Liver toxicity with clozapine

To the Editor

Sporadic cases of serious clozapine liver toxicity appeared in the 1990s (Kellner et al., 1993; Markowitz et al., 1997). Two reports of fatal acute liver failure have been published by MacFarlane et al. (1997) and Chang et al. (2009). We report a case of fulminant liver toxicity with good outcome.

The patient was a 21-year-old male university student. He had symptoms of social withdrawal and declining academic performance over the previous 2 years, and the development in recent months of auditory and visual hallucinations, as well as disorganised behaviour including posturing and catatonia. He had been tried on olanzapine with no clear benefit.

On admission he displayed extreme perplexity, thought blocking, flatness of affect, and detachment from those around him. He reported distressing hallucinations, torment and thoughts of suicide.

Investigations for pathology other than schizophrenia were negative. Liver function tests were normal.

Following trials of other antipsychotics, the use of clozapine was proposed. The patient was keen for this. He was enrolled in the clozapine monitoring program.

Clozapine was started at 12.5 mg daily and ramped up gradually in accordance with the recommended standard protocol. He tolerated this well apart from some early sedation and moderate tachycardia without electrocardiogram changes, effectively treated with atenolol. There was significant improvement in his mental state after a week, and this continued to improve.

About 3 weeks after commencing clozapine the patient reported somnolence, which increased over the ensuing week. Liver tests were done: alanine aminotransferase (ALT) was 794 U/L (normal 5–55), aspartate aminotransferase (AST) was 305 U/L (normal 5–55), bilirubin 25 μmol/L (normal 21 or less), albumin and gamma-GT were normal, white cell count (WCC) was 12.1 × 10⁹/L (normal 4–10) with neutrophilia, and haemoglobin (Hb) 120 g/L (normal 130–170).

These were repeated several days later with the following results: ALT 1914, AST 830, bilirubin 76, albumin 35 g/L (normal 38–48), gamma-GT 99 U/L (normal 60 or less), WCC 21.4 and Hb 107. Clozapine, which by this stage was at a dose of 250 mg per day, was stopped immediately and urgent gastroenterological advice sought, resulting in the patient’s transfer to a gastroenterology unit for observation.

Blood tests were repeated daily. Transaminase peaks (ALT 2282, AST 1056) occurred on the day after cessation, and gamma-GT peaked at 164 the following day. Bilirubin peaked at 112 on day 3, and international normalized ratio (INR) was raised to 1.3 for 5 days. Albumin was below normal for 2 weeks after drug cessation. C-reactive protein (CRP) peaked at 45.5 mg/L the day after cessation. Haemoglobin reached its nadir of 96 g/L 11 days after cessation. The WCC dropped to normal levels after 4 days and all enzymes had returned to normal levels 3–4 weeks after clozapine had been ceased, bilirubin being the last indicator to normalise.

The patient was restarted on amisulpride and within a few weeks he was discharged to the care of his parents on a dose of 200 mg twice daily with ongoing support in the community by the First Episode Psychosis team. Liver function tests were within normal limits.

The case illustrates the treacherous nature of this rare fulminant hepatotoxic reaction. Our patient complained only of sedation in the context of increasing doses of a very sedating drug. As with other cases reported, there was no history or biochemical evidence of pre-existing liver disease, or indeed any indicator of vulnerability, and the liver damage progressed very rapidly.

Our patient appears to be the youngest case described in the literature, which may have contributed to his good recovery after such marked liver test abnormalities. Had recovery not occurred, a liver transplant would need to have been considered. It is of interest that the two fatal cases in the literature were considered unsuitable for transplant, apparently at least partly because of their psychiatric diagnoses, a matter of concern in regard to the rights of the mentally ill.

Such cases are thankfully rare (suggested incidence in clozapine patients 0.001%) (MacFarlane et al., 1997), but they are obviously very serious and potentially fatal. Captured early, however, they can clearly have an extremely good outcome. Applying risk management principles to this situation, there is little to be said against adding liver function tests to the already mandatory weekly blood tests taken during the first months of clozapine administration.

Criteria for withdrawing the drug are not established. Elevated bilirubin appears very uncommon in benign transaminase elevations (Hummer et al., 1997) and could be proposed as one criterion. Other criteria would be liver enzyme levels at certain multiples of their normal upper limits (e.g. 3–5).