Comparison of percutaneous losses of sevoflurane and isoflurane in humans.

We studied the percutaneous losses of sevoflurane and isoflurane during administration and elimination in seven healthy male volunteers. Anesthesia was induced and maintained with fentanyl, midazolam, and/or thiopental, and nitrous oxide for 30 min, after which 1% sevoflurane and 0.4% isoflurane in 65% nitrous oxide were administered for 30 min. Inspired, end-tidal, and mixed-expired gas samples were collected during administration and for 5-7 days of elimination. To measure percutaneous loss, each subject's arm was enclosed in a glass cylinder sealed at both ends and with two ports, one for flushing with nitrogen and one for obtaining gas samples during the 30 min of administration and the first 150 min of elimination. Anesthetic concentrations in all samples were determined using gas chromatography. The surface area of the arm was measured and the total surface area was calculated. During administration and elimination, percutaneous loss of isoflurane was significantly greater than that of sevoflurane (P less than 0.05). For both volatile agents, losses during elimination were greater than during administration (P less than 0.05), but even when combined, these losses were too small to affect kinetic or metabolic studies based on mass balance.