Testing devices for the prevention and treatment of stroke and its complications.

We are entering a challenging but exciting period when many new interventions may appear for stroke based on the use of devices. Hopefully these will lead to improved outcomes at a cost that can be afforded in most parts of the world. Nevertheless, it is vital that lessons are learnt from failures in the development of pharmacological interventions (and from some early device studies), including inadequate preclinical testing, suboptimal trial design and analysis, and underpowered studies. The device industry is far more disparate than that seen for pharmaceuticals; companies are very variable in size and experience in stroke, and are developing interventions across a wide range of stroke treatment and prevention. It is vital that companies work together where sales and marketing are not involved, including in understanding basic stroke mechanisms, prospective systematic reviews, and education of physicians. Where possible, industry and academics should also work closely together to ensure trials are designed to be relevant to patient care and outcomes. Additionally, regulation of the device industry lags behind that for pharmaceuticals, and it is critical that new interventions are shown to be safe and effective rather than just feasible. Phase IV postmarketing surveillance studies will also be needed to ensure that devices are safe when used in the 'real-world' and to pick up uncommon adverse events.