OBJECTIVES: To evaluate the effects of short-term intravenous magnesium on the length of hospital stay and quality of life in children and adults with sickle cell disease. To determine the effects of long-term oral magnesium therapy on the frequency of painful crises and the quality of life in children and adults with sickle cell disease.

SEARCH METHODS: We searched the Cochrane Haemoglobinopathies Trials Register, compiled from electronic database searches and handsearching of journals and conference abstract books.Date of last search of the Cochrane Cystic Fibrosis and Genetic Disorders Group's Haemoglobinopathies Trials Register: 01 December 2016.Date of last search of other resources (clinical trials registries): 29 March 2017.

SELECTION CRITERIA: We searched for published and unpublished randomized controlled studies of oral or intravenous magnesium compared to placebo or no magnesium.

DATA COLLECTION AND ANALYSIS: Authors independently assessed the study quality and extracted the data using standard Cochrane methodologies.

MAIN RESULTS: We included five randomized placebo-controlled studies with a total of 386 participants (aged three to 53 years). Two shorter parallel studies (n = 306) compared intravenous magnesium sulphate to placebo (normal saline) for admission to hospital due to a vaso-occlusive crisis, for which we were able to analyse data. The quality of evidence was moderate for studies presenting this comparison mainly due to limitations due to risk of bias and imprecision. Two of the three longer-term studies comparing oral magnesium pidolate to placebo had a cross-over design. The third was a parallel factorial study which compared hydroxyurea and oral magnesium to each other and to placebo over a longer period of time; we only present the comparison of oral magnesium to placebo from this study. The quality of evidence was very low with uncertainty of the estimation.The eight-hourly dose levels in the two studies of intravenous magnesium were different; one used 100 mg/kg while the second used 40 mg/kg. Only one of these studies (n = 104) reported the mean daily pain score while hospitalised (a non-significant difference between groups, moderate quality evidence). The second study (n = 202) reported a number of child- and parent-reported quality of life scores. None of the scores showed any difference between treatment groups (low quality evidence). Data from one study (n = 106) showed no difference in length of stay in hospital between groups (low quality evidence). Both studies reported on adverse events, but not defined by severity as we had planned. One study showed significantly more participants receiving intravenous magnesium experienced warmth at infusion site compared to placebo; there were no differences between groups for other adverse events (low quality evidence).Three studies (n = 80) compared oral magnesium pidolate to placebo. None of them reported data which we were able to analyse. One study (n = 24) reported on the number of painful days and stated there was no difference between two groups (low quality evidence). None of the studies reported on quality of life or length of hospital stay. Two studies (n = 68) reported there were no differences in levels of magnesium in either plasma or red blood cells (moderate quality evidence). Two studies (n = 56) reported adverse events. One reported episodes of mild diarrhoea and headache, all of which resolved without stopping treatment. The second study reported adverse events as gastrointestinal disorders, headache or migraine, upper respiratory infections and rash; which were all evenly distributed across treatment groups (moderate quality evidence).