Vivian Sarmento Vasconcelos, Cristiane R Macedo, Alexsandra de Souza Pedrosa, Edna Pereira Gomes Morais, Gustavo JM Porfírio, Maria R Torloni

An estimated 1% to 3% of all individuals will receive a diagnosis of epilepsy during their lives, which corresponds to approximately 50 million affected people worldwide. The real prevalence is possibly higher because epilepsy is underreported in developing countries. Although most will achieve adequate control of their disease though the use of medication, approximately 25% to 30% of all those with epilepsy are refractory to pharmacological treatment and will continue to have seizures despite the use of two or more agents in adequate dosages. Over the last decade, researchers have tested the use of polyunsaturated fatty acid (PUFA) supplements for the treatment of refractory epilepsy, with inconsistent results. There have also been some concerns about the use of omega-3 PUFA compounds because they reduce platelet aggregation and could, in theory, cause bleeding. To assess the effectiveness and tolerability of omega-3 polyunsaturated fatty acids (eicosapentaenoic acid-EPA and docosahexanoic acid-DHA) in the control of seizures in people with refractory epilepsy. We searched the Cochrane Epilepsy Group Specialised Register (from inception up to November 2015), the Cochrane Central Register of Controlled Trials (CENTRAL) (2015, issue 11), MEDLINE (1948 to November 2015), EMBASE (1980 to November 2015), SCOPUS (1823 to November 2015); LILACS (Literatura Latino-Americana e do Caribe de Informação em Ciências da Saúde) (1982 to November 2015); ClinicalTrials.gov; World Health Organization (WHO) International Clinical Trials Registry Platform (November 2015). No language restrictions were imposed. We contacted study authors for additional and unpublished information and screened the reference lists of retrieved citations for potentially eligible studies not identified through the electronic search. All randomised and quasi-randomised studies using PUFAs for the treatment of drug-resistant epilepsy. Two review authors were involved in study selection, data extraction and quality assessment of the included trials. The following outcomes were assessed: seizure freedom, seizure reduction, improvement in quality of life, potential adverse effects, gastrointestinal effects, drop-out rates and changes in plasma lipid profile. Primary analyses were by intention to treat. Eight studies were identified as potentially relevant; three fulfilled the selection criteria and were included in the review. Two placebo-controlled, double blind trials involving adult participants were conducted in developed countries, while one placebo-controlled, single blind trial involving children was conducted in a developing country (Egypt). Bromfield 2008 randomised 27 American adults to receive 2.2 g/day of omega-3 PUFAs (EPA:DHA in a 3:2 ratio) or placebo. Yuen 2005 randomised 58 people in the UK to approximately 1.7 g/day omega-3 PUFAs (1g EPA and 0.7g DHA) or placebo. Reda 2015 randomised 70 Egyptian children to receive 3 ml/day of 1200 mg fish oil (providing 0.24 g DHA and 0.36 g EPA) or placebo. The three studies recruited a total of 155 subjects (85 adults and 70 children); 78 of them (43 adults and 35 children) were randomised to PUFAs and 77 (42 adults and 35 children) to placebo. All participants were followed for up to 12 weeks. Seizure freedom was reported by only one study, with a high risk of bias, involving exclusively children. The risk estimate for this outcome was significantly higher in the children receiving PUFA compared to the control group (risk ratio (RR) 20.00, 95% confidence interval (CI) 2.84 to 140.99, 1 study, 70 children). Similarly, PUFA supplementation was associated with a significant difference in the proportion of children with at least 50% reduction in seizure frequency (RR 33.00 95% CI 4.77 to 228.15, 1 study with a high risk of bias, 70 children). However, this effect was not observed when the data from two studies including adult participants were pooled (RR 0.57, 95% CI 0.19 to 1.75, I² 0%, 2 studies, 78 participants, low-quality evidence). One of our three primary outcomes (adverse effects related to bleeding) was not assessed in any of the studies included in this review. There were no significant differences between the PUFA and control groups in relation to gastrointestinal effects (RR 0.78, 95% CI 0.32 to 1.89, 2 studies, 85 participants, low-quality evidence).Supplementation with PUFA did not produce significant differences in mean frequency of seizures, quality of life or other side effects. In view of the limited number of studies and small sample sizes, there is not enough evidence to support the use of PUFA supplementation in people with refractory epilepsy. More trials are needed to assess the benefits of PUFA supplementation in the treatment of drug-resistant epilepsy.