- Email: [email protected]
Predictors for cardiovascular morbidity and overall mortality in Tunisian ESRD patients: A six year prospective study
Available online at www.sciencedirect.com
Clinical Biochemistry 42 (2009) 648 – 653
Predictors for cardiovascular morbidity and overall mortality in Tunisian ESRD patients: A six year prospective study☆ Hayet Fellah a , Mohamed Bassem Hammami a , Moncef Feki a,⁎, Karima Boubaker b , Taieb Ben Abdallah b , Bernard Lacour c , Abderraouf Mebazaa a , Neziha Kaabachi a a
Lab-SM-01 Research Laboratory and Service of Biochemistry, Rabta Hospital, Tunis, Tunisia b Service of Internal Medicine and Nephrology, Charles Nicole Hospital, Tunis, Tunisia c Laboratory of Biochemistry A, Necker Hospital, Paris, France
Received 22 October 2008; received in revised form 16 December 2008; accepted 17 December 2008 Available online 8 January 2009
Abstract Objectives: The study was aimed to test the predictive value of several potential cardiovascular factors and markers for non fatal cardiovascular events (CVE) and overall mortality in Tunisian patients with renal failure. Subjects and methods: One hundred and fifteen renal failure patients were followed-up from 2000 to 2006. At enrollment, each patient underwent clinical examination and blood collection for analysis of lipid parameters, albumin, C reactive protein (CRP), parathyroid hormone (PTH), homocysteine and hemoglobin. Multivariate Cox regression models were applied to identify the predictors for non fatal CVE and overall mortality. Results: During the follow up, seventeen patients were lost. Among the 98 remaining patients, 29 presented a non fatal CVE (21.5%) and 15 were deceased (11.1%). In univariate analyses, non fatal CVE were more frequent in smokers and in patients with high PTH concentrations and low HDL levels. Moreover, low albumin concentrations were univariately associated with overall mortality. In the multivariate analysis, non fatal CVE was significantly and independently associated with age [hazard ratio (95% confidence interval), 1.04 (1.01–1.08); p = 0.028] and the upper quartile of PTH concentrations [2.68 (1.24–5.81); p = 0.013]. Overall mortality was independently predicted by the bottom quartile of albumin concentrations [5.62 (2.02–15.6); p = 0.001] and the upper quartile of CRP concentrations [3.20 (1.14–8.79); p = 0.027]. Conclusion: Advanced age and high PTH levels are the main predictors of CVE, whereas low albumin and high CRP concentrations are the independent predictors of death in Tunisian renal patients. A better control of these factors would greatly increase the patient's survival rates. © 2009 The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved. Keywords: Cardiovascular disease; Chronic renal failure; Hemodialysis; Mortality; Renal transplantation; Risk factor
Introduction The incidence and prevalence of end stage renal disease (ESRD) continues to increase worldwide. Mortality in patients Abbreviations: CRF, chronic renal failure; CRP, C reactive protein; CVE, cardiovascular events ; ESRD, end stage renal disease; HD, hemodialysis; HDLC, HDL cholesterol ; HHC, hyperhomocysteinemia; PTH, parathyroid hormone; RT, renal transplantation. ☆ This work was supported by the Ministry of Higher Education, Scientific Research and Technology of Tunisia. ⁎ Corresponding author. Laboratory of Biochemistry, Rabta Hospital, 1007 Jebbari, Tunis, Tunisia. Fax: +216 71 570 506. E-mail address: [email protected] (M. Feki).
with ESRD is dramatically high and cardiovascular disease is the leading cause of mortality among them [1,2]. Atherosclerotic heart disease is believed to account for approximately 55% of mortality and contributes to a 20-fold increase in ischemic heart disease and to a 10-fold increase in risk of stroke among these patients [3–5]. Such findings are thought to be related to the high prevalence of several cardiovascular risk factors such as hypertension, smoking, diabetes, dyslipemia, inflammation, malnutrition, anemia, hyperhomocysteinemia [HHC], hypercoagulability and oxidative stress [5–12]. However, the exact reasons for the high mortality have not been completely elucidated and the relative contribution of each factor to poor clinical outcomes was not well understood. The
0009-9120/$ - see front matter © 2009 The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved. doi:10.1016/j.clinbiochem.2008.12.023
H. Fellah et al. / Clinical Biochemistry 42 (2009) 648–653
identification of patients at high risk and the precise evaluation of the causes of death are crucial to devise effective preventive strategies. We had already found a high prevalence of cardiovascular risk factors or markers in Tunisian patients with ESRD [13,14]. This study was aimed to test the predictive value of several clinical and biochemical factors for non fatal cardiovascular events (CVE) and overall mortality after a 6 year follow-up in Tunisian ESRD patients. Subjects and methods Patients One hundred fifteen consecutive ESRD patients (58 male and 57 female) aged 16 to 68 years (40.8 ± 10.8), were enrolled in the study during the year 2000, and followed up during 6 years. They consisted of 35 chronic renal failure (CRF) patients on conservative treatment, 50 hemodialyzed (HD) patients and 30 renal transplant (RT) recipients undergoing therapy at the Nephrology Department of Charles Nicole Hospital (Tunis). All patients are Arab descendants who resided in the north of Tunisia. Patients with diabetes, thyroid diseases, systemic and active inflammatory diseases or those under lipid lowering drugs were excluded. Hemodialysis was performed for 4 h three times a week, using bicarbonate buffer and polysulfone dialysis membranes. The dialysis dose (Kt/V), defined as the dialyzer clearance of urea (K, in mL/min) multiplied by the duration of the dialysis treatment (t, in minutes) divided by the volume of distribution of urea in the body (V, in mL), averaged 1.12 ± 0.26 with a range between 0.50 and 1.72. CRF and HD patients were supplemented with iron (50–100 mg), calcium carbonate (500 mg), and B group vitamins (thiamine, 400 mg; pyridoxine, 80 mg) daily. RT recipients were studied at least 2 years after the transplantation and were taking cyclosporine (3–4 mg/kg), prednisone (10– 15 mg), azathioprine (75–125 mg) daily as immunosuppressive treatment. No patient received erythropoietin treatment. At enrollment, collected demographic and clinical data included age, gender, smoking, etiology of ESRD, hypertension and drug consumption. The causes of renal impairment were chronic glomerulonephritis (n = 39), hypertensive nephropathy (n = 26), pyelonephritis and chronic interstitial nephritis (n = 17) and polycystic kidney disease (n = 6). In the remaining cases, the etiology was unspecified or unknown. The protocol of the survey was approved by the Ethics Committee on Human Research of the Charles Nicole Hospital (Tunis, Tunisia) and all participants gave their free informed consent. Cardiovascular events CVE included coronary, cerebral and abdominal aortic or peripheral vascular disease. Coronary heart disease was diagnosed by at least one of the following criteria: 1) myocardial infarction documented by a typical rise and fall of troponine I with ischemic symptoms and/or electrocardiogram changes indicative of myocardial necrosis or ischemia (Q waves, ST segment elevation or depression), 2) symptoms
649
consistent with angina confirmed by a positive exercise stress test result and/or abnormal coronarography, 3) acute heart failure documented by clinical and echographic findings, 4) arrhythmia, and 5) left ventricular hypertrophy. Cerebrovascular disease was defined as a stroke (acute onset irreversible neurological deficit such as aphasia, focal deficit or unilateral paresis confirmed by computed tomography) or symptomatic extra cranial carotid artery stenosis confirmed by ultrasonography. Peripheral vascular disease was diagnosed by the development of intermittent claudication accompanied by diminished pulses and confirmed by Doppler or lower extremity amputation or deep venous thrombosis confirmed by venous Doppler. Biochemical analysis Fasting blood samples were collected into EDTA-containing tubes for total homocysteine and hemoglobin and into freeanticoagulant tubes for other parameters. For HD patients, blood was collected before dialysis session. Plasma homocysteine was assessed by a fluorescent polarizing immunoassay method on an Axsym system (Abbott Diagnostics). Serum intact PTH was measured by radioimmunoassay (Cis BioInternational). Hemoglobin was measured on a globule counter (Coulter Beckman). Serum albumin and C-reactive protein (CRP) were measured by an immunoturbidimetric method, and total cholesterol and triglycerides were assessed by standard methods on a Hitachi 912 analyzer (Roche Diagnostics). HDL cholesterol (HDL-C) was determined by direct method. Cardiovascular risk factors or markers In this study, we considered the following potential cardiovascular risk factors or markers; hypertension (diastolic blood pressure N 90 mm Hg and/or systolic blood pressure N140 mm Hg, or the use of antihypertensive therapy) and smoking (patients who currently smoke at least 5 cigarettes per day). For biological markers, we considered as “at risk values” those of the upper quartile (compared to the three lower quartiles) for homocysteine (N32.6 μmol/L), CRP (N 12.8 mg/L) and PTH
Table 1 Main characteristics of study participants
Men (%) Smoking (%) Hypertension (%) Age (years) CCr (mL/min) BMI (kg/m2)
CRF patients (n = 35)
HD patients (n = 50)
RT patients (n = 30)
Controls (n = 31)
49 37 60 *** 42.3 ± 12.2 21.7 ± 12.6 *** 23.6 ± 4.6 **
50 30 44 *** 41.4 ± 10.4 8.0 ± 2.0 *** 21.8 ± 4.3 ***
53 30 43 *** 37.9 ± 9.4 60.1 ± 24.3 *** 22.5 ± 3.3 ***
48 26 – 38.5 ± 10.1 97.6 ± 19.2 26.8 ± 2.9
Results are expressed as means ± SD or percent; CRF: chronic renal failure; HD: hemodialysis; RT: renal transplant; CCr: clearance of creatinine; BMI: body mass index. ⁎⁎ p b 0.01. ⁎⁎⁎ p b 0.001 (towards controls).
650
H. Fellah et al. / Clinical Biochemistry 42 (2009) 648–653
Table 2 Unadjusted relative risk with 95% confidence interval of several potential cardiovascular risk factor for non-fatal cardiovascular events and overall mortality in end stage renal disease patients (n = 98) n
Non fatal cardiovascular events (n = 29)
72 26 66 32 71 27 74 24 71 27 69 29 81 17 77 21 72 26
27.8 34.6 21.2 46.9 28.2 33.3 28.4 33.3 31.0 25.9 23.2 50.0 24.7 52.9 28.6 33.3 29.2 30.8
% Hypertension Smoking Dyslipemia Upper quartile of homocysteine Upper quartile of CRP Upper quartile of parathormone Lower quartile of HDL cholesterol Lower quartile of albumin Lower quartile of hemoglobin
No Yes No Yes No Yes No Yes No Yes No Yes No Yes No Yes No Yes
RR (95% CI)
p
%
0.513
15.3 19.2 15.2 18.8 15.5 18.5 14.9 20.8 12.7 25.9 17.4 13.6 18.5 5.9 10.4 38.1 13.9 23.1
1.38 (0.53–3.59) 0.009 2.56 (1.31–4.34) 0.617 1.28 (0.49–3.31) 0.644 1.26 (0.47–3.39) 0.624 0.78 (0.29–2.11) 0.017 3.31 (1.21–9.05) 0.020 3.44 (1.16–10.00) 0.772 1.25 (0.44–3.44) 0.878 1.08 (0.41–2.86)
Overall mortality (n = 16) RR (95% CI)
p 0.425
1.32 (0.41–4.24) 0.651 1.29 (0.42–4.00) 0.465 1.24 (0.39–3.97) 0.344 1.51 (0.47–4.88) 0.113 2.41 (0.80–7.31) 0.483 0.75 (0.20–2.94) 0.182 0.27 (0.03–2.22) 0.005 5.31 (1.69–16.67) 0.215 1.86 (0.60–5.760)
CRP, C-reactive protein; RR, relative risk; CI, confidence interval.
(N337 μg/L), and those of the bottom quartile (compared to the three upper quartiles) for albumin (b35 g/L), HDL-C (b25 mg/ dL) and hemoglobin (b 84 g/L). Dyslipemia was considered in subjects with total plasma cholesterol N 250 mg/dL and/or plasma triglycerides N200 mg/dL. Statistical analysis Statistical analyses were carried out using SPSS for windows 10.0 software (SPSS Inc., Chicago, USA). Comparison between groups was performed using chi-square test for categorical variables. For continuous variables, comparison was performed using Student t-test for normally distributed variables and Wilcoxon rank sum test for non-normally distributed ones. The Kaplan-Meier method was used to compute observed survival.
Relative risks for CVE or all-cause mortality were determined by means of univariate analysis and multivariate Cox proportional hazards regression (backward stepwise model), including several potential cardiovascular risk factors (i.e. age, hypertension, smoking, dyslipemia, clearance of creatinine, upper quartile for plasma homocysteine, CRP and PTH concentrations, and bottom quartile for albumin, HDL-C and hemoglobin levels). Significance was established at a p value below 0.05, based on two-tailed calculations. Results Demographic and clinical patient characteristics at enrollment are shown in Table 1. During the follow-up, seventeen patients (12.6%) were lost. Among the 98 remaining patients,
Fig. 1. Kaplan-Meier estimates of observed survival during follow-up with regards to non fatal cardiovascular events in patients stratified by enrollment HDL cholesterol (HDL-C) and parathyroid hormone (PTH) concentrations.
H. Fellah et al. / Clinical Biochemistry 42 (2009) 648–653
651
Fig. 2. Kaplan-Meier estimates of observed survival during follow-up with regards to all-cause mortality in patients stratified by enrollment plasma albumin and Creactive protein (CRP) concentrations.
29 presented a non fatal CVE (21.5%) and 15 were deceased (11.1%). CVE consisted of heart failure (n = 10), angina (n = 6), stroke (n = 3), myocardial infarction (n = 3), arrhythmia (n = 3); left ventricular hypertrophy (n = 2), lower extremities vascular disease (n = 1) and femoral deep venous thrombosis (n = 1). Death was attributed to a cardiovascular event (n = 7), infectious processes (n = 4), cancer (n = 3), and severe malnutrition (n = 1). Non fatal CVE were more frequently observed in smokers and in patients with high PTH concentrations and low HDL-C concentrations. Overall mortality was significantly higher in patients with low albumin concentrations. We also observed a tendency towards increased mortality for high CRP levels (Table 2). Survival curves for non fatal CVE according to the upper quartile of PTH and the lower quartile of HDL cholesterol values and survival curves for overall mortality according to the upper quartile of CRP and the lower quartile of albumin values were shown in Figs. 1 and 2. When applying multivariate Cox regression analysis, non fatal CVE remained significantly associated with age [hazard ratio (HR), 95% confidence interval (95% CI), 1.04 (1.01– 1.08); p = 0.028] and the upper quartile of PTH concentrations [HR (95% CI), 2.68 (1.24–5.81); p = 0.013]. Overall mortality Table 3 Stepwise backward Cox regression models for non fatal cardiovascular events and for overall mortality in Tunisian ESRD patients (n = 98) B coefficient SE Non fatal cardiovascular events Age 0.042 Upper quartile of PTH 0.985 Overall mortality Bottom quartile of albumin 1.727 Upper quartile of CRP 1.152
HR (95% CI)
p
0.019 1.04 (1.01–1.08) 0.028 0.395 2.68 (1.24–5.81) 0.013 0.521 5.62 (2.02–15.6) 0.001 0.523 3.20 (1.14–8.79) 0.027
SE, standard error; HR, hazard ratio; CI, confidence interval; PTH, parathyroid hormone; CRP, C-reactive protein. Entered covariates are age, hypertension, smoking, dyslipemia, clearance of creatinine, upper quartile for plasma homocysteine, C-reactive protein and parathyroid hormone, and bottom quartile for albumin, HDL cholesterol and hemoglobin.
was associated with the bottom quartile of albumin concentrations [HR (95% CI), 5.62 (2.02–15.6); p = 0.001] and the upper quartile of CRP concentrations [HR (95% CI), 3.20 (1.14– 8.79); p = 0.027] (Table 3). Discussion This study showed that low albumin and high CRP levels are the main predictors for death and that advanced age and high plasma PTH concentrations are the independent predictors for non fatal CVE in Tunisian ESRD patients. These data provide a comprehensive evaluation of the predictors for CVE and death in Tunisian patients with ESRD, offering the opportunity to evaluate the prognostic strength of these factors when analyzed together. The finding that hypoalbuminemia is the main prognostic factor for death was largely reported [15–17]. It was suggested that chronic inflammation may be the missing link or factor that causally ties hypoalbuminemia to morbidity and mortality [18,19]. However, data consistent with the possibility that malnutrition may affect survival independently exist [20]. The role of high CRP concentration as an independent predictor for mortality found in this study was previously demonstrated in ESRD patients [16,21–23]. High PTH level was found to predict CVE in our patients. Excess PTH has adverse effects on cardiac function and structure, including cardiac interstitial fibrosis and left ventricular hypertrophy and dysfunction [24,25]. While an association between low PTH and increased mortality was reported in renal patients, these low PTH levels are secondary to a low protein and phosphorus intake resulting in malnutrition that may explain the excess mortality [26–28]. Our findings have important implications for clinical practice. In fact, ESRD patients with low albumin and/or high CRP levels should receive close follow-up and all sources of malnutrition and inflammation should be controlled. The control of excess PTH also remains an important goal to minimize cardiovascular morbidity and mortality in renal patients.
652
H. Fellah et al. / Clinical Biochemistry 42 (2009) 648–653
This study failed to detect significant associations between CVE and/or mortality and potential predictors such as smoking, hypertension, dyslipemia, HHC and anemia. In reality, the great diversity of perturbations and the complexity of interactions between multiple perturbations in renal patients may modify the relationship between each factor and outcomes. It is possible that the dominant influence of some factors prevented the detection of factors that are of less prognostic importance. The literature showed very confusing data on the role of various factors in poor outcomes in renal patients [7–13,29–35]. These discrepancies may be related to differences in population characteristics, sampling size, observation time, method of analysis and the impact of confounding covariates on the outcome of multivariate analysis. In this study, the association of smoking and low HDL with CVE lost its significance in multivariate analysis. The main limitations of the study are the small number and the heterogeneity of patients. The choice to focus on overall mortality and not on cardiovascular mortality is chiefly driven by the low number of cardiac deaths. Our sample cannot be considered as representative of Tunisian ESRD patients as patients with diabetes mellitus, a common cause of ESRD, were not enrolled. This is one reason of the low sample size and may explain the low rate of cardiovascular death in our series. In return, the proper effect of diabetes and its interaction with predictors on outcomes are excluded. Our data arose from multivariate regression models that simultaneously controlled for several important predictors of the outcomes, including clearance of creatinine, making the conclusions valid. However, it must be recognized that association does not imply causation. This study showed that low albumin and high CRP and PTH levels are the main predictors for poor outcomes in Tunisian renal patients. More vigorous measures to control these factors may contribute to reduce CVE and to improve survival in these patients. Other factors such as hypertension, dyslipemia, anemia, and HHC were not found to predict poor outcomes, suggesting a probable lower prognostic importance. It is also possible that complexity of interactions between factors and methodological issues have modified the relationship between predictors and outcomes. To define the relative contribution of each predictor to clinical outcome, large scale randomized prospective interventional studies are required and until the results of such studies are available, increased vigilance in cardiovascular risk factors management is warranted and may be useful to reduce the burden of ESRD morbidity and mortality. Acknowledgments The authors gratefully acknowledge Prof. Mohamed Salah Mekni for helpful advice in the preparation of this manuscript. References [1] Foley RN, Parfrey PS, Sarnak MJ. Clinical epidemiology of cardiovascular disease in chronic renal disease. Am J kidney Dis 1998;32:112–9.
[2] Gupta R, Birnbaum Y, Uretsky BF. The renal patient with coronary artery disease: current concepts and dilemmas. J Am Coll Cardiol 2004;44: 1343–53. [3] USRDS: the United States Renal data System. USRDS 2003 annual data report. Am J Kidney Dis 2003;42:1–230. [4] Cheung AK, Sarnak MJ, Yan G, et al, HEMO Study Group. Cardiac diseases in maintenance haemodialysis patients: results of the HEMO Study. Kidney Int 2004;65:2380–9. [5] Stenvinkel P, Heimburger O, Paultre F, et al. Strong association between malnutrition, inflammation, and atherosclerosis in chronic renal failure. Kidney Int 1999;55:1899–911. [6] Eikelboom JW, Hankey GJ. Associations of homocysteine, C-reactive protein and cardiovascular disease in patients with renal disease. Curr Opin Nephrol Hypertens 2001;10:377–83. [7] Mallamaci F, Zoccali C, Tripepi G, et al. Hyperhomocysteinemia predicts cardiovascular outcomes in haemodialysis patients. Kidney Int 2002;61: 609–14. [8] Longenecker JC, Coresh J, Powe NR, et al. Traditional cardiovascular disease risk factors in dialysis patients compared with the general population: the CHOICE Study. J Am Soc Nephrol 2002;13:1918–27. [9] Avram MM, Blaustein D, Fein PA, Goel N, Chattopadhyay J, Mittman N. Hemoglobin predicts long-term survival in dialysis patients: a 15-year single-center longitudinal study and a correlation trend between prealbumin and hemoglobin. Kidney Int 2003;87:S6–S11. [10] Zoccali C, Mallamaci F, Tripepi G. Novel cardiovascular risk factors in end stage renal disease. J Am Soc Nephrol 2004;15:S77–80. [11] Suliman M, Stenvinkel P, Qureshi AR, et al. The reverse epidemiology of plasma total homocysteine as a mortality risk factor is related to the impact of wasting and inflammation. Nephrol Dial Transplant 2007;22:209–17. [12] Portolés J, López-Gómez JM, Aljama P. A prospective multicentre study of the role of anaemia as a risk factor in haemodialysis patients: the MAR Study. Nephrol Dial Transplant 2007;22:500–7. [13] Fellah H, Feki M, Hsairi M, et al. Hyperhomocysteinemia and ESRD: determinants and association with cardiovascular disease in Tunisian patients. Clin Chem Lab Med 2003;41:675–80. [14] Fellah H, Feki M, Feki H, et al. Classical and non classical cardiovascular risk factors in Tunisian patients with end-stage renal disease: prevalence and association with cardio vascular events. Clin Lab 2004;5:447–53. [15] Combe C, Chauveau P, Laville M, et al. French Study Group for Nutrition in Dialysis. Influence of nutritional factors and haemodialysis adequacy on the survival of 1,610 French patients. Am J Kidney Dis 2001;37:S81–8. [16] Menon V, Greene T, Wang X, et al. C-reactive protein and albumin as predictors of all-cause and cardiovascular mortality in chronic kidney disease. Kidney Int 2005;68:766–72. [17] Honda H, Qureshi AR, Heimbürger O, et al. Serum albumin, C-reactive protein, interleukin 6, and fetuin a as predictors of malnutrition, cardiovascular disease, and mortality in patients with ESRD. Am J Kidney Dis 2006;47:139–48. [18] Kaysen GA. Malnutrition and the acute-phase reaction in dialysis patients—how to measure and how to distinguish. Nephrol Dial Transplant 2000;15:1521–4. [19] Yeun JY, Levine RA, Mantadilok V, Kaysen GA. C-reactive protein predicts all-cause and cardiovascular mortality in haemodialysis patients. Am J Kidney Dis 2000;35:469–76. [20] Leon JB, Majerle AD, Soinski JA, Kushner I, Ohri-Vachaspati P, Sehgal AR. Can a nutrition intervention improve albumin levels among haemodialysis patients? A pilot study. J Renal Nutr 2001;11:9–15. [21] Owen WF, Lowrie EG. C-reactive protein as an outcome predictor for maintenance haemodialysis patients. Kidney Int 1998;54:627–36. [22] Wang AY, Woo J, Lam CW, et al. Is a single time point C-reactive protein predictive of outcome in peritoneal dialysis patients? J Am Soc Nephrol 2003;14:1871–9. [23] Pearson TA, Mensah GA, Alexander RW, et al. Markers of inflammation and cardiovascular disease: application to clinical and public health practice: a statement for health-care professionals from the Centers for Disease Control and Prevention and the American Heart Association. Circulation 2003;107:499–511.
H. Fellah et al. / Clinical Biochemistry 42 (2009) 648–653 [24] Stefenelli T, Abela C, Frank H, et al. Cardiac abnormalities in patients with primary hyperparathyroidism: implications for followup. J Clin Endocrinol Metab 1997;82:106–12. [25] Strozecki P, Adamowicz A, Nartowicz E, Odrowaz-Sypniewska G, Wlodarczyk Z, Manitius J. Parathormone, calcium, phosphorus, and left ventricular structure and function in normotensive haemodialysis patients. Ren Fail 2001;23:115–26. [26] Akizawa T, Kinugasa E, Kurihara S, et al. Parathyroid hormone deficiency is an indicator of poor nutritional state and prognosis in dialysis patients. J Am Soc Nephrol 1999;10:615 A. [27] Chertow GM, Lazarus JM, Lew NL, Lowrie EG. Intact parathyroid hormone is directly related to mortality in haemodialysis. J Am Soc Nephrol 2000;11:573A (abstract). [28] Avram MM, Mittman N, Myint MM, Fein P. Importance of low serum intact parathyroid hormone as a predictor of mortality in haemodialysis and peritoneal dialysis patients: 14 years of prospective observation. Am J Kidney Dis 2001;38:1351–7. [29] Kalantar-Zadeh K, Block G, Humphreys MH, McAllister CJ, Kopple JD. A low, rather than a high, total plasma homocysteine is an indicator of
[30]
[31]
[32]
[33] [34] [35]
653
poor outcome in haemodialysis patients. J Am Soc Nephrol 2004;15: 442–53. Liu Y, Coresh J, Eustace JA, et al. Association between cholesterol level and mortality in dialysis patients: role of inflammation and malnutrition. JAMA 2004;291:451–9. Kopple JD. The phenomenon of altered risk factor patterns or reverse epidemiology in persons with advanced chronic kidney failure. Am J Clin Nutr 2005;81:1257–66. Ducloux D, Klein A, Kazory A, Devillard N, Chalopin JM. Impact of malnutrition-inflammation on the association between homocysteine and mortality. Kidney Int 2006;69:331–5. Silva RP, Barbosa PH, Kimura OS, et al. Prevalence of anaemia and its association with cardio-renal syndrome. Int J Cardiol 2007;120:232–6. Borrows R, Loucaidou M, Chusney G. Anaemia and congestive heart failure early post-renal transplantation. Nephrol Dial Transplant 2008;23:1728–34. Orth SR, Hallan SI. Smoking: a risk factor for progression of chronic kidney disease and for cardiovascular morbidity and mortality in renal patients-absence of evidence or evidence of absence? Clin J Am Soc Nephrol 2008;3:226–36.
Clinical Biochemistry 42 (2009) 648 – 653
Predictors for cardiovascular morbidity and overall mortality in Tunisian ESRD patients: A six year prospective study☆ Hayet Fellah a , Mohamed Bassem Hammami a , Moncef Feki a,⁎, Karima Boubaker b , Taieb Ben Abdallah b , Bernard Lacour c , Abderraouf Mebazaa a , Neziha Kaabachi a a
Lab-SM-01 Research Laboratory and Service of Biochemistry, Rabta Hospital, Tunis, Tunisia b Service of Internal Medicine and Nephrology, Charles Nicole Hospital, Tunis, Tunisia c Laboratory of Biochemistry A, Necker Hospital, Paris, France
Received 22 October 2008; received in revised form 16 December 2008; accepted 17 December 2008 Available online 8 January 2009
Abstract Objectives: The study was aimed to test the predictive value of several potential cardiovascular factors and markers for non fatal cardiovascular events (CVE) and overall mortality in Tunisian patients with renal failure. Subjects and methods: One hundred and fifteen renal failure patients were followed-up from 2000 to 2006. At enrollment, each patient underwent clinical examination and blood collection for analysis of lipid parameters, albumin, C reactive protein (CRP), parathyroid hormone (PTH), homocysteine and hemoglobin. Multivariate Cox regression models were applied to identify the predictors for non fatal CVE and overall mortality. Results: During the follow up, seventeen patients were lost. Among the 98 remaining patients, 29 presented a non fatal CVE (21.5%) and 15 were deceased (11.1%). In univariate analyses, non fatal CVE were more frequent in smokers and in patients with high PTH concentrations and low HDL levels. Moreover, low albumin concentrations were univariately associated with overall mortality. In the multivariate analysis, non fatal CVE was significantly and independently associated with age [hazard ratio (95% confidence interval), 1.04 (1.01–1.08); p = 0.028] and the upper quartile of PTH concentrations [2.68 (1.24–5.81); p = 0.013]. Overall mortality was independently predicted by the bottom quartile of albumin concentrations [5.62 (2.02–15.6); p = 0.001] and the upper quartile of CRP concentrations [3.20 (1.14–8.79); p = 0.027]. Conclusion: Advanced age and high PTH levels are the main predictors of CVE, whereas low albumin and high CRP concentrations are the independent predictors of death in Tunisian renal patients. A better control of these factors would greatly increase the patient's survival rates. © 2009 The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved. Keywords: Cardiovascular disease; Chronic renal failure; Hemodialysis; Mortality; Renal transplantation; Risk factor
Introduction The incidence and prevalence of end stage renal disease (ESRD) continues to increase worldwide. Mortality in patients Abbreviations: CRF, chronic renal failure; CRP, C reactive protein; CVE, cardiovascular events ; ESRD, end stage renal disease; HD, hemodialysis; HDLC, HDL cholesterol ; HHC, hyperhomocysteinemia; PTH, parathyroid hormone; RT, renal transplantation. ☆ This work was supported by the Ministry of Higher Education, Scientific Research and Technology of Tunisia. ⁎ Corresponding author. Laboratory of Biochemistry, Rabta Hospital, 1007 Jebbari, Tunis, Tunisia. Fax: +216 71 570 506. E-mail address: [email protected] (M. Feki).
with ESRD is dramatically high and cardiovascular disease is the leading cause of mortality among them [1,2]. Atherosclerotic heart disease is believed to account for approximately 55% of mortality and contributes to a 20-fold increase in ischemic heart disease and to a 10-fold increase in risk of stroke among these patients [3–5]. Such findings are thought to be related to the high prevalence of several cardiovascular risk factors such as hypertension, smoking, diabetes, dyslipemia, inflammation, malnutrition, anemia, hyperhomocysteinemia [HHC], hypercoagulability and oxidative stress [5–12]. However, the exact reasons for the high mortality have not been completely elucidated and the relative contribution of each factor to poor clinical outcomes was not well understood. The
0009-9120/$ - see front matter © 2009 The Canadian Society of Clinical Chemists. Published by Elsevier Inc. All rights reserved. doi:10.1016/j.clinbiochem.2008.12.023
H. Fellah et al. / Clinical Biochemistry 42 (2009) 648–653
identification of patients at high risk and the precise evaluation of the causes of death are crucial to devise effective preventive strategies. We had already found a high prevalence of cardiovascular risk factors or markers in Tunisian patients with ESRD [13,14]. This study was aimed to test the predictive value of several clinical and biochemical factors for non fatal cardiovascular events (CVE) and overall mortality after a 6 year follow-up in Tunisian ESRD patients. Subjects and methods Patients One hundred fifteen consecutive ESRD patients (58 male and 57 female) aged 16 to 68 years (40.8 ± 10.8), were enrolled in the study during the year 2000, and followed up during 6 years. They consisted of 35 chronic renal failure (CRF) patients on conservative treatment, 50 hemodialyzed (HD) patients and 30 renal transplant (RT) recipients undergoing therapy at the Nephrology Department of Charles Nicole Hospital (Tunis). All patients are Arab descendants who resided in the north of Tunisia. Patients with diabetes, thyroid diseases, systemic and active inflammatory diseases or those under lipid lowering drugs were excluded. Hemodialysis was performed for 4 h three times a week, using bicarbonate buffer and polysulfone dialysis membranes. The dialysis dose (Kt/V), defined as the dialyzer clearance of urea (K, in mL/min) multiplied by the duration of the dialysis treatment (t, in minutes) divided by the volume of distribution of urea in the body (V, in mL), averaged 1.12 ± 0.26 with a range between 0.50 and 1.72. CRF and HD patients were supplemented with iron (50–100 mg), calcium carbonate (500 mg), and B group vitamins (thiamine, 400 mg; pyridoxine, 80 mg) daily. RT recipients were studied at least 2 years after the transplantation and were taking cyclosporine (3–4 mg/kg), prednisone (10– 15 mg), azathioprine (75–125 mg) daily as immunosuppressive treatment. No patient received erythropoietin treatment. At enrollment, collected demographic and clinical data included age, gender, smoking, etiology of ESRD, hypertension and drug consumption. The causes of renal impairment were chronic glomerulonephritis (n = 39), hypertensive nephropathy (n = 26), pyelonephritis and chronic interstitial nephritis (n = 17) and polycystic kidney disease (n = 6). In the remaining cases, the etiology was unspecified or unknown. The protocol of the survey was approved by the Ethics Committee on Human Research of the Charles Nicole Hospital (Tunis, Tunisia) and all participants gave their free informed consent. Cardiovascular events CVE included coronary, cerebral and abdominal aortic or peripheral vascular disease. Coronary heart disease was diagnosed by at least one of the following criteria: 1) myocardial infarction documented by a typical rise and fall of troponine I with ischemic symptoms and/or electrocardiogram changes indicative of myocardial necrosis or ischemia (Q waves, ST segment elevation or depression), 2) symptoms
649
consistent with angina confirmed by a positive exercise stress test result and/or abnormal coronarography, 3) acute heart failure documented by clinical and echographic findings, 4) arrhythmia, and 5) left ventricular hypertrophy. Cerebrovascular disease was defined as a stroke (acute onset irreversible neurological deficit such as aphasia, focal deficit or unilateral paresis confirmed by computed tomography) or symptomatic extra cranial carotid artery stenosis confirmed by ultrasonography. Peripheral vascular disease was diagnosed by the development of intermittent claudication accompanied by diminished pulses and confirmed by Doppler or lower extremity amputation or deep venous thrombosis confirmed by venous Doppler. Biochemical analysis Fasting blood samples were collected into EDTA-containing tubes for total homocysteine and hemoglobin and into freeanticoagulant tubes for other parameters. For HD patients, blood was collected before dialysis session. Plasma homocysteine was assessed by a fluorescent polarizing immunoassay method on an Axsym system (Abbott Diagnostics). Serum intact PTH was measured by radioimmunoassay (Cis BioInternational). Hemoglobin was measured on a globule counter (Coulter Beckman). Serum albumin and C-reactive protein (CRP) were measured by an immunoturbidimetric method, and total cholesterol and triglycerides were assessed by standard methods on a Hitachi 912 analyzer (Roche Diagnostics). HDL cholesterol (HDL-C) was determined by direct method. Cardiovascular risk factors or markers In this study, we considered the following potential cardiovascular risk factors or markers; hypertension (diastolic blood pressure N 90 mm Hg and/or systolic blood pressure N140 mm Hg, or the use of antihypertensive therapy) and smoking (patients who currently smoke at least 5 cigarettes per day). For biological markers, we considered as “at risk values” those of the upper quartile (compared to the three lower quartiles) for homocysteine (N32.6 μmol/L), CRP (N 12.8 mg/L) and PTH
Table 1 Main characteristics of study participants
Men (%) Smoking (%) Hypertension (%) Age (years) CCr (mL/min) BMI (kg/m2)
CRF patients (n = 35)
HD patients (n = 50)
RT patients (n = 30)
Controls (n = 31)
49 37 60 *** 42.3 ± 12.2 21.7 ± 12.6 *** 23.6 ± 4.6 **
50 30 44 *** 41.4 ± 10.4 8.0 ± 2.0 *** 21.8 ± 4.3 ***
53 30 43 *** 37.9 ± 9.4 60.1 ± 24.3 *** 22.5 ± 3.3 ***
48 26 – 38.5 ± 10.1 97.6 ± 19.2 26.8 ± 2.9
Results are expressed as means ± SD or percent; CRF: chronic renal failure; HD: hemodialysis; RT: renal transplant; CCr: clearance of creatinine; BMI: body mass index. ⁎⁎ p b 0.01. ⁎⁎⁎ p b 0.001 (towards controls).
650
H. Fellah et al. / Clinical Biochemistry 42 (2009) 648–653
Table 2 Unadjusted relative risk with 95% confidence interval of several potential cardiovascular risk factor for non-fatal cardiovascular events and overall mortality in end stage renal disease patients (n = 98) n
Non fatal cardiovascular events (n = 29)
72 26 66 32 71 27 74 24 71 27 69 29 81 17 77 21 72 26
27.8 34.6 21.2 46.9 28.2 33.3 28.4 33.3 31.0 25.9 23.2 50.0 24.7 52.9 28.6 33.3 29.2 30.8
% Hypertension Smoking Dyslipemia Upper quartile of homocysteine Upper quartile of CRP Upper quartile of parathormone Lower quartile of HDL cholesterol Lower quartile of albumin Lower quartile of hemoglobin
No Yes No Yes No Yes No Yes No Yes No Yes No Yes No Yes No Yes
RR (95% CI)
p
%
0.513
15.3 19.2 15.2 18.8 15.5 18.5 14.9 20.8 12.7 25.9 17.4 13.6 18.5 5.9 10.4 38.1 13.9 23.1
1.38 (0.53–3.59) 0.009 2.56 (1.31–4.34) 0.617 1.28 (0.49–3.31) 0.644 1.26 (0.47–3.39) 0.624 0.78 (0.29–2.11) 0.017 3.31 (1.21–9.05) 0.020 3.44 (1.16–10.00) 0.772 1.25 (0.44–3.44) 0.878 1.08 (0.41–2.86)
Overall mortality (n = 16) RR (95% CI)
p 0.425
1.32 (0.41–4.24) 0.651 1.29 (0.42–4.00) 0.465 1.24 (0.39–3.97) 0.344 1.51 (0.47–4.88) 0.113 2.41 (0.80–7.31) 0.483 0.75 (0.20–2.94) 0.182 0.27 (0.03–2.22) 0.005 5.31 (1.69–16.67) 0.215 1.86 (0.60–5.760)
CRP, C-reactive protein; RR, relative risk; CI, confidence interval.
(N337 μg/L), and those of the bottom quartile (compared to the three upper quartiles) for albumin (b35 g/L), HDL-C (b25 mg/ dL) and hemoglobin (b 84 g/L). Dyslipemia was considered in subjects with total plasma cholesterol N 250 mg/dL and/or plasma triglycerides N200 mg/dL. Statistical analysis Statistical analyses were carried out using SPSS for windows 10.0 software (SPSS Inc., Chicago, USA). Comparison between groups was performed using chi-square test for categorical variables. For continuous variables, comparison was performed using Student t-test for normally distributed variables and Wilcoxon rank sum test for non-normally distributed ones. The Kaplan-Meier method was used to compute observed survival.
Relative risks for CVE or all-cause mortality were determined by means of univariate analysis and multivariate Cox proportional hazards regression (backward stepwise model), including several potential cardiovascular risk factors (i.e. age, hypertension, smoking, dyslipemia, clearance of creatinine, upper quartile for plasma homocysteine, CRP and PTH concentrations, and bottom quartile for albumin, HDL-C and hemoglobin levels). Significance was established at a p value below 0.05, based on two-tailed calculations. Results Demographic and clinical patient characteristics at enrollment are shown in Table 1. During the follow-up, seventeen patients (12.6%) were lost. Among the 98 remaining patients,
Fig. 1. Kaplan-Meier estimates of observed survival during follow-up with regards to non fatal cardiovascular events in patients stratified by enrollment HDL cholesterol (HDL-C) and parathyroid hormone (PTH) concentrations.
H. Fellah et al. / Clinical Biochemistry 42 (2009) 648–653
651
Fig. 2. Kaplan-Meier estimates of observed survival during follow-up with regards to all-cause mortality in patients stratified by enrollment plasma albumin and Creactive protein (CRP) concentrations.
29 presented a non fatal CVE (21.5%) and 15 were deceased (11.1%). CVE consisted of heart failure (n = 10), angina (n = 6), stroke (n = 3), myocardial infarction (n = 3), arrhythmia (n = 3); left ventricular hypertrophy (n = 2), lower extremities vascular disease (n = 1) and femoral deep venous thrombosis (n = 1). Death was attributed to a cardiovascular event (n = 7), infectious processes (n = 4), cancer (n = 3), and severe malnutrition (n = 1). Non fatal CVE were more frequently observed in smokers and in patients with high PTH concentrations and low HDL-C concentrations. Overall mortality was significantly higher in patients with low albumin concentrations. We also observed a tendency towards increased mortality for high CRP levels (Table 2). Survival curves for non fatal CVE according to the upper quartile of PTH and the lower quartile of HDL cholesterol values and survival curves for overall mortality according to the upper quartile of CRP and the lower quartile of albumin values were shown in Figs. 1 and 2. When applying multivariate Cox regression analysis, non fatal CVE remained significantly associated with age [hazard ratio (HR), 95% confidence interval (95% CI), 1.04 (1.01– 1.08); p = 0.028] and the upper quartile of PTH concentrations [HR (95% CI), 2.68 (1.24–5.81); p = 0.013]. Overall mortality Table 3 Stepwise backward Cox regression models for non fatal cardiovascular events and for overall mortality in Tunisian ESRD patients (n = 98) B coefficient SE Non fatal cardiovascular events Age 0.042 Upper quartile of PTH 0.985 Overall mortality Bottom quartile of albumin 1.727 Upper quartile of CRP 1.152
HR (95% CI)
p
0.019 1.04 (1.01–1.08) 0.028 0.395 2.68 (1.24–5.81) 0.013 0.521 5.62 (2.02–15.6) 0.001 0.523 3.20 (1.14–8.79) 0.027
SE, standard error; HR, hazard ratio; CI, confidence interval; PTH, parathyroid hormone; CRP, C-reactive protein. Entered covariates are age, hypertension, smoking, dyslipemia, clearance of creatinine, upper quartile for plasma homocysteine, C-reactive protein and parathyroid hormone, and bottom quartile for albumin, HDL cholesterol and hemoglobin.
was associated with the bottom quartile of albumin concentrations [HR (95% CI), 5.62 (2.02–15.6); p = 0.001] and the upper quartile of CRP concentrations [HR (95% CI), 3.20 (1.14– 8.79); p = 0.027] (Table 3). Discussion This study showed that low albumin and high CRP levels are the main predictors for death and that advanced age and high plasma PTH concentrations are the independent predictors for non fatal CVE in Tunisian ESRD patients. These data provide a comprehensive evaluation of the predictors for CVE and death in Tunisian patients with ESRD, offering the opportunity to evaluate the prognostic strength of these factors when analyzed together. The finding that hypoalbuminemia is the main prognostic factor for death was largely reported [15–17]. It was suggested that chronic inflammation may be the missing link or factor that causally ties hypoalbuminemia to morbidity and mortality [18,19]. However, data consistent with the possibility that malnutrition may affect survival independently exist [20]. The role of high CRP concentration as an independent predictor for mortality found in this study was previously demonstrated in ESRD patients [16,21–23]. High PTH level was found to predict CVE in our patients. Excess PTH has adverse effects on cardiac function and structure, including cardiac interstitial fibrosis and left ventricular hypertrophy and dysfunction [24,25]. While an association between low PTH and increased mortality was reported in renal patients, these low PTH levels are secondary to a low protein and phosphorus intake resulting in malnutrition that may explain the excess mortality [26–28]. Our findings have important implications for clinical practice. In fact, ESRD patients with low albumin and/or high CRP levels should receive close follow-up and all sources of malnutrition and inflammation should be controlled. The control of excess PTH also remains an important goal to minimize cardiovascular morbidity and mortality in renal patients.
652
H. Fellah et al. / Clinical Biochemistry 42 (2009) 648–653
This study failed to detect significant associations between CVE and/or mortality and potential predictors such as smoking, hypertension, dyslipemia, HHC and anemia. In reality, the great diversity of perturbations and the complexity of interactions between multiple perturbations in renal patients may modify the relationship between each factor and outcomes. It is possible that the dominant influence of some factors prevented the detection of factors that are of less prognostic importance. The literature showed very confusing data on the role of various factors in poor outcomes in renal patients [7–13,29–35]. These discrepancies may be related to differences in population characteristics, sampling size, observation time, method of analysis and the impact of confounding covariates on the outcome of multivariate analysis. In this study, the association of smoking and low HDL with CVE lost its significance in multivariate analysis. The main limitations of the study are the small number and the heterogeneity of patients. The choice to focus on overall mortality and not on cardiovascular mortality is chiefly driven by the low number of cardiac deaths. Our sample cannot be considered as representative of Tunisian ESRD patients as patients with diabetes mellitus, a common cause of ESRD, were not enrolled. This is one reason of the low sample size and may explain the low rate of cardiovascular death in our series. In return, the proper effect of diabetes and its interaction with predictors on outcomes are excluded. Our data arose from multivariate regression models that simultaneously controlled for several important predictors of the outcomes, including clearance of creatinine, making the conclusions valid. However, it must be recognized that association does not imply causation. This study showed that low albumin and high CRP and PTH levels are the main predictors for poor outcomes in Tunisian renal patients. More vigorous measures to control these factors may contribute to reduce CVE and to improve survival in these patients. Other factors such as hypertension, dyslipemia, anemia, and HHC were not found to predict poor outcomes, suggesting a probable lower prognostic importance. It is also possible that complexity of interactions between factors and methodological issues have modified the relationship between predictors and outcomes. To define the relative contribution of each predictor to clinical outcome, large scale randomized prospective interventional studies are required and until the results of such studies are available, increased vigilance in cardiovascular risk factors management is warranted and may be useful to reduce the burden of ESRD morbidity and mortality. Acknowledgments The authors gratefully acknowledge Prof. Mohamed Salah Mekni for helpful advice in the preparation of this manuscript. References [1] Foley RN, Parfrey PS, Sarnak MJ. Clinical epidemiology of cardiovascular disease in chronic renal disease. Am J kidney Dis 1998;32:112–9.
[2] Gupta R, Birnbaum Y, Uretsky BF. The renal patient with coronary artery disease: current concepts and dilemmas. J Am Coll Cardiol 2004;44: 1343–53. [3] USRDS: the United States Renal data System. USRDS 2003 annual data report. Am J Kidney Dis 2003;42:1–230. [4] Cheung AK, Sarnak MJ, Yan G, et al, HEMO Study Group. Cardiac diseases in maintenance haemodialysis patients: results of the HEMO Study. Kidney Int 2004;65:2380–9. [5] Stenvinkel P, Heimburger O, Paultre F, et al. Strong association between malnutrition, inflammation, and atherosclerosis in chronic renal failure. Kidney Int 1999;55:1899–911. [6] Eikelboom JW, Hankey GJ. Associations of homocysteine, C-reactive protein and cardiovascular disease in patients with renal disease. Curr Opin Nephrol Hypertens 2001;10:377–83. [7] Mallamaci F, Zoccali C, Tripepi G, et al. Hyperhomocysteinemia predicts cardiovascular outcomes in haemodialysis patients. Kidney Int 2002;61: 609–14. [8] Longenecker JC, Coresh J, Powe NR, et al. Traditional cardiovascular disease risk factors in dialysis patients compared with the general population: the CHOICE Study. J Am Soc Nephrol 2002;13:1918–27. [9] Avram MM, Blaustein D, Fein PA, Goel N, Chattopadhyay J, Mittman N. Hemoglobin predicts long-term survival in dialysis patients: a 15-year single-center longitudinal study and a correlation trend between prealbumin and hemoglobin. Kidney Int 2003;87:S6–S11. [10] Zoccali C, Mallamaci F, Tripepi G. Novel cardiovascular risk factors in end stage renal disease. J Am Soc Nephrol 2004;15:S77–80. [11] Suliman M, Stenvinkel P, Qureshi AR, et al. The reverse epidemiology of plasma total homocysteine as a mortality risk factor is related to the impact of wasting and inflammation. Nephrol Dial Transplant 2007;22:209–17. [12] Portolés J, López-Gómez JM, Aljama P. A prospective multicentre study of the role of anaemia as a risk factor in haemodialysis patients: the MAR Study. Nephrol Dial Transplant 2007;22:500–7. [13] Fellah H, Feki M, Hsairi M, et al. Hyperhomocysteinemia and ESRD: determinants and association with cardiovascular disease in Tunisian patients. Clin Chem Lab Med 2003;41:675–80. [14] Fellah H, Feki M, Feki H, et al. Classical and non classical cardiovascular risk factors in Tunisian patients with end-stage renal disease: prevalence and association with cardio vascular events. Clin Lab 2004;5:447–53. [15] Combe C, Chauveau P, Laville M, et al. French Study Group for Nutrition in Dialysis. Influence of nutritional factors and haemodialysis adequacy on the survival of 1,610 French patients. Am J Kidney Dis 2001;37:S81–8. [16] Menon V, Greene T, Wang X, et al. C-reactive protein and albumin as predictors of all-cause and cardiovascular mortality in chronic kidney disease. Kidney Int 2005;68:766–72. [17] Honda H, Qureshi AR, Heimbürger O, et al. Serum albumin, C-reactive protein, interleukin 6, and fetuin a as predictors of malnutrition, cardiovascular disease, and mortality in patients with ESRD. Am J Kidney Dis 2006;47:139–48. [18] Kaysen GA. Malnutrition and the acute-phase reaction in dialysis patients—how to measure and how to distinguish. Nephrol Dial Transplant 2000;15:1521–4. [19] Yeun JY, Levine RA, Mantadilok V, Kaysen GA. C-reactive protein predicts all-cause and cardiovascular mortality in haemodialysis patients. Am J Kidney Dis 2000;35:469–76. [20] Leon JB, Majerle AD, Soinski JA, Kushner I, Ohri-Vachaspati P, Sehgal AR. Can a nutrition intervention improve albumin levels among haemodialysis patients? A pilot study. J Renal Nutr 2001;11:9–15. [21] Owen WF, Lowrie EG. C-reactive protein as an outcome predictor for maintenance haemodialysis patients. Kidney Int 1998;54:627–36. [22] Wang AY, Woo J, Lam CW, et al. Is a single time point C-reactive protein predictive of outcome in peritoneal dialysis patients? J Am Soc Nephrol 2003;14:1871–9. [23] Pearson TA, Mensah GA, Alexander RW, et al. Markers of inflammation and cardiovascular disease: application to clinical and public health practice: a statement for health-care professionals from the Centers for Disease Control and Prevention and the American Heart Association. Circulation 2003;107:499–511.
H. Fellah et al. / Clinical Biochemistry 42 (2009) 648–653 [24] Stefenelli T, Abela C, Frank H, et al. Cardiac abnormalities in patients with primary hyperparathyroidism: implications for followup. J Clin Endocrinol Metab 1997;82:106–12. [25] Strozecki P, Adamowicz A, Nartowicz E, Odrowaz-Sypniewska G, Wlodarczyk Z, Manitius J. Parathormone, calcium, phosphorus, and left ventricular structure and function in normotensive haemodialysis patients. Ren Fail 2001;23:115–26. [26] Akizawa T, Kinugasa E, Kurihara S, et al. Parathyroid hormone deficiency is an indicator of poor nutritional state and prognosis in dialysis patients. J Am Soc Nephrol 1999;10:615 A. [27] Chertow GM, Lazarus JM, Lew NL, Lowrie EG. Intact parathyroid hormone is directly related to mortality in haemodialysis. J Am Soc Nephrol 2000;11:573A (abstract). [28] Avram MM, Mittman N, Myint MM, Fein P. Importance of low serum intact parathyroid hormone as a predictor of mortality in haemodialysis and peritoneal dialysis patients: 14 years of prospective observation. Am J Kidney Dis 2001;38:1351–7. [29] Kalantar-Zadeh K, Block G, Humphreys MH, McAllister CJ, Kopple JD. A low, rather than a high, total plasma homocysteine is an indicator of
[30]
[31]
[32]
[33] [34] [35]
653
poor outcome in haemodialysis patients. J Am Soc Nephrol 2004;15: 442–53. Liu Y, Coresh J, Eustace JA, et al. Association between cholesterol level and mortality in dialysis patients: role of inflammation and malnutrition. JAMA 2004;291:451–9. Kopple JD. The phenomenon of altered risk factor patterns or reverse epidemiology in persons with advanced chronic kidney failure. Am J Clin Nutr 2005;81:1257–66. Ducloux D, Klein A, Kazory A, Devillard N, Chalopin JM. Impact of malnutrition-inflammation on the association between homocysteine and mortality. Kidney Int 2006;69:331–5. Silva RP, Barbosa PH, Kimura OS, et al. Prevalence of anaemia and its association with cardio-renal syndrome. Int J Cardiol 2007;120:232–6. Borrows R, Loucaidou M, Chusney G. Anaemia and congestive heart failure early post-renal transplantation. Nephrol Dial Transplant 2008;23:1728–34. Orth SR, Hallan SI. Smoking: a risk factor for progression of chronic kidney disease and for cardiovascular morbidity and mortality in renal patients-absence of evidence or evidence of absence? Clin J Am Soc Nephrol 2008;3:226–36.