This issue of the Journal of Neurosurgery includes an excellent paper by Bervini et al. from Professor Morgan’s service at Macquarie University in Sydney, Australia. 2 This well-respected group has one of the largest experiences in the world with cerebral arteriovenous malformations (AVMs). In this study, the authors sought to ascertain the natural history of unruptured brain AVMs with regard to risk of rupture as well as the morbidity and mortality associated with resection. The authors retrospectively identified all patients with unruptured brain AVMs who were referred to their institution from a prospectively maintained AVM database consisting of both treated and untreated lesions. Over a 24-year period, a total of 427 patients with unruptured cerebral AVMs were identified. All patients with more than 1 day of follow-up (n= 377) were analyzed for risk of hemorrhage, irrespective of the management plan, from the date of initial consultation or referral to the date of last available follow-up or until the initiation of treatment. Radiologically confirmed hemorrhage unrelated to treatment occurred in 16 patients during a total of 279 case-years of follow-up. Permanent disability following hemorrhage resulted in a modified Rankin Scale (mRS) score> 1 in 88%(14 cases), mRS score> 2 in 69%, and death in 31% of the patients. The cumulative rate of hemorrhage using Kaplan-Meier analysis at 1 and 5 years was 8.1% and 11.5%, respectively. The annualized risk of hemorrhage over the first 5 years was 2.3%. The only variable associated with a significantly shorter time to hemorrhage was deep location by univariate analysis. The risk from resection was assessed in 2 ways. First, the authors analyzed the surgical outcomes in patients who actually underwent resection based on AVM grade. The authors classified AVMs according to Spetzler-Ponce Class where Class A comprised Spetzler-Martin Grade I and II lesions, Class B comprised Spetzler-Martin Grade III lesions, and Class C comprised Spetzler-Martin Grade IV and V lesions. Complications were defined as any permanent neurological deficits following surgery, including those related to preoperative embolization, that resulted in an mRS score> 1 at last follow-up. The authors then performed an interesting sensitivity analysis to estimate the true risk of surgery for all AVM grades and not only those selected based on a perceived favorable risk-benefit ratio. The authors included patients who were eligible for surgery but did not undergo resection because of perceived excess surgical morbidity with those who actually underwent resection. Patients were considered eligible for surgery if they were younger than 65 years and did not have significant comorbidities. These added unoperated cases were assumed to have an adverse outcome for the purpose of the sensitivity analysis. For patients with Spetzler-Ponce Class A lesions treated by surgery (n= 190), the risk of a new permanent deficit resulting in an mRS score> 1 or mRS score> 2 was 1.6% and 0.5%, respectively. None of the eligible patients with Spetzler-Ponce Class A lesions were excluded because of perceived excessive surgical risk and thus a sensitivity analysis was not performed. For patients with Spetzler-Ponce Class B lesions treated by surgery (n= 107), the risk of a new permanent deficit resulting in an mRS score> 1 was 14% and mRS score> 2 was 2.8%. Two eligible patients with Spetzler-Ponce Class B lesions were excluded because of perceived excessive surgical risk. The sensitivity analysis showed that if these cases had been included the risk of a new permanent neurological deficit resulting in an mRS score> 1 or mRS score> 2 would have …