Effect of Acetyl-L-carnitine in Migraine

The study will be a single-centre double-blinded, randomized, placebo controlled crossover study with Acetyl-L-carnitine or placebo. We plan to include 72 patients. The follow up duration will be 36 weeks. The study will be performed according to Good Clinical Practice, and relying partly on the International Headache Society Guidelines for controlled trials of drugs in migraine from 2012 and partly on the guidelines divulged by the International Headache Society task force on trial guidelines for chronic migraine. After a screening visit including a neurological consultation, eligible patients will sign an informed consent declaration before they enter a 4 week run-in (baseline) period when they keep a headache diary. After 4 weeks they return for the second visit. If they have had 2 or more migraine attacks they are allowed to proceed in the study. If they have less than 2 migraine attacks (required for proceeding in the study), they are allowed to extend the baseline period another 4 weeks. Those who then during the whole 8 week period have on average 2 or more migraine attacks per month are also allowed to proceed. Otherwise they are excluded from the study. Details of the treatment period The duration of each of the two treatment periods is 12 weeks. During each period there will be one telephone contact at the start of each treatment period to remind patients to start with medicines, and one after 2 weeks to check compliance and side effects. In the second last week of every treatment period there will be a doctor and nurse visit with drug accounting and dispensing of new medicines for the next period. At this visit one will ensure that the patient has just enough medicines left to finish the period before the wash-out. As recommended in crossover studies, the participants enter a washout period of 4 weeks between the two treatment periods, to reduce the risk of carryover effect. Randomization Randomization will be generated using a computerized procedure. A randomization list containing 72 patient numbers is made before the start of the study, and the patient number is then indicated on a package with medicines for that patient. The study has a crossover design, and the two different treatment periods (active or placebo) can arise to two different treatment sequences (AP or PA). Patients are therefore randomized in blocks of 4 where one of these two treatment sequences is assigned to each patient in random order. With 72 patients to be included, this means that 18 patients are randomized in each block. In each block, 50% patients have the treatment sequence AP, and 50% PA in a random order.