Published: April 1999

Publications

Potentially Serious Adverse Effects of Carbamazepine: Blood Dyscrasias and Skin Rash

This article is more than five years old. Some content may no longer be current.

Prescriber Update 18: 32–36
April 1999

Medsafe Editorial Team

The Centre for Adverse Reactions Monitoring recently received 3 reports of serious adverse reactions with carbamazepine: severe cholestatic jaundice, Stevens-Johnson syndrome, and multiorgan hypersensitivity with fulminant liver failure resulting in death.
Three published incidence studies have investigated the frequency and seriousness of cutaneous or haematological reactions with carbamazepine. Rash was found to occur in around 10% of patients. Most occurred in the first 2 weeks of treatment and were mild. In each of 2 studies one patient developed a serious reaction - erythema multiforme and Stevens-Johnson syndrome respectively.
Blood dyscrasias (moderate and severe leucopenia and 1 case of thrombocytopenia) occurred with an incidence of 2% with mild changes detected in up to 30% of patients. Most cases developed within the first month of therapy.
To reduce the risk of serious adverse effects, a blood screen and physical examination should be conducted during the first 4-6 weeks of therapy, and repeated where there are clinical reasons for concern. Carbamazepine should be withdrawn or the dose reduced if the white cell count falls below 3000/mm³ or the neutrophil count below 1000/mm³.

Reports of severe haematological, cutaneous and hepatic reactions

The Centre for Adverse Reactions Monitoring (CARM) has recently received three reports of serious adverse reactions with carbamazepine. Carbamazepine was originally approved 30 years ago for use as an anticonvulsant, but it is now also being used as a mood stabiliser and in the treatment of chronic pain. The Medicines Adverse Reactions Committee has questioned whether carbamazepine's adverse reaction profile was being adequately considered when it was used in these latter conditions.

The first case was of cholestatic jaundice in a 48-year-old woman treated with carbamazepine, amitriptyline and dihydrocodeine for chronic pain. Carbamazepine had been started a month prior to onset of the symptoms of cholestatic jaundice which included rash, reduced appetite, nausea, jaundice and fever. Investigations for infectious causes were negative. Peak liver function test results were bilirubin 495 µmol/L (normal 2-14 µmol/L), alkaline phosphatase 879 IU/L (40-130 IU/L), and gamma-glutamyltransferase 499 IU/L (1-35 IU/L). Recovery followed withdrawal of carbamazepine and treatment with prednisone.

The second case was of Stevens-Johnson syndrome in a 72-year-old man treated with carbamazepine for 3 weeks for seizures. The symptoms appeared 7 days after discontinuation of carbamazepine and indicated multiorgan hypersensitivity including generalised rash, swelling, cholestatic hepatitis, thrombocytopenia and nephritis. The patient recovered after one month.

The third case was fatal and also involved multiorgan hypersensitivity. The patient was a 9-year-old boy who had been taking sodium valproate for 4.5 months and carbamazepine for 2.5 weeks when the symptoms developed. He had a morbilliform rash, splenomegaly, conjunctivitis, purulent pharyngotonsillitis and lymphadenopathy. Subsequently, fulminant liver failure occurred and he died due to hepatic encephalopathy despite liver transplant.

The WHO database holds a significant number of reports of serious blood dyscrasias and skin reactions with carbamazepine. At February 1999 the WHO database held 208 reports of agranulocytosis, 306 of granulocytopenia, 126 of aplastic anaemia, 227 of erythema multiforme, 544 of Stevens-Johnson syndrome and 168 of toxic epidermal necrolysis out of a total of around 32,000 reports of adverse reactions with carbamazepine. In addition, 39 reports of hepatic failure have been collected by the WHO.

Frequency of rash approx. 10%; mostly mild

Three useful incidence studies have been conducted to estimate the frequency and seriousness of rash or blood dyscrasias with carbamazepine. The first study1 included 335 children with epilepsy treated with carbamazepine. The frequency of skin rash was 9.9% (33 children), with the younger children having a lower frequency (5.0% for those under 6 years).

The types of skin rash were as follows: erythema urticaria (n= 7), miliary exanthema (13), papular speckled rash (11), thrombocytopenic purpura (3), and Stevens-Johnson syndrome (2). One of the patients who developed Stevens-Johnson syndrome initially experienced miliary exanthema which progressed to the more serious reaction on continuation of carbamazepine.

Most rashes developed within 2 weeks & resolved 2 weeks after discontinuation

The skin rashes developed after 8-60 days with 88% of them appearing within 15 days of commencement of therapy. In 2 patients the rash was mild and resolved despite continuation of carbamazepine. The other 31 patients were switched to an alternative anticonvulsant with disappearance of the rash in 1-14 days. Three patients were recommenced on carbamazepine at a lower dose with reappearance of rash. Haematological (leucopenia, eosinophilia, thrombocytopenia) and hepatic effects each occurred in 30% of cases but no patients manifested clinical jaundice.

Another study2 investigated the incidence of rash with carbamazepine in a group of 113 adult psychiatric patients. A total of 13 patients (11.5%) developed skin rash after a mean of 12 (range 9-83) days. The mean age and carbamazepine dose was not significantly different between those with and without rash. The types of skin rash were as follows: erythema multiforme (1), macular (2), maculopapular (4), maculopapular with urticaria (2), maculopapular with pustules (1), and poorly described (3). Only 2 patients had generalised rash. Three patients, including the patient with erythema multiforme, developed a cluster of symptoms including fever, myalgia, pruritus, headache and lymphadenopathy. It took 5 weeks after discontinuation for the erythema multiforme to be resolved, but resolution was much more rapid in other patients (≤ 13 days).

Frequency of blood dyscrasias 2%; mostly mild

A third study3 was much larger and focussed on blood dyscrasias. This study included 977 inpatients treated with carbamazepine, 1251 with sodium valproate, and 65 with both agents simultaneously, compared with 1031 patients taking desipramine or imipramine. Twenty-five patients taking anticonvulsants developed moderate (WBC 3,000-4,000/mm³) or severe (< 3,000/mm³) leucopenia. The occurrence rates were 2% for carbamazepine, 0.4% for sodium valproate, 1.5% for both agents in combined use and 0.3% for the two tricyclic antidepressants. In addition, there was one case of significant thrombocytopenia (platelets < 100,000/mm³) with carbamazepine. 75% of the cases of leucopenia with carbamazepine occurred within 30 days of initiation of therapy, but 2 occurred after just over 100 days. Recovery after withdrawal of carbamazepine occurred after 2-14 days (mean 6.5 days), but in some patients carbamazepine was continued with resolution of leucopenia. Six patients were switched to valproate without recurrence of leucopenia. No mention was made of skin rashes in this study.

Haematological monitoring may reduce risk of serious reactions

From these studies the rate of skin rash with carbamazepine appears to be around 10%, with lower rates observed in young children. Severe rash is rare, perhaps around 0.5%. Rashes are usually of early onset (1-2 weeks) and appear to be allergic in character. To date it has not been possible to identify predictors (e.g. nature of rash, or haematological parameters) for rashes which will progress to severe and life-threatening events, perhaps accompanied by bone marrow suppression and multiorgan involvement. Indeed in some instances continued treatment has been safe. In the present studies it may be that the rashes did not progress to serious conditions because of the high level of monitoring.

Blood dyscrasias occur in about 2% of cases. They are most often transient leucopenia which resolves with continued treatment, but some cases may progress to serious bone marrow suppression. Cross-sensitivity with other anticonvulsants is not the norm, although it may occur, particularly when the event is severe and includes multiorgan involvement.4 A pragmatic approach to the risk of serious blood dyscrasias is to conduct one blood screen during the first 4-6 weeks of therapy, preferably at around 4 weeks, and have a repeat test if there are clinical reasons for concern. Greater vigilance may be appropriate in those with previous drug hypersensitivity or low pre-treatment blood count. Carbamazepine should be discontinued5 or the dose decreased if the white cell count falls below 3000/mm³ or the neutrophil count below 1000/mm³.

Carbamazepine is also associated with CNS effects including dizziness, headache, ataxia, drowsiness, fatigue and diplopia, and gastrointestinal effects, particularly nausea and vomiting.6 These effects often occur at the start of treatment and abate within a few days with continued treatment or dosage reduction. Carbamazepine may also cause hyponatraemia and decreased plasma osmolality by an anti-diuretic hormone-like effect, leading to lethargy, vomiting, headache, mental confusion and neurological abnormalities.

Carbamazepine is a relatively safe medicine, despite its propensity to cause serious and life-threatening blood dyscrasias and bone marrow suppression. The risk of these may be reduced by haematological monitoring and physical examination of patients in the first 4-6 weeks.

References
  1. Konishi T, Naganuma Y, Hongo K, Murakami M, Yamatani M, Okada T. Carbamazepine-induced skin rash in children with epilepsy. Eur J Pediatr 1993;152:605-8.
  2. Kramlinger KG, Phillips KA, Post RM. Rash complicating carbamazepine treatment. J Clin Psychopharmacology 1994;14:408-13.
  3. Tohen M, Castillo J, Baldessarini RJ, Zarate C, Kando JC. Blood dyscrasias with carbamazepine and valproate: a pharmacoepidemiological study of 2,228 patients at risk. Am J Psychiatry 1995;152:413-8.
  4. Havill S, Rademaker M. Anticonvulsant hypersensitivity syndrome. Prescriber Update No.16, p. 28-31, Apr 1998.
  5. Dukes MNG. Anticonvulsants. In Dukes MNG (Ed). Meyler’s Side Effects of Drugs 13th Edn. 1996: Amsterdam, p. 147.
  6. McNamara JO. Drugs effective in the therapies of the epilepsies. In Hardman JG, Limbird LE, Molinoff PB, Ruddon RW, Goodman Gilman A (Eds). Goodman & Gilman’s The Pharmacological Basis of Therapeutics. 9th Edn. 1996: New York, p.474.

 

Hide menus
Show menus
0 1 2 4 5 6 7 9 [ /