Anticonvulsant drugs can have various side effects on endocrine functions, such as impotence, hirsutism, infertility, and thyroid dysfunction. Gynaecomastia is caused by many types of drugs such as methyldopa, tricyclic antidepressant drugs, isoniazid, and spironolactone,1 but there have been only a few reports of gynaecomastia caused by anticonvulsant drugs, including phenytoin2 and zonisamide.3 We recently encountered a young man with partial seizures, who genetically had a heterozygous mutation of both CYP2C9 and CYP2C19,4normally responsible for biotransformation of phenytoin in the human liver microsomal P-450 system.5 He developed gynaecomastia after increasing the dose of phenytoin.

The patient was an 18 year old man with a diagnosis of left parietofrontal lobe epilepsy since the age of 2 years, until which time his developmental milestones were normal. He had complex partial seizures occurring at least once every day despite various anticonvulsant drugs of usually sufficient dose including clorazepate, phenytoin, and carbamazepine, and zonisamide since the age of 9 years, when he had an epilepsy surgery for a focal high intensity abnormality on T2 weighted MRI in the left parieto-occipital area.His seizures were not controlled. At the age of 18 years, the patient had chronic implantation of subdural electrodes for evaluating the intractable seizures, and based on the results of the invasive evaluation, he had focal resection in the left parietal and mesial parietofrontal areas. After surgery his seizures decreased in frequency to once every week, and became less severe, accompanied by loss of awareness in only a third. Before and after the surgery he was taking phenytoin (175 mg/ day), carbamazepine (900 mg/day), and zonisamide (400 mg/day) giving blood concentrations of 14.3 mg/l, 7.9 mg/l, and 13.2 mg/l, respectively. Two months after surgery, partly because of sleep deprivation, the patient had a cluster of complex partial seizures, some resulting in secondary generalisation, occurring seven times in 90 minutes. After a total of 375 mg phenytoin was intravenously loaded, his seizures were well controlled. Subsequently the medication was maintained at phenytoin (190 mg/ day), carbamazepine (1100 mg/day), and zonisamide (400 mg/day), and the steady state, basal blood concentrations of these anticonvulsant drugs were 16.6 mg/l, 5.2 mg/l, and 9.7 mg/l respectively.

About a month later, the patient noticed bilateral enlargement of his breasts associated with some tenderness restricted to the centre of the breast. The size was about 3 cm in diameter. No galactorrhea was seen. Blood concentrations of luteinising hormone, follicle stimulating hormone, prolactin, and oestradiol were all normal. His liver function was also within normal limits except for slight increases in γ-GTP. The patient had had moderate gingival hypertrophy and mild hypertrichosis. He was taking no medication other than the anticonvulsant drugs.

On rare occasions gynaecomastia has been reported in patients taking phenytoin, and in some of them it disappeared after stopping phenytoin.2 The mechanism as to how gynaecomastia is caused by this drug is uncertain. It was pointed out that some drugs interfere with testosterone synthesis as well as action by blocking the cytosol receptor of androgen in target tissues. Is has been shown that liver enzyme inducing antiepileptic drugs such as carbamazepine and phenytoin are associated with increased concentrations of serum sex hormone binding globulin and decreased amounts of free androgen available in the tissue.6 However, as for most drugs causing gynaecomastia, its mechanism is not well defined.

The anticonvulsive drug zonisamide has been clinically available in Japan for patients with seizures since 1989, and it is reported that three patients (age ranging from 3 years to 32 years) taking zonisamide (60 to 600 mg/day) developed gynaecomastia from 10 days to 1 year after taking zonisamide, and in all of them it disappeared after stopping zonisamide.3 Therefore, in the present patient, it is possible that both phenytoin and zonisamide played important parts in the development of gynaecomastia. However, the patient had been on zonisamide for the past 8 years, and its dosage was not increased before gynaecomastia developed. On the other hand, after phenytoin was loaded when he had frequent seizures, its maintenance dosage increased. Therefore, phenytoin is most likely a causative agent. Adolescent gynaecomastia usually occurs bilaterally, although asymmetric, in many boys as a physiological type, often during puberty with a mean onset age of 14 years, and it is correlated with transiently raised oestradiol of unknown origin before completion of puberty.1 However, there was no increase in oestradiol in the present patient.

The patient had a heterozygous point mutation in the defective allele of CYP2C9 and CYP2C19,4 and in such patients V_max values in the pharmacokinetics of phenytoin were 40% lower than those in patients with a normal type CYP2C subfamily (see subject No 43 in fig 1 of Odani et al 4) and thus our patient’s maintenance dose of phenytoin (175 mg/ day) was relatively small to achieve an appropriate blood concentration (14.3 mg/l). After having increased the maintenance dose of phenytoin by 15 mg/ day, the blood concentration was kept <20 mg/l, and clinically he did not show toxic symptoms such as cerebellar ataxia and gaze evoked nystagmus. The presence of a heterozygous CYP2C subfamily mutation might have played some additional part either in developing this rare side effect or in facilitating physiological gynaecomastia of the adolescent type.