Measurement of circulating troponin Ic enhances the prognostic value of C-reactive protein in haemodialysis patients

Size: px

Start display at page:

Download "Measurement of circulating troponin Ic enhances the prognostic value of C-reactive protein in haemodialysis patients"

Kory Strickland
6 years ago
Views:

1 NDT Advance Access published July 13, 2004 Nephrol Dial Transplant (2004) 1 of 6 DOI: /ndt/gfh365 Original Article Measurement of circulating troponin Ic enhances the prognostic value of C-reactive protein in haemodialysis patients Alexandre Boulier 1, Isabelle Jaussent 2, Nathalie Terrier 1, Franc ois Maurice 3, Jean-Pierre Rivory 3, Lotfi Chalabi 5, Anne-Marie Boularan 1,Cécile Delcourt 2, Anne-Marie Dupuy 1, Bernard Canaud 4,5,6 and Jean-Paul Cristol 1 1 Biochemistry Laboratory and 4 Nephrologie, Centre Hospitalier Universitaire de Montpellier, 2 INSERM Unite 500 Epidemiology of chronic diseases and ageing, 3 Centre He modialyse Languedoc Me diterrane e, 5 AIDER and 6 Renal Research and Training Institute, Montpellier, France Abstract Background. Cardiac Troponin I (ctni) levels are considered an important diagnostic tool in acute coronary events. They could be of predictive value in haemodialysis (HD) patients. However, the relationship between ctni and the HD-induced inflammatory state remains unclear. The aim of this study was to explore the prognostic relevance to all-cause and cardiovascular mortalities in HD patients of ctni, in combination with highly sensitive C-reactive protein (hs-crp) levels. Methods. We measured ctni and hs-crp at baseline (March 10 to November 16, 2001) in 191 HD patients without clinical signs of acute coronary artery disease [median age 66.7 years (range ), 94 females, 97 males]. We used a ctni concentration with a total imprecision of 10% (0.03 mg/l), determined in the laboratory, as the analytical threshold value. Patients were followed for mortality until 1 January, 2003 (median follow-up 418 days). The adjusted relative risks (RRs) of death and 95% confidence intervals (CIs) were estimated using Cox proportional hazard models. Results. A significant proportion (25.1%) of patients had elevated CTnl, 0.03 mg/l; 40.3% of patients had CRP concentrations 10 mg/l. During follow-up, 29 patients died, 44.8% due to cardiac causes. Elevated ctni or CRP levels were associated with increased mortality [RR adjusted for age, sex and duration of dialysis 4.2 ( ) for ctni 0.03 mg/l and 3.6 ( ) for CRP 10 mg/l], ctni being particularly predictive of cardiovascular death. Moreover, the combination of elevated hs-crp (10 mg/l) and circulating ctni (0.03 mg/l) dramatically impaired Correspondence and offprint requests to: Professor Jean-Paul Cristol, Biochemistry Laboratory, Hô pital Lapeyronie, 371 Av. Doyen Gaston Giraud, Montpellier cedex 5, France. jp-cristol@chu-montpellier.fr Nephrol Dial Transplant ß ERA EDTA 2004; all rights reserved the HD survival rate [adjusted RR for all-cause mortality 16.9 ( )]. Conclusion. Circulating ctni was associated with poor prognosis, especially when combined with elevated CRP, strongly supporting the adoption of regular ctni testing in HD patients. Keywords: cardiovascular disease; haemodialysis; mortality; troponin Introduction Troponins I and T are two components of the threemember protein complex that plays an essential role in the contraction of striated muscle. Since cardiac proteins differ in amino acid sequence from skeletal muscle proteins, plasma troponins have been proposed as reliable indexes of myocardial injury [1 3]. The measurement of changes in the concentration of circulating troponin is now recognized as useful in diagnosing acute myocardiac infarction, including non-st segment elevation events, and as a prognostic factor in unstable angina. Several assays have been developed by different groups allowing rapid and quantitative determination of troponin concentration. Unfortunately, the lack of standardization in cardiac troponin I (ctni) assays has led to different norms and to confusion about levels indicative of myocardial infarction. Since coronary artery disease is highly prevalent in patients with end-stage renal failure (ESRF), and vascular events account for more than half of deaths in them, haemodialysis (HD) patients have to be monitored carefully to detect subclinical signs of cardiac damage. Previous studies have suggested that troponins could be a predictive tool in coronary heart

2 2of6 disease in ESRF [4 8]. However, troponin levels in the HD population could be influenced by factors other than ischaemic damage, such as retention due to lack of epuration (their molecular weights being Da), or uraemic cardiomyopathy [9]. In addition, it has become clear from epidemiological studies that an elevation of C-reactive protein (CRP) is strongly associated with coronary heart disease and with poor prognosis in ESRF [10 12]. Due to the uncertainty of the use of ctni assays in the dialysis population, threshold values should be determined in this population and compared with the predictive value of CRP. Our present study analysed prospectively the relationship between ctni values and all-cause or cardiovascular disease mortality, and compared it with CRP levels. Materials and methods Study design Our cohort comprised 191 HD patients conscripted from the three dialysis facilities in Montpellier (France) a hospitalbased facility (Lapeyronie University Hospital), a public non-profit association (AIDER) and a private dialysis clinic (CHLM). The included patients were evaluated for inclusion from October to November 2001, and were followed prospectively until 1 January, Patients and baseline data The subjects were on either standard HD or haemodiafiltration (HDF) treatments with on-line ultrapure bicarbonatebased dialysate for h per week. Their medical charts were reviewed to collect data on age, sex, weight, height, primary renal disease, time when started on dialysis, previous transplantation and duration of transplantation, history of diabetes, smoking habits and current medications. Hypertension was defined as a blood pressure 140/90 mmhg at dry weight, or a documented history of hypertension. The clinical characteristics of the 191 patients are summarized in Table 1. The group included 94 women and 97 men with a median age of 66.7 years ( ). Their median time on dialysis was 4.8 years ( ), and their averaged Kt/V was 1.46±0.02; 51 of them were smokers. All patients were stable on HD, as documented by the low incidence of hypotensive episodes during HD (<4% of episodes requiring saline infusion). CAD defined as documented angina pectoris or history of myocardial infarction was present in 62 patients at baseline. End point On 1 January, 2003, all subjects were re-evaluated by the physicians in the dialysis centres. The dates of death, transplantation or transfer to another dialysis centre were documented. The causes of death were classified as: cardiac events (myocardial infarction, congestive heart failure and sudden death) and non-cardiovascular events (infection, neoplasm and unknown causes). Table 1. Clinical characteristics of 191 ESRF patients Age (years) Median (range) 66.7 ( ) Gender Male n (%) 97 (50.8) Female n (%) 94 (49.2) Coronaropathy No n (%) 129 (67.5) Yes n (%) 62 (32.5) Diabetes No n (%) 152 (79.6) Yes n (%) 39 (20.4) Smoking No n (%) 140 (73.3) Yes n (%) 51 (26.7) HD modality Polysulfone B n (%) 109 (57.1) Polysulfone H n (%) 82 (42.9) Hypertension No n (%) 148 (77.5) Yes n (%) 43 (22.5) Time HD (years) Median (range) 4.8 ( ) Total cholesterol (mmol/l) Mean±SE 5.02±0.09 Triglycerides (mmol/l) Median (range) 1.7 ( ) CRP (mg/l) <5 n (%) 69 (36.1) 5 10 n (%) 45 (23.6) 10 n (%) 77 (40.3) ctni (mg/l) <0.03 n (%) 143 (74.9) n (%) 22 (11.5) 0.04 n (%) 26 (13.6) Laboratory analysis and procedures At a single mid-week dialysis session pre- and post-dialysis, blood samples were collected at baseline as a part of our routine patient follow-up and quality assurance process. Routine parameters and highly sensitive CRP (hs-crp) were determined on an Olympus AU2700 apparatus (Rungis, France). Total serum cholesterol and triglyceride levels were measured by enzymatic methods (KonePro, Konelab). ctni was measured using an AccuTnI assay based on a chemiluminescence method on an Acces II Immunoassay system (Beckman-Coulter) [13]. Linearity was investigated by the serial dilution of different sera containing high amounts of ctni. As recommended by the joint European Society of Cardiology and the American College of Cardiology [3], an increased value for ctni was defined as values exceeding 0.04 mg/l, corresponding to the 99th percentile. In addition, functional sensitivity was defined as the lowest concentration with an acceptable imprecision [coefficient of variation (CV) <10%]. The 10% CV was determined through 20 replicated analyses of samples ranging from 0.01 to 0.04 mg/l. Statistical analysis A. Boulier et al. Data are expressed as mean±se for normally distributed variables and as median and range for non-normally distributed variables. Significance was set at P<0.05. The SAS software was used for statistical analyses. Differences in cumulative survival rates were compared first between patients with circulating ctni (10% functional

3 ctni enhances the prognostic value of hs-crp 3 of 6 sensitivity) and undetectable ctni levels, and secondly between patients with elevated CRP (10 mg/l) and normal CRP (<10 mg/l). Exposure was computed from the date of first blood draw until the date of death, with censoring for renal transplantation. Adjusted relative risks (RRs) of death and 95% confidence intervals (CIs) were estimated using Cox proportional hazard models in a two-step analysis: first, we estimated age-, sex- and duration of dialysis-adjusted RR using multivariate Cox models. Next, potential confounders (diabetes, hypertension, smoking, total cholesterol, triglycerides, dialysis centre and HD modality) were added to the model to obtain fully adjusted RRs. Results Analytical operations The functional sensitivity was tested in serum ranging from 0.01 mg/l (the detection limit) to 0.04 mg/l (the 99th percentile). At a total imprecision of 10%, the functional sensitivity was 0.03 mg/l. The coefficient of linearity obtained was r 2 ¼ Troponin and CRP distribution at baseline A significant number of patients had elevated ctni regardless of the cut-off criteria chosen: 13.6% 99th percentile (0.04 mg/l) and 25.1% 10% CV (0.03 mg/l) (Table 1). In comparison, in the reference population of the laboratory, those threshold values are exceeded only by 2 and 3%, respectively. No significant association could be detected between ctni levels and gender, coronary pathology, diabetes, smoking, time spent in dialysis, hypertension, dyslipidaemia or CRP Table 2. Adjusted and fully adjusted RRs for death by ctni and CRP concentrations at baseline levels. In contrast, ctni significantly increased with age (P ¼ ). The median of CRP level in our cohort was 7.3 mg/l ( ), compared with 1.8 mg/l in the reference population of the laboratory. Inflammation, based on a CRP >5 mg/l, was present in >60% of our subjects. In addition, 40.3% had CRP levels >10 mg/l. Follow-up The patients were followed-up for a median of 418 days (15 483). There were a total of 29 deaths, including 13 from cardiovascular causes. In addition, six major cardiovascular events (three myocardial infarctions, two vascular interventions for peripheral arteriopathy and one pulmonary oedema) occurred. Patient exposure was censored for renal transplantation (n ¼ 14). Of the 191 patients, eight were transferred to other renal dialysis units (seven from one to another dialysis facility in Montpellier and one to outside the Montpellier area). However, clinical data were available for all these patients. Troponin and CRP as prognostic indexes Patients with positive circulating ctni, defined by serum levels above the functional sensitivity at 10% CV (0.03 mg/l), had significantly worse outcomes than patients with negative ctni [RR adjusted for age, sex and time spent in dialysis 4.2 ( )]. After further adjustment for traditional risk factors (diabetes, smoking, hypertension, total cholesterol, triglycerides, dialysis centre and dialysis modality), all-cause mortality remained significantly associated with ctni levels 0.03 mg/l (Table 2). No significant interaction was Adjusted RR a 95% CI P-value Fully adjusted RR b 95% CI P-value All cause death (n ¼ 29) ctni (mg/l) < CRP (mg/l) < Composite criterion CRP <10 and ctni < CRP 10 or ctni CRP 10 and ctni < Cardiovascular death (n ¼ 13) ctni (mg/l) < CRP (mg/l) < NS NS Composite criterion CRP <10 and ctni < CRP 10 or ctni NS NS CRP 10 and ctni a Adjusted for age, sex and duration of dialysis. b Fully adjusted for diabetes, smoking, hypertension, dialysis centre, HD modality, total cholesterol and triglycerides.

4 4of6 evidenced between the level of ctni and time to death. In addition, of the six patients who had non-fatal cardiovascular events, only one had a ctni level >0.03 mg/l. The mortality observed in the high troponin group was not associated with previously identified cardiac ischaemic disease. Indeed, the relationship between elevated troponin and all-cause mortality was greater in patients without any history of coronary disease [(fully adjusted RR ¼ 9.3 ( )] than in those with known ischaemic disease [fully adjusted RR ¼ 1.3 ( )]. Inflammation was also predictive of all-cause mortality, as demonstrated by the significant RR associated with CRP levels 10 mg/l after full adjustment [RR ¼ 2.7 ( )] (Table 2). However, the threshold value of 5 mg/l could not be tested due to the insufficient number of deaths among patients with CRP levels between 5 and 10 mg/l. CRP >10 mg/l also tended to increase cardiovascular mortality [fully adjusted RR ¼ 1.3 ( )]. Combination of inflammation and circulating ctni reliably identified high-risk patients. Indeed, patients with high CRP levels (10 mg/l) or ctni-positive results (0.03 mg/l) had an enhanced mortality rate [adjusted RR ¼ 7.1 ( ); fully adjusted RR ¼ 6.5 ( )] compared with patients with both low CRP and ctni. The combination of CRP 10 mg/l and ctni 0.03 mg/l was associated with a dramatic increase in all-cause mortality [RR ¼ 16.9 ( ); fully adjusted RR ¼ 12.0 ( )]. Survival (%) Discussion Our study shows that the increase of circulating ctni in stable asymptomatic HD patients is associated with poor survival. Moreover, the combination of inflammation and elevated circulating ctni was associated with a dramatic increase in 1 year mortality (Figure 1). We measured ctni using the AccuTnI assay (Beckman-Coulter), which has been validated recently in a large population of non-uraemic patients [13]. We confirm that ctni, assayed by that method, is elevated in HD patients in the absence of acute coronary syndrome [8], as demonstrated by the high prevalence of patients with levels above the 99th percentile value determined by the manufacturer (0.04 mg/l). In addition, functional sensitivity, defined as an acceptable CV at 10%, should be used as a reference limit for elevated ctni [3,5]. The value obtained by us (0.03 mg/l) was close to the 99th percentile, and was in agreement with previous data [13]. Using this threshold value, 25.1% of our subjects had levels above the functional sensitivity, compared with only 1% identified with another less sensitive test [5]. The analytical capability and the high sensitivity of this commercially available kit allowed us to use the recommended functional sensitivity threshold value for our prospective study. Elevated ctni was associated with a significantly increased risk of all-cause and cardiovascular mortality. The relative risk of death remained significantly increased after full adjustment for recognized CRP<10 and ctni < 0.03 CRP 10 or ctni 0.03 CRP 10 and ctni 0.03 A. Boulier et al Number at risk : Survival time (in days) CRP<10 and ctni<0.03 : CRP 10 or ctni 0.03 : CRP 10 and ctni 0.03 : Fig. 1. Survival curves about CRP and CTnl Beckman.

5 ctni enhances the prognostic value of hs-crp 5 of 6 risk factors such as age, sex, diabetes, time spent in dialysis, hypertension, smoking, total cholesterol, triglycerides, dialysis centre and dialysis modality. Previous studies with special emphasis on ctnt have demonstrated that high circulating troponin concentrations are associated with poor survival [5,6]. Data concerning ctni are thought to be less consistent and are more difficult to interpret, since no standardization between assays using different antibodies exists [14]. Interestingly, ctni measured with a highly specific and sensitive test [13] was also predictive for all-cause and cardiovascular death in patients without a history of CAD documented in their medical charts, whether clinically determined or angiographically proven. Since ctni is highly specific to myocardial injury, ctni could originate from a silent ischaemic episode, which is observed frequently in ESRF and is associated with poor prognosis [15]. Alternatively, the release of ctni could be linked to non-ischaemic cardiac damage such as uraemic cardiomyopathy [9]. Indeed, interactions of metabolic abnormalities and toxin retention related to uraemia result in cardiomyopathy and left ventricular hypertrophy. Such cardiomyopathy has been recognized as a pejorative prognostic factor for arrhythmia, cardiovascular events and sudden death [9]. It was shown recently that left ventricular mass was a strong predictor of elevated ctnt levels in HD patients [16]. CRP has been recognized as an important predictor of morbidity and mortality in uraemic patients and is strongly associated with cardiovascular disease [11,12,17] and uraemic cardiopathy [18]. The development of highly sensitive CRP assay kits has shown that the increased mortality could be associated with 5 mg/l (or even lower CRP) levels [11,12]. However, the classical threshold of 10 mg/l, first proposed by Stevinkel et al. [10], has been confirmed recently by Chauveau et al. [19]. The statistically significant adjusted RR associated with CRP >10 mg/l, measured with a high sensitivity test, confirms previous reports of the epidemiological association between CRP and all-cause or cardiovascular deaths in HD populations. Moreover, this study provides evidence that the combination of high CRP and circulating ctni levels dramatically impairs the HD survival rate [fully adjusted RR 12.0 ( )]. This original finding suggests that CRP and elevated ctni are independent markers of uraemic cardiopathy [16,18] and that they could act synergistically to worsen cardiovascular prognosis. This hypothesis is supported further by the positive linear relationship recently observed in HD patients between ctnt and CRP or interleukin-6, which is a more upstream marker of the inflammatory cascade [20]. In conclusion, elevated circulating ctni was found to be associated with poor prognosis in HD patients, especially when occurring in concert with high CRP. These data present a plea for the regular measurement of both CRP and ctni in HD patients, to best stratify cardiovascular risk or to initiate more aggressive exploration and treatment. Prospective interventional studies are required in order to evaluate the benefit of regularly monitoring these markers and their potential role in therapeutic intervention. Acknowledgements. The authors would like to acknowledge with gratitude the nephrologists in various dialysis centres for their collaboration: A. Flavier, J. Y. Bosc, H. Leray-Moragues, C. Rouanet, I. Selcer and M. Delage, and M.H. Vernet for laboratory assistance. The ctni reagents used in this study were generously provided by Beckman-Coulter France. The reagents for hs-crp analysis were a generous gift from Olympus France. Conflict of interest stastement. None declared. References 1. Wu AHB, Apple FS, Gilbler WB, Jesse RL, Warshaw MM, Valdes R. National Academy of Clinical Biochemestry Standards of Laboratory Practice: recommendations for the use of cardiac markers in coronary artery diseases. Clin Chem 1999; 45: Apple FS, Wu AHB, Jaffe AS. European Society of Cardiology and American College of Cardiology guidelines for redefinition of myocardial infarction: how to use existing assays clinically and for clinical trials. Am Heart J. 2002; 144: Joint European Society of Cardiology/American College of Cardiology Committee. Myocardial infarction redefined a consensus document of the joint European Society of Cardiology/American College of Cardiology Committee for the redefinition of myocardial infarction. J Am Cardiol 2000; 36: Khan IA, Wattanasuwan N, Mehta NJ et al. Prognostic value of serum cardiac troponin I in ambulatory patients with chronic renal failure undergoing long-term hemodialysis: a twoyear outcome analysis. J Am Coll Cardiol 2001; 38: Apple FS, Murakami MM, Pearce LA, Herzog CA. Predictive value of cardiac troponin I and T for subsequent death in endstage renal disease. Circulation 2002; 106: Dierkes J, Domrose U, Westphal S, Ambrosch A, Bosselmann HP, Neumann KH, Luley C. Cardiac troponin T predicts mortality in patients with end-stage renal disease. Circulation 2000; 102: Stoffel MP, Pollok M, Baldamus CA. Troponin I is a better prognostic parameter of cardiovascular events in asymptomatic patients on haemodialysis than troponin T. Nephrol Dial Transplant 2000; 15: Chen P, Tanasijevic M. Use of serum cardiac troponin-i levels in dialysis patients. Semin Dial 2000; 13: London GM. Cardiovascular disease in chronic renal failure: pathophysiologic aspects. Semin Dial 2003; 16: Stenvinkel P, Heimbu rger O, Paultre F et al. Strong association between malnutrition, inflammation, and atherosclerosis in chronic renal failure. Kidney Int 1999; 55: Owen WF, Lowrie EG. C-reactive protein as an outcome predictor for maintenance hemodialysis patients. Kidney Int 1998; 54: Wanner C, Zimmermann J, Schwelder S, Metzger T. Inflammation and cardiovascular risk in dialysis patients. Kidney Int 2002; 61 [Suppl 80]: S99 S Uettwiller-Geiger D, Wu AH, Apple FS et al. Multicenter evaluation of an automated assay for troponin I. Clin Chem 2002; 48: Sheehan P, Blennerhassett J, Vasikaran SD. Decision limit for troponin I and assay performance. Ann Clin Biochem 2002; 39: Nakamura S, Uzu T, Inenaga T, Kimura G. Prediction of coronary artery disease and cardiac events using electrocardiographic changes during hemodialysis. Am J Kidney Dis 2000; 36:

6 6of6 16. Mallamaci F, Zoccali C, Parlongo S et al. Troponin is related to left ventricular mass and predicts all-cause and cardiovascular mortality in hemodialysis patients. Am J Kidney Dis 2002; 40: Brancaccio D, Tetta C, Gallieni M, Panichi V. Inflammation, CRP, calcium overload and a high calcium phosphate product: a liaison dangereuse. Nephrol Dial Transplant 2002; 17: Park CW, Shin YS, Kim CM et al. Increased C-reactive protein following hemodialysis predicts cardiac hypertrophy in A. Boulier et al. chronic hemodialysis patients. Am J Kidney Dis 2002; 40: Chauveau P, Level C, Lasseur C et al. C-reactive protein and procalcitonin as markers of mortality in hemodialysis patients: a 2-year prospective study. J Ren Nutr 2003; 13: Lowbeer C, Stenvinkel P, Pecoits-Filho R et al. Elevated cardiac troponin T in predialysis patients is associated with inflammation and predicts mortality. J Intern Med 2003; 253: Received for publication: Accepted in revised form:

Creatinine index and transthyretin as additive predictors of mortality in haemodialysis patients

Nephrol Dial Transplant (2008) 23: 345 353 doi:10.1093/ndt/gfm573 Advance Access publication 22 September 2007 Original Article Creatinine index and transthyretin as additive predictors of mortality in