Top ten research priorities for brain and spine cavernous malformations.

Cavernous malformations, also known as cavernomas, aff ect people at any age and occur throughout the CNS. In the brain, they can cause haemorrhagic stroke and epileptic seizures, and in the spinal cord, they can bleed and cause myelopathy. Despite the availability of microsurgical excision and stereotactic radiosurgery for cavernoma treatment and known genetic causes of most familial forms of cavernoma, uncertainties remain about cause, diagnosis, prognosis, treatment, and care. Therefore, to prioritise these uncertainties about brain and spine cavernomas for researchers and funding agencies, we undertook a James Lind Alliance priority setting partnership. This partnership included a multidisciplinary steering group of patients, carers, health-care professionals, representatives of patient support organisations, an information specialist, a James Lind Alliance adviser, and an administrator according to a protocol developed in August, 2014, and approved in January, 2015. The methods are described in full online . Between January and March 2015, we gathered uncertainties using a web-based survey that was distributed by professional and support organisations in the UK via email, post, and social media to patients, carers, and health-care professionals. We received 2268 uncertainties from 299 respondents (63% patients, 18% health-care professionals, and 19% others), and identifi ed a further 34 uncertainties from searches of the scientifi c literature. An information specialist subsequently de-duplicated these submissions, rejected submissions that were out of the scope of the priority setting partnership, rejected uncertainties if there was evidence in published systematic reviews that they had been answered, and added uncertainties identifi ed by these systematic reviews, resulting in a long list of 79 unique uncertainties. The steering group worked in pairs to further shorten the list to 54 uncertainties, which we circulated to 246 survey respondents who chose priorities from the long list of uncertainties. 136 (55%) respondents participated in the web-based prioritisation exercise, in which we used the rank order technique to generate a short list of 31 uncertainties. At a fi nal in-person workshop including 29 participants (41% patients, 31% healthcare professionals, and 28% others) and facilitated by three James Lind Alliance advisers, we achieved consensus on a fi nal list of 27 uncertainties (listed in the UK Database of Uncertainties about the Eff ects of Treatments [DUETs]), of which the top ten are immediate priorities for future research (panel). The top ten uncertainties refl ect the concerns of patients, carers, and health-care professionals in the UK: fi ve concerned prognosis, three concerned treatment or care, and two concerned cause. The James Lind Alliance process assures the internal validity and reliability of these priorities, but their generalisability to other populations is unknown. The 27 uncertainties identifi ed by this priority setting partnership, and in particular the top ten, can now inform the projects that the research community pursue and that funding bodies support in the UK and perhaps other parts of the world.