• valproate encephalopathy
• hyperammonaemia

In their excellent review of non-hepatic hyperammonaemia, Hawkes and colleagues acknowledge the diverse modes of presentation of, and the importance of a high index of suspicion for, encephalopathy secondary to raised blood ammonia concentration.¹ We would like to complement their review by reporting the case of a 78 year old woman with valproate encephalopathy associated with hyperammonaemia.

The patient presented to the accident and emergency department with a four week history of acute on chronic confusion, altered personality, and uncharacteristic aggressive behaviour. She had been taking sodium valproate (modified release) 500 mg twice daily and 300 mg at night, in addition to carbamazepine (modified release) 400 mg twice daily, for epilepsy diagnosed 20 years earlier.

On examination, her temperature was 37.5^oC. The patient was confused, extremely agitated, and uncooperative. Asterixis was noted. The deep tendon reflexes were symmetrically depressed but present in all limbs and the plantar responses were extensor. There were no signs of meningism. Physical examination was otherwise unremarkable.

Full blood count, renal, liver and clotting profiles were normal on admission. In addition, the random blood level of carbamazepine was 8 mg/l (therapeutic range 4–10 mg/l) and of valproate was 80 mg/l (therapeutic range 60–100 mg/l). Computed tomography of the head did not reveal any intracranial space occupying lesion or haemorrhage. Examination of the cerebrospinal fluid was unremarkable. Nevertheless, the patient was started empirically on acyclovir treatment for presumed herpes simplex encephalitis.

Over the subsequent three days, the patient's level of agitation became extreme and a neurological opinion was requested. The diagnosis of valproate encephalopathy was suspected and a blood ammonia concentration confirmed hyperammonaemia (174 μmol/l, normal range 0–59 μmol/l). Within 24 hours of discontinuing the sodium valproate, the patient's aggression and agitation resolved and her level of confusion improved. The blood ammonia fell to 39 μmol/l 11 days after stopping the sodium valproate, and the patient was eventually discharged home on an increased dose of carbamazepine alone (600 mg twice daily) for treatment of her epilepsy.

Since 1979, there have been at least 30 cases of sodium valproate associated encephalopathy reported in the specialist neurological and pharmacological literature, however, only two reports have appeared in the general medical literature in English.^2,3 Hyperammonaemia is an important and potentially reversible cause of encephalopathy, and should be suspected in any confused patient on sodium valproate therapy.