The Impact of Malnutrition on Mortality in Patients on Maintenance Hemodialysis in Serbia

Abstract

Numerous studies suggest a strong association between nutrition and clinical outcome in chronic hemodialysis (HD) patients. While determination of malnutrition is often based on objective measurements, such as biochemical parameters and anthropometric data, there is no single measurement that can reliably predict the risk for malnutrition or poor outcome. The aim of the present investigation was to determine the prevalence and severity of malnutrition among HD patients in a large university-affiliated HD center in Serbia, and to examine the relationship between various nutritional and nonnutritional factors, and the clinical outcome in the period of 20 months follow-up. The prospective observational cohort study included patients (n = 197) with more than 3 months on HD treatment before entering the study. Global nutritional status was evaluated by the dialysis malnutrition score (DMS). Body mass index (BMI), triceps skinfold thickness (TSF), mid-arm circumference (MAC), and mid-arm muscle circumference (MAMC), as well as several laboratory parameters and clinical examination, were recorded. Dose of HD and protein equivalence of nitrogen appearance normalized to ideal body weight (nPNA) were evaluated by the urea kinetic model. Mortality data were collected prospectively during the 20 months of follow-up. A moderate/severe degree of malnutrition was presented in 39.2% of female and 22.8% of male patients, while signs of mild malnutrition existed in 45.5% and 66.9% of patients, respectively. Kaplan-Meier survival analysis showed that in the entire group of patients with DMS score >10, the mortality rate was higher (log rank 5.61; P < 0.05) than in those with DMS score <or=10. Multiple logistic regression analysis (after adjusting for age and gender) showed that the predictors of malnutrition were: comorbidity expressed as index of coexistent diseases (ICED) (odds ratio [OR] 10.2; P < 0.001), length of time on HD treatment before entering the study (OR 1.29; P < 0.05), and nPNA (OR 23.62; P < 0.05). At the end of the 20 months follow-up period, a total of 40 patients died (20.3%) with cardiovascular diseases as the main cause of death (45% of the cases). After adjusting for age and sex, only comorbid medical problems expressed as ICED (OR 5.28; P < 0.01), DMS (OR 1.21; P < 0.01), and C-reactive protein (CRP) (OR 1.04; P < 0.05) were significant predictors of mortality. In conclusion, malnutrition in the Serbian cohort of patients on HD was very common. Nutritional status at enrollment (expressed as DMS), presence of comorbidity (ICED index), and CRP level were strong predictors of mortality during the 20 months of follow-up.