Malnutrition has been described in patients with chronic kidney disease as well as its association with cardiovascular risk and mortality in haemodialysis patients. Recently, the new term "protein energy wasting" has been proposed with new diagnostic criteria (biochemical and anthropometric markers) for early identification of patients at risk for protein energy wasting and mortality. The aim of this study was to examine the prevalence, evolution over time and prognostic significance of PEW in a Spanish dialysis centre for the first time in Spain. an observational study that included 122 prevalent haemodialysis patients at our centre. Between January 2010 and October 2012, three visits were carried out in which clinical, biochemical, anthropometric and body composition parameters were collected using BIS (bioelectrical impedance spectroscopy) along with their respective dialytic characteristics, in accordance with the criteria of the new definition. We analysed the prevalence of PEW in each visit, progression of the malnutrition parameters and factors potentially associated with PEW. After a mean follow-up period of 461 days, we analysed survival. Statistical analysis was performed using the R software. The prevalence of PEW remained constant over time: 37% at baseline visit, 40.5% at 12 months and 41.1% at 24 months. With the introduction of the dynamic variable muscle mass loss, included in the definition of PEW, prevalence increased to 50% at 24 months. The PEW situation is dynamic, as demonstrated by the fact that 26%-36% of patients without PEW develop it de novo each year and 12%-30% annually recover from this situation. The presence of PEW was associated with higher rates of resistance to erythropoietin (irEPO) and higher pulse pressure at the end of dialysis. In the multivariable regression model, PEW predictive clinical variables were over-hydration, irEPO, intracellular water and the extracellular water/intracellular water ratio. Twenty-six (21%) patients died. The Kaplan-Meier curve did not show any differences in mortality risk between patients with and those without PEW, but the loss of muscle mass was associated with increased mortality. The present observational study highlights the high prevalence of PEW, which has a dynamic nature in haemodialysis patients. Only the criterion of muscle mass loss (increased protein catabolism) was associated with increased mortality, while the other PEW criteria according to the ISRNM classification were not associated with increased mortality. We also observed a state of over-hydration in patients with PEW. This state of over-hydration (increased extracellular water due to occupation of muscle loss without an increase in total body water) cannot be evaluated by dry weight or the body mass index. Intervention studies are necessary in order to assess whether or not the prevention of sarcopaenia improves survival.