Impact of antibody specificity and calibration material on the measure of agreement between methods for cardiac troponin I.
After myocardial injury, there is a progressive rise in circulating immunoreactive cTnI. Empirically, the in vivo t112 of purified cTnI in dogs was found to be ~67 min (10), which agrees with the renal clearance of a protein possessing this molecular weight. Still unclear, however, is the [t.sub.1/2] of the immunoreactive form released from damaged myocardial tissue. Approximate calculations, generated from cardiopulmonary bypass studies in which reproducible injuries are made, suggested a much longer [t.sub.1/2], of several hours (e.g., 7 h), which supported the hypothesis that the major circulating form is complexed. Recent characterization studies have verified that circulating cTnI, released following myocardial damage, represents a combination of intact complex and its separate component parts (19-22). Other potentially important circulating forms of cTnI may also exist (in particular, phosphorylated, dephosphorylated, reduced, oxidized, and/or proteolytically degraded). For example, cTnI can be phosphorylated on two adjacent N-terminal serine residues (Ser 22 and 23), which have important regulatory roles in muscle contraction (23). These phosphorylated forms (apo-, mono-, and di-), which can be distinguished immunologically (24), influence the stability of the heterotrimer and its resistance to proteolytic degradation by the calcium-dependent protease [mu]-calpain (25). Similarly, the two cTnI cysteine residues have been shown to exist in either the reduced or oxidized forms, the latter producing an intrachain disulfide bond (21,22). Although the importance of these factors with respect to the immunoreactivity detected by different cTnI immunoassays has not been fully established, understanding these differences could be important when further characterizing current methods or developing new methods.
To compare and understand cTnI values obtained from four different cTnI methods, we first report the evaluation and analytical validation of immunoreagents developed for a noncommercial application that quantifies cTnI using the Dade Behring aca [R] plus immunoassay system. The assay is based on chromium dioxide particle technology (26) and uses two mAbs recognizing different epitopes on cTnI than the other commercial cTnI methods. After analytical validation of these immunoreagents, individual patient sample values were obtained and compared in four cTnI immunoassays: the Stratus [R] II assay, the Opus [R] II assay, the Access [R] assay, and the noncommercial Dade Behring aca plus cTnI assay. Finally, the extent of agreement between patient sample values obtained with these methods were compared after each method was recalibrated using human serum pools with cTnI values assigned using the Stratus II cTnI assay.
Materials and Methods
Reagents for the aca plus assay were supplied as a gift from Dade Behring (Newark, DE) as were the Stratus II cTnI assay kits. Analyses on the Sanofi Access analyzer were performed with the assistance of Mike Kemp and James Hooper at the Royal Brompton Hospital, London. Gurge Phull of Dade Behring UK gave us access to the Opus II analyzer to perform our analyses using this method. Reagents for the Opus and Access analyzers were purchased commercially and used according to the manufacturers' instructions.
aca plus CTNI ASSAY IMMUNOREAGENTS
Two mAbs (designated 144B3.63.5 and 144B5.2.1) were generated for, and used in, the prototype aca plus assay described in this study. Initially, to confirm the appropriate cTnI specificity of these reagents, immunoblot detection with purified mAbs 144B3.63.5, 144B5.2.1, and 3I-35 [known to recognize both cTnI and skeletal TnI (skTnI); Spectral Diagnostics] as the primary antibodies was performed. Briefly, either purified human cTnI or human skTnI (Biodesign International) was loaded into the large well of individual two-lane 4-20% gradient sodium dodecyl sulfate-polyacrylamide gel electrophoresis reducing gels (Novex Electrophoresis). Prestained protein markers (ISS) were electrophoresed in the second smaller well. After electrophoresis at a constant 125V for ~2 h, the proteins were transferred to nitrocellulose (Bio-Rad) at 150 mA for 2 h, using a TE22 Transphor Electrophoresis Unit (Hoeffer Scientific Instruments). Immunoblotting was performed with an Immunetics Miniblotter 25 such that 1 [micro]g of antigen was present in each lane. Retained primary mAbs were detected with alkaline phosphatase-conjugated goat anti-mouse antibodies followed by the addition of 5-bromo-4-chloro-3-indolylphosphate/nitro blue tetrazolium substrate.
CALIBRATION AND QUALITY CONTROL
Two types of material were used. The first (calibrator I) was obtained from the manufacturer and was specific to the test methodology. The second (calibrator II) was prepared from human serum samples that were selected on the basis of the creatine kinase MB isoenzyme concentration. Individual samples were pooled to five different target concentrations, and each pool was assigned a cTnI value, using the Stratus II cTnI method. Three quality-control (QC) pools were prepared using a partially purified cTnI fraction from human heart extracts (SCIPAC, Sittingbourne, UK). This material was added into a human serum pool (prescreened for low background cTnI concentrations using the Stratus II method) at ~1, 35, and 60 mg/L. Materials were stored aliquoted at -20[degrees]C.
Passing and Bablock analyses (27) were performed using the Astute Statistical Package (DDU Software, University of Leeds), with other statistical analyses performed using Statview II for the Macintosh (Abacus Concepts).
Assay range, hook effect, and detection limit. The assay working range and hook effect were determined by examination of the assay signal using increasing amounts of partially purified human cTnI, up to 1000 [micro]g/L. The assay detection limit was defined as two times the SD of 20 replicates of the zero calibrator.
Imprecision. The three human serum QC pools containing 1, 35, and 60 [micro]g/L cTnI that were prepared and frozen in aliquots were used to assess imprecision. The within-day imprecision (CV) was determined by analysis of 20 aliquots of each pool randomized in a single analytical run. The between-day CV was determined by ANOVA using the NCCLS protocol EP5-T with duplicate analysis of each pool performed on 20 nonconsecutive days.
Analytical recovery and parallelism. For recovery experiments, a human sample with high cTnI was added to 16 low-concentration cTnI patient samples to increase the baseline concentration by ~5 [micro]g/L, using one volume of high-concentration sample to nine volumes of low-concentration sample. All samples were analyzed by both the Stratus and aca plus methods. To determine parallelism, a series of 10 sera with increased cTnI concentrations, obtained from the routine laboratory, were diluted with the calibrator II zero calibrator to give 75%, 50%, 25%, and 12.5% of the original. Each dilution was analyzed in duplicate with the aca plus application. Linearity was assessed by linear regression analysis.
Method comparison. A total of 72 samples were analyzed in duplicate for cTnI using the aca plus, Stratus, Access, and Opus methods, each performed according to the manufacturer's instructions.
Interference. Two approaches were taken to study the potential interference from skTnI. In the first approach, purified skTnI (Spectral Diagnostics) was added to a human serum calibrator base pool at 1000 and 10 000 [micro]g/L. In the second approach, purified skTnI at 10 and 100 [micro]g/L was added to five patient samples, each with high endogenous cTnI. The cTnI concentrations were measured in all samples by both the aca plus and Stratus cTnI assays.
Interference from anticoagulants was studied by analysis of matched serum and citrated plasma from 12 patients with increased cTnI.
A typical aca plus calibration curve was demonstrated for 0-90 [micro]g/L cTnI, with an absorbance change equivalent to 1800 mA across this concentration range. No hook effect was observed up to 1000 [micro]g/L cTnI (data not shown), and the assay sensitivity was calculated as 0.14 [micro]g/L (2 X SD at zero cTnI). The intraassay imprecision was assessed over 20 days, with the results shown in Table 1. The total imprecision was ~5% above 1 [micro]g/L.
No significant cross-reaction was found in the aca plus method when up to 10 000 [micro]g/L skTnI added into nondiseased human serum was run as a sample (Fig. 1). In addition, Western blot analysis of the two antibodies 144B3.63.5 and 144B5.2.1 (both Ig[G.sub.1]) is shown in Fig. 2. Neither antibody recognized purified skTnI, but 144B3.63.5 recognized several low-molecular mass bands at 22, and 8 kDa, and a higher molecular mass band at ~40 kDa in the purified cTnI blot (Fig. 2A). Western blotting was also performed with both antibodies and troponin T together with an anti-troponin mAb as a control; neither of the anti-cTnI mAbs recognized troponin T (data not shown).
Finally, no interference from citrate as an anticoagulant was demonstrated. No significant difference between 12 matched serum and citrate plasma samples was observed, P >0.1 using a paired t-test (data not shown). The experimental protocol used in this study yielded a mean analytical recovery of 96.2% [+ or -] 17.3% on the aca plus analyzer and 93.9% [+ or -] 13.6% on the Stratus II system. Finally, no lack of parallelism was demonstrated, with the slope of the regression line not significantly different from 1 (P >0.05).
Method comparison data are shown in Fig. 3 along with regression analyses using the Passing and Bablock approach (27). Correlation coefficients were calculated separately, using standard linear regression (Table 2). Fig. 4 shows the difference plots for the same comparisons using Bland-Altman scatter analysis (28). To demonstrate the possible clinical implications of the variability between the experimental aca plus and Stratus data, we ranked the results as above or below the cutoff value of 1.5 [micro]g/L: for the 9 samples for which the aca plus results were <1.5 [micro]g/L, the Stratus assay results were <1.5 [micro]g/L for 6 samples and >1.5 [micro]g/L for 3 samples; for the 63 samples for which the aca plus results were >1.5 [micro]g/L, the Stratus assay results were <1.5 [micro]g/L for 5 samples and >1.5 [micro]g/L for 58 samples. The method comparison data were also analyzed using the same patient samples after each method was calibrated using calibrator II, as shown in Table 2. Within-assay imprecision profiles were calculated for all methods from the 72 duplicate measurements used in the method comparisons, as shown in Fig. 5.
We determined and compared patient sample cTnI values using four cTnI immunoassays, three commercial and one research, after initially calibrating each assay with the method-specific calibrator (calibrator I) and again after calibrating each assay with a patient sample-derived serum pool that had cTnI values assigned with the Stratus cTnI assay (calibrator 11). Before we performed these comparisons, however, we developed two noncommercial specific-specific mAbs for a prototype assay on the aca plus instrument. Extensive evaluation and analytical validation verified the utility of the research-only aca plus specific method, a precise and specific assay to quantify specific in serum and citrated plasma. These new antibodies, both individually and as a matched pair, did not recognize skTnI in Western blotting, nor did skTnI cross-react in the immunoassay (Figs. 1 and 2). In addition, the antibodies did not detect troponin T, which was at one stage questioned because of the appearance of a band in the blotting experiments at a position close to the 39-kDa marker. The aca plus specific assay had a working range of 0-90 [micro]g/L and a detection limit of 0.14 [micro]g/L. In addition, the aca plus application and the Stratus method had similar imprecision profiles (Fig. 5).
[FIGURE 1 OMITTED]
[FIGURE 2 OMITTED]
[FIGURE 3 OMITTED]
As part of the analytical validation of the aca plus cTnI assay, we performed a series of recovery experiments using a human serum pool containing a high concentration of endogenous cTnI, which was added to separate human serum samples. Because there was little published information concerning analytical recovery, we measured the same series of samples using the Stratus assay as the reference method. There was no significant difference in the mean recovery between the two methods: 96.2% [+ or -] 17.2% for the aca plus application and 93.9% [+ or -] 13.6% for the Stratus method. The SDs for both sets of data, however, indicated significant sample-to-sample variability in recovery. This variability in recovery might be attributed to differences (a) in the stability of the added specific in the different samples, as reported by Waskiewicz et al. (29) and Morjana (30); (b) in the relative concentrations of the cTnI forms in the individual samples (31); or (c) in the epitope specificity of the antibodies (32, 33). Katrukha et al. (32), in a careful study using three different combinations of paired mAbs shown to recognize different fragments of cTnI, demonstrated that the protein in serum is susceptible to proteolysis. Similar experiments on the protein extracted from tissue indicated that the central fragment of the protein (residues 30-110) was the most stable. The data also demonstrated that the degree of instability observed was method dependent. Katrukha et al. (19) used 15 mAbs against cTnI in a sandwich assay (a total of 196 combinations) to study the TnI released into the bloodstream after a myocardial infarction and found that the majority was released as a complex with troponin C. The authors showed that combination of TnI with cardiac troponin C led to a reduction in the interaction of the antibody pairs, indicating that intermethod variability of results could be attributed to the region of the cTnI recognized by the antibodies and the proportion of free and complexed protein present in the serum. Nevertheless, this observation was consistent with the recovery data, yet would still enable the samples to be diluted out with acceptable linearity.
[FIGURE 4 OMITTED]
Method comparison between the aca plus and the Stratus assays showed excellent agreement, although there was significant scatter around the regression line (Fig. 3A). This scatter, when viewed in the context of a cutoff value of 10 [micro]g/L for the Stratus assay as the reference procedure, indicates a sensitivity and a specificity of 79% and 86%, respectively, for the aca plus experimental method. Similar scatter was also evident when the Stratus method was compared with either the Opus or the Access methods (Fig. 3, B and C). When each assay was calibrated with the method-specific calibrator (calibrator I), the aca plus and Opus methods each showed good correlation with the Stratus assay. The aca plus assay showed good agreement (slope, 0.90) compared with the Stratus assay, whereas the Opus assay demonstrated a slope of 1.44 compared with the Stratus assay. A significant correlation slope bias, however, was observed between the Access and Stratus methods (slope, 0.07). Similar slope differences between the Stratus and Opus and the Stratus and Access methods were observed in a recently published abstract (34). Potential explanations for these slope differences are discussed below.
[FIGURE 5 OMITTED]
Unlike Bhayana et al. (34), however, we extended our study to compare the patient sample results from the same samples, obtained after calibrating each method with calibrator II. As summarized in Table 2, the correlation slopes between the Stratus II and Opus II cTnI methods and Stratus II and aca plus methods improved to ~1.0. More importantly, the correlation slope between the Stratus II and Access analyzers improved significantly, to 1.16. Although the correlation slopes improved after each method was recalibrated with a Stratus cTnI assay-assigned patient serum pool, the scatter around the correlation lines was unchanged and still significant.
Bland-Altman difference plots enabled a direct comparison of the bias and exploration of the concentration-dependent nature of that bias (28). These plots enable differences between methods to be evaluated, with respect to bias and scatter, in a more robust manner than regression analysis, even Passing and Bablock. In this study, in which the measure of agreement between several methods was sought, it was the only analysis that could facilitate this. If there is no systematic bias between methods, then the mean difference will be zero. If there is a fixed systematic bias between methods, then the mean difference will be more or less than zero. With a concentration-dependent (slope) bias between methods, the data scatter will show a nonrandom distribution with a slope significantly different to zero. Fig. 4A clearly shows a large degree of scatter around zero, which in most cases is random. However, there is a high bias for the Access assay relative to the others. Calibration of each method with calibrator II largely eliminated this bias, but the scatter essentially remained. In agreement with recent studies, these method comparison results suggest that a common reference preparation would improve the agreement between the available cTnI methods (34,35) but have little effect on the scatter.
The data around the regression lines (Fig. 4B) showed significant scatter, implying that the immunoreaction in each assay is different. Each assay has inherent precautions against nonspecific sample interferences and has been shown to be unaffected by such. In addition, these assays have been reported to be free of interference from skTnI isoforms. The most likely cause of method-related differences, therefore, is that each assay (i.e., antibody pair) recognizes the sample immunoreactive cTnI differently. The antibody pairs used in each method do not react with the same epitopes on cTnI. According to published results (8, 9,36), the Opus, Access, and Stratus assays recognize distinct epitopes N-terminal to the TnC binding site. Epitope mapping has determined that the aca plus antibodies recognize epitopes between residues 80 and 153 (unpublished data). Unique epitope recognition by each antibody pair, therefore, could lead to different cTnI values being reported because of variable detection of the many circulating forms of cTnI. For an example from the present study, Western blotting showed that mAb 144B3.63.5 recognized several lower molecular mass cTnI bands, in addition to intact cTnI, that were not recognized by mAb 144B5.2.1. When together in the immunoassay reaction cuvette, these mAbs will detect only cTnI fragments or complexes recognized by both. Other examples from the literature have shown that circulating immunoreactive cTnI is, in fact, a heterogeneous mixture of fragments and complexes (19-22, 30). The concentrations and identification of the circulating cTnI forms following myocardial damage are not yet clearly resolved. Because the three commercial assays investigated here appear to recognize basically the same fraction of circulating immunoreactive cTnI, however, the clinical significance seemingly is minimal. Further studies must be performed before this is verified.
In conclusion, there was an overall improvement in individual patient cTnI values between the four cTnI methods evaluated in this study, three commercial and one research, when a patient-derived serum pool calibrator was used. Scatter in the correlation plots between the different methods was still observed, presumably because each assay uses immunoreagent pairs that recognize different troponin epitopes. These individual patient sample differences were likely related to the different circulating forms of cTnI and the relative abilities of the different antibodies to recognize them. Despite the fact that different immunoreagent pairs were used, however, the commercial cTnI immunoassays still recognize similar, but not identical, circulating immunoreactive cTnI fraction(s). Further studies to fully characterize the many circulating forms of cTnI are warranted.
Y.O. and D.J.N. were supported by a research grant from Dade Behring Incorporated, Geneva, Switzerland. We thank Dr. Paul Collinson of the Mayday Hospital Croydon, London, UK, and Dr. Ying Foo of the Royal Free Hospital, London, UK, for their assistance in obtaining clinical samples. We also thank SCIPAC UK Ltd, Sittingbourne, Kent, UK, for the purified cTnI preparations, and Carol Davey, Fulya Yahioglu, Lisa Quann, and Ellen Magee for technical assistance.
Received December 8, 1998; accepted March 9, 1999.
(1.) Bhayana V, Hendersen AR. Biochemical markers of myocardial damage. Clin Biochem 1995;28:1-29.
(2.) Mair J, Wagner I, Puschendorf B, Mair P, Lechleitner P, Dienstl F. Cardiac troponin I to diagnose myocardial injury. Lancet 1993; 341:838-9.
(3.) Adams JE, Bodor GS, Davilaroman VG, Delmez JA, Apple FS, Ladenson JH, Jaffe AS. Cardiac troponin I a marker with high specificity for cardiac injury. Circulation 1993;88:101-6.
(4.) Bodor GS, Porterfield D, Voss EM, Smith S, Apple FS. Cardiac troponin I is not expressed in fetal and healthy or diseased adult human skeletal muscle tissue. Clin Chem 1995;41:1710-5.
(5.) Cummins P, Perry VS. Troponin I from human skeletal and cardiac muscles. Biochem J 1978;171:251-9.
(6.) Haller C, Zehelein J, Remppis A, Muller-Bardorf M, Katus HA. Cardiac troponin T in patients with end-stage renal disease: absence of expression in truncal skeletal muscle. Clin Chem 1998;44:930-8.
(7.) Ricchiuti V, Voss EM, Ney A, Odland M, Anderson PAW, Apple FS. Cardiac troponin T isoforms expressed in renal diseased skeletal muscle will not cause false-positive results by the second generation cardiac troponin T assay by Boehringer Mannheim. Clin Chem 1998;44:1919-24.
(8.) Bodor GS, Porter S, Landt Y, Ladenson JH. Development of monoclonal antibodies for an assay of cardiac troponin-I and preliminary results in suspected cases of myocardial infarction. Clin Chem 1992;38:2203-14.
(9.) Larue C, Defacque-Lacquement H, Calzolari C, Nguyen DL, Pau B. New monoclonal antibodies as probes for human cardiac troponin I: epitopic analysis with synthetic peptides. Mol Immunol 1992; 29:271-8.
(10.) Jaffe AS, Landt Y, Parvin CA, Abendschein DR, Geltman EM, Ladenson JH. Comparative sensitivity of cardiac troponin I and lactate dehydrogenase isoenzymes for diagnosing acute myocardial infarction. Clin Chem 1996;42:1770-6.
(11.) Etievent J-P, Chocron S, Toubin G, Taberlet C, Alwan K, Clement F, et al. Use of troponin I as a marker of perioperative myocardial ischaemia. Ann Thorac Surg 1995;59:1192-4.
(12.) Zaninotto M, Altinier S, Lachin M, Carraro P, Plebani M. Fluoroenzymometric method to measure cardiac troponin I in sera of patients with myocardial infarction. Clin Chem 1996;42:1460-6.
(13.) Larue C, Calzolari C, Bertinchant JP, Leclercq F, Grolleau R, Pau B. Cardiac-specific immunoenzymometric assay of troponin I in the early phase of acute myocardial infarction. Clin Chem 1993;39:972-9.
(14.) Lindahl B, Venge P, Wallentin L, for The BIOMACS Study Group. Early diagnosis and exclusion of acute myocardial infarction using biochemical monitoring. Coron Artery Dis 1995;6:321-8.
(15.) Bakker AJ, Koelemay MJW, Gorgels JPMC, van Vilies B, Smits R, Tijssen JGP, Haagen FDM. Failure of new biochemical markers to exclude acute myocardial infarction at admission. Lancet 1993; 342:1220-2.
(16.) Timmis AD. Will serum enzymes and other proteins find a clinical application in the early diagnosis of myocardial infarction? Br Heart J 1994;71:309-10.
(17.) Cummins P, Young A, Auckland ML, Michie CA, Stone PCW, Shepstone BJ. Comparison of serum cardiac specific troponin I with creatine kinase, creatine kinase-MB isoenzyme, tropomyosin, myoglobin and C-reactive protein release in marathon runners: cardiac or skeletal muscle trauma? Eur J Clin Investig 1987;17:317-24.
(18.) Vorderwinkler K-P, Mair J, Puschendorf B, Hempel A, Schluter K-D, Piper HM. Cardiac troponin I increases in parallel to cardiac troponin T, creatine kinase and lactate dehydrogenase in effluents from isolated perfused rat hearts after hypoxia-reoxygenation-induced myocardial injury. Clin Chim Acta 1996;251:113-7.
(19.) Katrukha AG, Bereznikova AV, Esakova TV, Pettersson K, Lovgren T, Severina ME, et al. Troponin I is released in bloodstream of patients with acute myocardial infarction not in free form but as complex. Clin Chem 1997;43:1379-85.
(20.) Morjana NA, Puia AM, Wong LG, Schreier WA. Cardiac troponin I exists as a complex in patient serum [Abstract]. Clin Chem 1997;43:S157.
(21.) Wu AHB, Feng YJ, Moore R, Apple FS, McPherson PH, Buechler KF, et al. Characterization of cardiac troponin subunit release into serum after acute myocardial infarction and comparison of assays for troponin T and I. Clin Chem 1997;44:1198-208.
(22.) McPherson PH, Buechler KF. Characterization and measurement of troponin I, troponin T and troponin complexes in blood from AMI patients [Abstract]. Clin Chem 1997;43:S136.
(23.) Jaquet K, Thieleczek R, Heilmeyer LMG. Pattern formation on cardiac troponin I by consecutive phosphorylation and dephosphorylation. Eur J Biochem 1995;231:486-90.
(24.) Cummins B, Russell GJ, Cummins P. A monoclonal antibody that distinguishes phosphorylated and dephosphorylated forms of cardiac troponin I. Biochem Soc Trans 1991;19:S-161.
(25.) Di Lisa F, De Tullio R, Salamino F, Barbato R, Melloni E, Siliprandi N, et al. Specific degradation of troponin T and I by [mu]-calpain and its modulation by substrate phosphorylation. Biochem J 1995; 308:57-61.
(26.) Birkmeyer RC, Diaco R, Hutson DK, Lau HP, Miller WK, Neelkantan NV, et al. Application of novel chromium dioxide magnetic particles to immunoassay development. Clin Chem 1987;33:1543-57.
(27.) Passing H, Bablock W. A new biometrical procedure for testing the equality of measurements from two different analytical methods. J Clin Chem Clin Biochem 1983;21:709-32.
(28.) Hollis S. Analysis of method comparison studies. Ann Clin Biochem 1996;33:1-4.
(29.) Waskiewicz D, Salhaney J, Lavigne L, Fagan G, Whitely G. Sample stability differences among troponin I kits [Abstract]. Clin Chem 1996; 42:S311.
(30.) Morjana NA. Degradation of human cardiac troponin I after myocardial infarction. Biotechnol Appl Biochem 1998;28:105-11.
(31.) Lavigne L, Waskiewicz D, Pervaiz S, Fagan G, Whiteley G. Investigation of serum troponin I heterogeneity and complexation to troponin T [Abstract]. Clin Chem 1996;42:S312.
(32.) Katrukha AG, Bereznikova AV, Filatov VL, Esakova TV, Kolosova OV, Pettersson K, et al. Degradation of cardiac troponin I: implication for reliable immunodetection. Clin Chem 1998;44: 2433-40.
(33.) Filatov VL, Katrukha AG, Bereznikova AV, Esakova TV, Bulargria TV, Kolosova OV, et al. Epitope mapping of anti-troponin I monoclonal antibodies. Biochem Mol Biol Int 1998;6:1179-87.
(34.) Bhayana V, Gougoulias T, Gawad Y. A comparison of four serum cardiac troponin I methods [Abstract]. Clin Chem 1997;43:S128.
(35.) Panteghini M, Bonora R, Pagani F. Automated immunoassay of cardiac troponin I in serum evaluated. Clin Chem 1997;43: 195-6.
(36.) Waskiewicz D, Lavigne L, Salhaney J, Annunziato M, Fagan G, Whitely G. Rapid self contained troponin I assay for the Behring Opus immunoassay system [Abstract]. Clin Chem 1995;41:S63.
DAVID J. NEWMAN,  * YEMI OLABIRAN,  WILLIAM D. BEDZYK,  SUZETTE CHANCE,  EILEEN G. GORMAN,  and CHRISTOPHER P. PRICE  ([dagger])
 Department of Clinical Biochemistry, St. Bartholomew's and the Royal London School of Medicine and Dentistry, Turner Street, London E1 2AD, UK.
 Dade Behring Incorporated, Glasgow Research Laboratory, Bldg. 700, P.O. Box 6101, Newark, DE 19714-6101.
* Present address: SW Thames Institute for Renal Research, St. Helier Hospital, Wrythe Lane, Carshalton, Surrey SM5 1AA, UK.
([dagger])Author for correspondence. Fax 44 171 377 1544; e-mail c.p.price@ mds.qmw.ac.uk.
 Nonstandard abbreviations: TnI, troponin 1; cTnI, cardiac TnI; mAb, monoclonal antibody; skTnI, skeletal TnI; and QC, quality control.
Table 1. Intraassay and total precision for the aca plus cTnl assay. Intraassay Material n Mean, [micro]g/L SD, [micro]g/L CV, % QC1 20 0.99 0.17 17 QC2 20 35.2 0.75 2.1 QC3 20 58.9 1.32 2.2 Total assay Material n n SD, mg/L CV, % QC1 20 20 0.2 20 QC2 20 20 1.42 4.0 QC3 20 20 3.09 5.2 Table 2. Regression analyses for intermethod comparisons vs Stratus. Slope using method-specific Slope using Stratus assigned Method calibrator I (95% CI) (a) calibrator II (95% CI) aca plus 0.90 (0.80-1.00) 1.00 (0.89-1.11) Opus II 1.41 (1.25-1.53) 0.99 (0.87-1.07) Access 0.07 (0.06-0.08) 1.19 (1.02-1.34) Intercept after recalibration Pearson Method (95% CI) correlation aca plus 0.23 (20.39 to 0.65) 0.9 Opus II 20.53 (21.04 to 0.04) 0.95 Access 0.47 (22.00 to 0.09) 0.83 (a) CI, confidence interval.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Enzymes and Protein Markers|
|Author:||Newman, David J.; Olabiran, Yemi; Bedzyk, William D.; Chance, Suzette; Gorman, Eileen G.; Price, Chr|
|Date:||Jun 1, 1999|
|Previous Article:||Measurement of the complex between prostate-specific antigen and [[alpha].sub.1]-protease inhibitor in serum.|
|Next Article:||Falsely increased immunoassay measurements of total and unbound phenytoin in critically ill uremic patients receiving fosphenytoin.|