Abstract
Objective
The objective of this study is to retrospectively determine the co-occurrence, associated characteristics, and risk factors for neurodevelopmental disorders in a pediatric sickle cell disease clinic population.
Method
We investigated the co-occurrence and features of neurodevelopmental disorders in pediatric sickle cell disease through a retrospective cohort study conducted between July 2017 and January 2019. The participants were patients with sickle cell disease under 18 years of age identified from our institutions’ clinic rosters and medical records databases.
Results
A total of 276 participants were eligible for study inclusion and 65 participants were found to have various neurodevelopmental disorders. Children with sickle cell disease and neurodevelopmental disorders were more likely to have a history of multiple sickle cell disease-related complications in comparison to children with sickle cell disease without neurodevelopmental disorders. Children with sickle cell disease and neurodevelopmental disorders were more likely to use disease-modifying therapies in comparison to children with sickle cell disease without neurodevelopmental disorders (chi2 27.2, p<0.001).
Conclusion
Children with sickle cell disease and neurodevelopmental disorders have higher odds of having certain disease-related complications and higher use of disease modifying treatments than children with sickle cell disease who do not have neurodevelopmental disorders. Screening and diagnoses of neurodevelopmental disorders may be relevant to clinical management of pediatric sickle cell disease.
Key Terms: sickle cell disease, neurodevelopmental disorders, language disorders, attention-deficit hyperactivity disorder, stroke
INTRODUCTION
Sickle cell disease (SCD) is an inherited hemoglobinopathy that impacts multiple organ systems.1 Children with SCD have an increased risk of ischemic stroke and/or silent cerebral infarction, affecting their attention and cognition.2,3 These difficulties with attention, memory, and executive function are common in certain neurodevelopmental disorders (NDD), such as attention deficit hyperactivity disorder (ADHD), learning disabilities, and language disorders. Children with SCD without stroke or silent cerebral infarction can also have NDD. The etiology of the NDD could be multi-faceted, including: 1) factors known to impact child development, specifically socio-economic status, parental education levels, nutrition, and sleep; 2) factors common to other chronic illnesses, such as frequent hospitalizations, school absences, and medication side effects; and, 3) factors specific to SCD, particularly brain injury, impaired cerebral perfusion and chronic anemia.
Prior studies have looked at the co-occurrence of multiple neuro-developmental and psychiatric disorders in SCD. One study found similar rates of anxiety, depressive, and disruptive behavior disorders in patients with SCD compared to a convenience sample of race-matched control participants.4 Another study performed neuropsychological evaluations in 15 children with SCD, most commonly diagnosing depression, anxiety, and somatic disorder.5 Increased difficulties with executive function, processing speed, graphomotor skills, and internalizing behaviors were noted in a group of 65 patients with SCD who received neuropsychological evaluation, in comparison to an age and gender matched control group.6 Other studies have evaluated isolated NDD in the pediatric SCD population, specifically language disorders,7–8 ADHD,9–12 cognitive disability,13–15, and cerebral palsy.16 Specific learning disabilities17,18 and autism10 may be understudied in this population.
Our group studied 59 children with SCD with documented NDD and found associations between typically milder SCD phenotypes and attention issues, as well as asthma diagnoses and behavioral issues.10 That study only included children with SCD and diagnosed NDD, with only 22% of participants having a history of stroke and/or silent cerebral infarction. In the current study, an entire cohort of pediatric SCD clinic patients, with or without NDD, was included for comparison purposes. The objective of this study is to determine the co-occurrence, associated characteristics, and risk factors for various NDD in a pediatric SCD clinic population through a retrospective cohort study. While these disorders have different symptoms and features, this study aims to look at children with SCD and NDD both as a combined group and for each individual NDD, to explore differences between different groups of children with SCD in the same clinical setting.
METHODS
Participants
This study was a retrospective chart review approved by our Institutional Review Board. Chart review occurred between July 2017 and January 2019. The participants were patients with SCD identified from our institutions’ clinic rosters and medical records database. This group was a relatively stable, local patient population. The entire group of patients with SCD from two affiliated inner city hospitals with shared personnel, a pediatric hematology clinic and a sickle cell neurodevelopmental clinic, were screened for study inclusion. Children under 18 years of age with SCD confirmed by a pediatric hematologist using laboratory values, genetic testing, and/or medical chart review were included in the study. Children who were adopted or in foster care were excluded from the study.
Chart Review
Medical records of participants from our institution who met inclusion criteria were reviewed. Charts were first reviewed to see which patients had received evaluations from relevant providers/disciplines (neuropsychology, psychology, neurology, developmental pediatrics, hematology, ophthalmology, etc.) over time, then specific disciplines notes were reviewed and chart search options were used to look for various NDD. Data were extracted from participants’ charts, including age at time of chart review, sex, race, SCD type, history of SCD-related complications, other medical diagnoses, NDD diagnoses (specifically defined as language disorders, ADHD, specific learning disabilities in reading, reading comprehension, and mathematics, global developmental delay, intellectual disability, cerebral palsy, and autism spectrum disorders), medication history, school support services, baseline laboratory values, neuroimaging results (transcranial Doppler (TCD), MRI, and MRA), and medical, therapeutic and psychological provider evaluations. Any charts with unclear information or that did not fit the pre-defined data coding were reviewed by the first author and an additional author, depending on the information type (hematological, neuropsychological, neurodevelopmental, etc). Other NDD included in the NDD group but not categorized above included 18 patients with various NDD diagnosis, including vision impairment, hearing impairment, disruptive behavior disorder, oppositional defiant disorder, developmental coordination disorder, obsessive compulsive disorder, other specific learning disabilities such as written language disorder, seizure disorder, epilepsy, Tourette’s syndrome, anxiety and depression. The majority of NDD identified by chart review were confirmed with neuropsychological or psychological testing and/or appropriate medical/therapeutic provider diagnostic evaluation available in the patients’ medical record. The most commonly used measures in these evaluations are listed in Table 1. As an example, ADHD was diagnosed by neuropsychological testing using parent and teacher questionnaires and/or clinician behavioral and testing observations. A minimal number of patients with unconfirmed NDD (4 patients with 9 total NDD altogether) per parent report without formal evaluation available in the chart were also included in the chart review. In these few cases, the NDD were simply listed in the patients’ past medical history or it was stated that “Parent reported child has a history of…” in the provider report with no additional information.
Table 1:
Testing Materials used to identify Neurodevelopmental Disorders
| Test Names |
|---|
| Wechsler Preschool and Primary Scale of Intelligence (WPPSI-IV) |
| Wechsler Intelligence Scale for Children (WISC-V) |
| Wide Range Achievement Test – 4 (WRAT-4) |
| Kennedy Krieger Institute School-Age and Preschool-Age Packets |
| The Beery-Buktenica Developmental Test of Visual-Motor Integration (VMI-6) |
| California Verbal Learning Test (CVLT-C) |
| Delis-Kaplan Executive Function System (D-KEFS) |
| Kaufman Test of Educational Achievement (KTEA-3) |
| Comprehensive Test of Phonological Processing – 2 (CTOPP-2) |
| Vanderbilt Assessment Scale |
| Conners Comprehensive Behavior Rating Scale |
| Test of Everyday Attention for Children (TEA-Ch) |
| Behavioral Rating Inventory of Executive Function (BRIEF, BRIEF-2) |
Statistical Analyses
Data were collected using Microsoft Access and exported into Stata IC-13 for analyses. Analyses were completed using non-parametric tests (Wilcoxon-Mann-Whitney test, chi-squared test) when appropriate depending on sample size. Multivariable and bivariate logistic regression models were performed to determine which SCD-related complications were associated with NDD.
RESULTS
Table 2 shows the characteristics of the clinic patient population. A total of 371 patients from one institution’s (pediatric hematology) clinic roster and 3 additional patients from the affiliated institution’s (SCD Neurodevelopmental Clinic) medical records were identified and reviewed for study inclusion. A total of 276 participants qualified for study inclusion (the majority of excluded participants, greater than 75%, were due to age over 18 years at time of chart review; the rest were due to uncertain SCD diagnosis or being adopted or in foster care). Sixty-five participants (24%) were found to have NDD; of these, 46 participants had NDD but no history of stroke, specifically ischemic, hemorrhagic stroke, or silent cerebral infarction, per neuroradiologist, hematologist, or neurologist report. Participants with NDD were significantly older than participants without NDD (median age 11 years, IQR 6–14 vs. 8 years, IQR 4–12, p = 0.004). A total of 53 participants received either a neurodevelopmental (n=44) and/or neuropsychological (n=27) evaluation at our institution, including 8 participants without NDD.
Table 2:
Descriptive Characteristics of Pediatric Sickle Cell Disease Clinic Patients.
| Group with Neurodevelopmental Disorders (n = 65) | Group without Neurodevelopmental Disorders (n = 211) | P-value | |
|---|---|---|---|
| Median age in years when chart reviewed (IQR) | 11 (6–14) | 8 (4–12) | 0.004 |
| Male Gender (%) | 35 (54) | 104 (49) | 0.52 |
| Race (%) | 0.078 | ||
| Black | 63 (97) | 199 (94) | |
| Other | 1 (1.5) | 12 (6) | |
| Multi-Racial | 1 (1.5) | 0 (0) | |
| Ethnicity (%) | |||
| Hispanic or Latino | 2 (3) | 3 (1.4) | 0.37 |
| Median BMI in kg/m2 (IQR) | 17.3 (15.6–20.5) | 16.5 (15.3–18.5) | 0.07 |
| SCD Type (%) | 0.26 | ||
| SS | 44 (68) | 127 (60) | |
| SC | 11 (17) | 61 (29) | |
| S Beta Thalassemia0 | 2 (3) | 4 (2) | |
| S Beta Thalassemia+ | 8 (12) | 19 (9) | |
| SCD Complications (%) | |||
| Stroke | 19 (29) | 14 (7) | <0.001 |
| Silent Cerebral Infarction | 10 (15) | 8 (4) | 0.001 * |
| Seizure | 8 (12) | 4 (2) | 0.001 * |
| Headache | 9 (14) | 14 (7) | 0.075* |
| Pain Crises | 55 (85) | 131 (62) | 0.001 |
| Dactylitis | 17 (26) | 25 (12) | <0.006 |
| Acute Chest Syndrome | 35 (54) | 93 (44) | 0.051 |
| Asthma | 23 (35) | 42 (20) | 0.005 |
| Priapism (male gender only) | 7 (20) | 3 (3) | 0.003 * |
| Splenic Sequestration | 7 (11) | 23 (11) | 1* |
| Avascular Necrosis | 1 (1) | 4 (2) | 1* |
| Retinopathy | 8 (12) | 18 (9) | 0.34* |
| History of SCD Treatments (%) | 0.001 * | ||
| None | 21 (32) | 121 (57) | |
| Hydroxyurea (HU) | 30 (46) | 71 (34) | |
| Chronic Transfusion Therapy (CTT) | 5 (8) | 9 (4) | |
| HU and CTT | 9 (14) | 7 (3) | |
| HU and Bone Marrow Transplant (BMT) | 0 (0) | 2 (1) | |
| HU, CTT, and BMT | 0 (0) | 1(1) | |
| Other Medical Diagnoses (%) | |||
| Seasonal allergies | 16 (25) | 32 (15) | 0.08 |
| Eczema | 7 (11) | 25 (12) | 1* |
| Obstructive Sleep Apnea | 10 (15) | 19 (9) | 0.14 |
| Constipation | 11 (17) | 20 (9) | 0.1 |
| School Support (%) | 30 (46) | 14 (7) | <0.001 |
| None | 35 (54) | 197 (93) | |
| IEP | 25 (38) | 5 (2.5) | |
| 504 Plan | 3 (5) | 5 (2.5) | |
| Other | 2 (3) | 4 (2) | |
| Neuroimaging Completed (%) | |||
| TCD | 39 (60) | 119 (56) | 0.25 |
| TCD (if SCD type SS or Sβ0) | 39 (85) | 116 (89) | 0.52 |
| Median Max TCD Velocity in cm/sec (IQR) | 120 (104–137) | 120 (102–135) | 0.83 |
| MRI | 41 (63) | 38 (18) | <0.001 |
| MRA | 26 (40) | 28 (13) | <0.001 |
| Laboratory Values | |||
| Median Hemoglobin in g/dl (IQR) | 9.4 (8.5–10.6) | 9.4 (8–10.7) | 0.85 |
| Median Fetal Hemoglobin % (IQR) | 8 (4.8–16.8) | 9.6 (4.9–16.7) | 0.64 |
| Median Reticulocyte Count % (IQR) | 5.3 (3.1–10.8) | 5.2 (2.9–10.3) | 0.43 |
| Median White-cell count per K/mm3 (IQR) | 9.31 (7.03–11.08) | 9.91 (7.13–12.17) | 0.35 |
| Median Platelets per K/mm3 (IQR) | 308 (238–418) | 300 (225–404) | 0.51 |
indicates Fisher’s exact test used when n<10.
Note: Significant results bolded. Chi squared test and Wilcoxon rank sum test used unless otherwise indicated. Column percentages presented in table. Categories are not exclusive for all variables
Sickle Cell Disease Characteristics Analyses
Children with SCD and NDD had significantly higher historical rates of stroke, silent cerebral infarction, seizure, pain crises, dactylitis, acute chest syndrome, asthma, and priapism (in male patients only), in comparison to children with SCD without NDD (see Table 1). Children with SCD and NDD were not significantly different from children without NDD with regards to specific stroke risk factors, including gender, hemoglobin levels, systolic and diastolic blood pressure, and most recent maximum transcranial Doppler velocity.
When compared to children with SCD without NDD, children with SCD and NDD had higher historical rates of receiving disease-modifying and/or curative therapies (Fishers exact 0.001). In particular, children with SCD and NDD had higher rates of hydroxyurea usage (60%), in comparison to children with SCD without NDD (38%) (chi2 9.44, p<0.003). These findings for disease-modifying therapies in general (chi2 16.8, p<0.006) and hydroxyurea specifically (58% vs. 78.3%, chi2 6.01, p<0.02) were still significant when children with SC and Sβ+ thalassemia were removed from the analyses, as they are not commonly offered these therapeutic options due to SCD guidelines.
Children with SCD and NDD were more likely to have brain MRIs than children with SCD without NDD (63% vs. 18%, chi2 49.4, p<0.001). There was no significant difference in the rates of abnormal MRIs per radiology report between the two groups (53.7% vs. 47.4%, chi2 0.31, p=0.58). Similarly, more children with SCD and NDD had completed brain MRAs than children with SCD without NDD (40% vs. 13.3%, chi2 22.6, p<0.001), but with no significant difference in the rates of abnormal MRAs per radiology report between the groups (34.6% vs. 35.7%, chi2 0.007, p=0.93).
Education Support Services Analyses
Children with SCD and NDD had significantly higher rates of receiving (in the past or currently) school support in comparison to children with SCD without NDD (46.2% vs. 6.6%, chi2 57.9, p<0.001) (see Table 1). Of note, only 46% of children with SCD and NDD reported receiving school supports. Of the 30 children with SCD and NDD receiving school support services, 25 (83.3%) had an Individualized Education Program (IEP - support requiring special education services), 3 (10%) had a 504 plan (support requiring educational classroom accommodations), and 2 (6.67%) had other/unspecified school support services, i.e. IEP or 504 were not specifically mentioned.
Specific Neurodevelopmental Disorders Analyses
Figure 1 shows the co-occurrence of reported NDD in this clinic population, alongside other populations in the United States. While our NDD data was based primarily on neuropsychological and physician evaluation with limited parent report data, the comparison groups in Figure 1 use data collection from national parent interviews and surveys, a less rigorous but larger scale comparison group. Additional statistical analyses were completed for specific NDD with a percentage greater than 5% (language disorders and ADHD). Specific learning disabilities were not included in this group, as multiple types of learning disabilities were grouped together for the table and comparison purposes.
Figure 1.
This figure shows the co-occurrence of specific NDD in four pediatric populations: our institutions’ pediatric sickle cell disease population, the general pediatric United States population, the black children in the general pediatric United States population, and the hispanic children in the general pediatric United States population.
Language Disorders: Parent Report Data from the 2012 National Health Interview Survey for United States Children Ages 3 to 17 years
ADHD: Parent Report Data from the 2016 National Survey of Children’s Health for United States Children Ages 2 to 17 years
Specific Learning Disability: Parent Report Data from the 2013 National Health Interview Survey for United States Children Ages 3 to 17 years
Global Developmental Delay: Parent Report Data from the 2014–2016 National Health Interview Survey for United States Children Ages 3 to 17 years
Intellectual Disability: Parent Report Data from the 2014–2016 National Health Interview Survey for United States Children Ages 3 to 17 years
Cerebral Palsy: Parent Report Data from the 2011–2012 National Survey of Children’s Health for United States Children Ages 2 to 17 years
Autism: Parent Report Data from the 2014–2016 National Health Interview Survey for United States Children Ages 3 to 17 years
Speech and Language Disorders
Speech and language disorders (language disorders was used for brevity in manuscript and figures), including receptive, expressive, and pragmatic language disorders, language delay, dysarthria, speech articulation disorder, and selective mutism, had a co-occurrence of 9% in this SCD clinic population. Of the 24 patients with language disorders, 16 (66.7%) had their diagnoses validated by psychological, speech and language, or neurodevelopmental evaluation; 4 (16.7%) were reported by their primary care provider and/or hematologist to have a language disorder and 4 patients (16.7%) were reported by their parents to have a language disorder. Five (20.8%) of the children with language disorders received speech therapy.
Fifteen (62.5%) children with language disorders were male (62.5% vs. 49.2%, Fisher’s exact 0.29). Fourteen (58.3%) children had HbSS, 5 (20.8%) children had HbSC, 1 (4.2%) child had HbSβ0, and 4 (16.7%) children had HbSβ+ subtype SCD. Thirteen (54.2%) children with SCD and language disorders were or had been on hydroxyurea (including one child with HbSβ+ SCD), one (4.2%) child with SCD and a language disorder was or had been on chronic transfusion therapy, and one (4.2%) child with SCD and a language disorder was or had been on both hydroxyurea and chronic transfusion therapy. Five (45.5%) of the 11 children with SCD and language disorders who had an MRI had an abnormal MRI.
There was no significant difference in the rates of having a history of stroke between children with a history of SCD and a language disorder and children with SCD without language disorders/other NDD (20.8% vs. 11.1%, p=0.2).
Attention Deficit Hyperactivity Disorder (ADHD)
ADHD had a co-occurrence of 7% in this SCD clinic population. Of the 19 patients with ADHD, 17 (89.5%) patients had their diagnoses validated by psychological, behavioral therapy, or neurodevelopmental evaluation, 1 (5.3%) patient’s diagnosis was by other physician report (pediatric neurologist), and 1 (5.3%) patient was reported by their parents to have ADHD. Eleven (47.4%) of the children with ADHD were seen by a behavioral therapist. Twelve (63.2%) children were on an appropriate medication for ADHD, either a stimulant medication (52.6%), alpha agonist (5.3%), or both (5.3%).
Thirteen (68.4%) children with ADHD were male (68.4% vs. 49%, Fishers exact 0.152). Fourteen (73.7%) children had HbSS, 2 (10.5%) children had HbSC, 1 (5.3%) child had HbSβ0, and 2 (10.5%) children had HbSβ+. Sixteen (84.2%) children with SCD and ADHD were or had been on hydroxyurea (including one child with HbSβ+ SCD) and three (15.8%) children with SCD and ADHD were or had been on chronic transfusion therapy and hydroxyurea.
Children with SCD and ADHD had higher rates of having a history of stroke in comparison to children with SCD without NDD (36.8% vs. 10.1%, p<0.002). Eighteen (94.7%) children with SCD and ADHD had an MRI completed. Ten (55.6%) of the children with SCD, ADHD, and a completed MRI had abnormal MRIs.
Other Neurodevelopmental Disorders
Specific learning disabilities had an overall co-occurrence of 7% in this SCD clinic population, with respective co-occurrence of 4.3% for reading, 1.5% for reading comprehension, 3.3% for mathematics, and 0.4% for other specific learning disabilities. Seven (38.9%) of the 18 children with specific learning disabilities had more than one specific learning disability. Also, 41.7% of participants with SCD and a specific learning disability in reading were male, compared to 75% with a specific learning disability in reading comprehension, and 22% with a specific learning disability in mathematics. Thirteen (72.2%) children had HbSS, 2 (11.1%) children had HbSC, 1 (5.6%) child had HbSβ0, and 2 (11.1%) children had HbSβ+.
Global developmental delay had an overall co-occurrence of 5%. Eight (61.5%) of the 13 children with global developmental delay were male. Eight (61.5%) children had HbSS, 2 (15.4%) children had HbSC, and 3 (23.1%) children had HbSβ+.
Intellectual disability had an overall co-occurrence of 2%. Four (80%) of the 5 children with intellectual disability were male. All five children had HbSS.
Cerebral palsy had an overall co-occurrence of 1%. One (50%) of the 2 children with cerebral palsy were male. One (50%) child had HbSS and 1 (50%) child had HbSC.
Autism spectrum disorder had an overall co-occurrence of 0.36%. The one child with SCD and autism spectrum disorder was male and had HbSβ0 subtype SCD.
Models of predictive variables for neurodevelopmental disorders
A multivariate logistic regression model was created to determine the SCD-related complications which could predict NDD. The first model (Model A) was constructed with all SCD related complications, excluding priapism (since it only included male patients) and avascular necrosis (sample size below 10 in both groups with and without NDD). The following variables had a p-value of < 0.05: seizure, stroke, dactylitis, and pain crises. A second adjusted logistic regression model (Model B) was run with only the above 4 variables. All 4 variables remained significant in the second model. The results for both models are shown in Table 3.
Table 3:
SCD-related complications as Predictive Variables for Neurodevelopmental Disorders
| Model A | Model B | ||||||
|---|---|---|---|---|---|---|---|
| Variable | N | Unadjusted Odds Ratio | 95% Confidence Interval | p-value | Adjusted Odds Ratio | 95% Confidence Interval | p-value |
| Seizure | 8 | 6.81 | 1.81 – 25.64 | 0.005 | 7.11 | 1.92 – 26.3 | 0.003 |
| Stroke | 19 | 4.75 | 2.02 – 11.14 | <0.001 | 3.13 | 1.48 – 6.69 | 0.003 |
| Dactylitis | 17 | 3.19 | 1.44 – 7.06 | 0.004 | 5.56 | 2.47 – 12.49 | <0.001 |
| Pain Crises | 55 | 2.67 | 1.21 – 5.9 | 0.015 | 2.67 | 1.23 – 5.77 | 0.013 |
| Asthma | 23 | 1.9 | 0.93 – 3.88 | 0.08 | |||
| Headache | 9 | 1.63 | 0.58 – 4.52 | 0.35 | |||
| Retinopathy | 8 | 1.36 | 0.51 – 3.62 | 0.54 | |||
| Acute Chest Syndrome | 35 | 0.99 | 0.51 – 1.93 | 0.97 | |||
| Splenic Sequestration | 7 | 0.86 | 0.31 – 2.35 | 0.76 | |||
Dependent variable: Presence or Absence of Neurodevelopmental Disorders
Independent variables: History of seizure, dactylitis, stroke, pain crisis, asthma, headache, retinopathy, acute chest syndrome, splenic sequestration
Models of risk factors for neurodevelopmental disorders
A bivariate logistic regression model (Model C) was created to determine which individual SCD-related complications were risk factors for NDD, again excluding priapism and avascular necrosis. The following variables had a p-value of < 0.05: stroke, seizure, pain crises, dactylitis, and asthma. Due to the small sample size, related variables were also grouped by similar etiologies/characteristics/organ system involvement and run in a separate model (Model D). Groups were characterized as Neurovascular (Stroke, Seizure, Headache, Retinopathy), Pain (Pain Crises, Dactylitis), Other (Splenic Sequestration, Avascular Necrosis), and Pulmonary (Acute Chest Syndrome, Asthma). Neurovascular and Pain Groups were associated with NDD with a p-value of < 0.05. Both models were then adjusted for age at time of chart review and gender. The association between Neurovascular and Pain Groups persisted in the adjusted model for the individual and grouped variables. Model results are shown in Table 4.
Table 4:
SCD-related complications as Risk Factors for Neurodevelopmental Disorders
| Model C | |||||||
|---|---|---|---|---|---|---|---|
| Variable | N | Unadjusted Odds Ratio | 95% Confidence Interval | p-value | Adjusted Odds Ratio | 95% Confidence Interval | p-value |
| Stroke | 19 | 5.81 | 2.71 – 12.44 | <0.001 | 4.75 | 2.15 – 10.46 | <0.001 |
| Seizure | 8 | 7.26 | 2.11 – 24.99 | 0.002 | 6.5 | 1.85 – 22.87 | 0.004 |
| Headache | 9 | 2.26 | 0.93 – 5.5 | 0.072 | 1.86 | 0.74 – 4.67 | 0.18 |
| Retinopathy | 8 | 1.51 | 0.62 – 3.64 | 0.37 | 0.88 | 0.34 – 2.32 | 0.8 |
| Pain Crises | 55 | 3.36 | 1.62 – 6.96 | 0.001 | 2.87 | 1.33 – 6.21 | 0.007 |
| Dactylitis | 17 | 2.64 | 1.32 – 5.27 | 0.006 | 3.42 | 1.64 – 7.14 | 0.001 |
| Splenic Sequestration | 7 | 0.99 | 0.4 – 2.42 | 0.98 | 0.97 | 0.39 – 2.4 | 0.94 |
| Acute Chest Syndrome | 35 | 1.48 | 0.85 – 2.59 | 0.17 | 1.28 | 0.72 – 2.28 | 0.41 |
| Asthma | 23 | 2.2 | 1.2 – 4.06 | 0.011 | 1.91 | 1.02 – 3.58 | 0.04 |
| Model D | |||||||
| Neurovascular | 76 | 4.41 | 2.44 – 7.98 | <0.001 | 3.84 | 1.99 – 7.4 | <0.001 |
| Pain | 196 | 3.16 | 1.48 – 6.74 | 0.003 | 2.68 | 1.22 – 5.92 | 0.014 |
| Other | 34 | 1 | 0.43 – 2.33 | 1 | 0.96 | 0.41 – 2.28 | 0.93 |
| Pulmonary | 147 | 1.56 | 0.88 – 2.74 | 0.13 | 1.3 | 0.72 – 2.35 | 0.38 |
Groups were characterized as Neurovascular (Stroke, Seizure, Headache, Retinopathy), Pain (Pain Crises, Dactylitis), Other (Splenic Sequestration, Avascular Necrosis), and Pulmonary (Acute Chest Syndrome, Asthma).
Dependent variable: History of seizure, dactylitis, stroke, pain crisis, asthma, headache, retinopathy, acute chest syndrome, splenic sequestration
Independent variables: Presence or Absence of Neurodevelopmental Disorder
Both models were adjusted for age at time of chart review and gender.
DISCUSSION
Children with SCD have an increased risk of neurological complications in comparison to the general pediatric population. These children may also be at increased risk for NDD, in part due to their increased co-occurrence of disease-related brain injury. This study was designed to further explore the co-occurrence and features of NDD in a pediatric SCD clinic population.
Approximately 24% of children in our SCD clinics had NDD; this number decreased to 19% when participants with a history of stroke were excluded from the group. The SCD and NDD group was significantly associated with higher rates of SCD-related complications involving multiple organ systems and received a higher level of treatment for their SCD symptoms; in addition, patients with one NDD often had multiple NDD. This finding may be explained by the number of complications having a direct negative effect on neurodevelopment. Alternatively, children with more complications would require additional treatment and thus it is possible, though unlikely, that the effect may be due to the treatments.
While children with SCD and NDD in this study had an increased risk of a history of seizure or stroke in comparison to other disease-related complications, they did not exhibit increased odds of having specific stroke risk factors, specifically gender, blood pressure, and elevated transcranial Doppler velocities. This might suggest differing mechanisms for NDD and seizure or stroke. Further investigation is required because we only compared children’s most recent transcranial Doppler velocities, which could be misleading in children with prior elevated velocities necessitating a change in treatment.
Children with SCD and ADHD had an increased likelihood of having a history of stroke, while children with SCD and language disorders did not show the same association. This finding may be due to the increased prevalence of frontal lobe brain injury from stroke and silent cerebral infarctions in SCD, as frontal lobe abnormalities are also linked to ADHD.19–21 Of note, autism co-occurrence is lower than expected in our SCD population. Past research has reported differences in autism diagnosis co-occurrence rates between black and white children, though concerns have been raised about delayed diagnosis of autism in certain racial and ethnic groups due to health care disparities.22 Additional studies are needed in other SCD populations regarding this finding.
Additional statistical analyses showed associations between SCD-related complications and NDD. Specifically, children with NDD had an increased risk of stroke, seizure, pain crises, dactylitis, and asthma. When these variables were grouped, similar associations were seen between the neurovascular and pain groups. These associations suggest that children with specific SCD-related complications may be at increased risk for NDD.
As shown in Figure 1, the general co-occurrence of NDD in this population of pediatric patients with SCD is similar to or less than the general pediatric population. It is possible that rates of NDD are being underestimated, given our reliance on the medical record for identification and the known low rates of routine developmental screening.23 There are participants in this study with educational support services with no known NDD. These participants may be receiving home and hospital services for health-related school absences or 504 accommodations or IEP services without undergoing a comprehensive diagnostic evaluation. In patients without psychological testing, parent report was relied upon, which could also contribute to co-occurrence underestimations, though only a small number of children with NDD from parent report were included. Other SCD clinics with neuropsychology support services have improved detection of and treatment for these disorders, with significant impact on successful transition to adult providers and hospitalization rates for pain crises.6,24 However, many NDD are diagnosed based on clinical judgement, using broad diagnostic criteria dependent on the clinician’s training background and existing resources. Given the opportunity for remediation, future guidelines should consider neurocognitive screening for all patients with SCD.
Children with SCD and NDD were more likely to receive educational support services in comparison to children with SCD without NDD; however, less than half of the patients with SCD and NDD in this clinic population were receiving school services at the time of this chart review. This finding suggests that even when diagnostic evaluation identifies deficits, educational interventions are not implemented in over half of eligible patients. In addition, while the majority of patients with SCD and ADHD had been seen by a neurodevelopmental physician, developmental behavioral physician, or pediatric neurologist, only approximately one-half (47%) had been seen by a behavioral therapist and only 63% were on appropriate medication management for ADHD, a standard of care for treatment of ADHD in children.25 This percentage is similar to the percentage of general pediatric patients on medication for ADHD, approximately 62%.26 Though stimulant medication has been found to be safe and effective for children with SCD and neurovascular disease, one hospital found that only 21% of patients with SCD and diagnosed ADHD were prescribed ADHD medication.11,12 Although the pediatric hematology clinic at our center has embedded behavioral psychology services and is able to refer to a local sickle cell neurodevelopmental comprehensive clinic, provider education, parental support, educational specialists, and school follow up may be required to reinforce the need for medical treatment and educational accommodations of NDD in this population. Other barriers may also impact diagnosis and treatment of these disorders in this predominantly underrepresented racial and ethnic patient population.27
Our study is subject to certain limitations. Being a retrospective study, we cannot discuss temporal/causal relationships between the associations found. We also relied on parent/patient report for certain patients, which may not be a reliable source and does not employ standardized definitions for medical terminology; however, this was a very low number (n=4) of patients. We also do not have age of diagnosis information for our participants, as this information was not reliably available in the chart. We did include age at time of chart review, which was notable for the NDD group being older than the non-NDD group. This finding could mean that the non-NDD group was not old enough to have been diagnosed with NDD yet; however, the non-NDD group still had a high enough median age (8 years old) to be diagnosed with the most common NDD in this SCD clinic, language disorders and ADHD. We also did not find differences in our statistical models when age was adjusted for in the analyses. While we know that certain NDD can be associated with brain injury in SCD, we do not have temporal information to determine which was identified first, NDD or brain injury. We also do not have socioeconomic data (or surrogate data such as insurance carrier information) for our participants, which has also been shown to impact development in SCD.9 Through this chart review, we found that many SCD providers do discuss educational and cognitive concerns with patients and we were also able to review primary care provider notes for certain patients. Consideration for adding educational queries to the clinic note template regarding school services may be prudent to more reliably obtain this information.
In conclusion, NDD are associated with certain SCD-related complications and increased use of disease-modifying treatments in children with SCD. Additional attention to cognitive screening, educational support services and treatment options for these patients is warranted. Additional research is needed in other SCD populations to obtain additional information on all NDD, especially autism and specific learning disabilities, with consideration of prospective data collection to obtain more information on causality.
Conflicts of Interest and Source of Funding:
JFC holds a patent for aptamers that are potential treatments for sickle cell disease. Under a license agreement between Immunarray, Ltd., and the Johns Hopkins University, JFC is entitled to royalties on a license for a brain biomarker panel. This arrangement has been reviewed and approved by the Johns Hopkins University in accordance with its conflict of interest policies. Neither the aptamers nor any of the analytes involved in this panel were studied in the current paper. For the remaining authors, none were declared. This study was supported by the National Institutes of Health National Heart Lung and Blood Institute, K23 HL133455-01A1 Lance (PI).
Footnotes
Author Disclosure Statement: None of the authors have any additional disclosures related to this paper.
References
- 1.Rees DC, Williams TN, Gladwin MT. Sickle-cell disease. Lancet 2010; 376(9757): 2018–31. [DOI] [PubMed] [Google Scholar]
- 2.Ohene-Frempong K, Weiner SJ, Sleeper LA, et al. Cerebrovascular accidents in sickle cell disease: Rates and risk factors. Blood 1998; 91(1): 288–94. [PubMed] [Google Scholar]
- 3.DeBaun MR, Armstrong FD, McKinstry RC, et al. Silent cerebral infarcts: A review on a prevalent and progressive cause of neurologic injury in sickle cell anemia. Blood 2012; 119(20): 4587–96. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Cepeda ML, Yang YM, Price CC, et al. Mental disorders in children and adolescents with sickle cell disease. South Med J 1997; 90(3): 284–7. [DOI] [PubMed] [Google Scholar]
- 5.Nunes S, Argollo N, Mota M, et al. Comprehensive neuropsychological evaluation of children and adolescents with sickle cell anemia: A hospital-based sample. Rev Bras Hematol Hemoter 2017; 39(1): 32–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Wills KE, Nelson SC, Hennessy J, et al. Transition planning for youth with sickle cell disease: Embedding neuropsychological assessment into comprehensive care. Pediatrics 2010; 126(Suppl 3): S151–9. [DOI] [PubMed] [Google Scholar]
- 7.Schatz J, Puffer ES, Sanchez C, et al. Language processing deficits in sickle cell disease in young school-age children. Dev Neuropsychol 2009; 34(1): 122–36. [DOI] [PubMed] [Google Scholar]
- 8.Schatz J, Schlenz A, Reinman L, et al. Developmental screening in pediatric sickle cell disease: Disease-related risk and screening outcomes in 4 year olds. J Dev Behav Pediatr 2017; 38(8): 654–62. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Bills SE, Schatz J, Hardy SJ, et al. Social-environmental factors and cognitive and behavioral functioning in pediatric sickle cell disease. Child Neuropsychol 2020; 26(1): 83–99. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Lance EI, Comi AM, Johnston MV, et al. Risk factors for attention and behavioral issues in pediatric sickle cell disease. Clin Pediatr (Phila) 2015; 54(11): 1087–93. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Acquazzino MA, Miller M, Myrvik M, et al. Attention deficit hyperactivity disorder in children with sickle cell disease referred for an evaluation. J Pediatr Hematol Oncol 2017; 39(5): 350–4. [DOI] [PubMed] [Google Scholar]
- 12.Daly B, Kral MC, Brown RT, et al. Ameliorating attention problems in children with sickle cell disease: A pilot study of methylphenidate. Journal of Developmental & Behavioral Pediatrics 2012; 33(3): 244. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Kawadler JM, Clayden JD, Clark CA, et al. Intelligence quotient in paediatric sickle cell disease: A systematic review and meta-analysis. Dev Med Child Neurol 2016; 58(7): 672–9. [DOI] [PubMed] [Google Scholar]
- 14.Chen R, Pawlak MA, Flynn TB, et al. Brain morphometry and intelligence quotient measurements in children with sickle cell disease. J Dev Behav Pediatr 2009; 30(6): 509–17. [DOI] [PubMed] [Google Scholar]
- 15.Hogan AM, Pit-ten Cate IM, et al. Physiological correlates of intellectual function in children with sickle cell disease: Hypoxaemia, hyperaemia and brain infarction. Dev Sci 2006; 9(4): 379–87. [DOI] [PubMed] [Google Scholar]
- 16.Ashley-Koch A, Murphy CC, Khoury MJ, et al. Contribution of sickle cell disease to the occurrence of developmental disabilities: A population-based study. Genetics in Medicine 2001; 3(3): 181. [DOI] [PubMed] [Google Scholar]
- 17.Boulet SL, Yanni EA, Creary MS, et al. Anonymous Health status and healthcare use in a national sample of children with sickle cell disease. Am J Prev Med 2010; 38(4): S528–35. [DOI] [PubMed] [Google Scholar]
- 18.Fowler MG, Whitt JK, Lallinger RR, et al. Neuropsychologic and academic functioning of children with sickle cell anemia. J Dev Behav Pediatr 1988; 9(4): 213–20. [PubMed] [Google Scholar]
- 19.Ford AL, Ragan DK, Fellah S, et al. Silent infarcts in sickle cell disease occur in the border zone region and are associated with low cerebral blood flow. Blood 2018; 132(16): 1714–23. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Weyandt L, Swentosky A, Gudmundsdottir BG. Neuroimaging and ADHD: FMRI, PET, DTI findings, and methodological limitations. Dev Neuropsychol 2013; 38(4): 211–25. [DOI] [PubMed] [Google Scholar]
- 21.Jacobson LA, Crocetti D, Dirlikov B, et al. Anomalous brain development is evident in preschoolers with attention-Deficit/Hyperactivity disorder. J Int Neuropsychol Soc 2018; 24(6): 531–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Christensen DL, Braun KVN, Baio J, et al. Prevalence and characteristics of autism spectrum disorder among children aged 8 years - autism and developmental disabilities monitoring network, 11 sites, united states, 2012. MMWR Surveill Summ 2018; 65(13): 1–23. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Hirai AH, Kogan MD, Kandasamy V, et al. Prevalence and Variation of Developmental Screening and Surveillance in Early Childhood. JAMA Pediatr 2018; 172(9): 857–866. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Janecek JK, Dorociak KE, Piper LE, et al. Integration of neuropsychology services in a sickle cell clinic and subsequent healthcare use for pain crises. Clin Neuropsychol 2019; 33(7): 1195–211. [DOI] [PubMed] [Google Scholar]
- 25.Wolraich ML, Hagan JF, Allan C, et al. , and the Subcommittee on Children and Adolescents with Attention-Deficit/Hyperactive Disorder. Clinical practice guideline for the diagnosis, evaluation, and treatment of attention-deficit/hyperactivity disorder in children and adolescents. Pediatrics 2019; 144(4). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Danielson ML, Bitsko RH, Ghandour RM, et al. Prevalence of Parent-Reported ADHD Diagnosis and Associated Treatment Among U.S. Children and Adolescents. Journal of Clinical Child & Adolescent Psychology 2018; 47:2, 199–212, [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Alvarado C, Modesto-Lowe V. Improving treatment in minority children with attention Deficit/Hyperactivity disorder. Clin Pediatr (Phila) 2017; 56(2): 171–6. [DOI] [PubMed] [Google Scholar]