Carbamazepine-Induced DRESS Syndrome During Epstein-Barr Virus Reactivation in an Adolescent

Introduction

Drug reaction with eosinophilia and systemic symptoms (DRESS) syndrome, referred to as drug-induced hypersensitivity syndrome, is a distinct, potentially life-threatening adverse reaction. It is seen in children and adults most often as a morbilliform cutaneous eruption with fever, lymphadenopathy, hematologic abnormalities, and multiorgan manifestations.¹ The initial event of this reaction is often viral reactivation; the virus in question is typically a member of the herpes family, such as human herpesvirus-6, Epstein-Barr virus (EBV), or cytomegalovirus.² DRESS syndrome is more common in adults and rarely seen in children. The most common clinical presentation of pediatric DRESS includes morbilliform rash in more than 99% of cases.³

Recognizing this syndrome is important as the mortality rate is around 5.4%.⁴ DRESS syndrome must be recognized promptly and the causative drug withdrawn. It has been reported that the earlier the drug withdrawal, the better the prognosis.⁵

Here, we present a case of DRESS syndrome in which the causative agent was carbamazepine (CBZ), which was associated with EBV reactivation, which was confirmed by a positive patch test.

Case Report

A 14-year-old female was admitted to the hospital with fever, facial edema, and a spreading rash, evolving over 3 days. She was seen 6 weeks prior by a child psychiatrist who prescribed a treatment consisting of CBZ 20 mg/kg/day and sertraline 50 mg/day for depressive syndrome. She had no significant past medical history, no prior medication use, and no known allergies. On physical examination, her temperature was 40°C, a diffuse erythematous rash was noted over the trunk (Figure 1), the face, and extremities with evident facial edema, and she weighed 40 kg. CBZ and sertraline were immediately withdrawn and replaced by fluoxetine with an initial dose of 10 mg/day. The patient received pulsed intravenous methylprednisolone 3 mg/kg/day for 3 days and oral prednisone 0.5 mg/kg/day for 3 weeks, followed by a gradual taper. Initial investigations were compatible with an inflammatory reaction and liver dysfunction as follows: C-reactive protein 65 mg/L (normal range [NR]: 0–8 mg/L), alanine aminotransferase 141 U/L (NR: 5–41 U/L), aspartate aminotransferase 114 U/L (NR: 0–50 U/L), and gamma-glutamyltransferase 110 U/L (NR: 9–50 U/L). Initial full blood count showed a white blood cell count of 6.8 × 10³/µL (NR: 4.3–10 × 10³/µL) and mild eosinophilia with 11% (NR: 0–5%). Serology tests were negative for viral hepatitis, cytomegalovirus, herpes simplex virus, and parvovirus B19. Recurrent EBV infection was demonstrated by the presence of IgM anti-VCA antibodies and IgG anti-EBNA antibodies in the serum taken 3 days after the resolution of the skin eruptions. A biopsy of the skin lesions revealed a combination of mild spongiosis and some necrotic keratinocytes with infiltration of lymphocytes and eosinophils, suggesting a drug reaction.

View Figure

• Download as Powerpoint
• Download Figure

Within 4 weeks, the outcome was favorable, with a resolution of symptoms. The abnormal laboratory test results described above had normalized. Over a follow-up period of 4 months, the patient had no further episodes of skin rash nor any symptoms of autoimmune disease.

To identify the inducing agent of the hypersensitivity reaction for this patient, patch tests for CBZ and sertraline were performed 6 months after complete recovery and induced a strongly positive skin reaction to CBZ in 48 hours but was negative to sertraline (Figure 2).

View Figure

• Download as Powerpoint
• Download Figure

In our case, the clinical, biological, and histological data are in accordance with the DRESS diagnosis criteria. According to the Registry of Severe Cutaneous Adverse Reactions scoring system established by Kardaun et al,⁶ our case yielded a score of 6. Thus, the DRESS diagnosis was “definite” (Table 1).

Table 1.Results of the RegiSCAR Scoring System

View Fullscreen

Discussion

Carbamazepine and sertraline were the suspected culprit drugs in view of a clear temporal relationship between their administration and the onset of the symptoms (6 weeks), as well as the improvement of the clinical and biological disorders some weeks after their withdrawal.

In our case, this adverse drug reaction was reported as “probable” by applying the Naranjo adverse drug reaction probability scale for the suspected drugs⁷ (Table 2). However, the role of CBZ in the occurrence of this adverse drug reaction was confirmed by skin tests. Therefore, the diagnosis of CBZ-induced DRESS was established.

Table 2.Naranjo Adverse Drug Reaction (ADR) Probability Scale

View Fullscreen

CBZ-induced DRESS is well documented in the adult literature, with many case reports.^8,9

In our review of the literature and according to data from MEDLINE, only 1 case of CBZ-induced DRESS during EBV infection was reported with an 8-year-old male during treatment with CBZ for epilepsy 9 weeks after drug withdrawal and corticosteroid treatment.¹⁰ Contrary to our case, this was a primary EBV infection and not reactivation, and patch tests were not performed.

Only 4 documented pediatric cases of CBZ-induced DRESS have been previously reported.^11–14 Patch tests were not performed in 3 cases, and viral reactivation was not causative.^11–13 One report of an 8-year-old female who developed a DRESS syndrome 5 weeks after starting CBZ did not have viral reactivation, but patch tests were positive for CBZ.¹⁴

In children, one of the major difficulties in the diagnosis of drug hypersensitivity is the differentiation of maculopapular eruption as an allergic reaction from a viral exanthema, which is very common. Peripheral blood eosinophilia may sometimes be helpful in the differentiation of drug reactions from viral infections. If an allergy is suspected, an allergy workup is recommended. Patch tests are a useful and safe tool for identifying the culprit drug for the DRESS syndrome.¹⁵

The most common causative agents in pediatric DRESS are antiepileptic drugs (50%), including carbamazepine, phenytoin, and phenobarbital.³ The aromatic ring in the chemical structure of carbamazepine leads to a higher risk of hyper sensitivity reactions. Age-related differences in drug metabolism may result in increased hypersensitivity to antiepileptic drugs in young children.¹⁵

Human herpes virus HHV-6, HHV-7, herpes simplex virus, cytomegalovirus, and EBV reactivation have been associated with DRESS syndrome.^16,17 The role of EBV in the pathogenesis of DRESS syndrome is still unclear. It is uncertain whether the commonly encountered viral infections during childhood play a role in or trigger hypersensitivity reactions to antiepileptic drugs in children.¹⁵ Viral infections may change drug metabolism or act as danger signals, leading to an immune response.¹⁸ Descamps et al¹⁹ proposed that EBV amplifies the T-cell activation induced by drugs and participates in developing visceral manifestations.

The concept that DRESS is no more than a viral disease triggered by a direct effect of drugs on virus reactivation and proliferation has been proposed.^16,20 Immunosuppression from treating DRESS with corticosteroids could contribute to viral reactivations.²¹

Conclusion

Even though DRESS is rare in childhood, we present here another case of carbamazepine-induced DRESS in an adolescent associated with EBV activation and confirmed by a positive patch test. Clinicians should be aware of the severe adverse effects that could be induced by CBZ in pediatric patients.