In Reply:—

In Dr. Wark's letter, he states, “hepatic failure after cardiopulmonary bypass is unlikely to be isoflurane hepatitis.” In response, I will borrow an Asian proverb:“A journey of a thousand miles must begin with a single step.”

Dr. Wark introduces alternative views of a classic argument. We both agree that hepatotoxicity can occur following halothane administration. However, we disagree about hepatotoxicity when other halogenated volatile anesthetics are considered. Nonetheless, hepatotoxicity following halogenated volatile anesthetics, including isoflurane, is believed to be idiosyncratic, 1and immune-mediated processes are believed to have a role in this injury, similar to other forms of idiosyncratic drug-induced hepatotoxicity. 2There are several reports of immune responses and hepatotoxicity following isoflurane administration. 1More importantly, in this case report, isoflurane was implicated after critically examining other possible causes of hepatotoxicity and was not based on histologic findings alone.

This case report takes forward steps in the journey of a thousand miles to the discovery of the etiologies of drug-induced hepatotoxicity. The first step was also taken by those before us. We demonstrated liver injury following isoflurane administration. Norepinephrine and nitroglycerin were administered, but hypotension or other evidence of ischemia was not demonstrated. The article from 1996 referenced by Dr. Wark clearly states the risk factors for gastrointestinal complications after cardiopulmonary bypass: advanced age, combined coronary artery bypass grafting–valve operation, postoperative low cardiac output, prolonged ventilation time, reexploration of the chest, sternal infection, and a history of peptic ulcer. In our report, the patient was 66 yr old, underwent a three-vessel coronary artery bypass graft, and maintained mean arterial pressures ≥ 70 mmHg for > 48 h intraoperatively and postoperatively during the initial demonstration of increased aspartate aminotransferase. Therefore, if Dr. Wark's probability was correct, the only risk factor was age.

By showing that the patient had several risk factors for volatile anesthetic-induced liver injury, additional steps were taken. The patient had received two anesthetics in 3 months; developed a relapsing fever postoperatively, which could suggest the presence of an immune-mediated process; and also developed significant increases in serum transaminases. The immunohistochemical evidence was supportive since the trifluoroacetyl-modified proteins were detected in centrilobular areas, suggesting drug-induced liver injury, and were associated with cytoplasmic organelles, suggesting specificity. Surprisingly, the trifluoroacetyl-modified proteins were detected after 72 h at concentrations previously not seen with isoflurane. Surely, the increased amount of trifluoroacetyl-modified proteins may have been caused by reduced hepatic clearance from global hepatic dysfunction induced by other causes. However, if this was the case, anesthetic metabolism should have been impaired as well.

Our gastroenterology colleagues completed another step in the journey by documenting identical histologic evidence of hepatotoxicity in a patient 5 days after isoflurane exposure for a different surgical procedure. 3We would like to thank Dr. Wark for his comments. We will only move forward by having these kinds of discussions.

Njoku DB, Shrestha S, Soloway R, Duray PR, Tsokos M, Abu-Asab M, Pohl LR, West AB: Subcellular localization of trifluoroacetyl liver proteins in association with hepatitis following isoflurane. A nesthesiology 2002; 96: 757–61
Van Pelt FN, Straub P, Manns MP: Molecular basis of drug-induced immunological liver injury. Semin Liver Dis 1995; 15: 283–300
Yilmaz AT, Arsian M, Demirkilc U, Ozal E, Kuralay E, Bingol H, Oz BS, Tatar H, Ozturk OY: Gastrointestinal complications after cardiac surgery. Eur J Cardiothorac Surg 1996; 10: 763–7