LC is a 68-year-old female who came to my Memory Disorders Clinic accompanied by her elderly parents. She was unable to provide a proper history due to her confusion. Her parents relate that she had an eightmonth history of progressive memory loss, increasing obsessive behaviors (checking the mail 10 times a day, going in and out of the garage every five minutes to check the temperature gauge). She was becoming increasingly agitated and unable to comprehend simple conversation. Her pertinent history revealed a bipolar disorder over the past 40 years requiring hospitalization for shock treatment and long-term psychiatric follow-up. The only medication she was taking at the time that I saw her was Risperdal 2mg in the morning, thyroid medication, and estrogen. One week before coming to see me she had been driving from her home to her psychiatric follow-up clinic and was found by the police walking on the highway not knowing where she was. She was admitted to a nearby hospital.

The ER physician stated she appeared alert and oriented and had a completely normal physical and neurological examination, including vital signs. He noted she had some lip smacking and a few involuntary arm movements. Her tests in the ER included a normal urine and serum drug screen; all basic labs, including liver and kidney function, EKG and chest x-ray were normal. She was given Ativan for agitation and Ambien for sleep and admitted overnight. After a short observation she was discharged home. The hospital physicians felt she was improving and decided her diagnosis was a dissociative reaction (fugue state).

She saw me one week later. Her personal history revealed she had a Bachelor's degree in music, taught piano, and was an accomplished pianist. Over the last five years this interest had waned. She did not smoke or drink. Her family history revealed other members with long histories of psychiatric disorders but no history of dementia.

During my evaluation I administered the Montreal Cognitive Assessment Test. This test is similar to the Mini Mental Test but is much more sensitive to memory and visual executive function. She scored 13 out of 30, which is very impaired. She was unable to do simple trail making, draw a clock and put in the time, do simple math or similarities, and she was unable to give me any words beginning with the letter "F." However, she was oriented except to time and place. Her general knowledge and information, and in particular about music and composers, was very poor despite her music degree. The neurological and physical portions of her examination were completely normal. Her parents filled out an Activities of Daily Living assessment form that allowed them to describe her simple and complex activities of daily living. They indicated she was having moderate memory loss, was more irritable, and had difficulty with conversation. She was unable to pay her bills or use the computer and had a lot of trouble with her speech. She had been driving her car without incident until this recent episode where she got lost.

It was my opinion that she clearly had a dementia, which had been progressing for at least eight months. I believed the onset and decline were much too rapid for Alzheimer's disease but considered other treatable disorders and ordered a number of basic studies including an MRI of the brain, EEG, and thyroid, B-12 and homocysteine blood levels.

The patient and her parents returned a month later so I could review the test results with them and determine a course of action.

L.C.'s EEG and MRI of the brain were normal. Her thyroid study showed a TSH of 5, which is the lower limit of normal, and a homocysteine level of 16.6 (normal is 8 to 11). Her B-12 level was 86, versus the normal range of 250 to 900. The obvious concern here was her very low B-12 level. I told the family she needed vitamin B-12 treatment, preferably injections, so as to overcome any possible absorption difficulties. I contacted her family physician who began giving her vitamin B-12 injections.

Three months later she returned with her parents to see me in follow-up. When I saw her she was a completely different person. She was pleasant, very appropriate, and less distant. I repeated the Montreal Cognitive Assessment Test, and she scored 26 out of 30, which is normal for her age and education. (Recall that her first test had showed 13 out of 30.) The trail-making test, cube and clock drawing, math, similarities, and giving a list of words beginning with "F" were all normal. She was now able to remember three out of five words after 15 minutes, where previously she was unable to identify any delayed words. Further history revealed she had had no other changes in her treatment other than the B-12 injections. Clearly, this was a dementia due to B-12 deficiency, which was corrected by B-12 injections.

A case-based discussion is a departure for "Dementia Insights" articles, but this is an interesting case from my practice. As a reader, you may think this is not an unusual case and that you have seen many others. However, in my 34 years of neurology practice, and in the last 15 years specializing in cognitive disorders, this is the only case I have ever seen where a dementia appeared to be purely due to B-12 deficiency and completely resolved with treatment. This is also the lowest B-12 level I have seen, although some patients have had B-12 levels ranging from 100 to 200. This case prompted me to review some of the literature on B-12 deficiency and cognitive function.

As far back as medical school we've been taught that when anyone has a dementia or an unclear cognitive disorder, the physician should always consider low B-12 and order a blood level because this is a very treatable disorder. This became even more important in the 1960s and 1970s when it was determined that a low B-12 level can be present despite a normal CBC and red cell morphology. Science has known for many years the role that B-12 and folic acid play in the body, especially the nervous system.

Figure 1 shows how dietary folate is converted after a few steps to L-methyltetrahydrafolate (L-MTHF), which has an important methyl group. L-MTHF and homocysteine, in the presence of B-12 and methionine synthetase, forms methionine. Methionine is converted to S-adenysol methionine (SAM). The methyl group from SAM is released when SAM is converted to S-adenysol homocysteine (SAH). This methyl group helps form DNA proteins, phospholipids, and neurotransmitters in the brain. SAH is converted to homocysteine, and the cycle continues.

It appears that low B-12 not only causes central nervous system (CNS) damage by decreasing availability of methyl groups, but also allows accumulation of homocysteine, which appears to be toxic to nerve cells and blocks N-methyl-dasparate receptors. It is also important to remember that for B-12 by itself to be functional, it needs to be converted to glutathionyl cobalamin, which requires glutathione for its synthesis. Oxidative stress can tie up glutathione and contribute further to lack of B-12 function.

What do we really know about low B-12 and the development of neuropsychiatric symptoms? Moretti, R. et al. thoroughly reviewed this subject (2004).1 This review mentioned that Larner, et al. in 1999 and 2001, after conducting a thorough literature search, stated the number of B-12 deficiency dementia cases overall is very small. Psychiatric symptoms have been attributable to low B-12 including cognitive slowing, confusion, memory changes, delirium, depression, and acute psychotic state. However, the studies on low B-12 and cognitive impairment and dementia are very mixed. There are many studies compared to normal control showing no correlation with B-12 levels and dementia, but there are some that do. Moretti, et al. concluded that B-12 treatment in the presence of low B-12 may improve frontal lobe and language function in patients with cognitive impairment but rarely reverses dementia.1

While we are on the subject of B-12, it is worth mentioning some points about folate deficiency in cognitive function, which can be tied into low B-12 levels. I did not measure folate in my patient, but it likely was not a factor because the patient recovered completely with B-12 injections. Moretti, et al. stated in their review that Hultberg, et al. in 2000 published in a foreign journal findings from a double-blind, placebo group that showed a direct correlation between low folate levels and impaired memory and attention, which in turn showed significant improvement with supplemental folate treatments compared to placebo.1 Many other studies have shown this. Moretti, et al. concluded that correlational studies show an association between low folate and cognitive and psychiatric dysfunction, even if the association does not always mean causation.

How do we diagnose B-12 deficiency? When measuring B-12 blood levels, traditionally we would consider deficiency to be the cause of the neurologic problem if levels were less than 200pg/ml and there was clinical evidence of a neurological disorder. However, there are some very important facts about B-12 levels in the normal population, especially the elderly. There is evidence that low B-12 levels occur in 10 percent of older people, and the prevalence increases with age from five percent at age 65 to 20 percent at age 85. Some of the reasons for this are:

  1. Development of gastric atrophy as we age, which decreases absorption of normal B-12,
  2. Medications can compete with absorption (Metformin),
  3. Dietary deficiency, due to social/economic reasons or heavy vegetarian diet,
  4. A small number with antibody evidence of pernicious anemia.

Studies have shown that measuring B-12 levels alone is very insensitive to the diagnosis of true B-12 deficiency, especially in the ranges of 200 to 400pg/ml. The lab that does my B-12 measurements here in Austin includes a stamped note on all B-12 reports that states, "Please note: although the reference range for Vitamin B-12 is 200 to 910pg/ml, it has been reported that between five and 10 percent of patients with values between 200 and 400pg/ml may experience neuropsychiatric symptoms with or without hematology abnormal results due to occult B-12 deficiency. Less than one percent of patients with values above 400pg/ml will have neuropsychiatric symptoms."

Are we really recognizing B-12 deficiency in many of our cognitive impaired patients? It has been shown that a more reliable test of B-12 deficiency is increased methylmalonic acid (MMA) levels (98 percent sensitivity). Levels are increased because MMA converts to succinyl co-enzyme A, which requires the presence of B-12. Succinyl co-A, of course, is important for pathways of ATP production. Elevated homocysteine does reflect B-12 deficiency, but it is much less sensitive as a marker than methylmalonic acid.

Using low B-12 levels as a sole measure of B-12 deficiency will miss 50 percent of cases. Increased homocysteine levels in the presence of a normal B-12 level—greater than 400— strongly suggests that the elevated homocysteine may be due to folate deficiency. Folate may be in the low range. There has been a lot of literature in the last few years in regard to the toxicity of elevated homocysteine. It has been known to increase stroke risks, and levels of 14 or higher increase the risk of Alzheimer's in normal elderly to twice the normal rate. Increasing homocysteine levels has also been reported to worsen Alzheimer's disease and other dementias. It has been shown to be a toxic agonist to the NMDA receptors. These receptors are necessary for long-term potentiation in medial temporal lobe neurons necessary to store memory. Homocysteine blocks the NMDA receptor and leads to intracellular calcium with release of proteases and eventually cell death.

Tangney, C., et al. in Neurology (January 2009) presented a study to look at B-12, its metabolite, and age related cognitive decline, over a 10-year period.2 They did this by measuring B- 12, methylmalonic acid, and homocysteine levels to see if these levels were related to cognitive decline in 516 study patients. They did cognitive tests including the Mini Mental Test, the East Boston Test of Immediate and Delayed Recall, etc. The study sample averaged 80 years of age. The team found that 19 percent of the study group had increased homocysteine levels greater than 13.9 micromoles per liter, and 36 percent had increased methylmalonic acid greater than 271 (mmol/L). Probable and definite B-12 deficiency was present in 15 percent of the study group. They concluded in their study:

  1. Higher serum B-12 appeared to be associated with a lower rate of cognitive decline,
  2. Higher concentration of methylmalonic acid was associated with a faster rate of cognitive decline over a six-year period, and
  3. Homocysteine was not related to cognitive changes.

Other studies have not shown a correlation between B-12 levels and cognitive decline. Hin, H., et al. in 2006 did a study of 1,000 individuals in the United Kingdom.3 They found that cognitive function best correlated with holotranscobalamin (a tissue transport protein for B-12), homocysteine, and methylmalonic acid, not serum B-12. In this study, homocysteine was higher (greater than 16) and B-12 was 274 -/+ 139.

How long should we give low B-12 and/or folate replacement treatment to improve cognitive function? Unfortunately, very little good study data are available. One study done in India in 2005 by Aaron, et al. evaluated 63 patients with B-12 deficiency with mean age of 46-years.4 The mean duration of symptoms was 10 months. Thirty-eight percent had neuropsychiatric symptoms (25 cases) and 19 percent of the neuropsychiatric cases had dementia (five patients). The mean MMSE score in the dementia cases was 15.5 -/+ 7.5. All of the dementia cases had memory impairment, 75 percent had language impairment, 55 percent disorientation, and 45 percent had poor judgment and visual spatial disturbance. Serum B- 12 was low (less than 200 pg/ml) and folate levels were normal. The mean follow-up was 13.4 months (six to 34 months) after starting therapy. All received parenteral B-12. This consisted of 1,000 micrograms per day for seven days, then once a week for a month, and then once a month long-term. Onehalf of all the cases improved (33/63), and nine cases (14 percent) did not change. However, 21 cases, or 32 percent, were lost to follow-up.

We do not know how many of those that improved were in the cognitive impaired group. The authors only mentioned that cognitive function showed significant improvement, and the mean MMSE level improved 2-4 points and the activities of daily living improved. They concluded that at least six months of therapy should be given for B-12 and/or folate deficiency. Even if serum levels of B-12 and folate increase, the central nervous system tissues themselves require at least six months for the metabolic machinery to improve.

Suggested guidelines for evaluating B-12, folate, MMA and homocysteine blood levels in cognitive impaired patients include the following:

  • It still seems reasonable to measure B-12 and folate levels in patients with changes in cognitive function (acute or chronic) unless another cause is very obvious and improvement is occurring.
  • If the B-12 level is less than 400, it seems prudent to measure MMA to determine if B-12 deficiency is present.
  • If the MMA is elevated, this verifies B-12 deficiency. and parenteral B-12 replacement is indicated.
  • If the B-12 level is greater than 400, B-12 deficiency is unlikely.
  • If the folate level is low-normal, serum homocysteine should be measured. If elevated in the presence of a normal B-12 level, suspect folate deficiency. Homocysteine is the only metabolic measurement of low folate in the presence of a normal B-12 level. In Figure 1, homocysteine plus LMTHF coverts to methionine. L-MTHF is reduced or absent in folate deficiency.
  • If folate and/or B-12 deficiency are being treated, repeat cognitive evaluation and blood levels in three months. If the blood levels are in the normal range, continue to evaluate the patient's cognitive function for at least another three to four months to see if improvement occurs.

Ronald Devere, MD is Director of the Taste & Smell Disorders Clinic and Alzheimer Disease & Memory Disorders Center in Austin, Texas.