Article Text

Alopecia with carbamazepine in two patients with focal seizures
  1. Department of Brain Pathophysiology
  2. Department of Neurology, Kyoto University School of Medicine, Shogoin, Sakyo-ku, Kyoto, 606, Japan
  1. Dr Akio Ikeda, Department of Brain Pathophysiology, Kyoto University School of Medicine, Shogoin, Sakyo-ku, Kyoto, 606, Japan.
  1. Department of Brain Pathophysiology
  2. Department of Neurology, Kyoto University School of Medicine, Shogoin, Sakyo-ku, Kyoto, 606, Japan
  1. Dr Akio Ikeda, Department of Brain Pathophysiology, Kyoto University School of Medicine, Shogoin, Sakyo-ku, Kyoto, 606, Japan.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

In the treatment of patients with epilepsy with antiepileptic drugs, alopecia was reported in association with valproic acid (VPA),1 and on rare occasions with carbamazepine (CBZ).2 3 We recently encountered two young women with partial seizures who developed alopecia after starting CBZ.

Patient 1 was a 32 year old woman with a diagnosis of left frontal lobe epilepsy since the age of 27. Her habitual seizures were simple partial seizures consisting of tonic contraction of her right hand and bilateral orbicularis oculi muscles, followed by eye deviation to the right and clonic convulsion of the right side of her face, which were not usually associated with disturbance of consciousness. A cranial MRI was normal and routine EEG showed normal background activity and intermittent irregular 2-3 Hz slow activity at the left frontotemporal area seen every 50 to 100 seconds. As the seizures had started occurring often, the patient had been started on VPA (1200 mg/day) and diazepam (6 mg/day). Because alopecia developed over a period of several months, these two drugs were discontinued, and then the hair loss stopped. As her seizures became less frequent despite discontinuing VPA and diazepam, no medication had been taken in the next year until she visited our hospital, at which time the seizures had started occurring once every month or every other month. The patient was then given CBZ, increased the dose gradually up to 300 mg/day in two months, and the steady state trough blood concentration was 5.0 mg/l. Three months after starting CBZ, the patient developed appreciable hair loss, as seen previously with VPA. After CBZ was switched to phenobarbitone, hair shedding stopped. The patient had no pregnancy, other dermatological problems, endocrine disorders, or autoimmune disturbance such as systemic lupus erythematosus throughout the clinical course.

Patient 2 was a 25 year old woman with a diagnosis of tuberous sclerosis associated with complex partial seizures since the age of 17. She had several small Pringle adenomas on her face and a depigmented macule on her trunk, but otherwise no skin lesions, including in her hair, were seen. Her intellectual function was within the normal range, and her social activity was normal. Multiple angiomyolipomas in both kidneys, plaques in the retina, and a small calcified tuber by the left lateral ventricle were found. Her habitual seizures consisted of loss of awareness, incoherent speech, staring, and right hand dystonic posture. Routine EEG showed normal background activity. Prolonged EEG monitoring recorded ictal rhythmic discharges mainly at the right posterior quadrant. The patient had been on VPA (200 mg/day) and phenobarbitone (80 mg/day) for more than six months, with steady state trough blood concentrations of 66.6 mg/l and 20.2 mg/l respectively. VPA was switched to CBZ to achieve better seizure control, that was increased up to 600 mg/day in one month. Two months afterwards, when the blood concentration of CBZ was 6.5 mg/l, the patient had appreciable hair loss, and especially over the front of her head her hair became very sparse. When the dose of CBZ was reduced to 200 mg/l in two weeks, moulting stopped, and new hair began to grow. No other skin lesions developed simultaneously, and no white tuft hair was seen. Throughout the clinical course, the patient had no pregnancy, endocrine disorders, or autoimmune disturbance. CBZ had been given in a different hospital one year before the present episode, but at that time the patient took the drug only for a week, because it caused dizziness and ataxia.

CBZ often causes allergic reactions such as skin rash, and also systemic lupus erythematosus, both of which can potentially cause hair loss.4 Pregnancy, fever, thyroid dysfunction, and malnutrition are also potential causes of hair loss.5 In the present two patients, there were no such accompanied causes to explain their hair loss. In the previous case report of hair loss associated with CBZ, it occurred a week after CBZ was initiated.2 Human hair follicles have two phases: an anagen (growth) phase of four years, and then a telogen (resting) phase of three months, followed by normal hair loss.5 Therefore, the onset of the hair loss within a week of CBZ administration was interpreted as a disturbance of growing of hair follicles (anagen phase).2 However, in the present report, it took two and three months in patients 1 and 2 respectively, until hair loss started. This can be explained by the premature ending of the anagen phase, as usually seen in treatment with antithyroid drugs, anticoagulants, and VPA.5 This type of hair loss occurs about three months after the insults—namely, at the end of the telogen phase,5 as seen in the present cases. This may explain why in patient 2 hair loss was not recognised when she had taken CBZ for a week previously. Both of the present patients had been on VPA previously, and patient 1 had hair loss most likely caused by VPA. This did not happen in patient 2. Thus it suggests that hair loss caused by CBZ and VPA does not necessarily occur by a common mechanism. The common characteristics in these two patients are that both were young women with partial seizures, which were treated by polytherapy with anticonvulsant drugs. A recent double blind, multi-institution trial to compare CBZ and VPA showed that changes in hair texture or hair loss were seen in 6% of the patients treated by CBZ.3 As the authors did not specifically differentiate between changes in hair texture and hair loss, it is uncertain how often hair loss occurred in association with CBZ, and what the patient population with this side effect was. Although alopecia resulting from treatment with CBZ is not often seen, attention must be given to epileptic patients, especially when precipitating factors of hair loss are present, and also especially in young women.


This study was supported by grants-in-aid for scientific research (A) 06404031 and (C) 08279106 from the Japan Ministry of Education, Science, Sport and Culture, a research grant for treatment of intractable epilepsy from the Japan Ministry of Health and Welfare, and a research grant from the Epilepsy Research Foundation.