BACKGROUND The underlying clinical condition and the Intensive Care Unit (ICU) environment make critical illness a stressful event. Although the usual management consists of sedation, non-pharmacological interventions such as music therapy have been suggested for their drug-sparing effect. Aim of the present review is to assess the current evidence on the effectiveness of music therapy in reducing stress and anxiety in critically ill, adult patients. METHODS A systematic review of publications was undertaken using MEDLINE, CINAHL, Cochrane Library, Scopus, Web of Science, Indice Italiano di Letteratura di Scienze Infermieristiche. We included studies of critically ill patients that assessed any effect of music therapy on stress and anxiety, which were variably assessed according to each study's definition. RESULTS Eleven studies were included (10 RCTs and one quasi-experimental design), for a total of 959 patients (range 17-373). The overall quality of the studies was satisfactory; several potential sources for bias were identified. Music therapy was generally provided as a single, 30'-intervention, ranging from 15 to 60'. Only in two studies was the intervention repeated more than once daily. The control groups were standard care, relaxation, headphones with no music or noise-cancelling headphones. Music therapy determined a significant reduction in the levels of anxiety and stress, as assessed by self-reported scales and physiologic parameters. Pooled analysis was not performed due to the heterogeneity of the interventions. CONCLUSIONS Despite significant heterogeneity in trial designs, timing and features of the intervention, music therapy is consistently associated with a reduction in anxiety and stress of critically ill patients.