| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Letters to the Editor |
Departments of1 Internal Medicine and, 3 Clinical Chemistry and Hematology, Bronovo Hospital, The Hague, The Netherlands, 2 Diagnostic Products Corporation, Breda, The Netherlands
aAddress correspondence to this author at: Department of Internal Medicine, Bronovo Hospital, Bronovolaan 5 2597 AX, The Hague, The Netherlands. E-mail tomvlasveld{at}hotmail.com or lvlasveld{at}bronovo.nl.
To the Editor:
More than 60 years after the identification of vitamin B12 [cobalamin (Cbl)], both analytical determination and clinical interpretation of its concentration in serum remain troublesome (1)(2)(3).
High serum Cbl concentrations frequently occur in myeloproliferative disorders and several hepatic diseases (4). These conditions are seldom associated with signs of Cbl deficiency, but in some cases high serum Cbl concentrations have been associated with clinical evidence of Cbl deficiency related to the presence of antibodies against the Cbl-binding protein transcobalamin (holoTC) or high concentrations of abnormal Cbl-binding proteins (5)(6)(7). We report a case in which an analytical problem within the Immulite 2000 assay [Diagnostic Products Corporation (DPC)] led to a falsely increased Cbl result for a patient presenting with classic hematologic and biochemical features of pernicious anemia.
We evaluated a 39-year-old man for macrocytic anemia. Laboratory examination results were as follows: hemoglobin, 6.8 mmol/L (reference interval, 8.4–10.9 mmol/L); mean corpuscular volume, 127 fL (80–100 fL); reticulocytes, 2.7% (0.5%–2.0%); lactate dehydrogenase, 2267 U/L (200–450 U/L); haptoglobin, <58 mg/L (160–2000 mg/L); folic acid, 9.7 nmol/L (6–39 nmol/L); and Cbl, 1199 pmol/L (130–700 pmol/L). Autoantibodies against parietal cells and intrinsic factor (IF) were positive. Because of the likelihood of pernicious anemia, we tested for Cbl deficiency. Increased concentrations of homocysteine [42.7 µmol/L (reference values <15.6 µmol/L)] and methylmalonic acid [2.46 µmol/L (reference values <0.30 µmol/L)] and a decreased concentration of holoTC [<10 pmol/L (reference interval, 55–170 L)] together with concentrations of circulating free holoTC and haptocorrin that were within the reference intervals provided metabolic evidence for Cbl deficiency at the cellular level. After the patient underwent 4 months of treatment with Cbl, the aberrant laboratory findings, including the homocysteine and holoTC concentrations, had returned to reference values.
We measured Cbl with a competitive chemiluminescence assay in which the endogenous Cbl and the Cbl present in the assay compete for binding to hog intrinsic factor (HIF). In this study we used the commercially available assays Immulite 2000 (DPC) and ADVIA Centaur (Bayer Corporation). In both assays, the initial step is alkaline hydrolysis, initiated by the addition of dithiothreitol (DTT) and sodium hydroxide/potassium cyanide solution, to denature Cbl-binding proteins and to inactivate IF-blocking antibodies. In the Immulite 2000 assay, the sample is incubated with immobilized Cbl-coated polystyrene beads, HIF, and alkaline phosphatase–labeled HIF-specific antibody. Unbound conjugate is removed by centrifugal washing. In the ADVIA assay, the sample is incubated with HIF coupled to immobilized paramagnetic particles and acridinium ester–labeled Cbl. After magnetic separation of bound and unbound Cbl and a washing step, the chemiluminescence activator is linked to anti-HIF antibody bound to the immobilized Cbl-coated beads in the Immulite 2000 assay and to the acridinium ester-labeled Cbl in the ADVIA assay. The amount of emission is inversely proportional to the Cbl concentration in the sample. The initially used Immulite 2000 assay was repeated in the hospital and in DPC laboratories and revealed a Cbl concentration >1200 nmol/L. The ADVIA assay gave a Cbl concentration of 114 nmol/L. A solid-phase IF-blocking antibody assay performed by DPC revealed an unusually high ratio of 4.75.
After the patient underwent 4 months of treatment with Cbl, his Cbl concentration was 511 nmol/L as measured by the Immulite 2000 and 460 nmol/L by the Centaur assay. In the initial test, we used Immulite 2000 assay reagent lot 159, and 4 months later we used lot 170. We reassayed the initial sample with lot 170 and obtained a Cbl concentration result of 90 nmol/L. We concluded that the initial high Cbl concentration was caused by an analytical problem within lot 159 of the Immulite 2000 assay reagents. Because of the unusually high anti-IF antibody concentration in the sample from our patient, we hypothesize that the initial step to inactivate the IF-blocking antibodies failed, probably because of diminished DTT activity. DTT is susceptible to oxidation, and oxidized DTT may lose its ability to inactivate the IF-blocking antibodies, particularly when these are present in high concentrations. On the basis of these findings and assumptions, the manufacturer adjusted the volumes of the reagents during the first step of the assay and added a DTT stabilizer to minimize the risk of instability.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
