Diagnosing vitamin B-12 deficiency on the basis of serum B-12 assay

I have been looking for information on this, as we are finding an
increasing number of patients with significant deficiency which would
appear to relate to diet (teenagers) or other conditions (alcoholism,
pregnancy) which respond readily to oral supplements (the only one I am
aware of that is easily available is Forceval - only 3mcg cyanocobalamin
daily though, so it would not be enough where malabsorption was the
problem). Yet hospital doctors and other GPs institute what is often
lifetime treatment with injections on the basis of only slight reduction
in B12, without consideration of whether this is a deficiency due to
malabsorption or not. I strongly suspect that there is a tendency to look
no further than injection treatment as it is safe, cheap and easy:
however, it distresses patients (it appears more painful than most other
injections) and ignores other causes.

Forceval is certainly more expensive on a drug cost basis, but when
the cost of a professional actually giving a B12 injection every three
months is taken into account, I suspect it would be no more expensive, and
may well be less. It is more appropriate when malnutrition is a cause.
Clearly it is important to check on response, but that is easy enough. I
have just had a result back on an alcoholic patient, with poor diet,
started on Forceval three months ago when his B12 level was 86ng/L: this
week it was 209 ng/L. MCV is not raised, so I am not concerned about
missing a true B12 deficiency.

I note the reference in one of the responses to "physiological"
deficiency in pregnancy - I would be interested to know how low can be
considered "physiological". It seems impossible locally to get midwives to
check folate and B12 when treating anaemia, but we often find low levels
of both. I have recently found a lady who had severe hyperemesis early on,
whose ferritin is low normal, but her B12 is only 66ng/L (and her folate
is low too). I treat pregnant women with B12 below 100ng/L with one or two
injections of B12, but then follow them up to make sure their levels
remain normal. This is based purely on what seems safe to me. I cannot
find any help locally - the haematology department here never agree to
Schilling tests if asked, and simply tell us to give B12 injections.

I note that DTB (Vol 47, No 2) recommends high dose B12 tablets (1mg)
instead of injections, which are simply not available here, but this high
dose is not really necessary for most younger people I think. They refer
to oral treatment with 50mcg-150mcg daily, but I am not sure what this is
based on - I do not know of any such tablets.

I would be interested in any response to this, even though I am
writing three years late.

Competing interests:
None declared

Dear Sir,

We have read with interest the paper entitled “Diagnosing vitamin B12
deficiency on the basis of serum B12 assay” from the last BMJ issue (1).

We have been aware that the paper has stirred some anxiety among British
colleges. To our opinion, we feel that "dysmyelopoietic appearance" might
be a greater problem than neurological lesions nowadays, especially in
countries in which homocysteine is not a routine test. In fact there is a
possible risk to confuse “atypical forms” of vitamin B12 (cobalamin)
deficiency with dysmyelopoetic stages, especially in elderly patients. In
our experience (of more than 300 patients with documented cobalamin
deficiency), this risk exists in at least 20% of the patient (2,3). Thus
to exclude this possibility of confusion, we usually propose a "traitement
d'épreuve" (exclusion therapy) with oral cobalamin therapy (1000 µg per
day of oral cyanocobalamin for at least 7 days). In fact, in several
studies, we have reported that one week of oral vitamin B12 therapy may be
effective to improve or cure cobalamin deficiency (4-6). The readers of
the BMJ may find several informations in a practical review of cobalamin
deficiency management we have recently published in the CAMJ (7).

Professor Emmanuel ANDRES

For the study group of cobalamin deficiency of the University Hospital of
Strasbourg,
Department of Internal Medicine,
University Hospital of Strasbourg,
France,
emmanuel andres@chru-strasbourg.fr

1. Devalia V. Diagnosing vitamin B-12 deficiency on the basis of
serum B12 assay. BMJ 2006; 333: 385-386.

2. Andrès E, Affenberger S, Vinzio S, Kurtz JE, Noel E, Kaltenbach G et
al. Food-cobalamin malabsorption in elderly patients: clinical
manifestations and treatment. Am J Med 2005; 118: 1154-1159.

3. Andrès E, Affenberger S, Zimmer J, Vinzio S, Grosu D, Pistol G et al.
Current hematological findings in cobalamin deficiency. A study of 201
consecutive patients with documented cobalamin deficiency. Clin Lab Haem
2006; 28: 50-56.

4. Kaltenbach G, Noblet-Dick M, Andrès E, Barnier-Figue G, Noel E, Vogel T
et al. Réponse précoce au traitement oral par vitamine B12 chez des sujets
âgés hypovitaminiques. Ann Med Interne (Paris) 2003; 154: 91-95.

5. Andrès E, Kaltenbach G, Noblet-Dick M, Noel E, Perrin AE, Vinzio S et
al. Hematological response to short-term oral cyanocobalamin therapy for
the treatment of cobalamin deficiencies in elderly patients. J Nutr Health
Aging 2006; 10: 3-6.

6. Andrès E, Kaltenbach G, Noel E, Noblet-Dick M, Perrin AE, Vogel T et
al. Efficacy of short-term oral cobalamin therapy for the treatment of
cobalamin deficiencies related to food-cobalamin malabsorption. A study of
30 patients. Clin Lab Haem 2003; 25: 161-166.

7. Andrès E, Loukili NH, Noel E, Kaltenbach G, Ben Abdelgheni M, Perrin AE
et al. Vitamin B12 (cobalamin) deficiency in elderly patients. CAMJ 2004;
171: 251-260.

Competing interests:
None declared

Sir,

One of the cases in the report by Devalia is a timely reminder that current “no boil” automated cobalamin immunoassays may produce normal results in the presence of severe cobalamin deficiency. Since 2004 the external quality assessment scheme for cobalamin assays UKNEQAS Haematinics has become aware of a further eight cases of sera from patients with severe megaloblastic anaemia or subacute combined degeneration of the cord in whom current commercial assays have failed to detect cobalamin deficiency. In all eight cases intrinsic factor antibody levels were strongly positive, suggesting that intrinsic factor antibody interference may be the cause of assay failure. These cases are not just a difficulty in the selection of a cut off point for the reference range but are “false normal results” with levels well within the manufacturers normal reference range and yet severe clinical deficiency. Manufacturers’ product literature stresses that diagnosis of cobalamin deficiency should not be based on the results of a single assay and that all clinical features and other laboratory tests including intrinsic factor antibody levels should be considered.

Devalia suggests heterophilic antibody interference may be relevant to the mechanism but the cases reported are more likely to be due to high titre anti-intrinsic factor antibody levels.

In two patients both with megaloblastic anaemia, which subsequently responded to cobalamin treatment, the initial cobalamin assay results were normal, but the clinical features aroused suspicion that the laboratory result was incorrect. Pre treatment sera were available in both cases for further study by UKNEQAS haematinics. Both sera had high titre anti-intrinsic factor, heterophilic antibodies were not detected in one case, and the pre-treatment sera were investigated further in a round robin exercise of cobalamin assays.

Table 1. Pretreatment serum cobalamin in two cases with false normal cobalamin results with current commercial assays.

The results demonstrate that only two of five methods found extremely low levels of cobalamin consistent with the clinical presentation of severe cobalamin deficiency in both cases. The Bayer centaur assay detected one of the two patient samples as low. The mechanism of assay failure is not yet clear, nor the frequency with which this may occur. The current commercial assays are competitive binding immunoassays which are no boil, and rely on alkaline hydrolysis and dithiothreitol or monothioglycerol treatment to release cobalamin from transcobalamin and other binders, and denature intrinsic factor antibody, prior to competitive binding against labelled cobalamin to assay intrinsic factor. We postulate that intrinsic factor antibody present in the patient serum at high titre fails to be denatured by the alkaline hydrolysis and dithiothreitol treatment and persists into the binding stage, resulting in assay failure. UKNEQAS Haematinics has alerted all manufacturers to this issue.

Ideally assays should be tested pre-commercial release with sera which contain high titre anti-intrinsic factor antibody to ascertain whether the assays will detect low cobalamin levels in this circumstance. It is not yet clear whether some assays are more vulnerable to assay interference that others.

The incidence of these”false normal” sera are rare but in our own clinical laboratory we detected four similar cases over a two year period from 12,000 cobalamin assay requests. Laboratories should not report cobalamin results in isolation from intrinsic factor antibody results or full blood count data. Clinicians are urged to be vigilant for these potentially dangerous “false normal” cobalamin results, and once suspected laboratories should store pre-treatment sera where available for repeat analysis on an alternative assay. Elevated homocysteine and methylmalonic acid levels which correct on B12 treatment, and positive anti-intrinsic factor antibody will confirm that the megaloblastic anaemia or neuropathy is indeed due to cobalamin deficiency.

These cases of assay failure should be reported to the in vitro diagnostics section of Medicines and health care related products agency (MHRA) as adverse events via http://www.mhra.gov.uk. UKNEQAS Haematinics would be pleased to assist laboratories to investigate any suspected false normal sera ( email: ukneqas.haematinics.org.uk) and collect data on any further cases.

A trial of therapy in suspected cobalamin therapy once serum has been stored will prevent delay in treatment and adverse clinical consequences.

Malcolm.S.Hamilton

Director UKNEQAS Haematinics Good Hope NHS Trust Rectory Rd. Sutton Coldfield B77 7RR

Sheena Blackmore

Deputy director UKNEQAS Haematinics

Anne Lee

UKNEQAS Haematinics Scheme scientist

Competing interests:
None declared

Based on a price of US$660 per kg of vitamin B12 [watsonii.com
product V013], 1 gram or $0.66 provides about 1 million daily required
doses of 1 mcg and would be sufficient for 34 eighty year long life-times,
with 1 gram arguably representing the only non-toxic true megadose of any
nutrient.

Conversely, $0.02 is the life-time cost for vitamin B12 for a human.
Even assuming worst case scenarios or
when elderly and that only 1/100th is absorbed, this likely represents a
fraction of the cost of the blood analysis of the 4.25% of the patient
population measured yearly, and found to contain 18% individuals 'below
normal' for B12.

Considering the current debate about folate fortification and its
theoretical harm from masking vitamin B12 deficiency, this is a golden
opportunity to fortify grain products with both vitamins with, for
example, targeting B12 at 10 mcg/d. A zero cost remedy to known deficiency
problems that simultaneously solves the majority of the diagnosis issues
raised by Devalia.

While scientists debate and before politicians act regarding food
fortification, my U.S. mail-ordered 'high potency' multivitamin / mineral
capsule contains 100 mcg B12 [and 800 mcg folic acid plus 6 other B-
vitamins] and I have yet to see science that this amount of B12 does not
eventually replete most worst case scenario patients at virtually no
cost, risk or effort. The vitamin bottle has a six month supply at
$0.12/day and comes without a child-proof cap; it's considered that safe!
vos{at}health-heart.org

Competing interests:
None declared