Effect of xanthine oxidase inhibitors on the renal clearance of uric acid and creatinine

To the Editor: We read with interest the paper by Ma et al. [1] on uratelowering therapies and specifically, their effects on the renal clearance of uric acid (CUA). Ma et al. reported that xanthine-oxidase inhibition (XOI), as achieved by allopurinol and febuxostat, promotes CUA and, in parallel, increases the renal clearance of creatinine (CCR). There are several aspects of the results of Ma et al. which we would like to comment on. Firstly, it should be noted that the mean changes in CCR and CUA observed by Ma et al. are small (CCR, 89.1 to 92.3 mL/min/1.73 m and CUA, 2.87 to 3.09 mL/min/ 1.73 m). The influence of allopurinol on renal function is inconsistent in the literature. A systematic review and meta-analysis by Bose et al. [2] found no significant effect of allopurinol on GFR (mean difference from placebo, 3.1 mL/min/1.73 m2, 95 % CI: −0.9–7.1, p= 0.13), although the effect size found in the meta-analysis is similar to that found in the study by Ma et al. [1, 2]. However, Bose et al. considered that allopurinol may slow the progression of chronic renal impairment [2]. Secondly, a confounding aspect of Ma et al.’s study is that all subjects in this study were given sodium bicarbonate (600 mg/day) to alkalize the urine. Alkalization of urine has been reported to increase the renal clearance of uric acid [3, 4]. Whether sodium bicarbonate was administered before the therapy with XOIs commenced must be answered in order to demonstrate any effect of ULT on the renal clearance of uric acid. Thirdly, Ma et al. observed a non-significant decrease in the fractional clearance of urate (FCU, CUA/CCR) in patients taking XOIs. The FCU is a much more reliable indicator of the efficiency of the kidney to clear uric acid [5] than CUA alone. However, Ma et al. found the FCU was maintained in patients who reached the treatment target serum UA of <6 mg/dL (0.36 mmol/L) but decreased in patients who did not achieve the target. Our research indicates that the patients who do not reach target are more likely to have a higher baseline UA [6]. Consistent with our findings, the target-failure patients in Ma et al. had higher baseline UA concentrations than those who reached target (baseline mean UA 10.3 vs 9.4 mg/dL, p< 0.05). It would have been of interest to relate the contrasts in CUA, CCR, and FCU values directly to the baseline serum UA. Finally, experimental methods are not ideal. Ma et al. used a 24-hour urine collection to quantify various indices of the renal handling of UA. The 24-hour urine collection, while generally deemed the Bgold-standard^ for renal function analysis, is highly prone to error [7]. This is because it is difficult for ambulatory patients to time the collection accurately and to remember to collect all voids of urine. To determine the effect of XOI inhibition on renal handling of UA, we agree that further study is needed. It would be important to take account of urine pH and baseline plasma urate and to match the groups on and off XOI for these variables.