The PubMed, Web of Science, Cochrane Library, and Embase databases were searched for relevant articles published by September 2025, with the detailed search strategy available in Supplementary Table S1. This study also referred to the reference list of published papers and literature reviews to supplement other related research.

Our study has been registered in the International Prospective Register of Systematic Reviews (PROSPERO). (https://www.crd.york.ac.uk/PROSPERO/recorddashboard, identifier CRD420251240886). As detailed in Supplementary Table S3 and Supplementary Table S4, this meta-analysis was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement (23).

The inclusion criteria were as follows: (1) participants: adult; (2) comparison: different levels of SUA; (3) outcomes: evaluated the relationship between the SUA levels and the risk of CHD; (4) types of studies: epidemiological studies; (5) reports of the hazard ratio (HR) estimates with the corresponding 95% confidence intervals (CIs) or having sufficient information to calculate them and provided useful data for the dose−response analysis; and (6) having clear sample size information.

The exclusion criteria were: (1) they were reviews, meta-analyses, case reports, case series, commentaries, letters and congress abstracts; (2) they were not conducted with humans; (3) no hazard ratio were reported or could not be extracted; (4) they were not related to SUA or CHD; (5) they were not published between 2015 to 2025; and (6) they used languages other than English.

Two researchers independently carried out the full process of our study from the screening for included studies to data analysis. The final articles included were selected by reviewing the title, abstract, and full text, with any disagreements resolved through discussion or by involving a third author if needed. When the article was unavailable or more information was needed for the analyses, an email was sent to the author requesting the article or additional details. The identification of duplicated manuscripts was done manually.

Data from all eligible studies were meticulously extracted by two authors based on the inclusion criteria. If two authors could not reach a consensus, a third author was consulted to evaluate the article. The following data were extracted: name of first author, year of publication, country, study design, number of cases and total population, gender, outcome type, SUA levels, variables of adjustments, hazard ratio (HR) with 95% confidence interval (CI) from the most adjusted model.

The risk of bias in the included articles was assessed by two authors utilizing Newcastle-Ottawa Scale (NOS) (24), disagreements were discussed and solved by the third author. The NOS rates stars based on three main criteria: selection (maximum of 4 points), comparability (maximum of 2 points), and outcome(maximum of 3 points), with a total of 8 items. Each study was assigned a NOS score (0–9), with a score of 7 or more indicating “high quality” and a score of less than 7 indicating “low quality”.

The strength of the association between SUA levels and the risk of CHD was assessed by using HR with 95% CI. Statistical heterogeneity was assessed using visual inspection of forest plots and statistical methods. The heterogeneity was evaluated by using the I² statistic, which can quantify the proportion of total variation among studies that is caused by heterogeneity rather than by chance factors. Although it is generally believed that an I² value greater than 50% indicates significant heterogeneity, we acknowledge that this threshold varies depending on the research context (25). Therefore, we also employed the Q test (with p<0.05 indicating statistical significance) for a comprehensive assessment to quantify the dispersion of the true effect. Additionally, we conducted subgroup analysis to verify the primary sources of heterogeneity and evaluate the stability of the combined results. The subgroups were defined as follows: gender(male, female), sample size(≤3000, >3000) the publication year(before2020, after 2020) and outcome type(broader CHD, myocardial infarction only). A dose-response meta-analysis was conducted by performing a trend analysis on the aggregated dose-response data using the generalized least squares trend estimation method. A linear relationship between SUA levels and the risk of CHD was initially assumed. Then, dose-response analyses for each study were performed using the glst command, and all dose-response data were pooled using restricted maximum likelihood. Publication bias was not assessed because fewer than ten studies were included in the meta-analysis, which renders statistical tests (e.g., Egger’s test) underpowered and unreliable. To evaluate the robustness of the pooled results, sensitivity analyses were performed using two complementary methods: first, a comparison of statistical models, in which the random-effects model was compared with a fixed-effects model to see if the results changed based on how we handled study differences; second, sequential exclusion of any single study method was used to assess its disproportionate influence on the overall estimate. Stata 12.0 was used for each statistical analysis(Stata Corp, College Station, Texas).

A flow chart was developed to illustrate the literature search and screening process (Figure 1). Based on the established search strategy, we conducted a search in four databases: PubMed, Web of Science, Cochrane Library, and Embase. A total of 9,585 articles were retrieved. After eliminating duplicate literature, 7,664 articles remained. According to the established inclusion and exclusion criteria, 7,419 articles were excluded after screening the titles and abstracts. Subsequently, 237 articles were further excluded after reading the full texts of the remaining documents. Finally, 8 cohort studies were included in this meta-analysis.

The main characteristics of the included studies are shown in Table 1. 8 cohort studies with 11, 009 CHD cases and 574, 815 participants were included in this meta-analysis. In general, the sample sizes of the included studies ranged from 203 to 63,014, while the publication year ranged from 2016 to 2023. Among these articles, one of them was from Europe, while the others were all from Asia(China).

The cohort studies were evaluated by NOS. All of the NOS scores of the included studies ranged from 8 to 9, with an average of 8.5, which indicated a good quality for this meta-analysis(Supplementary Table S2).

The heterogeneity test revealed a low level of heterogeneity (I² = 0.0%, p=0.557). Despite a low statistical heterogeneity, we employed a random-effects model for the primary analysis. This approach is more appropriate for observational epidemiology, as it accounts for inherent conceptual heterogeneity arising from differences in study design, population characteristics, and adjustment for confounders, which may persist even in the absence of significant statistical heterogeneity. The fixed-effects model results are also provided for comparison in a sensitivity analysis. We compared the highest and lowest levels of SUA to analyze the relationship between elevated SUA and the risk of CHD in eligible studies. After aggregating all the results (Figures 2, 3), there was a statistically significant association between elevated SUA and the risk of CHD (HR = 1.38, 95% CI = 1.30-1.47).

Subgroup analyses were carried out for the following variables: gender (male, female), sample size (≤3000, >3000), publication year(before 2020, after 2020) and outcome type (broader CHD, myocardial infarction only). Subgroup analyses by gender demonstrated a significant association between elevated SUA and CHD in males (HR = 1.32, 95% CI = 1.19-1.45) and females(HR = 1.46, 95% CI = 1.26-1.68) (Supplementary Figure S1). When stratified by sample size, both large sample subgroup (HR = 1.37, 95% CI = 1.28-1.47) and small sample subgroup (HR = 1.48, 95% CI = 1.26-1.73) revealed that elevated SUA increased the risk of CHD(Supplementary Figure S2). Analysis by publication year indicated a consistent association in studies published both before 2020 (HR = 1.30, 95% CI = 1.15-1.47) and after 2020 (HR = 1.41, 95% CI = 1.31-1.51) (Supplementary Figure S3). As for different outcome type, the summary HR for broader CHD was 1.41 (95% CI = 1.26-1.57), for the more specific outcome of myocardial infarction only, the pooled HR was 1.37 (95% CI = 1.25-1.50). Furthermore, no significant heterogeneity was observed across any of these subgroups.

The findings of the primary meta-analysis were robust across all pre-specified sensitivity analyses. The choice of meta-analytic model did not change the direction or statistical significance of the primary outcome. The pooled HR was 1.38 (95% CI = 1.30-1.47) under the random-effects model and 1.38 (95% CI = 1.30-1.47) under the fixed-effects model. The consistency between models strengthens confidence in the primary result (Table 2). In addition, sensitivity analysis using sequential exclusion of any single study method showed that the pooled effect estimate was not substantially altered by the removal of any single study (Figure 4).

7 studies (28–34) were further included in the dose-response meta-analysis to evaluate the dose-response relationship between different levels of SUA and the risk of CHD. Since the evaluation criteria for SUA differ between males and females, separate dose-response meta-analyses were conducted for males and females. The dose-response meta-analysis in males (Figure 5) indicated that there was a non-linear relationship between different SUA levels and the risk of CHD (P-non-linear < 0.05), and the higher the SUA level, the higher the risk of CHD. When the SUA level reached 324 μmol/L, the risk of CHD increased by 10% (HR = 1.10, 95% CI = 1.05-1.15, P < 0.05). The dose-response meta-analysis in females (Figure 6) suggested that there was a linear relationship between different SUA levels and the risk of CHD (P-linear < 0.05, P-non-linear > 0.05). The linear dose-response analysis found that each 94 μmol/L increase for SUA levels was associated with 10% higher CHD risk. The HR values at each SUA node for males and females are detailed in Supplementary Tables S5, S6.