OBJECTIVE: The objective of this study was to estimate the association between adverse drug reactions (ADRs) and exposure to allopurinol maintenance doses higher than those in the 1984 suggested limits of Hande et al. adjusted for level of renal function. METHODS: We conducted a retrospective review of electronic health records of patients prescribed allopurinol from January 1, 2004, to June 30, 2011, to identify those who had a definite or possible ADR to allopurinol. The associations of ADRs with maintenance doses of allopurinol 1 to 1.5 times and more than 1.5 times the suggested limits of Hande et al. compared with doses within the suggested limits of Hande et al. were estimated using logistic regression models. RESULTS: Of 4755 patients prescribed allopurinol, 2946 had a serum creatinine measured within 6 months of starting allopurinol, and of these, 1268 patients' records were reviewed. Forty-eight patients had a definite ADR to allopurinol, 2 of which were allopurinol hypersensitivity syndrome. The odds ratios of definite ADRs with maintenance doses of allopurinol 1.0 to 1.5 times and more than 1.5 times suggested compared with doses within suggested limits were, respectively, 1.42 (95% confidence interval [CI], 0.66-3.04) and 2.04 (95% CI, 0.87-4.77). Among those with an allopurinol maintenance dose more than 1.5 times suggested limits, the proportion of patients with a definite ADR was 2.6% (95% CI, 1.0%-5.2%). CONCLUSIONS: There is no significant association of high maintenance doses of allopurinol with ADRs, and the absolute risk of ADRs at doses higher than 1.5 times the 1984 suggested limits of Hande et al. is low. Cautious, gradual increases in allopurinol maintenance doses above the suggested limits of Hande et al. are warranted if necessary to achieve a serum uric acid level less than 6 mg/dL.