A 31-year-old housewife and mother of three, Mrs F had coped with epilepsy for many years, taking carbamazepine and phenytoin to control her condition.
When she became pregnant again in the mid 1990s, she experienced fits at least three times a week. Her neurologist adjusted her medication to phenobarbitone 60mg three times daily with phenytoin 400mg daily. Her epilepsy was reasonably well controlled on this regimen.
When Mrs F went into labour and attended her local hospital, the admitting doctor recorded and prescribed her medication incorrectly. Instead of phenytoin 400mg daily, he prescribed 400mg three times daily. Nobody noticed this error.
Mrs F had a healthy baby boy, but was kept in hospital due to an episode of fever. Two days later, the nursing notes record – ‘Paranoid ideation – strange behaviour’. These notes also document a complaint of buzzing in the left ear since parturition. By that evening Mrs F was obviously distressed and agitated; her obstetric team asked the on-call psychiatrist to see her and advise.
The duty psychiatrist diagnosed puerperal psychosis and ordered an intramuscular injection of zuclopenthixol with oral anticholinergics and regular chlorpromazine. Dr T, a specialist in psychiatry, reviewed Mrs F the next day, noting auditory third-person hallucinations. He recorded ideas of reference and delusions concerning the newborn baby and his father, and noted an absence of confusion, disorientation or features of an organic brain syndrome.
Over the next few days Mrs F was reviewed by several members of the psychiatry team. As her symptoms were not improving, she was transferred to the local psychiatric unit, where she was treated with haloperidol and ECT.
The ECT failed to produce any epileptiform movements.
Subsequently, Mrs F had slurred speech and visual hallucinations. Noting the presence of features consistent with an organic brain syndrome, Dr T recommended omitting the night-time chlorpromazine.
Over the next few days, Mrs F was restless and agitated. After a further ECT treatment where there were again absolutely no epileptiform movements, Dr T asked a neurologist to review Mrs F. The neurologist, who saw her the next day, identified the medication error and discontinued the phenytoin.
Mrs F recovered, but her coordination and walking were impaired, and she had difficulty regaining her original confidence; these symptoms took several months to resolve.
When Mrs F brought a claim against the hospital, we consulted an expert in psychiatry on behalf of Dr T.
In the expert’s opinion, although her original presentation was consistent with a diagnosis of puerperal psychosis, the failure to notice the error in the phenytoin dose and the evolving features of organic abnormality – namely the absence of any seizure activity during ECT and worsening clouding of consciousness – were indefensible.
One expert in psychiatry questioned the use of ECT altogether in a case such as this. Liability was shared between the psychiatric and obstetric institutes at 25% and 75% respectively.
The initial error in the prescription would probably have been brought to light sooner if (a) the doctors had reviewed the patient’s notes more closely, (b) the doctors had reviewed the medication chart for possible causes of Mrs F’s symptoms, and (c) the nurses administering thrice-daily doses of phenytoin had questioned the prescription.
Phenytoin has zero-order metabolism, so toxicity is a possible cause of acute illness in a patient taking this drug. For a useful tutorial on phenytoin toxicity see emedicine.medscape.com/emergency_medicine. Psychosis is a syndrome, not a diagnosis. Acute or chronic organic brain syndrome (delirium or dementia) can present with acute psychotic features.