Original

IMAJ • VOL 19articles

• may 2017 Original
IMAJ • VOL 19articles
• may 2017

low Borderline levels of serum vitamin B12 may Predict

cognitive Decline in elderly hip Fracture Patients
Eliyahu H. Mizrahi MD, Emilia Lubart MD and Arthur Leibovitz MD
Department of Geriatric Medicine and Rehabilitation, Shmuel Harofe Hospital, Beer Yaakov, affiliated with Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel

mia [4-7]. Most of the patients with biochemical vitamin B12

aBstract: background: The progression from cognitive impairment to deficiency have an intrinsic factor which helps in the absorption
dementia is a multifactorial process that involves genetic of vitamin B12 [8]. However as a result of achlorhydria or hypo-
and environmental factors. Vitamin B12 deficiency can be chlorhydria they may have difficulties in absorbing vitamin
an important factor in the progress from cognitive decline to B12 from food proteins [9]. In such cases, these biochemical
dementia. deficiencies are usually not diagnosed because they lack clinical
objectives: To examine the relationship between borderline features of vitamin B12 deficiency [8].The association between
low level of vitamin B12 (≤ 350 pg/ml) and cognitive decline serum vitamin B12 and cognitive decline is controversial. In
among a group of elderly hip fracture patients. the Longitudinal Aging Study Amsterdam, van den Kommer et
Methods: This retrospective chart review study was conducted
al. [10], did not find an association between vitamin B12 levels
in a geriatric rehabilitation ward of a university-affiliated
and cognitive decline. However, Clarke et al. [11] showed that
referral hospital. It comprised 91 elderly hip fracture patients.
more rapid cognitive decline was associated with low serum
Cognition was assessed by the Mini-Mental State Examination
vitamin B12 concentrations in community-dwelling elderly
tool. Fasting serum vitamin B12 levels were measured within
24 hours after admission to the rehabilitation ward.
people (n=1648) in Oxford, United Kingdom. According to the
results: Twenty-two of the patients had vitamin B12 levels ≤ general medical literature [12], normal serum levels of vitamin
350 pg/ml. In a multiple linear regression analysis, after B12 range between 300–900 pg/ml, with values below 200
adjusting for confounding variables, serum vitamin B12 pg/ml showing significant deficiency. However, as described in
levels ≤ 350 pg/ml were linked to a higher risk of developing the literature on geriatric patients [13], many elderly patients
cognitive decline (ß coefficient = -0.28, P = 0.008). have low borderline vitamin B12 serum levels (< 350 pg/ml)
conclusions: In our study, serum vitamin B12 levels ≤ 350 pg/ which improve with supplementation. Most of these patients
ml were independently associated with lower MMSE scores in are clinically asymptomatic. Low normal vitamin B12 levels,
elderly hip fracture patients. Serum vitamin B12 may assist in may be an early sign of a preclinical deficiency state. If this
identifying patients in the early stages of cognitive decline. assumption is correct then current laboratory norms for vita-
This study joins others that have reported on the association min B12 could be too low for the elderly population because
of low normal range vitamin B12 blood levels and conditions these levels may not identify patients with early deficiency.
like dementia, falls, fractures and frailty. We suggest a The aim of the present retrospective study was to investi-
reexamination of what is currently considered as the normal gate the association between cognitive decline and borderline
range of vitamin B12 in the elderly. serum vitamin B12 levels among older patients recovering
IMAJ 2017; 19: 305–308 from hip fracture. We hypothesized that serum vitamin B12
KeY worDs: vitamin B12, cognitive decline, elderly, hip fracture orders (e.g., myelopathy, neuropathy, brain
atrophy, depression and dementia),
cerebrovascular disease and megalobalstic ane-

he progression of cognitive impairment to dementia is a

t multifactorial process that involves genetic and environ-
mental factors [1]. One of the suggested important modifi-
able factors is vitamin B12 deficiency [2].
Among the elderly population, low serum levels of vitamin
B12 are common [3] and can be associated with neurologic dis-

305 305
Original
IMAJ • VOL 19articles
• may 2017 Original
IMAJ • VOL 19articles
• may 2017

levels lower than 350 pg/ml, as described in the literature on geriatric

patients [13], could be associated with cognitive decline among hip fracture
patients.

Patients anD methoDs

The study was approved by the local institutional review board. This retrospective
chart review analyzed consecutive patients between 2012 and 2014 who were
64 years of age or older and admitted to a geriatric rehabilitation ward of a
university- affiliated referral hospital with a primary diagnosis of hip
fracture. The standard rehabilitation course is based on an

306 306
Original
IMAJ • VOL 19articles
• may 2017 Original
IMAJ • VOL 19articles
• may 2017

interdisciplinary rehabilitative team approach and staff mem-

table 1. Clinical and demographic characteristics of patients
bers meet twice a week to evaluate the status of each patient. A

treatment plan is established and monitored with the purpose all B12 ≥ 350 B12 ≤ 350 P
variable patients pg/ml pg/ml value
of coordinating and integrating the various aspects of the staff
N 91 69 22
activities (medical, nursing, physical and occupational therapy,
Age, years 83.03 ± 6.34 83.07 ± 6.26 82.91 ± 6.71 0.91
and social work). These patients usually undergo a mean of 6
hours per week of physical and occupational therapy. Female gender 54 (59.3%) 41 (59.4%) 13 (59.1%) 0.98
The study sample included 109 consecutive patients admitted Education (years) 8.38 ± 5.41 8.64 ± 5.67 7.52 ± 4.5 0.41
with a diagnosis of recent hip fracture. We included all patients Diabetes mellitus, n (%) 35 (38.5%) 27 (39.1%) 8 (36.4%) 0.82
age ≥ 64 years (range: 64 to 96 years, mean: 83.03 ± 6.34) with Hypertension, n (%) 72 (79.1%) 56 (81.2%) 16 (72.7%) 0.39
pertrochanteric (extracapsular) or subcapital (intracapsular) hip Hyperlipidemia, n (%) 39 (42.9%) 32 (46.4%) 7 (31.8%) 0.23

fracture. The presence of ischemic heart disease (manifested as Ischemic heart disease, 59 (64.8%) 49 (71%) 10 (45.5%) 0.029
stable or unstable coronary syndrome), previous stroke, diabe- n (%)

tes mellitus, hypertension, hyperlipidemia, and atrial fibrilla- no statistically significant associations
tion had been established by medical history, or obtained by between low B12 vs. normal B12 patients by
interview or by a complete physical examination. We excluded age, gender, education, hypertension,
18 patients with serum vitamin B12 levels above the reference Parkinson’s disease, previous stroke, or diabetes
range (> 900 pg/ml) because high serum vitamin B12 concen- mellitus. MMSE
tration may be a laboratory sign of malignant hemopathies
or other serious clinical conditions [14]. As a result, the final
analysis included data of the remaining 91 patients.
Cognitive function was measured by the Mini-Mental State
Examination (MMSE) tool [15] within 1 week of admission.
Fasting serum levels of vitamin B12 were collected within
24 hours after admission to the rehabilitation ward. Vitamin
B12 concentration was determined using a radio assay kit
®
(COBAS 6000, Roche Diagnostics, USA).

statistical Methods
The comparison between patients with serum B12 levels,
dichotomized at 350 pg/ml were performed using t-tests for
continuous variables and chi-square tests for dichotomous
variables. Linear regression analyses was performed to simulta-
neously assess the independent relationships between vitamin
B12 and cognitive decline at admission and various comorbidi-
ties. A P value ≤ 0.05 was considered statistically significant.
Statistical analysis was performed using SPSS software (SPSS
Inc., version 21, Chicago, IL, USA).

results
The data of 91 consecutive hip fracture patients aged 64 years
and older admitted during a 2 year period (2012–2014) were
available. The clinical and demographic characteristics of these
patients are shown in Table 1. Mean age was 83.03 ± 6.34 years,
59.3% were women. The mean MMSE and mean serum vita-
min B12 levels were 16.86 ± 8.18 and 505.69 ± 200.17 pg/ml,
respectively. A total of 22 patients (24%) were found to have
serum vitamin B12 levels ≤ 350 pg/ml [Table 1]. There were

307 307
 Original
IMAJ • VOL 19articles
• may 2017 Original
IMAJ • VOL 19articles
• may 2017

Parkinson’s disease, n (%) 6 (6.6%) 5 (7.2%) 1 (4.5%) 0.66

Previous stroke, n (%) 15 (16.5%) 11 (15.9%) 4 (18.2%) 0.81
MMSE 16.86 ± 8.18 18.13 ± 7.46 12.86 ± 9.19 0.008
MMSE = Mini-Mental State Examination

table 2. Linear regression analysis predicting MMSE

independent predictors β P value
Vitamin B12 level -0.28 0.008
Age (years) -0.26 0.014
Gender (female) 0.052 0.65
Education (years) 0.26 0.012
Hypertension 0.087 0.42
Diabetes mellitus -0.102 0.43
Ischemic heart disease -0.041 0.77
Hyperlipidemia -0.049 0.64
Parkinson’s disease -0.11 0.27
Previous stroke 0.023 0.82
MMSE = Mini-Mental State Examination

score (18.13 ± 7.46 vs. 12.86 ± 9.19, P = 0.008) and ischemic heart disease [49
(71%) vs. 10 (45.5%), P = 0.029], emerged as the only statistically significant
parameters differing between low B12 vs. normal B12 patients [Table 1].
Because serum vitamin B12 levels higher than 350 pg/ml defined a group of
patients having higher MMSE scores, we performed a linear regression
analysis to test for independent predictors of MMSE scores. Higher serum
vitamin B12 levels (β = -0.28, P = 0.008) and higher education levels (β = 0.26,
P = 0.012) emerged as independently predictive of higher MMSE scores. Age
was independently and inversely associated with MMSE scores (β = -0.26, P
= 0.014) [Table 2]. None of the other variables that we tested, including gender,
hypertension, diabetes, ischemic heart disease, hyperlipidemia, Parkinson’s
disease, and previous stroke, were predictive of MMSE scores. A significant
correlation was found between serum vitamin B12 levels and MMSE scores
(Pearson’s correlation r = 0.206, P = 0.05), as shown in Figure 1.

308 308
Original
IMAJ • VOL 19articles
• may 2017 Original
IMAJ • VOL 19articles
• may 2017

A possible explanation for the association between low vita-

Figure 1. Correlation of cognitive status with serum vitamin B12
min B12 levels and cognitive decline is the inverse relationship
levels, using the MMSE tool (r = 0.206, P = 0.05)
that was found between serum vitamin B12 levels and plasma
30
total homocysteine [20]. High homocysteine levels are a risk
factor for cognitive decline [21] and have also been reported as
25
associated with a smaller hippocampus [22].
Another interesting finding is the association between
Mean MMSE at admission

20 the low borderline level of B12 with ischemic heart disease

which may hint to common risk factors and deserves further
15 investigation. Overall, as we mentioned earlier, as vitamin
B12 is involved in many essential biological processes more
10 attention should be accorded to the proper nutritional intake
of in general and to the elderly with cognitive deterioration
5 in particular.

0
study liMitations
Our study has several limitations. It is a retrospective study
0 200 400 600 800 1000

Serum vitamin B12 pg/ml

that took place in a single center. A comprehensive nutritional

assessment was not done and we used only one single cogni-
MMSE = Mini-Mental State Examination Our results suggest that elderly patients with serum vita- min B12 levels ≤
350 pg/ml should be screened routinely for cognitive decline.

Discussion
This study reports on the possible association of low serum
borderline vitamin B12 levels and cognitive decline in a
group of elderly hip fracture patients. Our data show that
serum vitamin B12 levels ≤ 350 pg/ml were associated with
lower MMSE scores, thus indicating cognitive decline. There
was an independent association between borderline vitamin
B12 levels and MMSE scores even after controlling for age,
gender, education, diabetes, ischemic heart disease, hyperten-
sion, previous stroke, Parkinson’s disease and hyperlipidemia.
The association between serum vitamin B12 levels and
cognitive decline remains controversial. Consistent with our
study, Nurk et al. [16], in the Hordaland Homocysteine Study,
which followed elderly community dwelling patients for 6 years,
reported an increased risk of cognitive decline with decreased
quintiles of baseline serum vitamin B12. In the Chicago Health
and Aging Project, Tangney and colleagues [17] found an
inverse associations between slower decline in cognition and
higher serum vitamin B12 levels and an inverse relationship
between MMA concentration and cognitive decline. However,
van den Kommer and co-authors in the Longitudinal Aging
Study Amsterdam (ages ≥ 65 years; n=1257, of whom n=1076
had longitudinal data) [10], Kang and co-authors in 635 women
> 70 years old from the Nurses’ Health Study [18], and Mooijaart
el al. in a population-based longitudinal study of 599 subjects
(Leiden 85-Plus Study, Netherlands) [19] did not find an asso-
ciation between serum vitamin B12 levels and cognitive decline.

309 309
Original
IMAJ • VOL 19articles
• may 2017 Original
IMAJ • VOL 19articles
• may 2017

tive screening tool. Holotranscobalamin, a

better indicator of vitamin B12 status, was
not measured [23,24]. In addition, despite
adjustments made for important
confounders, oth- ers could have been
considered, in particular those relating to
other metabolites, such as homocysteine, and
illnesses. In addition, caution must be
exercised with regard to extrapola- tion from
these findings and about drawing inferences
for dif- ferent populations. Despite the above
mentioned limitations, the present study is
advantageous because it points to the pos-
sible role of serum vitamin B12 on a cognitive
level among hip fracture patients. In clinical
practice, it is accepted that patients with serum
vitamin B12 levels < 200 pg/ml, should be
treated to prevent anemia, neuropathy and
cognitive decline [25]. However, in our
study we found an association between low-
normal serum vitamin B12 levels (≤ 350
pg/ml) and cognitive decline which might
suggest the need for earlier vitamin B12
supplementation.
Moreover, vitamin B12 is a multifaceted
neurotropic factor in the adult central nervous
system and has been reported both in the
blood and in the cerebrospinal fluid as
related to low cognition [7]. Such a
neurotropic factor could also be involved in
factors associated with the hip fracture
trauma of elderly patients. In view of these
accumulating reports as well as the low values
of what is now considered within the normal
range [11] and that the average age is now
considerably higher than it was when the
normal range was established, we consider a
revision of the normal range of serum vitamin
B12 levels to be necessary. A similar revision of
the normal range of vitamin D resulted in
adjusted values that now contribute to better
clinical guidance.
Further prospective studies are needed to
assess the causal relationship between serum
vitamin B12 levels and cognitive function.

310 310
Original
IMAJ • VOL 19articles
• may 2017 Original
IMAJ • VOL 19articles
• may 2017

12. Babior BM, Bunn HF. Megaloblastic anemias. In Isselbacher KJ, Adams RD,
correspondence
Brunwald E, et al. (eds.) Harrison’s Principles of Internal Medicine. New York:
dr. e.h. Mizrahi
McGraw-Hill, 2005: 601-7.
Dept. of Geriatric Medicine and Rehabilitation, Shmuel Harofe Hospital, Beer
Yaakov 70300, Israel 13. Sullivan DH, Johnson LE. Nutrition and aging. In Halter JB, Ouslander JG,
phone:(972-8) 925-8770 Tinetti ME et al. (eds.) Hazzard’s Geriatric Medicine and Gerontology, 6th Ed.
fax: (972-8) 925-8657 New York: McGraw-Hill, 2009: 439-68.
email: [email protected] 14. Ermens AA, Vlasveld LT, Lindemans J. Significance of elevated cobalamin
(vitamin B12) levels in blood. Clin Biochem 2003; 36 (8): 585-90.
references 15. Folstein M, Folstein S, McHugh PR. Mini-Mental State Examination. A practical
1. Migliore L, Coppede F. Genetics, environmental factors and the emerging role method for grading the cognitive state of patients for the clinician. J Psychiatr Res
of epigenetics in neurodegenerative diseases. Mutat Res 2009; 667 (1-2): 82-97. 1975; 12 (3): 189 -98.
2. Vogiatzoglou A, Smith AD, Nurk E, et al. Cognitive function in an elderly 16. Nurk E, Refsum H, Tell GS, et al. Plasma total homocysteine and memory in the
population: interaction between vitamin B12 status, depression, and elderly: the Hordaland Homocysteine Study. Ann Neurol 2005; 58 (6): 847-57.
apolipoprotein E ε4: the Hordaland Homocysteine Study. Psychosom Med 2013;
75 (1): 20-9. 17. Tangney CC, Tang Y, Evans DA, et al. Biochemical indicators of vitamin B12 and
folate insufficiency and cognitive decline. Neurology 2009; 72 (4): 361-7.
3. Lindenbaum J, Rosenberg IH, Wilson PW, et al. Prevalence of cobalamin deficiency
in the Framingham elderly population. Am J Clin Nutr 1994; 60 (1): 2-11. 18. Kang JH, Irizarry MC, Grodstein F. Prospective study of plasma folate, vitamin
4. Hin H, Clarke R, Sherliker P, et al. Clinical relevance of low serum vitamin B12 B12, and cognitive function and decline. Epidemiology 2006; 17 (6): 650-7.
concentrations in older people: the Banbury B12 study. Age Ageing 2006; 35 (4): 19. Mooijaart SP, Gussekloo J, Frolich M et al. Homocysteine, vitamin B12, and folic
416-22. acid and the risk of cognitive decline in old age: the Leiden 85-Plus study. Am J
5. Kim JM, Stewart R, Kim SW, et al. Changes in folate, vitamin B12 and Clin Nutr 2005; 82 (4), 866-71.
homocysteine associated with incident dementia. J Neurol Neurosurg Psychiatry 20. Selhub J, Miller JW. The pathogenesis of homocysteinemia: interruption of
2008; 79 (8): 864-8. the coordinate regulation by S-adenosylmethionine of the remethylation and
6. Selhub J, Bagley LC, Miller J, et al. B vitamins, homocysteine, and neurocognitive transsulfuration of homocysteine. Am J Clin Nutr 1992; 55 (1): 131-8.
function in the elderly. Am J Clin Nutr 2000; 71 (2): 614S-20S. 21. Tucker KL, Qiao N, Scott T, et al. High homocysteine and low B vitamins predict
7. Vogiatzoglou A, Refsum H, Johnston C, et al. Vitamin B12 status and rate of brain cognitive decline in aging men: the Veterans Affairs Normative Aging Study. Am J
volume loss in community-dwelling elderly. Neurology 2008; 71 (11): 826-32. Clin Nutr 2005; 82 (3): 627-35.
8. Carmel R. Mean corpuscular volume and other concerns in the study of vitamin 22. Williams JH, Pereira EA, Budge MM, et al. Minimal hippocampal width relates
B12 deficiency: epidemiology with pathophysiology. Am J Clin Nutr 2008; 87 to plasma homocysteine in community-dwelling older people. Age Ageing 2002;
(6): 1962-3. 31 (6): 440-4.
9. Allen LH. How common is vitamin B12 deficiency? Am J Clin Nutr 2009; 89(2): 23. Hvas AM, Nexo E. Holotranscobalamin–first choice assay for diagnosing early
693S-6S. vitamin B deficiency? J Intern Med 2005; 257 (3): 289-98.
10. Van den Kommer TN, Dik MG, Comijs HC, et al. Homocysteine and 24. Refsum H, Johnston C, Guttormsen AB, et al. Holotranscobalamin and total
inflammation: predictors of cognitive decline in older persons? Neurobiol Aging transcobalamin in human plasma: determination, determinants, and reference
2010; 31 (10): 1700-9. values in healthy adults. Clin Chem 2006; 52 (1): 129-37.
11. Clarke R, Birks J, Nexo E, et al. Low vitamin B12 status and risk of cognitive 25. Smith D, Refsum H. Vitamin B12 and cognition in the elderly. Am J Clin Nutr
decline in older adults. Am J Clin Nutr 2007; 86 (5): 1384-91. 2009; 89 (Suppl):707S-11S.

capsule

efficacy of a low-cost, heat-stable oral rotavirus vaccine in niger

Each year rotavirus gastroenteritis is responsible for about per 100 person-years, respectively), for a vaccine efficacy
37% of deaths worldwide from diarrhea among children of 66.7%. Similar efficacy was seen in the intention-to-treat
younger than 5 years of age, with a disproportionate effect analyses, which showed a vaccine efficacy of 69.1%. There
in sub-Saharan Africa. Isanaka et al. conducted a randomized, was no significant between-group difference in the risk of
placebo-controlled trial in Niger to evaluate the efficacy of a adverse events, which were reported in 68.7% of the infants
live, oral bovine rotavirus pentavalent vaccine (BRV-PV, Serum in the vaccine group and in 67.2% of those in the placebo
Institute of India) to prevent severe rotavirus gastroenteritis. group, or in the risk of serious adverse events (in 8.3% in the
Healthy infants received three doses of the vaccine or placebo vaccine group and in 9.1% in the placebo group); there were
at 6, 10, and 14 weeks of age. Among the 3508 infants who 27 deaths in the vaccine group and 22 in the placebo group.
were included in the per-protocol efficacy analysis, there were None of the infants had confirmed intussusception.
31 cases of severe rotavirus gastroenteritis in the vaccine N Engl J Med 2017; 376: 1121
group and 87 cases in the placebo group (2.14 and 6.44 cases Eitan Israeli

“change is the law of life. and those who look only to the past or present are certain to miss
the future”

311 311
Original
IMAJ • VOL 19articles
• may 2017 Original
IMAJ • VOL 19articles
• may 2017

John F. Kennedy (1917–1963), American politician who served as the 35th President of the United States
from January 1961 until his assassination in November 1963

312 312