Effect of Rheumatoid Factor on Cardiac Troponin I Measurement Using Two Commercial Measurement Systems

The Abbott AxSYM cardiac troponin I (cTnI) immunoassay may generate false-positive results in the presence of specific human anti-animal antibodies (HAAAs) (1). We use the term HAAAs rather than heterophile antibodies, a term used for weak, multispecific antibodies against diverse antigens (2), either human or animal. Techniques are available to detect and correct the interference attributable to HAAAs. We have used the Heterophile Blocking Tube (HBT; Scantibodies Inc.) to identify interference from suspected HAAAs. Anecdotal reports and experience in our laboratory have also suggested that the Abbott AxSYM cTnI may also be susceptible to interference from rheumatoid factor (RF), a heterophile antibody. Abbott Laboratories has recently modified its cTnI reagent system to address the issue of HAAA and heterophile antibody interference. The objective of this study was to compare the Beckman Access and Abbott AxSYM (original and modified) cTnI assays in serum samples containing RF, before and after pretreatment with HBT. We first designed a retrospective study on 23 sera with RF >80 kIU/L. We measured cTnI on both analytical systems with and without prior HBT treatment. In the second part of the study (during the phase-in period of the modified formulation), separated sera of 19 RF-positive samples (>80 kIU/L) were analyzed with Abbott’s original and modified cTnI immunoassay. Samples for both studies were obtained with all patient identifiers removed. In the first study (Table 1 A), the Beckman Access method gave cTnI results of 0 μg/L for 22 of 23 samples, with the remaining sample having a result of 0.08 μg/L. On the Abbott AxSYM, values ranged from 0 to 13.1 μg/L. Pretreatment of specimens with the HBT did not change the results with the Beckman Access method. For the Abbott AxSYM original formulation after pretreatment with the HBT, the results for eight samples remained unchanged, five showed increases in value from 1.2- to 4.2-fold, and seven showed decreases in value from 20% to 94%. Pretreatment with HBT reduced only one of the AxSYM cTnI results to 0 μg/L. In the second part of the study (Table 1B), the original Abbott formulation yielded results of 0 μg/L for 3 of 19 RF samples, with results for the remainder as high as 10.0 μg/L. In analysis of the same specimens with the modified formulation, 17 of 19 samples had values of 0 μg/L. One sample increased from 0.2 to 0.4 μg/L, and one decreased from 4.3 to 0.5 μg/L. Two interesting cases arose during these studies. The first was an individual with a cTnI of 74.1 μg/L with the original Abbott formulation in spite of little or no clinical evidence of cardiac ischemia. When the serum sample from that patient was reanalyzed with the enhanced formulation, the result fell to 0.4 μg/L. No additional sample was available for study. The second individual had little or no clinical evidence of cardiac ischemia, but had a cTnI of 25.3 μg/L with the original Abbott formulation and 0.8 μg/L with the modified assay. The results of pretreatment of the sample with the HBT before reanalysis with the original formulation were consistent with the presence of a HAAA. Our results suggest that the Beckman Access is less affected by RF interference than is the original formulation of the Abbott AxSYM system. The results further suggest that treatment of samples with HBT is not an appropriate step to eliminate RF interference in the original Abbott formulation. The introduction of the modified formulation of the cTnI assay kit by Abbott appears to have largely eliminated this problem. In two cases, the newer version did not reduce the cTnI value to 0 μg/L. This may, however, have reflected non-antibody interferents or cTnI in the sample. Furthermore, anecdotal evidence based on two samples suggests that the new formulation is more robust than the original with respect to interference by HAAAs. Comparison of cTnI results obtained with Beckman Access and Abbott AxSYM. ND, not determined. Comparison of cTnI results obtained with Beckman Access and Abbott AxSYM. ND, not determined. We thank Abbott Diagnostics (Missisauga, Ontario, Canada) and Beckman-Coulter (Missisauga, Ontario, Canada) for providing the cTnI kits.