The fundamental role of the clinical laboratory is to provide accurate test results, thus guiding patient care when clinical history and physical exam are insufficient. It is no wonder then that, when test results are called into question, it can lead to anxiety and mistrust of the laboratory on the part of both clinicians and patients. In many cases, however, questionable test results may not actually represent “laboratory error,” but rather unique, patient-specific idiosyncrasies in laboratory testing. In such scenarios, laboratory personnel have an opportunity to become vital members of the clinical team by investigating the mechanism of false or questionable results, suggesting alternative diagnostic methods, and ultimately guiding appropriate patient care. In this issue of The Journal of Applied Laboratory Medicine, Mattman et al. describe such a scenario in their article entitled “Grave Clinicopathologic Correlation: A Case of Hyperthyroxemia” (1). In the case they describe, an immunoassay interference affected multiple patient results in a way that suggested a consistent, but incorrect, diagnosis. Consultation with the clinical laboratory not only elucidated the underlying mechanism but also corrected the results to ensure that the patient was appropriately treated.
Mattman et al. describe the case of a young woman whose thyroid function testing, including an increased free thyroxine and borderline low thyroid-stimulating hormone (TSH)2, was suggestive of hyperthyroxemia and Graves' disease. However, her clinical signs and symptoms were those of a euthyroid patient, prompting clinicians to question the results. This scenario, in which the first warning of an analytical issue resulted from clinical correlation by an astute physician, illustrates an important mechanism by which the laboratory becomes aware of potential problems, and it emphasizes the importance of a close working relationship between clinicians and the laboratory. A number of standard methods to identify potential interferences can be undertaken under these circumstances, including serial dilution and linearity assessment, recovery studies, heterophile binding tubes, polyethylene glycol precipitation, or immunoadsorption with protein A/G (2). In this case, retesting using an alternative method on a Beckman Coulter platform resulted in analyte measurements in keeping with the patient's clinical status, leading to suspicion that the interference was specific to a component of the Roche immunoassay initially used. Of note, the Roche assays in question rely on a streptavidin-coated microparticle and biotinylated capture antibody or reagent analyte for sandwich and competitive immunoassays, respectively. This common strategy has been widely used for laboratory testing because multiple washing steps will not affect the high-affinity interaction and because molecular biological activity and immunologic specificity are not altered by biotinylation (3). However, as has been previously reported, patients may rarely develop endogenous anti-streptavidin antibodies that interfere with streptavidin-biotin–based immunoassays (4, 5). Knowledge of these previously reported findings perhaps influenced the laboratory clinicians in this case to pursue not only heterophilic antibody testing, but also pretreatment with streptavidin-coated magnetic beads, a much less common cause of immunoassay interference than heterophile antibodies (6). Interestingly, the free thyroxine (fT4) and TSH assays in question use competitive and sandwich methods, respectively, providing a mechanism by which a common method of interference could lead to reciprocal changes that suggest an incorrect diagnosis.
As illustrated in this case, the use of alternative testing platforms to investigate potential interferences has been a valuable and rapid way to identify obvious inconsistencies in assay-specific results. While the increasingly popular trend to adopt common platforms across hospital systems has well-known financial and standardization advantages, it is worth pointing out that diversity in the laboratory testing “ecosystem,” whether within a single system or in collaboration with reference laboratories, can be valuable in working up interferences. Ready access to nonstandardized testing menus and technology platforms across institutions allows for greater diversity of methodologies that may benefit certain patients.
In the present study, the authors conclude that an anti-streptavidin antibody was the most likely cause of analytical interference, based on the correction obtained by preincubation with streptavidin-coated beads. One confounding aspect of this case, however, is the set of normal results reported on Beckman Coulter fT4 and free triiodothyronine (fT3) assays that also used streptavidin-coated beads. In this case, it is necessary to postulate that the magnitude or significance of the observed interference is influenced by details of the mixing kinetics, number of available streptavidin molecules, or some other unknown factor. This is consistent with the previous report by Peltier et al. (5) but contrasts to the Rulander study (4), which showed that an interference correlated with the use of streptavidin across multiple analytical platforms. It is also interesting to note that Mattman et al. show normalization of patient results by a heterophile blocking, while in previous cases, heterophile blocking resulted in only partial correction of levels (4) or no significant change (5) in hormonal or antibody levels. Nevertheless, the authors do in fact demonstrate the same pattern of interference seen in the previously reported cases of anti-streptavidin antibodies, i.e., falsely low concentrations with noncompetitive assays and falsely high concentration levels with competitive immunoassays. This result highlights the inconsistent nature of antibody interferences, where no single strategy for laboratory workup (e.g., heterophile depletion) is sufficient to detect and/or correct all cases.
As the authors note, other sources of analytical error, such as biotin supplementation or endogenous antibodies to ruthenium, may affect the assays under consideration in this report. Additionally, antibiotin and antibiotinylated protein antibodies have been detected at rates of 3%–10% in adult subjects (7–9), thus providing another theoretical etiology of assay interference in streptavidin-biotin–based immunoassays.
As previously noted, while Mattman et al. focused on an effect in assays from one manufacturer, previous work demonstrated that multiple immunoassay platforms using streptavidin may show similar interferences due to endogenous anti-streptavidin antibodies. We therefore feel that the term “manufacturer-specific” analytical interferences, as used by the authors, may not always be robust enough when educating the clinicians and patients about the etiology of inaccurate laboratory values. Notwithstanding the fact that the antibody described in this case did not appear to affect another streptavidin-containing assay, a more encompassing term such as “methodology-specific” may sometimes be more appropriate. When possible and appropriate, definitively identifying and characterizing the culprit antibody can provide significant clinical value by formally documenting for the patient and provider the range of assays that may be affected.
This case report by Mattman et al. is yet another important example of how joint clinicopathologic correlation is an indispensable tool to prevent inappropriate patient care. Beyond the lab's traditional communication with an ordering physician, a formal letter or other documentation for the patient themselves can be provided to communicate potential effects of an interfering antibody on other laboratory testing and to suggest that future medical providers be informed because of the risk of inappropriate treatment based on false results. With comprehensive knowledge of assay methodology and the potential pitfalls of those assays, clinical laboratorians can quell physician and patient anxiety regarding false laboratory results and appropriately guide both immediate and long-term patient care.
- Received August 25, 2016.
- Accepted August 30, 2016.
- © 2016 American Association for Clinical Chemistry