Impact of protein energy wasting status on survival among Afro-Caribbean hemodialysis patients: a 3-year prospective study

Abstract

Background: We assessed the prognostic value of protein-energy wasting (PEW) on mortality in Afro-Caribbean MHD patients and analysed how diabetes, cardiovascular disease (CVD) and inflammation modified the predictive power of a severe wasting state.

Method: A 3-year prospective study was conducted in 216 patients from December 2011. We used four criteria from the nomenclature for PEW proposed by the International Society of Renal Nutrition and Metabolism in 2008: serum albumin 38 g/L, body mass index (BMI) ≤23 kg/m(2), serum creatinine ≤818 µmol/L and protein intake assessed by nPCR ≤0.8 g/kg/day. PEW status was categorized according the number of criteria. Cox regression analyses were used.

Results: Forty deaths (18.5 %) occurred, 97.5 % with a CV cause. Deaths were distributed as follows: 7.4 % in normal nutritional status, 13.2 % in slight wasting (1 PEW criterion), 28 % in moderate wasting (2 criteria) and 50 % in severe wasting (3-4 criteria). Among the PEW markers, low serum albumin (HR 3.18; P = 0.001) and low BMI (HR 1.97; P = 0.034) were the most significant predictors of death. Among the PEW status categories, moderate wasting (HR 3.43; P = 0.021) and severe wasting (HR 6.59; P = 0.001) were significant predictors of death. Diabetes, CVD, and inflammation were all additives in predicting death in association with severe wasting with a strongest HR (7.76; P < 0.001) for diabetic patients.

Conclusions: The nomenclature for PEW predicts mortality in our Afro-Caribbean MHD patients and help to identify patients at risk of severe wasting to provide adequate nutritional support.

Keywords: Afro-Caribbean patients; Body mass index; Hemodialysis; Mortality; Protein-energy wasting; Serum albumin.

Fig. 1

Survival curves according to protein-energy wasting status (P < 0.001, log-rank test)

Fig. 2

Severe wasting (at least three criteria for PEW) and additive markers (diabetes, previous cardiovascular event and inflammation) as predictors of all causes of mortality. PEW protein energy wasting, CVD previous cardiovascular disease, Inflam inflammation. Patients were divided into four groups according to PEW (presence/absence) and the additive factor (presence/absence). Model 1: adjustment for age, sex, diabetes and inflammation. Model 2: adjustment for age, sex, previous cardiovascular event and inflammation. Model 3: adjustment for age, sex, diabetes and previous cardiovascular event