The diagnosis of vitamin B-12 deficiency has always been problematic owing to the unavailability of a robust assay.2
Normal concentrations of vitamin B-12 have been reported in patients with overt deficiency,3
and the type of assay used may be relevant.4
We use the Access Immunoassay System (Beckman Coulter) which has shown a coefficient of variation of 6-10% under the United Kingdom National External Quality Assessment Scheme in the recent past. This implies that the results obtained in the cases presented are qualitatively and quantitatively in defined categories of “normal” in case 1 and “deficient” in case 2.
Case 1 highlights the dangers of a “false normal” vitamin B-12 result. The existence of a large “grey area” for vitamin B-12 assays with regard to interpreting deficiency is well recognised. However, the repeated (three) results showing vitamin B-12 values in the upper part of the normal range in this patient is unusual. There may be unique analytical reasons in this patient (for example, heterophil antibodies), although this remains speculative. Despite a normal result, a diagnosis of vitamin B-12 deficiency was considered possible, and so treatment started straightaway. The more “logical” diagnosis—of myelodysplasia—would have led to her having a much more aggressive and potentially detrimental therapy combining chemotherapy and bone marrow transplantation.
In case 2, the vitamin B-12 concentration was particularly low despite the absence of clinical symptoms and normal bone marrow appearances. Possibly the patient is developing vitamin B-12 deficiency.5
Last year about 8500 serum vitamin B-12 assays from primary and secondary care were done in our haematology laboratory, which serves a population of 200 000. The number of assays is increasing annually by an average of 9%. Testing may be done as part of an assessment of anaemia, which may be mild,6
or as part of a screening process to assess neuropsychiatric symptoms.7
Testing for these disorders is increasing as a result of an increasingly older population. Of the vitamin B-12 tests done in our laboratory, 8% were in the “deficient” range (< 107 pmol/l), 10% were in the “intermediate range” (107-132 pmol/l), and 42% were in the “normal” range (132-227 pmol/l). It is difficult to know how many of these may have true B-12 deficiency or functional B-12 deficiency despite normal B-12 levels, particularly if other coincident medical problems such as bleeding and liver or renal disease are present.
It is difficult to know how many patients worldwide are diagnosed with myelodysplasia influenced by the presence of a normal serum vitamin B-12 concentration. Use of serum homocysteine and methylmalonic acid concentrations have been reported to help in identifying elderly patients with vitamin B-12 deficiency in a screening situation8,9
and to demonstrate biochemical benefit in such patients given vitamin B-12 treatment empirically.10
However, a recent study of patients in an ambulatory setting also raised the question of whether homocysteine and methylmalonic acid were as reliable as generally believed.11
Measurement of holotranscobalamin II, another recent marker for functional vitamin B-12 concentrations, is claimed to be more sensitive than methylmalonic acid12
but its clinical utility is open to question.13
Ultimately, meticulous clinical assessment— including assessment of other autoimmune conditions and taking a family history—is important, given that a single ideal test is still not available. Testing for anti-intrinsic factor antibody, despite a normal serum vitamin B-12 concentration, can be particularly useful when underlying pernicious anaemia is strongly suspected, as in case 1. However, a reduced serum vitamin B-12 concentration associated with a negative anti-intrinsic factor antibody, as in case 2, may not help in arriving at a diagnosis as only half of patients with pernicious anaemia are positive for this. Empirical treatment, to assess any clinical response and to prevent neurological damage, may be pragmatically justifiable as the dangers of treatment are not as devastating as those of not treating.