For many years patients with chronic kidney disease have been advised to control the protein content of their diet. This advice has been given on the basis of a number of reported metabolic effects of lowering protein intake, such as lowering serum urea nitrogen levels, improving phosphocalcic metabolism and insulin resistance and, more recently, ameliorating proteinuria (independent of antiproteinuric medications). The effects on the progression of kidney disease, although spectacular in experimental studies, have been less convincing in humans. It is possible that flawed design of clinical trials is responsible for this discrepancy. In this Review, we comment on experimental findings that indicate that limiting protein intake protects the kidney and ameliorates uremic symptoms, outline how the body adapts to a reduction in protein intake, and describe the metabolic benefits to the patient. We then review the evidence from randomized controlled trials and meta-analyses that pertains to the effects of low-protein diets in adults with chronic kidney disease.