The prevalence and incidence of dementia increase dramatically with age. It affects 8% of people age over 65 and more than 60,000 new cases each year in Canada. Alzheimer’s disease accounts more than 50% of dementia cases in Canada  . Since 2002, Thailand has been an ageing society with older persons constituting more than 10% of population. In Thailand, studies showed that the prevalence of dementia is 1.8% - 10.2% in the age group of 55 years and above   . Health Systems Research Institute of Thailand surveyed 21,960 people aged over 60 years in 2008-2009. The findings showed that 12.4% of people aged over 60 years had dementia and 9.8% being male while 15.1% were female. According to the latest population census surveyed in 2010 by the Office of National Statistics, it reports that the ageing population accounted for 12% of the whole population, and it is expected that the number will increase to 17% by the year 2020. It is estimated that there are as huge number as at least 300,000 people who have been diagnosed as having dementia across the nation. Cognitive impairment is one major symptom of dementia. Older persons increase in our society, which means a big number of people with decreased cognitive function. Therefore, it is important to search for modifiable risk factors. The amino acid homocysteine may be such a risk factor  . Most longitudinal studies have found associations between cognitive function scores and homocysteine levels. As can be seen in Table A, studies involving healthy elderly people yielded conflicting results. Some have shown significant associations between homocysteine levels and cognitive function  -  , whereas others have not   .
Studies investigate the relation between homocysteine and cognitive function scores, as shown in Table A  .
Therefore, it is important to search for modifiable risk factors. The amino acid homocysteine may be such a risk factor  .
Homocysteine is metabolized through 2 different pathways (Figure 1)   . The effects of homocysteine in the brain are multiple but can be broadly divided into nurotoxic and vascular effects (Box 1)   .
2. Subjects and Methods
The 100 subjects in this study were recruited from the Baan Bangkae Social Welfare Development Center for Older Persons, Bangkok and the vicinity. The study was approved by the Medical Ethics Committee of Ramathibodi Hospital, Mahidol University, Bangkok, Thailand, and written informed consent was obtained from all participants. All resident who were aged 60 - 80 years were invited to participate. During a visit, trained interviewers administered a questionnaire covering, among other areas, sociodemographic background, medical history, and medication use. This was followed by 2 visits to the Baan Bangkae Social Welfare Development Center, where subjects underwent clinical examinations, including neuropsychologic testing.
Table A. Studies investigating relation between homocysteine and cognitive function scores  .
Box 1. Effects of elevated homocysteine levels in the brain
Figure 1. Homocysteine is metabolized through 2 different pathways: the methionine synthase pathway and the cystathionine pathway. It has been proposed that impaired remethylation of homocysteine produces an increase of intracellular homocysteine that is toxic to neurons, and a decrease in S-adenosyl methionine. Impairment of this reaction would occur in states of cerebral oxidative stress, which would augment oxidation of an intermediate form of vitamin B12 (cobalamin) produced in the methionine synthase reaction and compromise the conversion of the vitamin to its metabolically active form. By Canadian Medical Association, Angeles Garcia, and Katherine Zanibbi, CMAJ, 2004, 171, 897-904.
2.2. Laboratory Measurement
Venous blood sample after overnight fasting were drawn according to standard procedure. Plasma or serum was isolated and stored at −80˚C before analysis. Serum total homocysteine was measured at the clinical chemistry laboratory of the Ramathibodi Hospital, Mahidol University, Bangkok, using automated chemiluminescent enzyme immunoassay method (Diagnostic Products Corporation, Los Angeles, CA); the CV ranged from 4.1% to 10.2%.
Global cognitive function was assessed with the Thai language by using mini-Mental State Examination score (MMSE), and cognitive functions were assessed by a neuropsychological battery test for memory, executive function, attention, visual-spatial organization, information processing and motor speed.
Other measurements: The following variables were considered as possible confounders: age; sex; cigarette smoking (current, former, never); alcohol consumption, assessed with a semi quantitative food frequency questionnaire  ; level of education, group into 4 levels (complete primary education, lower vocational or general education, intermediate vocational or general education, and higher vocational training, college, or university)  ; and hypertension, which was defined in accordance with the WHO  as a systolic blood pressure of 160 mmHg or more, a diastolic blood pressure of 95 mmHg or more, or use of antihypertensive medication.
2.3. Statistical Analysis
Multiple linear regression analysis was used to examine the relations between homocysteine levels and neuropsychological test scores with control for potential confounding variables to evaluate whether the relations were altered by these other variables. All tests were two-sided, and a p value of less than 0.05 was considered to be statistically significant. All data analyses were done with SPSS version 17.0 (SPSS Inc., Chicago, IL).
In Table 1, the mean ± SD age of participants was 72.8 ± 4.6 y. About one-half of the population had a history of hypertension (54.6%), 24.1% had diabetes mellitus and 15.1% had cardiovascular disease, respectively.
Table 2 for neuropsychological test found that the highest % abnormality test of C-W stimulustest, follow by digit symboltest, C stimulus, trial marking test D-KEFS condition 5, verbal pair total (VP total), retention, verbal paired associates 2(VP2), and block design (73.1, 38.3, 34.4, 26.6, 22.3, 15.6, 13.8 and 13.8), respectively.
Table 3 in the model 1, homocysteine was inversely associated with scores for learning slope test (B = −0.048, p = 0.042) and verbal pair total test (B = −0.124, p = 0.032). After adjusting for confounders, no association was found between homocysteine and cognitive impairment. Age (B = −0.129, p = 0.007) was found to be a significant determinant of decreased learning slope score (Table 4, model 2). Similarly, age (B = −0.298, p = 0.009) and education (B = 0.267, p = 0.029) were found to be significant determinants of decreased verbal pair total score (Table 5, model 3).
High homocysteine levels have been associated with an increased risk of stroke and other cardiovascular events  , which, in turn have been related to decreased cognitive
Table 1. Demographic and laboratory characteristics of the study subjects.
Table 2. Cognitive performance of the study subjects.
Table 3. Association between homocysteine and cognitive test score (model 1).
Verbal pair total: VP total, *significant p < 0.05.
Table 4. Multiple linear regression between homocysteine, B12, folic, age, education, Cr and learning slope (model 2).
Table 5. Multiple linear regression between homocysteine, B12, folic, age, Cr and verbal pair total (model 3).
function and dementia    . Thus, we hypothesized that a high level of homocysteine was associated with cognitive function. However, in the present study, there was no significant association between high levels of homocysteine and cognitive impairment in a population of institutionalized subjects (Baan Bangkae). Several methodological
Considerations arise when we try to explain our negative findings. We used only one global measure of cognitive function, the MMSE. However, the MMSE is a valid and reliable test  . However, the MMSE was not developed to estimate change in cognitive impairment. In addition, with our definition of cognitive decline, we were able to find an association with known risk factors, such as age and education. Similar results from Seshadri et al.  which found that cognitive decline associated with age education, and stroke. Still, random misclassification may have diluted our results for cognitive decline  . Demographic variables, particularly age and education, were significantly associated with cognitive function scores than was homocysteine. The reasons may be the small number of sample size (100), normal level of homocysteine in blood samples (14 umol/l, 34.04% hyperhomocysteinemia) used cut-off point >15 umol/l compared with other study which had a higher homocysteine levels (cut-off point >13 umol/l)     than our study 1, age of elderly Baan Bangkae, higher than other communities and had many diseases (HT, hyperlipidemia, DM, CVD) that effect homocysteine level and cognitive impairment.
In summary, although an association between homocysteine and cognitive impairment was biologically plausible, homocysteine did not seem to be a risk factor for cognitive impairment in this general population of the elderly. However, the possibility that homocysteine is truly not related to cognitive impairment cannot be discarded. Further researches such as a large number of participants and subjects in community are needed to find out for better quality of life in elderly population.
The authors gratefully acknowledge the coworkers at the research center of Ramathibodi Hospital, Mahidol University, Bangkok, Thailand and coworkers at Baan Bangkae Social Welfare Development Center for Older Persons, Bangkok and the vicinity. Finally, to the 100 subjects, whose were participated in this study.