Although radiosurgery treatments began more than 30 years ago and thousands of patients have been treated, the role of radiosurgery in the treatment of intracranial lesions is continuing to evolve. For AVM several groups have reported favorable angiographic response rates. Further information, however, is needed to establish the relationship between this objective measure of response and the risk of subsequent hemorrhage. Ideally this relationship might be established through randomized trials. For ethical reasons, however, such trials are unlikely to be undertaken. Although radiosurgery may finally prove useful in the treatment of lesions other than AVM, currently available retrospective data do not allow firm conclusions in most cases. Various techniques can be used to deposit the dose accurately in a predetermined intracranial volume. Nonetheless, it frequently is difficult to define the three-dimensional configuration of the volume requiring treatment accurately. This difficulty is independent of the radiosurgical technique used and, together with dose prescription and patient selection, is probably of greater importance than possible differences in dose distribution produced by different systems. Because many patients will have radiosurgery in nontrial clinical settings, it is important that future reports provide sufficient information to allow meaningful interpretation. At minimum, reports must contain an accounting of all patients treated, including number followed and number lost to follow-up evaluation or otherwise lacking repeat radiologic studies or clinical evaluation. (The statement that 1000 patients were treated is not useful in the absence of this accounting and may cast doubt on reported results.) Beyond accounting it is helpful if reports provide (1) information on dimensions and intracranial location of lesions treated and correlation with dose and outcome, (2) quantitative information on dose prescribed, including minimum and maximum dose to the target, especially when multiple isocenters are used (when isodose contours from individual isocenters must be summed), (3) detailed information on patients developing complications (in what ways did their presentation or treatment differ from that of other patients?), and (4) thorough data analysis using modern statistical techniques. The number of radiosurgery facilities is increasing rapidly. Whether the number of patients treated in each facility will continue to rise each year as more new facilities are open is conjectural and will depend on the evolution of indications for radiosurgery and on the evolution of referral patterns as techniques, facilities, and results become established.
