Abstract
PURPOSE:
Preschool-aged children have a high rate of traumatic brain injuries (TBI). Anecdotally, caregivers of young children describe symptoms not typically associated with TBI. The objective of this study was to preliminarily describe symptoms reported following TBI in an outpatient sample of preschool-aged children.
METHODS:
Retrospective chart review was completed of the first specialty clinic visit by 28 children ages 3 through 5 years with TBI (32% mild, 50% moderate, 18% severe) to identify post-injury symptoms elicited by caregiver interview including querying of symptoms listed on the Acute Concussion Evaluation (ACE) plus an open-ended request to describe any other symptoms.
RESULTS:
Visits occurred on average 38 days post-injury. Caregivers endorsed typical post-concussive symptoms, with headache being most common. In addition, caregivers described other symptoms not captured by the ACE which were grouped into the following categories: Appetite Changes, Behavioral Dysregulation, Decreased Engagement, Disrupted Sleep, Enuresis, Increased Dependence, and Stomachaches.
CONCLUSIONS:
Caregivers of preschool-aged children with TBI endorse that young children experience many typical post-concussive symptoms but also a range of other symptoms which may not be typically associated with TBI. Additional work is needed to determine whether specialized evaluation tools and educational materials may be useful for this age group.
Introduction
Traumatic brain injury is prevalent in preschool-aged children and has important repercussions. The incidence of traumatic brain injury (TBI) is higher in children under the age of five years than in any other age group [1]. According to the Center for Disease Control, in 2009–2010, in the United States, the number of TBI-related emergency department (ED) visits in children 4 years of age or younger was 2193.8 per 100,000 [2]. Several studies have documented that preschool-aged children with moderate-severe TBI often show adverse cognitive-behavioral sequelae years after the injury [3, 4, 5, 6, 7]. Even mild brain injury in this population has been associated with weaknesses in cognitive and emotional functioning within the first year post-injury [8, 9, 10] and later [11, 12], highlighting the importance of identification and treatment of TBI in young children.
Given the importance of TBI in preschool-aged children, it is critical that TBI is accurately diagnosed in this age group. For children, parental report of post-injury symptoms plays an important role in diagnosis and management of TBI [13]. For older children with TBI, parents and providers typically reference symptoms listed on post-concussive symptom inventories, given that symptoms commonly associated with concussion, or mild TBI (mTBI), are applicable to children with more severe injuries as well [14, 15]. However, the parent-report pediatric TBI symptom inventories have only been validated with parents of children as young as 5 or 6 years old [16, 17], and applicability of common post-TBI symptom measures to preschool-aged children has not been fully explored.
While parent-report post-concussion symptom chec- klists have not been formally validated in preschool-aged children, some investigators have applied them to this age group, providing some preliminary information on how they perform. McKinlay and colleagues conducted a comparison of parent report of symptoms following mTBI in preschool versus school-aged children as assessed by a mailed version of the Rivermead Post-Concussion Symptom Questionnaire (RPCSQ) [18]. They found that parents of preschool-aged children reported fewer symptoms than parents of school-aged children. Using the Post Concussion Symptoms Inventory (PCSI) by telephone interview, Barlow et al. found that children under age 6 years with mTBI had faster resolution of post-concussive symptoms than older children, though in their study the younger children also had less severe injuries [19]. While it is possible that younger children have fewer symptoms and symptom duration is shorter after mTBI compared to older children, poor sensitivity of the current symptom inventories to the experience of younger children with mTBI could also underlie the findings reported by both McKinlay and Barlow.
The concern that standard pediatric post-concussion symptom inventories are not adequately sensitive for preschool-aged children was identified by Barlow et al. [19]. They identified that parents of younger children often report symptoms post-injury which are not typically thought of as TBI-associated symptoms and are not captured by the current standardized assessments. Barlow et al. specifically listed “change in feeding” and “being fussy” as symptoms identified by parents of preschool-aged children with mTBI; it is a gap in the literature that there is no further qualitative description of symptoms reported by caregivers of young children with TBI.
The purpose of the present study was to begin to address this gap by providing further qualitative description of symptoms experienced following TBI in preschool-aged children based not only on use of a standard post-concussive symptom inventory but also as reported by caregivers in response to open-ended questions, in order to capture symptoms that are not identified by current standardized assessments. We report post-injury symptoms manifested by preschool children with brain injury who received care in an outpatient brain injury specialty clinic based on interview with the child’s caregiver. We hypothesized that, in addition to endorsement of symptoms contained in a post-concussive symptom inventory, a constellation of other post-TBI symptoms would be identified through caregiver report.
Methods
This retrospective medical record review study was approved by the local Institutional Review Board.
Participants
Data from the first clinic visit for preschool-aged children who had been
evaluated in one of two specialty brain injury outpatient clinics at a large, urban,
rehabilitation center between February 2006 and May 2013 were included. Inclusion criteria
were: (1) age at the time of injury at least 3 years but less than 6 years and (2)
evidence of non-penetrating TBI. Exclusionary criteria included (1) pre-existing condition
that greatly impacted cognitive functioning (e.g., intellectual disability, Autism
Spectrum Disorder), (2) co-existing significant neurologic pathology (e.g. epilepsy, brain
tumor), and (3) admission to an inpatient rehabilitation unit, given that the need for
inpatient rehabilitation admission signifies ongoing impairment in cognitive,
communication, and/or motor skills at a level that may interfere with symptom
reporting/identification. Thirty-two children were identified who met the inclusion
criteria; three were excluded due to a pre- existing condition affecting cognitive
function (Autism Spectrum Disorder, mitochondrial disorder, and global developmental
delay), and one child was excluded due to suspected untreated epilepsy contributing to the
TBI and possibly to other reported symptoms. Twenty-eight participants (50% male) were
included in the analyses. They ranged in age from 3 years 0 month through 5 years
10 months at the time of injury (mean
Severity of brain injury was classified as mild, moderate, or severe. The
following criteria were used for classification as mTBI: Glasgow Coma Scale (GCS) score of
13–15 upon arrival to an emergency department (ED), no intracranial imaging findings, and
no depressed skull fracture. Moderate TBI was defined by GCS score between 9 and 12 in an
ED or GCS score of 13–15 with intracranial imaging findings or depressed skull fracture.
Severe TBI was defined as a GCS score of 8 and below in the ED. When documentation of GCS
was not available (
The most common mechanism of TBI was a fall. Following injury, all but
four children were briefly hospitalized. For those hospitalized, length of stay ranged
from 1 to 7 days (mean
Injury characteristics
Injury characteristics
Caregiver interview indicated that eight children (29%) had pre-injury behavioral problems such as difficulty with paying attention, regulating behavior, or following directions. Prior to the injury for which data were extracted, two boys who were otherwise healthy and typically developing had been diagnosed with a previous TBI secondary to a fall. One of these boys’ previous injury occurred several months prior to the current injury; the previous injury was diagnosed as a mTBI, and no changes were reported in his functioning after the first injury. The other of the two boys’ previous injury occurred approximately two weeks prior to the injury for which study data were collected. Per emergency department records related to the second injury, initial GCS score for the first injury (for which the child received care in a different hospital system) was 12, and he was admitted for observation for three days. Results of CT were unavailable. Per parent report he exhibited poor balance and behavioral difficulties initially after the first injury with near complete resolution of these symptoms at the time of the current injury.
Caregivers accompanying the child to the outpatient appointment and completing the interviews were all adults living in the home with the child who knew the child well. The caregiver interviewed was the mother for 21 children (75%), the father for 2 children (7%), both mother and father for 2 children (7%), and the grandmother for 3 children (11%, in two cases the grandmother was the legal guardian and in the third case the grandmother lived in the home with the child and the child’s parents). At the time of the evaluation, 71% of the participants were enrolled in a daycare or preschool setting.
Data were extracted from the brain injury clinic notes as well as any other available medical records and included basic demographic variables, medical history, injury related variables, and symptom report.
Child characteristics
At the time of the clinical evaluation, caregiver interview and review of relevant medical records were completed to obtain past medical history and demographic information including child’s age, sex, exposure to a daycare or school setting, and maternal education level. Report of pre-injury behavioral problems was obtained by querying caregivers about emotional or behavioral concerns (e.g., attentional difficulties, behavior dysregulation, separation anxiety) prior to the injury.
Post-injury medical course
Information regarding medical course was based on caregiver report (e.g., loss of consciousness following injury) as well as review of available medical records (e.g., GCS scores, interventions at scene, surgical interventions, and length of hospitalization). Radiology reports from head computerized tomography (Head CT) were reviewed, and findings were classified as normal, non-depressed skull fracture only, or intracranial findings (e.g., intraparenchymal hemorrhage, extra-axial hematomas), with or without skull fracture.
Symptom report
Symptom report was extracted from medical records documenting
information obtained from the caregiver during the clinical interview as described
below. A symptom was recorded as present if it was endorsed by the caregiver during
clinical interview or otherwise documented in the patient’s medical record. Caregivers
were instructed to only endorse symptoms that developed or were exacerbated post-injury.
Symptoms were further classified as
The clinical interview conducted with caregivers during the child’s visit had two parts. One portion of the interview was semi-structured and based on the Acute Concussion Evaluation [20]. The use of ACE elicited reports regarding the physical, cognitive, emotional, and sleep symptoms commonly experienced by older children. The ACE has been described and validated for use with parents of children aged three years and above [20], though the majority of the children in the validation sample were aged 13 years and above, with a total of 24 children aged 3 through 7 years included. During this part of the clinical interview, the providers listed each symptom from the ACE and asked the caregiver whether or not they felt the child had exhibited that symptom in increased frequency or intensity since the injury, compared to the child’s own pre-injury baseline. An author (S.R.) coded the presence or absence of these common symptoms as documented in the medical records.
Percentage of participants who displayed symptoms on the Acute Concussion
Evaluation at any time after injury (symptoms ever ) and during
the week prior to the evaluation (symptoms current )
Percentage of participants who displayed symptoms on the Acute Concussion
Evaluation at any time after injury (
Operational definitions of other symptoms
In addition to symptoms described on the ACE, in an open-ended fashion, caregivers were asked to describe any other new symptoms since injury. For example, during the clinical interview, after questioning about each of the ACE symptoms, caregivers were asked, “Have you noticed any other changes in the child since the injury?” These other symptoms were initially extracted from the medical records by S.R. Among the authors, consensus for definition and coding of these “other” symptoms was a multi-step process with the pre-set goal of achieving 75% agreement (3 of 4 authors) in coding each symptom based on chart review by each of the four raters. One feature of rater agreement was distinguishing the other symptoms from ACE symptoms such that coding of other symptoms did not overlap with any ACE symptoms. Initial operational definitions for the other symptoms, including the identification of a succinct term to describe sets of similar behaviors, were developed by consensus of all authors. Each rater then independently coded the presence or absence of the defined other symptoms for the same 5 children. As 75% agreement was not achieved at this step, the discrepancies were discussed, and the operational definitions were modified by consensus. Based on the updated definitions, the 4 raters each scored symptoms for another 5 patients; the goal for agreement was achieved at this point. Presence or absence of the other symptoms in the remainder of the sample was subsequently coded by S.R.
The data analysis was primarily descriptive in nature, including calculation of means, standard deviations, and percentages. All data analysis was conducted with Microsoft Excel 2008 for Mac.
Results
Table 2 summarizes the percentage of participants who experienced one or more symptoms in each of the ACE domains. Physical, emotional, and sleep symptoms as detailed by the ACE were commonly reported to have occurred at some point after injury, whereas the ACE cognitive symptoms were less commonly endorsed. Report of symptoms in all domains was less frequent when considering only the week preceding the clinic visit; however, the majority of caregivers reported at least one current emotional or physical symptom. Cognitive symptoms remained the least frequently reported symptoms.
Frequency of report of individual symptoms within the physical, cognitive, emotional, and sleep ACE domains are also displayed in Table 2. Within the physical domain, the most commonly reported symptom ever observed was headache, followed by fatigue and vomiting. Within the week prior to evaluation, headache remained the most commonly reported symptom and was reported for approximately one third of the children. The two children with repeat TBI comprised 40% of the children for whom balance concerns persisted at the time of evaluation. Within the emotional domain, irritability was most commonly reported ever after injury and within the last week prior to evaluation. Sleeping more than usual was the most commonly reported symptom within the sleep domain ever after injury; however, by the time of evaluation sleeping less than usual was more common. While few symptoms were endorsed within the cognitive domain, difficulty remembering was reported by 18% of the caregivers at some point after injury.
In addition to the ACE physical, cognitive, emotional, and sleep symptoms, 82% of caregivers reported other symptoms. The descriptions of these other symptoms are presented in Table 3 along with the percentage of children who displayed each symptom. The other symptoms which were most frequently reported as occurring ever after injury were decreased engagement, behavioral dysregulation, and increased dependence. In the week prior to the evaluation, behavioral dysregulation and increased dependence remained common (reported for 25% and 18% of children respectively); however, decreased engagement was not reported for any children.
Discussion
This study of children 3 to 5 years old presenting to an outpatient specialty brain injury clinic after sustaining a TBI indicates that preschool-aged children display many of the symptoms commonly exhibited by older children and adolescents across the spectrum of severity of TBI, such as headaches, fatigue, and irritability [15, 20]. In addition, as hypothesized, we found that caregivers of preschool-aged children with TBI commonly report other symptoms which are not captured on traditional TBI symptom inventories. This is consistent with the concern raised by Barlow et al. [19], and the current work advances the field by beginning to describe “atypical” symptoms reported across a small cohort of young children. It is possible that not accounting for these less traditional symptoms accounts for reports of quicker recovery/less symptom reporting in this age group [18, 19, 21] despite later adverse outcomes reported for even mild injuries in preschool-aged children [11, 12].
Many of the common post-TBI symptoms listed on the ACE form may represent abstract concepts that preschool-aged children would not be expected to understand or express, for example, “balance difficulties” or “difficulty concentrating”, such that in this age group, symptom report is dependent on caregiver observation/report. Some of the other symptoms reported for these young children may be relatively straightforward in representing a different presentation of an otherwise common post-TBI symptom. For example, stomachaches and appetite changes may be manifestations of nausea, anxiety, or more generalized malaise. Similarly, increased dependence, including behaviors such as being more clingy and requesting more help, may signify that a child is anxious or doesn’t feel well but may be unable to verbalize that information.
Of symptoms not captured by the ACE, behavioral concerns were most common. Atypical behaviors fell on both sides of a spectrum, from decreased engagement with being quieter than usual to behavioral dysregulation with temper tantrums, non-compliance, and physically aggressive behaviors. A pattern commonly observed was that decreased engagement was reported earlier post-injury; no caregivers reported that decreased engagement persisted in the week prior to evaluation. Following the phase of decreased engagement, behavior dysregulation often emerged and persisted, being among the mostly commonly reported symptoms still present at the time of the evaluation, on average one month post-injury. Given the association of TBI in this age group with later difficulties with cognitive-behavioral regulation [6, 22, 23], behavioral dysregulation may be particularly important to monitor closely over time. As with some of the physical symptoms reported, these behavioral symptoms too may overlap with more traditionally described symptoms, with decreased engagement perhaps representing fatigue or feeling slowed down and behavioral dysregulation including aspects of irritability and increased emotionality. Alternatively, these behavioral symptoms may be important markers of injury severity, with decreased engagement reflecting presentation of the decreased initiation observed early after more severe injury in older children and behavioral dysregulation heralding persisting difficulties with cognitive and behavioral control [24]. Another possibility is that these symptoms, as reported by the child’s caregiver, are a marker of changes in parent-child interactions following a child’s injury. As the quality of parent-child interactions has been proposed as a contributor to outcomes after injury [7, 25, 26], this too represents an important piece of the clinical picture. Further work is needed to understand the implication of these symptoms for later outcomes.
Some symptoms, such as enuresis representing a regression in toilet-training, may have multiple contributions. While difficulty concentrating was not strongly endorsed for this sample, perhaps due to limited expectations for the same in this age group, enuresis may occur because a child attends less or responds less quickly to their bodily signals regarding need to void. Alternatively, a diffuse brain injury, as occurs in TBI, may temporarily disrupt the relatively newly learned coordination required for continence.
Many of the typical post-TBI symptoms were reported with similar frequency to that described for preschoolers with mTBI by McKinlay et al. [18], despite the inclusion of children with more severe injuries in the current sample. The symptoms reported for the two children with repeat TBI generally appeared similar to those of the children with a single injury, though both children were among the 18% of children for whom balance remained a concern at the time of the clinic visit. There were some symptoms which are commonly reported by older children with TBI that were identified with low frequency in this sample. Physical symptoms that were infrequently reported included dizziness, sensitivity to light and noise, and numbness/tingling. The experience of these symptoms is likely hard for a young child to describe, and a caregiver may be more apt to notice generalized irritability rather than identifying a link between behavior and specific environmental stimuli. Additionally, cognitive symptoms, specifically ‘feeling mentally foggy’ and ‘difficulty concentrating’, were never/rarely reported, respectively, again likely due to these being concepts that a young child is unlikely to express and due to lack of demands/expectation for sustained attention in this age group.
While the descriptions of other symptoms reported by caregivers of preschool-aged children with TBI may reflect different expression of well-recognized post-TBI phenomena, it remains critical to raise awareness that young children and their caregivers may report physical symptoms and changes in behavior which clinicians and researchers may not readily associate with TBI. For example, in the absence of observed trauma, a mTBI may not be suspected in a child presenting with behavioral withdrawal and enuresis. Even in the setting of a known trauma, these symptoms may be attributed to other etiologies, such as a change in the child’s routine, rather than TBI. Failure to recognize a mTBI would result in not modifying a child’s activities as needed to manage symptoms and prevent repeat injury during recovery and thus may result in additional morbidity. Alternatively, it is possible that some or all of the identified symptoms represent a child’s response to trauma more globally, rather than being specific to TBI, although other recent work focused on sleep symptoms showed that sleep problems were more common in young children after TBI compared to children with orthopedic injuries [27]. Clarifying what symptoms in this age group are related to the experience of trauma but not TBI in particular will be a critical next step in this line of work. Future work including a non-TBI injury control group is needed to identify the specificity of these symptoms in order to optimize evaluation and management of TBI in this age group. Future work should also consider inclusion of a non-injury control group to determine the baseline frequency of these symptoms in this age group.
In addition to the primary limitation of the lack of a control group, the retrospective nature of this work limits the ability to provide a detailed timeline for presence and resolution of the symptoms described. Furthermore, the small, clinical, convenience sample did not allow exploration of the impact of child (e.g. age, pre-injury diagnoses/behaviors), family (e.g. caregiver demographics, caregiver/child interactions, litigation), and injury (e.g. severity) factors on symptom reporting. Symptom report was not obtained from other caregivers, such as daycare providers, though doing this in future work may help discern the possible role of change in parent-child interactions in parent-report of child’s symptoms.
Conclusion
Preschool-aged children with TBI exhibit many of the commonly described post-concussive symptoms. In addition, these children appear to show behavioral and physical symptoms which may not be readily associated with TBI. The current findings highlight the need for additional attention to the presentation of TBI in preschool aged children. Further work is needed to evaluate symptoms of TBI in this age group, including the specificity of symptoms for TBI in comparison to other forms of trauma, in order to optimize detection and management of TBI in young children.
Footnotes
Acknowledgments
Data analysis and manuscript preparation were supported in part by NIH grant K23HD061611.
Conflict of interest
The authors report no conflicts of interest.