Macrotroponin—Analytical Anomaly or Clinical Confounder

The term macrotroponin is usually applied to a complex of cardiac troponin (cTn) with an immunoglobulin (1). The resulting complex is cleared differently from uncomplexed cTn. The presence of macrotroponin can cause an artifactual increase of cTn values during routine measurement resulting in an actual false-positive cTn value. This complex may be with cTnT, cTnI, or a more elaborate complex with troponin C (I-C) or cTnT (T-I-C) (2–4). Typical presentation is a call from a confused clinician (5). A patient with an increased troponin has undergone a range of imaging studies, and no evidence of a myocardial infarction or anything consistent with increased cTn has been found. A detective story unfolds, and the laboratory consults a clinical chemistry Poirot who knows the relevant steps. Macrotroponin is revealed as the analytical villain. The prevalence of macrotroponin has been a mystery. Typically, occurrences of macrotroponin are conveyed as case reports, but to date, the prevalence and clinical impact have not been systematically investigated. In this issue of Clinical Chemistry, Lam et al. addressed this evidence gap (6). Building on a previous study, they systematically screened a community population with a cTn value above the upper reference limit (URL) for macrotroponin. Their results were somewhat startling: 99 of the 188 participants (52.6%) were found to have a macrotroponin. In a previous study from the same group, the prevalence was also found to be comparably high: 123 of the 223 participants (55.2%) in a similar community population were also shown to have macrotroponin, although participants were not selected among those with increased troponin (4). The prevalence of antitroponin antibodies (which may not necessarily produce macrotroponin) has been reported as 0.0% to 12.7% in healthy individuals and up to 13.4% post myocardial infarction (7). Other studies have estimated the prevalence by examining discordance between different troponin methods and reported the prevalence of macrotroponin as 54/1074 (5.0%) for high-sensitivity (hs) cTnI from Abbott Architect vs hs cTnI from Beckman Access (8) and 12/184 (6.5%) for hs cTnI from Abbott Architect vs cTnI from Vitros, although sample volume constraints only allowed confirmation in 7/9 (9). The higher rate of macrotroponin detection reported here probably reflects differences in study design, although subject selection may also play a role. Identification of macrotroponin by screening discordant results between methods will be systematically biased to underdetect compared to primary sample screening for macrotroponin. In the current study (6), 4641 cTnI measurements were performed with 311 increased results, of which 99/188 (52.7%) were confirmed as macrotroponin. Overall, an increase due to macrotroponin that manifests as a troponin value exceeding the URL can be estimated at 163/4641 (3.5%). In an unselected population (4), 54/223 (24.2%) were below the URL, 160/223 (71.7%) were from 1× to 10× the URL (median approximately twice the URL), and 9/223 (4.0%) were above the URL. The main consequence of macrotroponin is the produced increase in troponin values. The impact of macrotroponin on assay performance will be on assay specificity rather than sensitivity. The other factor to consider is the temporal stability of values. Macrotroponin typically presents with persistent increase in troponin concentrations and seems to show low short- and medium-term variations (5, 8, 9). Serial measurements therefore show elevated troponin but without the kinetic changes associated with acute cTn release. Different cTn methods have different rates of macrotroponin detection. There may also be a substantial difference when cTnT is subsequently remeasured on the same sample. Typically when a cTnI macrotroponin is present, it is not reflected in the cTnT result although there may be a T-I-C complex macrotroponin present (2). The discordances between cTnI and cTnT have an immediate and useful consequence. If a cTnI macrotroponin is suspected, a rapid exclusion is possible by measurement with a different cTnI assay but ideally by measurement with cTnT. The corollary is probably also true, but a systematic study has not been performed. Measurement of troponin is affected by hemolysis, phosphorylation, oxidation reduction, troponin fragmentation, and epitope masking. Macrotroponin is another factor to add to the list of potential analytical interferences for method evaluation studies. Discordance between assays has been used to estimate the frequency of macrotroponin occurrence but is not a direct measure. Assay manufacturers should attempt to obtain a panel of macrotroponin samples to estimate the degree of interference in their assay. The number of discordant results between assays will be affected by the prevalence of macrotroponin, the type of macrotroponin, and the ability of the assays under investigation to detect the type of macrotroponin present. In method comparison studies, discordance between methods will show as the scatter around the line of agreement and the number of outliers (9). The current study (6) shows clear differences when analytical performance is compared by receiver operating characteristic curve analysis. Comparison of the area under the curve (AUC) shows a significant difference overall between the assays according to the degree of macrotroponin detection. Impact on AUC is inversely proportional to the degree of detection of macrotroponin. When split into populations with and without macrotroponin, assays with a low rate of macrotroponin detection show a significant difference in the AUC, reflecting reduced specificity of the macrotroponin sensitive assay used to make the initial diagnostic selection. Analytical performance is not particularly affected by very high or very low values but is most affected by classification of results in the intermediate troponin range. The majority of increases are in the low range (1–10× URL), which will explain the impact on the AUC. The effect will also be exaggerated by using a population with a very low prior probability of acute coronary syndrome (ACS). To put these findings in context, diagnostic studies on cTn assays to date have not shown significant differences in diagnostic performance between assays in the ACS population, but macrotroponin will undoubtedly contribute to the difference between them at low troponin values. The need to exclude samples containing macrotroponin is clearly important if harmonization studies using sample pools is undertaken to avoid variable false-positive increase across methods. In reference interval studies, troponin values are disproportionally affected by the tail around the upper end of the distribution. The likelihood that macrotroponin contributes to this appears to be high, and this must be added to the criteria for exclusion. Pragmatically this would be best achieved by testing values in the upper 10% of the distribution. Figure 1 in Lam et al. (6) shows lower cTnI values in the non-ACS group when assays are less affected by macrotroponin or macrotroponin is not present. The findings from Lam et al. (6) need to be seen in the context of current clinical practice, which mandates serial sampling for the diagnosis of myocardial infarction. Patients with macrotroponin without myocardial infarction are very unlikely to show a significant change in troponin (delta troponin) due the macrotroponin itself. Indeed, it seems, if anything, the opposite is the case (8). Currently recommended rapid diagnostic algorithms combine admission testing with serial testing (10). Rule-out admission testing uses a low threshold and in theory would be compromised by macrotroponin, but in practice, this does not seem to occur. Rule-in, however, will be compromised as the thresholds are not as high as 10× the URL, so a significant proportion of immediately admitted patients might have a macrotroponin and no ACS. Indeed, the specificity of rule in algorithms is poor. Lam et al. (6) support the value of serial testing even when troponin is increased and, most important, that the troponin result must always be interpreted as part of the clinical picture. It is often overlooked that current ACS management guidelines do not require immediate (<2 h) intervention (10). There is always time to repeat the troponin. Lam et al. also may explain why there are so many patients with modest troponin increases in the emergency department with a multifactorial or indeterminate cause of troponin increase. Outcome studies in the asymptomatic population support the role of cTn measurements as a prognostic risk marker for long-term risk (11). Depending on the method, there is a risk of inappropriate classification of patients into a higher risk group (12). Methods selected for outcome studies need to avoid macrotroponin interference. Lam et al. (6) examined clinical outcomes in the population studied based on a single sample case. No adverse outcomes were seen, but hospital stay was longer in macrotroponin patients without ACS. Recently, a case-based review on interferences with cardiac biomarkers was able to identify clinical implications in 45% of cases of actual false-positive increases (13). The most common intervention was cardiac imaging including catheterization, and unnecessary interventions occurred in 10% of cases. The commonest causes were, in order, outliers (fliers), heterophile antibodies, and macrotroponin (22/222 case reports, 10 with clinical implications). Again, repeat the measurement (14)! So, what is the implication of macrotroponin for the laboratory and clinician? A rubeum allec? As Zhou Enlai is (incorrectly) reputed to have remarked when asked to comment on the significance of the French Revolution (of 1789), it’s too early to tell. This discussion underlies the importance of clinical assessment of the patient, serial sampling, and liaison with the lab. At the moment, it is analytically fascinating and potentially a confounder but apparently clinically inconvenient rather than catastrophic. The current rapid diagnostic algorithms require serial confirmation of increased values but should be modified so that all increased values require repeat. cTn, cardiac troponin; cTnI, cardiac troponin I; cTnT, cardiac troponin T; URL, upper reference limit; hs, high-sensitivity; AUC, area under the curve; ACS, acute coronary syndrome. All authors confirmed they have contributed to the intellectual content of this paper and have met the following 4 requirements: (a) significant contributions to the conception and design, acquisition of data, or analysis and interpretation of data; (b) drafting or revising the article for intellectual content; (c) final approval of the published article; and (d) agreement to be accountable for all aspects of the article thus ensuring that questions related to the accuracy or integrity of any part of the article are appropriately investigated and resolved. Upon manuscript submission, all authors completed the author disclosure form. Disclosures and/or potential conflicts of interest: P.O. Collinson, The Journal of Applied Laboratory Medicine, AACC. P.O. Collinson, consultant to IFCC Committee on Clinical Applications of Cardiac Bio-Markers (C-CB), Chair Clinical Advisory Board Radiometer, Psyros Diagnostics. None declared. None declared. None declared. None declared. None declared.