Background/Purpose: Metabolic syndrome, chronic kidney disease and hypertension are known to be associated with hyperuricemia and gout. Hypertension is associated with incident gout. The Normative Aging Study indicates five times the number of people with gout have asymptomatic hyperuricemia, and it is not understood what factors can explain this difference. Using data from two prospective cohorts, the Framingham Heart Study (FHS) and the Atherosclerosis Risk in Communities Study (ARIC), we tested the hypothesis that traits representing the comorbid conditions of hyperuricemia and gout, given hyperuricemia at baseline, could discriminate those that subsequently develop gout from those that do not during the follow-up period.

Methods: A total of 3,415 and 1,040 participants from ARIC and FHS, respectively met the inclusion criterion in which serum urate concentration was greater than 7.0 mg/dL at the 1^st exam in ARIC or exam 1, 2, or 8 of the offspring cohort in FHS. Prevalent gout cases were excluded. With ARIC, incident gout was defined only if self report gout was reported at the fourth exam (ARIC). With FHS, incident gout was defined only if clinical diagnostic impression of gout was reported at any exam subsequent to the exam with serum urate concentration greater than 7.0. We extracted the following variables at the exam where hyperuricemia was recorded: Age, sex, race (ARIC), systolic blood pressure, glomerular filtration rate (ARIC), BMI, triglycerides, low-density lipoprotein-LDL, WaistHipRatio (ARIC), glucose and serum urate level (7-8, > 8 mg/dL). Logistic regression was used to model incident gout with a linear predictor comprising the explanatory variables.

Results: A total (proportion) of 167 (0.05) and 99 (0.095) incident gout cases occurred during follow-up. The mean (SD) age at baseline was 55 (5.8) and 39.5 (11.5) with 10 and 16 years of follow-up for ARIC and FHS, respectively. Age at first exam (P-value=0.047-ARIC; 0.033-FHS), with odds ratio-OR (95% confidence interval [CI]), 0.97 (0.94, 1.0)-ARIC, 0.98 (0.96, 0.999)-FHS, and serum urate level (P-value=<0.0001-ARIC, 0.056-FHS), with OR 3.42 (2.38, 4.93)-ARIC, 1.53 (0.985, 2.38)-FHS, were the only consistent explanatory variables for incident gout between the two data sets. Increased LDL was associated with incident gout in FHS (P-value=0.02, mean (SD) in gout vs no gout 145.2 (33.9) vs 138.2 (36.1), with OR = 1.01 (1.001,1.014)). The model R² was 6.3% and 2.8% for the ARIC and FHS datasets respectively.

Conclusion: The traditional risk factors for hyperuricemia and gout do not differentiate people with hyperuricemia that develop incident gout from those that do not during the follow-up period. A possible explanation is that most of the risk factors predict the development of hyperuricemia but not gout. Individuals presenting with serum urate levels > 8 mg/dL had the highest risk for developing gout in the future. The low explanatory ability of these models indicates that additional features, possibly molecular (e.g., genomic, metabolomic), behavioral (e.g. exercise, smoking, alcohol use) and use of medications (diuretics, aspirin, ACE inhibitors), may be important predictors of gout given the state of hyperuricemia.

« Back to 2017 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/the-common-risk-factors-for-hyperuricemia-and-gout-do-not-predict-incident-gout-once-hyperuricemia-is-established/