May 1998

Absence of Biochemical Evidence for Renal and Hepatic Dysfunction after 8 Hours of 1.25 Minimum Alveolar Concentration Sevoflurane Anesthesia in Volunteers.
Ebert TJ, Frink EJ, Kharasch ED; Anesthesiology 1998;88:601-10.
[ see abstract below ]

The anesthetic sevoflurane undergoes degradation by carbon dioxide absorbents to compound A. Administration of compound A to rats produces renal injury characterized by proximal tubular necrosis, that results in increased urinary excretion of glucose, protein and the proximal tubular enzymes known as NAG and alpha-GST. In rats, higher compound A concentrations produce increases in serum creatinine and BUN. Studies in human volunteers and patients undergoing long sevoflurane anesthesia in low-flow (< 2 L/min fresh gas flow) or closed circuits have not shown renal injury based on serum BUN and creatinine measurements.

Recently, Eger et al. (Anesthesia & Analgesia 1997;84:160-8) administered 8 hours of 1.25 MAC sevoflurane at 2 L/min to volunteers and reported significant urinary glucose, protein and alpha-GST, as well as a distal tubular cell enzyme p-GST. To add to the controversy, two additional studies have appeared evaluating patients who received sevoflurane or isoflurane anesthesia at 1 L/min for procedures of up to 9 hours, which failed to identify adverse affects from sevoflurane despite comparable or higher compound A concentrations¹.

Because of the inconsistent findings across these studies, Ebert et al. repeated the 8 hour sevoflurane study in volunteers at two sites with blinded laboratory analyses. The study methods were duplicated from those of the Eger study. Noninvasive mean arterial blood pressure was monitored and mean pressure less than 50mmHg was treated with head down tilt, 250 ml saline bolus or both. Venous blood and urine samples were collected 24 hours before the experimental day and for 3 consecutive 24 hour periods after anesthesia. Additional samples were obtained 5 to 7 days after anesthesia.

The 13 volunteers in this study averaged age 25 and weight 86 kg. The mean blood pressure during the 8 hour period of anesthesia was 63 mmHg and total fluid administration during the study period was 1,050 ml saline. The average compound A concentration was 30 +/- 4 ppm. Tests of liver function including alanine aminotransferase, total bilirubin and alkaline phosphatase showed no significant changes from pre-anesthesia baseline and no values exceeded laboratory upper limit of normal. There were also no significant changes in BUN, creatinine or creatinine clearance. In this study protein excretion was corrected for 24 hour creatinine excretion (not done in the Eger study). There were no significant changes in average urinary glucose, protein or albumin excretion on any day after anesthesia. Individual values transiently exceeded the laboratory upper limits of normal, and these laboratory abnormalities were not associated with any other elevated renal marker. Urinary alpha-GST and pi-GST were elevated for 2 and 1 days after anesthesia, respectively. All laboratory values had returned to baseline by 5-7 days after anesthesia.

How did the results of this study differ from the earlier Eger study? Although the experimental design and methods were duplicated, the outcomes differed substantially. Eger et al. reported up to 4 gm albumin and 27 gm glucose in the 24 hour urine collections. Ebert et al. here report peak urine albumin of 131 mg per day and urine glucose of 1 gm per day. One difference between the studies was the mean level of compound A in the inspired gas, which was 41 ppm in the Eger study and 30 ppm in this study. The studies were matched for the key factors in compound A production including fresh gas flow rate, body mass, sevoflurane concentration, and CO2 absorbent temperatures. It is not known whether this difference in amount of compound A could produce the laboratory changes seen. Another difference was in the mean arterial blood pressures recorded during the 8 hour anesthetic administration. In the Eger study mean BP was 56 mmHg compared with 62 mmHg in the present study. Ebert and colleagues note that in another study where higher levels of concentration compound A were inspired by patients who were not hypotensive, there was no evidence of renal injury². Ebert and colleagues comment that low blood pressure should not be ruled out as a contributor or cofactor in the renal dysfunction reported previously.

What can be concluded from this study? First, as these authors point out, the relevance of this and previous volunteer studies to clinical anesthesia in surgical patients is not certain. Surgical patients are not routinely maintained on 1.25 MAC sevoflurane for 8 hours under hypotensive, volume-restricted conditions. In summary, assessments of renal and hepatic function after 8 hours of 1.25 MAC sevoflurane anesthesia did not show the abnormalities reported in the earlier study. The reasons for these differences are not entirely clear. We are still left with the sevoflurane-compound A controversy; we now have more data on which to make our individual clinical judgments.

¹ Anesthesiology 1997;86:1238-53 and Anesthesiology 1997;86:1231-7.

² Anesthesiology 1997;86:1238.

Return to the Current Literature Review Front Page , or read the abstract:

ABSTRACT