Association of Perinatal and Childhood Ischemic Stroke With Attention-Deficit/Hyperactivity Disorder

Key Points Question Does the risk of attention-deficit/hyperactivity disorder (ADHD) increase after pediatric stroke, and is that risk associated with family history of ADHD, adverse motor outcomes, or comorbid epilepsy? Findings This nationwide cohort study of 1320 patients who had perinatal or childhood stroke and 13 141 matched controls found a 2-fold increased risk of ADHD after stroke. Development of adverse motor outcomes and/or epilepsy were associated with a further elevated risk in children with perinatal stroke. Meaning These findings suggest that children may face an increased risk of ADHD after stroke and that surveillance for ADHD should be considered by pediatricians performing follow-up of children with pediatric stroke.

Symptoms of inattention and hyperactivity, as well as cognitive deficits in individuals with ADHD, often have a negative impact on academic achievements 12 and social functioning. 13The prevalence of childhood ADHD in the general population is estimated to be approximately 3% to 5%. 13,14Although the exact etiology of ADHD is often unclear, there is a strong hereditable component, 15 and associations have been found with low birth weight, prematurity, 16 male sex, 17 traumatic brain injuries, and brain tumors. 18,19 has been reported that 19% to 35% of children with cerebral palsy also have ADHD, 20 which is approximately 6 times higher than in the general population.Children with epilepsy also have a 2.5 to 5.5 times higher risk of ADHD. 21Numerous studies have reported that cerebral palsy and epilepsy are common comorbidities after pediatric stroke, [1][2][3][4][5][6][7]22 so we would also expect to see an increased risk of ADHD. Howver, it is unclear whether the risk of ADHD is increased in children with ischemic stroke who do not have these comorbidities.Previous studies assessing ADHD after pediatric stroke [8][9][10][11] have had limited sample sizes and used varying definitions of ADHD.They have also been hospital based and/or did not include general population controls, which restricted the generalizability of their results.
The main aim of our study was to address these limitations by using data from nationwide registers to evaluate the risk of ADHD after pediatric ischemic stroke in a large cohort of children.We included family history of ADHD, prematurity, small for gestational age (SGA), comorbid adverse motor outcomes, and epilepsy in our analysis.We hypothesized that children who had pediatric stroke face a higher risk of ADHD even if not born preterm or SGA or not having comorbid adverse motor outcomes and/or epilepsy.Our secondary aim was to explore whether ADHD and ischemic stroke have a shared heredity.Because diverse conditions affecting the brain have been reported to be genetically related, [23][24][25] we hypothesized that first-degree relatives of patients who had pediatric stroke would have an increased risk of ADHD.We explored this hypothesis by assessing the risk of ADHD in first-degree relatives of patients with pediatric stroke without ADHD.

Study Design and Study Population
This Swedish nationwide cohort study included children who were younger than 18 years with ischemic stroke between January 1, 1969, and December 31, 2016, and were alive 1 week after their stroke.They were identified using the National Patient Register 26,27 and the Medical Birth Register. 28dividuals with ADHD before their stroke were excluded.The Total Population Register 26 was used to identify 10 controls for each patient with stroke who were matched for sex, year of birth, and county of residence at the time of diagnosis.Parents, siblings, and offspring of patients and controls were identified from the Multigeneration Register, which is part of the Total Population Register.This cohort is part of PedStroke, a Swedish national study on pediatric stroke.Previous publications  29 risk of epilepsy, 22 and adverse motor outcomes 30 in the cohort.The study was approved by the regional ethics committee in Linköping.The authors had access to pseudonymized and encrypted data from Statistics Sweden only, therefore the ethics committee did not require individual informed consent. 31

Assessment Methods
Exposure consisted of ischemic stroke at younger than 18 years, defined by codes from the

Variables Used for Adjusted and Stratified Analyses
Perinatal stroke was defined as a diagnosis of an ischemic stroke 28 days after birth or earlier; childhood strokes were diagnosed after 28 days but before 18 years of age.Preterm birth was defined as birth before 37 plus 0 gestational weeks, and SGA was defined as birth weight of less than −2 SD for the mean birth weight for gestational age and sex according to Swedish reference data. 33verse motor outcomes consisted of a diagnosis of cerebral palsy, hemiparesis, tetraparesis, paraplegia, or any other paresis.These, together with epilepsy, were based on the ICD-8, ICD-9, and ICD-10 codes in the National Patient Register (eTable 1 in the Supplement).

Data Sources
The data from the Swedish national registers were linked using the unique personal registration number assigned to all residents at birth or immigration. 34The National Patient Register 27  The validity of the diagnoses of pediatric ischemic stroke in the National Patient Register and Medical Birth Register has been evaluated to be high.Review of medical records of parts of our cohort resulted in an overall positive predicted value of 89% for pediatric ischemic stroke, 96% for perinatal ischemic stroke, and 84% for childhood ischemic stroke. 37Diagnoses of ADHD in the National Patient Register and Prescribed Drug Register have also shown high validity. 38The registers were accessed on December 31, 2016.

Statistical Analysis
Follow-up started at the date of stroke diagnosis and the corresponding date for the controls and ended at the first ADHD diagnosis, death, or December 31, 2016, whichever came first.The follow-up for first-degree relatives started at birth and ended as described above.When stratifying by comorbidity, follow-up started at the date of the first diagnosis of adverse motor outcomes and/or epilepsy.
Cox proportional hazards regression was used to calculate hazard ratios (HRs) and assess the risk of ADHD after stroke.Each index individual was compared with his or her matched controls using an internal stratifying model.Analyses were adjusted for parental age at the birth of the child and for any ADHD diagnosis of parents and siblings.Unadjusted results can be found in eTables 2 and 3 in the Supplement.Separate analyses were performed for perinatal strokes and childhood strokes because these conditions have somewhat separate clinical symptoms and treatments.
6][17] We performed sensitivity analyses for individuals without adverse motor outcomes or epilepsy by censoring at the time of the first diagnosis of adverse motor outcomes or epilepsy after stroke.We also evaluated the risk of ADHD in parents, siblings, and offspring of children with stroke and controls.
The data analyses were performed from August 1 to 28, 2021, using SAS statistical software, version 9.4 (SAS Institute Inc), and Stata, version 16.0 (StataCorp LLC).A 95% CI that excluded 1.00 defined statistical significance.

Study Population and Baseline Characteristics
After excluding 7 children with ADHD before their stroke, 1320 individuals who were younger than 18

ADHD Risk After Pediatric Ischemic Stroke
The 1320 index individuals with pediatric stroke were twice as likely to have ADHD compared with controls when adjusting for parental age and a family history of ADHD (aHR, 2.00 [95% CI,  3).

Risk of ADHD After Perinatal Stroke
The 343 children with perinatal stroke also had an increased risk of ADHD (aHR,  3).

Risk of ADHD After Childhood Stroke
The risk of ADHD after childhood stroke was increased (aHR,  3).

Risk of ADHD in Relatives of Patients With Stroke
The siblings of patients with stroke had a higher risk of ADHD than the siblings of controls (aHR, 1.38 [95% CI, 1.09-1.74]),and the same result was seen for both perinatal and childhood stroke (Table 4).
When only including patients with pediatric stroke and without ADHD, the risk of ADHD in siblings was still elevated (aHR, 1.47 [95% CI, 1.14-1.90]).

Discussion
This nationwide cohort study of 1320 patients with pediatric stroke identified a 2-fold increased risk of ADHD after pediatric ischemic stroke.That risk remained increased after known risk factors for ADHD were considered, such as a family history of ADHD, 15 parental age, 39 prematurity, or SGA. 16e increased risk was seen after both perinatal and childhood stroke, which is in line with the study by Williams et al. 10 Other studies 9,11 did not distinguish between perinatal and childhood strokes.
Epilepsy and adverse motor outcomes have been associated with ADHD in the general population 21,40 and are common sequelae after pediatric stroke. 41,42Therefore, we also ran sensitivity analyses that censored individuals with these comorbidities.
11]43 However, most of these studies included small cohorts. 9,11,43The exception was the study by Williams et al, 10 which included 275 patients with pediatric stroke.To our knowledge, this is also the only study except ours that accounted for family history of ADHD.To our controls with ADHD, 324 had been prescribed ADHD medication.Individuals who died within the first week after a pediatric ischemic stroke or had a diagnosis of ADHD within 1 week after their stroke were excluded along with the data from their matched controls.
b Conditioned on matching set (age, sex, year of birth, and county of residence at the time of stroke) and adjusted for maternal and paternal age at the birth of the child and any ADHD in parents or siblings.
knowledge, our study is the largest to date to report an increased risk of a clinical diagnosis of ADHD after pediatric stroke.

Risk of ADHD After Perinatal Stroke
The risk of ADHD was increased after perinatal stroke and increased 6 times if the child also had adverse motor outcomes.The expected adverse motor outcome after perinatal stroke is cerebral palsy.Attention-deficit/hyperactivity disorder is more common in individuals with cerebral palsy than  Individuals who died within the first week after a pediatric ischemic stroke or had a diagnosis of ADHD within 1 week after their stroke were excluded along with the data from the matched controls.
b Conditioned on matching set (age, sex, year of birth, and county of residence at the time of stroke) and adjusted for maternal and paternal age at the birth of the child and any ADHD in parents or siblings.
in the general population, 20 and a Swedish population-based study 44 reported that half of the 200 children with cerebral palsy had positive results for ADHD.
As expected, children diagnosed with comorbid epilepsy after perinatal stroke had a higher risk of ADHD compared with controls than those with perinatal stroke without epilepsy.This result echoed the findings of Williams et al 10 and Auvin et al. 21Clinicians are already advised to screen for ADHD when children have epilepsy because this outcome is a well-known comorbid condition. 21netic and environmental factors can contribute to the comorbidity of ADHD in epilepsy, as well as an imbalance between excitation and inhibition in the brain. 45This imbalance could also contribute to the development of ADHD. 46

Risk of ADHD After Childhood Stroke
The risk of ADHD was also increased after childhood stroke, and the presence of adverse motor outcomes was not associated with the risk.Adverse motor outcomes after childhood stroke most commonly consist of hemiplegia and, for some of the younger children, cerebral palsy, in addition to less severe palsies and paresis that are often caused by a smaller insult to the brain.
Epilepsy seems to be a risk factor for ADHD after both childhood and perinatal strokes.
However, compared with controls, the risk of ADHD was also increased after childhood stroke for those without adverse motor outcomes or epilepsy.
It has been reported that the risk of ADHD is increased after other brain insults during childhood, such as severe traumatic brain injuries. 18In these cases, a decrease in the organization of white matter seems to play a crucial role in the development of ADHD, 47 and this decrease could also be important in other disorders.Whether this outcome is the case after pediatric stroke needs to be elucidated.

Risk of ADHD in First-Degree Relatives of Patients With Stroke
We found an increased risk of ADHD in the siblings of patients with pediatric stroke with and without ADHD.The rationale behind evaluating ADHD in relatives of these patients is that previous research 23 has demonstrated a high degree of genetic correlation between different psychiatric disorders.Thus, our results support the hypothesis that there is a genetic disposition that increases the risk of both ADHD and stroke.

Strengths and Limitations
0][11] Pediatric stroke is a relatively  11,43 which may overestimate the prevalence of ADHD. 49Stratified analyses of the risks of perinatal or childhood strokes and neurological comorbidities were performed.Data were retrieved from registers with high validity.The stroke diagnosis in our cohort has been validated.
This study has some limitations.There is a possibility of detection bias by misclassification underreporting, such as underreporting of stroke before the advent of modern imaging, and for surveillance bias if patients with stroke were more carefully observed.Some individuals might have had ADHD before the stroke, although the ADHD diagnosis was registered after the stroke.Fewer controls had ADHD (2.9%) than previously reported in Swedish national registers (4.3%), 50 possibly because the follow-up time was limited to the date of the stroke diagnosis in the matched index case and that a diagnosis code for ADHD is only available after ICD-9 was implemented in 1987.The number of individuals in the exposed group was relatively small for some analyses, with wide 95% CIs for some aHRs.If possible, our results should therefore be confirmed in other large cohorts of patients with pediatric stroke.
Owing to the limited number of individuals in some subgroups, we were also only able to report results compared with controls instead of making direct comparisons between children with and without comorbidities.We could not distinguish between different types of ischemic stroke, but a previous validation study 37 showed that 90% were arterial ischemic strokes.The proportion of perinatal strokes in our cohort was lower than expected.Some neonatal units may have used unspecific ICD codes (eg, cerebral ischemia) for perinatal stroke cases.Our study population was probably heterogenous in terms of the localization and size of the strokes.Outpatient diagnoses of ADHD have only been available since 2001.However, we estimate risks compared with controls, and the risks are probably not as influenced by this factor as the prevalence rates.We did not evaluate other mental health outcomes, socioeconomic status, parental health, or other child health variables that could have influenced our findings.In addition, we cannot rule out unmeasured and residual confounding variables that may have influenced the result.

Conclusions
The findings of this nationwide cohort study suggest that pediatric ischemic stroke was associated with a risk of ADHD irrespective of whether there was a family history of ADHD.This risk was even higher if there were adverse motor outcomes and/or epilepsy after perinatal stroke.Furthermore, the risk was increased in patients with childhood stroke who did not have these comorbidities.Ongoing surveillance for signs of ADHD should form an important part of follow-up programs after pediatric strokes, especially if individuals have comorbid adverse motor outcomes or epilepsy.

Table 2 .
Risk of ADHD After Pediatric Ischemic Stroke a a Attention-deficit/hyperactivity disorder was defined according to International Classification of Diseases, Ninth Revision, code 314 and International Statistical Classification of Diseases and Related Health Problems, Tenth Revision, code F90 and/or if the patient was in receipt of ADHD medication in the Prescribed Drug Register according to the Anatomical Therapeutic Chemical classification system codes N06BA01 to N06BA06, N06BA08 to N06BA12, and/or C02AC02 (guanfacine).Of the 75 index children with ADHD, 57 had been prescribed ADHD medication; of the 376

Table 3 .
Risk of ADHD After Pediatric Ischemic Stroke When Considering Perinatal Variables and Stroke Comorbidities a .28-14.20) a Attention-deficit/hyperactivity disorder was defined according to International Classification of Diseases, Ninth Revision, code 314 and International Statistical Classification of Diseases and Related Health Problems, Tenth Revision, code F90 and/or if the patient was in receipt of ADHD medication in the Prescribed Drug Register according to the Anatomical Therapeutic Chemical classification system codes N06BA01 to N06BA06, N06BA08 to N06BA12, and/or C02AC02 (guanfacine).

Table 4 .
48sk of ADHD in Parents, Siblings, and Offspring of Individuals With Pediatric Ischemic Stroke rare event, and this factor resulted in low power in most of the previous studies.Another major strength of the present study is its nationwide design.As recommended for follow-up studies,48we matched controls and compared risks of ADHD between patients with stroke and controls.All data were collected prospectively.We adjusted for important confounders, such as ADHD in first-degree relatives, parental age, prematurity, and SGA.A diagnosis of ADHD was identified based on clinical ICD-9 and ICD-10 codes in contrast to previous studies using diagnoses based on questionnaires or research tests, a Adjusted for maternal and paternal age at the birth of the child.JAMA Network Open.2022;5(4):e228884.doi:10.1001/jamanetworkopen.2022.8884(Reprinted) April 26, 2022 8/12 Downloaded From: https://jamanetwork.com/ by a Linkopings Universitet User on 06/10/2022