Malnutrition, defined as insufficient protein-calorie intake, is highly prevalent in haemodialysis (HD) patients. It is commonly associated with decreased body weight, depleted energy stores (fat tissue) and loss of somatic proteins. It has been suggested that there may be at least two fundamentally different types of malnutrition in HD patients. The first is related to low protein and energy intake. This type of malnutrition may be amenable to adequate nutritional and dialysis support. In contrast, the second type of malnutrition is associated with inflammation and atherosclerotic cardiovascular disease (MIA syndrome). This type of malnutrition is much more difficult to reverse with nutritional support and dialysis therapy, unless the underlying co-morbid conditions and chronic inflammatory response are treated adequately. There is no single measurement which can be used to determine the presence of malnutrition. Therefore, a panel of measurements is recommended, including a measure of body composition, a measure of dietary protein intake and at least one measure of serum protein status. Therapeutic interventions in malnutrition of HD patients include dietary prescription, correction of metabolic acidosis, therapeutic intervention for co-morbidities, an adequate dose of dialysis and finally interventions on dialysis techniques, membrane and dialysate quality. It is a widespread belief that convective treatments give a clinical advantage over standard diffusive HD, when considering the physiological outcomes, such as haemodynamic instability. Since convective treatments are usually performed utilizing synthetic biocompatible membranes, it is very difficult to separate the effects of convection from those of biocompatibility on a given outcome, e.g. the nutritional status of HD patients. There are, to date, no prospective randomized clinical trials with sufficient power in which the effects of convective therapies on the nutritional status of HD patients were compared with those of other dialysis modalities. Morbidity and mortality of HD patients are strictly associated, but not necessarily linked by a cause-effect relationship to some nutritional parameters, notably serum albumin levels. Thus, as the presence of malnutrition is one of the strongest predictors of mortality, the crucial point is to demonstrate the superiority of convective treatments over standard diffusive HD on hard outcomes: morbidity and mortality. Even though almost all available studies support the hypothesis that convection and high-flux biocompatible membranes are associated with reduced morbidity and mortality risks, there is, to date, no absolute proof showing a cause-effect relationship between convection and high-flux biocompatible membranes and the outcome of HD patients. In conclusion, convective therapies offer the opportunity for as yet unrivalled small and/or middle molecular clearance. Regarding short-term complications, there is ample evidence for the fact that haemodynamic stability is better maintained during haemofiltration than during standard HD. Although a beneficial effect of convective therapies was suggested mainly in uncontrolled studies, as far as the nutritional status, the morbidity and the mortality of HD patients are concerned, there appears to be a shortage of well-controlled prospective randomized clinical trials with sufficient power assessing the potential merits of convection.