See “Call to Action: Children With Intestinal Failure Deserve Routine Screening for Neurodevelopmental Disabilities” by Chan et al on page 157.
What Is Known/What Is New
What Is Known
- Survival in pediatric intestinal failure has dramatically improved in recent years.
- Children with intestinal failure often have complex surgical, medical, nutritional, and physical needs, with a growing awareness of heightened developmental risks.
- Existing outcome studies suggest motor deficits.
What Is New
- Within our sample, there was a high prevalence of diagnosed learning and attentional needs.
- Poorer cognitive outcomes were associated with length of first-year admission, hyperbilirubinemia, and intestinal failure diagnosis, with ≥2 first-year septic episodes predicting most cognitive variables when adjusting for prematurity.
- Managing early sepsis may be particularly important for reducing long-term cognitive risks.
Intestinal failure (IF) is defined as the inability to maintain fluid, nutrient, and electrolyte balance, because of congenital defect or disease, dysfunction or resection of the intestine. The most common cause of pediatric IF is short bowel syndrome, resulting from conditions such as necrotizing enterocolitis (NEC), gastroschisis, intestinal atresia or mid-gut volvulus (1–4). Despite an increased prevalence of IF (1–3), long-term survival has greatly improved (to around 90%), largely attributed to advancements in medical/surgical management (1), and the implementation of multidisciplinary Intestinal Rehabilitation Programs (IRP) (2–4). There has been increased awareness of developmental morbidities and improving quality of life. Thus, the present study aims to characterize the neurocognitive outcomes of children with various IF etiologies treated at a single IRP, with the goal of better understanding the impact of treatment on development and to facilitate timely and appropriate interventions.
Children with IF are at high risk of early cognitive and motor deficits because of the combined impact of preterm birth (5–9), low birth weight (BW), parenteral nutrition (PN) dependence, frequent infection, prolonged hospitalization, early medical/surgical treatments, and associated liver disease. Previous research in children with NEC and gastroschisis (10–17) indicate delays in early physical and/or intellectual development (16–18,20). In a sample of toddlers with IF (n = 33, ages 12–15 months) at our centre (19), poorer early cognitive outcomes were associated with: earlier gestational age (GA), lower BW, central nervous system morbidities, NEC diagnosis, greater length of stay in neonatal intensive care, greater number of surgeries, and conjugated hyperbilirubinemia.
Studies examining longer term cognitive outcomes indicate variable findings. Harris et al (20) yielded reassuring results in their sample of 39 children (aged 5–17 years), for which mean overall IQ estimates were average and on par with local (Australian) norms. However, in another sample of 19 children (aged 6–13) with NEC, almost 75% demonstrated borderline or clinically impaired motor skills, with more complex early surgery resulting in poorer intellectual functioning (14).
This study examines school-age neurocognitive outcomes in a sample of children with various IF etiologies, and also explores the associations between neurocognitive outcomes and potentially related medical and social factors. In this way, we address the following objectives:
Objective 1: Provide an overview of the neurocognitive outcomes of children with IF at ages 4 to 8 years. To do this, we calculated the proportion of children who met diagnostic criteria for ADHD, Specific Learning Disorder, or Intellectual Disability based on the most recent version of the DSM published at the time of the assessment (DSM-IV-TR or V) (21,22).
Objective 2: Explore the relationships between a broad range of neurocognitive outcomes and various medical variables or demographic variables. We calculated bivariate Pearson or Spearman correlations between our neurocognitive outcomes and a range of medical and demographic variables to identify the most influential factors worthy of further investigation.
Objective 3: Determine whether neurocognitive outcomes can be predicted by certain medical variables or demographic variables, adjusting for the impact of prematurity. The variables significantly correlated with the most outcomes in the previous analyses (from Objective 2) were used in linear regression analyses to determine if they could uniquely predict neurocognitive scores beyond what could be attributed to GA. For all analyses, P values >0.05 were considered significant.
Participants in this retrospective cohort study received treatment through the Group for Improvement of Intestinal Function and Treatment (GIFT) at the Hospital for Sick Children (Toronto, Canada). The GIFT Program is a multidisciplinary IRP constituted of health care providers from general surgery, gastroenterology, neonatology, transplantation, clinical nutrition, nursing, social work, physiotherapy, occupational therapy, speech language pathology, and psychology. To meet inclusion criteria for this study, participants must have attended a GIFT clinical appointment between 4 and 8 years of age between January 1, 2012 and March 31, 2016. Exclusion criteria included: children who had received a liver or intestinal transplant, as this is known to be associated with independent neurocognitive risks (23–25); children who could not speak English proficiently, were inpatients, or who had severe sensory or physical disability preventing their ability to complete an assessment.
Medical and Demographic Data
This retrospective study was approved by the Research Ethics Board at The Hospital for Sick Children, which allowed us to gather and analyze data from completed patient assessments. All medical and demographic information was collected from hospital records by an experienced neonatology nurse practitioner or through clinical interview by the medical team (see Table 1). Demographic variables collected included: sex, the child's English as a first language status, level of maternal education, and presence of a sibling at home. Medical variables collected included: GA, BW, etiology of IF, percentage of small and large bowel remaining after initial surgery, extent of prematurity (<37 weeks’ gestation), history of seizures, sensory issues (use of glasses or hearing aids), or central nervous system (CNS) abnormalities (intraventricular hemorrhage grades 3 or 4, microcephaly, or periventricular leukomalacia), and number on total PN at time of assessment. Within the first year of life, data on number of inpatient days, total PN days, total number of septic episodes (as defined by positive blood cultures with a change in clinical status, typically associated with a fever at presentation), and sustained (for a minimum of 2 weeks) conjugated bilirubin (>100 μmol/L) not associated with a septic or surgical event, was collected. All children would have been eligible to receive early rehabilitation intervention as part of their treatment through the GIFT program.
Our battery included measures of intellectual functioning, memory, visual-motor skills, language, and academics (see Table 2). Relevant cognitive or learning diagnoses were provided as per DSM-IV-TR or DSM-V criteria based on data from the clinical interview, standardized questionnaires, and age-normed neurocognitive scores. Assessment was completed by an experienced Psychologist (A.G., A.D.) or trained Psychometrist, under the supervision of a Psychologist (A.G.). Ages were not corrected for prematurity, in keeping with recommended clinical practice that scores not be corrected past 3 years of age (26). An important goal of these assessments was to identify children who would require special educational services at school, which requires comparing participants’ scores to their chronological-aged peers.
Within the study period, 43 children were deemed potentially eligible based upon their age. Ten children were excluded as they were transplant recipients and 5 children had severe neurodevelopmental disabilities that prevented them from participating in standardized assessment activities. The final sample consisted of 28 children. Table 1 provides a detailed summary of the population characteristics of our sample. The majority of participants (61%) were born premature. The most frequent etiology of IF was NEC (29%) followed by gastroschisis (21%). During the first year of life, the majority (68%) experienced at least 1 episode of sepsis. Most of the children were living with their biological parents (89%), had mothers with college/university education (68%), and had at least 1 sibling living with them (57%). About 25% of the sample had abnormal findings on brain imaging. Table 2 illustrates the number of children that were administered each measure; numbers varied somewhat across tests because of the version of certain cognitive measures appropriate to the age of the child and the time constraints of assessment, in keeping with typical clinical practice.
Study Objective 1: Cohort Cognitive and Diagnostic Profile
For the whole sample, the mean estimate of intellectual functioning was within 1 standard deviation (SD) of the mean population norm (mean Full Scale IQ standard score of 89). However, the range of intellectual functioning in our sample was large, with these estimates ranging from almost 4 SD below to 1.5 SD above the population mean (range = 53–123). Thirteen out of 28 children (46%) in our sample received a formal diagnosis; 8 were diagnosed with a learning disability (29%), 3 an intellectual disability (11%), and 2 had a dual diagnosis of learning disability and ADHD (7%).
Study Objective 2a: Neurocognitive Outcome and Medical Variables
See Table 3 for the results of all correlation analyses. Higher GA was related to better receptive vocabulary, visual-motor integration, and fine motor dexterity. Number of first-year septic episodes was negatively associated with the greatest number of neurocognitive outcomes at school age. Specifically, more frequent first year sepsis was related to lower scores on measures of intellectual functioning, working memory, receptive vocabulary, visual-motor integration, and visual memory.
Study Objective 2b: Neurocognitive Outcome and Demographic Variables
Having a sibling at home was associated with higher intellectual functioning, working memory, processing speed, literacy skills, and verbal learning at school age (see Table 3).
Study Objective 3a: Predicting Neurocognitive Outcome Based on Medical Variables (Adjusting for Prematurity)
We conducted separate linear regression analyses to determine the relative contribution of number of first year septic episodes in predicting the neurocognitive outcomes associated with this variable in the correlation analyses. First-year septic episodes significantly predicted school-age working memory (β = −0.51, P
= 0.009), visual-motor integration (β = −0.39, P = 0.038), and visual memory skills (β = −0.44, P = 0.042), even when adjusting for GA. First-year septic episodes did not significantly predict overall intelligence (P = 0.07) or receptive vocabulary (P = 0.17). GA was not a significant predictor in any of the models (P > 0.05). Given our relatively small sample size, we did not have sufficient power to include other medical variables in the regression model (beyond GA and first year sepsis) that were also found to be associated with the aforementioned neurocognitive outcomes in the correlation analyses.
Given the statistical significance of first-year septic episodes on several neurocognitive outcomes, we wanted to determine a “critical” number of first year septic episodes that might put a child at substantially more risk of adverse neurocognitive outcome at school age. We ran t-tests to compare the scores of children with different numbers of septic episodes on the neurocognitive outcomes used in the previous regression analyses. Children who had experienced 2 or more first year septic episodes had significantly lower working memory and visual-motor integration skills at school age than those who had less than 2 septic episodes (see Fig. 1).
Study Objective 3b: Predicting Neurocognitive Outcome Based on Demographic Variables (Adjusting for Prematurity)
Separate linear regression analyses were conducted using presence of a sibling to predict the neurocognitive outcomes initially identified in the correlational analyses. Adjusting for GA, having at least 1 sibling at home significantly predicted higher working memory (β = 0.38, P = 0.043), processing speed (β = 0.41, P = 0.031), word reading (β = 0.57, P = 0.033) and verbal learning abilities (β = 0.48, P = 0.026) at school age. Having a sibling did not significantly predict preliteracy skills (P = 0.08). GA was not a significant predictor in any of the models (P > 0.05).
Increased survival rates in pediatric IF has shifted the focus of clinical care and research towards understanding neurodevelopmental outcomes. Improved medical outcomes have largely been attributed to the inception of multidisciplinary IRPs. Examining long-term outcomes is critical to building a comprehensive understanding of the developmental morbidities in children with IF, and can facilitate the ability of treatment teams to implement appropriate interventions to promote optimal quality of life.
Data from initial studies within IF indicate that early cognitive delays are common in this population (19), but little is known about longer term neurocognitive functioning. Outcome studies in early brain injury, such as stroke, have shown that children may “grow into” deficits with increasing age as the rate of age-appropriate cognitive development increases (27,28). Therefore, restricting outcome research to the early developmental period (under age 4) would likely not reveal the scope of potential risks in pediatric IF. The objective of this study was to examine a range of neurocognitive outcomes in a sample of school age children treated for IF, and the medical and demographic variables associated with outcome. Our data suggests that school age children with IF are at risk of various cognitive and learning problems. We also provided early evidence of an association between first year septic episodes and increased neurocognitive risk that may be worth exploring further within the context of prospective longitudinal studies.
Previous research indicates that children with IF may be at particularly high risk of persistent cognitive difficulties because of the compounded impact of premature birth (and associated sequelae), as well as other medical treatment-related factors and complications (eg, prolonged hospitalization, sepsis, reliance on intravenous feeds) (29–31). In line with these findings, 46% of our sample met formal diagnostic criteria for learning disability (29%), ADHD (7%), or intellectual disability (11%), which is higher than Canadian prevalence rates of around 5% for ADHD (32), 4% for learning disability (33), and 1% for intellectual disability (34). Another recent investigation of neurodevelopmental outcomes in pediatric IF by Chesley et al (35) , however, provided a more optimistic outlook. In their sample of 15 children (ages 12 months to 5 years), 12 (80%) demonstrated “normal” functioning on developmental/cognitive testing. However, scores were considered “impaired” if they were 2 SD below the normative mean, whereas in clinical practice, children who receive standard scores 1 to 2 SD below normative mean are typically considered an “at-risk” group (ie, not “normal”). If these data were re-analyzed using this revised criteria, only 6 of 15 children (40%) would have demonstrated “normal” outcomes. This study also included a large (but fairly young) age range, thus potentially masking the developmental gaps often more apparent in school-aged children.
Although the average level of intellectual functioning in our sample was within 1 SD of the population mean, the range of functioning varied widely, with a higher than expected rate of specific learning diagnoses. This variability highlights the importance of identifying risk and protective factors that impact outcome. These results may be contrasted with the higher cognitively functioning sample of Harris et al (20), in which almost all of their children with gastroschisis scored within the average range on intellectual assessment. However, our sample included other IF etiologies with more variable neonatal characteristics (eg, GA, BW). Moreover, we adopted a broader clinical approach to assessing neurocognitive functioning by examining the prevalence of those with cognitive/learning diagnoses, as well as several additional domains beyond intellectual functioning, such as fine-motor skills, visual-motor integration, academics, and memory.
Several potential explanations have been proposed to explain the link between neonatal IF and subsequent delay including premature birth (36), infection (37), lengthy early hospitalization (38), nutritional status (including prolonged total PN) (37,39), disease type (13), and psycho-social factors (eg, parent education, age at childbirth, and socioeconomic status) (40–42). However, few of these factors have been empirically measured or directly explored in terms of their relationships with cognitive development. We identified several modifiable medical variables associated with poorer neurocognitive outcomes at school age, such as: higher number of first-year septic episodes, increased length of first-year admissions, and first-year sustained conjugated hyperbilirubinemia. Although we did not have sufficient power to investigate all potentially influential medical variables in our regression model, a helpful starting point was our examination of first-year sepsis in predicting school age neurocognitive functioning, while adjusting for prematurity. Critically, having at least 2 first-year septic episodes appears to significantly increase the extent of cognitive difficulties at school age, even when taking into account GA. Retrospective studies examining the impact of early sepsis on cognitive outcome in older school age children (>8 years of age, who were not age corrected), have found a greater prevalence of intellectual, attentional, and executive functioning difficulties compared with age-matched normative samples (43,44). Sepsis can increase permeability of the blood-brain barrier, which at its worst, leads to sepsis-associated encephalopathy (43). Pathophysiological processes of sepsis thought to be associated with cerebral dysfunction include inflammatory meditators, endotoxins, disturbance of the microcirculation, mitochondrial dysfunction, cell apoptosis, and hyperpyrexia (45,46).
Moreover, demographic factors are also known to impact cognitive outcomes, and may moderate the risks associated with early brain injury (47). Our data suggests that having a sibling at home (either older or younger) may be a protective factor, potentially mitigating some of the risks related to IF medical treatment (48,49). One hypothesis is that siblings may serve as role models, and help to provide a safe “training ground” for cognitive exploration. More broadly, this finding points to the role of an enriched home environment as a protective factor against adverse neurocognitive outcomes. To this end, our group has developed caregiver teaching initiatives to maximize opportunities for developmentally appropriate activities/interactions in hospital and following discharge (50).
Limitations and Directions for Future Research
Our study provides preliminary evidence that children with IF are at risk of neurocognitive problems at school age. We acknowledge that our study is limited by several factors; however, we are hopeful that our results may help to direct future prospective longitudinal study designs. As our sample size was relatively small, we did not have sufficient power to investigate the impact of other potentially influential medical variables (outside of first year sepsis and GA) on neurocognitive outcomes. Similarly, we could not statistically examine the impact of our sample's varied IF etiologies on outcome. Future multicentre collaborations would provide larger sample sizes, and allow for better examination of additional relevant medical variables on outcomes between IF etiologies. For example, comparison of means in our data suggested that NEC diagnosis might be associated with particularly poor neurocognitive development, although we lacked adequate power to statistically test this hypothesis. Given the specialized treatment provided by IRPs, all participants were treated at a single centre, possibly limiting generalizability of our findings to other institutions where certain clinical care practices may vary. We also excluded transplant recipients in an effort to capture IF outcomes more specifically; however, for this reason, we acknowledge that our study sample represents a subsample of children with IF. As this was a retrospective study, we relied on medical records to extract data on potentially influential risk factors; thus, we were only able to consider a limited range of both medical and social risk factors. There are undoubtedly other variables that would impact neurocognitive outcomes in pediatric IF. For example, future studies may give consideration to other potentially impactful risk factors, such as macronutrient or micronutrient deficiency, composition and volume of PN feeds, quality of parenting, socioeconomic status, as well as the inter-relatedness between different medical and social risk factors. Also, neurocognitive scores were only compared with published population norms. Future studies including clinically relevant control groups (eg, premature children without intestinal failure) will contribute to our understanding of the unique neurocognitive outcomes in this population.
We wish to thank Megan Vincent, Psychology Volunteer at The Hospital for Sick Children, for her assistance throughout this project.
1. Duggan CP, Jaksic T. Pediatric intestinal failure. N Engl J Med
2. Hess RA, Welch KB, Brown PI, et al. Survival outcomes of pediatric intestinal failure patients: analysis of factors contributing to improved survival over the past two decades. J Surg Res
3. Modi BP, Langer M, Ching YA, et al. Improved survival in a multidisciplinary short bowel syndrome program. J Pediatr Surg
4. Stanger JD, Oliveira C, Blackmore C, et al. The impact of multi-disciplinary intestinal rehabilitation programs on the outcome of pediatric patients with intestinal failure: a systematic review and meta-analysis. J Pediatr Surg
5. Baron IS, Rey-Casserly C. Extremely preterm birth outcome: a review of four decades of cognitive research. Neuropsychol Rev
6. Arpino C, Compagnone E, Montanaro ML, et al. Preterm birth and neurodevelopmental outcome: a review. Child's Nerv Syst
7. Allotey J, Zamora J, Cheong-See F, et al. Cognitive, motor, behavioral and academic performances of children born preterm: a meta-analysis and systematic review involving 64 061 children. BJOG
8. Aarnoudse-Moens CSH, Weisglas-Kuperus N, van Goudoever JB, et al. Meta-analysis of neurobehavioral outcomes in very preterm and/or very low birth weight children. Pediatrics
9. Mangin KS, Horwood LJ, Woodward LJ. Cognitive development trajectories of very preterm and typically developing children. Child Dev
10. Rees CM, Pierro A, Eaton S. Neurodevelopmental outcomes of neonates with medically and surgically treated necrotizing enterocolitis. Arch Dis Child Fetal Neonatal Ed
11. Hintz SR. Neurodevelopmental and growth outcomes of extremely low birth weight infants after necrotizing enterocolitis. Pediatrics
12. Bell MJ, Ternberg JL, Feigin RD, et al. Neonatal necrotizing enterocolitis: therapeutic decisions based upon clinical staging. Ann Surg
13. Walsh MC, Kliegman RM, Hack M. Severity of necrotizing enterocolitis: influence on outcome at 2 years of age. Pediatrics
14. Ta BDP, Roze E, Van Braeckel KNJA, et al. Long-term neurodevelopmental impairment in neonates surgically treated for necrotizing enterocolitis: enterostomy associated with a worse outcome. Eur J Pediatr Surg
15. Martin CR, Dammann O, Allred EN, et al. Neurodevelopment of extremely preterm infants who had necrotizing enterocolitis with or without late bacteremia. J Pediatr
16. Henrich K, Huemmer HP, Reingruber B, et al. Gastroschisis and omphalocele: treatments and long-term outcomes. Pediatr Surg Int
17. South AP, Marshall DD, Bose CL, et al. Growth and neurodevelopment at 16 to 24 months of age for infants born with gastroschisis. J Perinatol
18. Gorra AS, Needelman H, Azarow KS, et al. Long-term neurodevelopmental outcomes in children born with gastroschisis: the tiebreaker. J Pediatr Surg
19. So S, Patterson C, Gold A, et al. Early neurodevelopmental outcomes of infants with intestinal failure. Early Hum Dev
20. Harris EL, Hart SJ, Minutillo C, et al. The long-term neurodevelopmental and psychological outcomes of gastroschisis: a cohort study. J Pediatr Surg
21. American Psychiatric Association. Diagnositic and Statisitical Manual of Mental Disorders DSM-IV-TR. Washington DC: American Psychiatric Association; 2000.
22. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders: DSM-5. 5th ed.Arlington, VA: American Psychiatric Association; 2013.
23. Sorenson LG, Neighbors K, Martz K, et al. Longitudinal study of cognitive and academic outcomes after pediatric liver transplantation. J Pediatr
24. Ee LC, Lloyd O, Beale K, et al. Academic potential and cognitive functioning of long-term survivors after childhood liver transplantation. Pediatr Transplant
25. Thevenin DM, Baker A, Kato T, et al. Neurodevelopmental outcomes of infant multivisceral transplant recipients: a longitudinal study. Transplant Proc
26. Marlow N. Neurocognitive outcome after very preterm birth. Arch Dis Child Fetal Neonatal Ed
27. Levine SC, Kraus R, Alexander E, et al. IQ decline following early unilateral brain injury: a longitudinal study. Brain Cogn
28. Westmacott R, MacGregor D, Askalan R, et al. Late emergence of cognitive deficits after unilateral neonatal stroke. Stroke
29. Anderson PJ. Neuropsychological outcome of children born very preterm. Semin Fetal Neonatal Med
30. Schlapbach LJ, Aebischer M, Adams M, et al. Swiss Neonatal Network and Follow-Up Group. Impact of sepsis on neurodevelopmental outcome in a Swiss national cohort of extremely premature infants. Pediatrics
31. Leonberg BL, Chuang E, Eicher P, et al. Long-term growth and development in children after home parenteral nutrition. J Pediatr
1998; 132 (3 Pt 1):461–466.
32. Hauck TS, Lau C, Wing LL, et al. ADHD treatment in primary care. Can J Psychiatry
34. Statistics Canada. Canadian Survey on Disability, 2012. https://www150.statcan.gc.ca/n1/pub/89-654-x/89-654-x2015003-eng.htm
. Accessed December 20, 2018.
35. Chesley PM, Sanchez SE, Melzer L, et al. Neurodevelopmental and cognitive outcomes in children with intestinal failure. J Pediatr Gastroenterol Nutr
36. Neubauer AP, Voss W, Kattner E. Outcome of extremely low birth weight survivors at school age: the influence of perinatal parameters on neurodevelopment. Eur J Pediatr
37. Soraisham AS, Amin HJ, Al-Hindi MY, et al. Does necrotising enterocolitis impact the neurodevelopmental and growth outcomes in preterm infants with birthweight <1250 g or = 1250g? J Paediatr Child Health
38. Thompson RH. Where we stand: twenty years of research on pediatric hospitalization and health care. Child Health Care
39. Cole CR, Hansen NI, Higgins RD, et al. Very low birth weight preterm infants with surgical short bowel syndrome: incidence, morbidity and mortality, and growth outcomes at 18 to 22 months. Pediatrics
40. Chittleborough CR, Lawlor DA, Lynch JW. Young maternal age and poor child development: predictive validity from a birth cohort. Pediatrics
41. Eccles JS. Influences of parents’ education on their children's educational attainments: the role of parent and child perceptions. London Rev Educ
42. Terry-Humen E, Manlove J, A MK. Playing Catch-up: How the Children of Teen Mothers Fare. Washington DC: National Campaign to Prevent Teen Pregnancy; 2005.
43. Bronner MB, Knoester H, Sol JJ, et al. An explorative study on quality of life and psychological and cognitive function in pediatric survivors of septic shock. Pediatr Crit Care Med
44. Elison S, Shears D, Nadel S, et al. Neuropsychological function in children following admission to paediatric intensive care: a pilot investigation. Intensive Care Med
45. Widmann CN, Heneka MT. Long-term cerebral consequences of sepsis. Lancet Neurol
46. Schmutzhard E, Pfausler B. Neurologic complications of sepsis. Handb Clin Neurol
47. Harding JF. Increases in maternal education and low-income children's cognitive and behavioral outcomes. Dev Psychol
48. McHale SM, Updegraff KA, Witeman SD. Sibling relationships and influences in childhood and adolescence. J Marriage Fam
49. Howe N, Recchia H. Tremblay RE, Barr RG, Peters RD. Sibling relations and their impact on children's development. Encyclopedia on Early Childhood Development
. Montreal: Centre of Excellence for Early Childhood Development; 2014. 17–24.
50. So S, Rogers A, Patterson C, et al. Parental experiences of a developmentally focused care program for infants and children during prolonged hospitalization. J Child Health Care