Irrespective of the many available guidelines and recommendations , the measurement of cardiac troponins (cTn) has now outperformed whatever other cardiac biomarker proposed so far , thus becoming the mainstay for diagnosing a vast array of myocardial injuries, especially acute myocardial infarction (AMI).
This paradigm shift was mainly driven by the relatively recent development of the so-called highly-sensitive (hs) cTn immunoassays, which are characterized by an impressive improvement of analytical sensitivity, which now permits detecting cTn concentrations over 10-times lower than those measurable with the former and more conventional “contemporary-sensitive” techniques , . The remarkable improvement of the analytical performance characterizing the hs-cTn methods has hence revolutionized the diagnostic approach to patients with suspect AMI. Firstly, it has been demonstrated that cTn values between the analytical sensitivity and the upper limit of the reference range (URL) of the assay may allow ruling out AMI with up to 0.99 negative predictive value, thus potentially improving patient triage and/or discharge from the emergency department (ED) . The much lower imprecision at values close to the analytical sensitivity of the method has then enabled developing and implementing fast track protocols for excluding or diagnosing AMI, especially in patients with non-ST elevation myocardial infarction (NSTEMI), in whom the electrocardiogram is, by definition, non-diagnostic .
Despite the great expectations for routine implementation of hs-cTn immunoassays in clinical practice, the commercialization of these methods has been much slower than expected . Beside the cardiac troponin T (cTnT) hs immunoassay, which has been available for quite a long time , the number of commercially available cardiac troponin I (cTnI) hs immunoassays is now slowly but progressively increasing . As enhanced analytical sensitivity, lower imprecision and high accuracy are the main reasons supporting the use of these innovative techniques in early triage of patients with suspected AMI, the on-field evaluation represents an unavoidable step ahead of developing diagnostic pathways and clinical use.
A Beckman Coulter hs-cTnI assay has been in development for more than 8 years , but has never been commercialized worldwide. During this time, the research and development department of the company has considerably reformulated and refined the prototype assay, and has recently released a novel version, the performance of which have never been tested so far. Therefore, this study was aimed to perform a comprehensive analytical evaluation of the novel Beckman Coulter Access hsTnI on the Access 2 immunoassay system.
Materials and methods
The novel Access hsTnI (Beckman Coulter Inc., Brea, CA, USA) is a fully-automated chemiluminescent two-site (“sandwich”) immunoassay, which has been developed to be used on Access 2 and DxI systems. Briefly, a sample is added to a reaction vessel together with monoclonal anti-cTnI antibodies conjugated with alkaline phosphatase and paramagnetic particles coated with another monoclonal anti-cTnI antibody. The cTnI present in the sample binds to the anti-cTnI antibody on the solid phase, whereas the conjugate antibody reacts with different antigenic sites on the cTnI protein moiety. After incubation in the reaction vessel, the solid phase with bound cTnI is held in a magnetic field, whereas the remaining unbound material is energetically washed. The chemiluminescent substrate Lumi-Phos 530 is then added to the vessel and the light triggered by the reaction is measured with a luminometer. The generation of light is directly proportional to the concentration of cTnI in the test sample, and the final cTnI concentration is derived from a multi-point calibration curve. The modifications of the novel Access hsTnI immunoassay compared to the former prototype method are: (i) introduction of a different solid phase, (ii) a different antibody pair (only one of the original antibodies of the former prototype assay has been retained) and (iii) use of a different buffer formulation (i.e. protein and surfactant concentrations), which has enabled to further increase the analytical sensitivity.
The evaluation of the analytical performance of the novel Beckman Coulter Access hsTnI on the immunochemistry platform Access 2 (Beckman Coulter Inc.) included the assessment of limit of blank (LOB), limit of detection (LOD), functional sensitivity, linearity, imprecision (within-run, between-run and total), calculation of 99th percentile URL, and comparison with two other commercial cTnI immunoassays. All measurements were performed using routine heparin-plasma samples. Specifically, primary evacuated blood tubes (Vacutest Kima, containing gel and lithium-heparin; reference number, 12550; Kima, Padova, Italy) referred by the ED to the laboratory of clinical chemistry and hematology of the University Hospital of Verona (Italy) for diagnostic cTn measurement were centrifuged at 2000×g for 10 min at room temperature. Lithium-heparin plasma was then divided in four aliquots of 0.4 mL each. The first aliquot was used for routine cTn measurements, whereas the other three aliquots were used for the study. The entire study was carried out by using the same calibration curve and an identical lot of reagents.
The calculation of LOB and LOD was performed according to Armbruster and Pry , as follows: [LOB]=[mean value]+1.645×[standard deviation, (SD)] of 20 replicates of Access 2 sample buffer; [LOD]=[LOB]+1.645×[SD] of 20 replicates of a routine plasma sample with the lowest measurable cTnI value (i.e. 0.15 ng/L). The functional sensitivity of the assay was defined as the lowest cTnI value which could be measured with an imprecision ≤10%. This value was calculated by measuring six consecutive 1:2 scalar dilutions in Access 2 sample buffer (i.e. from 1:2 to 1:64) of a routine plasma sample with a cTnI value of 18.35 ng/L. All dilutions were assayed in 10 consecutive runs and the imprecision was finally calculated for each dilution. A model fit was constructed to extrapolate the cTnI plasma concentration associated with 10% imprecision.
The linearity of the assay was evaluated by serially diluting at fixed ratios (0.1:9.9; 0.5:9.5; 1:9; 2:8; 3:7; 4:6; 5:5; 6:4; 7:3, 8:2; 9:1) a routine plasma sample with a high cTnI value (i.e. 4195 ng/L) with another routine plasma sample with a low cTnI value (i.e. 0.95 ng/mL). The serial dilutions were then measured in duplicate. The theoretical cTnI values were calculated from cTnI measured values of undiluted specimens and the linearity was estimated as Spearman’s correlation coefficient.
The within-run, between-run and total imprecision of Access hsTnI was calculated using three plasma pools prepared to display low (approx. 15 ng/L), intermediate (approx. 100 ng/L) and high (approx. 1590 ng/L) cTnI plasma concentrations. The within-run and between-run (i.e. between-day) imprecision was measured in 20 consecutive runs and 20 consecutive working days, respectively. The final data was reported in terms of coefficient of variation (CV%). The total imprecision was calculated with the formula proposed by Krouwer and Rabinowitz .
Calculation of the 99th percentile of the URL
The 99th percentile URL of the immunoassay was calculated according to joint International Federation of Clinical Chemistry (IFCC) and Clinical and Laboratory Standards Institute (CLSI) document EP28-A3C , by measuring blood donor samples received during three consecutive days from the local transfusion medicine service. All blood donors underwent accurate medical visit and comprehensive laboratory testing before blood donation, to rule out the presence of pathological conditions. Difference between genders was evaluated with the Mann-Whitney U-test; cTnI values were shown as median and interquartile range (IQR).
Comparison studies were performed using routine plasma samples referred for diagnostic assessment of cTnI. All samples were tested within 2 h from collection. The results of Access hsTnI were compared with those obtained on the same samples using the AccuTnI+3 immunoassay (Beckman Coulter Inc.) on the same Access 2 analyzer, as well as with Siemens Dimension Vista cTnI (Siemens Healthcare Diagnostics). The Beckman Coulter AccuTnI+3 assay is a paramagnetic particle, chemiluminescent immunoassay, characterized by LOB and functional sensitivity of 10 ng/L and 30 ng/L, respectively . The Siemens Dimension Vista cTnI is a contemporary-sensitive immunoassay based on luminescent oxygen channeling (LOCI) technology, characterized by LOB and functional sensitivity of 15 ng/L and 36 ng/L, respectively . The agreement of values obtained with Access hsTnI and those obtained with the other two immunoassays was assessed with the Mann-Whitney U-test for paired samples, Spearman’s correlation, Passing and Bablok regression analysis, whereas the absolute bias was calculated using Bland and Altman plot analysis.
The statistical analysis was performed using Analyse-it (Analyse-it Software Ltd, Leeds, UK). The complete study, based on pre-existing samples referred for diagnostic cTn testing, was carried out for purposes of ISO 15189:2012 accreditation, in accordance with the Declaration of Helsinki, under the terms of relevant local legislation. The study was approved by the local Institutional Review Board (University Hospital of Verona, Verona, Italy; SOPAV2, protocol number: 971CESC, date of approval: 25 July, 2016).
The LOB and LOD of Access hsTnI were 0.14 ng/L and 0.34 ng/L, respectively. The estimation of functional sensitivity is shown in Figure 1. A 10% imprecision corresponded to a cTnI concentration of 1.35 ng/L. The linearity of the Access hsTnI was excellent in the range of concentrations comprised between 0.95 and 4195 ng/L (Spearman’s correlation coefficient, 1.000; p<0.001). The equation of the Passing and Bablok regression was as follows: y=1.05x+15.3. The results of imprecision studies using lithium-heparin pools with low, medium and high cTnI values are summarized in Table 1. The within-run imprecision was between 2.17% and 2.94%, the between-run imprecision was between 4.61% and 5.44%, whereas the total analytical imprecision ranged between 5.44% and 6.09%.
The 99th percentile URL of Access hsTnI, calculated in samples of 175 ostensibly healthy blood donors (mean age, 36±12 years; 82 women and 93 men), was 15.8 ng/L. Notably, the assay imprecision at the 99th percentile URL was lower than 3.5%. Measurable cTnI values (i.e. higher than the LOB) were found in 173/175 subjects (98.9%), whereas values higher than the LOD were found in 170/175 subjects (97.1%). A significant association was found between cTnI values and age (Spearman’s correlation, 0.17; p=0.022), and the median concentration of cTnI was found to be nearly double in men (1.60 ng/L; IQR, 1.20–2.40 ng/L) than in women (0.90 ng/L; IQR, 0.60–1.40 ng/L; p=0.013).
As regards the comparison of cTnI values obtained with Access hsTnI and AccuTnI+3 in plasma samples with Access hsTnI values comprised between 2 and 19,995 ng/L, the difference was not found to be statistically significant (p=0.236). The Spearman’s correlation was 0.97 (95% CI, 0.96–0.98; p<0.001), whereas the equation of the Passing and Bablok regression analysis was [Access hsTnI] =1.10×[AccuTnI+3]−13.2. The mean bias estimated with Bland and Altman plot analysis was −9.3% (95% CI, from −18.3% to 0%) (Figure 2). The correlation (r=0.97; 95% CI, 0.96–0.98; p<0.001) and the bias (−10.8%; 95% CI, −1.6% to −20%) remained virtually unchanged after eliminating the 16 plasma samples with cTnI values below the analytical sensitivity of the AccuTnI+3 immunoassay.
As regards the comparison between Access hsTnI and Dimension Vista cTnI, although the difference of values was found to be statistically significant (p<0.001) and the mean bias was large (−55%; 95% CI, from −64% to −46%), an acceptable agreement was also observed. Specifically, the Spearman’s correlation was 0.95 (95% CI, 0.92–0.96; p<0.001) and the equation of the Passing and Bablok regression analysis was [Access hsTnI]=0.71×[Siemens Vista cTnI]−14.7. The comparison between Access AccuTnI+3 and Dimension Vista cTnI yielded the following data: bias, −48% (95% CI, −40% to −57%); Spearman’s correlation, 0.97 (95% CI, 0.96–0.98; p<0.001).
The evaluation of the analytical performance of Beckman Coulter Access hsTnI immunoassay shows remarkable advancements over the former prototype method. The novel hs technique displays substantially better values of LOB (0.14 vs. 1.03 ng/L), LOD (0.34 vs. 2.06 ng/L) and functional sensitivity (1.35 vs. 8.66 ng/L) , but also retains the excellent total precision of the previous immunoassay (5.4%–6.1% vs. 3.5%–7.1%). Further support to the improved analytical sensitivity of Access hsTnI is provided by the number of measurable values in samples collected from ostensibly healthy subjects, which has increased from 70% to 97% over the LOD, and from 92% to approximately 99% above the LOB, respectively. In accordance with other investigators’ findings for hscTn, we also found a significant correlation of Access hsTnI values with age and gender , , , .
Furthermore, the results of our evaluation confirm an acceptable agreement of values between Access hsTnI and the contemporary-sensitive AccuTnI+3 immunoassay, with a mean bias <10%. Unlike this finding, the comparison with Siemens Vista cTnI yielded less satisfactory data, showing a correlation of 0.95 (i.e. slightly worse than that obtained by Storti et al. between AccuTnI and Siemens Advia Centaur cTnI; r=0.98) , but a considerable bias (i.e. higher than 50%). This is not surprising, as it is well known that the current harmonization of cTnI immunoassays is poor and, even after recalibration, the variability remains as high as 40% . This has been particularly attributed to the use of cocktails of different cTnI antibodies reacting differently with both intact cTnI and the different degradations products released after AMI .
According to current recommendations, the ratio between the 99th URL and the functional sensitivity of cTn immunoassays should always be greater than 1 . Interestingly, the 99th percentile URL value of Access hsTnI (15.8 ng/L) was found to be nearly 12-fold higher than the functional sensitivity (i.e. 1.35 ng/L), so perfectly matching these indications . These excellent results are also confirmed by the fact that the assay imprecision at the 99th percentile URL was found to be approximately 3.5%, which is an even better performance than that of Abbott ARCHITECT STAT hs cTnI immunoassay (i.e. approx. 5%) obtained in a multicenter study .
One of the most important consequences of the excellent analytical and functional sensitivity of the novel Access hsTnI is the possibility to evaluate the use of this immunoassay for developing fast-track protocols , . Although further clinical studies are needed to validate this assumption, previous evidence showed that hs immunoassays capable to detect extremely low plasma cTnI concentrations may represent a paradigm shift for appropriate triage and management of patients with suspected acute coronary syndrome, from both a clinical and an economic perspective .
There are some limitations in this study. First, the analytical evaluation was carried out as a single-center study, and additional multicenter validation may be advisable for confirming these preliminary results. Then, the population used for calculating the 99th percentile only included 175 healthy subjects, so that this value should be further confirmed using a larger sample.
In conclusion, the results of our evaluation of the novel Beckman Coulter hsTnI indicate that the analytical performance of this fully-automated immunoassay is excellent, thus representing a substantial improvement over the former prototype.
Shah AS, Anand A, Sandoval Y, Lee KK, Smith SW, Adamson PD, et al. High-sensitivity cardiac troponin I at presentation in patients with suspected acute coronary syndrome: a cohort study. Lancet 2015;386:2481–8. PubMedWeb of ScienceCrossrefGoogle Scholar
Ferraro S, Dolci A, Panteghini M. Fast track protocols using highly sensitive troponin assays for ruling out and ruling in non-ST elevation acute coronary syndrome. Clin Chem Lab Med 2017;55:1683–9. Web of SciencePubMedGoogle Scholar
Giannitsis E, Kurz K, Hallermayer K, Jarausch J, Jaffe AS, Katus HA. Analytical validation of a high-sensitivity cardiac troponin T assay. Clin Chem 2010;56:254–61. Web of ScienceCrossrefPubMedGoogle Scholar
Kavsak PA, MacRae AR, Yerna MJ, Jaffe AS. Analytic and clinical utility of a next-generation, highly sensitive cardiac troponin I assay for early detection of myocardial injury. Clin Chem 2009;55:573–7. PubMedWeb of ScienceCrossrefGoogle Scholar
Clinical and Laboratory Standards Institute. Defining, Establishing, and Verifying Reference Intervals in the Clinical Laboratory, 3rd ed. Document EP28-A3C. CLSI, Wayne, PA, USA, 2010. Google Scholar
Lippi G, Dipalo M, Avanzini P, Formentini A, Aloe R. Analytical assessment of the Beckman Coulter Unicel DxI AccuTnI+3 immunoassay. Diagnosis 2014;1:195–7. Google Scholar
Arrebola MM, Lillo JA, Diez De Los Ríos MJ, Rodríguez M, Dayaldasani A, Yahyaoui R, et al. Analytical performance of a sensitive assay for cardiac troponin I with loci technology. Clin Biochem 2010;43:998–1002. Web of ScienceCrossrefPubMedGoogle Scholar
Clerico A, Fortunato A, Ripoli A, Prontera C, Zucchelli GC, Emdin M. Distribution of plasma cardiac troponin I values in healthy subjects: pathophysiological considerations. Clin Chem Lab Med 2008;46:804–8. PubMedWeb of ScienceGoogle Scholar
Zeller T, Ojeda F, Brunner FJ, Peitsmeyer P, Münzel T, Binder H, et al. High-sensitivity cardiac troponin I in the general population – defining reference populations for the determination of the 99th percentile in the Gutenberg Health Study. Clin Chem Lab Med 2015;53:699–706. Web of SciencePubMedGoogle Scholar
Krintus M, Kozinski M, Boudry P, Lackner K, Lefèvre G, Lennartz L, et al. Defining normality in a European multinational cohort: Critical factors influencing the 99th percentile upper reference limit for high sensitivity cardiac troponin I. Int J Cardiol 2015;187:256–63. CrossrefWeb of ScienceGoogle Scholar
Schofer N, Brunner FJ, Schlüter M, Ojeda F, Zeller T, Baldus S, et al. Gender-specific diagnostic performance of a new high-sensitivity cardiac troponin I assay for detection of acute myocardial infarction. Eur Heart J Acute Cardiovasc Care 2017;6:60–8. PubMedCrossrefWeb of ScienceGoogle Scholar
Storti S, Masotti S, Prontera C, Franzini M, Buzzi P, Casagranda I, et al. Evaluation of analytical performance and comparison of clinical results of the new generation method AccuTnI+3 for the measurement of cardiac troponin I using both patients and quality control plasma samples. Clin Chim Acta 2015;451:129–34. Web of SciencePubMedCrossrefGoogle Scholar
Clerico A, Ripoli A, Masotti S, Prontera C, Storti S, Fortunato A, et al. Pilot study on harmonization of cardiac troponin I immunoassays using patients and quality control plasma samples. On behalf of the Italian Section of the European Ligand Assay Society (ELAS) and of the Study Group on Cardiovascular Biomarkers of the Società Italiana di Biochimica Clinica (SIBioC). Clin Chim Acta 2016;456:42–8. Google Scholar
Panteghini M, Pagani F, Yeo KT, Apple FS, Christenson RH, Dati F, et al. Evaluation of imprecision for cardiac troponin assays at low-range concentrations. Clin Chem 2004;50:327–32. CrossrefPubMedGoogle Scholar
Krintus M, Kozinski M, Boudry P, Capell NE, Köller U, Lackner K, et al. European multicenter analytical evaluation of the Abbott ARCHITECT STAT highly sensitive troponin I immunoassay. Clin. Chem Lab Med 2014;52:1657–65. Google Scholar
Casagranda I, Cavazza M, Clerico A, Galvani M, Ottani F, Zaninotto M, et al. Proposal for the use in emergency departments of cardiac troponins measured with the latest generation methods in patients with suspected acute coronary syndrome without persistent ST-segment elevation. Clin Chem Lab Med 2013;51:1727–37. Web of SciencePubMedGoogle Scholar
About the article
Published Online: 2017-07-12
Published in Print: 2017-11-27
Author contributions: All the authors have accepted responsibility for the entire content of this submitted manuscript and approved submission.
Research funding: None.
Employment or leadership: None declared.
Honorarium: None declared.
Competing interests: The reagents used in this study were provided by Beckman Coulter, but the company played no role in the study design; in the collection, analysis, and interpretation of data; in the writing of the report; or in the decision to submit the report for publication.