Relationship between serum protein and mortality in adults on long-term hemodialysis: exhaustive review and meta-analysis

SOURCE: Nutrition
OUTPUT TYPE: Journal Article
TITLE AUTHOR(S): M.Herselman, N.Esau, J.-M.Kruger, D.Labadarios, M.R.Moosa
Print: HSRC Library: shelf number 6156
HANDLE: 20.500.11910/4452

If you would like to obtain a copy of this Research Output, please contact Hanlie Baudin at


The aim of this exhaustive review and meta-analysis was to explore the relation among serum protein, inflammatory markers, and all-cause and cardiovascular mortalities in adult patients on maintenance hemodialysis. We searched the Medline, Science Citation Index, Academic Search Premier, Cochrane Library, and Embase electronic data bases. Data extraction and quality assessment were done independently by two reviewers and results were pooled using the random effects model. Cochran's Q was used to identify heterogeneity and a funnel plot was used for assessment of publication bias. A meta-analysis was performed on 38 studies (265 330 patients) reporting on serum proteins, inflammatory markers, and mortality. A significant inverse relation was found between serum albumin and all cause (hazard ratio [HR] 0.7038, 95% confidence interval [CI] 0.6367-0.7781) and cardiovascular (HR 0.8726, 95% CI 0.7909-0.9628) mortalities, with a significantly stronger relation with all-cause mortality (P ? 0.0014). Pooled results for C-reactive protein showed a weak but significant direct relation with all-cause mortality (HR 1.0322, 95% CI 1.0151-1.0496), but there was not a significant relation between C-reactive protein and cardiovascular mortality (HR 1.0172, 95% CI 0.9726- 1.0639). A high degree of heterogeneity was identified among studies especially in the case of all cause mortality. An asymmetrical funnel plot for serum albumin is suggestive of publication bias. From the meta-analysis it is concluded that serum albumin showed a significant inverse relation with all-cause and cardiovascular mortalities but the relation between prealbumin and all-cause mortality was not significant. C-reactive protein showed a significant direct relation with all-cause mortality but not with cardiovascular mortality. The potential adverse effects of malnutrition and infections in relation to mortality highlight the need for continued treatment of infections and correction of malnutrition in patients on dialysis.