BACKGROUND: Noisy breathing (death rattle) occurs in 23 to 92% of people who are dying. The cause of death rattle remains unproven but is presumed to be due to an accumulation of secretions in the airways. It is therefore managed physically (repositioning and clearing the upper airways of fluid with a mechanical sucker) or pharmacologically (with anticholinergic drugs). OBJECTIVES: To describe and assess the evidence for the effectiveness of interventions used to treat death rattle in patients close to death. SEARCH STRATEGY: Randomised controlled trials (RCTs), before and after studies and interrupted time series (ITS) studies in adults and children with death rattle were sought by MEDLINE (1966 to 2007), EMBASE (1980 to 2007), CINAHL (1980 to 2007), the Cochrane Pain, Palliative and Supportive Care Trials Register and the Cochrane Central Register of Controlled Trials. In addition, the reference lists of all relevant trials and reports were checked and investigators who are known to be researching this area were contacted for unpublished data or knowledge of the grey literature. SELECTION CRITERIA: RCTs, controlled before and after studies and ITS reporting the outcome of pharmacological and non-pharmacological interventions for treating death rattle. DATA COLLECTION AND ANALYSIS: Data was extracted by two independent review authors and trials were quality scored. There was insufficient data to carry out an analysis. MAIN RESULTS: Thirty studies were identified, of which only one study met the inclusion criteria. This small study was a randomised placebo-controlled trial of the use of hyoscine hydrobromide in patients with death rattle. Hyoscine hydrobromide tended to reduce death rattle compared to placebo but this was not significant. A larger randomised study, comparing atropine, hyoscine butylbromide and scopolamine, is in progress. AUTHORS’ CONCLUSIONS: There is currently no evidence to show that any intervention, be it pharmacological or non-pharmacological, is superior to placebo in the treatment of death rattle. We acknowledge that in the face of heightened emotions when death is imminent, it is difficult for staff not to intervene. It is therefore likely that the current therapeutic options will continue to be used. However, patients need to be closely monitored for lack of therapeutic benefit and adverse effects while relatives need time, explanation and reassurance to relieve their fears and concerns. There is a need for more well-designed multi-centre studies with objective outcome measures and the ability to recruit sufficient numbers.