Traumatic brain injury (TBI) is the most common cause of neurological morbidity, occurring more frequently in childhood and adolescence than at any other time of life.[1, 2] By the age of 16, as many as one in five children will experience a mild traumatic brain injury (mTBI), and in the USA, an estimated 799 out of 100 000 children under 14 years visit the Emergency Department with mTBI every year.[1, 2] The current literature suggests that significant morbidity is associated with mTBI in children due to postconcussion syndrome (PCS).[3-5] Persistent PCS symptoms (>3mo) occur in 11% of children and include physical (i.e. headaches, nausea, dizziness, double vision), cognitive (i.e. difficulties with attention, concentration, thinking speed, and memory), and mood (i.e. irritability, sadness, nervousness, sleep difficulties) changes.[4-6]
Although the complex pathophysiology of mTBI is well described,[7, 8] there is considerably less pathophysiological evidence to explain prolonged PCS. Preinjury factors including substance abuse, psychiatric problems, and life stresses have been associated with persistent PCS.[3, 4, 9-11] The lack of biological explanations for PCS together with the non-specific nature of PCS symptoms has led many researchers to question its validity, and to highlight the potential for misdiagnosis, especially where depression is present.
Depression is common after TBI. One study found that 35% of adults were depressed after mTBI. When present, depression is associated with more severe PCS symptoms and with poorer outcomes. Similarly, children also have an increased prevalence of psychiatric disorders after TBI, with attention deficit and depressive disorders occurring most frequently.[2, 14-18] A family history of psychiatric problems is a risk factor for depression after pediatric TBI, although specific genetic determinants have not been identified.
Further, previous negative childhood experiences are a risk factor for a major depressive disorder after exposure to a stressful event such as mTBI.
A disturbance in the serotonin neurotransmitter system is a potential neuropathological substrate involved in depressive symptoms. Serotonin (5-HT) is a neurotransmitter associated with the regulation of mood, emotion, and stress. The frontal lobes are a major area of projection for the serotonergic system and injury to the left anterior frontal lobe is associated with depression after TBI. A single nucleotide polymorphism (rs6295; NC_000005.9:g.63258565C>G) in the promoter region of the serotonin HTR1A gene results in two different alleles, C(-1019) and G(-1019). The G-allele has been associated with major depression and suicide.
We sought to investigate the relationship between a specific genetic (HTR1A G(-1019) allele) factor, depression, childhood life stressors, and PCS after pediatric mTBI. We hypothesized that as TBI is associated with an increased incidence of depression, and that depression may be misdiagnosed as PCS, that children who have PCS after sustaining a mild TBI are more likely to carry the HTR1A G(-1019) allele, and to report more preinjury life stressors.
- Top of page
Participants in this study included 47 symptomatic children (32 males, 15 females; mean age 14y [SD 3y 3mo]) who experienced post-concussive symptoms for 7 or more days and 42 asymptomatic children (26 males 16 females; mean age 13y 6mo [SD 3y 1mo]) after mTBI. Age, Ethnicity, and other clinical details of the participants is shown in Table 1. Five of 42 of children in the symptomatic group and none of the asymptomatic group were hospitalized (p<0.001). The symptomatic group experienced PCS symptoms for 110.7 (SD 85.6) days. The asymptomatic group had no increase in symptoms compared with baseline when telephoned at 11.2 (SD 4.2) days post injury. Outcome measures were collected at 1.72 (SD 0.69) years post injury, with similar timing between groups (p=0.19).
Table 1. Demographic characteristics and survey results of asymptomatic and symptomatic groups
| ||Asymptomatic n=42||Symptomatic n=47|| p |
|Mean age in years (SD)||13.6 (3.1)||14.0 (3.3)||0.26|
|Caucasian ethnicity (%)||40 (95)||46 (98)||0.60|
|Time in years since injury (SD)||1.62 (0.60)||1.83 (0.77)||0.19|
|Initial PCSI after mTBI (SD)||10.52 (8.5)||21.63 (15.2)||<0.001a|
|Current PCSI (SD)||8.9 (7.3)||20.9 (15.01)||<0.001a|
|Children's Depression Inventoryb (SD)|| || ||0.520|
|Life events score (range)||1 (0–6)||3 (0–7)||0.004a|
Children in the symptomatic group had higher PCSI scores (mean 20.9, SD 15.01) compared with the asymptomatic group (mean 8.9, SD 7.3, t(76)=4.56, p<0.001). The most common persistent symptoms reported included ‘irritability’ (18 out of 42), ‘acting more emotional’ (18 out of 42), and ‘headache’ (18 out of 42). The CDI instrument is validated for ages 7 to 17 years, thus was only administered to participants within this age range. The mean total CDI score (children aged 7–17y) for the symptomatic cohort (n=37) was 7.16 (SD 6.83) and 5.92 (SD 7.64) for the asymptomatic group (n=24). The CDI scores in the symptomatic and asymptomatic groups did not differ (t(45)=0.648, p=0.520). One participant in each group, 3% overall, had clinically significant depression symptoms (a score of ≥19). One participant in the symptomatic group had been diagnosed with depression before her mTBI.
The symptomatic group reported more stressful life events (median score 3, range 0–7) compared with the asymptomatic group (median score 1, range 0–6). Multiple logistic regression analysis examining age and stressful life events score showed a significant difference between groups (χ2 (1)=9.71, p=0.002). After controlling for the effects of age, the SLEQ score was a significant predictor of being symptomatic after mTBI (Wald (1)=8.51, p=0.004).
Of the 89 DNA collection kits returned (two DNA samples were not returned with surveys because of parental reservations regarding DNA collection: one symptomatic and one asymptomatic), 90.8% of the samples were successfully analyzed. Table 2 shows the relative frequencies of the alleles and genotypes for both groups. The asymptomatic (n=38) and symptomatic (n=41) groups did not differ significantly in allelic frequency from each other), or from known frequencies of a reported control population from Ontario.
Table 2. Frequency of the HTR1A C(-1019) and G(-1019) genotypes and alleles
- Top of page
We report a low prevalence of depression (3%) in children after mTBI. Our results are in contrast to previous studies that report an increased prevalence of depressive disorders in children after TBI, ranging from 11 to 36%.[16-18] Most of the previous studies, however, had small sample sizes,[16-18] and they report depression in children with a wide range of TBI severity, including subsets of hospitalized mTBI patients. Although transient depressive symptoms in the first 6 months after injury could have been missed, the low prevalence in mTBI we report is in keeping with earlier observations that mood disorders increase in prevalence with increasing severity of injury.
There was no relationship between the HTR1A C(-1019)G allele and the development of PCS after mTBI in children. The genotype and allele frequencies in our sample are consistent with those seen in a population from Ontario (Table 2), and therefore are representative of predominantly Caucasian English-speaking individuals in Canada. Although there are no previous studies examining the HTR1A (C-1019)G allelic polymorphisms in TBI, this result is in keeping with recent genetic studies examining the role of serotonin transporter gene-linked polymorphisms (5-HTTLPR) in depression after TBI.
Previous life stressors do appear to be associated with PCS symptoms after mTBI. We found that children who were symptomatic after mTBI had an increased number of stressful life events before their injury compared with those children who recovered quickly after mTBI. Our findings are consistent with previous literature linking premorbid stressors and low socio-economic status to persistent PCS symptoms and cognitive and behavioral problems after mTBI in children. One explanation for this could be that stress from the injury may summate with the premorbid stress and overwhelm psychological coping mechanisms, resulting in increased perception of PCS symptoms. Alternatively, we hypothesize that significant preinjury life stresses at a young age can precondition or sensitize a child to become susceptible to the development of persistent PCS symptoms after a TBI. Post's behavior sensitization model of mood disorders suggests repeated stress leads to underlying neuropathological changes in the brain; that is, premorbid stress may prime the brain by creating a neurobiological substrate for the subsequent development of PCS symptoms after the trigger of an mTBI. Indeed the relative absence of PCS symptoms in orthopedic control groups suggests that the brain injury itself is vital to the development of symptoms.[3, 6] Although our finding needs to be replicated in other studies with larger sample sizes, identifying preinjury life stressors as a potential risk factor for the development of PCS has important implications for future management of mTBI. Early interventions including stress counseling may benefit patients with mTBI who have a history of significant premorbid stress burden and early PCS symptoms.
This study has several limitations, including a small sample size. Given that there are no previous studies examining the HTR1A C(-1019)G allelic polymorphisms in TBI on which to base the sample size estimation, the results should be interpreted somewhat cautiously. Although the CDI has been shown to discriminate between controls and depressed children, and is able to assess symptom severity, a clinical interview is needed to diagnose depression. The prevalence of the depression may have been underestimated, as a clinical interview was not performed at the time of the survey. The SLEQ represents another limitation. Self-report checklists are the most widely used method for assessing life stressors; however, currently there are no standardized measures of stressors in children. The SLEQ used in this study has not been validated although it is similar to other such questionnaires, for example LES-C, and has good face validity. Developing our own questionnaire enabled us to include events occurring over the lifetime of the child, that is, it allowed for summation of events and did not assume a linear relationship between severity or number of events (potentially inflating the effect of stressful events) and probability of depression. The stressful life event survey does not assess the temporal impact of events. For example, does a grandparent's death have the same impact on an infant as on an older child? Stressor checklists cannot distinguish between independent stressors versus those dependent on subsequent behavior and symptom expression; prospective studies are needed to investigate the relationship between life stressors and development of PCS symptoms over time. Retrospective self-reporting of the SLEQ does introduce the possibility of attribution bias; however, this risk was minimized by the inclusion of predominantly external objective SLEQ items reflecting environmental conditions rather than subjective assessment of the stressors. Both the CDI and SLEQ were co-administered 1 to 3 years post injury. No evidence of reporting bias was found in the CDI results, which were not significantly different in patients with persistent PCS compared with those without.
In summary, although the results of this pilot study must be interpreted cautiously given the sample size, they are suggestive that depressive symptoms and HTR1A polymorphisms may not play a role in the expression and persistence of PCS symptoms after mTBI in children. This is in contrast to preinjury life stressors which have been associated with the development of PCS. Larger prospective studies are needed to confirm and validate these preliminary findings. It is likely that prolonged PCS may be the result of a combination of preinjury factors, i.e. stressful life events and postinjury neuropathological and psycho-social factors, although specific genetic determinants have not yet been identified.[3, 4, 11] Screening pediatric patients with mTBI for a history of previous stressful life events may help to identify patients at risk of developing PCS.