Apolipoprotein E polymorphism and outcome after mild to moderate traumatic brain injury: A study of patient population in India
Background: The nature and extent of recovery after traumatic brain injury (TBI) is heterogeneous. Apolipoprotein E (APOE) plays a major role in repair of cell membrane and growth of neurites following injury to cells. Studies done on the western population have shown that the APOE e4 genotype is associated with poor survival following neurotrauma. Aim: To explore the association of APOE polymorphism and outcome following TBI in a patient population from a tertiary care hospital exclusive for neurological diseases in south India. Patients and Methods: Ninety eight patients who sustained mild to moderate TBI (computed tomography (CT) scan brain showing traumatic parenchymal contusions) were the subjects of the study and the study period was from November 2003 to December 2008. APOE polymorphism status was determined by PCR technique using venous blood. Patients were assessed on follow-up with a battery of four neuropsychological tests as well as Glasgow outcome scale. Results: Of the 98 patients, 20 (20%) patients had at least one APOE e4 allele. A follow-up of minimum six months was available for 73 patients. None of the12 patients who had at least one APOE e4 allele had a poor outcome at six-month follow-up whereas 11(18%) patients without an APOE e4 allele had a poor outcome (Fisher's Exact test, P=0.192). On the neuropsychological tests, performance of patients with APOE e4 allele did not differ significantly from those without these alleles. Conclusion: This study does not support the current contention that the presence of APOE e4 allele should have a significant negative effect on the outcome after TBI.
Keywords: Apolipoprotein E, Glasgow outcome scale, neurotrauma, neuropsychological test
The pathophysiology of closed brain injury is complex and outcome depends on various factors like mechanism, type, severity, and location of injury in addition to various demographic variables (especially the age of the individual). Among patients who sustain similar injuries, the nature and extent of recovery after traumatic brain injury (TBI) is heterogeneous and is not fully explained by known and apparent demographic and injury related prognostic factors. These observations suggest that additional factors such as genetic and environmental influences modulate the secondary injury and recovery pathways.
Apolipoprotein E (APOE) plays a major role in repair of cell membrane and growth of neurites following injury to cells. , The synthesis of this protein is controlled by a gene on chromosome 19 with three primary alleles (ε2, ε3 and ε4). Studies carried out on western population have shown that the APOE genotype may predict recovery from TBI. APOE ε4 genotype is associated with a poor survival (acute as well as chronic) following TBI., Human APOE exhibits genetic polymorphism with varying prevalence rates reported among different ethnic groups. The proportion of subjects with at least one or more APOE ε4 alleles in various study populations is quite variable: 0% in tribal population of Koch, India; 29.5%  in Caucasian population); and 40.9%  in South African population. The largest study from India reported a APOE ε4 prevalence rate of 14.2%, which is the lowest reported rate in the world. The only African study has failed to find a positive association between APOE ε4 allele and poor outcome in patients with TBI. A positive association between APOE4 genetic expression and Alzheimer's disease and high prevalence of cognitive deficits progressing to Alzheimer's disease following head injury have also been reported. 
Mild to moderate TBI accounts for nearly 80% of all TBI. , However, most of the studies, clinical and basic research have been focused on severely brain injured individuals. The studies on the influence of APOE on cognitive function have been limited mostly to individuals who have had severe brain trauma. However, the observations in patients with severe TBI cannot be generalizable to majority of the individuals with mild and in some cases with moderate TBI.  In this study we explore the possible association between APOE polymorphism and outcome in a south Indian patient population with mild to moderate TBI.
The study period was from November 2003 to December 2008. One hundred and twenty eight patients were recruited for the study. Patients in the age group 16-65 (Median 44) years, who sustained mild or moderate (GCS 9 to 15) TBI and computed tomography (CT) scan showing cerebral contusion with or without associated injuries (extradural hematoma (EDH), subdural hematoma (SDH), diffuse edema, fracture etc) were included in the study. Patients who sustained severe TBI, penetrating injuries and polytrauma were excluded from the study. The protocol of the study has the approval of the Institutional Ethics Committee. Informed consent was taken from the patient or close relative in their own language. Severity of injury (GCS), demographic data along with the treatment given was documented on a structured proforma.
The APOE genotypic analysis was carried out using standard PCR method at Department of Genetics, University of Delhi, South Campus, New Delhi, India under supervision of Prof. B.L. Thelma Kuttappa, who has standardized the test for reproducibility and sensitivity.
Patients were clinically evaluated at six-month follow-up by Glasgow Outcome Scale (GOS) and a battery of four neuropsychological tests: Digit vigilance test (DVT sustained attention), Digit symbol substitution test (DSST mental speed), Complex figure test (CFT visuoconstructive ability and memory) and Token test (verbal comprehension). These tests were selected because in an ongoing project on head injured patients in our institute, National Institute of Mental Health and Neurological Sciences (NIMHANS), Bangalore, more than 25% of patients were found to have deficits on these tests (unpublished data). These tests have been standardized and validated with internationally accepted tests and are relevant to the local population in content, form and language. 
Of the 128 patients recruited for the study, because of technical reasons results of APOE polymorphism were available in 98 patients. Six months follow-up period was available for 81 patients and 73 patients with a known APOE genotype status had a minimum six month follow-up. Among these patients, neuropsychological assessment could be carried out in 51 patients.
a) APOE distribution in study population
Of the 98 patients tested for APOE polymorphism, 20 (20.3%) patients had one or more APOE ε4 allele. APOE ε33 was the most commonly found genotype followed by APOE ε34. There were only two patients with ε44 genotype and none with ε22 genotype. The allele frequencies for ε2, ε3 and ε4 alleles were 0.086, 0.801 and 0.112 respectively. Out of 81 patients with a minimum six-month follow-up, results of APOE polymorphism were available for 73 patients. Twelve patients had at least one APOE ε4 allele [Table 1].
b) Comparison of patients with and without APOE ε4 allele with clinical outcome
Among 73 patients who attended the follow-up clinic and had available APOE polymorphism data, 12 had at least one APOE ε4 allele. None of the patients with ε4 as one of the allele had poor outcome (Fisher's Exact test, P=0.192) [Table 2]. The two groups, however, did not significantly differ with regards to other demographic, radiological and treatment related parameters [Table 3]. The mean follow up was 10.9 months in patients with at least one APOE ε4 allele and 10.3 months in those without it. There were only two patients with APOE ε44 alleles. Both of them had a GOS of five at six months follow-up.
c) Comparison of GOS in patients with APOE ε33 and the rest of the patient population.
APOE ε33 group represented 68.4% of the patient population and 18.9% of the patients with this allelic combination had a poor outcome as compared with 5% in the group without it (Fisher's Exact test P =0.270) [Table 4].
d) Analysis of APOE alleles in patients who died
At six-month follow-up, of the 81 patients, 12 patients died. The results of the APOE genotype were available for 10 patients. None of these patients had an APOE ε4 allele [Table 5].
e) Analysis of neuropsychological assessment
Neuropsychological assessment at six months could be carried out in 51 patients. Illiterate patients could not perform a few of the tests used, hence the number of patients who could complete each of the four tests were different. DSST test was completed by 44 patients; DVT by 45 patients; CFT by 45 patients and Token test was performed by 51 patients.
Digit symbol substitution test (DSST - Test of mental speed)
On comparing the results of APOE genotype between the two groups, patient with APOE ε4 alleles had slightly lower scores but this difference was not statistically significant. On this test there was no significant difference between APOE ε33 group and the rest of the study population as well [Table 5].
Digit vigilance test (DVT - Test of sustained attention and visual scanning)
No significant differences were noticed in the test performance of patients with or without APOE 4 and APOE ε33 alleles [Table 6].
Complex figure test (CFT - Test of visuoconstructive ability and memory)
Performances of patients with APOE ε4 (ε4ε4, ε2ε4, ε3ε4) or ε33 alleles did not differ from the rest of the study group [Table 7].
Token test (Test of verbal comprehension)
Performance in this test by patients with APOE ε4 alleles was almost similar to that of patients without it. However, performance of patients in the APOE ε33 group was significantly inadequate on the Token test as compared to the rest of the study population (P =0.02, chi square test) [Table 8].
In view of the central role of APOE molecule in neuronal repair and regeneration, its genetic expression and protein synthesis pattern may predict recovery from TBI. In the nervous system, apolipoprotein E is engaged in the redistribution of cholesterol from cells during membrane synthesis, neuritic extension, growth and repair.  In various cell lines, apoε3 has been shown to increase growth and branching of neurites, where as apoε4 was found to have opposite affect., Most of the investigators have found a positive correlation between the presence of APOE ε4 allele and poor outcome after TBI. , These studies support the notion that a genetic factor can explain the variability in short and long term recovery patterns from central nervous system (CNS) insults.
Most of the studies evaluating the role of APOE polymorphism in TBI have been carried out in Caucasian population. The reported proportion of patients having at least one or more APOE ε4 alleles in various study populations was quite variable: 0%, 22.6%,  27.2%,  29.5%,  32%,  39.1%.  In a study from the African continent, this proportion was 40.9% (Black Zulu speaking patients).  In the present study from south India, 20.3% of the patients had at least one APOE ε4 allele. The prevalence is marginally less as compared to the Caucasian population and is significantly less as compared to African population. In a large community based sample (n=4450) from Northern India, Thelma et al.  reported a APOE ε4 prevalence rate of 14.2% with overall allele frequency of ε2, ε3 and ε4 being 0.039, 0.884 and 0.073 respectively. Other Indian studies with much smaller sample have reported APOE ε4 allele frequencies of 0.101 and 0.127.  The allele frequency in our limited sample of head injury patients for ε2, ε3 and ε4 was 0.086, 0.801 and 0.112 respectively.
Presence of APOE ε4 allele has been suggested in various clinical studies to have an negative effect on the outcome following closed head injury., In the study by Teasdale et al.  the patients with APOE ε4 allele were younger (34.3 years vs. 41.9 years) and a higher proportion had a low GCS scores than those without APOE ε4. In their study 57% patients with APOE ε4 had unfavorable outcomes compared with 16/59 patients (27%) without it. Studies from in the western population have also shown that in addition to having a poor outcome on GOS, patients with APOE ε4 allele have also performed poorly on various neuropsychological tests during follow-up.,, On the basis of poor performance on neuropsychological tests, Crawford et al.  concluded that patients with APOE ε4 had a significantly poorer memory function than those without APOE ε4.
In the present study, patients with APOE ε4 allele were younger (35.7 Vs 42.5 years) than those without APOE ε4. However, none among the 12 patients with APOE 4 allele had a poor outcome at six months of follow-up compared to 11 (23.5%) patients without it. This difference was not found to be statistically significant (p=0.192). Similarly, the performance of patients with APOE ε4 allele was not found to be significantly different on any of the four neuropsychological tests  as compared to patients without it at six-month follow-up. The results of our study are in contrast to the findings of the earlier studies in western population. This is the first study in Indian population evaluating the role of APOE polymorphism in predicting clinical outcome after closed head injury. The results demonstrate that patients with an APOE ε4 allele have a trend towards a slightly better outcome at six month follow up in Indian population.
There are a few studies that partially support our results. The Rotterdam study  (a population based prospective study) concluded that the APOE genotype was not an effective predictor of outcome after head injury. The study in African population also failed to find a positive correlation between presence of APOE ε4 allele and poor outcome. In a detailed neuropsychological assessment of 90 adult patients with mild and moderate TBI at six-month follow-up, Chamelian et al had found no effect of APOE ε4 allele on the poor outcome.
Teasdale et al,  conducted one of the largest studies till date (U.K. based patient population) enrolling 1000 head injury patients. Their results revealed that 36% of APOE ε4 carriers had an unfavorable outcome (GOS 1, 2, 3) compared to 33% of non-carriers of APOE ε4 (p=0.23) at six-month follow-up. They, however, found an association between ε4 allele and age. Children and young adults with ε4 allele had less favorable outcomes. Brichtova et al  in a study of 70 pediatric patients with TBI concluded that children with ε4 allele were more likely to have severe clinical symptomatology and unfavorable outcome. Han et al  studied the role of APOE genotype on neuropsychological outcomes one month following mild to moderate head injury in a group of 78 patients from military population. Their results revealed that ε4 carriers performed better on measures of attention, executive functioning and episodic memory encoding as compared to non-carriers. Hiekkanen et al also failed to find a positive correlation between presence of APOE ε4 allele and poor outcome after mild head injury.
There are few studies which have concentrated on the subgroup of patients within the APOE4 group. ,, The proportion of patients having ε2ε4 profile as compared to all patients with one or more ε4 allele ranges from 6.3%, to 22.5%.  There are fewer studies which have looked into the outcome in this group. Of the ten patients with ε2ε4 genotype in the study by Nathoo et al.  six had a good outcome. In the present study, three (25%) patients in the APOE4 group had a ε2ε4 genotype. All the three patients had a good outcome at follow-up. The higher number of ε2ε4 genotype and a slightly lesser number of ε3ε4 genotype patients in our patient population could be one of the factors to explain the different outcome seen in the present study.
This preliminary study indicates that APOE genotype does not influence the outcome after mild to moderate TBI. The theoretical explanation for our findings could be:
(1) high susceptibility of APOE gene product to environmental influences;
(2) presence of modifier genes in this ethnic population; and
(3) exclusion of patients sustaining severe head injury from our study population.
The limitation of the present study include: the distribution of APOE allele in the regional population is not known; the mean follow up duration was only 10 months; pediatric and severe head injury patients have been excluded; neuropsychological evaluation could not be done for all the patients; and the inclusion of ε2ε4 and ε3ε4 in the ε4 group ( this could act as a confounding variable and possibly minimize the effect of the e4 allele as, the presence of e2 and e3 alleles have been associated with opposite effects , as compared to e4 allele). Lastly, the present sample size though comparable to the previously published reports, ,,, is small. The study is a pilot project to spur further research in this field. Ideally, a large community based study with longer follow-up period and covering various geographical areas in India needs to be undertaken to give a clearer picture.
To conclude that our study suggests that in this part of the world patients with mild to moderate TBI with ε4 allele have a better outcome as compared to patients without it. However, it needs to be confirmed in larger population including people from different regions of India.
The authors are especially thankful to Ms Prachi Semwal-Research Scholar, Department of Human Genetics, Delhi University for her assistance in carrying out the genotypic analyses for APOE.
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5], [Table 6], [Table 7], [Table 8]