The current study population was obtained from the UK Biobank dataset, consisting of 502,409 participants aged 40 to 69, recruited from across the UK between 2006 and 2010. All participants provided written informed consent at the time of recruitment. During the initial assessment, participants completed questionnaires regarding sociodemographic characteristics and health behaviors. They also underwent interviews conducted by trained healthcare professionals to gather health-related information and provided biological samples. In this study, a total 39,704 participants were excluded due to missing SUA level measurements (n = 33,001), having a history of VHD (n = 6,636) prior to recruitment, or being diagnosed with congenital heart disease (n = 67). The participant selection process is illustrated in Supplementary Figure 1. The UK Biobank received ethical approval from the Northwest Multicenter Research Ethics Committee (reference No. 16/NW/0274). The present research using the UK Biobank Resource was approved under Application Number 67855.

Information on demographics, lifestyle habits, and medical history was collected at enrollment in the UK Biobank through self-administered touchscreen questionnaires and personal interviews. During the interviews, trained staff used standardized procedures to measure participants’ height, weight, and waist circumference. Body mass index (BMI, kg/m²) was calculated by dividing weight (kg) by the square of height (m²). Obesity and abdominal obesity statuses were determined based on the World Health Organization (WHO) classification and the International Diabetes Federation (IDF) consensus report. To minimize errors arising from racial differences, race-specific criteria were applied [17,18]. Smoking status was categorized as non-smoker, former smoker, and current smoker. Physical activity was divided into two groups: those reporting engaging in moderate activity for at least five days a week (classified as active) and those not meeting this criterion (considered inactive). Dietary patterns were defined based on recommended guidelines for cardiovascular health. If participants failed to meet more than half of the recommended items on the food frequency questionnaire, their diet pattern was considered unfavorable [19]. Alcohol consumption was categorized by the frequency of alcohol intake: never or special occasions only, 1–2 times per week, 3–4 times per week, and daily or almost daily. More detailed information on the biological sampling procedures can be found in related publications [20]. The components of metabolic syndrome were determined in accordance with the criteria specified in the IDF consensus report [17]. Baseline metabolic and cardiovascular diseases were defined through three sources: 1) self-reported information obtained during in-person interviews or via touchscreen questionnaires during enrollment, 2) diagnostic or procedure codes found in the electronic health records database linked to hospital admission records, and 3) the initial occurrence of the comorbidity in the health outcomes database, which is interconnected with hospital in-patient records and primary care records.

The primary exposure variable was SUA level. SUA was measured using the Uricase PAP analysis conducted on a Beckman Coulter AU5800 instrument (Beckman Coulter, Brea, CA, USA). It was noted that later measurements of SUA collected in 2012–2013 were strongly correlated with the baseline level [21]. SUA level was categorized into clinically interpretable groups defined as normal (6.0 mg/dL), high (6.0–8.9 mg/dL), and very high (≥ 9.0 mg/dL). Although universally accepted SUA thresholds for cardiovascular outcomes in the general population have not been established, these categories were informed by widely recognized gout research. The threshold of 6.0 mg/dL reflects the established therapeutic target in gout management, while ≥ 9.0 mg/dL corresponds to levels associated with a markedly increased risk of gout onset and recurrence in large cohort studies [22,23].

We confirmed the prevalence of VHD at enrollment using diagnostic International Classification of Diseases (ICD)-9 or 10 codes and self-report information. Cases of incident VHD during the follow-up period were ascertained based on first occurrence of disease and on hospitalization records. Congenital heart diseases were also identified through the database of first occurrences, which was likewise encoded using ICD-10 diagnostic codes. The classification of VHD outcomes based on ICD codes was informed by previously published UK Biobank research [24]. Detailed information, including the definitions of mitral and aortic VHD and aortic stenosis, is provided in Supplementary Table 1.

To investigate differences in baseline characteristics between groups, we conducted t-tests and chi-squared tests. Continuous variables were reported as mean ± standard deviation, while categorical variables were presented as percentages. The primary exposure variable, SUA level, was analyzed both as a continuous and categorical variable. Multivariable-adjusted Cox proportional hazard regression models were constructed to assess the association of SUA level with the risk of future VHD occurrence. Multicollinearity among covariates included in the multivariable models was assessed using generalized variance inflation factors (GVIFs) (Supplementary Table 2). To examine changes in VHD risk associated with changes in SUA level, restricted spline curves were adjusted for age, sex, and race, evaluated using the rms R package. The Cox proportional hazard regression models were adjusted for potential confounding factors related to overall lifestyle, metabolic indicators, and medications. Individuals with missing data were excluded from each model (Supplementary Table 3). To further assess the robustness of the findings, the multivariable Cox regression models were subsequently reconstructed after excluding participants using diuretics or urate-lowering therapy at baseline. Subgroup analyses were performed based on age, gender, race, obesity, metabolic disorders, and the presence of cardiovascular diseases. To assess the synergistic effect of obesity and abdominal obesity with increased SUA, we conducted a joint effect analysis. Additionally, to examine the impact of obesity on the risk of VHD based on SUA levels, we stratified VHD risk by obesity and abdominal obesity within SUA level categories. Statistical analyses were performed using R (version 3.9.0). The significance level was set at p <0.05, and values below this threshold were considered statistically significant.

After exclusion of individuals with missing SUA levels, baseline VHD, and congenital heart disease, a total of 462,705 participants were included in the analysis. Of those, 340,793 (73.7%) had SUA levels below 6.0 mg/dL, 118,861 (25.7%) had levels ranging from 6.0 to 8.9 mg/dL, and 3,051 (0.7%) had levels of 9.0 mg/dL or higher. The mean age of participants was 56.0 years, with females accounting for 54.2% of the study population. The baseline characteristics of the participants categorized by SUA level are presented in Table 1. Patients with higher SUA level exhibited older age, a higher proportion of males, and a higher prevalence of obesity. Hyperuricemia was also associated with elevated systolic blood pressure, reduced estimated glomerular filtration rate, and increased HbA1c. Accordingly, the prevalence of metabolic diseases such as hypertension, dyslipidemia, and type 2 diabetes demonstrated an upward trend with increasing SUA level.

Among the 462,705 participants without a baseline diagnosis of VHD, a total of 14,573 subjects (3.1%) developed VHD during a mean follow-up period of 12.3 years. The incidence of all VHD demonstrated a significant correlation with SUA level. Using SUA < 6.0 mg/dL as reference group, we observed a proportional rise in the hazard ratio of VHD with each 1.0 mg/dL increase in SUA level (Fig. 1A, Table 2, Supplementary Table 4). In the sensitivity analysis excluding participants using diuretics or urate-lowering therapy (n = 9,487), the association between SUA level and incident VHD remained consistent, demonstrating similar effect sizes and statistical significance across all models (Supplementary Table 5).

When we categorized participants into groups with normal (< 6.0 mg/dL), high (6.0–8.9 mg/dL), and very high SUA (≥ 9.0 mg/dL), we observed a stepwise increase in VHD risk that corresponded to the SUA categories. The association between SUA level and VHD risk remained consistent after adjusting for demographic characteristics, lifestyle behaviors, metabolic parameters, and treatment for metabolic diseases.

The risk of VHD was assessed separately by VHD subtype. Mitral and aortic VHD, as well as aortic stenosis, showed associations with SUA level. Following adjustment for potential confounding factors, the hazard ratio for mitral VHD exhibited the most notable statistical significance.

In the restricted cubic spline analysis adjusted for age, sex, and race, the lowest estimated risk of incident VHD was observed at SUA levels around 4.0–4.9 mg/dL (Supplementary Fig. 2). Based on this observation, we performed an additional multivariable-adjusted analysis using SUA 4.0–4.9 mg/ dL as the reference group. In this analysis, both the lower SUA group (< 3.0 mg/dL) and those with higher SUA levels demonstrated elevated VHD risk (Supplementary Table 6).

To address the potential confounding effects of age, sex, obesity, and metabolic profile on the association between SUA level and VHD, we performed subgroup analyses assessing VHD risk based on SUA level. These analyses involved stratification by age, sex, obesity status, metabolic disorders, and cardiovascular disease (Fig. 1B, Table 3).

In line with our overall findings, subgroups stratified by age, sex, and obesity status demonstrated a consistent and significant increasing trend in VHD risk with higher SUA level. Moreover, the association between SUA level and VHD risk remained consistent across subgroups defined by the presence of hypertension, dyslipidemia, type 2 diabetes, metabolic syndrome, and cardiovascular disease. Significant interactions between SUA level and clinical profiles were found only within subgroups stratified by sex and type 2 diabetes (p for interaction < 0.001 and 0.02, respectively).

Obesity has been identified as a causal risk factor for the development of numerous cardiovascular conditions, including aortic stenosis, heart failure, and coronary artery disease [25,26]. To access the impact of weight management on incident VHD, we investigated VHD risk based on obesity status as determined by BMI and waist circumference, which is a critical anthropometric parameter (Fig. 2, Table 4).

In the overall population, both BMI and abdominal obesity showed significant correlation with incident VHD. Given the documented association between obesity and hyperuricemia in previous studies, we analyzed the risk of VHD based on obesity status in different SUA categories. Among participants with SUA levels below 9.0 mg/dL, we consistently observed an upward trend in VHD risk with higher BMI and abdominal obesity, which aligns with the findings observed in the overall population. However, this trend was not apparent among participants with SUA levels exceeding 9.0 mg/dL, indicating that VHD risk is not stratified based on obesity status in those with very high SUA levels. In formal interaction analyses, the association between BMI and incident VHD differed significantly accordingly to SUA category (p for interaction < 0.0001). A similar pattern was observed for abdominal obesity (p for interaction = 0.008), suggesting effect modification by SUA level.