COMT Val 158 Met and Cognitive and Functional Outcomes after Traumatic Brain Injury

Abstract

There is significant variability in long-term outcomes after traumatic brain injury (TBI), making accurate prognosis difficult. In seeking to enhance understanding of outcomes, this study aimed to investigate whether COMT Val¹⁵⁸Met allele status was associated with performance on neuropsychological measures of attention and working memory, executive functioning, learning and memory, and speed of information processing in the early rehabilitation phase. The study also aimed to examine whether the COMT polymorphism was associated with longer-term functional outcomes. A total of 223 participants (71.3% male) with moderate-to-severe TBI were recruited as rehabilitation inpatients to participate in a prospective, longitudinal head injury outcome study. The three COMT genotype groups (Val/Val, Val/Met, and Met/Met) were well matched for estimated full-scale IQ, years of education, age at injury, and injury severity. Results showed no significant difference between genotypes on neuropsychological measures (all p>0.05) or functional outcome, as measured by the Glasgow Outcome Scale-Extended (GOS-E), after controlling for age, education, and severity of injury. The presence of frontal lobe pathology was also not associated with cognitive performance. Those with greater injury severity (i.e., longer duration of post-traumatic amnesia) performed more poorly on measures of processing speed and verbal new learning and recall. It was concluded that there was little support for the influence of COMT Val¹⁵⁸Met on cognitive function, or functional outcome measures, in the acute rehabilitation phase after TBI.

Introduction

Despite recent advances in traumatic brain injury (TBI) rehabilitation, it is estimated that half of those who sustain severe brain injury will still have unfavorable outcomes.¹ Cognitive impairments are observed as one of the most debilitating consequences of TBI, with difficulties persisting many years after injury. These deficits have been found to predict long-term functional outcomes.²

Despite characteristic patterns of impairment, there remains substantial variability in outcomes across individuals, even within injury severity categories after TBI. This makes accurate clinical prognosis difficult in individual cases.³ To date, the most reliable and consistent outcome predictors are age at injury, and injury severity, as measured by Glasgow Coma Scale (GCS) scores or duration of post-traumatic amnesia (PTA).⁴ Recent studies suggest that PTA is a more reliable predictor of long-term outcome than GCS.⁵ Extended PTA duration has been found to predict persistent cognitive deficits, poorer vocational outcomes, and difficulties returning to study.³ Given the high prevalence of TBI, and relatively young age of those injured, there is a need for greater understanding of other predictors of long-term outcomes.

Recent research has examined individual genetic predispositions to assist in explaining variance in outcomes. Polymorphisms in genes modulating neurotransmitter systems may influence the manifestation of injury related cognitive impairments.⁶ One candidate gene is Catechol-O-Methyltransferase (COMT). The COMT gene is located on the long arm of chromosome 22 (band 22q11.2) and codes for the production of the enzyme, catechol-O-methyltransferase.⁷ The role of this enzyme is the metabolic degradation of the catecholamines, dopamine (DA) and norepinephrine (NE), achieved through 3-O-methylation of the benzene ring.⁸ In light of the scarcity of DA transporters, COMT acts as a significant determinant of prefrontal DA flux, accounting for more than 60% of DA degradation.⁹

Research has focused on a common single-nucleotide polymorphism (SNP) within the COMT gene, rs4680 (Val¹⁵⁸Met), resulting from a valine to methionine substitution at codon 158.¹⁰ Thus, individuals homozygous for the Val form metabolize catecholamines more rapidly than those with the Met form, resulting in a lower bioavailability of DA within the prefrontal cortex (PFC). As a result of the codominance of alleles, heterozygotes are thought to have an intermediate level of activity, accounting for the observed trimodal distribution of COMT.¹¹ However, there is some evidence to suggest that the Val¹⁵⁸ form is a predominant factor that determines the higher COMT activity within the PFC. Thus, the Val allele exerts a degree of dominance, determining level of enzyme activity.¹²

Catecholamines play a critical role in normal and pathological cognitive function, with optimal levels of neurotransmitters necessary for efficient performance.¹³ Evidence for this can be found in pharmacological studies, with depletion of local DA and NE found to disrupt attention, ¹⁴ memory and learning,¹⁵ and executive functioning (EF),¹⁶ with disruptions, in some instances, similar to those observed in the ablation of neural tissues.¹⁷ Further, experimentally induced cognitive constraints have been reversed using DA and NE agonists.^16,17

Altered levels of catecholamines have been well documented after TBI. Studies using cortical impact injuries in rats have reported increased levels of tyrosine hydroxlase (TH), the enzyme involved in synthesis of catecholamines, in the frontal cortex at 14¹⁸ and 28 days postinjury and a decrease in DA transporter protein.^19,20 More recently, enhanced microglia COMT expression has been identified in the hippocampus at 14 days postinjury.²¹

In clinical studies, reduced levels of homovanillic acid (HVA; a DA metabolite) have been found in lumbar cerebrospinal fluid after severe TBI (sTBI).^22,23 Reduced HVA levels have been found to correlate with duration of unconsciousness.²³ Other human studies have focused on plasma neurotransmitter levels and activity of the sympathetic nervous system after TBI. Within the first week postinjury, normal plasma levels of NE, epinephrine, and DA have been found after mild-to-moderate TBI.^24,25 In contrast, elevated plasma levels of these neurotransmitters have been identified after sTBI,²⁶ and these levels correlated with GCS in many cases.^24,27 Higher NE levels were associated with poorer outcomes, as measured on the GCS²⁵ and the Glasgow Outcome Scale (GOS).²⁷ Recovery has been associated with normalization of catecholamine activation.²⁸ Other studies have, however, demonstrated more persistent dysregulation over time.²⁹ Even in mild TBI, pharmacological challenges with agents such as the DA D₂ receptor agonist, bromocriptine, have demonstrated altered activation in working memory (WM) tasks, implicating changes in catecholaminergic systems in WM performance after injury.³⁰ Functional imaging has also been used to demonstrate deficits in neurotransmission after moderate-to-severe TBI.³¹ Such fluctuations in neurotransmission may contribute to the common deficits observed in PFC-mediated cognitive functions, including impairments in attention,^32,33 EF,^34,35 speed of processing,^36,37 and memory³⁸ after TBI.

The higher bioavailability of catecholamines in the PFC among Met homozygotes has been shown to confer a cognitive advantage over that of Val homo- or heterozygotes in line with the inverted U relationship.¹¹ Neuroimaging studies have found increased blood flow to the PFC in Val allele carriers, compared to Met allele carriers, during the performance of cognitive tasks.³⁹ This increase in blood flow may indicate inefficient neural responses or neural compensatory responses in individuals with the Val allele in order to maintain performance at equal levels to that of Met carriers.⁴⁰

Numerous studies have confirmed the association of COMT with cognition. Healthy individuals with the Val form demonstrated poorer performance on measures of WM⁴¹ and EF.^42,43 COMT has also been shown to influence cognition within attention deficit hyperactivity disorder populations, with Val allele carriers performing more poorly than Met allele carriers on tasks of WM⁴² and problem-focused activities.⁴⁴ To date, no studies have examined the influence of COMT Val¹⁵⁸Met on functional outcomes.

Despite this evidence, few studies have investigated the effects of COMT allele status on cognitive performance after TBI. One study by Lipsky and colleagues⁴⁵ examined the effect of COMT Val¹⁵⁸Met on EF among 113 individuals with moderate-to-severe TBI. Investigations showed that Val homozygotes exhibited significantly more perseverative responses, indicative of cognitive inflexibility, compared to Met homo- and heterozygotes on the Wisconsin Card Sorting Test (WCST). It was concluded that the COMT polymorphism could influence EF in those with brain injury. Though the authors ensured genotype groups were well matched with regard to presence of frontal lobe pathology, the possible relationship between frontal lobe pathology and COMT genotype was not explored.

Similarly, Flashman and colleagues⁴⁶ examined the effect of COMT gene status on EF in 39 mild-to-moderate TBI patients and 27 demographically matched healthy controls. The TBI group performed significantly worse on several measures of EF, when compared to healthy controls. Additionally, there was a significant association between COMT gene status and performance on the reaction time distractibility subtest, with Met allele holders demonstrating superior performance, when compared with Val allele holders. Together, these preliminary studies support the need for larger-scale investigation of the genetic modulators of cognitive functions, and functional outcomes, after TBI.

The aim of the current study was to investigate whether COMT Val¹⁵⁸Met allele status was associated with performance on widely used neuropsychological measures in the acute rehabilitation stage of recovery. The relationship between COMT genotype and the presence of frontal lobe pathology was also explored. The study also aimed to examine whether the COMT polymorphism was associated with longer-term functional outcomes after TBI. It was hypothesized that Met allele holders would perform significantly better than Val allele holders on measures of attention and WM, EF, learning and recall, and speed of processing as a result of the higher bioavailability of catecholamines in the PFC. It was also expected that the presence of frontal lobe pathology would confer additional cognitive disadvantage in Val homozygotes. Additionally, it was hypothesized that Met allele carrier status would manifest in better functional outcomes, as demonstrated by higher Glasgow Outcome Scale-Extended (GOS-E) scores in Met homozygotes than in participants with Val/Val and Val/Met allele status.

Methods

Participants

Participants were adolescents and adults admitted to Epworth HeathCare (Melbourne, VIC, Australia) for rehabilitation after sustaining a moderate-to-severe TBI. Injury severity was determined by a combination of lowest GCS score recorded and duration of post-traumatic amnesia (PTA) as measured by the Westmead PTA Scale.⁴⁷ Those with a GCS of 13–15 had duration of PTA of more than 24 h, and those with duration of PTA of less than 24 h had a GCS of less than 13, thereby ensuring that all participants had an overall injury severity falling within the moderate-to-severe range. Epworth provides rehabilitation to approximately 30–50% of all moderate-to-severe TBI patients within the state of Victoria, in the context of an accident compensation system administered by the Transport Accident Commission (TAC). The TAC provides funding for all hospital and rehabilitation costs and for continuing supports postinjury, regardless of fault and socioeconomic status. Participants were recruited during their inpatient admission to participate in a prospective longitudinal head injury outcome study. Participants completed a neuropsychological assessment after emerging from PTA and consented to provide a saliva sample for genetic analysis. Participants were required to have sufficient understanding of English and adequate physical and cognitive abilities to undertake cognitive assessment, as determined by their treating neuropsychologist, and provide written, informed consent. Exclusion criteria included any previous history of head injury or other pre- or coexisting neurological or psychiatric disturbances.

Measures

The Glasgow Coma Scale

⁴⁸The GCS was used to measure coma depth as a basis for determining injury severity. The lowest preintubation score obtained in the first 24 h post-trauma was used. GCS scores were divided into three injury severity categories: mild (13–15); moderate (9–12); and severe (3–8).

The Westmead PTA Scale

⁴⁷The Westmead PTA Scale was used as a prospective measure of PTA duration, assessing orientation, and the ability to lay down new memories at 24-h intervals until a score of 12/12 is obtained for three consecutive days.

Wechsler Adult Intelligence Scale III and IV

^49,50Subtests included Arithmetic, Block Design, Coding, Digit Span, and Matrix Reasoning. Raw scores were utilized in data analysis.

Trail Making Test

⁵¹Total time to complete Parts A and B was recorded. The Trail Making Test (TMT) was used to assess WM, speed of processing, and EF. Normative data were derived from Tombaugh.⁵²

The Rey Auditory Verbal Learning Test

⁵³Total Learning Score Trials 1–5 (Rey Auditory Verbal Learning Test [RAVLT] Total) and Trial 7 Delayed recall score (RAVLT T7 Delay) raw scores were used to examine learning and retention of verbal material.

The Rey Complex Figure Test

⁵⁴Raw score total was used to examine visuospatial recall after a 30-min delay (Rey Complex Figure Test 30-min delay).

The above cognitive measures have been shown to be sensitive to TBI.^55
–57

Wechsler Test of Adult Reading

The Wechsler Test of Adult Reading (WTAR)⁵⁸ was used to estimate premorbid IQ (full-scale IQ; FSIQ), using the WTAR-Demographics predicted tables.

The Glasgow Outcome Scale-Extended

⁵⁹The GOS-E was used to assess global functional outcomes after TBI. The GOS-E is a semistructured interview consisting of 19 questions examining injury-related changes in consciousness, independence in and out of the home, work, social, and leisure activities, family and friendships, return to normal life, and epilepsy. Scores are rated on an 8-point scale ranging from 1 (dead) to 8 (upper good recovery). Scores were coded into the following categories: 7–8=good recovery and <7=moderate-to-severe disability, on the basis of previous studies in our group.⁶⁰

Procedure

Ethics approval was obtained from both Epworth Healthcare and Monash University ethics committees, and informed consent was provided in accord with the Helsinki Declaration. Participants were recruited into the longitudinal head injury outcome study while they were inpatients at Epworth Hospital. Cognitive tests were completed by a treating neuropsychologist as a part of the rehabilitation team clinical assessment, once participants had emerged from PTA, as assessed by the Westmead PTA Scale.⁴⁷ TBI participants also provided a saliva sample for genotyping. Computed tomography (CT) brain scans from the acute hospital were acquired on a GE Lightspeed VCT scanner (64-slice; GE Healthcare, Waukesha, WI) and were coded by an independent rater for the presence or absence of frontal lobe pathology. The GOS-E was administered at follow-up clinic appointments 1 and/or 2 years postinjury.

COMT genotyping

Genomic DNA was extracted from patient saliva samples by protinease K (Promega, Madison, WI) digestion for amplification by polymerase chain reaction (PCR). The COMT VAL158MET polymorphism was tested by restriction fragment-length polymorphism analysis. A 109-base-pair (bp) fragment was PCR amplified with GoTaq Hotstart DNA Polymerase (Promega) using the forward primer, Comt1nt1881 (5′CTCATCACCATCGAGATCAA), and the reverse primer, Comt2nt1989 (5′CCAGGTCTGACAACGGGTCA). This PCR product was digested for 4 h with NlaIII followed by electrophoresis using 10% Ready Gel TBE gels (Bio-Rad, Hercules, CA). Restriction digest products were visualized by ethidium bromide staining. VAL/VAL homozygotes have bands of 86 and 23 bp, VAL/MET heterozygotes have bands of 86, 68, 23, and 18 bp, and MET/MET homozygotes have bands of 68 and 18 bp. The three genotypes were distinguished by the presence or absence of the 86- and 68-bp bands.

Statistical analysis

All statistical analysis was performed using Statistical Package for the Social Sciences software (version 20.0; SPSS, Inc., Chicago, IL). Data were screened for missing values and outliers, and the assumptions of normality, linearity, and homogeneity of variance were examined.

A one-way analysis of variance was used to examine differences between genotype groups (Val/Val, Val/Met, and Met/Met) on age at injury, PTA duration, years of education, and WTAR estimated FSIQ. Chi-square tests examined differences between groups on GCS and percentage of frontal lobe pathology.

A series of 3 (Val/Val, Val/Met, Met/Met)×2 (FL pathology, no FL pathology) way analyses of covariance (ANCOVAs) was conducted for each cognitive outcome variable, controlling for age at injury (AAI), years education (Yrs Ed), and duration of PTA (PTA), with Bonferroni's corrections applied for multiple comparisons. Previous studies have also divided the groups into Met/Met versus any Val allele,⁶¹ resulting from the dominance of the Val allele over the Met allele, with comparable findings to the three-group analysis. Such analysis was undertaken in the present study with similar findings to those within the three-group comparison outlined above. For the purpose of brevity, only the three-group analysis is reported. Given the absence of a healthy control group in the present study, performance on cognitive measures for the three TBI genotype groups was compared to that of normative samples (data derived from test manuals or sources outlined under Measures, with the standard reference group for comparison set at the mean age of the current sample [36 years]). Logistic regressions were conducted for the 12- and 24-month GOS-E analyses.

Results

Participants

The study included 223 participants, of whom 71.3% were male. Participants had an average age of 36.2 years (standard deviation [SD], 16.4; range, 16–90) at the time of injury. The majority of participants (54.7%) were involved in car accidents, followed by pedestrian (16.1%) and motorcycle accidents (14.8%). Duration of PTA: <24 h, mild (n=23; 10.3%); PTA <7 days, moderate (n=24; 10.8%); PTA 7–28 days, severe (n=116; 52.0%); and PTA >28 days, very severe (n=55; 24.7%); and unknown duration (n=5; 2.2%). The distribution of GCS for participants was as follows: severe 3–8 (n=106), moderate 9–12 (n=37), mild 13–15 (n=70), and unknown (n=10). Thus, on the basis of combined GCS/PTA data, the sample had predominantly moderate-to-severe injuries. CT scan results showed that 89.7% (n=200) of participants were classified as having abnormal scans, with 30.5% (n=68) of these showing frontal lobe pathology and the remaining 65.5% (n=146) showing other forms of pathology. Participants were mainly Caucasian (n=210; 94.2%), with 12 (5.4%) of Asian (mongoloid) race and 1 (0.4%) of African (negroid) race. Given the relatively small percentage of non-Caucasian ethnicity (5.8%), these data were not analyzed separately, as would be recommended in samples with greater ethnic diversity, given findings in the literature suggesting that expression of the Val allele may manifest variable risks across diverse ethnic populations.⁶² The TBI genotype distribution was as follows: Val/Val n=63; Val/Met n=116; and Met/Met n=44. The genotype frequencies did not depart from the Hardy-Weinberg equilibrium (chi-square [χ²]=0.62; df=2). Neuropsychological assessment occurred, on average, at 29.0 (SD, 26.8) days after injury.

A series of independent measures t-tests revealed that the genotype groups (Met/Met, Val/Met, and Val/Val) did not differ significantly in terms of: age at injury, PTA duration, estimated FSIQ, or years of education (p>0.05 for all; see Table 1). Similarly, chi-square analysis showed no significant difference between groups for distribution of GCS score (χ²=27.39; df=24, p=0.29) or percentage with frontal lobe pathology evident on CT brain scan (χ²=1.50; df=2; p=0.47). The three groups were therefore well matched in terms of background demographics and injury severity.

Table 1.

Demographics and Injury Severity by COMT Allele Status

Variable	MM n (%)	VM n (%)	VV n (%)	Total n (%)	p value
GCS^a
Severe (3–7)	21 (52.5)	52 (47.3)	33 (52.4)	106 (49.5)
Moderate (8–12)	4 (10.0)	25 (22.7)	8 (12.7)	37 (17.3)
Mild (13–15)	15 (37.5)	33 (30.0)	22 (34.9)	71 (33.2)
Total	40 (100)	110 (100)	63 (100)	213 (100)	0.29
FL pathology^a
Present	16 (37.2)	36 (32.7)	16 (26.2)	68 (31.6)
Absent	27 (62.8)	74 (67.3)	45 (73.8)	147 (68.4)
Total	43 (100)	110 (100)	61 (100)	214 (100)	0.47

	M (SD)	M (SD)	M (SD)	M (SD)
AAI, years	38.9 (16.8)	35.4 (15.5)	35.7 (17.7)	36.2 (16.4)	0.47
PTA duration, days	19.3 (18.8)	20.6 (21.2)	24.1 (23.6)	21.4 (21.4)	0.46
WTAR FSIQ	100.7 (10.9)	99.0 (11.2)	100.0 (8.8)	99.6 (10.6)	0.67
Yrs Ed	12.0 (2.7)	11.6 (2.1)	11.7 (2.4)	11.7 (2.3)	0.67

Chi-square analysis.

MM, Met/Met; VM, Val/Met; VV, Val/Val; GCS, Glasgow Coma Scale; FL, frontal lobe; M, mean; SD, standard deviation; AAI, age at injury; PTA, post-traumatic amnesia; WTAR FSIQ, Wechsler Test of Adult Reading predicted premorbid IQ; Yrs Ed, years of education.

Association of genotype with cognitive and outcome measures

Means and SDs of group performance on individual cognitive outcome measures are presented in Table 2. Individual ANCOVAs revealed that, for each cognitive outcome variable, there was no significant main effect of the between-subjects factor genotype (all p>0.05) or for the presence of frontal lobe pathology (all p>0.05), nor was there any significant interaction between these two factors. The three covariates were each found to be significantly associated with a number of the dependent variables, after Bonferroni's corrections were applied. Longer PTA duration was associated with lower scores on: Block Design (F(1, 160)=10.00; p=0.002); Coding (F(1, 127)=21.57; p<0.0005); RAVLT Total (F(1,144)=14.83; p<0.0005); RAVLT T7 Delay (F(1,165)=21.95; p<0.0005); and with slower completion times on TMT-A (F(1,123)=19.77; p<0.0005) and TMT-B (F(1,123)=8.67; p=0.004). Older age at injury was associated with lower scores on: Coding (F(1, 127)=16.72; p<0.0005); RAVLT Total (F(1,144)=10.80; p=0.001); and RAVLT T7 Delay (F(1, 165)=8.67; p=0.004). Fewer years of education was associated with lower scores on Matrix Reasoning (F(1,81)=12.78; p=0.001) and Digit Span (F(1,170)=11.54; p=0.001).

Table 2.

Raw Score Means and Standard Deviations by COMT Allele Status

Cognitive measure	MM	VM	VV
Block Design	39.3 (13.3)	37.6 (12.8)	35.9 (12.9)
Coding	54.5 (17.3)	51.0 (14.9)	52.0 (16.4)
Digit Span Total	19.3 (6.5)	19.9 (6.2)	19.7 (7.8)
Arithmetic	14.1 (4.2)	12.8 (3.7)	13.4 (3.8)
Matrix Reasoning	14.6 (5.6)	16.2 (5.7)	17.6 (4.9)
TMT-A (sec)	38.7 (19.1)	42.3 (21.3)	42.3 (18.2)
TMT-B (sec)	92.4 (59.8)	99.1 (48.3)	92.9 (43.5)
RCFT 30-min Delay	16.7 (5.7)	16.6 (6.6)	15.6 (6.7)
RAVLT Total	44.3 (10.0)	46.3 (11.0)	44.3 (11.0)
RAVLT T7 Delay	7.6 (3.3)	8.2 (4.1)	7.5 (4.0)

Means (standard deviations) are shown.

MM, Met/Met; VM, Val/Met; VV, Val/Val; TMT, Trail Making Task; RAVLT Total, Rey Auditory Verbal Learning Test Total score from Learning trials 1–5; RAVLT T7 Delay, Rey Auditory Verbal Learning Test Trial 7 Delayed recall score; RCF, Rey Complex Figure test; GOS-E, Glasgow Outcome Scale-Extended.

Examination of participant scores, relative to normative samples, revealed that TBI participants scored below normative samples on the following cognitive outcome measures. Average performance across all three genotypes for the TMT-A and TMT-B fell between the 10th and 20th percentile. Similarly, mean scores on the RCFT 30-min delay fell below the 10th percentile. Performance on the RAVLT Total and T7 Delay fell 1 SD below the mean. On the Wechsler Adult Intelligence Scale subtests, all three groups were impaired, relative to age-based norms, on Coding, Digit Span Forward, and Digit Span Backward. Performance on Block Design, Matrix Reasoning, and Arithmetic was consistent with that of normative samples across the three genotypes. In summary, all three TBI groups demonstrated cognitive impairment primarily in the domains of attention and WM, speed of information processing, new learning, and recall.

Distribution of GOS-E category by genotype is outlined in Table 3. As can be seen in Table 3, outcome at 12 months postinjury fell within the moderate-to-severe disability category (i.e., a GOSE score of less than 7) for approximately two thirds of participants, and a similar pattern was evident at 2 years postinjury. Two logistic regression analyses were performed with GOS-E at 12 and 24 months as the dependent variables and Years of Education, AAI, PTA, and genotype as predictor variables. At 12 months postinjury, a total of 128 cases were analyzed and the total model was unable to discriminate good recovery from moderate-to-severe disability (χ²=6.62; df=5; p=0.250). At 24 months postinjury, a total of 119 cases were analyzed and the total model was significant (χ²=11.45; df=5; p=0.43); however, the model only accounted for between 5% and 7% of the variance in outcomes, with 81.9% of those with moderate-to-severe disability successfully predicted, but only 36.2% of predictions for good recovery being accurate. Overall, 63.9% of predictions were accurate. Only PTA duration accurately predicted GOS-E at 24 months (p=0.032).

Table 3.

GOSE Outcomes by COMT Allele Status

	MM n (%)	VM n (%)	VV n (%)	Total n (%)
GOS-E 12 months
Good recovery (7–8)	5 (11.9)	24 (57.1)	13 (31)	42 (30.2)
Moderate-to-severe disability (<7)	23 (23.7)	48 (49.5)	26 (26.8)	97 (69.8)
Total	28 (20.1)	72 (51.8)	39 (28.1)	139 (100.0)
GOS-E 24 months
Good recovery (7–8)	9 (17.3)	27 (51.9)	16 (30.8)	52 (39.4)
Moderate-to-severe disability (<7)	17 (21.2)	39 (48.8)	24 (30.3)	80 (60.6)
Total	26 (19.7)	66 (50.0)	40 (30.3)	132 (100.0)

MM, Met/Met,; VM, Val/Met,; VV, Val/Val,; GOS-E, Glasgow Outcome Scale-Extended.

Discussion

The present study sought to determine whether COMT Val¹⁵⁸Met allele status was associated with performance on a range of cognitive measures after TBI and, additionally, whether there was an association with the presence of frontal lobe pathology. The study also aimed to examine whether the COMT polymorphism was associated with functional outcomes.

Results from a series of independent measures ANCOVAs did not support the study hypotheses, with no significant main effect for group (Met/Met, Val/Met, and Val/Val), or frontal lobe pathology (present/absent), and no significant interactions between the two factors, on cognitive outcome measures after TBI. Other studies that have found a significant effect of the COMT polymorphism on cognitive functioning after TBI^45,46 differ from the current study with respect to the nature of measures employed. In these studies, significant effects were reported on measures that perhaps more selectively tap into catecholamine-mediated cognitive functions (e.g., WCST, n-back), as opposed to the multi-factorial, clinical measures that were employed in the current study. The multi-dimensional nature of the more-standard clinical neuropsychological measures used in the current study may have resulted in a loss of sensitivity for detecting subtle genetic effects. In particular, the significant finding of Lipsky and colleagues,⁴⁵ of Val homozygotes making more-perseverative responses on the WCST, was not examined in the current study. Future research should aim to include those measures that do selectively tap into areas that are impaired after TBI and draw more directly on functions controlled by the PFC.

Time since injury may also have influenced the likelihood of detecting genetic effects. Time from injury to assessment was not documented in the study by Flashman and colleagues.⁴⁶ Lipsky and colleagues⁴⁵ recruited participants within 1 year postinjury, with mean time since injury varying from 48 to 57 days. Because participants in the present study were in the relatively acute stage of recovery when cognitive testing was conducted, on average, 29 days postinjury, factors other than cognitive impairment might have influenced test performance (e.g., pain, fatigue, or medication effects). Further investigations in which participants are assessed acutely should control for these potentially confounding factors. Cognitive assessment at 6–12 months postinjury may be more likely to detect subtle group differences.

Similarly, acute fluctuations in concentrations of catecholamines, found to persist for weeks after injury,⁶³ could have acted to mask the influence of COMT Val¹⁵⁸Met on cognitive processing. The well-documented acute rise in levels of catecholamines within the PFC would act to disrupt the already delicate homeostatic balance of catecholamines, necessary for optimum cognitive processing, in line with the inverted U hypothesis. Fluctuations in these neurotransmitter concentrations could therefore have made small genetic effects difficult to detect.

An additional discrepancy between this and previous studies is that of injury severity. The current study primarily included participants who had sustained moderate-to-severe injuries, whereas Lipsky and colleagues⁴⁵ and Flashman and colleagues⁴⁶ included significant numbers of individuals with relatively mild injuries. Inclusion of individuals with more-serious injuries in the present study, with extensive and heterogeneous brain pathology, may have obscured or masked the effects of a single genetic polymorphism. In attempting to control for this, groups were well matched on measures on GCS and PTA duration, and percentage of frontal lobe pathology did not differ across the groups. This categorization, with respect to the independent rating of the CT brain scans, was somewhat rudimentary because the limitations of CT, as opposed to magnetic resonance imaging, in identifying more-diffuse pathology after TBI is acknowledged.⁶⁴ Longer PTA duration was associated with poorer performance on a number of measures, many of which are timed tasks (Block Design, Coding, TMT-A, and TMT-B). Interestingly, there was no such association on Matrix Reasoning, a task without time limitations. Length of PTA was also associated with learning and delayed recall on the RAVLT. Thus, severity of injury (as measured by duration of PTA), though controlled for across groups, was significantly associated with cognitive outcome after TBI in this moderate-to-severely injured sample. This is consistent with these participants demonstrating impaired performance on a number of tasks, relative to normative samples. However, as in the Lipsky and colleagues study,⁴⁵ the lack of a healthy control comparison group limits the interpretation of the comparison made with normative data, as well as ability to determine the direct contribution of the TBI to any observed relationship between COMT and any outcome variables in genetics studies. Future research should aim to further examine the relationship between injury severity and genetic contributions to recovery, with the inclusion of a matched healthy control group who undertake cognitive assessment and COMT genotyping.

Although cognitive testing was not undertaken beyond the inpatient rehabilitation period, functional outcome measures were taken in the postacute stage of recovery (12 and 24 months postinjury), by which stage many factors may have coalesced to influence function. Nonsignificant findings may be attributed to the multi-faceted nature of the GOS-E, which is determined, only in part, by cognitive function. Functional outcome scales are inherently complex in nature and are mediated by a range of factors, including physical disability and emotional coping. Consequently, the GOS-E may lack sensitivity to detect any influence of subtle genetic effects.

The Val¹⁵⁸Met SNP forms just one part of many functional SNPs found within the COMT gene, which is, in turn, one of the many genes that influence different aspects of cognitive function and recovery after TBI. Moreover, gene functions can be modified by other genes, proteins, and environmental factors, with modifications rendering the detection of their effect more difficult.¹ Thus, examination of just one of these alleles in isolation (candidate gene approach) does not allow for any underlying relationships between SNPs and genes to be studied and does not take into account environmental modifications of gene expression.² A number of recent reviews^6,65 and studies investigating influence of other polymorphisms on outcome after TBI⁶⁶ have advocated for the examination of the interactions between a range of candidate alleles likely to contribute to cognition. Full screening of the COMT gene and other dopaminergic genes, such as dopamine receptor D2 gene or ankyrin repeat and kinase domain-containing 1, which are known to be associated with cognition, is recommended.⁶⁵ Whether studying the influence of a single gene, interaction between a range of candidate alleles or multiple genes, or taking a genome-wide approach, it is clear that very large samples are required to confer sufficient statistical power to detect effects. This would perhaps be best facilitated by large-scale collaborative, multi-institutional studies.

Despite nonsignificant findings, the current study has addressed limitations within the existing literature. The study employed the largest sample size to date to examine the influence of COMT Val¹⁵⁸Met SNP on performance on commonly used neuropsychological measures in a TBI population. Further, to our knowledge, this study was the first to examine the influence of COMT Val¹⁵⁸Met on a functional outcome measure in TBI patients. Last, the study has controlled for the influence of injury severity by examining a range of severity-related variables.

Examining the potential influence of genetic polymorphisms represents a promising, novel approach to examining predictors of outcomes after brain injury. In order to achieve sufficient power to detect the effect of a single SNP, hundreds of participants may be required, bringing into question the feasibility of conducting such research.¹ Therefore, future research should aim to adopt a meta-analytic approach to examining the influence of the Val158Met SNP or allow for pooling of data through a multi-center collaborative approach. Further examination of genetic influences through such studies will potentially contribute to understanding variance in outcomes after TBI.

Footnotes

Acknowledgments

This project was supported by the Victorian Government Transport Accident Commission and the Institute for Safety Compensation and Recovery Research. The authors thank Drs. Katrina Simpson Gershon Spitz for their advice regarding statistical analysis as well as the participants who gave so generously of their time.

Author Disclosure Statement

No competing financial interests exist.

References

Dardiotis

, Fountas

K.N.

, Dardioti

, Xiromerisious

, Kapsalaki

, Tasiou

, and Hadjigergiou

G.M.

(2010). Genetic association studies in patients with traumatic brain injury. Neurosurg. Focus, 28, E9.

Weaver

S.M.

, Chau

, Portelli

J.N.

, and Grafman

(2012). Genetic polymorphisms influence recovery from traumatic brain injury. Neuroscientist, 18, 631–644.

Ponsford

, Draper

, and Schönberger

(2008). Functional outcome 10 years after traumatic brain injury: its relationship with demographic, injury severity and cognitive and emotional status. J. Int. Neuropsychol. Soc., 14, 233–242.

Schönberger

, Ponsford

, Reutens

, Beare

, and O'Sullivan

(2009). The relationship between age, injury severity and MRI findings following traumatic brain injury. J. Neurotrauma, 26, 2157–2167.

Brown

A.W.

, Malec

J.F.

, McClelland

R.L.

, Diehl

N.N.

, Englander

, and Cifu

D.X.

(2005). Clinical elements that predict outcome after traumatic brain injury: a prospective multicenter recursive partitioning (decision-tree) analysis. J. Neurotrauma, 22, 1040–1051.

McAllister

T.W.

(2009). Polymorphisms in genes modulating the dopamine system: do they influence outcome and response to medication after traumatic brain injury. J. Head Trauma Rehabil., 24, 65–68.

Grossman

M.H.

, Emanuel

B.S.

, and Budarf

M.L.

(1992). Chromosomal mapping of the human catechol-O-methyltransferase gene to 22q11.1-q11.2. Genomics, 12, 822–825.

Mier

, Kirsch

, and Meyer-Lindenber

(2010). Neural substrates of pleiotropic action of genetic variation of COMT: a meta-analysis. Mol. Pyschiatry, 15, 918–927.

Malhotra

A.K.

, Kestler

L.J.

, Mazzanti

, Bates

J.A.

, Goldberg

, and Goldman

(2002). A functional polymorphism in the COMT gene and performance on a test of prefrontal cognition. Am. J. Psychiatry, 159, 652–654.

10.

Grossman

M.H.

, Szumlanski

, Littrell

J.B.

, Weinstein

, and Weinshilboum

R.M.

(1992). Electrophoretic analysis of low and high activity forms of catechol-O-methyltransferase in human erythrocytes. Life Sci., 50, 473–480.

11.

Turnbridge

E.M.

, Harrison

P.J.

, and Weinberger

D.R.

(2006). Catechol-O-methyltransferase, cognition, and psychosis: Val158Met and beyond. Biol. Pyschiatry, 60, 141–151.

12.

Chen

, Lipska

B.K.

, Halim

, Ma

Q.D.

, Matsumoto

, Melhem

, Kolachana

B.S.

, Hyde

T.M.

, Hermen

M.M.

, Apud

, Egan

M.F.

, Kleinman

J.E.

, and Weinberg

R. D.

(2004). Functional analysis of genetic variation in catechol-O-methyltransferase (COMT): effects on mRNA, protein, and enzyme activity in postmortem human brain. Am. J. Hum. Genet., 75, 807–821.

13.

Robbins

T.W.

, and Arnsten

A.F.

(2009). The neuropsychopharmacology of front-executive function: monaminergic modulation. Annu. Rev. Neurosci., 32, 267–287.

14.

Arnsten

A.F.

(2011). Catecholamine influences on dorsolateral prefrontal cortical networks. Biol. Psychiatry, 69, 89–99.

15.

Bethus

, Tse

, and Morris

R.G.M.

(2010). Dopamine and memory: modulation of the persistence of memory for novel hippocampal NMDA receptor-dependent paired association. J. Neurosci., 30, 1610–1618.

16.

Newman

L. A.

, Darling

, and McGaughy

(2008). Atomoxetine reverses attentional deficits produced by noradrenergic deafferentation of medial prefrontal cortex. Psychopharmacology, 200, 39–50.

17.

Brozoski

T.J.

, Brown

R.M.

, Rosvold

H.E.

, and Goldman

P.S.

(1979). Cognitive deficit caused by regional depletion of dopamine in prefrontal cortex of rhesus monkey. Science, 205, 929–932.

18.

Kobori

, Clifton

G.L.

, and Dash

P.K.

(2007). Enhanced catecholamine synthesis in the prefrontal cortex after traumatic brain injury: implications for prefrontal dysfunction. J. Neurotrauma, 23, 1094–1102.

19.

Yan

H.Q.

, Kline

A.E.

, Ma

, Hooghe-Peters

E.L.

, Marion

D.W.

, and Dixon

C.E.

(2001). Tyrosine hydroxylase, but not dopamine beta-hydroxylase, is increased in rat frontal cortex after traumatic brain injury. Neuroreport, 12, 2323–2327.

20.

Yan

H.Q.

, Kline

A.E.

, Ma

, Li

, and Dixon

C.E.

(2002). Traumatic brain injury reduces dopamine transporter protein expression in the rat frontal cortex. Neuroreport, 13, 1899–1901.

21.

Redell

J.B.

, and Dash

P.K.

(2007). Traumatic brain injury stimulates hippocampal catechol-O-methyl transferase expression in microglia. Neurosci. Lett., 413, 36–41.

22.

Bareggi

S.R.

, Porta

, Selenati

, Assael

B.M.

, Calderini

, Collice

, Rossanda

, and Morselli

P.L.

(1975). Homovanillic acid and 5-hydroxyindole-acetic acid in the csf of patients after a severe head injury. I. Lumbar csf concentration in chronic brain post-traumatic syndromes. Eur. Neurol., 13, 528–544.

23.

Vecht

C.J.

, van Woerkom

C.A.

, Teelken

A.W.

, and Minderhoud

J.M.

(1975). Homovanillic acid and 5-hydroxyindoleacetic acid cerebrospinal fluid levels. A study with and without probenecid administration of their relationship to the state of consciousness after head injury. Arch. Neurol., 32, 792–797.

24.

Clifton

G.L.

, Ziegler

M.G.

, and Grossman

R.G.

(1981). Circulating catecholamines and sympathetic activity after head injury. Neurosurgery, 8, 10–14.

25.

Hamill

R.W.

, Woolf

P.D.

, McDonald

J.V.

, Lee

L.A.

, and Kelly

(1987). Catecholamines predict outcome in traumatic brain injury. Ann. Neurol., 21, 438–443.

26.

Nayak

A.K.

, Mohanty

, Singh

R.K.

, and Chansouria

J.P.

(1980). Plasma biogenic amines in head injury. J. Neurol. Sci., 47, 211–219.

27.

Woolf

P.D.

, Hamill

R.W.

, Lee

L.A.

, Cox

, and McDonald

J.V.

(1987). The predictive value of catecholamines in assessing outcome in traumatic brain injury. J. Neurosurg., 66, 875–882.

28.

Markianos

, Seretis

, Kotsou

, and Christopoulos

(1996). CSF neurotransmitter metabolites in comatose head injury patients during changes in their clinical state. Acta Neurochir. (Wien) 138, 57–59.

29.

Aquilani

, Iadarola

, Boschi

, Pistarini

, Arcidiaco

, and Contardi

(2003). Reduced plasma levels of tyrosine, precursor of brain catecholamines, and essential amino acids in patients with severe traumatic brain injury after rehabilitation. Arch. Phys. Med. Rehabil., 84, 1258–1265.

30.

McAllister

T.W.

, Flashman

L.A.

, McDonal

B.C.

, Ferrell

R.B.

, Tosteson

T.D.

, Yanofsky

N.N.

, Grove

M.R.

, and Saykin

A.J.

(2011). Dopaminergic challenge with bromocriptine one month after mild traumatic brain injury: altered working memory and BOLD response. J. Neuropsychiatry Clin. Neurosci., 23, 277–286.

31.

Donnemiller

, Brenneis

, Wissel

, Scherfler

, Poewe

, Riccabona

, and Wenning

G.K.

(2000). Impaired dopaminergic neurotransmission in patients with traumatic brain injury: a SPECT study using ¹²³I-beta-CIT and ¹²³I-IBZM. Eur. J. Nucl. Med. Mol. Imaging, 27, 1410–1414.

32.

Mathias

, and Wheaton

(2007). Changes in attention and information-processing speed following severe traumatic brain injury: a meta-analytic review. Neuropsychology, 21, 212–223.

33.

Willmott

, Ponsford

, Hocking

, and Schönberger

(2009). Factors contributing to attentional impairment after traumatic brain injury. Neuropsychology, 23, 424–432.

34.

Millis

S.R.

, Rosenthal

, Novack

T.A.

, Sherer

, Nick

T.G.

, Kreutzer

J.S.

, High

W.M.J.

, and Ricker

J.H.

(2001). Long-term neuropsychological outcome after traumatic brain injury. J. Head Trauma Rehabil., 16, 343–355.

35.

Wallesch

C.W.

, Curio

, Kutz

, Jost

, Bartels

, and Synowitz

(2001). Outcome after mild-to-moderate blunt head injury: effects of focal lesions and diffuse axonal injury. Brain Inj., 15, 401–412.

36.

Draper

, and Ponsford

(2008). Cognitive functioning and outcome 10 years following traumatic brain injury and rehabilitation. Neuropsychology, 22, 618–625.

37.

Kinsella

(2008). Traumatic brain injury and processing speed, in: Information Processing Speed in Clinical Populations. DeLuca

and Kalmar

J.H.

(eds). Taylor & Francis: New York, pps. 173–194.

38.

Vakil

(2005). The effect of moderate to severe traumatic brain injury (TBI) on different aspects of memory: a selective review. J. Clin. Exp. Neuropsychol., 27, 977–1021.

39.

Meyer-Lindenberg

, Nichols

, Callicott

J.H.

, Kolachana

, Buckholtz

, Mattay

V.S.

, Egan

, and Weinberg

D.R.

(2006). Impact of complex genetic variation in COMT on human brain function. Mol. Psychiatry, 11, 867–877.

40.

Bertolino

, Blasi

, Latorre

, Rubino

, Rampino

, Sinibaldi

, Caforio

, Petruzzella

, Pizzuti

, Scarabino

, Nardini

, Weinberg

D.R.

, and Dallapiccola

(2006). Additive effects of genetic variation in dopamine regulating genes on working memory cortical activity in the human brain. J. Neurosci., 26, 3918–3922.

41.

Aguilera

, Barrantes-Vidal

, Arias

, Moya

, Villa

, Ibanez

M.I.

, Ruiperez

M.A.

, Oret

, and Fananas

(2008). Putative role of the COMT gene polymorphism (Val158Met) on verbal working memory functioning in a healthy population. Am. J. Med. Genet. B Neurospychiatr. Genet., 147, 898–902.

42.

Goldberg

T.E.

, Egan

M.F.

, Gscheidle

, Coppola

, Weickert

, Kolachana

B.S.

, Goldman

, and Weinberger

D.R.

(2003). Executive subprocesses in working memory: relationship to catechol-O-methyltransferase Val158Met genotype and schizophrenia. Arch. Gen. Psychiatry, 60, 889–896.

43.

Sheldrick

A.J.

, Krug

, Markov

, Leube

, Michel

T.M.

, Zerres

, Eggermann

, and Kircher

(2008). Effect of COMT val158met genotype on cognition and personality. Eur. Psychiatry, 23, 385–389.

44.

Sengupta

, Grizenko

, Schmitz

, Schwartz

, Bellingham

, Polotskaia

, Stephanian

M.T.

, Goto

, Grace

A.A.

, and Joober

(2008). COMT Val108/158Met polymorphism and the modulation of task-oriented behavior in children with ADHD. Neuropsychopharmacology, 33, 3069–3077.

45.

Lipsky

R.H.

, Sparling

M.B.

, Ryan

L.M.

, Xu

, Salazar

A.M.

, Goldman

, and Warden

D.L.

(2005). Association of COMT Val158Met genotype with executive functioning following traumatic brain injury. J. Neuropsychiatry Clin. Neurosci., 17, 465–471.

46.

Flashman

L.A.

, Saykin

A.J.

, Rhodes

C.H.

, and McAllister

T.W.

(2004). Effect of COMT Val/Met genotype on frontal lobe functioning in traumatic brain injury. J. Neuropsychiatry Clin. Neurosci., 16, 238–239.

47.

Shores

E.A.

, Marosszeky

J.E.

, Sandanam

, and Batchelor

(1986). Preliminary validation of a scale for measuring the duration of post-traumatic amnesia. Med. J. Aust., 144, 569–572.

48.

Teasdale

, and Jennett

(1974). Assessment of coma and impaired consciousness: a practical scale. Lancet, 2, 81–84.

49.

Wechsler

(1997). Wechsler Adult Intelligence Scale—Third Edition. The Psychological Corporation: San Antonio, TX.

50.

Wechsler

(2008). Wechsler Adult Intelligence Scale—Fourth Edition. The Psychological Corporation: San Antonio, TX.

51.

Reitan

R.M.

, and Wolfson

(1995). Category teset and trail making test as measures of frontal lobe functions. Clin. Neuropsychol., 9, 50–56.

52.

Tombaugh

T.N.

(2004). Trail Making Test A and B: normative data stratified by age and education. Arch. Clin. Neuropsychol., 19, 203–214.

53.

Schmidt

(1996). Rey Auditory Verbal Learning Test: A Handbook. Western Pyschological Services: Torrance, CA.

54.

Meyers

J.E.

, and Meyers

K.R.

(1995). Rey Complex Figure Test and Recognition Trial: Professional Manual. Psychological Assessment Resources: Lutz, FL.

55.

Langeluddecke

P.M.

, and Lucas

S.K.

(2003). Wechsler Adult Intelligence Scale—Third Edition findings in relation to severity of brain injury in litigants. Clin. Neuropsychol., 17, 273–284.

56.

Lezak

M.D.

, Howieson

, Bigler

, and Tranel

(2012). Neuropsychological Assessment (5th ed.). Oxford University Press: New York.

57.

Sloan

, and Ponsford

(2012). Managing cognitive problems following TBI, in: Traumatic Brain Injury: Rehabilitation for Everyday Adaptive Living. Ponsford

, Sloan

, and Snow

(eds). Psychology Press: New York.

58.

Wechsler

(2001). Wechsler Test of Adult Reading. The Psychological Corporation: New York.

59.

Wilson

J.T.L.

, Pettigrew

L.E.L.

, and Teasdale

G.M.

(1998). Structured interviews for the Glasgow Outcome Scale and the Extended Glasgow Outcome Scale: guidelines for their use. J. Neurotrauma, 15, 573–585.

60.

Ponsford

, McLaren

, Schönberger

, Burk

, Rudzki

, Oliver

, and Ponsford

(2011). The association between apolipoprotein E and traumatic brain injury severity and functional outcome in a rehabilitation sample. J. Neurotrauma, 28, 1683–1692.

61.

Willmott

, Ponsford

, McAllister

, and Burke

(2013). Effect of COMT Val158Met genotpye on attention and response to methylphenidate following traumatic brain injury. Brain Inj., 27, 1281–1286.

62.

Glatt

S.J.

, Faraone

S.V.

, and Tsuang

M.T.

(2003). Meta-analysis identifies an association between the dopamine D2 receptor gene and schizophrenia. Mol. Psychiatry, 8, 911–915.

63.

McIntosh

T.K.

, Yu

, and Gennarelli

T.A.

(1994). Alterations in regional brain catecholamine concentrations after experimental brain injury in the rat. J. Neurochem., 63, 1426–1433.

64.

Bigler

ED.

(2005). Structurual imaging, in: Textbook of Traumatic Brain Injury. Silver

J.M.

and Yudofsky

S.C.

(eds). American Psychiatric Publishing: Airlington, VA.

65.

Weaver

S.M.

, Chau

, Portelli

J.N.

, and Grafman

(2012). Genetic polymorphisms influence recovery from traumatic brain injury. Neuroscientist, 18, 631–644.

66.

Waters

R.J.

, Murray

G.D.

, Teasdale

G.M.

, Stewart

, Day

, Lee

R.J.

, and Nicoll

A.R.

(2013). Cytokine gene polymorphisms and outcome after traumatic brain injury. J. Neurotrauma, 30, 1710–1716.