Background/Purpose: Dermatologic side effects to use of gout treatments are concerning to patients. The goal of the study was to assess the risk of occurrence of Drug Rash with Eosinophilia and Systemic Symptoms (DRESS) with gout medications.

Methods: This retrospective study utilized the Veterans Affairs (VA) administrative and clinical databases from fiscal year 2001 to 2012.  Patients with considered to have prevalent gout based on the presence of >=1 International classification of diseases, ninth revision (ICD-9) codes for gout during an inpatient visit or during ³2 codes during outpatient visits. We defined exposure to allopurinol, febuxostat, colchicine as at least 30-day filled prescription; since pegloticase is given as a set of infusions, exposure was defined 1 or more infusion. Control population included patients not exposed to any of the four medications, i.e., Urate-lowering therapies (ULT) including Xanthine oxidase inhibitors (allopurinol, febuxostat) and pegloticase or anti-inflammatory medication, colchicine.  DRESS was defined as the 1^st presence of a code 995.27 in either inpatient or outpatient encounter, that occurred on or after filling the index prescription.  Each patient contributed to exposed and non-exposed periods.  Switching from one ULT to another led to censoring the patient from contributing to that medication and beginning to contribute the person-time to the current/new ULT.  ULT and colchicine did not censor each other, since both can be used together. We used Cox proportional hazards models that assessed hazards of DRESS adjusted for the following factors: Model 1:  drug exposure, age, gender, body mass index, race, marital status, region; Model 2: variables in model 1, plus baseline Charlson comorbidities.

Results: There was 198839/220184.7 patients/PYs exposed to allopurinol, 3116/2093.7 patients/PYs to febuxostat; 30/10.2 patients/PYs to pegloticase and 150623/144225.4 patients/PYs to colchicine. 

In the multivariable-adjusted Cox proportional hazards model that adjusted for demographics (Model 1), compared to controls allopurinol, febuxostat, colchicine and pegloticase were each associated with statistically significantly higher risk of DRESS (Table 1).  Results were similar when models were additionally adjusted for comorbidity (Model 2).

Conclusion: Allopurinol and colchicine were associated with a 2-fold increase in hazards of DRESS.  With much smaller numbers, pegloticase was associated with dramatic increase in the risk. Febuxostat did not reach statistical significance for increased risk of DRESS in full model, although the odds were similar to allopurinol and it was significantly associated with the risk of DRESS in smaller model. More research is needed to define the risk factors for these reactions with gout medications to allow a more judicious and safer use in high-risk populations.

Table 1. Risk of DRESS with gout medications