Gout is the most common inflammatory arthritis in men over 40 years and has an increasing prevalence among postmenopausal women. Lowering serum uric acid levels remains one of the primary goals in the treatment of chronic gout. In clinical trials, febuxostat has been shown to be effective in lowering serum uric acid levels to < 6.0 mg/dL.
To evaluate the benefits and harms of febuxostat for chronic gout.
We searched The Cochrane Library, MEDLINE, EMBASE, and International Pharmaceutical Abstracts from inception to July 2011. The ClinicalTrials.gov website was searched for references to trials of febuxostat. Our search did not include any restrictions.
Two authors independently reviewed the search results and disagreements were resolved by discussion. We included any controlled clinical trial or open label trial (OLT) using febuxostat at any dose.
Data collection and analysis
Data and risk of bias were independently extracted by two authors and summarised in a meta-analysis. Continuous data were expressed as mean difference and dichotomous data as risk ratio (RR).
Four randomised trials and two OLTs with 3978 patients were included. Risk of bias differed by outcome, ranging from low to high risk of bias. Included studies failed to report on five to six of the nine outcome measures recommended by OMERACT. Patients taking febuxostat 120 mg and 240 mg reported more frequent gout flares than in the placebo group at 4 to 28 weeks (RR 1.7; 95% CI 1.3 to 2.3, and RR 2.6; 95% CI 1.8 to 3.7 respectively). No statistically significant differences were observed at 40 mg and 80 mg. Compared to placebo, patients on febuxostat 40 mg were 40.1 times more likely to achieve serum uric acid levels < 6.0 mg/dL at 4 weeks (95% CI 2.5 to 639), with an absolute treatment benefit of 56% (95% CI 37% to 71%). For febuxostat 80 mg and 120 mg, patients were 68.9 and 80.7 times more likely to achieve serum uric acid levels < 6.0 mg/dL at their final visit compared to placebo (95% CI 13.8 to 343.9, 95% CI 16.0 to 405.5), respectively; with an absolute treatment benefit of 75% and 87% (95% CI 68 to 80% and 81 to 91%), respectively. Total discontinuation rates were significantly higher in the febuxostat 80 mg group compared to placebo (RR 1.4; 95% CI 1.0 to 2.0, absolute risk increase 11%; 95% CI 3 to 19%). No other differences were observed.
When comparing allopurinol to febuxostat at 24 to 52 weeks, the number of gout flares was not significantly different between the two groups, except for febuxostat 240 mg (RR 2.3; 95% CI 1.7 to 3.0). Patients on febuxostat 40 mg showed no statistically significant differences in benefits or harms. Patients on febuxostat 80 mg and 120 mg were 1.8 and 2.2 times more likely to achieve serum uric acid levels < 6.0 mg/dL at their final visit (95% CI 1.6 to 2.2, 95% CI 1.9 to 2.5) with an absolute treatment benefit of 29% and 44% (95% CI 25% to 33%, 95% CI 38% to 50%), respectively, at 24 to 52 weeks. Total discontinuation rates were higher for febuxostat 80 mg and 120 mg compared to allopurinol (RR 1.5; 95% CI 1.2 to 1.8, absolute risk increase 11%; 95% CI 6% to 16%; and RR 2.6; 95% CI 2.0 to 3.3, absolute risk increase 20%; 95% CI 3% to 14%, respectively). Discontinuations due to adverse events were similar across groups. Total adverse events were lower for febuxostat 80 mg and 120 mg compared with allopurinol (RR 0.93; 95% CI 0.87 to 0.99, absolute risk increase 6%; 95% CI 0.7% to 11%; and RR 0.90; 95% CI 0.84 to 0.96, absolute risk increase 8%; 95% CI 3% to 13%, respectively). No other relevant differences were noted.
After 3 years of follow-up there were no statistically significant differences regarding effectiveness and harms between febuxostat 80 mg or 120 mg and allopurinol groups (adverse event rate per 100 patient-years 227, 216, and 246, respectively).
Although the incidence of gout flares requiring treatment may be increased in patients taking febuxostat compared to placebo or allopurinol during early treatment, no such increase in gout flares was observed in the long-term follow-up study when compared to allopurinol. Febuxostat at any dose was shown to be beneficial in achieving serum uric acid levels < 6.0 mg/dL and reducing serum uric acid levels in the period from baseline to final visit when compared to placebo and to allopurinol. However, the grade of evidence ranged from low to high, which indicates that further research is needed.