SEARCH

SEARCH BY CITATION

Keywords:

  • children;
  • CYP2C19*2;
  • drug rash with eosinophilia and systemic symptoms (DRESS);
  • phenobarbital;
  • severe cutaneous adverse drug reactions;
  • Stevens–Johnson syndrome (SJS);
  • Thai;
  • toxic epidermal necrosis (TEN)

Abstract

Background

Aromatic anticonvulsant–induced severe cutaneous adverse drug reactions (SCARs), including Stevens–Johnson syndrome (SJS), toxic epidermal necrosis (TEN), and drug rash with eosinophilia and systemic symptoms (DRESS), are fatal immune-mediated adverse drug reactions. CYP2C19, a cytochrome P450 isoform, plays a role in metabolic rate of aromatic anticonvulsant. HLA-B*1502 has also been demonstrated to be associated with carbamazepine-induced SJS-TEN.

Methods

Forty case patients who were diagnosed with SCARs after initiation of phenobarbital (PB), phenytoin (PHT), or carbamazepine (CBZ) for 1–8 wk and forty control patients who received PB, PHT, or CBZ at least 2 months with no adverse drug reactions were enrolled in the study. The genotypes of CYP2C19*1, CYP2C19*2, and HLA-B*1502 were analyzed using allele-specific polymerase chain reaction technique. Clinical characteristics of SCARs patients who used different drugs were also analyzed.

Results

There was no significant difference in sex, onset of symptoms, laboratory results, treatment, and length of stay among patients with SCARs due to PB, PHT, or CBZ. The patients with CYP2C19*2 variant had a trend to have a likelihood to develop SCARs more than the patients with CYP2C19 wild type (OR = 2.5, 95% CI (0.96–67.3) p = 0.06). In subgroup analysis, the patients with CYP2C19*2 variant were at four times increased risk of SCARs from phenobarbital more than the patients with CYP2C19 wild type (OR = 4.5, 95% CI (1.17–17.37) p < 0.03). There was no association between the HLA-B*1502 and aromatic anticonvulsant–induced severe cutaneous adverse reactions (SCARs).

Conclusion

CYP2C19*2 variant may play a role in the genetic predisposition of SCARs from phenobarbital.