Anticonvulsant hypersensitivity syndrome study of 60 cases

Introduction

Anticonvulsant hypersensitivity syndrome (AHS) is a potentially fatal drug reaction with cutaneous and systemic reaction to the areneoxide producing anticonvulsants (phenytoin, carbamazepine and phenobarbital sodium). The hallmark features include fever, cutaneous rash and lymphadenopathy, accompanied by multisystem abnormalities. It was earlier recognised as dilantin hypersensitivity syndrome when phenyl ethyl hydantoin (phenytoin sodium) was used in the treatment of Sydenham′s chorea,[1] subsequently characteristic constellation of signs and symptoms were reported with carbamazepine and phenobarbital.[2] The incidence of AHS has been estimated to be between one in 1000 and one in 10000 exposures.[2] Although no sex or age predilection has been observed, the overall incidence is higher in blacks.[3] In vitro analysis suggests that siblings of those with a history of AHS are at an increased risk.[4] It is believed that phenytoin, carbamazepine and phenobarbital are metabolized

by cytochrome P-450 to arene oxide metabolites. The metabolites are detoxified by epoxide hydrolase enzyme.[5],[6] This enzyme may be lacking or mutated in persons in whom AHS develops.[7]

The accumulated metabolites are directly toxic to keratinocytes or may form antigen hapten complex with proteins triggering an immune response.[7] Others postulate that the reaction is secondary to circulating antibodies or is a form of allergic hypersensitivity.[8],[9]

Subjects and Methods

Sixty indoor patients of AHS (Male-49, Female--11,Age group 23-65 years) referred to us from neurosurgery departments during August 1995- August 1997 were studied. A total of 33.000 patients, including 4864 of head injury received anti-convulsants during this period. All patients were subjected to detailed clinical history and examination. Primary underlying pathology, constitutional symptoms, nature of onset, distribution, duration and progression of the rash, lymphadenopathy and exposure to offending drugs were noted. Investigations included complete hemogram, liver function tests, blood sugar, histopathologic examination of the rash, patch testing with offending and cross reactive drugs and other relevant investigations were done.

Results

Total number of cutaneous drug reaction during study period were 558. There were 60 patients with anticonvulsant hypersensitivity syndrome constituting 11% of the total patients.

Fever was present before the onset of cutaneous rash and was continuous and high grade (38°C-40.5°C). It persisted for about 1 week after the offending drug was withdrawn. Skin eruption was initially erythematous and maculo papular, involving chest, back and upper arms. It was mildly pruritic and soon evolved into a dusky red confluent erythema with fine scaling. While this process was occurring in upper parts of the body, maculo papular rash progressed below the knees and on the dorsa of hands. In those patients (39 patients, phenytoin-27, carbamazepine-12) who presented to us early, as soon as the offending drug was stopped the progression of rash also stopped. In other patients where early features of AHS were not recognised the rash showed progression to either erythroderma/ exfoliative dermatitis (8 patients, common with carbamazepine-6, phenytoin-1, phenobarbital sodium-1) or SJ syndrome(13 patients, common with phenytoin-11, carbamazepine-1, phenobarbital sodium-1).

One interesting feature was the presence of periorbital and perioral erythema and oedema in 18 (30%) of these patients which is usually not present in other forms of drug reactions.

Tender bilaterally symmetrical cervical lymphadenopathy was present in 12 (20%) patients. Although abnormal LFTs (SGOT,PT) were seen in 15 (25%) patients, there was no evidence of icterus. One patient progressed to acute renal failure.

Histopathology done from the early lesions of the maculo-papular rash showed lymphocytic vasculitis. We performed patch tests with the offending drug and cross reactive drugs i.e. with all the three drugs in 5 patients, but they were negative.

Phenytoin was the commonest offending drug (39 patients) followed by carbamazepine (19 patients) and phenobarbital sodium (2 patients). There were 3 patients on phenytoin therapy who were receiving radiotherapy also for their brain tumours.

Patients on phenytoin developed AHS after a period of 2-6 weeks AV(3), carbamazepine after 2-12 weeks AV (3) and phenobarbital sodium after 6 weeks. [Table - 1]

Of the primary conditions for which drugs were used, head injury was present in 43 patients, brain tumors in 9, epilepsy in 5, neurocysticercosis in 2 patients and encephalitis in I patient. Incidence of AHS in patients of head injury was 1 in 100 ( 1%,43/4864) or much higher than expected. Incidence of isolated cutaneous rash to anticonvulsant was 6% in patients of head injury. [Table - 2]

Discussion

Acute neurotrauma was present in 70% (43 patients). Review of literature showed that the protein binding of anticonvulsant drugs is decreased in patients of head injury with an increase in free levels.[10] Further, head injury and radiotherapy act as an inducer of cytochrome P-450 system leading to increased formation and accumulation of arene-oxide metabolites,[11],[12],[13] these then may cause AHS by toxic or immunologic mechanisms. This may explain the higher incidence in patients of head injury, as well as in patients receiving phenytoin and radio therapy simultaneously.

We found that the rash progressed in 3 patients of head injury with AHS due to phenytoin, when phenytoin was substituted with carbamazepine which is the usual practice in clinical side. In vitro testing has shown that there may be upto 75% cross-reactivity between these arene oxide producing offending drugs.[4] Thus these drugs shouldn′t be substituted with each other. Valproic acid is the safest alternative for seizure control in these patients but because of its hepatic metabolism, it shouldn′t be used during acute or convalescent phase.

Prognosis of AHS seems to be good. In our patients progression of rash stopped as soon as the offending drugs were withdrawn. Patients were treated with topical corticosteroids and anti-histamines with attention paid to hydration and electrolyte balance.

Studies have shown that hepatitis is the cause of liver failure-the most common cause of death in patients of AHS. The overall mortality being 18-40% when the liver is involved.[14],[11] Twenty five percent of our patients showed abnormal LFTs but all recovered well as these patients were diagnosed earlier and offending drugs stopped.