Should data monitoring committees assess efficacy when considering safety in trials in acute stroke?