| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Letters to the Editor |
Clinical Biochemistry Department, Complejo Hospitalario, Universitario de Vigo, Vigo, Pontevedra, Spain
aAddress correspondence to this author at: Clinical Biochemistry Department, Complejo Hospitalario Universitario de Vigo, C/ Pizarro No. 22, 36204 Vigo, Pontevedra, Spain. E-mail milagros.blanco.perez{at}sergas.es.
To the Editor:
The effect of a substance present in a sample that alters the correct value of the result is referred to as analytical interference (1). Heterophilic antibodies consist of both natural antibodies and autoimmune antibodies that exhibit weak binding and polyspecificity, reacting with multiple heterogeneous antigens. Some heterophilic antibodies can react with human and animal immunoglobulins by binding to epitopes located on either the F(ab) or Fc region, although in healthy persons there is a predominance of idiotypic interactions (2).
We detected what appeared to be this type of interference in a study comparing the Olympus® immunoturbidimetric assay (ITA) for ferritin on the AU-5400 automated analyzer with the Roche® electrochemiluminescence immunoassay for ferritin on the Modular E-170 analyzer. Ferritin concentrations were determined by both methods in samples from 806 patients, and 2 samples were found to have discrepant results: 4 and 215 µg/L for one sample and 7 and 97 µg/L for the other sample, as measured by the Roche and Olympus methods, respectively. During 6 subsequent months of using the Olympus ferritin method for routine work, our laboratory measured ferritin concentrations in 13 143 patient samples, of which 2310 were selected according to the following criteria: result with an alarm; hemoglobin <11.0 g/L and ferritin >16 µg/L without inflammation, infection, or hepatopathy; and suspicion by the analyst or clinician. These samples were reanalyzed with both the Olympus and Roche methods. Of these 2310 samples, 13 appeared to exhibit similar interference, giving a prevalence of at least 0.1%. The mean ferritin concentration in the samples containing interferent was 76 µg/L (range, 9–256 µg/L) when assayed by the Roche method and was 244 µg/L (47–568 µg/L) when assayed by the Olympus method (P <0.05). These samples were treated with Scantibodies Laboratory® heterophilic blocking tubes (HBTs), in which the active blocker is a monoclonal antibody specific against a unique variable region of heterophilic antibodies that bind mouse immunoglobulin and are effective in blocking this interference in immunoassays (3). After treatment with HBT, we found that the heterophilic antibodies were either partly or completely neutralized, with a decrease in the ferritin concentration measured by the Olympus method. In contrast, the ferritin result obtained with the Roche method did not change after pretreatment with the HBT reagent. Rheumatoid factor (RF) was unlikely to be the cause of interference, as the RF concentration was <20 kIU/L in 12 of the samples and 34 and 78 kIU/L in 2 others (Table 1
).
|
Benoist et al. (4) have reported falsely increased C-reactive protein values measured with an turbidimetric assay on the Hitachi 911 analyzer (Boehringer®), which they attributed to heterophilic antibodies. We did not detect any heterophilic interference in other ITA methods from Olympus (antistreptolisin antibodies, C-reactive protein, transferrin), not even when analyzing the above samples.
We hypothesize that the interfering antibody in the Olympus ferritin assay is a natural antiidiotypic antibody that binds to a unique idiotype on the rabbit antihuman ferritin antibody linked to latex (5). The commercial ferritin reagent from Olympus does contain normal rabbit serum to minimize these interferences, but this is not a sufficiently specific blocking agent for the interfering heterophilic antibodies we identified. Thus, the accuracy of the results is compromised in patients with antibodies of this kind. Careful clinical validation by laboratory staff and correct interpretation of the results by the physician requesting the process may not always be not sufficient to detect this type of interference, and we would recommend that the Olympus ferritin method be reformulated by incorporating specific heterophilic blocking agents in the assay systems to minimize heterophilic interference.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
