|Home | About | Journals | Submit | Contact Us | Français|
A case of antiepileptic hypersensitivity syndrome presenting as an acute septic illness is reported. A 67-year-old man, with a history of essential hypertension (treated with amlodipine) and recently diagnosed nocturnal epilepsy (treated with phenytoin), presented initially with an acute pneumonic illness with periorbital oedema. This was treated successfully with antibiotics and his phenytoin was changed to carbamazepine due to concerns regarding the periorbital oedema. He was re-admitted the following month with a history of weight loss, rigors, night sweats and high temperatures. Investigations revealed very high inflammatory markers. Despite extensive tests, no source of infection, autoimmune disease or malignancy was identified. The carbamazepine was then stopped and over the next 5 days the patient’s symptoms gradually resolved. The patient’s haematological and biochemical abnormalities returned to normal and he was discharged home. Since discharge he has been very well with no further sequelae.
This is a case of antiepileptic hypersensitivity syndrome (AHS) presenting as an acute infective illness. AHS is caused by the aromatic (benzene ring containing) antiepileptic drugs: carbamazepine, phenytoin, primidone, and phenobarbitol. Due to their shared chemical structure they have a high cross-reactivity. AHS usually begins with fevers and high temperatures before developing into a maculopapular rash with cervical lymphadenopathy. There can be much variation in its presentation and frequently it can mimic infections and neoplastic disease, particularly lymphoma.1 Without prompt withdrawal of the offending drug severe complications may ensue. Physicians are frequently aware of antiepileptic drugs and their side effects, but rarely come across the high cross-reactivity and potential for devastating consequences of this family of drugs.
This case is of great importance as it highlights the potentially fatal consequences of AHS that can occur with these widely used antiepileptic drugs. Without the knowledge of its presentation and management it is unlikely to be considered early enough to withdraw the drug to minimise potential complications.
This case also highlights a wider point, namely that drugs are often overlooked as potential causes of signs and symptoms. It demonstrates that we should all be more aware of the side effects of drugs and their complications, and always consider them as potential contributors in a disease process.
The patient, a 67-year-old Caucasian man, is a retired office worker who lives with his wife and enjoys an active, independent life. He is a lifelong non-smoker and drinks alcohol on occasions. His medical history includes a 5 year history of essential hypertension for which he is taking amlodipine 10 mg once a day. He also is on phenytoin 250 mg once a day for nocturnal epilepsy which had been diagnosed 3 months previously. There is no family history of any illnesses.
One month after the patient started taking phenytoin he was admitted to hospital as a medical emergency with shortness of breath, high temperature, dry cough, bilateral oedema and erythematous rash in the periorbital region. Blood tests revealed a raised white cell count and C reactive protein (CRP); arterial blood gases showed a type I respiratory failure. Chest x-ray showed right lower lobe shadows consistent with infection. He was diagnosed with pneumonia and treated with intravenous amoxicillin-clavulanic acid (Augmentin), oxygen and intravenous fluids. The antibiotics were converted to oral after 48 h and continued orally for another 5 days. The patient was reviewed by his neurologist on admission who stopped his phenytoin due to concerns over whether it may have caused the periorbital oedema and rash. Phenytoin values were assessed on admission and were found to be therapeutic. It was felt that the patient had a reaction to phenytoin and so his antiepileptic medication was changed to carbamazepine on the day of discharge. With treatment his pneumonia resolved and he was discharged within a week with neurology follow-up.
After discharge the patient continued to improve but never became fully well. His appetite was poor and he suffered with generalised malaise. Three weeks following discharge he developed episodic fevers and rigors. These were occurring three to four times each day with a predominance throughout the night. His appetite was very poor and he reported that he had lost 5 kg in weight. He then began vomiting and through his general practitioner was referred back to the acute medical services for assessment.
On admission he complained of vomiting, fevers and rigors. He denied any change in his bowels or abdominal pains. There were no respiratory symptoms, rash, myalgia or arthalgia. His ears and throat were normal and he denied any other symptoms. He had no recent travel history, no exposure to animals, and no history of tuberculosis.
On examination he was clinically well and orientated. He was not clinically jaundiced or anaemic, and there was no palpable lymphadenopathy. His observations showed: temperature 40°C, blood pressure 128/69 mm Hg, heart rate 120 beats/min regular, respiratory rate 16 breaths/min, and saturations were 91% on air. Ear, nose and throat examination were normal, as were heart and chest sounds. His abdomen was soft and non-tender. There was no rash or nuchal rigidity and joints and muscles were all normal. Neurological examination was also normal.
The patient’s presentation was highly suggestive of an infective process with the high temperatures, neutrophilia and high CRP. The history and examination did not, however, identify any obvious focus of infection, suggesting the possibility of a hidden source of infection such as an intra-abdominal abscess or bacterial endocarditis.
The second possible diagnosis was an underlying neoplasm such as lymphoma. This diagnosis would be consistent with the episodic fevers and weight loss, but there was no lymphadenopathy clinically or radiologically. Less likely were the possibilities of autoimmune disease and drug effects.
The patient was given paracetamol, intravenous fluids and nutritional supplements. Antibiotics were withheld because no clear source of infection had been identified. On day 3 the possibility of a drug reaction to carbamazepine was considered and it was decided in conjunction with neurology advice to replace the carbamazepine with clonazepam.
For the first 2 days of admission the patient had frequent spiking temperatures of 39°C both day and night. These temperature spikes were occurring up to three times each day. He continued to feel very unwell and nauseous. On day 3 his carbamazepine was withdrawn. From day 4 he had no further episodes of pyrexia, and over the next few days showed daily improvement in his symptoms. His white cell count and CRP values returned to normal within 5 days, by which time he was feeling well enough to go home and was therefore discharged on clonazepam. He was followed up in clinic after 2 months, by which time he was clinically very well, had resumed his normal lifestyle, and was putting on weight. His liver function tests also returned to normal over this 2 month period. He was reviewed by the neurologist and discharged on long term clonazepam. Six months later he is still well with no further repercussions.
We believe that our patient had AHS. His first admission was characterised by a pneumonic type illness. At this time he never produced any sputum and no organism was isolated. His full blood count at this time revealed an eosinophilia of 0.94×109 (normal range 0.02–0.5). At this time he also had periorbital oedema and associated rash—a characteristic of AHS. Taking this into consideration, it seems likely that this first episode may have been a reaction to phenytoin with early eosinophilic pneumonia/pneumonitis, thereby sensitising the body to the aromatic antiepileptic drugs.
The subsequent admission had all the hallmarks of an acute septic illness, with spiking temperatures, high CRP and neutrophilia. However, despite multiple cultures and various imaging and blood tests, we were unable to identify any organisms. While in hospital and taking carbamazepine the patient’s symptoms and abnormal blood tests persisted, and it was not until the drug was withdrawn that there was a rapid and complete resolution of symptoms and inflammatory markers. The timing of the withdrawal and rapid resolution would suggest that it was the drug causing the problem rather than any other possible aetiology such as a viral infection.
We believe that this was an early presentation of AHS, which might have progressed on to more severe complications including widespread rash if the drug had not been discontinued. Most characteristically, rash is an early feature in 90% of cases after a week and is the most common feature leading clinicians to suspect drug involvement. In the early stages rash may be absent. Physicians need a high degree of suspicion and awareness to consider the diagnosis while excluding alternative diagnoses to prevent serious complications developing.
AHS is multisystem disorder that typically starts with high fevers which can be present for up to a week2 before progressing to an erythematous maculopapular rash (on the torso) and in 20% of cases lymphadenopathy.3 Periorbital oedema is a particular feature of AHS1,4,5 which occurs in 30% of cases, and is absent in other drug hypersensitivity reactions.3 Symptoms usually occur from 2–12 weeks after the aromatic antiepileptic drug has been started.3 If the antiepileptic drug is not withheld then severe complications may ensue such as toxic epidermal necrolysis, hepatitis, renal failure and pnemonitis.6
The incidence of AHS is approximately 1 in 1000 to 1 in 10000 exposures.5 Studies have shown that aromatic antiepileptic drugs are all detoxified through a similar enzyme system. Certain people have a deficiency in the enzyme epoxide hydrolase which leads to an accumulation of toxic arene oxide intermediates in organs where the enzyme systems are concentrated.7 These toxic elements lead to cytotoxicity and immune mediated damage. Lymphocyte transformation tests that demonstrate individual patient susceptibility to various drugs have shown that there may well be an autosomal co-dominant pattern of inheritance. These studies have also shown that there is a cross reactivity of up to 80% between the aromatic antiepileptic drugs.7
The key to managing the illness is suspicion of potential diagnoses and hence early identification of the problem, stopping the drug, and instituting supportive care. Benzodiazepines are widely used as an alternative antiepileptic in the acute illness.1,2 The reason for this is that they are effective antiepileptic drugs but do not share many of the side effects which are common with the other drugs, such as skin rashes and haematological abnormalities. Once the acute phase is over then alternative non-aromatic antiepileptics can be used such as sodium valproate and levetiracetam. The clinical signs and symptoms can take weeks to resolve even after the drug has been stopped,7 and therefore early use of other antiepileptics (which can cause skin rashes and abnormal liver tests) is not advisable as side effects and reactions would be difficult to differentiate from the syndrome.1–3
Competing interests: None.
Patient consent: Patient/guardian consent was obtained for publication.