet 2014;57:440–52. [12] de Blaauw I, Wijers CH, Schmiedeke E, et al. First results of a European multi-center registry of patients with anorectal malformations. J Pediatr Surg 2013;48:2530–5. [13] Boemers TM, Beek FJ, Bax NM. Review. Guidelines for the urological screening and initial management of lower urinary tract dysfunction in children with anorectal malformations—the ARGUS protocol. BJU Int 1999;83:662–71. [14] Solomon BD, Baker LA, Bear KA, et al. An approach to the identification of anomalies and etiologies in neonates with identified or suspected VACTERL (vertebral defects, anal atresia, tracheo-esophageal fistula with esophageal atresia, cardiac anomalies, renal anomalies, and limb anomalies) association. J Pediatr 2014;164:451–7. [15] Solomon BD, Bear KA, Kimonis V, et al. Clinical geneticists' views of VACTERL/VATER association. Am J Med Genet A 2012;158A:3087–100. [16] Garabedian C, Sfeir R, Langlois C, et al. Does prenatal diagnosis modify neonatal management and early outcome of children with esophageal atresia? Am J Obstet Gynecol 2014 [epub ahead of print]. [17] Perlman S, Bilik R, Leibovitch L, et al. More than a gut feeling—sonographic prenatal diagnosis of imperforate anus in a high-risk population. Prenat Diagn 2014;34: 1307–11. [18] Goldin AB, Sawin RS, Garrison MM, et al. Aminoglycoside-based triple-antibiotic therapy versus monotherapy for children with ruptured appendicitis. Pediatrics 2007;119:905–11. [19] Fox D, Morrato E, Campagna EJ, et al. Outcomes of laparoscopic versus open fundoplication in children's hospitals: 2005-2008. Pediatrics 2011;127:872–80.
[20] Nah SA, Ong CC, Lakshmi NK, et al. Anomalies associated with anorectal malformations according to the Krickenbeck anatomic classification. J Pediatr Surg 2012;47:2273–8. [21] HCUP Kids' Inpatient Database (KID). Healthcare Cost and Utilization Project (HCUP). Rockville, MD: Agency for Healthcare Research and Quality; 2009. [22] Brown AK, Roddam AW, Spitz L, et al. Oesophageal atresia, related malformations, and medical problems: a family study. Am J Med Genet 1999;85:31–7. [23] Driver CP, Shankar KR, Jones MO, et al. Phenotypic presentation and outcome of esophageal atresia in the era of the Spitz classification. J Pediatr Surg 2001;36:1419–21. [24] Chern JJ, Aksut B, Kirkman JL, et al. The accuracy of abnormal lumbar sonography findings in detecting occult spinal dysraphism: a comparison with magnetic resonance imaging. J Neurosurg Pediatr 2012;10:150–3. [25] Pedersen RN, Calzolari E, Husby S, et al. Oesophageal atresia: prevalence, prenatal diagnosis and associated anomalies in 23 European regions. Arch Dis Child 2012; 97:227–32. [26] Iuchtman M, Brereton R, Spitz L, et al. Morbidity and mortality in 46 patients with the VACTERL association. Isr J Med Sci 1992;28:281–4. [27] Nelson CP, Johnson EK, Logvinenko T, et al. Ultrasound as a screening test for genitourinary anomalies in children with UTI. Pediatrics 2014;133:e394–403. [28] Goossens WJ, de Blaauw I, Wijnen MH, et al. Urological anomalies in anorectal malformations in The Netherlands: effects of screening all patients on long-term outcome. Pediatr Surg Int 2011;27:1091–7. [29] Kim SM, Chang HK, Lee MJ, et al. Spinal dysraphism with anorectal malformation: lumbosacral magnetic resonance imaging evaluation of 120 patients. J Pediatr Surg 2010;45:769–76.
Please cite this article as: Lautz TB, et al, VACTERL associations in children undergoing surgery for esophageal atresia and anorectal malformations: Implications for pediatric surgeons, J Pediatr Surg (2015), http://dx.doi.org/10.1016/j.jpedsurg.2015.02.049
Contents lists available at ScienceDirect
Journal of Pediatric Surgery journal homepage: www.elsevier.com/locate/jpedsurg
VACTERL associations in children undergoing surgery for esophageal atresia and anorectal malformations: Implications for pediatric surgeons☆,☆☆ Timothy B. Lautz a,⁎, Ankur Mandelia a, Jayant Radhakrishnan b a b
Division of Pediatric Surgery, Department of Surgery, Ann & Robert H. Lurie Children's Hospital of Chicago and Northwestern University Feinberg School of Medicine, Chicago, IL, USA University of Illinois College of Medicine, Chicago, IL USA
a r t i c l e
i n f o
Article history: Received 6 August 2014 Received in revised form 8 February 2015 Accepted 14 February 2015 Available online xxxx Key words: Trachea-esophageal fistula Esophageal atresia Anorectal malformation VACTERL
a b s t r a c t Purpose: The aim of this study was to compare the frequency and nature of VACTERL associations between children who underwent surgery for esophageal atresia/tracheoesophageal fistula (EA/TEF) and anorectal malformation (ARM). Methods: We identified all children who underwent surgery for EA/TEF and/or ARM at hospitals participating in the Pediatric Health Information System (PHIS) database between 2004 and 2012. PHIS is an administrative database of free-standing children's hospitals managed by the Child Health Corporation of America (Overland Park, KS) that contains patient-level care data from 43 hospitals. The complete records of patients in this cohort were cross-referenced for diagnoses of vertebral, cardiac, renal and limb anomalies. Results: 2689 children underwent repair of esophageal atresia. Mean gestational age was 36.5 ± 3.2 weeks and mean birth weight was 2536.0 ± 758.7 g. Associated VACTERL diagnoses included vertebral anomaly in 686 (25.5%), ARM in 312 (11.6%), congenital heart disease in 1588 (59.1%), renal disease in 587 (21.8%) and limb defect in 192 (7.1%). 899 (33.4%) had 3 or more anomalies and met criteria for a VACTERL diagnosis. 4962 children underwent repair of ARM. Mean gestational age was 37.4 ± 2.7 weeks and mean birth weight was 2895.2 ± 765.1 g. Associated VACTERL diagnoses included vertebral anomaly in 1562 (31.5%), congenital heart disease in 2007 (40.4%), EA/TEF in 348 (7.0%), renal disease in 1723 (34.7%) and limb defect in 359 (7.2%). 1795 (36.2%) had 3 or more anomalies and met criteria for a VACTERL diagnosis. Conclusion: VACTERL associations are relatively common in children with EA/TEF and ARM and occur in specific clusters, with cardiac disease more common in EA/TEF and spinal and renal/urinary anomalies more common in ARM. © 2015 Elsevier Inc. All rights reserved.
Anomalies of the spine or vertebrae (V), anorectal malformations (A), congenital cardiac anomalies (C), esophageal atresia/tracheoesophageal fistula (TE), renal and urinary abnormalities (R), and limb lesions (L) frequently occur together and are referred to as VACTERL anomalies. This association was first described by Quan and Smith in 1972 and 1973 [1,2], while the cardiac component was incorporated by Temtamy and Miller in 1974. In more recent years, a number of publications from single centers and birth defect registries have further delineated the patterns and frequency of these associations [3–12]. However, the diagnostic
☆ Funding source: There are no sources of funding to report for this study. ☆☆ Financial disclosure: The authors have no financial relationships relevant to this article to disclose. ⁎ Corresponding author at: Division of Pediatric Surgery, Ann & Robert H. Lurie Children's Hospital of Chicago, 225 E. Chicago Ave, Box 63, Chicago, IL 60611. Tel.: +1 312 227 4210. E-mail address: [email protected] (T.B. Lautz).
criteria for VACTERL, the list of specific diagnoses included within this spectrum, and the optimal evaluation of children born with one or more apparent components remain debated [13–15]. In the field of pediatric surgery, VACTERL associations are encountered most frequently in patients who require an operation for esophageal atresia/tracheoesophageal fistula (EA/TEF) or anorectal malformation (ARM). Pediatric surgeons play an important role in counseling parents of children with EA/TEF and ARM in the prenatal and early postnatal periods. A prenatal diagnosis of EA/TEF can now be established in approximately 25% of cases [16]. Prenatal diagnosis of ARM, though still rare, is increasingly possible, especially in high risk patients [17]. Comprehensive, modern, multiinstitutional data on the frequency and significance of VACTERL associations in these patients can better inform these conversations and guide the patients' diagnostic evaluations. To this end, this study was designed to compare the frequency of VACTERL anomalies between patients with EA/TEF and ARM using a national database of large children's hospitals and to determine predictors of mortality.
http://dx.doi.org/10.1016/j.jpedsurg.2015.02.049 0022-3468/© 2015 Elsevier Inc. All rights reserved.
Please cite this article as: Lautz TB, et al, VACTERL associations in children undergoing surgery for esophageal atresia and anorectal malformations: Implications for pediatric surgeons, J Pediatr Surg (2015), http://dx.doi.org/10.1016/j.jpedsurg.2015.02.049
2
T.B. Lautz et al. / Journal of Pediatric Surgery xxx (2015) xxx–xxx
1. Methods We performed a retrospective cohort study to identify all patients who underwent surgery for EA/TEF or ARM at hospitals participating in the Pediatric Health Information System (PHIS) database between 2004 and 2012. PHIS is an administrative database of free-standing children's hospitals managed by the Child Health Corporation of America (Overland Park, KS) that contains patient-level care data. The PHIS hospitals are 43 of the largest children's hospitals in America. The participating hospitals are located in noncompeting markets in 27 states and the District of Columbia. Participating hospitals provide patientlevel data including demographics, diagnoses, and procedures, as well as billing data which includes all medication, diagnostic imaging, laboratory, and supply charges. Data in PHIS are deidentified, but contain an encrypted medical record number that allows tracking of individual patients across multiple inpatient and outpatient encounters. In accordance with PHIS policies, analysis was blinded to the identity of the hospitals. The application of the PHIS database to important issues in the field of pediatric surgery has considerable precedent [18,19]. Patients were identified based on International Classification of Diseases 9th Revision (ICD-9) diagnosis codes for the various VACTERL anomalies shown in Table 1. Inclusion in the study was limited to patients who also had an ICD-9 procedure code indicative of surgery for EA/TEF (31.73 or 42.89) or ARM (48.49, 49.79, 46.11, 46.10, 46.03, 46.13). All hospital encounters for these patients were captured, regardless of whether a VACTERL diagnosis was addressed during that encounter. Data were aggregated across all encounters for a given patient to determine the total number of visits (inpatient, emergency department, ambulatory surgery or outpatient) and the range of time (and age) between those visits. Not all types of encounters were captured for all hospitals in every year.
Table 1 ICD-9 codes used to identify VACTERL associations. Spinal and vertebral anomalies Congenital absence of vertebra Hemivertebra Spina bifida Tethered cord Absent rib Congenital vertebral fusion Other Anal atresia Cardiac anomalies ASD VSD Tetralogy of Fallot Anomalies of the great veins Transposition of the great vessels Congenital anomalies of the aorta Truncus arteriosis Hypoplastic left heart Endocardial cushion defect Common ventricle Other Esophageal atresia Renal anomalies Agenesis/dysgenesis Obstructive uropathy Vesicoureteral reflux Dysplastic kidney Posterior urethral valves Other Limb deformities Reduction deformity upper limb Reduction deformity lower limb Polydactyly Syndactyly Other anomalies Trisomy 21 Single umbilical artery Duodenal atresia
Demographic data and birth information (birth weight and gestational age) were captured from the patient's earliest encounter. Birth weight and gestational age were generally only available for patients whose first admission was in the neonatal period. The median age at first encounter was 0 days (interquartile range 1 day) for those with a birthweight reported compared to 151 days (interquartile range 454 days) for those whose birth weight was not recorded. Race was summarized as white, black, Asian, and other/unknown. The complete records of patients in this cohort were searched for ICD-9 diagnosis codes indicative of the other VACTERL anomalies listed in Table 1. Selection of the exact diagnoses meeting criteria for inclusion as a VACTERL association was based on review of prior literature. The “vertebral” category was broadened to include both bony vertebral anomalies and spinal dysraphisms [7,20]. The “anorectal malformations” category was limited to atresia of the rectum and not expanded to include other intestinal atresias as has been suggested by some [7]. For purposes of comparing to other studies, these more proximal gastrointestinal atresias were reported separately. “Cardiac” defects included major structural anomalies of the heart, but not a patent ductus arteriosus or noncardiac aortic anomalies. The “renal” anomalies category was broadened to include all renal and internal urinary anomalies but not external defects such as hypospadias. Finally the “limb anomalies” included both upper and lower extremity reduction deformities, polydactyly and syndactyly. The study was approved with exempt status by the Institutional Review Board of the Ann & Robert H. Lurie Children's Hospital of Chicago (#2013-15531). Statistical analysis was performed using IBM SPSS Statistics v20. Q-Q plots were examined to determine variation from a normal distribution in this large sample. Variables with a clear nonnormal distribution were reported as a median (interquartile range [IQR]). Otherwise, continuous variables were reported as a mean ± standard deviation. A logistic regression was performed to identify factors associated with mortality. For all analyses, statistical significance was set at a twotailed p b 0.01 because of the large sample size. 2. Results
756.13 756.14 741.X 742.59 756.3 756.15 756.10, 756.19 751.2 745.5 745.4 745.2 747.4X 745.1X 747.2X 745.0 746.7 745.6X 745.3 745.8, 745.9 750.3 753.0 753.2 593.7X 753.15 753.6 753.3, 753.4 755.2X 755.3X 755.0X 755.1X
2.1. Esophageal atresia/tracheoesophageal fistula 2689 children underwent repair of EA/TEF. Demographics are shown in Table 2. The majority of patients (n = 1429, 53.1%) were admitted to their respective PHIS hospital on the day of birth, while 708 (26.3%), 113 (4.2%), and 48 (1.8%) were admitted/transferred on the 1st, 2nd, and 3rd days of life, respectively. For the remaining 391 (14.5%) patients, the first admission was at a later age. Mean gestational age was 36.5 ± 3.2 weeks among the 1649 (61.3%) patients in whom it was Table 2 Patient demographics.
Gender Male Female Race Caucasian African-American Asian Other Unknown Ethnicity Hispanic Not Hispanic Unknown Birth weight (mean ± SD) Gestational age (mean ± SD)
758.0 747.5 751.1
Hospital encounters, median (interquartile range)
EA/TEF
ARM
55.4% 44.6%
54.0% 46.0%
67.9% 9.6% 1.3% 16.5% 4.6%
62.5% 12.0% 3.8% 16.5% 5.1%
15.8% 36.0% 48.2% 2536.0 ± 758.7 g (n = 2287) 36.5 ± 3.2 weeks (n = 1649) 4 (IQR 8)
19.4% 28.9% 51.6% 2895.2 ± 765.1 g (n = 3267) 37.4 ± 2.7 weeks (n = 2443) 4 (IQR 7)
Please cite this article as: Lautz TB, et al, VACTERL associations in children undergoing surgery for esophageal atresia and anorectal malformations: Implications for pediatric surgeons, J Pediatr Surg (2015), http://dx.doi.org/10.1016/j.jpedsurg.2015.02.049
T.B. Lautz et al. / Journal of Pediatric Surgery xxx (2015) xxx–xxx
reported. Gestational age was between 24 and 29 weeks in 57 (3.5%), between 30 and 36 weeks in 632 (38.3%), and between 37 and 42 weeks in 960 (58.2%). Associated VACTERL diagnoses included spinal or vertebral anomaly in 683 (25.4%), ARM in 312 (11.6%), congenital heart disease in 1588 (59.1%), renal or internal urinary disease in 587 (21.8%) and limb defect in 171 (6.4%). Details of these associated diagnoses are shown in Table 3. Associated procedures for these VACTERL diagnoses were performed at the respective PHIS hospital in 218 (8.1%) patients with ARM and 372 (13.8%) with cardiac disease. Eight hundred and ninety-two patients (33.2%) had 3 or more anomalies and met criteria for a VACTERL diagnosis. Overall, 760 (28.3%) had isolated EA/TEF while 1037 (38.6%) had two total anomalies, 528 (19.6%) had three, 235 (8.7%) had four, 102 (3.8%) had five, and 27 (1.0%) had all six categories of VACTERL anomalies. Other conditions included trisomy 21 in 89 (3.3%), single umbilical artery in 78 (2.9%), and duodenal atresia in 127 (4.7%). 2.2. Anorectal malformation 4962 children underwent repair of ARM. The majority of patients (n = 1980, 39.9%) were admitted to their respective PHIS hospital on the day of birth, while 937 (18.9%), 231 (4.7%), and 53 (1.1%) were admitted/transferred on the 1st, 2nd, and 3rd days of life, respectively. For the remaining 1761 (35.5%) patients, the first admission was at an older age. Mean gestational age was 37.4 ± 2.7 weeks among the 2443 (49.2%) patients in whom it was reported. Gestational age was between 23 and 29 weeks in 45 (1.8%), between 30 and 36 weeks in 653 (26.7%), and between 37 and 42 weeks in 1745 (71.4%).
Table 3 VACTERL associations in patients undergoing repair of EA/TEF(n = 2689).
Spinal and vertebral anomalies Congenital absence of vertebra Hemivertebra Spina bifida Tethered cord Absent rib Congenital vertebral fusion Other Anorectal malformation Cardiac disease ASD VSD Tetralogy of Fallot Anomalies of the great veins Transposition of the great vessels Congenital anomalies of the aorta Truncus arteriosis Hypoplastic left heart Endocardial cushion defect Common ventricle Other Renal anomalies Agenesis/dysgenesis Obstructive uropathy Vesicoureteral reflux Dysplastic kidney Posterior urethral valves Other Limb deformities Reduction deformity upper limb Reduction deformity lower limb Polydactyly Syndactyly Other anomalies Trisomy 21 Single umbilical artery Duodenal atresia a
Diagnosis
Diagnosis and surgerya
683 (25.4%) 31 (1.2%) 139 (5.2%) 32 (1.2%) 223 (8.3%) 172 (6.4%) 26 (1.0%) 369 (13.7%) 312 (11.6%) 1588 (59.1%) 1252 (46.6%) 569 (21.2%) 164 (6.1%) 124 (4.6%) 86 (3.2%) 151 (5.6%) 24 (0.9%) 29 (1.1%) 66 (2.5%) 21 (0.8%) 6 (0.2%) 587 (21.8%) 129 (4.8%) 254 (9.4%) 212 (7.9%) 65 (2.4%) 22 (0.8%) 191 (7.1%) 171 (6.4%) 107 (4.0%) 15 (0.6%) 43 (1.6%) 22 (0.8%)
n/a
218 (8.1%) 372 (13.8%) 267 (9.9%) 211 (7.8%) 119 (4.4%) 59 (2.2%) 70 (2.6%) 59 (2.2%) 20 (0.7%) 24 (0.9%) 50 (1.9%) 18 (0.7%) 2 (0.1%) n/a
n/a
n/a 89 (3.3%) 78 (2.9%) 127 (4.7%)
Captures surgery at the same institution only.
3
Associated VACTERL diagnoses included spinal or vertebral anomaly in 1558 (31.4%), congenital heart disease in 2007 (40.4%), EA/TEF in 348 (7.0%), renal or internal urinary disease in 1723 (34.7%) and limb defect in 280 (5.6%). Details of these associated diagnoses are shown in Table 4. Associated procedures for these VACTERL diagnoses were performed at the respective PHIS hospital in 218 (4.4%) patients with EA/TEF and 404 (8.1%) with cardiac disease. 1774 (35.8%) had 3 or more anomalies and met criteria for a VACTERL diagnosis. Overall, 1802 (36.3%) had isolated anal atresia while 1386 (27.9%) had two total anomalies, 1000 (20.2%) had three, 593 (12.0%) had four, 154 (3.1%) had five, and 27 (0.5%) had all six categories of VACTERL anomalies. Other conditions included trisomy 21 in 245 (4.9%), single umbilical artery in 100 (2.0%), and duodenal atresia in 176 (3.5%). 2.3. Comparison of EA/TEF and arm cohorts The frequency of VACTERL associations was compared between the EA/TEF and ARM groups. Cardiac disease, with or without an association operation, was more common in the EA/TEF cohort (p b 0.0001 both for diagnosis and for diagnosis with or without an associated surgery). Spinal or vertebral anomalies in general, and the specific diagnosis of tethered cord, spina bifida, and absent vertebra were all more common in the ARM group (p b 0.001). An absent rib was more common in EA/ TEF (p b 0.001). Renal and internal urinary anomalies as a whole, as well as the specific diagnoses of agenesis/dysgenesis, obstructive uropathy, VUR, dysplastic kidney, and posterior urethral valve were all more common in the ARM group (all p b 0.001). Limb anomalies in general were equally distributed between the two cohorts (p = 0.20), Table 4 VACTERL associations in patients undergoing repair of anorectal malformation (n = 4962).
Spinal and vertebral anomalies Congenital absence of vertebra Hemivertebra Spina bifida Tethered cord Absent rib Congenital vertebral fusion Other Cardiac disease ASD VSD Tetralogy of Fallot Anomalies of the great veins Transposition of the great vessels Congenital anomalies of the aorta Truncus arteriosis Hypoplastic left heart Endocardial cushion defect Common ventricle Other Esophageal atresia/tracheoesophageal fistula Renal anomalies Agenesis/dysgenesis Obstructive uropathy Vesicoureteral reflux Dysplastic kidney Posterior urethral valves Other Limb deformities Reduction deformity upper limb Reduction deformity lower limb Polydactyly Syndactyly Other anomalies Trisomy 21 Single umbilical artery Duodenal atresia a
Diagnosis
Diagnosis and surgerya
1558 (31.4%) 216 (4.4%) 196 (4.0%) 211 (4.3%) 897 (18.1%) 192 (3.9%) 48 (1.0%) 752 (15.2%) 2007 (40.4%) 1674 (33.7%) 675 (13.6%) 148 (3.0%) 152 (3.1%) 86 (1.7%) 105 (2.1%) 28 (0.6%) 35 (0.7%) 69 (1.4%) 18 (0.4%) 4 (0.1%) 348 (7.0%) 1723 (34.7%) 466 (9.4%) 756 (15.2%) 794 (16.0%) 229 (4.6%) 107 (2.2%) 496 (10.0%) 280 (5.6%) 115 (2.3%) 50 (1.0%) 107 (2.2%) 58 (1.2%) 245 (4.9%) 100 (2.0%) 176 (3.5%)
n/a
404 (8.1%) 303 (6.1%) 248 (5.0%) 111 (2.2%) 78 (1.6%) 67 (1.4%) 57 (1.1%) 23 (0.5%) 29 (0.6%) 43 (0.9%) 15 (0.3%) n/a 218 (4.4%)
n/a
n/a
n/a
Captures surgery at the same institution only.
Please cite this article as: Lautz TB, et al, VACTERL associations in children undergoing surgery for esophageal atresia and anorectal malformations: Implications for pediatric surgeons, J Pediatr Surg (2015), http://dx.doi.org/10.1016/j.jpedsurg.2015.02.049
4
T.B. Lautz et al. / Journal of Pediatric Surgery xxx (2015) xxx–xxx
Table 5 Predictors of mortality among patients with EA/TEF or ARM. Odds ratio (95% confidence interval) Gender Male Female Race White Black Asian Other/unknown Ethnicity Hispanic Not-Hispanic Unknown Birthweight b2.0 kg N2.0 kg Unknown First admission to PHIS hospital ≤30 days of life N30 days of life TEFa ARMa Vertebral or spinal anomalyb Cardiac disease None Diagnosis without surgery Diagnosis and surgery Renal/urinary anomalyb Limb anomalyb a b
p 0.27
1.0 0.87 (0.68–1.11) b0.001 1.0 2.04 (1.46–2.86) 1.78 (0.92–3.47) 1.24 (0.92–1.66) 1.0 0.80 (0.55–1.16) 1.20 (0.86–1.68)
0.017 b0.001
3.15 (1.82–5.48) 0.89 (0.52–1.52) 1.0 b0.001 4.50 (2.13–9.51) 1.0 1.39 (0.80–2.42) 1.15 (0.66–2.00) 0.72 (0.54–0.95) 1.0 1.97 (1.46–2.65) 5.77 (4.14–8.03) 1.12 (0.85–1.47) 1.53 (1.02–2.30)
0.25 0.64 0.02 b0.001
0.42 0.04
Diagnosis and procedure. Diagnosis only.
although reduction deformity of the upper limb was more common in the EA/TEF group (p b 0.001). Trisomy 21 was more common in the ARM group (p = 0.001). There was no clinical or statistical difference in the frequency of a VACTERL diagnosis between the EA/TEF (33.2%) and ARM (35.8%) groups (p = 0.024). Inpatient mortality during any encounter was 6.1% among patients with EA/TEF and 3.3% among patients with ARM (p b 0.001). Mortality was driven by comorbid conditions, especially congenital heart disease. Taken together, mortality was 1.3% with isolated EA/TEF or ARM, 5.3% with 2 conditions, 6.0% with 3 conditions, 5.8% with 4 conditions, 6.7% with 5 conditions and 7.9% with all 6 conditions. Most notably, mortality was 3.2% for those who did not undergo cardiac surgery compared to 13.5% for those who had a cardiac operation. Predictors of mortality were assessed in a multivariate logistic regression analysis as shown in Table 5. Independent predictors of mortality included black race, birthweight b 2 kg, congenital heart disease (greater for those who required cardiac surgery than in those with a diagnosis but no operation). Patients whose first encounter at the PHIS hospital was in the first month of life also had a higher mortality, likely because those who were seen at a later age were a self-selected population who came for an elective operation later in life. 3. Discussion The associations between EA/TEF, ARM and other VACTERL diagnoses have important implications not only for the management of
children born with these anomalies, but also for prenatal and postnatal counseling of their parents. It is essential for pediatric surgeons, obstetricians and pediatricians to have current data on the frequency and nature of these associations, derived from a diverse and generalizable patient population, to facilitate optimal management and counseling. We analyzed more than 2500 children with EA/TEF and nearly 5000 with ARM managed at large children's hospitals across the USA. Based on weighted estimates from the Kids' Inpatient Database, there were an estimated 1254 children born with EA/TEF and 1937 with ARM in the year 2009, which corresponds to approximately 10,000 cases of EA/TEF and 15,500 cases of ARM over an 8 year period [21]. Therefore, the current cohort captured a significant fracture of children with these conditions during this period, and these data represent some of the most current and comprehensive information on the frequency of VACTERL associations. Modern descriptions of the frequency of VACTERL associations in children with EA/TEF and ARM have largely come from single institutional reports with differing criteria, making comparisons difficult [5,7,22,23]. For children with EA/TEF, the frequency of associated anomalies identified in the current study was similar to those reported by Quan and Smith in their initial description, by Keckler et al in a review of 112 patients, as well as Brown et al who contacted 579 families of children with EA/TEF from two support groups (Table 6) [2,7,24]. Results are also similar to those reported by Pedersen et al in an analysis of a large European birth defects registry [25]. The higher frequency of renal/urinary conditions in the current study is likely related to the fact that other analyses did not all include nonrenal internal urinary anomalies in their categorization. There is still considerable debate over whether functional conditions such as VUR should be included with structural conditions such as renal agenesis. Likewise, the higher rate of cardiac anomalies in this study may relate to the broad definition used, the long duration of follow-up available beyond the neonatal period for many patients in PHIS, and/or the inclusion of patients with a diagnosis of a small ASD or VSD that didn't require intervention. Similarly, for those children with ARM, the frequency of associated VACTERL anomalies is comparable to those reported by Quan and Smith, as well as Cho et al in an analysis of 103 patients at three medical centers in a single city, de Blaauw et al in a registry analysis of 203 ARM patients, and Rittler et al in a study of 887 children (Table 7) [2,5,8,12]. Differences in the rates of renal/urinary and cardiac anomalies again likely relate to variable inclusion criteria. Internal urinary anomalies were included by Rittler but not de Blaauw, while Cho also included external genitourinary anomalies. Iuchtman et al in 1992 demonstrated that mortality is worse in babies with multiple VACTERL anomalies and often related to their cardiac condition and very low birth weight [26]. However, at the time of their study in the late 1980s, the mortality rate for these patients was much higher (24%). Nonetheless, the trends were reproduced in the current study. Mortality was significantly higher in patients with EA/ TEF compared to ARM, but this was driven by the comorbid conditions, especially congenital heart disease requiring surgery, but also low birth weight and black race. It has been proposed that the VACTERL associations actually occur in specific clusters [6,9]. Kallen et al identified upper and lower groups [6]. The upper group includes a cluster of EA/TEF, cardiac defects, upper
Table 6 Frequency of VACTERL associations in children with TEF.
Spinal and vertebral Anorectal Cardiac Renal and internal urinary Limb
Quan and Smith [2] (n = 19)
Brown et al [22] (n = 579)
Keckler et al [7] (n = 112)
Brosens et al [11] (n = 582)
Pedersen et al [25] (n = 1222)
Current study (n = 2689)
23% 10%
17% 12% 20% 16%
24% 14% 32% 17%
21% 14% 27% 16%
16% 29% 16%
25% 12% 59% 22%
7%
10%
16%
12%
13%
6%
Please cite this article as: Lautz TB, et al, VACTERL associations in children undergoing surgery for esophageal atresia and anorectal malformations: Implications for pediatric surgeons, J Pediatr Surg (2015), http://dx.doi.org/10.1016/j.jpedsurg.2015.02.049
T.B. Lautz et al. / Journal of Pediatric Surgery xxx (2015) xxx–xxx
5
Table 7 Frequency of VACTERL associations in children with ARM.
Spinal and vertebral Cardiac EA/TEF Renal and internal urinary Limb
Quan and Smith [2] (n = 19)
Cho et al [5] (n = 103)
Rittler et al [8] (n = 524)
de Blaauw et al [12] (n = 203)
Current study (n = 4962)
27%
9%
20%
31%
27% 11% 49% 4%
16% 20% 21% 11%
18% spinal 18% vertebral 31% 9% 29% 7% upper 11% lower
7% 5%
costovertebral defects and upper preaxial limb reduction defects. In contrast, the lower group includes ARM and kidney malformations, as well as EA/TEF but not cardiac anomalies or upper limb reduction defects. These groupings were generally supported by our data. Cardiac conditions were diagnosed in 59.1% of neonates with EA/TEF compared to 40.4% of those with ARM, and required operation in 13.8% and 8.1%, respectively. While an echocardiogram is advised for all of these patients in the neonatal period, parents can be advised that the likelihood of significant findings is higher in the EA/TEF group. Likewise, as described above, these cardiac comorbidities drive the higher mortality in the upper group. Conversely, renal and internal urinary anomalies occurred in 34.7% with ARM compared to only 21.8% with EA/TEF. Importantly, among the ARM group, vesicoureteral reflux was the most common renal/urinary anomaly, occurring in 16.0% compared to obstructive uropathy in 15.2% and renal agenesis/dysgenesis in 9.4%. Prior reports have demonstrated previously high frequencies of vesicoureteral reflux among children with ARM. For instance, Nah et al reported a rate of 12% in a cohort of 99 neonates with ARM [20]. While renal ultrasound to screen for structural anomalies is widely performed as a screening tool in any neonate with a VACTERL condition, voiding cystourethrogram (VCUG) to assess for reflux is utilized less frequently. Given the frequency of reflux in neonates with ARM, combined with recent data showing that ultrasound is a poor screening test for this condition [27], VCUG must be considered an important screening tool in this population. Recommendations for screening have been described elsewhere. Based on an analysis of renal and urinary anomalies in 331 children born with ARM, Goossens et al recommended performing ultrasonography in all patients but reserving VCUG for patients with dilated upper urinary tracts, lumbosacral and spinal abnormalities or recurrent urinary tract infections [28]. Similarly, in an older study, Boemers et al recommended reserving VCUG for patients with hydronephrosis and those without a perineal fistula [13]. Tethered cord occurred in 18% of patients in this study with ARM. In a prior review of 99 patients with ARM, Nah et al found that tethered cord was the most common spinal anomaly, occurring in 20% of patients [20]. Tethered cord was seen in patients with all types of ARM. Furthermore, Kim et al found that the rate of tethered cord correlated with the level of the ARM [29]. The frequency of tethered cord was 9% with perineal fistula, 13% with vestibular fistula, 25% with rectouretheral fistula, 27% with rectovesicular fistula, and 47% with cloacal anomaly. In that study, the frequency of tethered cord correlated with the sacral ratio, and a cutoff of b 0.6 was an optimal cutoff for predicting the need for MRI. Lumbar ultrasound is also frequently used as a screening tool for tethered cord and other occult spinal dysraphisms. While abnormal findings on lumbar US have good specificity, the sensitivity of this study is poor [24]. Because ultrasound is an imperfect screening tool for tethered cord, MRI screening should be performed liberally in children with ARM, especially those with sacral ratio b 0.6 and high imperforate anus. While benefiting from the large patient cohort drawn from many institutions, the current study does have several important limitations. Because of the large sample size, even small differences between the EA/TEF and ARM cohorts were statistically significant. This was partially addressed by utilizing a stricter definition of statistical significance (p b 0.01). Nonetheless, differences between the cohorts should be
40% 7% 35% 6%
considered based on both clinically significant and statistically significant differences. Furthermore, our analysis was based on ICD-9 diagnosis codes and thus was limited in specificity. For example these codes could not be used to distinguish the type of fistula in neonates with EA/TEF nor the level of the fistula in ARM. It has been demonstrated that the frequency of associations is different in children with low versus high lesions and these differences may be further defined in the future [5,28]. In addition, we could not distinguish children with VACTERL association from those with the same anomalies owing to an underlying genetic disorder such as CHARGE syndrome or Fanconi anemia. This limitation may have resulted in an overestimation of the frequency of the VACTERL association in these patients. The analysis was also limited by the availability of birthweight and gestational ages in the database. These data were generally only available for those patients whose first encounter at a PHIS hospital was in the first few days of life. Finally, while PHIS allows for tracking of the cohort of patients across multiple encounters at the same hospital, it does not capture diagnoses made or operations performed at other hospitals (including other PHIS hospitals). As such, the frequency of associations, especially those not necessarily made during the index hospitalization (such as vesicoureteral reflux) is expected to be underestimated. In conclusion, the data reported herein represent a comprehensive and generalizable description of the frequencies of VACTERL associations in children with EA/TEF and ARM. They are a valuable resource for guiding the assessment of children born with one of these conditions and for performing prenatal and postnatal counseling of parents. At that present time, comprehensive guidelines for the assessment of children with suspected VACTERL association are available which include parameters for initial evaluation in all cases and selective subsequent investigation [14]. With further insight into the patterns of the VACTERL associations, it may be possible to optimize the initial evaluation of these patients based on the presenting diagnoses.
References [1] Quan L, Smith DW. The VATER association: vertebral defects, anal atresia, tracheoesophageal fistula with esophageal atresia, radial dysplasia. The clinical delineation of birth defects of the cardiovascular system; 1972 75–8. [2] Quan L, Smith DW. The VATER association. Vertebral defects, Anal atresia, T-E fistula with esophageal atresia, Radial and Renal dysplasia: a spectrum of associated defects. J Pediatr 1973;82:104–7. [3] Barnes JC, Smith WL. The VATER association. Radiology 1978;126:445–9. [4] Botto LD, Khoury MJ, Mastroiacovo P, et al. The spectrum of congenital anomalies of the VATER association: an international study. Am J Med Genet 1997;71:8–15. [5] Cho S, Moore SP, Fangman T. One hundred three consecutive patients with anorectal malformations and their associated anomalies. Arch Pediatr Adolesc Med 2001;155: 587–91. [6] Kallen K, Mastroiacovo P, Castilla EE, et al. VATER non-random association of congenital malformations: study based on data from four malformation registers. Am J Med Genet 2001;101:26–32. [7] Keckler SJ, St Peter SD, Valusek PA, et al. VACTERL anomalies in patients with esophageal atresia: an updated delineation of the spectrum and review of the literature. Pediatr Surg Int 2007;23:309–13. [8] Rittler M, Paz JE, Castilla EE. VACTERL association, epidemiologic definition and delineation. Am J Med Genet 1996;63:529–36. [9] Solomon BD, Pineda-Alvarez DE, Raam MS, et al. Analysis of component findings in 79 patients diagnosed with VACTERL association. Am J Med Genet A 2010;152A: 2236–44. [10] Temtamy SA, Miller JD. Extending the scope of the VATER association: definition of the VATER syndrome. J Pediatr 1974;85:345–9.
Please cite this article as: Lautz TB, et al, VACTERL associations in children undergoing surgery for esophageal atresia and anorectal malformations: Implications for pediatric surgeons, J Pediatr Surg (2015), http://dx.doi.org/10.1016/j.jpedsurg.2015.02.049
6
T.B. Lautz et al. / Journal of Pediatric Surgery xxx (2015) xxx–xxx
[11] Brosens E, Ploeg M, van Bever Y, et al. Clinical and etiological heterogeneity in patients with tracheo-esophageal malformations and associated anomalies. Eur J Med Genet 2014;57:440–52. [12] de Blaauw I, Wijers CH, Schmiedeke E, et al. First results of a European multi-center registry of patients with anorectal malformations. J Pediatr Surg 2013;48:2530–5. [13] Boemers TM, Beek FJ, Bax NM. Review. Guidelines for the urological screening and initial management of lower urinary tract dysfunction in children with anorectal malformations—the ARGUS protocol. BJU Int 1999;83:662–71. [14] Solomon BD, Baker LA, Bear KA, et al. An approach to the identification of anomalies and etiologies in neonates with identified or suspected VACTERL (vertebral defects, anal atresia, tracheo-esophageal fistula with esophageal atresia, cardiac anomalies, renal anomalies, and limb anomalies) association. J Pediatr 2014;164:451–7. [15] Solomon BD, Bear KA, Kimonis V, et al. Clinical geneticists' views of VACTERL/VATER association. Am J Med Genet A 2012;158A:3087–100. [16] Garabedian C, Sfeir R, Langlois C, et al. Does prenatal diagnosis modify neonatal management and early outcome of children with esophageal atresia? Am J Obstet Gynecol 2014 [epub ahead of print]. [17] Perlman S, Bilik R, Leibovitch L, et al. More than a gut feeling—sonographic prenatal diagnosis of imperforate anus in a high-risk population. Prenat Diagn 2014;34: 1307–11. [18] Goldin AB, Sawin RS, Garrison MM, et al. Aminoglycoside-based triple-antibiotic therapy versus monotherapy for children with ruptured appendicitis. Pediatrics 2007;119:905–11. [19] Fox D, Morrato E, Campagna EJ, et al. Outcomes of laparoscopic versus open fundoplication in children's hospitals: 2005-2008. Pediatrics 2011;127:872–80.
[20] Nah SA, Ong CC, Lakshmi NK, et al. Anomalies associated with anorectal malformations according to the Krickenbeck anatomic classification. J Pediatr Surg 2012;47:2273–8. [21] HCUP Kids' Inpatient Database (KID). Healthcare Cost and Utilization Project (HCUP). Rockville, MD: Agency for Healthcare Research and Quality; 2009. [22] Brown AK, Roddam AW, Spitz L, et al. Oesophageal atresia, related malformations, and medical problems: a family study. Am J Med Genet 1999;85:31–7. [23] Driver CP, Shankar KR, Jones MO, et al. Phenotypic presentation and outcome of esophageal atresia in the era of the Spitz classification. J Pediatr Surg 2001;36:1419–21. [24] Chern JJ, Aksut B, Kirkman JL, et al. The accuracy of abnormal lumbar sonography findings in detecting occult spinal dysraphism: a comparison with magnetic resonance imaging. J Neurosurg Pediatr 2012;10:150–3. [25] Pedersen RN, Calzolari E, Husby S, et al. Oesophageal atresia: prevalence, prenatal diagnosis and associated anomalies in 23 European regions. Arch Dis Child 2012; 97:227–32. [26] Iuchtman M, Brereton R, Spitz L, et al. Morbidity and mortality in 46 patients with the VACTERL association. Isr J Med Sci 1992;28:281–4. [27] Nelson CP, Johnson EK, Logvinenko T, et al. Ultrasound as a screening test for genitourinary anomalies in children with UTI. Pediatrics 2014;133:e394–403. [28] Goossens WJ, de Blaauw I, Wijnen MH, et al. Urological anomalies in anorectal malformations in The Netherlands: effects of screening all patients on long-term outcome. Pediatr Surg Int 2011;27:1091–7. [29] Kim SM, Chang HK, Lee MJ, et al. Spinal dysraphism with anorectal malformation: lumbosacral magnetic resonance imaging evaluation of 120 patients. J Pediatr Surg 2010;45:769–76.
Please cite this article as: Lautz TB, et al, VACTERL associations in children undergoing surgery for esophageal atresia and anorectal malformations: Implications for pediatric surgeons, J Pediatr Surg (2015), http://dx.doi.org/10.1016/j.jpedsurg.2015.02.049
