Study population

Study design

Cognitive assessment



Budge et al., 2002 [5]

158 community dwelling people age 60 - 91 yr

Cross section of prospective cohort


Higher tHcy levels associated with lower memory scores per umol/L (OR 1.15, 95% CI 1.10 - 1.27)

OR adjusted for age, sex, serum cystatin C level and systolic blood pressure

Duthie et al., 2002 [6]

334 community dwelling people who had participated in Scottish Mental Survey of 1932 and 1947

Cross section


tHcy levels negatively associated with scores on RPM, WAIS in older cohort with higher tHcy levels (mean 10.9 umol/l, 95% CI 10.1 - 11.5)

Results adjusted for childhood intelligence quotient

Pins et al., 2002 [7]

1077 people aged 60 - 90 yr in Rotterdam Scan Study

Cross section of prospective cohort

Abbreviated Stroop tesatl Letter-Digit Substitution Task, Verbal fluency test, PPMST, Modified Rey’s test

Patients with tHcy > 14 umol/l had lower scores for global cognitive function (diference-0.20, 95%CI-0.30 - 0.11)

Results adjusted for age, sex, education level, depression, serum creatinine level

Miller et al., 2003 [8]

1789 community dwelling people aged > 60 yr in Sacramento Area Latino Study on Aging

Cross section of prospective cohort

3MSE, verbal and visual memory tests, object naming conceptualization and attention span tests

Inverse relation between tHcy levels and scores on 3MSE (p = 0.02), picture association (p = 0.05), verbal attention span (p = 0.04), and recognition tests (p = 0.001),

Multiple linear regression model included folate, cobalamin, age creatinine, sex, education and acculturation

Ravaglia et al., 2003 [9]

650 community dwelling people aged 65 - 91 yr (mean 73 yr) with normal cognitive function in Conselice Study

Population based study


Inverse relation between odds of tHcy level > 15 umol/l and MMSE scores

Results adjusted for age, income, education level, serum creatinine level, serum vitamin B index, active lifestyle, coffee and meat consumption

Garcia et al., 2004 [10]

281 community dwelling people aged >65 yr

Cross section

Stroop, Mattis DRS, CVLT

Subjects with elevated tHcy levels (>13.9 umol/l) had lower stroop scores than those with normal tHcy levels in univariate analysis (p < 0.05)

Strongest association found between methylcitric acid and cognitive scores

Dufouil et al., 2003 [11]

1241 people aged >60 yr in Epidemiology of Vascular Aging Study

prospective cohort; 4-yr follow-up

MMSE, Trail

Making Test Part B, Digit Symbol Substitution Test from the WAIS, Finger Tapping Test

Odds of cognitive decline 2.8 (95% CI 1.2 - 6.2) in patients with tHcy level > 15 umol/l

OR adjusted for age, sex, education level, baseline cognition, BMI, alcohol consumption, smoking, hypertension, hypercholesterolemia,

Glycemix status, history of vascular disease, and folate and B12 levels

Kalmijn et al., 1999 [12]

702 community dwelling people aged > 55 yr in Rotterdam Study

prospective cohort; mean follow-up 2.7 yr


No association between tHcy and cognitive impairment (highest v. lowest tertile, OR 0.91, 95% CI 0.52 - 1.58)

OR adjusted for age, education level, and baseline MMSE score

Ravaglia et al., 2000 [13]

54 people aged > 65 yr in Conselice Study

Cross section of prospective cohort

MMSE, clock drawing test, prose memory test, Corsi block tapping task, Mental Deterioration Battery

No association between tHcy and cognitive test scores

Results adjusted for age, sex, education level, smoking status, alcohol or coffee consumption, and previous cardiovascular disease