BACKGROUND During automated closed-circuit anesthesia (CCA), the Zeus (Dräger, Lübeck, Germany) uses a high initial fresh gas flow (FGF) to rapidly attain the desired agent and carrier gas concentrations, resulting in a desflurane consumption well above patient uptake. Because both FGF and carrier gas composition can affect consumption, we determined the Zeus' agent consumption with automated CCA and with automated low flow anesthesia (LFA) (= maintenance FGF of 0.7 L min(-1)) with 3 different carrier gases. METHODS After IRB approval, 65 ASA PS I or II patients undergoing general surgery received desflurane in either O2, O2/air, or O2/N2O, with the Zeus to maintain the end-expired concentration (FA) at 6, 6, and 4% and the F1O2 at 1.0, 0.6, and 0.4, respectively. In addition, patients were assigned to either automated CCA (O2 n = 11; O2/air n = 11; O2/N2O n = 11) or automated LFA (selected FGF 0.7 L min(-1)) (O2 n = 12; O2/air n = 11; O2/N2O n = 9). Demographics and desflurane consumption at 2, 4, 6, 8, 10, 20, 30, 40 and 50 min were compared. RESULTS With the same carrier gas, desflurane consumption was lower with the CCA mode than with LFA mode after 4 min in the O2 groups, 6 min in the O2/air groups, and 30 min in the O2/N2O groups. Within each mode, desflurane consumption in the O2 and O2/air groups was identical at all times. Despite the use of a lower FA in the N2O groups, initial desflurane consumption was higher than in the O2 and O2/air groups, but it was lower later (> or = 15 min) only with LFA. CONCLUSIONS After 50 min, desflurane consumption with automated CCA is lower than with automated LFA. However, initial agent consumption is complex, and N2O in particular may increase initial desflurane consumption (though ultimately resulting in lower desflurane usage because of its MAC sparing effect) because initial FGF is increased to rapidly reach the target concentrations. Differences in desflurane consumption only become apparent after FGF has stabilized to the target FGF.