Aim: In this study, we aimed to compare the effects of desflurane-N2O and desflurane-fentanyl combinations on hemodynamics, recovery times, volatile anesthetic consumption and costs in low-flow desflurane anesthesia by bispectral index (BIS) monitoring of depth of anesthesia.,Methods: After approval of ethics committee and obtaining patient consents, 60 patients were divided into two equal groups randomly. Non-invasive blood pressure measurement, ECG, SpO2 and BIS were monitored. All patients received 10 L.min(-1) 100% oxygen with mask for 5 minute before intubation. 2 mg.kg(-1) propofol, 2 mu g.kg(-1) fentanyl and 0.6 mg.kg(-1) rocuronium bromide were administered at induction in both groups. Desfluran 6% was chosen for anesthesia maintenance. Group 1 received 50% O2-N2O mixture in 6 L.min(-1) and Group 2 received 50% O2-air mixture in 6 L.min(-1) as carrier gas. Low-flow anesthesia (1 L.min(-1)) was started after a 10-min period of initial high flow (6 L.min(-1)). In Group 2, infusion of fentanyl was begun in 1 mu g. kg.hour(-1) rate. Desflurane level was adjusted at a main BIS value of 40-60. Blood pressure, heart rate, FiO(2), etO(2), FiN2(2), EtN2O, FiCO(2), EtCO2, Fidesfluran and Etdesflurane were recorded.,Results: There were no significant differences between the two groups in terms of heart rate, arterial blood pressure, settings of desfluran and recovery time. BIS values (p<0.001) and anesthetic agent costs (p<0.001) were higher in Group 2.,Conclusion: Using fentanyl infusion instead of nitrous oxide in low flow-anesthesia with desflurane did not alter the hemodynamic parameters. Fentanyl infusion with medical air-oxygen as carrier gas is an alternative technique, but increases BIS values and anesthetic agent costs.