Abstract
Objective: To investigate the relation between head trauma and incidence of dementia in a prospective population-based study.
Background: Whether head trauma increases the risk of dementia and AD remains controversial. It has been suggested that the risk might be particularly increased for carriers of the APOE-ε4 allele.
Methods: The study population included 6645 participants of the prospective population-based Rotterdam Study, aged 55 years or older, who were free of dementia at baseline. Head trauma with loss of consciousness was measured at baseline by a self-report to a physician and detailed the number of head traumas, time since head trauma, and duration of loss of consciousness. The cohort was followed for incident dementia that was diagnosed according to international criteria. Logistic regression was used to calculate the risk of dementia after adjusting for age, gender, and education.
Results: No increased risk of dementia or AD was found for persons with a history of head trauma with loss of consciousness (relative risk [RR] for dementia = 1.0, 95% CI, 0.5–2.0; RR for AD = 0.8, 95% CI, 0.4–1.9). Multiple head traumas, time since head trauma, and duration of unconsciousness did not significantly influence the risk of dementia. In addition, the APOE-ε4 allele did not modify the relationship.
Conclusions: This study suggests that mild head trauma is not a major risk factor for dementia or AD in the elderly. In addition, this study does not concur with previous cross-sectional studies suggesting an interaction with the APOE genotype.
The association between head trauma and AD remains controversial. Although some reports suggested a positive relation with AD,1-6 other studies could not confirm head trauma as a risk factor for dementia or AD.7-14
An association between head trauma and AD is biologically plausible. Head trauma can lead to overexpression of the β-amyloid precursor protein, resulting in deposition of β-amyloid in the brain, similar to that seen in brains of AD subjects.15 Recently, it was hypothesized that head trauma may potentiate β-amyloid neurotoxicity, since it was found that brain injury in transgenic mice increased β-amyloid levels and exacerbated neuron death in the hippocampus without increasing plaque formation in the brain.16 The interest in head trauma as a putative risk factor for dementia and AD has been renewed by some authors, who reported a possible interaction between the APOE-ε4 allele and head trauma in increasing the risk of AD.17,18 The presence of APOE-ε4 seems to augment the formation of β-amyloid in the brain after head trauma.15
Thus far, most studies have been based on prevalent cases or on selected populations. We studied the relation between head trauma and dementia and its subtype AD in the prospective population-based cohort of the Rotterdam Study. Furthermore, we assessed the influence of APOE-ε4.
Methods.
Study population.
The Rotterdam Study is a prospective population-based cohort study that investigates chronic diseases in older adults. The study was approved by the Medical Ethics Committee of the Erasmus University, Rotterdam. The baseline survey occurred between 1990 and 1993. All persons 55 years of age and older, residing in a suburb of Rotterdam, The Netherlands, were invited to participate.19 Of those invited, 7983 (78%) accepted and gave informed consent.20 Participants were interviewed in their home, and persons living independently were subsequently examined during two visits at a research center. Institutionalized persons were examined in their institution. Dementia status at baseline was assessed in 7528 participants (94%), and the dementia-free cohort at baseline consisted of 7046 persons. Among these, information on previous head trauma with loss of consciousness was obtained for 6645 persons. Follow-up assessments occurred in 1993 and 1994 after an average of 2.1 years (SD = 0.8).
Head trauma.
At baseline, trained research physicians interviewed individuals with a detailed questionnaire at the research center. This questionnaire contained specific questions about previous head trauma, including number of head traumas, age at first and last head trauma, occurrence and duration of loss of consciousness, and post-traumatic amnesia.
Dementia assessment.
A three-phase process was used for dementia screening and diagnosis, both at baseline and follow-up.20 In the first phase, individuals were screened by trained research assistants using the Mini-Mental State Examination (MMSE)21 and the Geriatric Mental State Schedule (GMS-A).22 In the second phase, persons scoring below 26 on the MMSE or above 0 on the GMS-A were subsequently examined by a physician using the Cambridge Examination for Disorders of the Elderly (CAMDEX).23 In the third phase, individuals suspected of dementia were examined by a neurologist, neuropsychologist, and, if possible, underwent MRI of the brain. To complete follow-up, dementia status of those who died before examinations or those who were not re-examined in person was obtained from family practitioner medical files and ambulatory psychiatric services on an ongoing basis.24
Diagnosis of dementia was made by a panel that reviewed all respondent information using the American Psychiatric Association’s criteria (DSM-III-R).25 A subdiagnosis of AD was based on NINCDS-ADRDA criteria.26 Vascular dementia was evaluated on the basis of the NINDS-AIREN criteria.27
Apolipoprotein E.
APOE-ε4 was available in a large sample of 4070 participants with complete information on head trauma and dementia at follow-up (n = 102). APOE typing used genomic DNA primers that were amplified under specified conditions.28 The amplified PCR product was digested by Hhal, a restriction enzyme, and the resulting fragments were separated by 5% agarose gel electrophoresis.
Statistical analyses.
Multiple logistic regression was used to estimate the ORs of total dementia and AD for individuals who reported a previous head trauma with loss of consciousness. The ORs obtained with logistic regression were considered to represent estimates of relative risks. All relative risk estimates were adjusted for gender, age, and education. Stratified analyses were performed according to gender, number of head traumas, time since head trauma, and duration of loss of consciousness. Time since head trauma was divided into two time periods: head traumas that occurred >10 years and ≤10 years before baseline assessment. To assess a possible interaction between APOE-ε4 and previous head trauma with loss of consciousness, we computed, through the use of dummy variables, the relative risks for the following groups compared to the risk for individuals without an APOE-ε4 allele and without a history of head trauma: persons with at least one APOE-ε4 allele and without a history of head trauma; persons without an APOE-ε4 allele with a history of head trauma; and persons with at least one APOE-ε4 allele and a history of head trauma.
Results.
Baseline characteristics of individuals without dementia, with incident dementia, and with incident AD are listed in table 1. The mean age at baseline was 68.9 (SD = 8.7) years and 59.1% of the persons were women. During the follow-up period, 129 persons developed dementia, 91 (71%) of whom developed AD. Patients with incident dementia or AD were older, were more often women, and were less educated.
Baseline characteristics of the study population
Table 2 shows the adjusted relative risk of dementia and AD for persons with a previous head trauma with loss of consciousness. The relative risk was 1.0 (95% CI, 0.5–2.0) for dementia and 0.8 (95% CI, 0.4–1.9) for AD. There were no major differences in the relative risks between men and women. Exclusion of persons with cognitive impairment (MMSE score < 26) at baseline did not change the results (RR for dementia = 1.0; 95% CI, 0.5–2.1).
Estimated relative risk (RR) with corresponding 95% CI of total dementia and of AD for head trauma with loss of consciousness
The influence of the number of head traumas, time since head trauma, and duration of loss of consciousness is shown in table 3. . There were no significant differences between the relative risks for one head trauma compared to multiple head traumas, for head traumas experienced >10 years compared to within 10 years before baseline interview, or for duration of loss of consciousness of >15 minutes compared to ≤15 minutes. There was a trend, however, toward a higher risk of dementia and AD with increasing number of head injuries and duration of loss of consciousness.
Estimated relative risk (RR) with corresponding 95% CI of total dementia and of AD for number of head traumas, time since head trauma, and duration of loss of consciousness
Table 4 shows the combined effects of the APOE-ε4 allele and head trauma with loss of consciousness. The presence of at least one APOE-ε4 allele increased the risk of dementia. We found, however, no indication for a positive interaction between the APOE-ε4 allele and history of head trauma in increasing the risk of dementia or AD.
Independent and combined effects of APOE-ε4 and head trauma on the risk of dementia and AD (relative risk with corresponding 95% CI)*
Discussion.
The results of this study suggest that mild head trauma is not a major risk factor for incident dementia or AD, irrespective of the interval between head trauma and incidence of dementia. We could not confirm the findings from previous studies suggesting that head trauma would especially increase the risk of AD in carriers of the APOE-ε4 allele.
Our study has several methodologic strengths, which lend credence to our findings. The study is a prospective population-based cohort study, and exposure information was obtained before the occurrence of dementia. Therefore, recall bias that may invalidate cross-sectional case–control studies is less likely. Because baseline information was available for almost all participants and complete follow-up was obtained for the entire cohort, selection bias does not seem to play a role either. The study included a rigorous diagnostic workup of incident dementia among a general population, whereas previous prospective studies concerned with this relation were limited to register-based cases.7,8 Furthermore, most studies were based on relatively few patients.8,10,29 Patient numbers and the number of persons with head trauma in our study were larger, providing 80% power to detect a relative risk of at least 1.6.
There are a number of limitations of this study that have to be considered. Follow-up period was short. One might argue that some of the persons with incident dementia were subclinically dement at baseline. This could have resulted in underreporting of head trauma among those who became demented at follow-up and thereby could have biased the results away from a significant association between head trauma and dementia. However, exclusion of individuals who were cognitively impaired at baseline did not change the results, suggesting that recall bias due to subclinical dementia is not a likely explanation for our findings. Although we tried to maximize exposure recall by asking only about head trauma with loss of consciousness and by letting trained general physicians ask the questions, we recognize that people may fail to recall mild head trauma and that we therefore may have missed a number of head trauma cases, which would lead to attenuation of our results toward the null.
Since the early 1980s, case–control studies, cohort studies, and meta-analyses have sought to elucidate the specific role of head trauma as a determinant of dementia. The results were controversial. Some studies reported a significantly elevated relative risk1-5; others also reported a positive, but not statistically significant, association.30,31 More recent case–control studies and meta-analyses are inconsistent with the prior studies.9-14 The number of prospective studies on this topic is limited and most do not find an association.7,8,10 The primary exception is the Washington Heights study, which reported that the risk of AD was more than threefold increased for individuals with a history of head trauma with loss of consciousness.29 A recent population-based study of patients with AD showed that traumatic brain injury reduced the time to onset of AD but did not increase overall risk.32 Previously, duration of loss of consciousness and time elapsed between head trauma and onset of dementia have been suggested as risk factors for AD.2,3,29 These findings could not be confirmed by our prospective study, although there seemed to be a trend toward a higher risk of dementia among persons with multiple head traumas and a longer duration of loss of consciousness. The sample size of this study was, however, not sufficient to judge whether these trends were meaningful.
A few recent articles have addressed the issue concerning the independent and combined effects of head trauma and the APOE-ε4 allele on AD in cross-sectional analysis. First in 1995, it was shown in a community-based case–control study that the combined effect of head trauma and the APOE-ε4 allele on the risk of dementia was higher than expected based on the independent effects, suggesting synergism.17 Unfortunately, the number of incident cases was too small to study this prospectively.29 In a volunteer cohort, identified via the Alzheimer’s Disease Research Center, persons with at least one APOE-ε4 allele and a history of head trauma had an especially high risk of AD, although there was no evidence for a synergistic effect.18 A case–control study among enrollees of a health maintenance organization showed an increased risk of AD for individuals with a history of head injury but no modification by the APOE-ε4 allele.6 We also found no modification of the relationship between head trauma and incident dementia by the APOE-ε4 allele.
Acknowledgments
Supported by the NESTOR stimulation program for geriatric research in The Netherlands (Ministry of Health and Ministry of Education), The Netherlands Organization for Scientific Research (NWO), the Health Research and Development Council (ZON), and the municipality of Rotterdam.
Acknowledgment
The authors thank the general practitioners of Ommoord and the RIAGG Rijnmond Noord–Oost for providing medical information of participants and Hubert Backhovens, Marleen Van den Broeck, Anita Wehnert, and Sally Serneels for APOE-ε4 genotyping. The authors also thank Maarten de Rijk, Alewijn Ott, and Frans van Harskamp for diagnosing patients with dementia.
- Received January 20, 1999.
- Accepted July 28, 1999.
References
Letters: Rapid online correspondence
REQUIREMENTS
You must ensure that your Disclosures have been updated within the previous six months. Please go to our Submission Site to add or update your Disclosure information.
Your co-authors must send a completed Publishing Agreement Form to Neurology Staff (not necessary for the lead/corresponding author as the form below will suffice) before you upload your comment.
If you are responding to a comment that was written about an article you originally authored:
You (and co-authors) do not need to fill out forms or check disclosures as author forms are still valid
and apply to letter.
Submission specifications:
- Submissions must be < 200 words with < 5 references. Reference 1 must be the article on which you are commenting.
- Submissions should not have more than 5 authors. (Exception: original author replies can include all original authors of the article)
- Submit only on articles published within 6 months of issue date.
- Do not be redundant. Read any comments already posted on the article prior to submission.
- Submitted comments are subject to editing and editor review prior to posting.
