We read with great interest the article by Li et al. (1) on the association of serum uric acid (SUA) with all-cause and cardiovascular mortality in adults with diabetes. Using nationally representative data from the U.S. National Health and Nutrition Examination Survey (NHANES), 1999–2018, this study enrolled a total of 7,101 patients with diabetes, with follow-up through 31 December 2019 (median follow-up 7.3 years). Based on a prospective cohort design, this study demonstrated that higher levels of SUA were associated with higher risks of all-cause and cardiovascular disease (CVD) mortality in diabetes.
We endorse the findings of Li et al. (1). Several previous Mendelian randomization studies have shown a causal relationship between SUA and cardiovascular events (2). In addition, lower SUA levels is associated with improved cardiovascular outcomes (3). The study by Li et al. (1) further expands the understanding of SUA and the potential health risks of elevated SUA. However, we have some concerns about the methodology of the article.
First, the definition of diabetes in the study was not accurate enough, owing to the lack of oral glucose tolerance test results. This had a substantial impact on the prevalence of diabetes in NHANES (4), which in turn affected the study inclusion population. The vital role of oral glucose tolerance tests in the comprehensive monitoring of diabetes should be emphasized. Second, there were some issues with the definitions of covariates. 1) Dyslipidemia in this study did not include LDL cholesterol, which is generally considered to be more associated with cardiovascular outcomes than triglyceride and HDL cholesterol and has received the most attention. 2) The definition of hypertension did not include self-reported history of hypertension, which identified a significant number of hypertensive patients in NHANES. 3) We have concerns with the definition of CVD. In NHANES, CVD was defined more as a composite of five (rather than two) cardiovascular events (congestive heart failure, coronary heart disease, angina pectoris, heart attack, and stroke) (5). A sensitivity analysis that excludes participants with CVD at baseline also may help to examine the robustness of the findings. Third, in the stratified analysis, multiplicity adjustments should be performed to control the type I error rate (the false-positive rate), and the stratified analysis in this study should be considered exploratory.
In conclusion, the authors have provided valuable and important findings. However, we believe that addressing the above issues will make this study more convincing and rigorous.
Article Information
Duality of Interest. No potential conflicts of interest relevant to this article were reported.
