Inflammation, Malnutrition, and Cardiac Disease as Predictors of Mortality in Hemodialysis Patients
A. Rashid Qureshi,Anders Alvestrand,José C. Divino-Filho,Alberto Gutierrez,Olof Heimbürger,Bengt Lindholm,Jonas Bergström
DOI: https://doi.org/10.1681/asn.v13suppl_1s28
IF: 14.978
2002-01-01
Journal of the American Society of Nephrology
Abstract:ABSTRACT. Various studies suggest a strong association between nutrition and clinical outcome in hemodialysis (HD) patients. Several morbidity factors that per se increase the risk of a poor outcome, such as cardiovascular disease (CVD) and inflammation, may also cause malnutrition. Among laboratory parameters used to assess nutritional status, serum albumin appears to be a particularly strong predictor of morbidity and mortality. This study assessed the importance of nutritional status and inflammation and other comorbidity factors as predictors of mortality in HD patients. Nutritional status was evaluated in 128 HD patients by subjective global nutritional assessment (SGNA) and by measuring several anthropometric markers (actual body weight, percentage of actual body weight to desirable body weight, midarm muscle circumferences, triceps skinfold thickness), and serum albumin, plasma insulin such as insulin growth factor-1 and as a marker of inflammation, serum C-reactive protein (s-CRP) levels. The mortality during the next 36 mo was analyzed in relation to age, gender, CVD, SGNA, serum albumin, CRP, and several other factors by Kaplan-Meier analysis multivariate. Cox proportional hazard analysis was used to identify independent predictors of mortality. After 36 mo, 58 patients were still on HD treatment, 57 patients (45%) had died while receiving treatment, and 13 had received a kidney transplant. The main cause of death was CVD (58%), followed by infection (18%); malnutrition/cachexia was a rare direct cause of death (5%). Kaplan-Meier analysis showed that age, female gender, CVD, diabetes, SGNA, all anthropometric parameters, serum albumin, plasma insulinlike growth factor-1, and s-CRP were significant predictors of mortality. Analysis by the Cox model showed that age, gender, CVD, nutritional status (SGNA), and CRP were independent predictors of mortality at 36 mo. A low albumin level was not an independent predictor, although it was strongly associated with a reduced survival rate in the Kaplan-Meier analysis. Inflammation, malnutrition, and CVD appeared to contribute to increased mortality in a stepwise manner. The mortality at 36 mo was 0% when none of these complications was present, whereas the mortality was 75% in those patients with all three risk factors present at baseline. It is concluded that in addition to malnutrition and comorbidities (CVD, diabetes mellitus), inflammation (elevated s-CRP) is a significant independent risk factor for mortality in HD patients. Inflammation, malnutrition, and CVD appear to be interrelated, each additionally contributing to the high mortality in these patients.
urology & nephrology