Nephrology Dialysis Transplantation

Transcription

1 Nephrol Dial Transplant (1999) 14: Original Article Nephrology Dialysis Transplantation Prognostic value of cardiac troponin T and I elevations in renal disease patients without acute coronary syndromes: a 9-month outcome analysis Martin Möckel1, Ralf Schindler1, Ludwig Knorr1, Christian Müller1,2, Günther Heller Jr1,3, Thomas V Störk1,4 and Ulrich Frei1 1Department of Nephrology/Intensive Care Medicine and 2Clinical Chemistry Institute, Charité/Virchow-Klinikum, Berlin, 3Institute of Medical Sociology and Social Medicine, Philipps-University of Marburg, and 4Department of Cardiology/ Angiology/Intensive Care Medicine, Karl Olga Hospital, Stuttgart, Germany Abstract patients are of questionable value for risk stratification, Background. Moderate elevations of cardiac troponin most probably due to unspecific elevations. (Tn) T, up to levels presumably diagnostic for minor myocardial damage, are suspected to be false positive Key words: haemodialysis; outcome; renal disease; tro- in nearly 0.3 of end-stage renal disease ( ESRD) patients ponin T; troponin I undergoing haemodialysis ( HD). It is not clear whether cardiac TnI is superior to TnT in those patients, if differences between ESRD and pre-esrd occur, and Introduction what the prognostic meaning of these troponin elevations might be. Subjects and methods. We examined 40 chronic renaldiagnosis of acute coronary syndromes (ACS) and Cardiac troponins have recently been utilized in the disease patients [56.4 SD 13.9 years; 22 male, 18 female) without evidence of an acute coronary synmarkers such as CK and CKMB [1]. Troponin T, C provide major advantages compared to conventional drome (ACS) for at least 28 days prior to the investiand I are part of the troponin tropomyosin complex gation. Cardiac status was determined by history, physical examination, ECG and echocardiography. of the contractile apparatus in myocytes [2]. TnI Patients were divided into subgroups with HD (n=20) ( Da) and T ( Da) are coded by different and without HD (n=20). Patients without HD had a genes in skeletal and cardiac muscle, and therefore the mean creatinine clearance (CC) of ml/min. Tn development of cardiospecific antibodies is possible. were measured by immunoassay techniques. TnT was To overcome the lack of specificity of the conventional compared to two different TnI tests ( TnID, TnIB), cardiac markers (creatine kinase, myoglobin), troponin CK/CKMB activity and myoglobin (MYO) concentraspecificity for cardiac muscle was shown. T [3,4] and troponin I tests [5 10] were developed and tions. In all patients, a 9-month follow-up for acute myocardial infarction, re-hospitalization, and death The clinical value of cardiac troponin measurements was completed. were evaluated in large populations with acute Results. None of the troponins significantly predicted myocardial infarction [5,11 13] with and without conpatient outcome. Tn did not correlate with CC comitant skeletal muscle damage [14,15]. Reference (r<0.6). Applying the lowest reported threshold values values of troponin T and I are very low, mostly below for all tests in the HD group, 0.3 patients were positive the detection limit. for TnT, 0.55 patients were positive for TnID, and Troponins were shown to be useful for the diagnosis 0.15 for TnIB. In the group without HD, 0.2 patients of perioperative AMI [ 14,15]. Nevertheless, the diawere positive for TnT and TnID and 0.1 for TnIB. gnostic specificity of the troponin tests is far below 1 Conclusions. Moderate elevations of cardiac troponins in unselected patient groups [11,16]. This is mainly are common in clinically stable patients with renal due to the fact that patients with severe, unstable disease and are neither diagnostic for an acute coronary angina pectoris can show elevations of cardiac tropon- syndrome nor predictive of outcome. It is concluded ins without having myocardial infarction according to that increased troponins in asymptomatic renal WHO criteria. This observation led to the introduction of minor myocardial damage ( MMD) diagnosis. These MMD patients have a worse prognosis compared to other angina pectoris groups who are troponin Correspondence and offprint requests to: DrM.Möckel MD, Charité/ Campus Virchow-Klinikum, Department of Nephrology/Intensive negative [17 21]. Other markers, such as CKMB, are Care Medicine, Augustenburger Platz 1, Berlin, Germany. not as effective in the identification of MMD patients European Renal Association European Dialysis and Transplant Association

2 1490 M. Möckel et al. Patients with renal disease, and especially those laboratory. For cardiac troponin I and T determination, 5 ml undergoing haemodialysis therapy, are likely to have lithium-heparin blood was centrifuged at 5 C (10 min, 3000 coronary artery disease [ 22]. Since patients with r.p.m.). Three aliquots of at least 500 ml were immediately chronic renal failure frequently have elevated CK or frozen at 80 C. The dialysis regimen was at the discretion of the physician in charge, who was not involved in the study. myoglobin levels [23], measurement of cardiac troponins may be advantageous. Clinical observations suggest that troponin T yields false-positive values for the Analytical methods diagnosis of acute coronary syndromes [ 24 33] but Plasma concentrations of sodium, potassium, and creatinine less false-positive troponin I values [12,28 31,33,34] were measured with a DAXA 72 clinical chemistry analyser were observed in patients with end-stage renal disease (Bayer) using routine procedures. CK, CKMB, and myoglo- ( ESRD) undergoing haemodialysis therapy ( HD) or bin were measured with a Hitachi 911A system using system in pre-esrd patients. Most of these references are test kits (Roche Diagnostics). The CK activity was deter- merely case reports and not controlled studies and do mined according to the standard procedures of the German not indicate crucial aspects such as patient characterto 70 U/l for females and 80 U/l for males. Normal levels Society for Clinical Chemistry with upper reference limits up istics, how acute coronary syndromes were excluded, for CKMB, determined kinetically after immunoinhibition, and the threshold for the troponin values. It remains are up to 10 U/l or less than 0.06 of total CK activity. uncertain whether the false-positive diagnosis of min- Myoglobin was measured by a latex amplified immunoturbidimal myocardial damage is due to the threshold values imetric test with a reference range up to 70 mg/l. The biochemchosen and whether cardiac troponins are of prognostic ical criteria for acute myocardial infarction (AMI ) are 100 value in renal patients. The aim of the present study U/l CK with more than 0.06 of CKMB and 90 mg/l was to evaluate whether there are elevations in cardiac myoglobin. troponin T and troponin I (measured by two different Cardiac troponins were measured from frozen samples tests) in patients with renal disease without evidence (see above) which were rapidly thawed at 21 C, centrifuged of an acute coronary syndrome (ACS) using the latest for 5 min at 2500 r.p.m. and immediately processed with the generation immunotests. Subgroup analysis of creatinine 3different troponin assays. clearance correlation with troponin values in patients with chronic renal failure before dialysis and Cardiac troponin T the influence of the dialysis modality was performed. This was measured with an Elecsys 2010A analyser using a A 9-month follow up for acute myocardial infarction, newly developed immunoassay employing two monoclonal re-hospitalization and death addressed the question murine antibodies specific for cardiac troponin T and electrowhether troponin elevations are of diagnostic/pro- chemiluminescence as the detection technique ( Roche gnostic value. Diagnostics; 4). The sensitivity was 0.01 mg/l, the detection range was mg/l and the interassay CV was at low level concentrations (mean 0.1 mg/l ), at inter- Subjects and methods mediate levels (mean 0.33 mg/l ) and at high levels (mean 5.9 mg/l ). The analytical procedure is completed Patients within 9 min. Cardiac troponin I was measured by means of two different We included 40 consecutive hospitalized patients with plasma commercially available tests: DADE (TnID), and Behring creatinine elevations above 1.5 mg/dl (pre-end-stage renal Diagnostics ( TnIB). disease, pre-esrd; n=20) or who were on chronic haemo- The immunofluorescence assay, Stratus IIA, employing a dialysis therapy for at least 1 month ( ESRD; n=20). monoclonal capture antibody and a alkaline phosphatase- Exclusion criteria were: haemoglobin below 8.0 g/dl, age labelled secondary antibody was used for TnID determinaabove 80 years or severely compromised clinical status, tion. The analytical sensitivity was 0.35 mg/l, the measuring known neoplastic disease, acute renal failure, any angina range was mg/l and the interassay CV was at pectoris symptoms occuring during the last 14 days or an low level concentrations (mean 5.1 mg/l ), at interacute coronary syndrome (unstable angina, acute myocardial mediate levels (mean 18.5 mg/l ) and at high levels infarction; ACS) in the last 4 weeks. All patients gave (mean 33.4 mg/l ). The analytical procedure is completed informed consent. Table 1 shows the clinical characteristics within 10 min. of the patients and the renal and concomitant diseases are The immunofluorescence assay, Behring OPUS PlusA, listed in Table 2. The mode of haemodialysis, especially the containing two polyclonal goat antibodies against cardiac type of membrane used was left to the discretion of the troponin I and was used to measure TnIB. The analytical attending physician. The membranes used were recorded and sensitivity was 0.5 mg/l, the measuring range was listed below mg/l, and the interassay CV was at low level concentrations (mean 3.7 mg/l ), at intermediate levels Protocol (mean 19.5 mg/l ) and at high levels (mean 99.1 mg/l). The analytical procedure is completed within 20 min. Cardiac history, 12-lead ECG and echocardiography were performed in all patients to determine the cardiac status. Threshold values Blood samples were taken immediately before and after haemodialysis, and in pre-esrd patients once at 8.00 a.m. The selection of cut-off values is essential for the comparison Plasma concentrations of sodium, potassium, creatinine, of different analytical methods. As described above, the creatine-kinase and its isoenzyme MB (CKMB), and myoglo- characteristics of the three tests differ widely. First, the range bin were measured within 2 h in our central clinical chemistry of troponin T is different from that of TnI due to assay

3 Cardiac troponins in renal disease 1491 Table 1. Clinical characteristics of the 40 patients Parameter ESRD Pre-ESRD Age (years) 51.5 (28 75) 63.5 (27 78)** Gender (m5f) :8 Body-mass-index (kg/m2) ( ) ( )* ESRD duration (years) 2.5 ( ) Creatinine-clearance (ml/min) 13.4 ( ) Sodium (mmol/l ) 138 ( ) 138 ( ) Potassium (mmol/l ) 5.0 ( ) 4.3 ( )** Creatinine (mg/dl ) 7.7 ( ) 2.8 ( )*** Admission diagnoses Shunt surgery 0.4 ( ) 0.05 ( ) Parathyroid resection 0.25 ( ) 0 ( )# General surgery 0.15 ( ) 0.1 ( ) PA-PTA 0.05 ( ) 0.1 ( ) Progression of renal insufficiency 0 ( )# 0.35 ( ) Hypertension+ 0 ( )# 0.2 ( ) Diabetes mellitus+ 0 ( )# 0.15 ( ) Pneumonia 0.1 ( ) 0 ( )# Acute pancreatitis 0.05 ( ) 0.05 ( ) Median and range in quantitative variables; decimal fraction and 0.95 confidence interval in qualitative variables; ESRD, end-stage renal disease; *P<0.05; **P<0.01; ***P<0.001 vs ESRD (values before haemodialysis); PA-PTA, percutaneous transluminal angioplasty of peripheral arteries; +, severely aggravated; #, one-sided, confidence interval. frequencies were too small for Chi-square testing. The distribution of quantitative variables were not normally distrib- uted, and therefore non-parametric tests were chosen. The values are listed as medians and range. To determine whether troponin values independently predict outcome, a logistic regression was performed including the variables age, sex, group ( ESRD or pre-esrd) and troponin. The sample size was estimated using MedCalcA statistical software (alpha-error 0.05, beta-error 0.2) on the basis of event rates for troponin T positive and negative patients reported in the literature. Event rates lay between and for troponin T positive and between and for troponin T negative patients [37,38], depending on the population selected. The calculations resulted in sample sizes of patients. Therefore, we decided to include at least 40 patients. Further calculations and graphics were performed on a personal computer using SPSSA Inc. statistical software. callibration issues. Second, the plasma of healthy blood donors does usually not contain measurable concentrations of TnT. Third, the two different TnI tests are not standardized. Thus, the analytical results do not represent the same amount of substance and it is not possible to calculate a fixed relation between the test results. Three different threshold values for the troponin markers were used as described in the literature and as outlined by the manufacturers of the assays. The cut-off values indicating minor myocardial damage (MMD) and increased risk of future cardiac events were 0.1 mg/l for TnT [17 19], 0.4 mg/l for TnID [20] and 2.0 mg/l [21,35] and 1.6 mg/l [36 ] for TnIB. The threshold for acute myocardial infarction (AMI) was 0.5 mg/l for TnT [11], 3.1 mg/l for TnID [12] and 2.5 mg/l for TnIB [13]. The generally suggested cut-off values for myocardial injury (including both AMI and MMD) which could be obtained by the companies at the time of analytical testing were 0.1 mg/l for TnT, 1.5 mg/l for TnID, and 2.0 mg/l for TnIB. Unpublished data from patients with suspected AMI (A. Wu, personal communication) suggest an infarction cut-off of 0.2 mg/l for TnT which was additionally included in the analysis. Follow up The composite primary end-point: acute myocardial infarction, death from all causes, and re-hospitalization was recorded for all patients. The median follow up time was 9 (7 11) months. Statistics Differences between quantitative variables were analysed by Wilcoxon s test. Linear correlation coefficients (Pearson) were calculated. Qualitative variables are displayed as decimal fraction and 0.95 confidence intervals and were analysed by chi-square test. No statistically approved result regarding the differences between decimal fractions of the various troponin tests could be obtained because the expected Results The levels of cardiac markers are listed in Table 3. The type of dialyser membrane did not influence the cardiac markers which were not different before and after dialysis therapy or between the ESRD/pre-ESRD groups. Low-flux membranes were used in 11 patients (cuprophane, n=3; polycarbonate n=2; haemophane, n=2; polysulphone, n=4) and high-flux membranes in nine individuals (polysulphone, n=3; polyamide, n=6). Table 4 shows the decimal fraction of values above the different cut-off values for TnI and TnT tests. In the ESRD group, the troponin test was regarded as positive if at least one value before or after haemodialysis was above the threshold. The diagnostic meaning of different troponin tests in relation to each other are demonstrated in Figures 1 and 2. Troponin

4 1492 M. Möckel et al. Table 2. Patient history and illnesses ( ; P=0.22) for TnIB, 3.22 (0.6 17; P=0.168) for TnID and 1.03 ( ; P=0.969) for TnT. ESRD Pre-ESRD Analysis of the three components of the combined end-point and a combination of two did not show any Angina pectoris significant results. No The creatinine clearance correlates with TnID (r= ( ) ( ) 0.5, P=0.023) only. Less than 0.5 of values were found Stable ( ) ( ) above the detection limit. The significance of the CAD correlation coefficients mentioned above depends on ( ) ( ) one extreme value with high CC and Tn. Therefore, Previous MI ( ) ( ) this correlation was not conclusive. Hypertension ( ) ( ) Diabetes mellitus Discussion ( ) ( ) HLP The present study demonstrates that troponin T and ( ) ( ) HFBG troponin I are moderately and frequently elevated in ( ) ( ) uraemic patients without clinical evidence of an acute HU coronary syndrome. Troponin T [ 17 19], troponin I ( ) ( ) by Dade ( TnID, [20]) and troponin I by Behring Family history of CAD (TnIB, [21]) are shown to be useful in risk stratification ( ) ( ) Smoking of unstable angina pectoris patients by diagnosing ( ) ( ) minor myocardial damage and to have high sensitivity Package median years 15 [0 240] 20 [0 50] and specificity for acute myocardial infarction. All [range] three cardiac markers are elevated beyond the minor Renal diagnoses Diabetic nephropathy myocardial damage threshold in uraemic patients ( ) ( ) (Tables 3, 4, Figures 1, 2), but only a few values were Goodpasture syndrome above the acute myocardial infarction cut-off ( Figures ( ) ( )# 1, 2). IgA nephritis ( ) ( )# To evaluate the predictive value of these elevations, Post-streptococcal-GN a 9-month follow up of patients was completed. No ( ) ( ) significant prediction of death, acute myocardial infarc- Chronic GN tion or rehospitalization by either troponin T or I ( ) ( )# could be shown by logistic regression analysis. In Alport syndrome ( ) ( )# contrast, Apple et al. [39] investigated 16 dialysis PKD patients retrospectively and found an association ( ) ( ) between troponin T, I and CKMB elevations and fatal Renal artery stenosis acute myocardial infarction in four of those patients. ( )# ( ) Analgesic nephropathy The 12 surviving patients of his study had TnT eleva- ( ) ( ) tions at lower levels but no TnI values above the minor Hypertensive nephros myocardial damage cut-off. In our prospective study a clerosis ( ) ( ) 9-month follow-up was completed in ESRD and pre- ESRD patients for the first time. We could not show Values are decimal fractions and 0.95 confidence intervals; #, one- that these markers have a significant prediction sided, confidence interval; ESRD, end-stage renal disease; of mortality, acute myocardial infarction, or re- CAD, coronary artery disease; MI, myocardial infarction; HLP, hyperlipoproteinemia (serum cholesterol>200 mg/dl ); HFBG, high hospitalization. Although the number of patients fibrinogen (>300 mg/dl); HU; high uric acid (>10 mg/dl); GN, evaluated is higher than in previous studies, the number glomerulonephritis; PKD, polycystic kidney disease. is still relatively low for outcome analysis. Table 5 contains the cardiac troponin levels of seven patients T values are plotted against TnID (Figure 1) and TnIB who died or suffered from acute myocardial infarction ( Figure 2). in the follow-up. As shown in Figures 1, 2 and Table 5, Table 5 shows the troponin values of seven patients the different troponins do not correlate and no test is (0.18) who died or experienced acute myocardial clearly superior. This may be due to the fact that infarction during the follow up period. Twenty-two troponins are released as T-I-C complex, followed by patients (0.55) were re-hospitalized; five (0.13) for degradation in the circulation. The different test systems cardiac causes (acute myocardial infarction, pulmonary have various affinities for the T-I-C, I-C complex oedema, hypotension, pleural effusion). Using the or free troponin T [40]. Some events were correctly different cut-off values defined above, the logistic predicted by troponin elevations and possibly larger regression showed no significant prediction of the trials would be able to show significance. primary endpoint by troponin values. The lowest Although more patients have to be observed for a applicable cut-offs odds ratios (0.95 CI) were: 4.57 longer period to gain more evidence regarding the

5 Cardiac troponins in renal disease 1493 Table 3. Cardiac markers Parameter ESRD (n=20) Pre-ESRD (n=20) Before HD After HD Creatine kinase [ U/L] 13 (5 116) 13 (6 87) 25.5 (3 80) CKMB [U/L] 0(0 6) 0(0 0) 0(0 0) Myoglobin [mg/l] (58 496) (75 438) 94.5 (52 284)# Troponin T [mg/l] (0 1.6) (0 1.85) ( ) Troponin I D [mg/l] 0.1 (0 1.9) 0.15 (0 2) 0 (0 4.1) Troponin I B [mg/l] 0 (0 8.78) 0 (0 12.6) 0 (0 4.64) ESRD, end-stage renal disease, #P= vs ESRD. Table 4. Decimal fraction of cardiac troponin values above the different threshold values Method TnID TnIB TnT Cut-off [mg/l ] ESRD Pre-ESRD 0.20* ESRD, end-stage renal disease; TnI, troponin I; D, DADE; B, Behring; TnT, Troponin T; *P<0.05 vs ESRD. Fig. 1. Scattergram of troponin T and troponin I D (DADE) values. The lines represent the lowest minor myocardial damage (dotted) and infarction (solid) threshold values. The figure shows that troponin elevations above minor myocardial damage cut-off values are common but higher values are rare. Troponin T and I elevations do not significantly correlate with each other. Fig. 2. Scattergram of troponin T and troponin I B (Behring) values. The lines represent the minor myocardial damage (dotted) and infarction (solid) threshold values. The figure shows that troponin elevations above minor myocardial damage cut-off values are common but higher values are rare. Troponin T and I elevations do not significantly correlate with each other. prognostic implications of troponins in renal disease, dialysis patients in their study, 0.17 had elevations the explanation of our findings could be a molecular using a second-generation TnT assay [4], but only 0.04 change in muscle properties in uraemia. No cardiac in the TnID test. The cut-off in the McLaurin study troponin I isoform (fetal or adult) has been reported was defined as 0.8 mg/l. This is not appropriate, because for skeletal muscle, but several reports exist for troponin a 0.4 mg/l cut-off value of TnID was shown to be of T. Bodor et al. [41] described cardiac troponin prognostic value in the study by Antman and col- T expression in regenerating (polymyositis, Duchenne leagues [20]. The second major concern regarding the muscular dystrophy) and normal skeletal muscle. study by McLaurin et al. [33] is that four of the five Unfortunately the antibody used by Bodor and colleagues patients had serum TnT values below the detection was different from the monoclonal antibodies limit. The fifth patient had a serum concentration of in the improved second-generation test used in our 0.1 mg/l, which is exactly the threshold value for minor study. These antibodies have been shown to be nearly myocardial damage. Therefore, skeletal muscle expression 100% cardiospecific in Western blot analysis by the of cardiac TnT is not shown to be the cause of company. McLaurin et al. [33] have shown that cardiac plasma troponin T elevations in asymptomatic HD troponin T, but not troponin I was expressed in skeletal patients. This is supported by new work of Ricchiutti muscle biopsies in four of five dialysis patients. Of 24 et al. [42], who showed that the second-generation

6 1494 Table 5. Cardiac troponins of seven patients who died or suffered from acute myocardial infarction M. Möckel et al. Patient Group Death Cause of death AMI TnID [mg/l ] TnIB [mg/l] TnT [mg/l] SK HD yes AMI yes <0.35/< /< /0.012 ZH No HD no yes <0.35 <0.5 <0.01 AH No HD yes SD no < SI No HD yes Pneumonia no <0.35 < RI No HD no yes 0.6 < SG No HD yes ICH no <0.35 <0.5 <0.01 LH No HD yes ICH no HD, haemodialysis (values before/after); AMI, confirmed acute myocardial infarction ( WHO criteria); SD, sudden death; ICH, intracranial haemorrhage; TnI, troponin I; D, DADE; B, Behring; TnT, troponin T, Roche Diagnostics; values above the minor myocardial damage cut-off are shown in bold. troponin T assay does not detect fetal and regenerative difference was only significant for TnID with its lowest troponin T. Another important issue is addressed by cut-off ), mortality was higher in the pre-esrd group. work of Anderson et al. [43] who show that the failing This may be due to a selection bias because the age human heart expresses more of the troponin T isoform was higher and the degree of illness seemed to be more 2 compared to normal cardiac tissue. Again in this severe in the pre-esrd group ( Table 1). study, no cardiac troponin T was found to be expressed Although we could show that several patients with in skeletal muscle specimens [43]. Therefore, it is renal failure have non-diagnostic elevations of troponin, unlikely that detection of troponins in asymptomatic determination of cardiac troponins is of possible uraemic patients is due to molecular changes in skeletal benefit in patients with acute coronary syndromes. A or cardiac muscle. recently published paper [46] shows a sensitivity of 94% and specificity of 100% for cardiac troponin I in Are troponin elevations in asymptomatic uraemic renal insufficiency patients. Unfortunately the conclupatients an analytical problem? sions from this paper by Martin et al. are limited due to incomplete diagnostic studies in several of the 56 As it is known from other biochemical analyses (e.g. patients. Larger studies including consecutive uraemic parathyroid hormone), especially in HD patients, the patients, with and without angina pectoris symptoms, metabolism of troponins may be altered, and detection have to be performed to evaluate cardiac troponin use of troponin degradation products may lead to analyt- in renal disease patients. ical results judged to be false positive [44]. However, troponin degradation products have not been described in the literature up to now. Additionally, binding sites Conclusions of test antibodies could be altered due to changes in protein kinase C (PKC) properties in renal patients as In uraemic patients without signs of acute cardiac PKC phosphorylates troponin T and I at many differated disease, cardiac troponin I and T are frequently elevent sites [45]. above threshold values which are usually dia- Another explanation for higher troponin levels in gnostic for minor myocardial damage. Moderate uraemic patients may be asymptomatic chronic troponin elevations do not predict outcome in these uraemic myopathy with continuous release of troponasymptomatic patients. It is concluded that increased troponins in ins. This explanation is speculative and no clinical or renal patients are of questionable value experimental data are available at present to support for risk stratification most probably due to unspecific this hypothesis. elevations. We have shown that TnT and TnI may be asymptomatically elevated in renal patients. Figures 1 and 2 Acknowledgements. We thank Dr N. Hegner from Behring Diagnostics and Dr E. Spanuth from Roche Diagnostics Germany illustrate that only a few renal patients remained for the supplementation of the analytical procedures; we are also troponin positive when the threshold values for acute grateful to P. Levett, Berlin, and Prof. A. Wu from Hartford myocardial infarction were used. These high cut-off Hospital, CT, USA for carefully reading the manuscript and for values were not predictive of prognosis. The decimal their helpful comments. fraction of positive patients varies among the different troponins and between patient groups ( ESRD/pre- ESRD). Nevertheless, only TnID values of the ESRD References and the pre-esrd group were significantly different 1. Katus HA. Cardiac troponins in different disease conditions. when a 0.4 mg/l cut-off was used. Editorial comment. Intensivmed 1997; 34: The grade of renal function loss did not relevantly 2. Katus HA, Scheffold T, Remppis A, Zehlein J. Proteins of the troponin complex. Lab Med 1992; 23: correlate with troponin values (r<0.6). Although 3. Katus HA, Looser S, Hallermayer K et al. Development and in ESRD patients seem to have relatively more troponin vitro characterization of a new immunoassay of cardiac troponin elevations compared to the pre-esrd group (the T. Clin Chem 1992; 38:

7 Cardiac troponins in renal disease Baum H, Braun S, Gerhardt W et al. Multicenter evaluation of 26. Li D, Jialal I, Keffer J. Greater frequency of increased cardiac a second-generation assay for cardiac troponin T. Clin Chem troponin T than increased troponin I in patients with chronic 1997; 43: renal failure. Clin Chem 1996; 42: (letter) 5. Bertinchant JP, Larue C, Pernel I et al. Release kinetics of 27. Collinson PO, Stubbs PJ, Rosalki SB. Cardiac troponin T in serum cardiac troponin I in ischemic myocardial injury. Clin renal disease. Clin Chem 1995; 41: (letter) Biochem 1996; 29: Hafner G, Thome-Kromer B, Schaube J et al. Cardiac troponins 6. Bodor GS, Porter S, Landt Y, Ladenson J. Development of in serum in chronic renal failure. Clin Chem 1994; 40: monoclonal antibodies for an assay of cardiac troponin-i and ( letter) preliminary results in suspected cases of myocardial infarction. 29. Bhayana V, Gougoulias T, Cohoe S, Henderson AR. Clin Chem 1992; 38: Discordance between results for serum troponin T and troponin 7. Zaninotto M, Altinier S, Lachin M, Carraro P, Plebani M. I in renal disease. Clin Chem 1995; 41: Fluoroenzymometric method to measure cardiac troponin I in 30. McLaurin MD, Apple FS, Herzog CA, Sharkey SW. Cardiac sera of patients with myocardial infarction. Clin Chem 1996; 42: troponin I, T and CK-MB in chronic hemodialysis patients Circulation 1995; 92 [Supplement I]: I Larue C, Calzolari C, Bertinchant JP, Leclercq F, Grolleau R, 31. Apple FS, Dahlmeier B, Sharkey SW. Serum cardiac troponin Pau B. Cardiac-specific immunoenzymometric assay of troponin I and T concentrations in end stage renal disease, type I diabetes I in the early phase of acute myocardial infarction. Clin Chem patients. Clin Chem 1995; 41: Frankel WL, Herold DA, Ziegler TW, Fitzgerald RL. Cardiac 1993; 39: troponin T is elevated in asymptomatic patients with chronic 9. Fagan GJ. Entwicklung und Charakterisierung der Troponinrenal failure. Am J Clin Pathol 1996; 106: Antikörper für das Opus Immunoassay-System. In: 33. McLaurin MD, Apple FS, Voss EM, Herzog CA, Sharkey SW. Hossmann V, Thomas L (eds). Diagnostik und Therapie Kardialer Cardiac troponin I, cardiac troponin T and creatine kinase Erkrankungen. TH-Books-Verlagsgesellschaft, Frankfurt, 1997; Mbyte in dialysis patients without ischemic heart disease Evidence of cardiac troponin T expression in skeletal muscle. 10. Cummins B, Auckland ML, Cummins P. Cardiac-specific tro- Clin Chem 1997; 43: ponin I radioimmunoassay in the diagnosis of acute myocardial 34. Trinquier S, Flécheux O, Bullenger M, Castex F. Highly specific infarction. Am Heart J 1987; 113: immuno-assay for cardiac troponin I assessed in noninfarct 11. Katus HA, Remppis A, Neumann FJ et al. Diagnostic efficiency patients with chronic renal failure or severe polytrauma. Clin of troponin T measurements in acute myocardial infarction. Chem 1995; 41: (letter) Circulation 1991; 83: Hegner N, Baum H, Eller T et al. Multizentrische Evaluierung 12. Adams JE III, Bodor GS, Dávila-Román VGet al. Cardiac von OPUS Troponin I im Vergleich zu Myoglobin und der troponin I. A marker with high specifity for cardiac injury. CK-MB-Konzentration bei kardialen Erkrankungen. Clin Lab Circulation 1993; 88: ; 43: Wu AHB, Feng YJ, Contois JH, Pervaiz S. Comparison of 36. Wu AH, Feng YJ, Contois JH, Azar R, Waters D. Prognostic myoglobin, creatine kinase-mb, and cardiac troponin I for value of cardiac troponin I in patients with chest pain. Clin diagnosis of acute myocardial infarction. Ann Clin Lab Sci 1996; Chem 1996; 42: (letter) 26: Lopez-Jimenez F, Goldman L, Sacks DB et al. Prognostic value 14. Adams JE III, Sigard GA, Allen BT et al. Diagnosis of perioperup of cardiac troponin T after noncardiac surgery. 6-month followative myocardial infarction with measurement of cardiac troponin data. JACC 1997; 29: I. N Engl J Med 1994; 330: Lindahl B, Venge P, Wallentin L for the FRISC study group. 15. Hake U, Schmid FX, Iversen S et al. Troponin T a reliable Troponin T identifies patients with unstable coronary artery marker of perioperative myocardial infarction? Eur J Cardio- disease who benefit from long-term antithrombotic protection. thorac Surg 1993; 7: JACC 1997; 29: Möckel M, Maske J, Danne O et al. Prospective validation of 39. Apple FS, Sharkey SW, Hoeft P et al. Prognostic value of serum a new protocol for the diagnosis of acute myocardial infarction cardiac troponin I and T in chronic dialysis patients. A 1-year using myoglobin, troponin Ic and CK-MB mass. Final results. outcomes analysis. Am J Kidney Dis 1997; 29: Eur Heart J 1997; 18: Wu AHB, Feng Y-J, Moore R et al. For the AACC Subcommittee on ctni standardization. Characterization of 17. Hamm CW, Ravkilde J, Gerhardt W et al. The prognostic value cardiac troponin subunit release into serum after acute myocardof serum troponin T in unstable angina. N Engl J Med 1992; ial infarction and comparison of assays for troponin T and I. 327: Clin Chem 1998; 44: Lindahl B, Venge P, Wallentin L for the FRISC Study Group. 41. Bodor GS, Survant L, Voss EM, Smith S, Porterfield D, Apple Relation between troponin T and the risk of subsequent cardiac FS. Cardiac troponin T composition in normal and regenerating events in unstable coronary artery disease. Circulation 1996; 93: human skeletal muscle. Clin Chem 1997; 43: Ricchiuti V, Voss EM, Ney A, Odland M, Anderson PAW, 19. Ohman EM, Armstrong PW, Christenson RH et al. Cardiac Apple FS. Fetal and regenerative cardiac troponin T isoforms troponin T levels for rise stratification in acute myocardial expressed in renal diseased skeletal muscle are not detected by ischemia. N Engl J Med 1996; 335: the second generation cardiac troponin T assay by Boehringer 20. Antman EA, Tanasijevic MJ, Thompson B et al. Cardiac-specific Mannheim. Clin Chem 1998; 44 [Supplement]: A116 troponin I levels to predict the risk of mortality in patients with 43. Anderson PAW, Malouf NN, Oakeley AE, Pagani ED, Allen acute coronary syndromes. N Engl J Med 1996; 335: PD. Troponin T isoform expression in humans. A comparison 21. Lüscher MS, Thygesen K, Ravkilde J, Heickendorff L for the among normal and failing adult heart, fetal heart, and adult TRIM study group. Applicability of cardiac troponin T and I and fetal skeletal muscle. Circ Res 1991; 69: for early risk stratification in unstable coronary artery disease. 44. Burckhardt P, Jaeger P, Jacquet AF, Wauters JP. Fate of Circulation 1997; 96: parathyroid hormone during hemodialysis and ultrafiltration. 22. Wheeler DC. Cardiovascular disease in patients with chronic Horm Res 1985; 21: renal failure. Lancet 1997; 348: Jideama NM, Noland TA, Raynor RL et al. Phosphorylation 23. Jaffe AS, Ritter C, Meltzer V, Harter H, Roberts R. Unmasking specificities of protein kinase C isoenzymes for bovine cardiac artifactual increases in creatine kinase isoenzymes in patients troponin I and troponin T and sites within these proteins and with renal failure. J Lab Clin Med 1984; 104: regulation of myofilament properties. J Biol Chem 1996; 271: 24. Katus HA, Haller C, Müller-Bardorf M, Scheffold T, Remppis A. Cardiac troponin T in end-stage renal disease 46. Martin GS, Becker BN, Schulman G. Cardiac troponin-i accurpatients undergoing chronic maintenance haemodialysis. Clin ately predicts myocardial injury in renal failure. Nephrol Dial Chem 1995; 41: ( letter) Transplant 1998; 13: Li D, Keffer J, Corry K, Vazquez M, Jialal I. Nonspecific elevation of troponin T levels in patients with chronic renal Received for publication: failure. Clin Biochem 1995; 28: Accepted in revised form: