Overall, there were 10,537 observations in the NHANES 2009–10, from which we excluded 4,319 participants less than 20 years of age, 99 women who were pregnant or breastfeeding and 530 observations due to missing values of gout, hyperuricemia or serum creatinine, leaving 5,589 observations in the analysis dataset. The 99th percentile of serum creatinine measured was 1.86 mg/dL (164 µmol/L); 22 participants in the dataset reported receiving hemodialysis or peritoneal dialysis in the preceding 12 months. shows the characteristics of the study sample by renal impairment status.
Comparison of participants by renal function.a
shows the estimated number of gout cases in the US by renal function, age, gender, and ethnicity.
Estimates of number of people with gout and hyperuricemia, in millions, by renal impairment status.a
Overall, the unadjusted prevalence of gout among those with any renal impairment was five-fold higher than among those without any renal impairment [7.3%; (6.0%, 8.8%) vs. 1.4% (0.9%, 2.1%)]. The unadjusted rates among men with and without any renal impairment were 9.3% (6.8%, 12.5%) and 2.5% (1.6%, 3.9%), respectively. Among women, the corresponding rates were 5.3% (3.8%, 7.2%) and 0.3% (0.1%, 0.8%). shows the estimated crude and age-adjusted prevalence of gout, hyperuricemia, and categories of renal impairment that were comparable to the estimates from the 2007–8 NHANES cycle. The present analyses estimate that there were 7.58 million with gout in the US in 2009–10. Among these, there were 5.86 million people with eGFR <90 ml/min/1.73 m2. shows the increasing prevalences of gout and hyperuricemia with worsening renal function. The age standardized prevalence of gout was 2.9% (1.6%, 5.1%) among those with no renal impairment compared to 24.0% (20.3%, 28.2%) among those with moderate or severe renal impairment. Similarly there was a five-fold increase in prevalence of hyperuricemia among those with severe renal impairment compared to those with no renal impairment. shows the increase in mean serum urate with decrease in renal function among men and women. Urate concentrations among men were higher in the subset of the population with higher eGFR but the gender differences disappeared at low eGFR (<30 ml/min/1.73 m2).
Overall prevalence of gout, hyperuricemia and renal impairment in the NHANES 2009–10.a
Age-standardized prevalence rates of gout and hyperuricemia.
Bivariate association between estimated glomerular filtration rate and serum urate concentrations.
The prevalence rates of gout were lower in women than in men (). Among those with no renal impairment, Hispanics had a prevalence rate almost a tenth of Whites and African Americans. This difference prevailed in all categories of renal impairment except in the severe renal impairment category, where there were too few Hispanics with gout for a reliable estimate. Among the other ethnic categories, increasing severity of renal impairment was associated with higher prevalence rates of gout, except in those with diabetes and those with metabolic syndrome.
Age-standardized prevalence (%) of gout by renal glomerular function.a
Among those with gout, the age-standardized proportion of participants with normal renal function was 50% (43%, 58%) among men and 26% (16%, 38%) among women (). The proportion of people with severe renal impairment among women was more than 6 times the corresponding proportion among men. Such a gender disparity was not apparent with respect to hyperuricemia.
Distribution of severity of renal impairment among people with gout and hyperuricemia.a
Results of logistic regression models are shown in . A substantial increase in the odds ratios for hyperuricemia and gout was noted with lower eGFR. Information on alcohol consumption was only available for 3,600 participants. Additional adjustment for this variable in logistic regression did not change the findings. Results of the ordinary least squares (OLS) models are shown in . In multivariable OLS regressions that adjusted for the same variables as in the logistic regression models, each standard deviation decrease of eGFR (27.6 ml/min/1.73 m2) was associated with a 0.64 mg/dL (0.56, 0.72) increase of serum urate. Analyses including alcohol consumption information and excluding the 22 NHANES participants who reported utilizing dialysis did not alter these results.
Results of logistic regression analyses of the risk for gout and hyperuricemia by renal impairment.a
Results of ordinary least squared regression analyses of the relationship between serum urate and eFGR.a