- Email: [email protected]
Determinants of quality of life in children with colorectal diseases
Journal of Pediatric Surgery xxx (2016) xxx–xxx
Contents lists available at ScienceDirect
Journal of Pediatric Surgery journal homepage: www.elsevier.com/locate/jpedsurg
Determinants of quality of life in children with colorectal diseases☆ Victoria A. Lane a,b,c, Kristine M. Nacion b, Jennifer N. Cooper b, Marc A. Levitt a,c, Katherine J. Deans b,c, Peter C. Minneci b,c,⁎ a b c
Center for Colorectal and Pelvic Reconstruction, Nationwide Children's Hospital, Columbus, OH, USA Center for Surgical Outcomes Research, The Research Institute at Nationwide Children's Hospital, Columbus, OH, USA Department of Pediatric Surgery, Nationwide Children's Hospital, Columbus, OH, USA
a r t i c l e
i n f o
Article history: Received 1 April 2016 Received in revised form 2 August 2016 Accepted 4 August 2016 Available online xxxx Key words: Anorectal malformation Hirschsprung disease Quality of life
a b s t r a c t Background: Health related Quality of Life (HRQoL) is an important outcome in medical care. The aim of our study was to identify characteristics associated with lower HRQoL scores in children with anorectal malformation (ARM) and Hirschsprung disease (HD). Methods: Patients younger than 18 years, with HD or ARM, who were evaluated at our center from April 2014 to August 2015, were identified. The results of comprehensive questionnaires regarding diagnosis, symptoms, comorbidities and previous medical/surgical history, and validated tools to assess urinary status, stooling status and HRQoL were evaluated. Results: In children aged 0–4 years, vomiting and abdominal distension were found to be associated with a significant reduction in total HRQoL scores. In children N4 years of age, vomiting, nausea, abdominal distension, and abdominal pain were also associated with a significantly lower HRQoL. The strongest predictor of lower HRQoL scores on regression tree analysis, in all age groups, was the presence of a psychological, behavioral or developmental comorbidity. Conclusion: Patients with either HD or an ARM who have a psychological, behavioral or developmental problem experience significantly lower HRQoL than children without such problems, suggesting that provision of behavioral/developmental support as part of the multidisciplinary care of these children may have a substantial impact on their HRQoL. © 2016 Elsevier Inc. All rights reserved.
Health related Quality of Life (HRQoL) has been established as an important patient reported outcome in medical care, especially for patients with chronic diseases where complete recovery may never be achieved [1,2]. Patients with anorectal malformations (ARMs), Hirschsprung Disease (HD) or other complex colorectal diagnoses are a particular focus in the pediatric surgical literature as they are known to encounter diverse problems affecting HRQoL. These problems include fecal incontinence, constipation, problems with toilet training, increased stool frequency, abdominal pain, and gynecological and sexual difficulties, all of which may lead to marked limitations in social functioning and emotional wellbeing in addition to the development of behavioral problems [2,3]. The surgical outcomes of HD and ARM patients have improved with modern surgical techniques, and there is increased understanding of the anatomic and physiologic derangements. Despite
☆ Funding Support: This project was supported by intramural funding from the Research Institute and the Department of Surgery at Nationwide Children's Hospital. The authors report no proprietary or commercial interest in any product mentioned or concept discussed in this article. ⁎ Corresponding author at: Center for Surgical Outcomes Research, The Research Institute at Nationwide Children's Hospital, 700 Children's Drive, J West–4th Floor, Columbus, OH 43205. Tel.: +1 614 722 3066; fax: +1 614 722 6980. E-mail address: [email protected] (P.C. Minneci).
this however, many patients have impaired fecal and urological control, which may negatively impact their health related quality of life [4]. Information on HRQoL in children with ARMs and HD is limited owing to the rarity and heterogeneous nature of these diseases [2,5,6]. Past studies have used a variety of HRQoL assessment tools, including nonvalidated tools, and tools that do not measure at least three domains of HRQoL (e.g. physical, mental, and social functioning) [2]. Some of these issues can now be addressed because of the availability of numerous validated tools to measure HRQoL in children. Currently, the most widely used tool to measure HRQoL in children is the PedsQL™ Generic Core Scales developed in the 1990s by James Varni [7]. In addition to these HRQoL scores, validated tools have been developed to assess bowel and urinary function [8,9]. The aim of our study was to identify characteristics associated with lower HRQoL scores in children with ARM and HD, thereby facilitating an understanding of which interventions might improve HRQoL in these patients. 1. Methods 1.1. Study population The Center for Colorectal and Pelvic Reconstruction at our institution is a quaternary referral center for children with complex colorectal
http://dx.doi.org/10.1016/j.jpedsurg.2016.08.004 0022-3468/© 2016 Elsevier Inc. All rights reserved.
Please cite this article as: Lane VA, et al, Determinants of quality of life in children with colorectal diseases, J Pediatr Surg (2016), http://dx.doi.org/ 10.1016/j.jpedsurg.2016.08.004
2
V.A. Lane et al. / Journal of Pediatric Surgery xxx (2016) xxx–xxx
diagnoses including HD, ARM and functional constipation. The center has a comprehensive prospective clinical registry into which patients are enrolled that interfaces with the hospital electronic health record to allow systematic reporting of patient data during routine patient care interactions. Prior to their initial appointment, patients and their families are asked to complete an extensive questionnaire providing information on their underlying diagnosis, current symptoms, comorbidities and previous medical and surgical history. They also complete a series of validated tools to assess functional urinary and stooling status and HRQoL (parent-reported and self-reported surveys). Only parent-proxy surveys were examined in this study owing to the young age of most patients and concerns about the quality of the self-reported survey data as most surveys were completed online, so it could not be verified that the patient rather than the parent truly completed the self-report survey. All patients with the diagnosis of HD and ARM who were younger than 18 years and seen in the outpatient clinic between April 2014 and August 2015 were included. All patients were consented at the time of enrollment into the Patient Registry using an opt-out consent process. This study was approved by the Institutional Review Board at our institution. 1.2. Quality of life measures Parent-reported HRQoL scores were assessed using the PedsQL™ Pediatric Quality of Life inventory [7,10]. The PedsQL is a modular instrument designed to measure HRQoL in healthy and acutely or chronically ill children and adolescents, aged 0–18 years. These PedsQL Generic Score Scales are age specific and include parent reported versions for all age groups: PedsQL Infant Scales (Infant 1–12 months, Infants 13–24 months) and PedsQL Generic Score Scales (Toddlers 2–4 years, Young children 5–7 years, Children 8–12 years and Teens 13–17 years). Therefore for children aged 0–2 years, the PedsQL Infant Scales were used which measure 5 domains: physical, emotional, social, cognitive functioning, and physical symptoms. For children aged 2–17 years, the PedsQL 4.0 Generic Core Scales-Child and Parent Report module was used which measures four domains: physical, emotional, social and school functioning. For the purpose of this analysis, total parent-reported HRQoL scores as well as physical (physical function) and psychosocial (sum of emotional, social and school functioning) HRQoL scores were examined.
1.4. Other patient characteristics evaluated Patient characteristics evaluated in this study included age, gender, adoption status and the diagnosis of HD or ARM. In addition, parents were asked whether their child currently had specific gastrointestinal symptoms including nausea, vomiting, abdominal pain and distension, constipation and fecal impaction, stool and urinary accidents, early satiety and urinary tract infection. The presence of a variety of comorbidities was also evaluated and these are categorized in Table 1. 1.5. Statistical analysis In order to account for the fact that many normal children younger than 4 years are not yet toilet trained, all analyses were stratified by age. Separate analyses were performed in patients aged 0–4 years (infant and toddler) and patients 5–17 years (young child, child and teen). Parent-reported HRQoL scores were described using means and standard deviations (SDs) for the overall patient population and across groups defined by the various patient demographic and clinical characteristics evaluated. In order to evaluate whether HRQoL differed across groups defined by patient demographic and clinical characteristics, HRQoL scores were compared across patient characteristics using ttests or analysis of variance (ANOVA). In order to identify the most important predictors of QOL among all patient characteristics evaluated, regression tree analysis was performed in lieu of traditional multivariable regression modeling, owing to expected collinearity and interactions among the evaluated variables. A regression tree involves recursive binary partitioning such that, at each split, all predictors and all possible cut point values for each of the predictors are considered, then the predictor and cut point that minimize the residual sum of squares (i.e. explain the largest amount of remaining variability) of the resulting tree are chosen. This is done repeatedly until each “leaf” in the tree reaches a minimum number of observations. Finally, 10fold cross validation is used to reduce the size of the tree in order to achieve an optimal model without overfitting the data [11]. Since this was an exploratory study, no adjustments were made for multiple comparisons and a p-value of 0.05 was considered to be statistically significant. SAS version 9.3 (SAS Institute Inc., Cary, NC) and R version 2.15.1 (R Foundations for Statistical Computing, Vienna Austria) were used for analyses. 2. Results
1.3. Bowel and urinary function scales 2.1. Cohort description The Vancouver dysfunctional elimination syndrome/non-neurogenic lower urinary tract dysfunction (NLUTD/DES) questionnaire was administered to assess urinary functional outcomes, including urinary accidents, nocturnal enuresis, dysuria and hesitancy [8]. The Vancouver NLUTD/DES questionnaire is a 14 item survey designed to be selfadministered by children 9 years and older, and self-administered by parents of children less than 9 years of age. The questionnaire uses a 5-point Likert scale. Scores range from 0 to 52 with lower scores indicating better urinary function. The developers of the questionnaire suggest a total score of 11 as the optimal cutoff score to diagnose NLUTD/DES [8]. The Baylor Continence Scale (BCS) was used to assess bowel function because it is a validated tool that is able to differentiate between normal children and those with various underlying conditions, including anorectal malformations and idiopathic constipation/enuresis [9]. This scale is a self-administered 23-item parent reported survey which measures social continence in children with ARM after surgical correction. It uses a 4-point Likert response scale. Scores range from 2 to 92 with lower scores indicating better social continence. In the report describing the development and validation of this survey, normal children without anatomic or functional continence problems were found to have an average BCS score of 11.5 whereas those with an anorectal malformation had an average BCS of 28.
A total of 549 patients younger than 18 years were enrolled in the patient registry during the study period, 353 of which had a known diagnosis of HD or ARM. Of these, 325 (92.1%) had parentreported HRQoL data and were thus included in this study. The overall mean (SD) of parent reported total HRQoL in this cohort was 74.2 (17.5). 244/325 (75.1%) were in the 0–7 year age range. 2.2. Patients aged 0–4 years Average total HRQoL scores across groups based on key demographic and clinical characteristics in the 185 patients between age 0 and 4 years are shown in Table 2. Only demographic information, and symptoms and comorbidities for which significant differences emerged in total HRQoL are shown in Table 1. The overall mean (SD) of parent reported total HRQoL in this entire cohort was 80.2 (14.4). Total HRQoL scores differed significantly by age with children aged 2–4 having higher parent-reported QOL than children aged 0–24 months, but no differences were seen in total HRQoL scores by gender, adoption status or HD vs. ARM diagnosis. Significant differences emerged in the total HRQoL score in children with vomiting and abdominal distension; however, no significant
Please cite this article as: Lane VA, et al, Determinants of quality of life in children with colorectal diseases, J Pediatr Surg (2016), http://dx.doi.org/ 10.1016/j.jpedsurg.2016.08.004
V.A. Lane et al. / Journal of Pediatric Surgery xxx (2016) xxx–xxx
3
Table 1 Comorbidities at intake. Head/Face Respiratory Cardiac
Gastrointestinal Urologic
• Cleft lip and/or palate, Micrognathia, Ear abnormality, Choanal atresia, Deviated septum, Craniosynostosis • Laryngomalacia, Tracheomalacia, Bronchomalacia, Wheezing, Asthma, Sleep apnea • Patent Ductus Arteriosus, Patent foramen ovale, Ventricular septal defect, Aortic stenosis, Aortic double arch, Coarctation of aorta, Atrioventricular can defect, Hetrotaxy, Double outlet right ventricle, Double outlet left ventricle, Tetralogy of Fallot, Transposition of great vessels, Total anomalous pulmonary venous drainage, Other valve abnormality, Pathological heart murmur • Esophageal atresia with/without trachea-esophageal fistula, Duodenal atresia, Duodenal stenosis, Duodenal web, Jejunal atresia, Ileal atresia, Colonic atresia, Malrotation, Gastroschisis, Omphalocele, Meckel's diverticulum • Kidney Renal agenesis, Crossed fused ectopic kidney, Horseshoe kidney, Kidney malrotation, Dysplastic kidney, Cystic kidney, Renal stones, Chronic kidney disease, Hydronephrosis • Bladder Bladder stones, Bladder duplication, Bladder exstrophy, Bladder neck outflow obstruction, Bladder neck incompetence, Urachal cyst, Bladder diverticulum, Neurogenic bladder • Ureter Vesicoureteric reflux, Ectopic ureter, Duplicated ureter, Mega-ureter, Ureterocele, Uretero-pelvic junction (UPJ) obstruction
• Urethra Posterior urethral valves, Urogenital sinus, Urethral diverticulum, Urethral stricture, Duplicated urethra, Urethral stenosis, Urethral atresia Genitals/Reproductive Disorder of sexual development (DSD), Hydrocolpos, Hematocolpos, Absent uterus, Uterine didelphys, Bicornate uterus, Unicornate uterus, Vaginal atresia, Vaginal septum, Cervical atresia, Clitoromegaly, Penoscrotal transposition, Bifid scrotum, Undescended testicle(s), Absent testis (es), Hypospadias, Micropenis, Mullerian, vaginal or other gynecologic anomaly Skeletal/Vertebral Oligodactyly, Polydactyly, Syndactyly, Absent radius, Talipes, Absent radius, Congenital dislocation of hips, Kyphosis, Scoliosis, Rib anomalies, Cervical/thoracic/lumbar hemivertebrae, Sacral dysplasia, Absent coccyx, Pre-sacral mass/tumor, Meningocele, Myelomeningocele, Tethered cord, Occult spinal dysraphism Brain/Neurologic Hydrocephalus, Hydrocephaus with shunt, Seizures, Abnormal hearing, Deafness, Abnormal vision, Blindness, Brain injury Endocrine/Metabolic Diabetes Type 1, Diabetes Type 2, Thyroid disease, Electrolyte abnormality, Malignant hyperthermia Chromosomal/Genetic Trisomy 21, Trisomy 18, Mitochondrial disease, Currarino syndrome, Pallister Hall, Mowat Wilson, Waardenburg Shah (I–IV), Cat-Eye Syndrome, Inherited metabolic disorder Hematologic/Lymphatic/Oncologic Anemia, Bleeding/clotting disorder, Thalassemia-alpha and beta, Sickle cell disease, Malignancy/cancer Psychological/behavioral/developmental Developmental delay, Attention Deficit Disorder (ADD), Attention Deficit hyperactivity Disorder (ADHD), Obsessive Compulsive Disorder (OCD), Autism Spectrum Exposure/Infectious disease Tuberculosis, Methicillin Resistant Staphylococcus aureus (MRSA), Resistant Gram-Negative rods (RGNR), Vancomycin-Resistant Enterococcus, Pertussis
differences emerged in those reported to have gastrointestinal and urinary tract comorbidities, abdominal pain, constipation, stool and urinary accidents, early satiety and fecal impaction (data not shown). No difference in total HRQoL was seen between HD and ARM patients; however, significantly lower total HRQoL was reported in patients with any comorbidity (Table 2, and more specifically in patients with comorbidities involving diagnoses relating to the head/ face such as cleft lip/palate and micrognathia, and diagnoses related to the airway/trachea/esophagus and respiratory tract such as asthma and tracheomalacia. Other comorbidities including cardiac diagnoses, gastrointestinal, renal, skeletal, brain/neurologic and endocrine were not found to be associated with significant changes in total HRQoL scores. Patients reported to have a chromosomal anomaly or a psychological, behavioral, or developmental comorbidity also had significantly lower total HRQoL than patients without (p b 0.001 for both). When analyzing whether patients with and without a history of previous surgical procedures had different HRQoL, total HRQoL did not differ according to either the number of previous surgical procedures (including urological procedures or previous history of an ostomy), or the presence of an ostomy (p N 0.10 for all, data not shown). In regression tree analysis, the factor most useful for distinguishing young children with ARM or HD with higher versus lower HRQoL scores was whether they had a psychological, behavioral, or developmental comorbidity (Mean (SD) 66.4 (14.1) if present vs. 81.8 (13.5) if not present). The addition of other patient characteristics did not significantly improve model fit. When patients with HD and ARM were evaluated separately, in both groups, those with a psychological, behavioral, or developmental comorbidity had significantly lower HRQoL than those without this comorbidity (65.8 (13.0) vs. 82.7 (11.6) in ARM patients, 66.9 (15.8) vs. 80.1 (16.5) in HD patients, p b 0.05 for both).
2.3. Patients aged 5–17 years: Total HRQoL score Average total HRQoL scores across groups based on key demographic and clinical characteristics in the 140 patients aged 5–17 years are shown in Table 3. The overall mean (SD) of the parent reported total HRQOL score in this age group was 72.5 (18.8). Among patients aged 5–17 years, HRQoL scores did not differ by age, gender or adoption status. The overall HRQoL score was not significantly different between the ARM group and the HD group. When considering the symptoms associated with a reduction in the total HRQoL score vomiting, nausea, abdominal distension, and abdominal pain were found to be statistically significant. Total HRQoL did not differ between patients with and without stooling accidents and urinary accidents, and thus these are not reported in Table 3. Psychological/behavioral/developmental diagnoses, chromosomal anomalies, cardiac diagnoses, and brain/neurological diagnoses led to lower reported total HRQoL score (Table 3). No significant difference in total HRQoL score was seen in children with or without head/face diagnoses, respiratory, gastrointestinal, urologic, skeletal, endocrine, hematologic diagnoses or infectious diseases. In this age group, past surgical history once again was not associated with lower total HRQoL scores nor was having a current colostomy or ileostomy (data not shown). Regression tree analysis for total HRQoL (Fig. 1) revealed that the factor most useful for predicting total HRQoL in school age children with HD or ARM was whether they had a psychological, behavioral, or developmental comorbidity. In addition, in those patients who had a psychological, behavioral, or developmental comorbidity, the presence of a chromosomal/genetic comorbidity was an important predictor of total HRQoL score.
Please cite this article as: Lane VA, et al, Determinants of quality of life in children with colorectal diseases, J Pediatr Surg (2016), http://dx.doi.org/ 10.1016/j.jpedsurg.2016.08.004
4
V.A. Lane et al. / Journal of Pediatric Surgery xxx (2016) xxx–xxx
Table 2 Parent reported total HRQoL scores across various patient characteristics in children 0–4 years of age. Patient characteristic/Symptoms/Comorbidity
Patient Characteristic Age group b1 year 1 to b2 years 2–4 years Gender Male Female Adoption status Not adopted Adopted Diagnosis ARM (includes all cloaca variants and imperforate anus) Hirschsprung disease Current Symptoms Vomiting Yes No Abdominal distension Yes No Comorbidities Any comorbidity Yes No Head/Face Yes No Airway/Trachea/Esophagus/Respiratory Yes No Chromosomal/Genetic Yes No Psychological/Behavioral/Developmental Yes No
N
HRQOL score Mean (SD)
-
P value
0.02 57 32 96
77.9 (15.1) 76.1 (15.0) 82.9 (13.2)
114 71
80.0 (14.0) 80.5 (15.0)
143 41
79.5 (14.5) 82.5 (13.8)
119
81.3 (12.6)
66
78.1 (17.0)
12 173
69.0 (17.3) 81.0 (13.8)
50 135
74.4 (16.2) 82.3 (13.0)
137 48
78.9 (14.3) 83.8 (13.9)
8 174
67.6 (17.0) 80.9 (14.0)
17 163
71.5 (13.4) 81.2 (14.3)
19 158
69.2 (16.3) 82.0 (13.2)
20 162
66.4 (14.1) 82.0 (13.4)
behavioral, or developmental comorbidity were also seen to have significantly lower psychosocial HRQoL scores. The only variable that significantly discriminated between patients with lower and higher psychosocial HRQoL scores in regression tree analysis was the presence of a psychological, behavioral, or developmental comorbidity. When other variables were added, none significantly improved the fit of the model. 2.6. Patients aged 5–17 years: fecal and urinary continence
0.81
0.23
0.14
0.005
b.001
0.04
0.01
Fecal and urinary functional statuses were assessed in this age group using the Baylor Continence Scale and the Vancouver dysfunctional elimination syndrome/non-neurogenic lower urinary tract dysfunction (NLUTD/DES) questionnaire respectively. Among all school age children in our study, the median Vancouver score was 14 (IQR 8–19) and the median Baylor score was 23 (IQR 17–33). Low scores for both the Baylor and the Vancouver questionnaires are indicative of better urinary and fecal control. A Vancouver score greater than or equal to 11 was seen to be significantly associated with lower psychosocial HRQoL score. Baylor Continence scores above the median were not associated with a significant decrease in the total HRQoL score overall, or in either particular domain of QoL, though the differences were in the expected direction (Table 4). Neither Vancouver nor Baylor scores differed between patients with HD and ARM (p = 0.09 and p = 0.72 respectively). In addition, when interactions between these scores and diagnosis were evaluated, they were not found to be significant for either total, psychosocial health, or physical health HRQoL scores (p N 0.10 for all), suggesting that differences in HRQoL by these scores did not vary between HD and ARM patients.
0.008
3. Discussion
b.001
This study has focused on the evaluation HRQoL in patients with HD and ARM (aged 0–17 years), and has identified a number of patient characteristics, comorbidities and symptoms associated with a significant reduction in the parent-reported HRQoL score. In all age groups, an associated psychological, behavioral, or developmental comorbidity was found, on regression tree analysis, to be the most important factor to be predictive of a lower total HRQoL. When analyzing the HRQoL in children reported to have specific symptoms in the 0–4 year age group, those with vomiting and abdominal distention were reported to have a lower total HRQoL. In children aged 5–17 years, those reported to have the symptoms of vomiting, nausea, abdominal distension, and abdominal pain were found to have significantly lower parent reported total HRQoL, than those without these symptoms. This may be explained by the fact that the older child is able to express feelings of nausea and abdominal pain to their guardians more consistently. Surprisingly, stooling accidents, a history of multiple previous surgical interventions, and the presence of an ostomy were not found to have a significant negative impact on parent reported total HRQoL scores. In 2008, Hartmann et al. [12] investigated factors affecting quality of life of children and adolescents, aged 8–16 years, with ARM and HD. The authors hypothesized that quality of life is affected by background characteristics including age and clinical factors (e.g. disease severity) but these are likely mediated by disease specific factors including disease functioning (e.g. fecal continence) and perceived self-confidence (e.g. school attitude). They identified a lower quality of life in those reporting psychosocial problems in all domains and global disease specific functional problems, which would include fecal incontinence. The older patients (12–16 years), females, and those with more severe forms of the disease were found to have lower overall global functioning and lower self-confidence, and this was reflected in a lower quality of life score. In our study the older age group of 5–17 years, the parent reported psychosocial HRQoL scores were significantly lower for those children with vomiting, abdominal pain, cardiac diagnoses,
b.001
2.4. Patients ages 5–17 years: physical HRQoL score When considering the variables associated with a reduction in the physical health HRQoL score, having HD, any gastrointestinal symptoms (vomiting, nausea, abdominal distension and abdominal pain), an associated chromosomal/genetic comorbidity, and a psychological, behavioral or developmental problem were all found to be significant as shown in Table 3. Patients with a genital/reproductive comorbidity (e.g. undescended testes, vaginal septum, hypospadias, and vaginal atresia) were also seen to have significantly lower physical HRQoL scores. Regression tree analysis revealed that the factor most useful for predicting physical health HRQoL scores in children 5–17 years of age was whether or not they had vomiting. Children with no vomiting, no abdominal pain and no chromosomal/genetic anomaly had particularly high physical HRQoL scores, and patients experiencing vomiting had particularly low physical HRQoL scores regardless of the presence of other symptoms or comorbidities. The results of this analysis are demonstrated in Fig. 2. 2.5. Patients aged 5–17 years: psychosocial HRQoL score Patients reported to have vomiting or abdominal pain were seen to have significantly lower psychosocial HRQoL score, compared to patients not reported to have these symptoms. In addition, patients with a cardiac, brain/neurologic, chromosomal/genetic, or psychological,
Please cite this article as: Lane VA, et al, Determinants of quality of life in children with colorectal diseases, J Pediatr Surg (2016), http://dx.doi.org/ 10.1016/j.jpedsurg.2016.08.004
V.A. Lane et al. / Journal of Pediatric Surgery xxx (2016) xxx–xxx
5
Table 3 Parent reported total, physical and psychosocial HRQoL scores across patient characteristics in children 5–17 years of age. Patient characteristic/Symptoms/Comorbidity Age group 5–7 years 8–12 years 13–17 years Gender Male Female Adoption status Not adopted Adopted Diagnosis ARM (includes all cloaca variants and imperforate anus) Hirschsprung's disease Symptoms Vomiting Yes No Nausea Yes No Abdominal distension Yes No Abdominal pain Yes No Cardiac diagnoses Yes No Genitals and Reproduction Organs Yes No Brain/Neurologic Yes No Chromosomal/Genetic Yes No Psychological/Behavioral/Developmental Yes No
N
Total HRQOL score Mean (SD)
59 53 28
76.0(18.1) 70.2(19.9) 69.6(17.5)
82 58
73.9(19.5) 70.5(17.7)
115 25
72.0(19.5) 75.1(14.9)
88 52
74.6(17.7) 69.1(20.2)
9 131
53.7(14.4) 73.8(18.4)
14 126
59.9(19.4) 73.9(18.2)
22 118
63.4(19.3) 74.2(18.3)
31 109
62.8(16.9) 75.3 (18.4)
34 102
67.7(20.4) 74.9(17.2)
43 92
76.1(16.1) 71.8(19.0)
36 100
66.0(22.7) 75.1(16.6)
15 117
54.0(20.2) 75.0(16.9)
42 95
62.5(21.2) 77.6(15.50
P value
Physical Health HRQOL score Mean (SD)
0.17
Psychosocial Health HRQOL score Mean (SD)
0.39 79.0 (25.7) 77.1 (25.0) 71.0 (26.0)
0.29
0.06
0.16
0.55 71.1 (18.5) 69.3 (16.9)
0.053 74.8 (26.3) 85.6 (19.6)
0.09
0.80 70.6 (18.6) 69.5 (14.2)
0.04 80.2 (23.2) 70.8 (28.2)
0.26 71.7 (17.6) 68.1 (18.2)
b0.001
0.002 45.5 (18.0) 78.9 (24.5)
0.04 58.7 (16.4) 71.2 (17.7)
b0.001
0.007 54.3 (26.7) 79.2 (24.2) 0.01
0.11 63.2 (18.1) 71.2 (17.7)
0.004 62.4 (29.5) 79.4 (23.90
b.001
0.08 64.2 (16.5) 71.5 (17.9)
0.001 63.6 (23.4) 80.4 (24.9)
0.049
0.004 62.3 (16.3) 72.7 (17.7)
0.13 71.7 (25.7) 79.2 (24.2)
0.20
0.047 65.7 (20.1) 72.6 (16.4)
0.02 83.4 (18.1) 74.3 (27.3)
0.01
0.59 72.2 (17.8) 70.5 (17.1)
0.08 69.4 (31.6) 79.5 (22.8)
b0.001
0.04 64.4 (20.9) 72.7 (16.0)
b0.001 56.0 (28.8) 79.1 (23.5)
b0.001
b0.001 53.0 (18.5) 72.8 (16.3) b0.001
0.03 69.3 (30.4) 80.9 (22.1)
P value
74.5 (16.4) 66.6 (19.3) 68.8 (16.5)
79.3 (24.9) 73.1 (26.0) 0.45
brain/neurological diagnoses, genetic/chromosomal diagnoses and psychological/behavioral and developmental concerns. It seems reasonable to conclude that vomiting would have a negative impact on day-to-day functioning, schooling, and interaction with friends and abdominal pain could also influence school attendance. The diagnoses of brain/neurological, genetic/chromosomal and psychological/ behavioral/development would likely all have a negative impact in the ability of the child to interact in social situations, form friendships, and acquire independence and this is reflected in a lower psychosocial HRQoL score. In our study, we have analyzed patients with ARM and HD combined. It has been considered reasonable to assume that patients with ARM and HD have comparable disease specific problems, with both groups having a congenital problem, with a chronic disease profile and both requiring surgical correction in early childhood [12]. However, we recognize that children with HD and ARM do not have the same voluntary urinary and fecal continence potential. Within the ARM subgroup specifically, the continence potential varies with the severity of the ARM and other associated factors including spinal and sacral anomalies [13]. More objective measures of continence would be beneficial to help clarify its effect on quality of life [12]. In our study we have used the Baylor Continence Scores and Vancouver NLUTD/DES to objectively measure continence in the 5–17 year age group [8,9]. Statistically significant lower psychosocial HRQoL scores were seen in those with higher scores/worse urinary function.
P value
58.9 (19.2) 75.9 (14.4)
In 2012, Grano et al. reported the findings of their study investigating how fecal incontinence may influence difference aspects of QoL in children and adolescents with an ARM [5]. In their study the effect of fecal incontinence was studied at two different time points, 4 years apart, to establish if fecal incontinence at the first time point, predicted different aspects of QoL at the second time point. Their study identified that fecal incontinence had a strong influence on several areas of QoL. The parental perception of the severity of the type of anorectal malformation was found to influence the reporting of fecal incontinence in the child; however for the anorectal malformation subgroup, this can be explained from an anatomical standpoint. The parental perception of disease severity was, however, also predictive of the child having problems with emotional and body image QoL. There has been a trend in recent years, following the introduction of bowel management programs, to establish social fecal continence at an earlier age [14] and this suggests that both the surgeon and the family are aware of the potential negative consequences fecal incontinence can have on a child's QoL [5]. Bai et al. [3] reported that behavioral problems were significantly higher in children with poor fecal continence than those with good fecal continence; however, one can argue that the fecal incontinence may lead to the behavioral issue, secondarily. In our study we do not see a significant difference in total, physical or psychosocial HRQoL in children with high/worse Baylor scores. Hartman et al. postulated that in patients with HD and ARM, the symptoms of fecal incontinence and constipation drive an emotional
Please cite this article as: Lane VA, et al, Determinants of quality of life in children with colorectal diseases, J Pediatr Surg (2016), http://dx.doi.org/ 10.1016/j.jpedsurg.2016.08.004
6
V.A. Lane et al. / Journal of Pediatric Surgery xxx (2016) xxx–xxx
All patients Mean (SD) = 72.5 (18.8) 140 patients (100%)
Psychological, behavioral or developmental comorbidity
Yes Mean (SD) =62.5 (21.2) 42 patients (30.0%)
No/Do not know Mean (SD) = 76.9 (15.9) 98 patients (70.0%)
Chromosomal/ genetic comorbidity
Yes Mean (SD) = 44.6 (16.0) 9 patients (6.4%)
No/Do not know Mean (SD) = 67.4 (19.9) 33 patients (23.6%)
Fig. 1. Regression tree for parent reported total HRQOL score in children aged 5 to 17 years. This regression tree demonstrates that for children aged 5 to 17 years, in hierarchical branching order, the variables seen to predict a lower total HRQoL score are having a psychological, behavioral or developmental comorbidity, followed by a chromosomal or genetic comorbidity.
response that influences the individual's ability to cope with these stressors, and may explain why other authors have found that patients are able to maintain a good quality of life [12]. Studies have demonstrated that adolescents and young adults with fecal incontinence have limited psychiatric morbidity, but experienced condition specific psychosocial problems affecting day-to-day life. In those with chronic constipation specifically, the chronic nature of the condition appeared to have stimulated psychological protective factors, such as positive coping strategies to deal with disease related stressors [6,12,15–18]. Previous studies have found that behavioral problems were significantly higher in children with poor fecal continence compared to those without [3]. Some groups advocate that the treatment of the psychiatric disorder is the primary concern whereas others argue that these psychological disturbances are secondary to the physiological abnormalities and suggest that the treatment of constipation and fecal incontinence should be in the pediatric setting [19,20]. The question as to whether fecal incontinence results in psychological disorders, or vice versa is difficult to establish; however, successful treatment of defecation disorders in children has been shown to normalize Child Behavior Checklist (CBCL) scores in the majority of patients, with only a minority of patients requiring referral to a mental health professional [21]. In 2012, Grano et al. reported on the long-term disease-specific quality of life in children and adolescent patients with ARM. In their conclusion, they stated that families may benefit from psychological and emotional support, considering the potential impact fecal incontinence may have on social, emotional and psychological adjustment [5]. Given that the factor found to be most predictive of lower HRQoL in our study population was an associated psychological, developmental or behavioral problem, then providing psychosocial support may improve the QoL in these patients. The definition of QoL varies, but there is a general consensus regarding two central aspects. First, QoL should be regarded as a multidimensional construct incorporating at least three domains that can be
affected by one's disease or treatment. These domains include physical, social and mental functioning. Second, QoL where possible, should be assessed from the patient's perspective; however, as in our study, it is recognized that this is not always possible [11,22]. Recommendations regarding the minimum age of child-administered QoL instruments varies from 7 to 9 years [23]. In our study, the majority (75.1%) of patients was aged 0–7 years, and we were unable to verify if surveys in the older age group were reliably completed by the child, and we therefore used parent reported surveys only. When using parent reported information, one must be aware of the influence that parent's perspective may have on the quality of life scores [24]. Previous HRQoL studies have been limited by the inability to subjectively analyze different patient age groups with regards to quality of life [12]. In our study we have used the validated PedsQL questionnaires, all of which have a parent reported version [7]. The age groups include, Infant 1–12 months, Toddlers 2–4 years, Young children 5–7 years, Children 8–12 years and Teens 13–17 years. Our patient group was divided into two groups: Infants and Toddlers (0–4 years) and Young children, Children and Teens (5–17 years). The decision for this cut off was two-fold. First, in our center, the age of 4 years is considered to be an appropriate age for potty training recognizing that in children with complex past medical and surgical history developmental milestones may be delayed. Secondly, the PedsQL surveys for children aged 5 years and above can be analyzed for total, psychosocial, and physical HRQoL. There are several limitations to this cross sectional study. Firstly, the sample size did not allow for analysis of HRQoL in the different sub groups of anorectal malformations or extent of Hirschsprung's disease; these subgroup analyses are important and will be performed after additional patient accrual. In addition, our study used parent reported HRQoL owing to the young age of our patients and the questionable validity of the child-reported data collected via online surveys. Whenever possible, HRQoL should be assessed from the patient's perspective. Finally, this is a single center study, which may not be generalizable to other centers.
Please cite this article as: Lane VA, et al, Determinants of quality of life in children with colorectal diseases, J Pediatr Surg (2016), http://dx.doi.org/ 10.1016/j.jpedsurg.2016.08.004
V.A. Lane et al. / Journal of Pediatric Surgery xxx (2016) xxx–xxx
7
All patients Mean (SD) = 76.7 (25.5) 140 patients (100%)
Vomiting
Yes Mean (SD) = 45.5 (18.0) 9 patients (6.4%)
No Mean (SD) = 78.9 (24.5) 131 patients (93.6%)
Chromosomal/genetic comorbidity
No / Do not know Mean (SD) = 81.2 (23.2) 117 patients (83.6%)
Yes Mean (SD) = 59.2 (27.1) 14 patients (10.0%)
Abdominal pain
No Mean (SD) = 84.1 (21.4) 98 patients (70.0%)
Yes Mean (SD) = 66.2 (26.9) 19 patients (13.6%)
Fig. 2. Regression tree for parent reported physical health summary HRQOL score in children age 5 to 17 years. This regression tree demonstrates that for children aged 5 to 17 years, in hierarchical branching order, the variables seen to predict a lower physical HRQoL score are vomiting, followed by an associated chromosomal or genetic comorbidity and lastly, abdominal pain.
4. Conclusions This study demonstrates that the strongest predictor of low HRQoL in patients with either Hirschsprung disease or an anorectal malformation is the presence of psychological, behavioral or developmental problems. This suggests that providing behavioral and developmental support as part of the multidisciplinary care of these patients may have a substantial impact on their quality of life. In our center, we have dedicated social workers and child life specialists to provide this type of support. The next stage in improving the care of patients treated
at our center will include screening of all children/families to identify those that may benefit from psychological, behavioral and developmental assessment and support.
References [1] Quality of life and clinical trials (editorial)Lancet 1995;346:1–2. [2] Hartman EE, Oort FJ, Aronson DC, et al. Quality of life and diseasespecific functioning of patients with anorectal malformations or Hirschsprung's disease: a review. Arch Dis Child 2011;96:398–406.
Table 4 Baylor Social Continence Scale and Vancouver DES at intake for in children aged 5–17 years old. Baylor/Vancouver Scores
Vancouver Dysfunctional Elimination Syndrome Survey score b11 ≥11 (more symptoms) Baylor Social Continence Scale score b median score ≥ median score (more symptoms)
N
Total HRQOL score Mean (SD)
P value
Physical Health HRQOL score Mean (SD)
P value
Psychosocial Health HRQOL score Mean (SD)
P value
42 21
75.8 (17.8) 66.7 (17.8)
0.06
79.6 (23.2) 71.9 (27.2)
0.24
73.7 (17.4) 63.9 (15.7)
0.03
28 29
77.8 (18.2) 68.8 (15.8)
0.051
84.5 (19.7) 72.2 (26.2)
0.051
74.3 (20.1) 67.0 (12.8)
0.11
Please cite this article as: Lane VA, et al, Determinants of quality of life in children with colorectal diseases, J Pediatr Surg (2016), http://dx.doi.org/ 10.1016/j.jpedsurg.2016.08.004
8
V.A. Lane et al. / Journal of Pediatric Surgery xxx (2016) xxx–xxx
[3] Bai Y, Yuan Z, Wang W, et al. Quality of life for children with fecal incontinence after surgically corrected anorectal malformation. J Pediatr Surg 2000;35:462–4. [4] Rintala RJ, Pakarinen MP. Outcome of anorectal malformations and Hirschsprung's disease beyond childhood. Semin Pediatr Surg 2010;19:160–7. [5] Grano C, Aminoff D, Lucidi F, et al. Long-term disease-specific quality of life in children and adolescent patients with ARM. J Pediatr Surg 2012;47:1317–22. [6] Athanasakos EP, Kemal KI, Malliwal RS, et al. Clinical and psychosocial functioning in adolescents and young adults with anorectal malformations and chronic idiopathic constipation. Br J Surg 2013;100:832–9. [7] Varni JW, Seid M, Kurtin PS. PedsQL 4.0: reliability and validity of the pediatric quality of life inventory version 4.0 generic core scales in healthy and patient populations. Med Care 2001;39(8):800–12. [8] Afshar K, Mirbagheri A, Scott H, et al. Development of a symptom score for dysfunctional elimination syndrome. J Urol 2009;182(4):1939–44. [9] Brandt ML, Daigneau C, Graviss EA, et al. Validation of the Baylor continence scale in children with anorectal malformations. J Pediatr Surg 2007;42(6):1015–21. [10] Varni JW, Limbers CA, Neighbors K, et al. The PedsQL™ infant scales: feasibility, internal consistency reliability, and validity in healthy and ill infants. Qual Life Res 2011;20:45–55. [11] Constitution of the World Health Organization. Geneva: WHO 1947. Preamble to the constitution of the World Health Organization as adopted by the International Health Conference, New York, June 19–22, 1946; signed on July 22, 1946, by the representatives of 61 states (official records of the World Health Organization, No.2, p100) and entered into force on April 7, 1948. Geneva, Switzerland: World Health Organization; 1947. [12] Hartman EE, Oort FJ, Sprangers MA, et al. Factors affecting quality of life of children and adolescents with anorectal malformations or Hirschsprung disease. J Pediatr Gastroenterol Nutr 2008 Oct;47(4):463–71.
[13] Torre M, Martucciello G, Jasonni V. Sacral development in anorectal malformations and in normal population. Pediatr Radiol 2001;31(12):858–62. [14] Bischoff A, Levitt MA, Bauer C, et al. Treatment of fecal incontinence with a comprehensive bowel management program. J Pediatr Surg 2009;44(6):1278–83. [15] Belsey J, Greenfield S, Candy D, et al. Systematic review: impact of constipation on quality of life in adults and children. Aliment Pharmacol Ther 2010;31:938–49. [16] Gabel S, Am H, Wald A, et al. Prevalence of behavior problems and mental health utilization among encopretic children: implications for behavioral pediatrics. J Dev Behav Pediatr 1986;7:293–7. [17] Friman PC, Mathews JR, Finney JW, et al. Do encopretic children have clinically significant behavioral problems? Pediatrics 1988;82:407–9. [18] Benninga MA, Voskuijl WP, Akkerhuis GW, et al. Colonic transit times and behavior profiles in children with defecation disorders. Arch Dis Child 2004;89: 13–6. [19] Levine MD, Mazonson P, Bakow H. Behavioral symptom substitution in children cured of encopresis. Am J Dis Child 1980;134:663–7. [20] Landman GB, Rappaport L, Fenton T, et al. Locus of control and self esteem in children with encopresis. J Dev Behav Pediatr 1896;7:111–3. [21] Van der Plas RN, Benninga MA, Redekop WK, et al. Randomised trial of biofeedback training for encopresis. Arch Dis Child 1996;75:367–74. [22] Aaronson NK, Meyerowitz BE. Bard me et al. Quality of life research in oncology. Past achievements and future priorities. Cancer 1991;67:839–43. [23] Eiser C, Mohay H, Morse R. The measurement of quality of life in young children. Child Health Care Dev 2000;26:401–14. [24] Grano C, Bucci S, Aminoff D, et al. Does mothers' perception of social support mediate the relationship between fecal incontinence and quality of life of the child? Pediatr Surg Int 2013;29(9):919–23.
Please cite this article as: Lane VA, et al, Determinants of quality of life in children with colorectal diseases, J Pediatr Surg (2016), http://dx.doi.org/ 10.1016/j.jpedsurg.2016.08.004
Contents lists available at ScienceDirect
Journal of Pediatric Surgery journal homepage: www.elsevier.com/locate/jpedsurg
Determinants of quality of life in children with colorectal diseases☆ Victoria A. Lane a,b,c, Kristine M. Nacion b, Jennifer N. Cooper b, Marc A. Levitt a,c, Katherine J. Deans b,c, Peter C. Minneci b,c,⁎ a b c
Center for Colorectal and Pelvic Reconstruction, Nationwide Children's Hospital, Columbus, OH, USA Center for Surgical Outcomes Research, The Research Institute at Nationwide Children's Hospital, Columbus, OH, USA Department of Pediatric Surgery, Nationwide Children's Hospital, Columbus, OH, USA
a r t i c l e
i n f o
Article history: Received 1 April 2016 Received in revised form 2 August 2016 Accepted 4 August 2016 Available online xxxx Key words: Anorectal malformation Hirschsprung disease Quality of life
a b s t r a c t Background: Health related Quality of Life (HRQoL) is an important outcome in medical care. The aim of our study was to identify characteristics associated with lower HRQoL scores in children with anorectal malformation (ARM) and Hirschsprung disease (HD). Methods: Patients younger than 18 years, with HD or ARM, who were evaluated at our center from April 2014 to August 2015, were identified. The results of comprehensive questionnaires regarding diagnosis, symptoms, comorbidities and previous medical/surgical history, and validated tools to assess urinary status, stooling status and HRQoL were evaluated. Results: In children aged 0–4 years, vomiting and abdominal distension were found to be associated with a significant reduction in total HRQoL scores. In children N4 years of age, vomiting, nausea, abdominal distension, and abdominal pain were also associated with a significantly lower HRQoL. The strongest predictor of lower HRQoL scores on regression tree analysis, in all age groups, was the presence of a psychological, behavioral or developmental comorbidity. Conclusion: Patients with either HD or an ARM who have a psychological, behavioral or developmental problem experience significantly lower HRQoL than children without such problems, suggesting that provision of behavioral/developmental support as part of the multidisciplinary care of these children may have a substantial impact on their HRQoL. © 2016 Elsevier Inc. All rights reserved.
Health related Quality of Life (HRQoL) has been established as an important patient reported outcome in medical care, especially for patients with chronic diseases where complete recovery may never be achieved [1,2]. Patients with anorectal malformations (ARMs), Hirschsprung Disease (HD) or other complex colorectal diagnoses are a particular focus in the pediatric surgical literature as they are known to encounter diverse problems affecting HRQoL. These problems include fecal incontinence, constipation, problems with toilet training, increased stool frequency, abdominal pain, and gynecological and sexual difficulties, all of which may lead to marked limitations in social functioning and emotional wellbeing in addition to the development of behavioral problems [2,3]. The surgical outcomes of HD and ARM patients have improved with modern surgical techniques, and there is increased understanding of the anatomic and physiologic derangements. Despite
☆ Funding Support: This project was supported by intramural funding from the Research Institute and the Department of Surgery at Nationwide Children's Hospital. The authors report no proprietary or commercial interest in any product mentioned or concept discussed in this article. ⁎ Corresponding author at: Center for Surgical Outcomes Research, The Research Institute at Nationwide Children's Hospital, 700 Children's Drive, J West–4th Floor, Columbus, OH 43205. Tel.: +1 614 722 3066; fax: +1 614 722 6980. E-mail address: [email protected] (P.C. Minneci).
this however, many patients have impaired fecal and urological control, which may negatively impact their health related quality of life [4]. Information on HRQoL in children with ARMs and HD is limited owing to the rarity and heterogeneous nature of these diseases [2,5,6]. Past studies have used a variety of HRQoL assessment tools, including nonvalidated tools, and tools that do not measure at least three domains of HRQoL (e.g. physical, mental, and social functioning) [2]. Some of these issues can now be addressed because of the availability of numerous validated tools to measure HRQoL in children. Currently, the most widely used tool to measure HRQoL in children is the PedsQL™ Generic Core Scales developed in the 1990s by James Varni [7]. In addition to these HRQoL scores, validated tools have been developed to assess bowel and urinary function [8,9]. The aim of our study was to identify characteristics associated with lower HRQoL scores in children with ARM and HD, thereby facilitating an understanding of which interventions might improve HRQoL in these patients. 1. Methods 1.1. Study population The Center for Colorectal and Pelvic Reconstruction at our institution is a quaternary referral center for children with complex colorectal
http://dx.doi.org/10.1016/j.jpedsurg.2016.08.004 0022-3468/© 2016 Elsevier Inc. All rights reserved.
Please cite this article as: Lane VA, et al, Determinants of quality of life in children with colorectal diseases, J Pediatr Surg (2016), http://dx.doi.org/ 10.1016/j.jpedsurg.2016.08.004
2
V.A. Lane et al. / Journal of Pediatric Surgery xxx (2016) xxx–xxx
diagnoses including HD, ARM and functional constipation. The center has a comprehensive prospective clinical registry into which patients are enrolled that interfaces with the hospital electronic health record to allow systematic reporting of patient data during routine patient care interactions. Prior to their initial appointment, patients and their families are asked to complete an extensive questionnaire providing information on their underlying diagnosis, current symptoms, comorbidities and previous medical and surgical history. They also complete a series of validated tools to assess functional urinary and stooling status and HRQoL (parent-reported and self-reported surveys). Only parent-proxy surveys were examined in this study owing to the young age of most patients and concerns about the quality of the self-reported survey data as most surveys were completed online, so it could not be verified that the patient rather than the parent truly completed the self-report survey. All patients with the diagnosis of HD and ARM who were younger than 18 years and seen in the outpatient clinic between April 2014 and August 2015 were included. All patients were consented at the time of enrollment into the Patient Registry using an opt-out consent process. This study was approved by the Institutional Review Board at our institution. 1.2. Quality of life measures Parent-reported HRQoL scores were assessed using the PedsQL™ Pediatric Quality of Life inventory [7,10]. The PedsQL is a modular instrument designed to measure HRQoL in healthy and acutely or chronically ill children and adolescents, aged 0–18 years. These PedsQL Generic Score Scales are age specific and include parent reported versions for all age groups: PedsQL Infant Scales (Infant 1–12 months, Infants 13–24 months) and PedsQL Generic Score Scales (Toddlers 2–4 years, Young children 5–7 years, Children 8–12 years and Teens 13–17 years). Therefore for children aged 0–2 years, the PedsQL Infant Scales were used which measure 5 domains: physical, emotional, social, cognitive functioning, and physical symptoms. For children aged 2–17 years, the PedsQL 4.0 Generic Core Scales-Child and Parent Report module was used which measures four domains: physical, emotional, social and school functioning. For the purpose of this analysis, total parent-reported HRQoL scores as well as physical (physical function) and psychosocial (sum of emotional, social and school functioning) HRQoL scores were examined.
1.4. Other patient characteristics evaluated Patient characteristics evaluated in this study included age, gender, adoption status and the diagnosis of HD or ARM. In addition, parents were asked whether their child currently had specific gastrointestinal symptoms including nausea, vomiting, abdominal pain and distension, constipation and fecal impaction, stool and urinary accidents, early satiety and urinary tract infection. The presence of a variety of comorbidities was also evaluated and these are categorized in Table 1. 1.5. Statistical analysis In order to account for the fact that many normal children younger than 4 years are not yet toilet trained, all analyses were stratified by age. Separate analyses were performed in patients aged 0–4 years (infant and toddler) and patients 5–17 years (young child, child and teen). Parent-reported HRQoL scores were described using means and standard deviations (SDs) for the overall patient population and across groups defined by the various patient demographic and clinical characteristics evaluated. In order to evaluate whether HRQoL differed across groups defined by patient demographic and clinical characteristics, HRQoL scores were compared across patient characteristics using ttests or analysis of variance (ANOVA). In order to identify the most important predictors of QOL among all patient characteristics evaluated, regression tree analysis was performed in lieu of traditional multivariable regression modeling, owing to expected collinearity and interactions among the evaluated variables. A regression tree involves recursive binary partitioning such that, at each split, all predictors and all possible cut point values for each of the predictors are considered, then the predictor and cut point that minimize the residual sum of squares (i.e. explain the largest amount of remaining variability) of the resulting tree are chosen. This is done repeatedly until each “leaf” in the tree reaches a minimum number of observations. Finally, 10fold cross validation is used to reduce the size of the tree in order to achieve an optimal model without overfitting the data [11]. Since this was an exploratory study, no adjustments were made for multiple comparisons and a p-value of 0.05 was considered to be statistically significant. SAS version 9.3 (SAS Institute Inc., Cary, NC) and R version 2.15.1 (R Foundations for Statistical Computing, Vienna Austria) were used for analyses. 2. Results
1.3. Bowel and urinary function scales 2.1. Cohort description The Vancouver dysfunctional elimination syndrome/non-neurogenic lower urinary tract dysfunction (NLUTD/DES) questionnaire was administered to assess urinary functional outcomes, including urinary accidents, nocturnal enuresis, dysuria and hesitancy [8]. The Vancouver NLUTD/DES questionnaire is a 14 item survey designed to be selfadministered by children 9 years and older, and self-administered by parents of children less than 9 years of age. The questionnaire uses a 5-point Likert scale. Scores range from 0 to 52 with lower scores indicating better urinary function. The developers of the questionnaire suggest a total score of 11 as the optimal cutoff score to diagnose NLUTD/DES [8]. The Baylor Continence Scale (BCS) was used to assess bowel function because it is a validated tool that is able to differentiate between normal children and those with various underlying conditions, including anorectal malformations and idiopathic constipation/enuresis [9]. This scale is a self-administered 23-item parent reported survey which measures social continence in children with ARM after surgical correction. It uses a 4-point Likert response scale. Scores range from 2 to 92 with lower scores indicating better social continence. In the report describing the development and validation of this survey, normal children without anatomic or functional continence problems were found to have an average BCS score of 11.5 whereas those with an anorectal malformation had an average BCS of 28.
A total of 549 patients younger than 18 years were enrolled in the patient registry during the study period, 353 of which had a known diagnosis of HD or ARM. Of these, 325 (92.1%) had parentreported HRQoL data and were thus included in this study. The overall mean (SD) of parent reported total HRQoL in this cohort was 74.2 (17.5). 244/325 (75.1%) were in the 0–7 year age range. 2.2. Patients aged 0–4 years Average total HRQoL scores across groups based on key demographic and clinical characteristics in the 185 patients between age 0 and 4 years are shown in Table 2. Only demographic information, and symptoms and comorbidities for which significant differences emerged in total HRQoL are shown in Table 1. The overall mean (SD) of parent reported total HRQoL in this entire cohort was 80.2 (14.4). Total HRQoL scores differed significantly by age with children aged 2–4 having higher parent-reported QOL than children aged 0–24 months, but no differences were seen in total HRQoL scores by gender, adoption status or HD vs. ARM diagnosis. Significant differences emerged in the total HRQoL score in children with vomiting and abdominal distension; however, no significant
Please cite this article as: Lane VA, et al, Determinants of quality of life in children with colorectal diseases, J Pediatr Surg (2016), http://dx.doi.org/ 10.1016/j.jpedsurg.2016.08.004
V.A. Lane et al. / Journal of Pediatric Surgery xxx (2016) xxx–xxx
3
Table 1 Comorbidities at intake. Head/Face Respiratory Cardiac
Gastrointestinal Urologic
• Cleft lip and/or palate, Micrognathia, Ear abnormality, Choanal atresia, Deviated septum, Craniosynostosis • Laryngomalacia, Tracheomalacia, Bronchomalacia, Wheezing, Asthma, Sleep apnea • Patent Ductus Arteriosus, Patent foramen ovale, Ventricular septal defect, Aortic stenosis, Aortic double arch, Coarctation of aorta, Atrioventricular can defect, Hetrotaxy, Double outlet right ventricle, Double outlet left ventricle, Tetralogy of Fallot, Transposition of great vessels, Total anomalous pulmonary venous drainage, Other valve abnormality, Pathological heart murmur • Esophageal atresia with/without trachea-esophageal fistula, Duodenal atresia, Duodenal stenosis, Duodenal web, Jejunal atresia, Ileal atresia, Colonic atresia, Malrotation, Gastroschisis, Omphalocele, Meckel's diverticulum • Kidney Renal agenesis, Crossed fused ectopic kidney, Horseshoe kidney, Kidney malrotation, Dysplastic kidney, Cystic kidney, Renal stones, Chronic kidney disease, Hydronephrosis • Bladder Bladder stones, Bladder duplication, Bladder exstrophy, Bladder neck outflow obstruction, Bladder neck incompetence, Urachal cyst, Bladder diverticulum, Neurogenic bladder • Ureter Vesicoureteric reflux, Ectopic ureter, Duplicated ureter, Mega-ureter, Ureterocele, Uretero-pelvic junction (UPJ) obstruction
• Urethra Posterior urethral valves, Urogenital sinus, Urethral diverticulum, Urethral stricture, Duplicated urethra, Urethral stenosis, Urethral atresia Genitals/Reproductive Disorder of sexual development (DSD), Hydrocolpos, Hematocolpos, Absent uterus, Uterine didelphys, Bicornate uterus, Unicornate uterus, Vaginal atresia, Vaginal septum, Cervical atresia, Clitoromegaly, Penoscrotal transposition, Bifid scrotum, Undescended testicle(s), Absent testis (es), Hypospadias, Micropenis, Mullerian, vaginal or other gynecologic anomaly Skeletal/Vertebral Oligodactyly, Polydactyly, Syndactyly, Absent radius, Talipes, Absent radius, Congenital dislocation of hips, Kyphosis, Scoliosis, Rib anomalies, Cervical/thoracic/lumbar hemivertebrae, Sacral dysplasia, Absent coccyx, Pre-sacral mass/tumor, Meningocele, Myelomeningocele, Tethered cord, Occult spinal dysraphism Brain/Neurologic Hydrocephalus, Hydrocephaus with shunt, Seizures, Abnormal hearing, Deafness, Abnormal vision, Blindness, Brain injury Endocrine/Metabolic Diabetes Type 1, Diabetes Type 2, Thyroid disease, Electrolyte abnormality, Malignant hyperthermia Chromosomal/Genetic Trisomy 21, Trisomy 18, Mitochondrial disease, Currarino syndrome, Pallister Hall, Mowat Wilson, Waardenburg Shah (I–IV), Cat-Eye Syndrome, Inherited metabolic disorder Hematologic/Lymphatic/Oncologic Anemia, Bleeding/clotting disorder, Thalassemia-alpha and beta, Sickle cell disease, Malignancy/cancer Psychological/behavioral/developmental Developmental delay, Attention Deficit Disorder (ADD), Attention Deficit hyperactivity Disorder (ADHD), Obsessive Compulsive Disorder (OCD), Autism Spectrum Exposure/Infectious disease Tuberculosis, Methicillin Resistant Staphylococcus aureus (MRSA), Resistant Gram-Negative rods (RGNR), Vancomycin-Resistant Enterococcus, Pertussis
differences emerged in those reported to have gastrointestinal and urinary tract comorbidities, abdominal pain, constipation, stool and urinary accidents, early satiety and fecal impaction (data not shown). No difference in total HRQoL was seen between HD and ARM patients; however, significantly lower total HRQoL was reported in patients with any comorbidity (Table 2, and more specifically in patients with comorbidities involving diagnoses relating to the head/ face such as cleft lip/palate and micrognathia, and diagnoses related to the airway/trachea/esophagus and respiratory tract such as asthma and tracheomalacia. Other comorbidities including cardiac diagnoses, gastrointestinal, renal, skeletal, brain/neurologic and endocrine were not found to be associated with significant changes in total HRQoL scores. Patients reported to have a chromosomal anomaly or a psychological, behavioral, or developmental comorbidity also had significantly lower total HRQoL than patients without (p b 0.001 for both). When analyzing whether patients with and without a history of previous surgical procedures had different HRQoL, total HRQoL did not differ according to either the number of previous surgical procedures (including urological procedures or previous history of an ostomy), or the presence of an ostomy (p N 0.10 for all, data not shown). In regression tree analysis, the factor most useful for distinguishing young children with ARM or HD with higher versus lower HRQoL scores was whether they had a psychological, behavioral, or developmental comorbidity (Mean (SD) 66.4 (14.1) if present vs. 81.8 (13.5) if not present). The addition of other patient characteristics did not significantly improve model fit. When patients with HD and ARM were evaluated separately, in both groups, those with a psychological, behavioral, or developmental comorbidity had significantly lower HRQoL than those without this comorbidity (65.8 (13.0) vs. 82.7 (11.6) in ARM patients, 66.9 (15.8) vs. 80.1 (16.5) in HD patients, p b 0.05 for both).
2.3. Patients aged 5–17 years: Total HRQoL score Average total HRQoL scores across groups based on key demographic and clinical characteristics in the 140 patients aged 5–17 years are shown in Table 3. The overall mean (SD) of the parent reported total HRQOL score in this age group was 72.5 (18.8). Among patients aged 5–17 years, HRQoL scores did not differ by age, gender or adoption status. The overall HRQoL score was not significantly different between the ARM group and the HD group. When considering the symptoms associated with a reduction in the total HRQoL score vomiting, nausea, abdominal distension, and abdominal pain were found to be statistically significant. Total HRQoL did not differ between patients with and without stooling accidents and urinary accidents, and thus these are not reported in Table 3. Psychological/behavioral/developmental diagnoses, chromosomal anomalies, cardiac diagnoses, and brain/neurological diagnoses led to lower reported total HRQoL score (Table 3). No significant difference in total HRQoL score was seen in children with or without head/face diagnoses, respiratory, gastrointestinal, urologic, skeletal, endocrine, hematologic diagnoses or infectious diseases. In this age group, past surgical history once again was not associated with lower total HRQoL scores nor was having a current colostomy or ileostomy (data not shown). Regression tree analysis for total HRQoL (Fig. 1) revealed that the factor most useful for predicting total HRQoL in school age children with HD or ARM was whether they had a psychological, behavioral, or developmental comorbidity. In addition, in those patients who had a psychological, behavioral, or developmental comorbidity, the presence of a chromosomal/genetic comorbidity was an important predictor of total HRQoL score.
Please cite this article as: Lane VA, et al, Determinants of quality of life in children with colorectal diseases, J Pediatr Surg (2016), http://dx.doi.org/ 10.1016/j.jpedsurg.2016.08.004
4
V.A. Lane et al. / Journal of Pediatric Surgery xxx (2016) xxx–xxx
Table 2 Parent reported total HRQoL scores across various patient characteristics in children 0–4 years of age. Patient characteristic/Symptoms/Comorbidity
Patient Characteristic Age group b1 year 1 to b2 years 2–4 years Gender Male Female Adoption status Not adopted Adopted Diagnosis ARM (includes all cloaca variants and imperforate anus) Hirschsprung disease Current Symptoms Vomiting Yes No Abdominal distension Yes No Comorbidities Any comorbidity Yes No Head/Face Yes No Airway/Trachea/Esophagus/Respiratory Yes No Chromosomal/Genetic Yes No Psychological/Behavioral/Developmental Yes No
N
HRQOL score Mean (SD)
-
P value
0.02 57 32 96
77.9 (15.1) 76.1 (15.0) 82.9 (13.2)
114 71
80.0 (14.0) 80.5 (15.0)
143 41
79.5 (14.5) 82.5 (13.8)
119
81.3 (12.6)
66
78.1 (17.0)
12 173
69.0 (17.3) 81.0 (13.8)
50 135
74.4 (16.2) 82.3 (13.0)
137 48
78.9 (14.3) 83.8 (13.9)
8 174
67.6 (17.0) 80.9 (14.0)
17 163
71.5 (13.4) 81.2 (14.3)
19 158
69.2 (16.3) 82.0 (13.2)
20 162
66.4 (14.1) 82.0 (13.4)
behavioral, or developmental comorbidity were also seen to have significantly lower psychosocial HRQoL scores. The only variable that significantly discriminated between patients with lower and higher psychosocial HRQoL scores in regression tree analysis was the presence of a psychological, behavioral, or developmental comorbidity. When other variables were added, none significantly improved the fit of the model. 2.6. Patients aged 5–17 years: fecal and urinary continence
0.81
0.23
0.14
0.005
b.001
0.04
0.01
Fecal and urinary functional statuses were assessed in this age group using the Baylor Continence Scale and the Vancouver dysfunctional elimination syndrome/non-neurogenic lower urinary tract dysfunction (NLUTD/DES) questionnaire respectively. Among all school age children in our study, the median Vancouver score was 14 (IQR 8–19) and the median Baylor score was 23 (IQR 17–33). Low scores for both the Baylor and the Vancouver questionnaires are indicative of better urinary and fecal control. A Vancouver score greater than or equal to 11 was seen to be significantly associated with lower psychosocial HRQoL score. Baylor Continence scores above the median were not associated with a significant decrease in the total HRQoL score overall, or in either particular domain of QoL, though the differences were in the expected direction (Table 4). Neither Vancouver nor Baylor scores differed between patients with HD and ARM (p = 0.09 and p = 0.72 respectively). In addition, when interactions between these scores and diagnosis were evaluated, they were not found to be significant for either total, psychosocial health, or physical health HRQoL scores (p N 0.10 for all), suggesting that differences in HRQoL by these scores did not vary between HD and ARM patients.
0.008
3. Discussion
b.001
This study has focused on the evaluation HRQoL in patients with HD and ARM (aged 0–17 years), and has identified a number of patient characteristics, comorbidities and symptoms associated with a significant reduction in the parent-reported HRQoL score. In all age groups, an associated psychological, behavioral, or developmental comorbidity was found, on regression tree analysis, to be the most important factor to be predictive of a lower total HRQoL. When analyzing the HRQoL in children reported to have specific symptoms in the 0–4 year age group, those with vomiting and abdominal distention were reported to have a lower total HRQoL. In children aged 5–17 years, those reported to have the symptoms of vomiting, nausea, abdominal distension, and abdominal pain were found to have significantly lower parent reported total HRQoL, than those without these symptoms. This may be explained by the fact that the older child is able to express feelings of nausea and abdominal pain to their guardians more consistently. Surprisingly, stooling accidents, a history of multiple previous surgical interventions, and the presence of an ostomy were not found to have a significant negative impact on parent reported total HRQoL scores. In 2008, Hartmann et al. [12] investigated factors affecting quality of life of children and adolescents, aged 8–16 years, with ARM and HD. The authors hypothesized that quality of life is affected by background characteristics including age and clinical factors (e.g. disease severity) but these are likely mediated by disease specific factors including disease functioning (e.g. fecal continence) and perceived self-confidence (e.g. school attitude). They identified a lower quality of life in those reporting psychosocial problems in all domains and global disease specific functional problems, which would include fecal incontinence. The older patients (12–16 years), females, and those with more severe forms of the disease were found to have lower overall global functioning and lower self-confidence, and this was reflected in a lower quality of life score. In our study the older age group of 5–17 years, the parent reported psychosocial HRQoL scores were significantly lower for those children with vomiting, abdominal pain, cardiac diagnoses,
b.001
2.4. Patients ages 5–17 years: physical HRQoL score When considering the variables associated with a reduction in the physical health HRQoL score, having HD, any gastrointestinal symptoms (vomiting, nausea, abdominal distension and abdominal pain), an associated chromosomal/genetic comorbidity, and a psychological, behavioral or developmental problem were all found to be significant as shown in Table 3. Patients with a genital/reproductive comorbidity (e.g. undescended testes, vaginal septum, hypospadias, and vaginal atresia) were also seen to have significantly lower physical HRQoL scores. Regression tree analysis revealed that the factor most useful for predicting physical health HRQoL scores in children 5–17 years of age was whether or not they had vomiting. Children with no vomiting, no abdominal pain and no chromosomal/genetic anomaly had particularly high physical HRQoL scores, and patients experiencing vomiting had particularly low physical HRQoL scores regardless of the presence of other symptoms or comorbidities. The results of this analysis are demonstrated in Fig. 2. 2.5. Patients aged 5–17 years: psychosocial HRQoL score Patients reported to have vomiting or abdominal pain were seen to have significantly lower psychosocial HRQoL score, compared to patients not reported to have these symptoms. In addition, patients with a cardiac, brain/neurologic, chromosomal/genetic, or psychological,
Please cite this article as: Lane VA, et al, Determinants of quality of life in children with colorectal diseases, J Pediatr Surg (2016), http://dx.doi.org/ 10.1016/j.jpedsurg.2016.08.004
V.A. Lane et al. / Journal of Pediatric Surgery xxx (2016) xxx–xxx
5
Table 3 Parent reported total, physical and psychosocial HRQoL scores across patient characteristics in children 5–17 years of age. Patient characteristic/Symptoms/Comorbidity Age group 5–7 years 8–12 years 13–17 years Gender Male Female Adoption status Not adopted Adopted Diagnosis ARM (includes all cloaca variants and imperforate anus) Hirschsprung's disease Symptoms Vomiting Yes No Nausea Yes No Abdominal distension Yes No Abdominal pain Yes No Cardiac diagnoses Yes No Genitals and Reproduction Organs Yes No Brain/Neurologic Yes No Chromosomal/Genetic Yes No Psychological/Behavioral/Developmental Yes No
N
Total HRQOL score Mean (SD)
59 53 28
76.0(18.1) 70.2(19.9) 69.6(17.5)
82 58
73.9(19.5) 70.5(17.7)
115 25
72.0(19.5) 75.1(14.9)
88 52
74.6(17.7) 69.1(20.2)
9 131
53.7(14.4) 73.8(18.4)
14 126
59.9(19.4) 73.9(18.2)
22 118
63.4(19.3) 74.2(18.3)
31 109
62.8(16.9) 75.3 (18.4)
34 102
67.7(20.4) 74.9(17.2)
43 92
76.1(16.1) 71.8(19.0)
36 100
66.0(22.7) 75.1(16.6)
15 117
54.0(20.2) 75.0(16.9)
42 95
62.5(21.2) 77.6(15.50
P value
Physical Health HRQOL score Mean (SD)
0.17
Psychosocial Health HRQOL score Mean (SD)
0.39 79.0 (25.7) 77.1 (25.0) 71.0 (26.0)
0.29
0.06
0.16
0.55 71.1 (18.5) 69.3 (16.9)
0.053 74.8 (26.3) 85.6 (19.6)
0.09
0.80 70.6 (18.6) 69.5 (14.2)
0.04 80.2 (23.2) 70.8 (28.2)
0.26 71.7 (17.6) 68.1 (18.2)
b0.001
0.002 45.5 (18.0) 78.9 (24.5)
0.04 58.7 (16.4) 71.2 (17.7)
b0.001
0.007 54.3 (26.7) 79.2 (24.2) 0.01
0.11 63.2 (18.1) 71.2 (17.7)
0.004 62.4 (29.5) 79.4 (23.90
b.001
0.08 64.2 (16.5) 71.5 (17.9)
0.001 63.6 (23.4) 80.4 (24.9)
0.049
0.004 62.3 (16.3) 72.7 (17.7)
0.13 71.7 (25.7) 79.2 (24.2)
0.20
0.047 65.7 (20.1) 72.6 (16.4)
0.02 83.4 (18.1) 74.3 (27.3)
0.01
0.59 72.2 (17.8) 70.5 (17.1)
0.08 69.4 (31.6) 79.5 (22.8)
b0.001
0.04 64.4 (20.9) 72.7 (16.0)
b0.001 56.0 (28.8) 79.1 (23.5)
b0.001
b0.001 53.0 (18.5) 72.8 (16.3) b0.001
0.03 69.3 (30.4) 80.9 (22.1)
P value
74.5 (16.4) 66.6 (19.3) 68.8 (16.5)
79.3 (24.9) 73.1 (26.0) 0.45
brain/neurological diagnoses, genetic/chromosomal diagnoses and psychological/behavioral and developmental concerns. It seems reasonable to conclude that vomiting would have a negative impact on day-to-day functioning, schooling, and interaction with friends and abdominal pain could also influence school attendance. The diagnoses of brain/neurological, genetic/chromosomal and psychological/ behavioral/development would likely all have a negative impact in the ability of the child to interact in social situations, form friendships, and acquire independence and this is reflected in a lower psychosocial HRQoL score. In our study, we have analyzed patients with ARM and HD combined. It has been considered reasonable to assume that patients with ARM and HD have comparable disease specific problems, with both groups having a congenital problem, with a chronic disease profile and both requiring surgical correction in early childhood [12]. However, we recognize that children with HD and ARM do not have the same voluntary urinary and fecal continence potential. Within the ARM subgroup specifically, the continence potential varies with the severity of the ARM and other associated factors including spinal and sacral anomalies [13]. More objective measures of continence would be beneficial to help clarify its effect on quality of life [12]. In our study we have used the Baylor Continence Scores and Vancouver NLUTD/DES to objectively measure continence in the 5–17 year age group [8,9]. Statistically significant lower psychosocial HRQoL scores were seen in those with higher scores/worse urinary function.
P value
58.9 (19.2) 75.9 (14.4)
In 2012, Grano et al. reported the findings of their study investigating how fecal incontinence may influence difference aspects of QoL in children and adolescents with an ARM [5]. In their study the effect of fecal incontinence was studied at two different time points, 4 years apart, to establish if fecal incontinence at the first time point, predicted different aspects of QoL at the second time point. Their study identified that fecal incontinence had a strong influence on several areas of QoL. The parental perception of the severity of the type of anorectal malformation was found to influence the reporting of fecal incontinence in the child; however for the anorectal malformation subgroup, this can be explained from an anatomical standpoint. The parental perception of disease severity was, however, also predictive of the child having problems with emotional and body image QoL. There has been a trend in recent years, following the introduction of bowel management programs, to establish social fecal continence at an earlier age [14] and this suggests that both the surgeon and the family are aware of the potential negative consequences fecal incontinence can have on a child's QoL [5]. Bai et al. [3] reported that behavioral problems were significantly higher in children with poor fecal continence than those with good fecal continence; however, one can argue that the fecal incontinence may lead to the behavioral issue, secondarily. In our study we do not see a significant difference in total, physical or psychosocial HRQoL in children with high/worse Baylor scores. Hartman et al. postulated that in patients with HD and ARM, the symptoms of fecal incontinence and constipation drive an emotional
Please cite this article as: Lane VA, et al, Determinants of quality of life in children with colorectal diseases, J Pediatr Surg (2016), http://dx.doi.org/ 10.1016/j.jpedsurg.2016.08.004
6
V.A. Lane et al. / Journal of Pediatric Surgery xxx (2016) xxx–xxx
All patients Mean (SD) = 72.5 (18.8) 140 patients (100%)
Psychological, behavioral or developmental comorbidity
Yes Mean (SD) =62.5 (21.2) 42 patients (30.0%)
No/Do not know Mean (SD) = 76.9 (15.9) 98 patients (70.0%)
Chromosomal/ genetic comorbidity
Yes Mean (SD) = 44.6 (16.0) 9 patients (6.4%)
No/Do not know Mean (SD) = 67.4 (19.9) 33 patients (23.6%)
Fig. 1. Regression tree for parent reported total HRQOL score in children aged 5 to 17 years. This regression tree demonstrates that for children aged 5 to 17 years, in hierarchical branching order, the variables seen to predict a lower total HRQoL score are having a psychological, behavioral or developmental comorbidity, followed by a chromosomal or genetic comorbidity.
response that influences the individual's ability to cope with these stressors, and may explain why other authors have found that patients are able to maintain a good quality of life [12]. Studies have demonstrated that adolescents and young adults with fecal incontinence have limited psychiatric morbidity, but experienced condition specific psychosocial problems affecting day-to-day life. In those with chronic constipation specifically, the chronic nature of the condition appeared to have stimulated psychological protective factors, such as positive coping strategies to deal with disease related stressors [6,12,15–18]. Previous studies have found that behavioral problems were significantly higher in children with poor fecal continence compared to those without [3]. Some groups advocate that the treatment of the psychiatric disorder is the primary concern whereas others argue that these psychological disturbances are secondary to the physiological abnormalities and suggest that the treatment of constipation and fecal incontinence should be in the pediatric setting [19,20]. The question as to whether fecal incontinence results in psychological disorders, or vice versa is difficult to establish; however, successful treatment of defecation disorders in children has been shown to normalize Child Behavior Checklist (CBCL) scores in the majority of patients, with only a minority of patients requiring referral to a mental health professional [21]. In 2012, Grano et al. reported on the long-term disease-specific quality of life in children and adolescent patients with ARM. In their conclusion, they stated that families may benefit from psychological and emotional support, considering the potential impact fecal incontinence may have on social, emotional and psychological adjustment [5]. Given that the factor found to be most predictive of lower HRQoL in our study population was an associated psychological, developmental or behavioral problem, then providing psychosocial support may improve the QoL in these patients. The definition of QoL varies, but there is a general consensus regarding two central aspects. First, QoL should be regarded as a multidimensional construct incorporating at least three domains that can be
affected by one's disease or treatment. These domains include physical, social and mental functioning. Second, QoL where possible, should be assessed from the patient's perspective; however, as in our study, it is recognized that this is not always possible [11,22]. Recommendations regarding the minimum age of child-administered QoL instruments varies from 7 to 9 years [23]. In our study, the majority (75.1%) of patients was aged 0–7 years, and we were unable to verify if surveys in the older age group were reliably completed by the child, and we therefore used parent reported surveys only. When using parent reported information, one must be aware of the influence that parent's perspective may have on the quality of life scores [24]. Previous HRQoL studies have been limited by the inability to subjectively analyze different patient age groups with regards to quality of life [12]. In our study we have used the validated PedsQL questionnaires, all of which have a parent reported version [7]. The age groups include, Infant 1–12 months, Toddlers 2–4 years, Young children 5–7 years, Children 8–12 years and Teens 13–17 years. Our patient group was divided into two groups: Infants and Toddlers (0–4 years) and Young children, Children and Teens (5–17 years). The decision for this cut off was two-fold. First, in our center, the age of 4 years is considered to be an appropriate age for potty training recognizing that in children with complex past medical and surgical history developmental milestones may be delayed. Secondly, the PedsQL surveys for children aged 5 years and above can be analyzed for total, psychosocial, and physical HRQoL. There are several limitations to this cross sectional study. Firstly, the sample size did not allow for analysis of HRQoL in the different sub groups of anorectal malformations or extent of Hirschsprung's disease; these subgroup analyses are important and will be performed after additional patient accrual. In addition, our study used parent reported HRQoL owing to the young age of our patients and the questionable validity of the child-reported data collected via online surveys. Whenever possible, HRQoL should be assessed from the patient's perspective. Finally, this is a single center study, which may not be generalizable to other centers.
Please cite this article as: Lane VA, et al, Determinants of quality of life in children with colorectal diseases, J Pediatr Surg (2016), http://dx.doi.org/ 10.1016/j.jpedsurg.2016.08.004
V.A. Lane et al. / Journal of Pediatric Surgery xxx (2016) xxx–xxx
7
All patients Mean (SD) = 76.7 (25.5) 140 patients (100%)
Vomiting
Yes Mean (SD) = 45.5 (18.0) 9 patients (6.4%)
No Mean (SD) = 78.9 (24.5) 131 patients (93.6%)
Chromosomal/genetic comorbidity
No / Do not know Mean (SD) = 81.2 (23.2) 117 patients (83.6%)
Yes Mean (SD) = 59.2 (27.1) 14 patients (10.0%)
Abdominal pain
No Mean (SD) = 84.1 (21.4) 98 patients (70.0%)
Yes Mean (SD) = 66.2 (26.9) 19 patients (13.6%)
Fig. 2. Regression tree for parent reported physical health summary HRQOL score in children age 5 to 17 years. This regression tree demonstrates that for children aged 5 to 17 years, in hierarchical branching order, the variables seen to predict a lower physical HRQoL score are vomiting, followed by an associated chromosomal or genetic comorbidity and lastly, abdominal pain.
4. Conclusions This study demonstrates that the strongest predictor of low HRQoL in patients with either Hirschsprung disease or an anorectal malformation is the presence of psychological, behavioral or developmental problems. This suggests that providing behavioral and developmental support as part of the multidisciplinary care of these patients may have a substantial impact on their quality of life. In our center, we have dedicated social workers and child life specialists to provide this type of support. The next stage in improving the care of patients treated
at our center will include screening of all children/families to identify those that may benefit from psychological, behavioral and developmental assessment and support.
References [1] Quality of life and clinical trials (editorial)Lancet 1995;346:1–2. [2] Hartman EE, Oort FJ, Aronson DC, et al. Quality of life and diseasespecific functioning of patients with anorectal malformations or Hirschsprung's disease: a review. Arch Dis Child 2011;96:398–406.
Table 4 Baylor Social Continence Scale and Vancouver DES at intake for in children aged 5–17 years old. Baylor/Vancouver Scores
Vancouver Dysfunctional Elimination Syndrome Survey score b11 ≥11 (more symptoms) Baylor Social Continence Scale score b median score ≥ median score (more symptoms)
N
Total HRQOL score Mean (SD)
P value
Physical Health HRQOL score Mean (SD)
P value
Psychosocial Health HRQOL score Mean (SD)
P value
42 21
75.8 (17.8) 66.7 (17.8)
0.06
79.6 (23.2) 71.9 (27.2)
0.24
73.7 (17.4) 63.9 (15.7)
0.03
28 29
77.8 (18.2) 68.8 (15.8)
0.051
84.5 (19.7) 72.2 (26.2)
0.051
74.3 (20.1) 67.0 (12.8)
0.11
Please cite this article as: Lane VA, et al, Determinants of quality of life in children with colorectal diseases, J Pediatr Surg (2016), http://dx.doi.org/ 10.1016/j.jpedsurg.2016.08.004
8
V.A. Lane et al. / Journal of Pediatric Surgery xxx (2016) xxx–xxx
[3] Bai Y, Yuan Z, Wang W, et al. Quality of life for children with fecal incontinence after surgically corrected anorectal malformation. J Pediatr Surg 2000;35:462–4. [4] Rintala RJ, Pakarinen MP. Outcome of anorectal malformations and Hirschsprung's disease beyond childhood. Semin Pediatr Surg 2010;19:160–7. [5] Grano C, Aminoff D, Lucidi F, et al. Long-term disease-specific quality of life in children and adolescent patients with ARM. J Pediatr Surg 2012;47:1317–22. [6] Athanasakos EP, Kemal KI, Malliwal RS, et al. Clinical and psychosocial functioning in adolescents and young adults with anorectal malformations and chronic idiopathic constipation. Br J Surg 2013;100:832–9. [7] Varni JW, Seid M, Kurtin PS. PedsQL 4.0: reliability and validity of the pediatric quality of life inventory version 4.0 generic core scales in healthy and patient populations. Med Care 2001;39(8):800–12. [8] Afshar K, Mirbagheri A, Scott H, et al. Development of a symptom score for dysfunctional elimination syndrome. J Urol 2009;182(4):1939–44. [9] Brandt ML, Daigneau C, Graviss EA, et al. Validation of the Baylor continence scale in children with anorectal malformations. J Pediatr Surg 2007;42(6):1015–21. [10] Varni JW, Limbers CA, Neighbors K, et al. The PedsQL™ infant scales: feasibility, internal consistency reliability, and validity in healthy and ill infants. Qual Life Res 2011;20:45–55. [11] Constitution of the World Health Organization. Geneva: WHO 1947. Preamble to the constitution of the World Health Organization as adopted by the International Health Conference, New York, June 19–22, 1946; signed on July 22, 1946, by the representatives of 61 states (official records of the World Health Organization, No.2, p100) and entered into force on April 7, 1948. Geneva, Switzerland: World Health Organization; 1947. [12] Hartman EE, Oort FJ, Sprangers MA, et al. Factors affecting quality of life of children and adolescents with anorectal malformations or Hirschsprung disease. J Pediatr Gastroenterol Nutr 2008 Oct;47(4):463–71.
[13] Torre M, Martucciello G, Jasonni V. Sacral development in anorectal malformations and in normal population. Pediatr Radiol 2001;31(12):858–62. [14] Bischoff A, Levitt MA, Bauer C, et al. Treatment of fecal incontinence with a comprehensive bowel management program. J Pediatr Surg 2009;44(6):1278–83. [15] Belsey J, Greenfield S, Candy D, et al. Systematic review: impact of constipation on quality of life in adults and children. Aliment Pharmacol Ther 2010;31:938–49. [16] Gabel S, Am H, Wald A, et al. Prevalence of behavior problems and mental health utilization among encopretic children: implications for behavioral pediatrics. J Dev Behav Pediatr 1986;7:293–7. [17] Friman PC, Mathews JR, Finney JW, et al. Do encopretic children have clinically significant behavioral problems? Pediatrics 1988;82:407–9. [18] Benninga MA, Voskuijl WP, Akkerhuis GW, et al. Colonic transit times and behavior profiles in children with defecation disorders. Arch Dis Child 2004;89: 13–6. [19] Levine MD, Mazonson P, Bakow H. Behavioral symptom substitution in children cured of encopresis. Am J Dis Child 1980;134:663–7. [20] Landman GB, Rappaport L, Fenton T, et al. Locus of control and self esteem in children with encopresis. J Dev Behav Pediatr 1896;7:111–3. [21] Van der Plas RN, Benninga MA, Redekop WK, et al. Randomised trial of biofeedback training for encopresis. Arch Dis Child 1996;75:367–74. [22] Aaronson NK, Meyerowitz BE. Bard me et al. Quality of life research in oncology. Past achievements and future priorities. Cancer 1991;67:839–43. [23] Eiser C, Mohay H, Morse R. The measurement of quality of life in young children. Child Health Care Dev 2000;26:401–14. [24] Grano C, Bucci S, Aminoff D, et al. Does mothers' perception of social support mediate the relationship between fecal incontinence and quality of life of the child? Pediatr Surg Int 2013;29(9):919–23.
Please cite this article as: Lane VA, et al, Determinants of quality of life in children with colorectal diseases, J Pediatr Surg (2016), http://dx.doi.org/ 10.1016/j.jpedsurg.2016.08.004