BACKGROUND: In studies of methoxyflurane-induced nephrotoxicity, renal-concentrating impairment has been observed only when serum inorganic fluoride concentrations exceed 50 microM. Prolonged sevoflurane anesthesia can result in serum inorganic fluoride concentrations in excess of 50 microM. The authors compared renal function after prolonged sevoflurane anesthesia with that after isoflurane anesthesia. In addition, they measured urinary excretion of N-acetyl-beta-glucosaminidase (NAG), a sensitive index of renal tubular damage, during the 3-day period after anesthesia. METHODS: Thirty-four healthy patients who underwent either sevoflurane (23 patients) or isoflurane (11 patients) anesthesia at a total gas flow of 61/min for orthopedic surgery scheduled to last at least 5 h were studied. At 16.5 h after cessation of anesthesia, patients were administered 10 units of vasopressin and urine was collected frequently thereafter for evaluation of urinary osmolality. In addition, urinary excretion of NAG was measured before and on days 1-3 after anesthesia. Based on whether peak fluoride concentrations exceeded 50 microM, 23 patients anesthetized with sevoflurane were assigned to a sevofluranehigh group (> 50 microM) or a sevofluranelow (< 50 microM) group. RESULTS: The eight patients in the sevofluranehigh group had a mean peak fluoride concentration of 57.5 +/- 4.3 microM. A significant, albeit weak, inverse correlation was found between peak fluoride concentration and maximal urinary osmolality after the injection of vasopressin (r = -0.42, P < 0.05). Mean maximum urinary osmolality tended to be lower in the sevofluranehigh group (681 +/- 60 mOsm/kg) than in the other two groups after administration of vasopressin, although the difference among the three groups did not quite reach a statistical significance (P = 0.068). One patient had a transient concentrating defect (maximum urinary osmolality = 390 mOsm/kg) on day 1 after anesthesia. Urinary excretion of NAG in both the sevofluranehigh and sevofluranelow groups was greater on days 2 and 3 after anesthesia than before anesthesia. The increase in urinary NAG excretion was dose related with sevoflurane, but there was no difference in results of routine laboratory renal tests on days 2 and 3 after anesthesia among the three groups. CONCLUSIONS: The authors concluded that sevoflurane anesthesia results in increased serum fluoride concentration, a tendency toward decreased maximal ability to concentrate urine, and increased excretion of NAG. However, the increase in urinary NAG excretion was not indicative of clinically significant renal damage in these patients with no preexisting renal disease.