ABSTRACT
To study the effect of general anesthesia with diazepam, fentanyl and nitric oxide, a common combination during the intraoperative recording of visual evoked potentials (VEP) in transsphenoidal surgery, we compared the amplitude and latency of VEP components before anesthesia and at four moments after induction during the first hour of elective breast surgery in 20 patients with no neurological deficits. The results for these patients (group I) before anesthesia and 15 min after induction were also compared to presurgical recordings for 19 patients with hypophysial tumors and histories visual field and acuity involvement (group II). Latency increased significantly as a result of anesthesia whereas amplitude was affected to a lesser degree. Changes in latency of the main positive component was after anesthesia was the only parameter that was significantly different for the two group studied, variation being greater in group II.