Januray 1998
Dose-related biochemical markers of renal injury after sevoflurane versus desflurane anesthesia in volunteers.

Eger EI, Gong D, Koblin DD, Bowland T, Ionescu P, Laster MJ, Weiskopf RB; Anesthesia and Analgesia 1997;85:1154-63.

[ see abstract below ]

1997 might well be remembered as the year of the Great Sevoflurane Controversy.

In the March 1997 issue of AnesthesiaWeb, I reported on an article by Dr Eger on volunteers exposed to 1.25 MAC sevoflurane or desflurane at 2 L/min for 8 hr . He and his colleagues demonstrated that patients receiving sevoflurane developed multiple biochemical markers of nephrotoxic injury, apparently related to the production of Compound A, formed by the interaction of a breakdown product of sevoflurane (inorganic fluoride) with sodalime at low gas flow rates. Although no volunteer subsequently developed increases in BUN or creatinine, his findings were instrumental in the FDA maintaining its warnings on the use of low gas flows (2 L/min) with sevoflurane.

In the August 1997 issue of AnesthesiaWeb, Drs Philip, Lubarsky, and I reviewed two papers that appeared to refute Dr Eger's claims. Bito et al compared low flow sevoflurane, high flow sevoflurane and low flow isoflurane in patients exposed to these agents for about 6 hr, and found no differences in BUN, creatinine or tubular enzymes for three days postoperatively. Kharasch et al studied the use of sevoflurane or isoflurane at 1 L/min in 73 patients undergoing procedures lasting longer than 2 hr, with similar results and no correlation between indices and Compound A exposure. Their conclusions were that moderate duration low flow sevoflurane anesthesia, even with the formation of Compound A, is as safe as low flow isoflurane anesthesia. In the November 1997 issue of Anesthesia and Analgesia, Dr Eger and his colleagues are back! In another study on volunteers, they present findings consistent with a dose-response relationship between biochemical markers of glomerular and tubular injury (urinary albumin, alpha-glutathione-S-transferase) and the "dose" of Compound A.

The threshold of renal injury is between 80 and 168 ppm/h (i.e. parts per million-hours) of exposure to Compound A. This was achieved with the use of 1.25 MAC sevoflurane at 2 L/min gas flow for four hours, but not for two hours, and was not seen at all with desflurane. In the discussion, Eger et al. acknowledge that clinically significant renal injury with the use of low flow sevoflurane anesthesia has not been reported in patients. However, the potential for renal injury appears to be increased by the use of higher concentrations of sevoflurane given for longer periods to larger patients, with hyperventilation in the presence of a lower gas flow. Also, they postulate that the threshold for renal injury may vary idiosyncratically.

Where does this leave us? After all this, it appears as if we can reach the following conclusions:

1. Compound A is generated by the use of sevoflurane with low gas flow (<2L/min).

2. Compound A appears to induce subclinical biochemical injury to the glomerulus and tubules, depending on the "dose", with a threshold of 80-168 ppm/hr.

3. The relationship between Compound A formation, biochemical injury and clinically relevant renal dysfunction remains unclear and unproven.

4. In the light of this information, it appears prudent to avoid situations of possible increased renal injury with sevoflurane (see above) in patients with abnormal renal function or who are at risk to perioperative acute renal failure.

This is obviously a half way position between those such as Kharasch et al who suggest that the renal risk with sevoflurane is negligible, and those such as Eger et al who suggest that the renal risk is ever present under the right circumstances. The future will tell.



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

ABSTRACT