People with fragile X syndrome (FXS) have an intellectual dysfunction that can range from very mild to severe. Symptoms can include speech and language delays and behavioural difficulties such as aggression or self injurious behaviours, emotional lability, and anxiety-related problems (for example obsessive-compulsive symptoms and perseverative behaviours). In some cases, affected people may have an additional diagnosis of attention deficit hyperactivity disorder or an autism spectrum disorder. To review the efficacy and safety of L-acetylcarnitine in improving the psychological, intellectual, and social performance of people with FXS. In May 2015 we searched the Cochrane Central Register of Controlled Trials (CENTRAL), MEDLINE, Embase, PsycINFO, Web of Science, and two other databases. We also searched three trials registers, four theses databases, and the reference lists of relevant studies and reviews. Randomised controlled trials (RCTs) that assessed the efficacy of L-acetylcarnitine, at any dose, in people of any age diagnosed with FXS compared with placebo. For each trial, two review authors independently extracted data on the children included and interventions compared, and assessed the risk of bias of the studies across the following domains: randomisation sequence generation, allocation concealment, blinding (of participants, personnel, and outcome assessors), incomplete outcome data, selective outcome reporting, and other potential sources of bias. We found only two RCTs that compared oral L-acetylcarnitine (LAC) with oral placebo in children with FXS. The studies included a total of 83 participants, all of them male, who were treated and followed for one year. The age of participants at the start of treatment ranged from 6 to 13 years, with a mean age of 9 years. Neither study provided information on randomisation, allocation concealment procedures, or blinding of outcome assessment, and we received no responses from the authors we emailed for clarification. We therefore rated studies as being at unclear risk of bias on these domains. We judged both studies to be at low risk of bias for blinding of participants and personnel, incomplete outcome data, and selective reporting, but to be at high risk of other bias, as at least one study was funded by a drug company, and in both studies people working for the company were part of the research team.We used the Grading of Recommendations Assessment, Development and Evaluation (GRADE) approach to rate the quality of the available evidence. Overall, the quality of the evidence was low due to the imprecision of results and high risk of other bias.Regarding the primary outcome of psychological and learning capabilities, both studies assessed the effect of interventions on children's verbal and non-verbal intellectual functioning using the Wechsler Intelligence Scale for Children - Revised. The authors did not provide detailed data on those results but said that they found no important differences between treatment and placebo.Both studies evaluated the impact of the treatment on hyperactive behaviour using the Conners' Abbreviated Parent-Teacher Questionnaire. In one study, teachers' assessments of the children found no clear evidence of a difference (mean difference (MD) 0.50, 95% confidence interval (CI) -5.08 to 6.08, n = 51; low-quality evidence). The other study stated that there were no differences between treated and untreated participants, but did not provide detailed data for inclusion in the meta-analysis.Parents' assessments favoured LAC in one study (MD -0.57, 95% CI -0.94 to -0.19, n = 17; low-quality evidence), but not in the other (MD -2.80, 95% CI -7.61 to 2.01, n = 51; low-quality evidence), though changes were not large enough to be considered clinically relevant.Regarding social skills, one study reported no clear evidence of a difference in Vineland Adaptive Behavior composite scores (MD 8.20, 95% CI -0.02 to 16.42, n = 51; low-quality evidence), yet results in the socialisation domain favoured LAC (MD 11.30, 95% CI 2.52 to 20.08, n = 51; low-quality evidence).Both studies assessed the safety of the active treatment and recorded no side effects. Neither of the included studies assessed the secondary outcome of caregiver burden. Low-quality evidence from two small trials showed that when compared to placebo, LAC may not improve intellectual functioning or hyperactive behaviour in children with FXS.