Attention should be paid to false-positive results due to heterophilic antibodies interfering with Abbott high-sensitivity cardiac troponin I assay

Introduction

Cardio-specific troponins T and I (cTnT and cTnI) are considered as the most effective biomarkers of myocardial infarction [1]. In recent years, chemiluminescent microparticle immunoassay (CMIA) has been widely used for quantitative determination of troponin I [2, 3]. However, the testing can be affected by macrotroponin, heterophilic antibodies and other endogenous substances [4].

Several research papers reported the influence of heterophilic antibodies on the determination concentrations of high- or ultra-sensitive troponins [5, 6]. Though human/mouse chimeric antibodies in two-site immunoassays was proved to be more effective than mouse IgG for reducing interference from heterophilic antibodies, the false-positive rate was 2.8 % in healthy individuals. In this case-report, we described a further case of a 28-year-old female misdiagnosed with myocarditis due to falsely elevated troponin I with initial test result of 595.0 ng/L considered as critical value caused by heterophilic antibody, although the assay method applied a chimeric antibody.

Case presentation

On June 14th, 2023 (day 0), a 28-year-old female patient visited the emergency room due to palpitations and chest tightness. Heart rate of the patient was 85 beats per min, the blood pressure was 129/77 mmHg, the oxygen saturation was 99 %, and the respiratory rate was 20 breaths per minute. The patient denies past medical history, hepatitis history, tuberculosis history, surgery history, trauma history, and blood transfusion history. However, the high-sensitivity troponin I (hs-cTnI) in serum sharply increased at a concentration of 595.0 ng/L measured by Abbott Architect i2000SR Analyzer based on CMIA (Abbott Diagnostics, Longford, Ireland). Therefore, the patient was timely admitted to the hospital for treatment with suspected myocarditis. Immediately after admission, an electrocardiogram showed no ischemic changes, and echocardiography showed no local wall motion abnormalities. The left ventricular ejection fraction ranged from 57 to 62 %. Aortic dissection, acute pulmonary embolism as well as acute coronary syndrome were also ruled out.

On the morning of June 15th, 2023 (day 1), hs-cTnI remained significantly increased (500.1 ng/L). Subsequently, hs-cTnI were detected on June 18th (day 4), June 24th (day 10), June 27th (day 13), and June 28th, 2023 (day 14), and the concentrations were 462.6, 468.7, 462.5, 458.0 ng/L, respectively.

On June 28th (day 14), the laboratorian received feedback from an attending physician that the hs-cTnI levels in the patient did not match the clinical condition, and immediately investigate and analyze the possibility of false-positive results in the detection of hs-cTnI. The tested sample was free of hemolysis, lipemia or icterus, so potential interference factors was considered. Several common procedures were used for interference analysis: method comparison, serial dilution, and blocking reagent studies.

Firstly, the further used platforms (Siemens ADVIA Centaur XP and Roche Cobas e 801) resulted in low concentrations within reference values (Table 1). Secondly, we conducted a sample dilution test. In this case, the value of R² (coefficient of determination) from the serial dilution of the patient sample was 0.8517, while R² value from the control sample was 0.9985 by regressive analysis. The lack of dilution linearity was observed (Figure 1). Thirdly, to determine the kinds of endogenous interfering substances, rheumatoid factor, antinuclear antibody, and anti-extracted nuclear antigens autoantibodies from remaining samples of the patient were further tested. The concentration of rheumatoid factor was 5.6 IU/mL within the reference interval (0–20.0 IU/mL), determined by immunoturbidimetry using LABOSPECT008AS (Hitachi, Japan). The antinuclear antibody (ANA) IgG was negative (1:100), detected by indirect immunofluorescence assay (EUROIMMU, Germany). Anti-extracted nuclear antigens autoantibodies including nucleosome, dsDNA, Histone, SmD1, U1-snRNP, SS-A/Ro 60, SS-A/Ro 52, SS-B/La, Scl70, CENP-B, Jo-1, P0(RPP) were negative, detected by Line Immuno Assay (HUMAN, Germany). Therefore, the interferences from the above substances could be excluded.

Table 1:

Hs-cTnI/hs-cTnT concentrations determined by different platforms.

Regent kit, instrument (company)	Parameter	Result, ng/L
		Sample 1#	Sample 2#
Hs-cTnI, Architect i2000SR Analyzer (Abbott Diagnostics, Longford, Ireland)	hs-cTnI	462.5	458.0
Hs-cTnI, ADVIA Centaur XP Immunoassay Analyzer (Siemens Diagnostics, Germany)	hs-cTnI	8.3	7.0
Hs-TnT, Cobas e 801 Analyzer (Roche Diagnostics GmBH, Mannheim, Germany)	hs-cTnT	5.1	3.9

1. Sample 1# was from the patient on day 13; sample 2# was from the patient on day 14; Architect i2000SR Analyzer was from the author’s laboratory; ADVIA Centaur XP Immunoassay Analyzer and Cobas e 801 Analyzer were from the laboratory of Zigong Fourth People’s Hospital and West China Hospital of Sichuan University, respectively.

Figure 1:

The determination results of hs-TnI in the patient and control sample by serial dilution test. The patient and control sample were serially diluted with zero calibrator by a ratio of 1:1 (primary fold), 1:2, 1:4, 1:6, 1:8 and 1:10, and then hs-cTnI was tested respectively. The value of R² (coefficient of determination) was obtained by regressive analysis.

Furthermore, the heterophilic blocking tube (HBT, Scantibodies Laboratory, Inc.) was used to preprocess samples of the patient (day 13 and day 14) and five control samples from the remaining samples of other patients stored at 2–8 °C. The results showed that the HBT procedure caused a decrease in hs-cTnI concentrations of the patient samples within the reference values which suggests the presence of interference (Table 2).

Table 2:

Determination results of hs-cTnI before or after HBT pretreatment of the samples.

Samples	hs-cTnI before pretreatment, ng/L	hs-cTnI after pretreatment, ng/L	Recovery, %
Patient sample (day 14)	458.0	10.0	2.2
Patient sample (day 13)	462.5	8.9	1.9
Control sample 1#	1,538.7	1,667.7	108.4
Control sample 2#	284.4	294.2	103.4
Control sample 3#	192.6	186.1	96.6
Control sample 4#	99.7	95.8	96.1
Control sample 5#	75.3	71.6	95.1

1. All samples were determined by Architect i2000SR analyzer; HBT, heterophilic blocking tube; the control samples were from other patients in the author’s laboratory.

Discussion

In addition to acute myocardial infarction, elevated levels of troponin can occur in acute or chronic myocardial injury related diseases, such as stress cardiomyopathy and pulmonary embolism [7]. It is worth noting that in some cases, the increase in troponin may be due to analysis interference in laboratories. Analytical interference is a common problem that affects almost all immunoassays, producing false-positive or false-negative results [8].

One of the most common endogenous interferences is heterophilic antibody. The production of heterophilic antibodies in the body is caused by heterophilic antigens. Exposure to animals or ingestion of animal products, viral infections, autoimmune diseases, and receiving intravenous immunoglobulin therapy can all lead to the production of heterophilic antibodies. The frequency of this interference ranges between 0.1 and 3.1 % [9, 10].

In this case, the patient has extremely elevated concentrations of hs-cTnI without any clinical or laboratory evidence of infection, autoimmune diseases, skeletal muscle damage, or myocardial injury, which could highly suggest a false-positive result.

In order to confirm the false-positive results of hs-cTnI, we sent the sample to two other laboratories for determination of hs-cTnI and hs-cTnT using two different platforms. And the inconsistent results suggested that there were one or more factors affecting the measurement of hs-cTn I in our laboratory.

Heterophilic antibodies can significantly affect immunochemical assays, especially assays based on a sandwich principle. On an Architect i2000SR Analyzer, the principle for the determination of hs-cTnI is shown in Figure 2A. When heterophilic antibodies are present in the sample, a bridge between cTnI antibody coated magnetic beads and acridine ester labeled cTnI antibody is formed and then the luminescent sandwich complex can be further formed after washing, which results in a false-positive result (Figure 2B). However, the detection of hs-cTnI was not interfered using ADVIA Centaur XP immunoassay analyzer. Different results may result from the different antibodies used by the manufactures. According to the instructions of Abbott hs-cTnI assay, anti-cTnI coated magnetic beads were mouse antibodies against cTnI and acridine ester labeled anti-cTnI were human-mice chimeric antibodies. But on ADVIA Centaur XP immunoassay analyzer, the labeled antibodies were produced by recombinant goat anti-human cTnI. Based on this, it could be inferred that there were mouse-derived heterophilic antibodies present in the sample, which resulted in the recurrence of false elevated values of hs-cTnI.

Figure 2:

The schematic diagram of Abbott hs-cTnI immunoassay. (A) The determination principle of hs-cTnI based on a sandwich immunoassay with two steps. (B) The mechanism of a false-positive result of hs-cTnI caused by heterophilic antibody with two steps. cTnI antibody coated magnetic bead: mouse antibody against cTnI; acridine ester labeled anti-cTnI: human-mouse chimeric antibody.

Conclusions

In conclusion, the Architect assay used chimeric antibodies to overcome the interference with heterophilic antibodies, but nevertheless this interference can still occur. When laboratorians encounter a significant increase of hs-cTnI without clinical manifestations of myocardial injury, potential interference should be considered. It requires a close and strong collaboration between clinicians and laboratorians to manage these similar cases. To prevent false-positive results caused by interferences from being used in clinical practice, the clinicians are suggested to contact the laboratorians whenever the clinical picture, historical data and laboratory values are not conclusive. Meanwhile, laboratorians should take a series of appropriate methods to eliminate interference in a timely manner to obtain accurate results.