It has been argued that the oedema of kwashiorkor is not caused by hypoalbuminaemia because the oedema disappears with dietary treatment before the plasma albumin concentration rises. Reanalysis of this evidence and a review of the literature demonstrates that this was a mistaken conclusion and that the oedema is linked to hypoalbuminaemia. This misconception has influenced the recommendations for treating children with severe acute malnutrition. There are close pathophysiological parallels between kwashiorkor and Finnish congenital nephrotic syndrome (CNS) pre-nephrectomy; both develop protein-energy malnutrition and hypoalbuminaemia, which predisposes them to intravascular hypovolaemia with consequent sodium and water retention, and makes them highly vulnerable to develop hypovolaemic shock with diarrhoea. In CNS this is successfully treated with intravenous albumin boluses. By contrast, the WHO advise the cautious administration of hypotonic intravenous fluids in kwashiorkor with shock, which has about a 50% mortality. It is time to trial intravenous bolus albumin for the treatment of children with kwashiorkor and shock.