| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Letters |
1
Laboratoire Universitaire de Biophysique, Unité d'Analyses Endocriniennes, CNRS UPRES-A 7004, Université Louis Pasteur, Faculté de Médecine, 67085 Strasbourg Cedex, France, ,
2
Service Central de Médecine Nucléaire, Hôpital Salengro, CHRU, 59037 Lille Cedex, France, ,
3
Service de Médecine Interne, Hôpital de Hautepierre, 67098 Strasbourg Cedex, France, ,
4
Clinique Marc Linquette, Unité de Soins Normalisés A CHRU, 59037 Lille Cedex, France
a Address correspondence to this author at: Institut de Physique Biologique, Faculté de Médecine, 67085 Strasbourg Cedex, France. Fax 00 33 3 88 14 48 61; e-mail sapin{at}alsace.u-strasbg.fr.
To the Editor:
We read with interest the paper by Després and Grant (1) on antibody interference in thyroid assays. Thyroid hormone autoantibodies, heterophile antibodies, and rheumatoid factors are certainly the main sources of artifacts. As mentioned by the authors, anti-thyrotropin (anti-TSH) antibodies are more uncommon but may nevertheless deserve additional comments.
The existence of anti-TSH antibodies in patient sera has been reported after injections of bovine TSH (2)(3). The antibodies also appear in autoimmune thyroid diseases such as Graves disease, Hashimoto thyroiditis, silent thyroiditis, and subacute thyroiditis (4)(5)(6)(7), and nonthyroid autoimmune disease (6). In sera from patients with Graves disease, the possibility that thyrotropin receptor antibodies (TRAbs) may be anti-idiotype antibodies against anti-TSH antibodies or that anti-TSH antibodies may be anti-idiotype antibodies against TRAbs is controversial (8)(9)(10)(11)(12). Most of the reported anti-TSH antibodies reacted against bovine TSH; however, some also reacted against human TSH (4)(12)(13)(14).
The results of published studies on anti-TSH antibody interference in TSH assays concerned mainly RIAs. In those cases, depending on the assay design and the antibody specificity, interference may yield lower or increased values. Increased results were found with the double antibody techniques (4)(5)(13)(15)(16)(17). Single antibody techniques with polyethylene glycol (PEG) precipitation yielded low values (14)(15). Fewer results have been reported with the widely used, "sandwich" immunometric assays (IMAs). IMA results have been found to be lower (5) or similar to double antibody results (6). Moreover, different IMA kits may yield discrepant values (14).
We previously reported (18) TSH concentrations that we measure (19) with eight different third-generation IMAs in four serum samples that contained anti-TSH antibodies as determined by increased precipitation of protein-bound 125I bovine or human TSH. Two samples from patients with autoimmune thyroid disorders (Graves disease and postpartum thyroiditis) contained only anti-bovine TSH antibodies. The results of the different TSH kits were not grossly discrepant, ranging from 0.36 to 0.60 mIU/L and from 2.9 to 4.7 mIU/L for the two samples, respectively. The other two sera contained both anti-bovine and anti-human TSH antibodies. In the first case, our suspicion was aroused because the high serum TSH contrasted with an apparently healthy clinical picture. The second case was from a euthyroid woman who had given birth to two children with transient neonatal hyperthyrotropinemia, a case similar to those presented in previous reports (17)(20). This patient's serum contained neither stimulating nor blocking TRAbs. The thyroid-stimulating TRAbs were checked by measuring the stimulating activity (cAMP production) of the serum on thyroid cells. Blocking TRAbs were ruled out because, after preincubation of the serum with thyroid cells followed by two wash steps, added TSH displayed normal stimulating activity. This serum blocked the stimulating activity of TSH on thyroid cells only when TSH and the serum were incubated simultaneously. We considered that this serum contained anti-TSH antibodies capable of inhibiting the stimulating activity of TSH. Measurements with the different kits yielded, for these two euthyroid patient sera containing anti-human TSH antibodies, TSH results that were highly discrepant, ranging from 2.2 to 36.6 mIU/L and from 2.1 to 13.9 mIU/L, respectively [see Sapin et al. (18) for details]. Of 12 values, 9 were in the hypothyroid range. As measured by the BeriLux kit, estimates of TSH concentrations in these samples were within the health-related reference interval after immunoglobulins were eliminated through pretreatment with 250 mL/L PEG and centrifugation for 30 min at 4 °C (18). After this treatment, the mean TSH recovery was 66% in control sera without anti-TSH antibodies. This decreased recovery value was taken into account when the results after PEG treatment were calculated (1.2 mIU/L for both patients).
Our observations underline the possibility of misinformation in TSH assays because of anti-TSH antibodies, even with recent kits. For the first two patients, variation between kits could be explained in part by the bias we observed between the results of the different kits in euthyroid control patients (19)(21). The median ranged from 1.0 to 1.9 mIU/L. This is definitely not the case for the last two patients. During this study, which has spanned >5 years, 300 000 TSH determinations have been performed by the two laboratories in which these four cases were found. The incidence of anti-TSH antibodies may exceed 4 in 300 000 because some cases may have been missed. However, the incidence of TSH results showing a gross discordance either with the clinical status of the patient or between methods because of anti-TSH antibodies could be only 2 in 300 000.
Despite the forthcoming preclusion of bovine TSH injections (22) and despite the fact that recombinant human TSH (Thyrogen) does not seem to be a potent immunogen (23), anti-bovine and anti-human TSH antibodies may both be spontaneously present in the sera of thyroidal ill or clinically healthy patients. Moreover, it is worth noting that anti-TSH antibodies have been consistently reported to be a source of interference in TRAb assays, yielding negative TRAb values (5)(6)(7)(12)(13)(24). This interference frequently contributed to the discovery of anti-TSH antibodies.
This work was supported by the Hôpitaux Universitaires de Strasbourg. We thank H. Bornet for the anti-human TSH antibody determinations and A.M. Madec for measuring the stimulating and blocking activities on thyroid cells. We also thank N. Heider for carefully reviewing the manuscript.
References
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  R. Sapin, A. Agin, and F. Gasser Misleading High Thyrotropin Results Obtained with a Two-Site Immunometric Assay Involving a Chimeric Antibody Clin. Chem., May 1, 2004; 50(5): 946 - 948. [Full Text] [PDF]
|
|
|
|
|
|
J. G. H. Vieira, S. K. Nishida, M. T. Faria de Camargo, O. M. Hauache, R. Monteiro de Barros Maciel, and V. Guimaraes "MacroLH": Anomalous Molecular Form That Behaves as a Complex of Luteinizing Hormone (LH) and IgG in a Patient with Unexpectedly High LH Values Clin. Chem., December 1, 2003; 49(12): 2104 - 2105. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
