- Email: [email protected]
Diagnostic potential of phosphorylated cardiac troponin I as a sensitive, cardiac-specific marker for early acute coronary syndrome: Preliminary report
Clinica Chimica Acta 362 (2005) 65 – 70 www.elsevier.com/locate/clinchim
Diagnostic potential of phosphorylated cardiac troponin I as a sensitive, cardiac-specific marker for early acute coronary syndrome: Preliminary report David Bar-Or a,b,*, Gregory W. Thomas b, Raphael Bar-Or b, Leonard Rael b, James V. Winkler b a
Swedish Medical Center, Trauma Research Dept., 501 East Hampden Avenue, Room 4-454, Englewood, CO 80113, United States b DMI BioSciences, Inc. 3601 South Clarkson Street #420, Englewood, CO 80113, United States Received 18 January 2005; received in revised form 24 May 2005; accepted 25 May 2005 Available online 21 June 2005
Abstract Background: Cardiac troponin I (cTnI) has low sensitivity in the early hours of acute coronary syndrome (ACS). For patients with early ACS symptoms, we determined the diagnostic potential of an immunoassay for phosphorylated cTnI (PO4-cTnI). Methods: In a prospective study of 61 emergency department patients with suspected ACS, we compared a novel plasma immunoassay for PO4-cTnI to cTnI overall and in a subgroup with symptoms V4 h duration (n = 31). Admission PO4-cTnI and cTnI assays (thresholds determined by ROC curve) were analyzed in a blinded fashion against the clinical, ECG and coronary angiographic diagnosis of ACS. Results: Overall, PO4-cTnI sensitivity was significantly higher than cTnI (82% vs. 50%, respectively, P b 0.05) and PO4-cTnI specificity was 81% (n = 61). Addition of PO4-cTnI to cTnI improved sensitivity to 91% vs. 50% for cTnI alone ( P b 0.001). In the V 4 h subgroup (n = 31), PO4-cTnI sensitivity was significantly higher than cTnI (79% vs. 26%, respectively, P b 0.01) and PO4-cTnI specificity was 75%. In the same V 4 h subgroup, addition of PO4-cTnI to cTnI improved sensitivity to 84% vs. 26% for cTnI alone ( P = 0.001). Conclusions: The results suggest that PO4-cTnI, alone or in combination with cTnI, warrants further investigation as a sensitive, cardiac-specific diagnostic tool for early ACS. D 2005 Elsevier B.V. All rights reserved. Keywords: Phosphorylation; Inflammation; Troponin; Cardiac markers; Acute coronary syndrome; Myocardial infarction
Abbreviations: cTnI, cardiac troponin I; PO4-cTnI, phosphorylated cardiac troponin I; CV, coefficient of variation; Sens, sensitivity; Spec, specificity; STEMI, ST elevation myocardial infarction; Non-STEMI, non-ST elevation myocardial infarction; UA, unstable angina. * Corresponding author. Swedish Medical Center, Trauma Research Dept., 501 East Hampden Avenue, Room 4-454, Englewood, CO 80113, United States. Tel.: +1 303 788 4089; fax: +1 303 788 4064. E-mail address: [email protected] (D. Bar-Or). 0009-8981/$ - see front matter D 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.cccn.2005.05.015
66
D. Bar-Or et al. / Clinica Chimica Acta 362 (2005) 65–70
1. Introduction Cardiac troponins are primary myocardial tissuespecific biomarkers of acute coronary syndrome (ACS); however, cardiac troponins are unreliable in the first 4–6 h after the onset of symptoms due to low sensitivity [1–4]. Non-specific diagnostic markers, such as myoglobin or C-reactive protein (CRP), are often recommended to assist risk stratification in the first few hours of ACS symptoms [1,2,5,6]. Substantial evidence now supports a primary role for inflammation in ACS and elevated inflammatory markers can indicate higher risk of ischemic events in suspected ACS [7]. Phosphorylation of proteins by kinases is involved in inflammation and activation of signal transduction pathways [8,9] that can include kinase phosphorylation of cardiac troponin I (cTnI) which is reported to cause cardiomyocyte dysfunction [10]. In this report, we examined the potential utility of phosphorylated cTnI (PO4-cTnI) in the diagnosis of ACS.
2. Materials and methods 2.1. Clinical population Sixty-one adult patients admitted to the ED of Swedish Medical Center, Englewood, CO, with symptoms highly suggestive of ACS participated in this study and completed IRB-approved informed consent. Exclusion criteria were recent trauma, active systemic inflammatory disease, corticosteroid use during the previous 3 weeks, and any use of NSAID, COX-2 inhibitor or N 1 aspirin per day in the week before suspected ACS in order to limit this preliminary study to inflammatory responses associated primarily with ACS. Blood samples for PO4cTnI and cTnI and the initial ECG were obtained at the time of ED admission. 2.2. PO4-cTnI immunoassay description Blood samples for PO4-cTnI were collected in lithium heparin tubes, centrifuged and the plasma was frozen at 80 8C for immunoassays performed later by 1 operator blinded to clinical outcomes.
Affinity-purified goat polyclonal antibody reactive with the C-terminus of human cTnI (Santa Cruz Biotechnology, product sc-8118) was obtained and investigated for its specificity to human cTnI. Briefly, samples of human plasma and serum were spiked with increasing amounts (approximately 0.1 to 5.0 ng/ml concentrations) of recombinant human cTnI (BiosPacific, Emeryville, CA) and analyzed by western blot and immunoprecipitation. Results demonstrated the anti-cTnI antibody exhibited strong recognition of the human cTnI and no apparent cross reaction with other serum proteins (data not shown). The PO4-cTnI ELISA method was as follows: the anti-cTnI antibody was diluted to 1 Ag/ml in PBS, pH 7.4, with 100 Al added to each ELISA strip well (Nunc, F16 Maxisorp strips). After incubation overnight (room temperature), antibody solution was aspirated, and 200 Al of PBS containing 4% BSA added to each well. Following 18–24 h incubation (room temperature), BSA solution was aspirated, the strips blotted, and 50 Al of patient plasma added. Background consists of 50 Al of PBS containing 4% BSA added to duplicate wells. Following 45 min incubation (room temperature) with gentle shaking, samples are aspirated, and wells washed (4 with wash buffer of 50 mmol/ l Tris, pH 7.9–8.1, 0.2% Tween 20). Then 100 Al of a mixture of 1 Ag/ml rabbit polyclonal anti-phosphoprotein antibody (Zymed) in PBS containing 4% BSA mixed with a 1: 10,000 dilution in PBS containing 4% BSA of goat anti-rabbit IgG conjugated to HRP (Pierce) was added to each well. Following 45-min incubation (room temperature) with gentle shaking, wells are again washed 4 with wash buffer and 100 Al TMB substrate (Pierce) was added. Following 30-min incubation, the reaction was stopped with 100 Al of 0.18 mol/ l H2SO4. Optical density (OD) of each well was read at 450 nm and 530 nm on a spectrophotometer plate reader. OD at 530 nm was subtracted from OD at 450 nm, and the background (OD at 530 nm OD at 450 nm) was subtracted from this number. 2.3. cTnI immunoassay description cTnI assays were performed at the central clinical laboratory of Swedish Medical Center by Dimension
D. Bar-Or et al. / Clinica Chimica Acta 362 (2005) 65–70
RxLR troponin I (Dade Behring) colorimetric 1-step sandwich assay. A cTnI of 0.6–1.5 ng/ml is suggested by the manufacturer as being consistent with WHO criteria for acute myocardial infarction (AMI). At this study institution the lowest reported cTnI level has been established as b0.4 ng/ml based upon increased mortality for UA z 0.4 ng/ml [3]. Coefficient of variations (CV) for the laboratory cTnI assay at concentrations of 0.39 and 1.23 ng/ml were 6.03% and 1.89%, respectively. 2.4. Clinical evaluations Clinical outcomes for this study were determined after discharge by reviewing patient medical records including, laboratory, ECG, ED, coronary catheterization, surgical and discharge reports. The diagnosis of ACS at this study institution is routinely based upon a combination of clinical symptoms, ECG and cTnI levels at initial presentation followed by serial cTnI levels and additional cardiac diagnostic studies (e.g., coronary angiography) for patients requiring further observation. ACS included ST-segment elevation myocardial infarction (STEMI), non-ST-segment elevation myocardial infarction (non-STEMI), or unstable angina pectoris (UA) determined according to American College of Cardiology/ American Heart Association guidelines. As blood troponin levels appear to have lowest sensitivity in the first 4–6 h after the onset of symptoms [1,2], evaluation of PO4-cTnI vs. cTnI results in patients with b 4 h of symptoms was considered clinically important for purposes of this investigation. All hospital staff and attending physicians were blinded to PO4-cTnI immunoassay results during this study.
3. Results Study patient demographic data is listed in Table 1. Although a normal range for the PO4-cTnI immunoassay is not yet known, ROC curve optimal diagnostic thresholds were 0.115 OD for PO4-cTnI and 0.59 ng/ml for cTnI with area under the curve (AUC) of 0.80 and 0.72, respectively (Fig. 1). Mean CV for the PO4-cTnI immunoassay was 8.2% (95% CI, 4.6–11.7%) at the threshold. Overall, 34 out of 61 (56%) had ACS and no patient in this study had cTnI between b 0.4 and 0.59 ng/ml. Using the optimal diagnostic thresholds of 0.115 OD for PO4-cTnI and 0.59 ng/ml for cTnI, PO4-cTnI sensitivity and specificity were 82% and 81%, respectively (Table 2) compared to cTnI sensitivity and specificity of 50% and 92%, respectively. PO4-cTnI sensitivity was significantly higher than cTnI (82% vs. 50%, respectively, P b 0.05) and PO4-cTnI specificity (82%) was not statistically different than cTnI ( P = NS). Addition of PO4-cTnI to cTnI improved sensitivity to 91% vs. 50% for cTnI alone ( P b 0.001). Addition of PO4-cTnI to cTnI and ECG improved sensitivity to 94% vs. 76% for cTnI and ECG alone but the difference did not reach statistical significance ( P = 0.088). The addition of PO4-cTnI to cTnI alone and the addition of PO4-cTnI to cTnI and ECG caused a 4.5- and 3.1fold improvement respectively in the negative likelihood ratios. Intermittent or vague symptoms resulted in approximately one-third of the study patients admitted
Table 1 Baseline characteristics of 61 patient cohort and 35 patients with acute coronary syndrome (ACS)
2.5. Statistical analysis Threshold values for PO4-cTnI and cTnI were determined to maximize the sum of sensitivity and specificity. Diagnostic indices, receiver operating characteristic (ROC) curve analysis and area under the ROC curve were conducted using MatlabR 7.0 R14 (MathWorks, Natick, MA). Data was analyzed by the chi-square test using MedCalcR 7.6 (Medcalc, Mariakerke, Belgium). Significance is reported at P b 0.05.
67
Age (years), mean F S.D. Male, n (%) STEMI, n (%) UA, n (%) Non-STEMI, n (%) Symptoms V 4 h, n (%) Symptoms V12 h, n (%) Symptoms z24 h, n (%)
Overall (n = 61)
ACS (n = 34)
61.1 F12.9 34 (55.7) 16 (26.2) 7 (11.5) 11 (18.0) 31 (50.8) 42 (68.9) 19 (31.1)
62.4 F 11.3 26 (76.5) 16 (48.6) 7 (20.6) 11 (32.4) 19 (55.9) 24 (70.6) 9 (26.5)
Abbreviations: STEMI = ST-segment elevation myocardial infarction; UA= unstable angina; Non-STEMI = non-ST-segment elevation myocardial infarction.
68
D. Bar-Or et al. / Clinica Chimica Acta 362 (2005) 65–70 100
Sensitivity
80
60 cTnI PO4-cTnI
40
20
0 0
20
40
60
80
100
100-Specificity Fig. 1. ROC curves for phosphorylated cTnI (PO4-cTnI) and clinical laboratory troponin (cTnI).
to the ED N24 h after the onset of suspected ACS which resulted in an overall mean symptom onset time (approximate) of N16 h (n = 61). Time of onset appeared to be more precise for patients with recent onset of symptoms and patients with b 4 h of symptoms (n = 31) averaged 2.1 h (+ 1.0 h S.D.) between symptom onset and ED admission. In the subgroup of 31 patients (Table 2) with b4 h of symptom
duration, 19 (61%) had ACS and PO4-cTnI sensitivity was significantly higher than cTnI (79% vs. 26%, respectively, P b 0.01) and PO4-cTnI specificity (75%) was not statistically different than cTnI ( P = 0.25). In the same V 4 h subgroup, addition of PO4-cTnI to cTnI improved sensitivity to 84% vs. 26% for cTnI alone ( P = 0.001). The correlation coefficient between PO4-cTnI and cTnI was 0.0858 ( 0.3345 to 0.1741), implying that the PO4-cTnI and cTnI assays used in this study are independent. Although precise population prevalence was not known for this cohort, we estimated the prevalence using the sample population and calculated positive and negative predictive values (PPV, NPV) which are included with the diagnostic indices listed in Table 2.
4. Discussion Inflammation in ACS is reported to involve leukocytes, endothelial cells, complement, cytokines [7], and activation of signal transduction pathways that can include protein kinase phosphorylation of cTnI [10]. Protein phosphorylation can occur rapidly
Table 2 Diagnostic indices for overall cohort and a subgroup with V4 h of symptoms Sensitivity (95% CI) Specificity (95% CI) PPV (95% CI)
Overall PO4-cTnI (n = 61) cTnI Either PO4-cTnI or cTnI Either cTnI or ECG Either PO4-cTnI, cTnI, or ECG V4 h PO4-cTnI (n = 31) cTnI Either PO4-cTnI or cTnI Either cTnI or ECG Either PO4-cTnI, cTnI, or ECG
NPV (95% CI)
Positive Negative likelihood likelihood ratio ratio
82% (70.0–90.5%) 50% (36.8–63.2%) 91% (80.2–96.6%)
81% (69.1–89.9%) 92% (81.2–97.1%) 72% (58.6–82.5%)
85% (72.9–92.3%) 79% (65.9–87.7%) 4.45 89% (78.1–95.6%) 58% (44.0–70.0%) 6.25 82% (68.9–90.1%) 86% (73.6–93.1%) 3.26
0.21 0.54 0.12
76% (63.3–86.1%)
84% (71.7–91.8%)
87% (74.7–93.7%) 72% (59.0–82.9%) 4.78
0.28
94% (83.9–98.3%)
64% (50.4–75.8%)
78% (65.0–87.4%) 89% (77.4–95.2%) 2.61
0.09
79% (60.1–90.8%) 26% (12.7–45.8%) 84% (65.5–94.2%)
75% (55.9–88.1%) 100% (85.9–99.8%) 73% (53.2–86.6%)
68% (44.9–83.4%)
82% (62.9–92.8%)
87% (68.4–95.7%) 60% (40.8–76.8%) 3.76
0.39
89% (71.7–97.1%)
55% (35.7–72.3%)
77% (57.9–89.8%) 75% (55.6–88.2%) 1.97
0.19
83% (64.9–93.6%) 69% (50.0–83.8%) 3.16 100% (85.9–99.8%) 44% (26.5–63.0%) Infinite 84% (65.5–94.2%) 73% (53.2–86.6%) 3.09
0.28 0.74 0.22
Diagnostic indices for phosphorylated cardiac troponin I immunoassay (PO4-cTnI); cardiac troponin I (cTnI); PO4-cTnI integrated with cTnI alone; and cTnI or electrocardiogram (ECG) results for the diagnosis of acute coronary syndrome (ACS) overall and in a subgroup with V4 h of symptoms. Abbreviations: positive predictive value (PPV); negative predictive value (NPV); confidence interval (CI).
D. Bar-Or et al. / Clinica Chimica Acta 362 (2005) 65–70
and change the functional properties of proteins [8]. Phosphorylated cTnI is associated with decreased myocardial calcium responsiveness and contractility and has been identified by indirect techniques in the sera of patients with myocardial infarction [10–13]. However, it has been recommended that the measurement of specific forms of cTnI, including phosphorylated cTnI, be avoided in order to obtain reliable cTnI assays for the diagnosis of myocardial infarction [13]. Results of this investigation suggest that phosphorylated cTnI may detect early ACS and that an immunoassay for PO4-cTnI could potentially have higher diagnostic sensitivity than cTnI in patients with early symptoms of ACS. The mechanism by which PO4-cTnI appears to be detected earlier than cTnI is unknown. However, phosphorylated cTnI has reduced affinity for the troponin C complex and may be free to leak into plasma during early inflammatory responses or the PO4-cTnI assay may detect a phosphorylated cTnI fragment that appears in plasma before the intact protein [14,15]. Results presented here are of limited scope and other possibilities such as decreased dephosphorylation, potential non-specificity of the PO4-cTnI assay and the absence of reported laboratory cTnI levels b 0.4 ng/ml could have artificially improved PO4cTnI results. Limitations of the study design include an ED study population that is skewed towards ACS, exclusion of patients with inflammatory diseases and anti-inflammatory therapies, and limited availability of performance characteristics for the PO4-cTnI immunoassay. Lower specificity, PPV and positive likelihood ratios for PO4-cTnI compared to cTnI were expected as cardiac troponins are considered a primary diagnostic criterion for ACS. Future studies will investigate additional PO4-cTnI immunoassay performance characteristics (e.g., normal range and clinical cutoff concentrations), comparison to cTnI levels b0.4 ng/ml, comparison to other ischemia and inflammation markers, and serial PO4-cTnI levels in larger subsets of patients with UA, non-STEMI, non-cardiac symptoms, inflammatory diseases, and anti-inflammatory therapies. We conclude that an immunoassay for phosphorylated cTnI warrants further investigation as a readily detectable cardiac-specific biomarker of early ACS.
69
Acknowledgements The authors thank Diane Schleufer, MTASCP, for her assistance in preparing the manuscript. This work was supported by DMI BioSciences, Inc. Each of the authors is an employee of DMI BioSciences, Inc. and has a financial interest in DMI BioSciences, Inc.
References [1] Wu AH, Apple FS, Gibler WB, Jesse RL, Warshaw MM, Valdes Jr R. National Academy of Clinical Biochemistry Standards of Laboratory Practice: recommendations for the use of cardiac markers in coronary artery diseases. Clin Chem 1999;45:1104 – 21. [2] Jesse RL. Impact of the measurement of troponin on the triage, prognosis, and treatment of patients with chest pain. Clin Chim Acta 1999;284:213 – 21. [3] Antman EM, Tanasijevic MJ, Thompson B, Schactman M, McCabe CH, Cannon CP, et al. Cardiac-specific troponin I levels to predict the risk of mortality in patients with acute coronary syndromes. N Engl J Med 1996;335:1342 – 9. [4] Rao SV, Ohman EM, Granger CB, Armstrong PW, Gibler WB, Christenson RH, et al. Prognostic value of isolated troponin elevation across the spectrum of chest pain syndromes. Am J Cardiol 2003;91:936 – 40. [5] de Winter RJ. C-reactive protein and cardiac troponin for early risk stratification in patients with acute coronary syndromes. Clin Chim Acta 2001;311:53 – 6. [6] Mockel M, Gerhardt W, Heller Jr G, Klefisch F, Danne O, Maske J, et al. Validation of NACB and IFCC guidelines for the use of cardiac markers for early diagnosis and risk assessment in patients with acute coronary syndromes. Clin Chim Acta 2001;303:167 – 79. [7] Mulvihill NT, Foley JB. Inflammation in acute coronary syndromes. Heart 2002;87:201 – 4. [8] Hunter T. Protein kinases and phosphatases: the yin and yang of protein phosphorylation and signaling. Cell 1995;80(27): 225 – 36. [9] Shanley TP. Phosphatases: counterregulatory role in inflammatory cell signaling. Crit Care Med 2002;30(Suppl. 1):S80 – 8. [10] Takeishi Y, Chu G, Kirkpatrick DM, Li Z, Wakasaki H, Kranias EG, et al. In vivo phosphorylation of cardiac troponin I by protein kinase Cbeta2 decreases cardiomyocyte calcium responsiveness and contractility in transgenic mouse hearts. J Clin Invest 1998;102:72 – 8. [11] Labugger R, Organ L, Collier C, Atar D, Van Eyk JE. Extensive troponin I and T modification detected in serum from patients with acute myocardial infarction. Circulation 2000; 102:1221 – 6. [12] Burkart EM, Sumandea MP, Kobayashi T, Nili M, Martin AF, Homsher E, et al. Phosphorylation or glutamic acid substitution at protein kinase C sites on cardiac troponin I differen-
70
D. Bar-Or et al. / Clinica Chimica Acta 362 (2005) 65–70
tially depress myofilament tension and shortening velocity. J Biol Chem 2003;278:11265 – 72. [13] Katrukha A, Bereznikova A, Filatov V, Esakova T. Biochemical factors influencing measurement of cardiac troponin I in serum. Clin Chem Lab Med 1999;37:1091 – 5. [14] Lindhout DA, Li MX, Schieve D, Sykes BD. Effects of T142 phosphorylation and mutation R145G on the interaction of the
inhibitory region of human cardiac troponin I with the Cdomain of human cardiac troponin C. Biochemistry 2002;41:7267 – 74. [15] McDonough JL, Arrell DK, Van Eyk JE. Troponin I degradation and covalent complex formation accompanies myocardial ischemia/reperfusion injury. Circ Res 1999;84:9 – 20.
Diagnostic potential of phosphorylated cardiac troponin I as a sensitive, cardiac-specific marker for early acute coronary syndrome: Preliminary report David Bar-Or a,b,*, Gregory W. Thomas b, Raphael Bar-Or b, Leonard Rael b, James V. Winkler b a
Swedish Medical Center, Trauma Research Dept., 501 East Hampden Avenue, Room 4-454, Englewood, CO 80113, United States b DMI BioSciences, Inc. 3601 South Clarkson Street #420, Englewood, CO 80113, United States Received 18 January 2005; received in revised form 24 May 2005; accepted 25 May 2005 Available online 21 June 2005
Abstract Background: Cardiac troponin I (cTnI) has low sensitivity in the early hours of acute coronary syndrome (ACS). For patients with early ACS symptoms, we determined the diagnostic potential of an immunoassay for phosphorylated cTnI (PO4-cTnI). Methods: In a prospective study of 61 emergency department patients with suspected ACS, we compared a novel plasma immunoassay for PO4-cTnI to cTnI overall and in a subgroup with symptoms V4 h duration (n = 31). Admission PO4-cTnI and cTnI assays (thresholds determined by ROC curve) were analyzed in a blinded fashion against the clinical, ECG and coronary angiographic diagnosis of ACS. Results: Overall, PO4-cTnI sensitivity was significantly higher than cTnI (82% vs. 50%, respectively, P b 0.05) and PO4-cTnI specificity was 81% (n = 61). Addition of PO4-cTnI to cTnI improved sensitivity to 91% vs. 50% for cTnI alone ( P b 0.001). In the V 4 h subgroup (n = 31), PO4-cTnI sensitivity was significantly higher than cTnI (79% vs. 26%, respectively, P b 0.01) and PO4-cTnI specificity was 75%. In the same V 4 h subgroup, addition of PO4-cTnI to cTnI improved sensitivity to 84% vs. 26% for cTnI alone ( P = 0.001). Conclusions: The results suggest that PO4-cTnI, alone or in combination with cTnI, warrants further investigation as a sensitive, cardiac-specific diagnostic tool for early ACS. D 2005 Elsevier B.V. All rights reserved. Keywords: Phosphorylation; Inflammation; Troponin; Cardiac markers; Acute coronary syndrome; Myocardial infarction
Abbreviations: cTnI, cardiac troponin I; PO4-cTnI, phosphorylated cardiac troponin I; CV, coefficient of variation; Sens, sensitivity; Spec, specificity; STEMI, ST elevation myocardial infarction; Non-STEMI, non-ST elevation myocardial infarction; UA, unstable angina. * Corresponding author. Swedish Medical Center, Trauma Research Dept., 501 East Hampden Avenue, Room 4-454, Englewood, CO 80113, United States. Tel.: +1 303 788 4089; fax: +1 303 788 4064. E-mail address: [email protected] (D. Bar-Or). 0009-8981/$ - see front matter D 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.cccn.2005.05.015
66
D. Bar-Or et al. / Clinica Chimica Acta 362 (2005) 65–70
1. Introduction Cardiac troponins are primary myocardial tissuespecific biomarkers of acute coronary syndrome (ACS); however, cardiac troponins are unreliable in the first 4–6 h after the onset of symptoms due to low sensitivity [1–4]. Non-specific diagnostic markers, such as myoglobin or C-reactive protein (CRP), are often recommended to assist risk stratification in the first few hours of ACS symptoms [1,2,5,6]. Substantial evidence now supports a primary role for inflammation in ACS and elevated inflammatory markers can indicate higher risk of ischemic events in suspected ACS [7]. Phosphorylation of proteins by kinases is involved in inflammation and activation of signal transduction pathways [8,9] that can include kinase phosphorylation of cardiac troponin I (cTnI) which is reported to cause cardiomyocyte dysfunction [10]. In this report, we examined the potential utility of phosphorylated cTnI (PO4-cTnI) in the diagnosis of ACS.
2. Materials and methods 2.1. Clinical population Sixty-one adult patients admitted to the ED of Swedish Medical Center, Englewood, CO, with symptoms highly suggestive of ACS participated in this study and completed IRB-approved informed consent. Exclusion criteria were recent trauma, active systemic inflammatory disease, corticosteroid use during the previous 3 weeks, and any use of NSAID, COX-2 inhibitor or N 1 aspirin per day in the week before suspected ACS in order to limit this preliminary study to inflammatory responses associated primarily with ACS. Blood samples for PO4cTnI and cTnI and the initial ECG were obtained at the time of ED admission. 2.2. PO4-cTnI immunoassay description Blood samples for PO4-cTnI were collected in lithium heparin tubes, centrifuged and the plasma was frozen at 80 8C for immunoassays performed later by 1 operator blinded to clinical outcomes.
Affinity-purified goat polyclonal antibody reactive with the C-terminus of human cTnI (Santa Cruz Biotechnology, product sc-8118) was obtained and investigated for its specificity to human cTnI. Briefly, samples of human plasma and serum were spiked with increasing amounts (approximately 0.1 to 5.0 ng/ml concentrations) of recombinant human cTnI (BiosPacific, Emeryville, CA) and analyzed by western blot and immunoprecipitation. Results demonstrated the anti-cTnI antibody exhibited strong recognition of the human cTnI and no apparent cross reaction with other serum proteins (data not shown). The PO4-cTnI ELISA method was as follows: the anti-cTnI antibody was diluted to 1 Ag/ml in PBS, pH 7.4, with 100 Al added to each ELISA strip well (Nunc, F16 Maxisorp strips). After incubation overnight (room temperature), antibody solution was aspirated, and 200 Al of PBS containing 4% BSA added to each well. Following 18–24 h incubation (room temperature), BSA solution was aspirated, the strips blotted, and 50 Al of patient plasma added. Background consists of 50 Al of PBS containing 4% BSA added to duplicate wells. Following 45 min incubation (room temperature) with gentle shaking, samples are aspirated, and wells washed (4 with wash buffer of 50 mmol/ l Tris, pH 7.9–8.1, 0.2% Tween 20). Then 100 Al of a mixture of 1 Ag/ml rabbit polyclonal anti-phosphoprotein antibody (Zymed) in PBS containing 4% BSA mixed with a 1: 10,000 dilution in PBS containing 4% BSA of goat anti-rabbit IgG conjugated to HRP (Pierce) was added to each well. Following 45-min incubation (room temperature) with gentle shaking, wells are again washed 4 with wash buffer and 100 Al TMB substrate (Pierce) was added. Following 30-min incubation, the reaction was stopped with 100 Al of 0.18 mol/ l H2SO4. Optical density (OD) of each well was read at 450 nm and 530 nm on a spectrophotometer plate reader. OD at 530 nm was subtracted from OD at 450 nm, and the background (OD at 530 nm OD at 450 nm) was subtracted from this number. 2.3. cTnI immunoassay description cTnI assays were performed at the central clinical laboratory of Swedish Medical Center by Dimension
D. Bar-Or et al. / Clinica Chimica Acta 362 (2005) 65–70
RxLR troponin I (Dade Behring) colorimetric 1-step sandwich assay. A cTnI of 0.6–1.5 ng/ml is suggested by the manufacturer as being consistent with WHO criteria for acute myocardial infarction (AMI). At this study institution the lowest reported cTnI level has been established as b0.4 ng/ml based upon increased mortality for UA z 0.4 ng/ml [3]. Coefficient of variations (CV) for the laboratory cTnI assay at concentrations of 0.39 and 1.23 ng/ml were 6.03% and 1.89%, respectively. 2.4. Clinical evaluations Clinical outcomes for this study were determined after discharge by reviewing patient medical records including, laboratory, ECG, ED, coronary catheterization, surgical and discharge reports. The diagnosis of ACS at this study institution is routinely based upon a combination of clinical symptoms, ECG and cTnI levels at initial presentation followed by serial cTnI levels and additional cardiac diagnostic studies (e.g., coronary angiography) for patients requiring further observation. ACS included ST-segment elevation myocardial infarction (STEMI), non-ST-segment elevation myocardial infarction (non-STEMI), or unstable angina pectoris (UA) determined according to American College of Cardiology/ American Heart Association guidelines. As blood troponin levels appear to have lowest sensitivity in the first 4–6 h after the onset of symptoms [1,2], evaluation of PO4-cTnI vs. cTnI results in patients with b 4 h of symptoms was considered clinically important for purposes of this investigation. All hospital staff and attending physicians were blinded to PO4-cTnI immunoassay results during this study.
3. Results Study patient demographic data is listed in Table 1. Although a normal range for the PO4-cTnI immunoassay is not yet known, ROC curve optimal diagnostic thresholds were 0.115 OD for PO4-cTnI and 0.59 ng/ml for cTnI with area under the curve (AUC) of 0.80 and 0.72, respectively (Fig. 1). Mean CV for the PO4-cTnI immunoassay was 8.2% (95% CI, 4.6–11.7%) at the threshold. Overall, 34 out of 61 (56%) had ACS and no patient in this study had cTnI between b 0.4 and 0.59 ng/ml. Using the optimal diagnostic thresholds of 0.115 OD for PO4-cTnI and 0.59 ng/ml for cTnI, PO4-cTnI sensitivity and specificity were 82% and 81%, respectively (Table 2) compared to cTnI sensitivity and specificity of 50% and 92%, respectively. PO4-cTnI sensitivity was significantly higher than cTnI (82% vs. 50%, respectively, P b 0.05) and PO4-cTnI specificity (82%) was not statistically different than cTnI ( P = NS). Addition of PO4-cTnI to cTnI improved sensitivity to 91% vs. 50% for cTnI alone ( P b 0.001). Addition of PO4-cTnI to cTnI and ECG improved sensitivity to 94% vs. 76% for cTnI and ECG alone but the difference did not reach statistical significance ( P = 0.088). The addition of PO4-cTnI to cTnI alone and the addition of PO4-cTnI to cTnI and ECG caused a 4.5- and 3.1fold improvement respectively in the negative likelihood ratios. Intermittent or vague symptoms resulted in approximately one-third of the study patients admitted
Table 1 Baseline characteristics of 61 patient cohort and 35 patients with acute coronary syndrome (ACS)
2.5. Statistical analysis Threshold values for PO4-cTnI and cTnI were determined to maximize the sum of sensitivity and specificity. Diagnostic indices, receiver operating characteristic (ROC) curve analysis and area under the ROC curve were conducted using MatlabR 7.0 R14 (MathWorks, Natick, MA). Data was analyzed by the chi-square test using MedCalcR 7.6 (Medcalc, Mariakerke, Belgium). Significance is reported at P b 0.05.
67
Age (years), mean F S.D. Male, n (%) STEMI, n (%) UA, n (%) Non-STEMI, n (%) Symptoms V 4 h, n (%) Symptoms V12 h, n (%) Symptoms z24 h, n (%)
Overall (n = 61)
ACS (n = 34)
61.1 F12.9 34 (55.7) 16 (26.2) 7 (11.5) 11 (18.0) 31 (50.8) 42 (68.9) 19 (31.1)
62.4 F 11.3 26 (76.5) 16 (48.6) 7 (20.6) 11 (32.4) 19 (55.9) 24 (70.6) 9 (26.5)
Abbreviations: STEMI = ST-segment elevation myocardial infarction; UA= unstable angina; Non-STEMI = non-ST-segment elevation myocardial infarction.
68
D. Bar-Or et al. / Clinica Chimica Acta 362 (2005) 65–70 100
Sensitivity
80
60 cTnI PO4-cTnI
40
20
0 0
20
40
60
80
100
100-Specificity Fig. 1. ROC curves for phosphorylated cTnI (PO4-cTnI) and clinical laboratory troponin (cTnI).
to the ED N24 h after the onset of suspected ACS which resulted in an overall mean symptom onset time (approximate) of N16 h (n = 61). Time of onset appeared to be more precise for patients with recent onset of symptoms and patients with b 4 h of symptoms (n = 31) averaged 2.1 h (+ 1.0 h S.D.) between symptom onset and ED admission. In the subgroup of 31 patients (Table 2) with b4 h of symptom
duration, 19 (61%) had ACS and PO4-cTnI sensitivity was significantly higher than cTnI (79% vs. 26%, respectively, P b 0.01) and PO4-cTnI specificity (75%) was not statistically different than cTnI ( P = 0.25). In the same V 4 h subgroup, addition of PO4-cTnI to cTnI improved sensitivity to 84% vs. 26% for cTnI alone ( P = 0.001). The correlation coefficient between PO4-cTnI and cTnI was 0.0858 ( 0.3345 to 0.1741), implying that the PO4-cTnI and cTnI assays used in this study are independent. Although precise population prevalence was not known for this cohort, we estimated the prevalence using the sample population and calculated positive and negative predictive values (PPV, NPV) which are included with the diagnostic indices listed in Table 2.
4. Discussion Inflammation in ACS is reported to involve leukocytes, endothelial cells, complement, cytokines [7], and activation of signal transduction pathways that can include protein kinase phosphorylation of cTnI [10]. Protein phosphorylation can occur rapidly
Table 2 Diagnostic indices for overall cohort and a subgroup with V4 h of symptoms Sensitivity (95% CI) Specificity (95% CI) PPV (95% CI)
Overall PO4-cTnI (n = 61) cTnI Either PO4-cTnI or cTnI Either cTnI or ECG Either PO4-cTnI, cTnI, or ECG V4 h PO4-cTnI (n = 31) cTnI Either PO4-cTnI or cTnI Either cTnI or ECG Either PO4-cTnI, cTnI, or ECG
NPV (95% CI)
Positive Negative likelihood likelihood ratio ratio
82% (70.0–90.5%) 50% (36.8–63.2%) 91% (80.2–96.6%)
81% (69.1–89.9%) 92% (81.2–97.1%) 72% (58.6–82.5%)
85% (72.9–92.3%) 79% (65.9–87.7%) 4.45 89% (78.1–95.6%) 58% (44.0–70.0%) 6.25 82% (68.9–90.1%) 86% (73.6–93.1%) 3.26
0.21 0.54 0.12
76% (63.3–86.1%)
84% (71.7–91.8%)
87% (74.7–93.7%) 72% (59.0–82.9%) 4.78
0.28
94% (83.9–98.3%)
64% (50.4–75.8%)
78% (65.0–87.4%) 89% (77.4–95.2%) 2.61
0.09
79% (60.1–90.8%) 26% (12.7–45.8%) 84% (65.5–94.2%)
75% (55.9–88.1%) 100% (85.9–99.8%) 73% (53.2–86.6%)
68% (44.9–83.4%)
82% (62.9–92.8%)
87% (68.4–95.7%) 60% (40.8–76.8%) 3.76
0.39
89% (71.7–97.1%)
55% (35.7–72.3%)
77% (57.9–89.8%) 75% (55.6–88.2%) 1.97
0.19
83% (64.9–93.6%) 69% (50.0–83.8%) 3.16 100% (85.9–99.8%) 44% (26.5–63.0%) Infinite 84% (65.5–94.2%) 73% (53.2–86.6%) 3.09
0.28 0.74 0.22
Diagnostic indices for phosphorylated cardiac troponin I immunoassay (PO4-cTnI); cardiac troponin I (cTnI); PO4-cTnI integrated with cTnI alone; and cTnI or electrocardiogram (ECG) results for the diagnosis of acute coronary syndrome (ACS) overall and in a subgroup with V4 h of symptoms. Abbreviations: positive predictive value (PPV); negative predictive value (NPV); confidence interval (CI).
D. Bar-Or et al. / Clinica Chimica Acta 362 (2005) 65–70
and change the functional properties of proteins [8]. Phosphorylated cTnI is associated with decreased myocardial calcium responsiveness and contractility and has been identified by indirect techniques in the sera of patients with myocardial infarction [10–13]. However, it has been recommended that the measurement of specific forms of cTnI, including phosphorylated cTnI, be avoided in order to obtain reliable cTnI assays for the diagnosis of myocardial infarction [13]. Results of this investigation suggest that phosphorylated cTnI may detect early ACS and that an immunoassay for PO4-cTnI could potentially have higher diagnostic sensitivity than cTnI in patients with early symptoms of ACS. The mechanism by which PO4-cTnI appears to be detected earlier than cTnI is unknown. However, phosphorylated cTnI has reduced affinity for the troponin C complex and may be free to leak into plasma during early inflammatory responses or the PO4-cTnI assay may detect a phosphorylated cTnI fragment that appears in plasma before the intact protein [14,15]. Results presented here are of limited scope and other possibilities such as decreased dephosphorylation, potential non-specificity of the PO4-cTnI assay and the absence of reported laboratory cTnI levels b 0.4 ng/ml could have artificially improved PO4cTnI results. Limitations of the study design include an ED study population that is skewed towards ACS, exclusion of patients with inflammatory diseases and anti-inflammatory therapies, and limited availability of performance characteristics for the PO4-cTnI immunoassay. Lower specificity, PPV and positive likelihood ratios for PO4-cTnI compared to cTnI were expected as cardiac troponins are considered a primary diagnostic criterion for ACS. Future studies will investigate additional PO4-cTnI immunoassay performance characteristics (e.g., normal range and clinical cutoff concentrations), comparison to cTnI levels b0.4 ng/ml, comparison to other ischemia and inflammation markers, and serial PO4-cTnI levels in larger subsets of patients with UA, non-STEMI, non-cardiac symptoms, inflammatory diseases, and anti-inflammatory therapies. We conclude that an immunoassay for phosphorylated cTnI warrants further investigation as a readily detectable cardiac-specific biomarker of early ACS.
69
Acknowledgements The authors thank Diane Schleufer, MTASCP, for her assistance in preparing the manuscript. This work was supported by DMI BioSciences, Inc. Each of the authors is an employee of DMI BioSciences, Inc. and has a financial interest in DMI BioSciences, Inc.
References [1] Wu AH, Apple FS, Gibler WB, Jesse RL, Warshaw MM, Valdes Jr R. National Academy of Clinical Biochemistry Standards of Laboratory Practice: recommendations for the use of cardiac markers in coronary artery diseases. Clin Chem 1999;45:1104 – 21. [2] Jesse RL. Impact of the measurement of troponin on the triage, prognosis, and treatment of patients with chest pain. Clin Chim Acta 1999;284:213 – 21. [3] Antman EM, Tanasijevic MJ, Thompson B, Schactman M, McCabe CH, Cannon CP, et al. Cardiac-specific troponin I levels to predict the risk of mortality in patients with acute coronary syndromes. N Engl J Med 1996;335:1342 – 9. [4] Rao SV, Ohman EM, Granger CB, Armstrong PW, Gibler WB, Christenson RH, et al. Prognostic value of isolated troponin elevation across the spectrum of chest pain syndromes. Am J Cardiol 2003;91:936 – 40. [5] de Winter RJ. C-reactive protein and cardiac troponin for early risk stratification in patients with acute coronary syndromes. Clin Chim Acta 2001;311:53 – 6. [6] Mockel M, Gerhardt W, Heller Jr G, Klefisch F, Danne O, Maske J, et al. Validation of NACB and IFCC guidelines for the use of cardiac markers for early diagnosis and risk assessment in patients with acute coronary syndromes. Clin Chim Acta 2001;303:167 – 79. [7] Mulvihill NT, Foley JB. Inflammation in acute coronary syndromes. Heart 2002;87:201 – 4. [8] Hunter T. Protein kinases and phosphatases: the yin and yang of protein phosphorylation and signaling. Cell 1995;80(27): 225 – 36. [9] Shanley TP. Phosphatases: counterregulatory role in inflammatory cell signaling. Crit Care Med 2002;30(Suppl. 1):S80 – 8. [10] Takeishi Y, Chu G, Kirkpatrick DM, Li Z, Wakasaki H, Kranias EG, et al. In vivo phosphorylation of cardiac troponin I by protein kinase Cbeta2 decreases cardiomyocyte calcium responsiveness and contractility in transgenic mouse hearts. J Clin Invest 1998;102:72 – 8. [11] Labugger R, Organ L, Collier C, Atar D, Van Eyk JE. Extensive troponin I and T modification detected in serum from patients with acute myocardial infarction. Circulation 2000; 102:1221 – 6. [12] Burkart EM, Sumandea MP, Kobayashi T, Nili M, Martin AF, Homsher E, et al. Phosphorylation or glutamic acid substitution at protein kinase C sites on cardiac troponin I differen-
70
D. Bar-Or et al. / Clinica Chimica Acta 362 (2005) 65–70
tially depress myofilament tension and shortening velocity. J Biol Chem 2003;278:11265 – 72. [13] Katrukha A, Bereznikova A, Filatov V, Esakova T. Biochemical factors influencing measurement of cardiac troponin I in serum. Clin Chem Lab Med 1999;37:1091 – 5. [14] Lindhout DA, Li MX, Schieve D, Sykes BD. Effects of T142 phosphorylation and mutation R145G on the interaction of the
inhibitory region of human cardiac troponin I with the Cdomain of human cardiac troponin C. Biochemistry 2002;41:7267 – 74. [15] McDonough JL, Arrell DK, Van Eyk JE. Troponin I degradation and covalent complex formation accompanies myocardial ischemia/reperfusion injury. Circ Res 1999;84:9 – 20.