|
|
||||||||
Ann Thorac Surg 2005;79:698-699
© 2005 The Society of Thoracic Surgeons
a Papworth Hospital, Papworth Everard, Cambridge, England
Accepted for publication August 28, 2003.
* Address reprint requests to Dr Thomasset, Leicester General Hospital, Gwendolen Rd, Leicester, UK, LE5 4PW
sarahthomasset{at}hotmail.com
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
In March 2002 a 75-year-old woman was admitted to Papworth Hospital for elective coronary artery bypass grafting (CABG). She was a nonsmoker with severe symptoms of both angina and heart failure. An angiogram showed severe triple-vessel disease and mildly impaired left ventricular function.
The patient had severe rheumatoid arthritis for which she was taking leflunomide and cocodamol. She had been taking 20 mg of leflunomide daily, which is the maximum recommended maintenance dose, for 14 months before surgery. Concomitant medication included lisinopril, diltiazem, aspirin, glyceryl trinitrate (GTN) tablets, atorvastatin, budesonide aerosol spray, quinine sulfate, and lansoprazole. These were for hypertension, angina, hypercholesterolemia, asthma, nocturnal leg cramps, and a previous duodenal ulcer. Treatment with atorvastatin was initiated in 1999. Aspirin was stopped 5 days before surgery. All other medications were continued up until the day of surgery. The patient's preoperative liver function test results and coagulation screen were normal.
A combination of the left internal mammary artery and saphenous vein were used to perform a threefold CABG without event. The bypass duration was 46 minutes and the aortic cross-clamp time was 24 minutes. Anesthesia was induced with midazolam and fentanyl. It was maintained with propofol. The muscle relaxant used was pancuronium. This is a frequently used anesthetic regimen for cardiac surgical procedures. All anesthetic agents were administered intravenously. No inhaled anesthetic agents were given.
Leflunomide, aspirin, atorvastatin, budesonide, quinine sulfate, and lansoprazole were all recommenced within 2 days of surgery. Cocodamol and GTN were continued as required. Lisinopril was recommenced on day 5. The patient's postoperative critical care stay was uneventful and she was transferred to the ward on the first postoperative day. Her vital signs (blood pressure, apex rate, temperature, oxygen saturation, and respiratory rate) were monitored every 4 hours in accordance with the hospital's observation protocol. All observations were within normal limits with reference to the protocol until day 6, when the patient developed hematemesis and she was transferred back to intensive care.
An upper gastrointestinal (GI) endoscopy showed a collection of blood and water in the stomach but no obvious lesion. Liver function was deranged. The serum level of alanine aminotransferase (ALT) was 737 U/L (3-35) and alkaline phosphatase was 189 U/L (30-300), suggestive of hepatocellular injury. Coagulation was also abnormal, with an activated partial thromboplastin time of 70 seconds and prothrombin time of 18 seconds. The serum amylase level was within normal limits. Hematemesis continued throughout day 6, and the patient's blood pressure became increasingly labile. On admission to intensive care, the patient's central venous pressure measured 2 mm Hg. Intravenous fluids were administered, and changes in the central venous pressure were characteristic of hypovolemia. The patient's renal function began to deteriorate despite attempts to restore circulating volume and stopping lisinopril. Renal replacement therapy using hemofiltration was commenced late on day 6. Renal failure was attributed to one or more of three causes: hypovolemia, recommencement of lisinopril, and hepatorenal syndrome.
On day 7 the clinical picture deteriorated further, with continued bleeding from the GI tract and suture sites despite attempts to correct the coagulopathy with fresh-frozen plasma. An abdominal computed tomographic scan suggested hepatic ischemia, probably a consequence of hypovolemia. The patient died in the evening of day 8. A postmortem examination revealed a congested liver and multiple hemorrhagic sites throughout the GI tract. The mucosal surface of the stomach was hyperemic, with an area of superficial ulceration. No chronic ulceration was seen.
| Comment |
|---|
|
|
|---|
This patient had been prescribed a number of drugs that are common to many patients before CABG: lisinopril, diltiazem, aspirin, GTN tablets, atorvastatin, budesonide, lansoprazole, and cocodamol. No reports have linked any of these drugs to post-CPB hepatic dysfunction. Similarly, no reports link midazolam, fentanyl, or pancuronium to such liver problems, although a small number of reports have linked propofol to post-CPB hepatic dysfunction. Considering that propofol is usually included in anesthetic regimens for cardiac surgical procedures, any association with post-CPB hepatic dysfunction must be weak.
Leflunomide and quinine sulfate are not so commonly prescribed to patients before CABG. Quinine-induced hepatotoxicity has been reported rarely. However, leflunomide-induced hepatotoxicity is well recognized, so much so that the monitoring of liver function is recommended throughout therapy [1]. Aspartate aminotransferase and ALT levels can both be greater than twice the upper limit of normal in 11% of patients treated with leflunomide [1]. In addition, since leflunomide was first marketed in September 1998, it has been associated with six times more cases of fatal liver toxicity than methotrexate, an antimetabolite that is also prescribed for troublesome rheumatoid arthritis [2].
The mechanism of leflunomide-induced hepatotoxicity is unknown. Leflunomide, an isoxazole derivative, is a member of the disease-modifying class of antirheumatic drugs. It is rapidly converted in the GI tract and plasma to its active metabolite, A771726. The primary mode of action of A771726 is to prevent T-lymphocyte proliferation by reversible inhibition of dihydroorotate dehydrogenase, a key enzyme involved in pyrimidine synthesis [1]. Most cases of leflunomide-induced hepatotoxicity have occurred within the first 6 months of treatment [5].
Cardipulmonary bypass alters the clinical effects of many drugs. A number of studies, for example, have reported adverse perioperative events in patients receiving chronic angiotensin-converting enzyme inhibitor therapy before CPB. Potentiated vasodilatation leads to an increased requirement for vasoconstrictors, which on occasion may lead to problems that probably arise from graft spasm [6]. In addition, increased digoxin levels following CPB have been blamed for the development of postoperative arrhythmias [7]. It may well be that CPB acts to potentiate the unwanted action of leflunomide and its class of drugs. The metabolism of leflunomide is incompletely understood, and pharmacokinetic changes during and after CPB have not been described. However, we propose that CPB could potentiate the action of leflunomide in any or all of the following three ways:
Although we cannot rule out the possibility that a combination of drugs caused liver dysfunction, it does seem possible that leflunomide was involved. As use of this disease-modifying drug increases in the cardiac surgical population, research is required to define the risk it poses for these patients. Until such evidence is available, we recommend that leflunomide treatment should be discontinued for at least 2 weeks (approximately one half-life) before cardiac surgical procedures that require the use of CPB and emphasize the importance of reporting possible future cases.
| References |
|---|
|
|
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |