Background: Friedreich ataxia is a rare inherited autosomal recessive neurological disorder, characterised initially by unsteadiness in standing and walking, slowly progressing to wheelchair dependency usually in the late teens or early twenties. It is associated with slurred speech, scoliosis, and pes cavus. Heart abnormalities cause premature death in 60% of people with the disorder. There is no easily defined clinical or biochemical marker and no known treatment. This is the second update of a review first published in 2009 and previously updated in 2012. Objectives: To assess the effects of pharmacological treatments for Friedreich ataxia. Search methods: On 29 February 2016 we searched The Cochrane Neuromuscular Specialised Register, CENTRAL, MEDLINE, EMBASE and CINAHL Plus. On 7 March 2016 we searched ORPHANET and TRIP. We also checked clinical trials registers for ongoing studies. Selection criteria: We considered randomised controlled trials (RCTs) or quasi-RCTs of pharmacological treatments (including vitamins) in people with genetically-confirmed Friedreich ataxia. The primary outcome was change in a validated Friedreich ataxia neurological score after 12 months. Secondary outcomes were changes in cardiac status as measured by magnetic resonance imaging or echocardiography, quality of life, mild and serious adverse events, and survival. We excluded trials of duration shorter than 12 months. Data collection and analysis: Three review authors selected trials and two review authors extracted data. We obtained missing data from the two RCTs that met our inclusion criteria. We collected adverse event data from included studies. We used standard methodological procedures expected by Cochrane. Main results: We identified more than 12 studies that used antioxidants in the treatment of Friedreich ataxia, but only two small RCTs, with a combined total of 72 participants, both fulfilled the selection criteria for this review and published results. One of these trials compared idebenone with placebo, the other compared high-dose versus low-dose coenzyme Q10 and vitamin E (the trialists considered the low-dose medication to be the placebo). We identified two other completed RCTs, which remain unpublished; the interventions in these trials were pioglitazone (40 participants) and idebenone (232 participants). Other RCTs were of insufficient duration for inclusion. In the included studies, the primary outcome specified for the review, change in a validated Friedreich ataxia rating score, was measured using the International Co-operative Ataxia Rating Scale (ICARS). The results did not reveal any significant difference between the antioxidant-treated and the placebo groups (mean difference 0.79 points, 95% confidence interval -1.97 to 3.55 points; low-quality evidence). The published included studies did not assess the first secondary outcome, change in cardiac status as measured by magnetic resonance imaging. Both studies reported changes in cardiac measurements assessed by echocardiogram. The ejection fraction was not measured in the larger of the included studies (44 participants). In the smaller study (28 participants), it was normal at baseline and did not change with treatment. End-diastolic interventricular septal thickness showed a small decrease in the smaller of the two included studies. In the larger included study, there was no decrease, showing significant heterogeneity in the study results; our overall assessment of the quality of evidence for this outcome was very low. Left ventricular mass (LVM) was only available for the smaller RCT, which showed a significant decrease. The relevance of this change is unclear and the quality of evidence low. There were no deaths related to the treatment with antioxidants. We considered the published included studies at low risk of bias in six of seven domains assessed. One unpublished included RCT, a year-long study using idebenone (232 participants), published an interim report in May 2010 stating that the study reached neither its primary endpoint, which was change in the ICARS score, nor a key cardiological secondary endpoint, but data were not available for verification and analysis. Authors' conclusions: Low-quality evidence from two small, published, randomised controlled trials neither support nor refute an effect from antioxidants (idebenone, or a combination of coenzyme Q10 and vitamin E) on the neurological status of people with Friedreich ataxia, measured with a validated neurological rating scale. A large unpublished study of idebenone that reportedly failed to meet neurological or key cardiological endpoints, and a trial of pioglitazone remain unpublished, but on publication will very likely influence quality assessments and conclusions. A single study of idebenone provided low-quality evidence for a decrease in LVM, which is of uncertain clinical significance but of potential importance that needs to be clarified. According to low-quality evidence, serious and non-serious adverse events were rare in both antioxidant and placebo groups. No non-antioxidant agents have been investigated in RCTs of 12 months' duration.