1

Journal of Alzheimer’s Disease xx (20xx) x–xx DOI 10.3233/JAD-131265 IOS Press

3

4

roo f

2

Among Vitamin B12 Deficient Older People, High Folate Levels are Associated with Worse Cognitive Function: Combined Data from Three Cohorts

or P

1

8

a The

12 13 14 15 16 17 18 19 20 21 22 23 24

dA

11

cte

10

University of Melbourne, Department of Psychiatry, Parkville, VIC, Australia Health, Geelong, VIC, Australia c National Ageing Research Institute, Parkville, VIC, Australia d Deakin University, School of Medicine, Waurn Ponds, VIC, Australia e The University of New South Wales, Department of Psychogeriatrics, NSW, Australia f Aged Care Psychiatry, Prince of Wales Hospital, Randwick, NSW, Australia g Austin Health, Heidelberg Repatriation Hospital, Heidelberg, VIC, Australia h The Queen Elizabeth Hospital, Woodville South, SA, Australia i The University of Melbourne, Mental Health Research Institute, Parkville, VIC, Australia j The University of Melbourne, Department of Pathology, Parkville, VIC, Australia k Edith Cowan University, Centre of Excellence for Alzheimer’s Disease Research & Care, School of Exercise, Biomedical and Health Sciences, Joondalup, WA, Australia l Sir James McCusker Alzheimer’s Disease Research Unit (Hollywood Private Hospital), Neurosciences Unit, Health Department of Western Australia, Perth, WA, Australia m The University of Melbourne, Centre for Neuroscience, Parkville, VIC, Australia n Austin PET centre, Austin Hospital, Heidelberg, VIC, Australia o Preventative Health Flagship, Commonwealth Science Industry Research Organization, Parkville, VIC, Australia b Barwon

rre

9

25 26 27 28 29 30

Un

Accepted 2 October 2013

co

6

uth

7

Eileen Moorea,b,∗ , David Amesa,c , Alastair Manderb , Ross Carneb,d , Henry Brodatye,f , Michael Woodwardg , Karyn Boundyh , Kathryn Ellisa,c,i , Ashley Bushi,j , Noel Fauxi , Ralph Martinsk,l , Colin Mastersi,m , Christopher Rowen , Cassandra Szoekec,o and David Wattersb,d

5

Abstract. Background: Folate fortification of food aims to reduce the number of babies born with neural tube defects, but has been associated with cognitive impairment when vitamin B12 levels are deficient. Given the prevalence of low vitamin B12 levels among the elderly, and the global deployment of food fortification programs, investigation of the associations between cognitive impairment, vitamin B12, and folate are needed. Objective: To investigate the associations of serum vitamin B12, red cell folate, and cognitive impairment.

∗ Correspondence

to: Dr. Eileen Moore, CNR Bellarine and Ryrie Street, PO Box 281, Level 4, Douglas Hocking Research Institute, The Geelong Hospital, Geelong, VIC 3220, Australia. Tel.: +61 420693726; Fax: +61 342152023; Emails: dr.eileenmoore@ gmail.com; [email protected].

ISSN 1387-2877/13/$27.50 © 2013 – IOS Press and the authors. All rights reserved

2

33 34 35 36 37 38 39 40 41 42 43

Methods: Data were collected on 1,354 subjects in two studies investigating cognitive impairment, and from patients attending for assessment or management of memory problems in the Barwon region of south eastern Australia between 2001 and 2011. Eligible subjects who had blood measurements of vitamin B12 and red cell folate taken within six months of cognitive testing were included. Subjects with stroke or neurodegenerative diseases other than Alzheimer’s disease were excluded. A Mini-Mental State Examination score of 45.3 nmol/L) was reported to increase from 7% before 1994 to 38% in 1999 in over-60 year olds [9] in a study that compared levels before and after food fortification in the US. Two large studies from the US; namely, the NHANES [6] and the Framingham Heart Study [10], reported that high folate or folic acid supplementation may exacerbate cognitive impairment associated with low serum vitamin B12 levels. The NHANES study investigated cognitive impairment and vitamin B12 levels in 1,301 community-

dA

35

cte

34

rre

33

Vitamin B12 is an enzyme cofactor in two reactions in humans; namely (1) the re-arrangement of methylmalonyl-CoA to succinyl-CoA, and (2) the methylation of homocysteine (Hcy) to form methionine. These reactions allow the correct formation of DNA during cell division, lipid metabolism, and methylation reactions to occur. The latter reaction is dependent on the availability of methyltetrahydrofolate (a form of vitamin B9) to act as a methyl donor. The metabolite Hcy accumulates during deficiency of either vitamin B12 or folate; whereas elevated methylmalonic acid (MMA) is a specific marker for vitamin B12 deficiency. MMA is a neurotoxic organic acid which may cause dysfunctional myelination [1]. The condition of elevated Hcy levels is called hyperhomocysteinemia (HHcy) and is a risk factor for cardiovascular disease [2] and Alzheimer’s disease (AD) [3]. Liver and other animal products are rich dietary sources of vitamin B12, whereas methyltetrahydrofolate is found in green leafy vegetables. Supplemental vitamin B9 is usually given as folic acid, an oxidized form of vitamin B9, due to its superior stability. Folic acid must first be converted to tetrahydrofolate before entering the methylation cycle via the activity of the dihydrofolate reductase enzyme. The activity of this enzyme is low in humans [4], leading to accumulation of unmetabolized folic acid in circulation when supplemental folic acid is taken. Data from the 2001–2002 cycle of the National Health and Nutritional Examination Survey (NHANES) of

co

32

INTRODUCTION

Un

31

uth

Keywords: Aged, Alzheimer’s disease, cognition, folic acid, vitamin B12

64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97

E. Moore et al. / Folic Acid, Vitamin B12, and Cognition

106 107 108 109 110 111 112 113 114 115 116 117 118

119

120

121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146

METHODS Participants and settings

Study design

roo f

105

PRIME study subjects were recruited during routine patient care at nine memory clinics and included people with AD (National Institute of Neurological and Communicative Disorders and Stroke – Alzheimer’s Disease and Related Disorders NINCDS-ADRDA criteria) and individuals with mild cognitive impairment (MCI, Petersen criteria). The aim of the PRIME study was to assess the clinical outcomes of patients with dementia at memory clinics across Australia. Participants were assessed at baseline and then after 3, 6, 12, 24, and 36 months. At each visit, participants underwent assessment by a neuropsychological battery for cognition, and participants’ primary caregivers completed Zarit burden interviews. AIBL study subjects volunteered after promotion of the study on television in late 2006, or were referred to the study by the clinician treating them for a memory problem, and included people aged over-60 years with normal cognition, MCI, or AD. The aim of the AIBL study was to investigate the markers that predict future AD. Participants were allocated to one of three diagnostic groups by clinical panel review. Participants of the AIBL study underwent a neuropsychological battery assessing the domains of cognition that are affected in AD and other dementias. Assessment was at baseline and then after 18 months. Eligible participants assessed at the Barwon region CDAMS or geriatrician for memory complaints were recruited at a routine visit. Eligible subjects had a MiniMental State Examination (MMSE) score performed within six months of blood being drawn. Participants with diagnoses of stroke or neurodegenerative diseases other than AD were excluded.

or P

104

uth

103

dA

102

years reported that mean serum folate levels increased by 38% in the years after voluntary food fortification was introduced [17]. Serum folate levels of the sample studied varied with the consumption of foods fortified with folic acid, but not with age or gender.

cte

101

Vitamin B12, folate, and cognitive test scores were examined for 1,354 participants. Blood samples for biochemical markers were taken within six months of cognitive testing. Participants were recruited from two large studies examining cognition in late life, namely the Prospective Research in Memory (PRIME) clinics study and the AIBL study. The PRIME study recruited 970 participants from nine sites in Australia, including three each in Victoria and New South Wales, and one each in Queensland, Western Australia, and South Australia. The AIBL study recruited 1,112 participants in Victoria (60%) and Western Australia (40%). The study cohorts and methods of the PRIME study and the AIBL study are described elsewhere [15, 16]. Additional participants who resided in the Barwon region of south east Australia between 2001 and 2011 were recruited through the local Cognitive, Dementia and Memory Services (CDAMS), and a geriatrician’s private practice. Most (83%) of the blood samples were taken between 2001 and September 2009; well after the introduction of voluntary fortification of foods with folic acid in Australia in 1995. The remainder was taken after the introduction of mandatory fortification in September 2009. A survey of Australians aged 27–77

rre

100

dwelling volunteers aged 60 years and older. Participants with low serum vitamin B12 levels (59 nmol/L) had higher levels of Hcy and MMA, and were four times more likely to experience cognitive impairment (odds ratio 4.3, 95% confidence interval (CI): 2.3–8.0) than individuals with normal levels of vitamin B12 and folate [6]. Vitamin B12 deficiency is common in older adults: 6.1% of over-65 year old Finns [11], 7.8% of Israelis [12], 15.3% of Canadians [13], and 23.8% of 74–80 year old Dutch people [14]. Given the global scale of deployment of food fortification programs, and the prevalence of vitamin B12 deficiency among older adults, the effect on cognition of high folate or folic acid supplementation merits immediate investigation. Previous studies have investigated serum folate or folic acid; whereas our aim was to investigate the associations of red cell folate, serum vitamin B12, and cognitive impairment in a large cohort from the Australian population.

co

99

Un

98

3

Biochemical measurements The lower reference value for serum vitamin B12 varies with the assay used, but approximates to 150 pmol/L. Megaloblastic anemia and sub-acute degeneration of the spinal cord occurs with serum vitamin B12 levels below this cut-off. A subclinical low-normal range (∼150–250 pmol/L) has been widely investigated [18]. Carmel et al. identified that metabolic abnormalities can occur (such as elevated

147 148 149 150 151

152

153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185

186

187 188 189 190 191 192 193 194

4

203 204 205 206 207 208 209 210 211 212 213 214 215 216 217

PRIME Study

451 AD

191 MCI

Statistical analyses

328 Other

82 MCI

682 HC

A binary logistic regression model was formed with cognitive performance as the response variable. Participants who scored below 24 on the MMSE were considered to have impaired cognitive performance. The MMSE is sensitive to age, level of education, and depression, so the model was adjusted for these CDAMS memory clinic

935 AD

141 Other

219 220 221 222 223 224 225 226 227 228 229 230 231

232

Private Geriatrician’s Practice

AIBL study

197 AD

roo f

202

or P

201

Institutional review was performed at each study host site. Reviewing committees included the Barwon Health Human Research Ethics Committee (HREC, Victoria), Austin Health HREC (Victoria), St Vincent’s Hospital Governance Review Unit (Victoria), Hunter New England HREC (New South Wales), Northern Hospital Network HREC (New South Wales), Northern Sydney Central Coast HREC (New South Wales), Metro North Health Service District HREC (Queensland), South Metropolitan Area Health Service HREC (Western Australia), and The Queen Elizabeth Hospital Ethics of Human Research Committee (South Australia).

uth

200

218

dA

199

Ethical approval

cte

198

rre

197

homocysteine or methylmalonic acid levels, or altered deoxyuridine suppression) when vitamin B12 levels are in the subclinical low-normal range, without immediate neurological or hematological signs [19]. Serum vitamin B12 levels 1,594 nmol/L) were considered ‘high’, whereas RCF levels of 1,594 nmol/L or lower were considered ‘normal’. These cut-offs for biochemical marker status approximate those used by the Framingham Heart Study investigators [10]. Serum vitamin B12 and RCF levels were measured using the ADVIA Centaur chemiluminescent microparticle immunoassay at most sites. Exceptions were as follows: Prime Site 2 used the Tosoh immunoassay analyser AIA600, Prime Site 4 used the Roche Cobas 8000 electrochemiluminescence immunoassay, and Prime Site 6 used a Siemens Healthcare Diagnostic Immulite 2000 immunoassay.

426 AD

238 MCI

198 HC

3,879 study participants’ or patients’ records were screened

co

196

291 records that were incomplete were excluded. Missing data included medication use, diagnosis, co-morbid conditions, pathology results, and date of birth.

3,588 participants records with complete medical histories were reviewed

Un

195

E. Moore et al. / Folic Acid, Vitamin B12, and Cognition

532 duplicate records for single participants were identified. These were identified from those having at least four out of the following five identifiers in common: 1. Date of birth, 2. Gender, 3. At least two matching MMSE scores/dates or serum vitamin B12 measurements/dates, 4. Place of residence, 5. At least two matching medications or co-morbid conditions.

Data from 3,056 individual participants were assessed for inclusion Excluded: - 1,015 participants, biochemical measurements not taken within six months of cognitive assessment, -121 participants with ‘stroke’ and -566 participants with diagnoses other than AD or mild cognitive impairment.

Data from 1,354 participants were included in the statistical analysis

Fig. 1. Recruitment, inclusion, and exclusion of the study cohort. PRIME study, Primary Research in MEmory (PRIME) study; AIBL, Australian Imaging, Biomarkers and Lifestyle (AIBL) study of aging; CDAMS, Cognitive, Dementia and Memory Services memory clinic; AD, Alzheimer’s disease; HC, healthy controls; MCI, mild cognitive impairment; MMSE, Mini-Mental State Examination.

233 234 235 236 237 238

E. Moore et al. / Folic Acid, Vitamin B12, and Cognition

247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268

269

270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287

DISCUSSION

RESULTS

The characteristics of the study cohort are shown in Table 1. Participants with low vitamin B12 levels (1,594 nmol/L) were older, and a greater proportion had a diagnosis of AD, than participants with biochemical measurements within the normal ranges. Both of the groups with high RCF (groups B and D) had a much higher proportion of folic acid supplement users (15% and 23%, respectively). Participants with low serum vitamin B12 levels (1,594 nmol/L) were more likely to have an impaired cognitive performance (MMSE 1,594 nmol/L) were also at increased risk for cognitive impairment (AOR 1.74, 95% CI: 1.03–2.95). This important new finding suggests that high folate levels may be an independent risk factor for cognitive impairment. Folic acid given in supplements and fortified foods may inhibit the shared biochemical pathway between folate and vitamin B12, leading to accumulation of homocysteine or to lowered methylation potential. This possible disease mechanism was suggested by Selhub et al. [22], and later by Faux et al. [8], in which high homocysteine was associated with both high or low folate levels. Evidence from the NHANES [23] and Framingham cohort [10] studies suggest that it is unmetabolized folic acid that is associated with poor cognitive function. In the former study, higher levels of 5-methyltetrahydrofolate were associated with better cognition, whereas higher folic acid was associated with poorer cognitive function. Both of the groups with the highest proportion of folic acid supplement users showed the greatest risk of cognitive impairment, so our data are consistent with the idea that high levels of folic acid may be detrimental to cognitive function. Frequently, clinicians have advised patients in the early stages of cognitive decline to take extra folate on the basis of early work reporting less dementia with high folate intake [24], so this is a possible source of confounding in observational studies that show association between high folate levels and poor cognition. In this study, information regarding a number of

uth

244

dA

243

tion of biochemical status with respect to cognitive classification, though confidence intervals were wide due to the small number in this group. Both the groups with low B12 and normal folate (C) and with high folate and normal B12 (B) also showed increased risk of cognitive impairment.

cte

242

rre

241

variables. Level of education was classified as ‘primary’ (less than six years of schooling), ‘secondary’ (between six and thirteen years of schooling), and ‘tertiary’ (more than 13 years of schooling). The reference category for level of education was ‘primary’. Participants were classified as having a history of depression if depression or depressive symptoms were noted in their patient medical history. Four categories for biochemical status were formed: namely, category (A) normal vitamin B12 and RCF, category (B) normal vitamin B12 and high RCF, category (C) low vitamin B12 and normal RCF, and category (D) low vitamin B12 and high RCF. The reference level for biochemical status used was category (A). The grouping and analyses technique used in this study is similar to that first described by Selhub and colleagues [6], and later used by Miller and colleagues [21]. MMSE scores and biochemical measurements were examined for 480 patients with AD, 187 with MCI, and 687 participants who were cognitively-intact. Participants were recruited during two controlled studies: (1) the PRIME clinics study (n = 970), and (2) the AIBL study (n = 1,112). An additional 1,797 patients treated for memory complaints in the Barwon region of south eastern Australia were also screened. There were 532 participants that were recruited from more than one source; their data were merged. Figure 1 describes recruitment, inclusion, and exclusion of participants forming the study cohort.

co

240

Un

239

5

288 289 290 291 292 293

294

295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336

6

E. Moore et al. / Folic Acid, Vitamin B12, and Cognition Table 1 Characteristics of cohort

Characteristic

Biochemical status (C)