Onyx injection is a new technique for embolization of cerebral aneurysms that is involved in a controversy about the 'toxicity' of its solvent, dimethyl sulfoxide (DMSO). We retrospectively studied 38 patients treated for aneurysms with the liquid polymer, Onyx. Induction was with propofol, fentanyl and vecuronium, and anesthesia was maintained with isoflurane in O-2 and N2O. The patients were given 500 ml of fluid after induction, and bradycardia was prevented in order to keep patients hyperdynamic. Electrocardiography (ECG), non-invasive blood pressure (NIBP), pulse oximetry, core temperatures, invasive blood pressure (BP), etCO(2), and urine output were monitored throughout the intervention. Heart rate and BP changes in response to balloon inflation, DMSO injection, Onyx injection and balloon deflation were recorded. The patients were followed with serial neurological examinations, computerized tomography and/or magnetic resonance imaging postoperatively for evidence of any neurological injury. Cumulative DMSO doses were always well under previously implicated doses for systemic toxicity. No changes implicating toxic reactions were observed during DMSO and Onyx injections. Balloon-induced changes returned to baseline within 1 min of balloon deflation. Technique-related permanent morbidity occurred in two patients (worsening of cranial nerve palsies in one and monocular blindness in another) and intracranial hemorrhage with resulting death in one patient. All patients showed a tendency to oxygen desaturation, but this finding did not cause any clinical consequence. Anesthesiologists need to be vigilant in monitoring patients treated with techniques that are new or are being developed. We have seen no evidence of toxicity or any anesthetic complications in our group of patients, our only clinical concern being a tendency to oxygen desaturation, which may be explained by the inhalational elimination of DMSO.