Global and cardiovascular mortality and risk factors in patients under hemodialysis treatment

Abstracts

FUNDAMENTO: Mortalidade global e cardiovascular (CV) elevada de pacientes em hemodiálise. OBJETIVO: Avaliação da mortalidade global e CV e identificação do risco de pacientes em hemodiálise. MÉTODOS: Estudo observacional, prospectivo. Estudados 334 pacientes em três anos. Desfechos primários: mortalidade global e CV. Sobrevida avaliada pelo método de Kaplan-Meier. Identificação de variáveis de risco pela Regressão de Cox, bi e multivariada. RESULTADOS: Foram estudados 189 (56,6%) homens, idade 48,8 ± 14,2 anos, maioria de não brancos (295[88,3%]) e com escolaridade de 0 a menor que 8 anos (211[63,2%]). Mortalidade total de 21,6% (72/334), 50% sobrevivendo 146 meses, e mortalidade CV de 41,7%(30/72), 75% sobrevivendo 141 meses. Na análise bivariada, o RR de óbito não cardiovascular (ONCV) e CV aumentou com Idade >60 anos, Hb < 9,0 g/dl e glicemia de jejum >126 mg/dl; de ONCV apenas, com baixa escolaridade, viuvez, Hb<11,0 g/dl, Ht<33,0%, glicemia de jejum >100 mg/dl, produto Ca x P <42 e creatinina >9,2 mg/dl; diminuiu com PA>140/90 mmHg (antes da sessão de HD) e Ht>36%; de óbito CV apenas, aumentou com creatinina >9,4 mg/dl. Na análise multivariada, o RR de ONCV e CV aumentou com idade >60 anos e Hb<9 g/dl; o RR de óbito CV aumentou com glicemia>126 mg/dl e o de ONCV com taxa de remoção de ureia na hemodiálise (Kt/V) <1,2. CONCLUSÃO: A mortalidade global e CV de pacientes em hemodiálise é elevada. Os fatores de risco independentes para ONCV e CV foram idade >60 anos e Hb<9 g/dl, para óbito CV apenas glicemia >126 mg/dl e ONCV Kt/V<1,2. Vale assinalar a importância do monitoramento na correção e prevenção desses três últimos fatores.

Mortalidade; letalidade; diálise renal; falência renal crônica

FUNDAMENTO: Mortalidad global y cardiovascular (CV) elevada de pacientes en hemodiálisis OBJETIVO: Evaluación de la mortalidad global y CV e identificación del riesgo de paciente en hemodiálisis. MÉTODOS: Estudio observacional, prospectivo. Estudiados 334 pacientes en tres años. Desenlaces primarios: mortalidad global y CV. Sobrevida evaluada por el método de Kaplan-Meier. Identificación de variables de riesgo por la Regresión de Cox, bi y multivariada. RESULTADOS: Se estudiaron 189 (56,6%) hombres, edad 48,8 ± 14,2 años, mayoría no blancos (295[88,3%]) y con escolaridad de 0 a menor que 8 años (211[63,2%]). Mortalidad total de 21,6% (72/334), 50% sobreviviendo 146 meses, y mortalidad CV de 41,7% (30/72), 75% sobreviviendo 141 meses. En el análisis bivariado, el RR de óbito no cardiovascular (ONCV) y CV aumentó con edad > 60 años, Hb < 9,0 g/dl y glucemia en ayunas > 126 mg/dl; sólo de ONCV, con baja escolaridad, viudez, Hb < 11,0 g/dl, Ht < 33,0%, glucemia en ayunas > 100 mg/dl, producto Ca x P < 42 y creatinina > 9,2 mg/dl; disminuyó con PA > 140/90 mmHg (antes de la sesión de HD) y Ht > 36%; de óbito CV solamente, aumentó con creatinina > 9,4 mg/dl. En el análisis multivariado, el RR de ONCV y CV aumento con edad > 60 años y Hb < 9 g/dl; el RR de óbito CV aumentó con glucemia > 126 mg/dl y el de ONCV con tasa de remoción de urea en la hemodiálisis (Kt/V) < 1,2. CONCLUSIÓN: La mortalidad global y CV de pacientes en hemodiálisis es elevada. Los factores de riesgo independientes para ONCV y CV fueron edad > 60 años y Hb < 9 g/dl, para óbito CV sólo glicemia > 126 mg/dl y ONCV Kt/V < 1,2. Cabe señalar la importancia del monitoreo en la corrección y prevención de estos tres últimos factores.

Mortalidad; letalidad; diálisis renal; falla renal crónica

ORIGINAL ARTICLE

Escola Bahiana de Medicina e Saúde Pública, BA - Brazil

Mailing address

ABSTRACT

BACKGROUND: There is a high global and cardiovascular mortality rate among patients who need hemodialysis.

OBJECTIVE: To assess global and cardiovascular mortality and to identify the risk factors in patients who undergo hemodialysis.

METHODS: Observational. prospective study. A total of 334 patients were studied within three years. Primary outcomes: global and cardiovascular mortality. Survival was assessed through Kaplan-Meier method. and the risk variables were identified by means of bivariate and multivariate Cox regression.

RESULTS: A total of 189 men (56.6%). aging 48.8 ± 14.2. majority non-white (295. 88.3%) and who did finished the elementary school (211. 63.2%). Global mortality rate was 21.6%. with a 50% rate of 146-month survival; cardiovascular mortality rate was 41.7% (30/72). with a 75% rate of 141-month survival. In the bivariate analysis. the relative risk (RR) for non-cardiovascular and cardiovascular death increased when age >60 years old was Hb <9.0 g/dL and fast glycemia >126 mg/dL. Only non-cardiovascular death with low school grade and widow. Hb<11.0 g/dL. Ht<33.0%. fast glycemia>100 mg/dL. Ca product x P<42 and creatinine >9.2 mg/dL decreased with blood pressure (BP) >140/90 mmHg (before hemodialysis session) and Ht>36%; Obit due only to cardiovascular factors increased with creatinine >9.4 mg/dL. In the multivariate analysis. non-cardiovascular and cardiovascular RR increased with age >60 years old and Hb<9 g/dL; cardiovascular death RR increased with glycemia >126 mg/dL. and non-cardiovascular death RR increased with urea removal rate in hemodialysis (Kt/V) <1.2.

CONCLUSION: Global and cardiovascular mortality of patients who need hemodialysis is high. Independent risk factors for non cardiovascular and cardiovascular causes of death were age >60 years old and Hb<9 g/dL. for cardiovascular cause of death only. was fasting blood glucose >126 mg/dL. and for non-cardiovascular cause of death. Kt/V<1.2.

Key words: Mortality; letality/cardiovascular diseases; renal dialysis; kidney failure chronic.

Introduction

Chronic kidney disease is considered to be a public health problem, and its incidence has been consequently increasing due to diabetes and hypertension cases and to the ageing of population. In Brazil, according to the 2006 Census of Sociedade Brasileira de Nefrologia, there were 70,872 patients under dialysis treatment, with 64,306 under hemodialysis treatment¹. Cardiovascular mortality among hemodialysis patients is usually elevated (40 to 50% of the population with chronic kidney failure), a rate that is 10 to 20 times higher than in general population². Traditional risk factors, as well as those related to renal insufficiency and to hemodialysis process, such as anemia, chronic inflammation, malnutrition, left ventricle hypertrophy, increase in calcium-phosphorus product and low Kt/v, participate in this rate³.

In order to identify the frequency of such events and the associated risk factors, a cohort prospective study was carried out during three years in a hemodialysis center of Salvador, Bahia, Brazil.

Methods

Longitudinal study which included 334 patients submitted to clinical and laboratorial assessment from February 2nd 2004 to January 1st 2007. Patients with 18 years old or more and who signed the informed consent were studied. The study included patients who were already under treatment with hemodialysis and those who started the treatment after that date. However, the survival curves for global and cardiovascular mortality were gauged from the day of the first hemodyalisis session, for all. All patients were submitted to clinical examination, as the blood pressure was checked before and during the hemodialysis session, according the Brazilian Society of Hemodyalisis criteria, by means of an aneroid sphygmomanometer (Tycos^®). Routine laboratorial assessment included hemoglobin, hematocrit, leucogram, serum iron, glycemia, albumin, creatinine, urea, potassium, Ca x P product and Kt/v (which assesses the efficacy of hemodialysis based on urea emotion rate, as regulating the volume to be cleaned in the session). Moreover, 284 patients took total cholesterol (TC) dosage, HDL cholesterol and triglycerides (TG). LDL cholesterol was calculated through Friedewald formula [LDL-C = (TC-HDL-C) - (TG/5)] for levels of TG lower than 400 mg/dL. A total of 158 patients went through posterior-anterior thoracic radiography, and the cardiothoracic index (CTI) was considered altered when superior to 50%. In 179 patients, electrocardiogram was carried out in the 12 derivations, and left ventricle hypertrophy (LVH) was diagnosed based on Sokolow-Lyon criteria and on Cornell University⁴. Continuum variables were expressed in mean ± standard deviation (SD) or median and interquartile interval, according to normality or asymmetry of its distribution, respectively, and the categorical variables were expressed in percentiles. Global, cardiovascular and non-cardiovascular mortality were assessed. Survival curves were built through Kaplan-Meier method. For the evaluation of mortality-related risk factors, non-cardiovascular and cardiovascular mortalities were compared. Bivariate analysis was carried out through Cox regression model, with 95% confidence interval. For the adjustment of Cox multivariate models, the backward algorithm was used, as variables of bivariate analysis which presented p<0.15 were included. Epi-Info version 6 was the utilized database, and Cox regression analysis was made through STATA software, version 10. The study was approved by the Ethics Committee of Fundação Bahiana para o Desenvolvimento das Ciências.

Results

A total of 334 patients were assessed, with mean age of 48.8 ± 14.2 years old, 189 (56.6%) males, 295 (88.8%) mulattos and black individuals, and 211 (63.2%) with a low education level, including illiterates (33), and those with incomplete first grade (178). Table 1 shows the patients' demographic, clinical and laboratorial data according to their conditions of alive or dead, after a three-year observation. Death cases were significantly older (56.9±14.6 [23-87] versus 46.6±14.9 [18-88] years old, p<0.001), and presented with a higher prevalence of antecedents of acute myocardial infarction (7 [19.7%] versus 3 [3.4%], p=0,027).88] years old, p<0.001). The alive and dead people did not differ with regard to skin color, marital status, educational level and other cardiovascular risk antecedents. The same occurred in relation to blood pressure and weight. The laboratorial assessment showed, in obits, a higher anemia grade (9.8 ± 2.5 g/dL versus 10.7 ± 2.1 g/dL, p=0.001 e Ht 30.6 ± 6.6% versus 32.9 ± 7.0%, p=0.001), hyperglycemia (121.5 [88.0] mg/dL versus 94.0 [40.0] mg/dL, p<0,001), lower plasmatic creatinine concentration (7.6±4.1 mg/dL versus 9.5±3.8 mg/dL, p<0.001) and tendency to higher cardiomegaly in the 158 patients that went through thoracic X-ray (CTI), 57.2 [5.9] versus 53.1 [9.8], p<0.055).

In Table 1, it is possible to observe that the majority of the 72 obits (30, 41.7%) occurred due to cardiovascular cause, which is divided into cardiac, 26/30 (86.7%), which is the main one, and cerebral vascular causes, 4/30 (13.3%), followed by infectious cause, 16/72 (22.2%). Approximately a quarter of these patients, 20 (27.8%), died by causes qualified as CID-10, R 68.8 (other causes-related symptoms).

Survival curve regarding obit by all causes shows that 50% of the patients survive 146 months (12.2 years) after the beginning of hemodialysis treatment (Figure 1), while 75% of the patients survived for approximately 141 months with regard to cardiovascular obits (11.7 years) (Figure 2).

The comparison through bivariate analysis of risk factor associated with non-cardiovascular and cardiovascular mortality, respectively, showed that age >60 years old, Hb<9.0 g/dL and fast glycemia >126 mg/dL significantly increased both kind of obits (Table 2). Non-cardiovascular obit risk also increased significantly with incomplete first grade school, widowers, Hb<11.0 g/dL, Ht<33%, fast glycemia >100 mg/dL, Ca x P product <42 and Kt/V <1.2, and decreases with BP>140/90 mmHg before hemodialysis session, and Ht>36%. Creatinine >9.2 mg/dL, on the other hand, significantly increased non-cardiovascular obit risk, while >9.4 mg/dL increased cardiovascular obit risk, respectively. Therefore, in multivariate model, in which these risk variables are comprised (Table 3), only age >60 years and Hb<9.0 g/dL remained as independent risk variables for non-cardiovascular and cardiovascular obit; glycemia >126 mg/dL remained as independent risk for cardiovascular obit and Kt/V<1,2 as independent risk for non-cardiovascular obit.

Discussion

The numerous obit occurrences by all causes, specifically by cardiovascular cause, is in compliance with literature data, indicating that chronic kidney failure, even at the period of advanced chronic hemodialysis treatment, is a high risk condition for global and cardiovascular obit³. In private, the analysis of the survival curves indicates that the cardiovascular disease, besides its elevated frequency, was associated with a shorter period of hemodialysis treatment. The risk factors, the frequent cardiovascular antecedents among obit cases and the presence of cardiomegaly, as indicated by increase in CTI, are suggestive of severity of vascular disease in these patients and, certainly, contributed to this outcome. Among risk factors, the emphasis in given the following variables as independent factors: age equal or superior to 60 years old and Hb< 9,0 g/dL for non-cardiovascular as much as cardiovascular risk, hyperglycemia equal or superior to 126 mg/dL for cardiovascular obit and urea removal rate (Kt/V) inferior to 1.2 for non-cardiovascular obit. Along with these variables, a series of co-variables showed to be influent with regard to mortality, although not in an isolate way, but by a conjunct action that, however, must be considered from the clinical point of view. This context comprises widowers, normal blood pressure values before hemodialysis session, lower grades of anemia (Hb >9.0 g/dL and <11.0 g/dL), fast hyperglycemia between 100 and 126 mg/dL and reduced phosphatemia values. Despite their ethiopathogenic importance, dislipidemia, lower serum albumin levels, hyperphosphatemia, hyperkalemia and eccentric left ventricle hypertrophy are nor presented as a differential risk in this sample, which evidently does not reduce the related preventive and therapeutic care. It is worth to emphasize the small reduction of serum albumin concentration that did not present a significant difference between life and death cases, though it was higher in the latter. These median albumin levels equal or superior to 3.4 g/dL, in these hemodialysis patients, are suggestive of a jeopardized nutritional status, which may contribute to a low mortality rate.

Among the three mentioned independent risk factors, the most important one, due to its pathogenic actuation amplitude and control possibility, is anemia. In patients with end stage renal disease, it represents the most frequent and neglected risk factor for non-cardiovascular and cardiovascular obit, which may not be adequately valorized because it is common and is not placed among traditional ones³. In the present paper, for each diminution Hb gram inferior to 9.0 g/dL, the relative risk for non-cardiovascular obit would increase in approximately 2.37 times, and for cardiovascular obit, in 3.4 times. Still, this condition remains inadequately treated during predialysis phase of renal disease^5,6, with aggravation in dialysis phase⁶, despite the current availability of erythropoietin by SUS. It is important to point out that, beside the systemic effects of chronic anemia, it is an important cause of eccentric hypertrophy of the heart and of myocardial fibrosis, factors that may lead to congestive heart failure^6,7, to which an elevate prevalence for hypertension is related (70% of the patients in this study), chronic inflammation, insulin resistance (altered fast glycemia) and type 2 diabetes⁸, all factors that contribute to the atherosclerosis process⁹. Though low educational level is a matter of social amplitude and complex in its short and medium-term solution, it must be taken into account, for it affects significantly the portion of the society that comprises the majority of the patients in hemodialysis treatment. A continuum educational support offered by the interdisciplinary team, which is necessary for supporting this type of program, may constitute a feasible solution. Age, as a progressive risk factor and in its biological characteristics, indicates the necessity of a more rigorous control of the multiple risk factors present in the elderly and already mentioned in the literature³.

With regard to the other risk co-variables identified by the bivariate analysis, the matters of blood pressure with reduced relative risk in 45% when equal to or higher than 140/90 mmHg before hemodialysis session, as well as the increase of relative risk with Ca x P product <42 and lipidic profile neutrality deserve a specific comment. The presence of a U correlation between systolic blood pressure (SBP) and mortality of patients under hemodialysis treatment was reported by Zager et al¹⁰, which is in compliance with current data and shows that the optimum level for SBP is not yet determined, and should be motif for longitudinal studies. The increased risk of global mortality by the reduced Ca x P product is not reported in the literature, even in patients that were given sevelamer chloride^11-14, a current medication that is specific for hyperphophatemia control. Face to that, the findings should be registered for the purpose of future assessment. One of the limitations of this study was the fact that the complementary examinatios (lipidic profile, magnesium dosage, echocardiogram and thoracic X-ray) were initiated a period after the beginning of the research, which kept some patients from participating in the evaluation due to obit or refusal to go through the examinations. Another limitation, naturally, comes from the observational characteristics of the research, as data were obtained by means of the routine protocol of the institution, and not of the applicability of a research protocol. At last, a financial limitation did not allow cardiac alterations to be also assessed by echocardiography, and the high sensibility C-reactive protein to be determined in order to assess the grade of the inflammatory process that happens in these patients.

Conclusions

The mortality of patients under hemodialysis treatment as substitutive renal therapy is still elevated, and cardiovascular disease strongly contributes for such rates. Independent non-cardiovascular and cardiovascular mortality risks were: age >60 years old and Hb <9.0 g/dL; fast glycemia >126 mg/dL was an independent cardiovascular mortality risk and urea removal rate during hemodialysis (Kt/V) lower than 1.2 of non-cardiovascular mortality. It is important to emphasize the possibility of monitoring, correction and prevention of such independent, modifiable mortality factors.

Acknowledgments

The authors would like to thank Carlos Teles for the statistical analysis and Gabrielita C. Machado for the help in the bibliographical review.

Potential Conflict of Interest

No potential conflict of interest relevant to this article was reported.

Sources of Funding

This study was funded by FAPESB.

Study Association

This article is part of the thesis of doctoral submitted by Fátima Aparecida Afonso Almeida, from Escola Bahiana de Medicina e Saúde Pública.

References

• 1. Sociedade Brasileira de Nefrologia. Censos. [Acesso em 2007 jul 7]. Disponível em: http://www.sbn.org.br/censos.htm
• 2. Silva Jr ACC, Abensur H, Lotaif LD, Amodeo C, Piegas LP. Novos fatores de risco cardiovascular. Rev Soc Cardiol Estado de São Paulo. 2007; 17 (1): 50-9.
• 3. Canziani ME. Doenças cardiovasculares na doença renal crônica. J Bras Nefrol. 2005; 26 (supl. 1): 20-1.
• 4. Gasperin CA, Germiniani H, Facin AR, Souza AM, Cunha CLP. Análise dos critérios eletrocardiográficos para determinação da sobrecarga ventricular esquerda. Arq Bra Cardiol. 2002; 78 (1): 59-71.
• 5. Isek K, Kohagura K. Anemia as a risk factor for chronic kidney disease. Kidney Int. 2007; 72: 54-9.
• 6. Zalunardo N, Levin A. Anemia and the heart in chronic kidney disease. Semin Nephrol. 2006; 26: 290-5.
• 7. Li S, Collins AJ. Association of hematocrit value with cardiovascular morbidity and mortality in incident hemodialysis patients. Kidney Int. 2004; 65 (2): 626-33.
• 8. Dogra G, Irish A, Chan D, Watts G. Insulin resistance, inflammation, and blood pressure determine vascular dysfunction in CKD. Am J Kidney Dis. 2006; 48: 926-34.
• 9. Grundy SM, Cleeman JI, Daniels SR, Donato KA, Eckel RH, Franklin BA, et al. Diagnosis and management of the metabolic syndrome. An American Heart Association/National Heart, Lung, and Blood Institute Scientific Statement. Circulation. 2005; 112: 2735-52.
• 10. Zager PG, Nikolic J, Brown RH, Campbell MA, Hunt WC, Peterson D, et al. "U" curve association of blood pressure and mortality in hemodialysis patients. Kidney Int. 1998; 54: 561-9.
• 11. Chertow GM, Burke SM, Dillon MA, Slatolposky E, for the RenaGel Study Group. Long-term effects of sevelamer hydrochloride on calcium x phosphate product and lipid profile of hemodialysis patients. Nephrol Dial Transplant. 1999; 14: 2907-14.
• 12. Carvalho AB, Cuppari L. Controle da hiperfosfatemia na DRC. J Bras Nefrol. 2008; 30 (supl 2): 4-8.
• 13. Delmez, J, Block G, Robertson J, Chasan-Taber S, Blair A, Dillon M, et al. A randomized, double-blind crossover design study of sevelamer hydrochloride and sevelamer carbonate in patients on hemodialysis. Clin Nephrol. 2007; 68 (6): 386-91.
• 14. Shantouf R, Budoff MJ, Ahmadi N, Tiano J, Flores F, Kalantar-Zadeh K. Effects of sevelamer and calcium-based phosphate binders on lipid an inflammatory markers in hemodyalisis patients. Am J Nephrol. 2008; 28: 275-9.

Global and cardiovascular mortality and risk factors in patients under hemodialysis treatment

Fátima Aparecida A. Almeida; Felipe Carrhá Machado; José Andrade Moura Junior; Armênio Costa Guimarães

Publication Dates

• Publication in this collection
  15 Jan 2010
• Date of issue
  Feb 2010

History

• Accepted
  15 May 2009
• Reviewed
  14 Dec 2008
• Received
  20 Aug 2008