Study Objective: To determine compound A concentrations in a low-flow circuit containing Dragersorb Free(R) (Drager, Lubeck, Germany), Amsorb(R) (Armstrong, Coleraine, Northern Ireland), and Sodasorb II(R) (W. R. Grace, Lexington, MA). Design: Randomized study. Setting: Hamamatsu University Hospital. Patients: 24 ASA physical status I and II patients scheduled for general anesthesia greater than 3 hours' duration. Interventions: Patients were allocated to three groups of eight patients each to receive either using either Dragersorb Free(R), Amsorb(R), or Sodasorb II(R). Immediately before anesthesia induction, 1 kg of fresh absorbent was placed in the anesthesia canister. Anesthesia was maintained with sevoflurane (end-tidal concentration 1% to 3%) in oxygen and nitrous oxide (FIO2 > 0.3) at a total flow of 1 L/min. Measurements: Inspiratary compound A concentration in the circuit was measured once every hour. Main Results: Maximum compound A concentrations for Dragersorb Free(R), Amsorb(R), and Sodasorb II(R) were 2.4 +/- 0.8 (mean +/- SD) ppm, 3.1 +/- 0.5 ppm, and 28.0 +/- 10.0 ppm (p < 0.01 vs. Drdgersorb (R) and Amsorb(R)). Concentrations with Drdgersorb Free(R) and Amsorb(R) remained at less than 4 ppm throughout the study. Conclusions: Because compound A concentrations in the circuit with Drdgersorb Free(R) and Amsorb(R) were negligible, sevoflurane can be used at afresh gas flow of 1 L/min with these two absorbents. (C) 2003 by Elsevier Science Inc.