OBJECTIVE To determine if the previously described single-step O(2)/N(2)O fresh gas flow (FGF) sequence could be combined with a simple desflurane vaporizer (F(D)) sequence to maintain the end-expired desflurane (F(A)des) at 4.5% with the anesthesia delivery unit machine (ADU Anesthesia Machine(R); General Electric, Helsinki, Finland). METHODS Prospective randomized clinical study. METHODS Onze Lieve Vrouw Hospital, Aalst, Belgium, a large teaching hospital. METHODS 42 ASA physical status I and II patients requiring general endotracheal anesthesia and controlled mechanical ventilation. METHODS In 18 patients undergoing general anesthesia with controlled mechanical ventilation, F(D) was determined to maintain F(A)des at 4.5% with O(2)/N(2)O FGF of two and 4 L per minute for three minutes and 0.3 and 0.4 L per minute thereafter. Using the same FGF sequence, we prospectively tested the F(D) schedule that approached this observed F(D) pattern with the fewest possible adjustments in another 24 patients. RESULTS F(D) of 6.5% for 15 minutes followed by 5.5% thereafter approximated the observed F(D) course well. When it was prospectively tested, the median (25th, 75th percentiles) performance error was -1% (-5.1%, 5.2%); absolute performance error, 7.1% (3.9%, 9.5%); divergence, -6.6% per hour (23.1%, 3.1%/h); and wobble, 2.2% (1.8%, 3.2%). Because F(A)des increased above 4.9%, F(D) was decreased in 5 patients after 23 minutes (0.5% decrement once or twice); in two patients, F(D) was temporarily increased. In one patient, FGF was temporarily increased because the bellows volume became insufficient. CONCLUSIONS One O(2)/N(2)O rotameter FGF setting change from 6 to 0.7 L per minute after three minutes and one desflurane F(D) change from 6.5% to 5.5% after 15 minutes maintained anesthetic gas concentrations within predictable and clinically acceptable limits during the first 20 minutes.