Utility of the China-PAR Score in predicting secondary events among patients undergoing percutaneous coronary intervention

To the Editor: Coronary artery disease (CAD) is the leading cause of mortality worldwide. Effective treatments, such as percutaneous coronary intervention (PCI), have reduced the mortality of patients with CAD and have transitioned most of them to the chronic phase of cardiovascular disease (CVD). Consequently, there is a high demand for secondary prevention strategies to reduce adverse outcomes of CAD. Risk assessment can quantify the residual risk, identify patients at high secondary risk, and facilitate informed preventive treatment decisions.[1] However, the existing scoring systems for secondary risk, such as the Global Registry of Acute Coronary Events (GRACE) and Patterns of Non-Adherence to Anti-Platelet Regimen in Stented Patients (PARIS) scores, usually involve certain clinical indicators that are unfamiliar to the public and demonstrate poor performance when applied to Chinese patients; Thus, it limited their utility, especially as self-assessment tools. The Prediction for Atherosclerotic Cardiovascular Disease Risk in China (China-PAR) score is a guideline-recommended pragmatic tool for primary prevention using several conventional risk factors.[2] The risk factors related to primary CVD are likely implicated in the recurrent events because of the same underlying pathophysiology of atherosclerosis. Furthermore, primary and secondary CVDs prevention shares similar risk prevention strategies.[3] Hence, the risk score for primary prevention may be speculated to be extended to secondary prevention. Nevertheless, the effectiveness of the China-PAR score in risk assessment for secondary events among patients with CAD is unclear. This study aimed to evaluate the utility of the China-PAR score in predicting long-term secondary events in 10,724 Chinese patients with CAD undergoing PCI at Fuwai Hospital in Beijing, China in 2013. All patients were followed up according to the standard protocol until 2019, and 9843 patients (91.8%) completed the 5-year follow-up survey. After excluding 482 patients with incomplete data for the China-PAR score assessment, 9361 patients were included in the final analysis. This study was approved by the Institutional Review Board at Fuwai Hospital (No. 2013‑449), and written informed consent was obtained from each patient. Data on admission were acquired from inpatient medical records, including demographic characteristics, lifestyle information, medical history, and anthropometric measurements. The follow-up surveys were conducted using standard protocols. Endpoints were obtained from medical records and telephone interviews and adjudicated by two independent cardiologists, with discrepancies resolved by an additional cardiologist. The endpoints included all-cause death and major adverse cardiovascular and cerebrovascular events (MACCEs). MACCEs were defined as the occurrence of myocardial infarction, stroke, stent thrombosis, or cardiac death. The China-PAR score is a sex-specific tool for assessing the 10-year risk of first CVD events in the Chinese population using age, residence, waist circumference, systolic blood pressure, total cholesterol, high-density lipoprotein cholesterol, current smoking, antihypertensive treatment, diabetes mellitus, and family history of CVD.[4] Considering the unavailability of data on waist circumference and family history of CVD in our study, the score was revised using the derivation cohort in the China-PAR study that included body mass index and family history of CAD instead. When the revised China-PAR score was applied to the derivation cohort, C-indices of 0.792 (95% confidence interval [CI]: 0.772–0.812) and 0.807 (95% CI: 0.783–0.832) for male and female participants, respectively, were similar to those of the original score (0.794 and 0.811).[4] Moreover, the difference between the original and revised scores among the derivation cohort was 0, with a median (5th and 95th percentile) of 0.03% (−1.02%, 1.03%). The revised China-PAR score was calculated for each patient with CAD and divided into four categories, including low (<5.0%), moderate (5.0–<10.0%), high (10.0–<20.0%), and very high (≥20.0%) according to the guideline.[2] Cox proportional hazard models were used to estimate hazard ratios (HR) and 95% CIs for secondary events associated with the revised China-PAR score. Dose–response relationships were assessed using the restricted cubic spline analyses with five knots at scores of 5%, 10%, 15%, 20%, and 25%, with 5% as the reference. C-indices were computed to estimate the discrimination capacity of the revised China-PAR score on the endpoints. Subgroup analysis was conducted to assess the association between the endpoints and the revised China-PAR score stratified by previous CAD, including myocardial infarction, PCI, and coronary artery bypass grafting. Differences across subgroups were evaluated using P values for the interactions of previous CAD and the score categories. Sensitivity analysis was performed by excluding patients who experienced the endpoints within the first month or 3 months after PCI. All analyses were conducted using the SAS statistical package (version 9.4; SAS Institute, Inc., Cary, NC, USA). Among the 9361 patients with CAD undergoing PCI, the average age was 58.4 years, and 77.4% were males. According to the revised China-PAR score, there were 2259 (24.1%), 2929 (31.3%), 3143 (33.6%), and 1030 (11.0%) patients with low, moderate, high, and very high scores, respectively. During an average follow-up of 4.9 years, 365 all-cause deaths and 1048 MACCEs were identified, with event rates of 7.9 and 23.7 per 1000 person-years, respectively. Supplementary Table 1, https://links.lww.com/CM9/C296 presents the HRs and 95% CIs for the associations between the endpoints and the revised China-PAR score. The risk of all-cause death and MACCE gradually increased with elevated score. Compared with patients who underwent PCI and had a low score, the HRs of all-cause death were 1.39 (95% CI: 0.96–2.01), 2.44 (95% CI: 1.74–3.41), and 4.95 (95% CI: 3.46–7.07) for patients with moderate, high, and very high score, respectively. The corresponding HRs of MACCE were 1.25 (95% CI: 1.03–1.50), 1.78 (95% CI: 1.49–2.12), and 2.28 (95% CI: 1.85–2.82), respectively. The associations between MACCE subtypes and revised China-PAR score were further estimated. The score was significantly and positively related to stroke, stent thrombosis, and cardiac death, but not to myocardial infarction [Supplementary Table 2, https://links.lww.com/CM9/C296]. Similar findings were observed among male and female patients [Supplementary Table 3, https://links.lww.com/CM9/C296]. However, most HRs for males were higher than those for females across the score categories. The dose–response curves showed that the HRs for all-cause death and MACCE increased dramatically with the revised China-PAR score, indicating monotonic and linear dose–response relationships [Supplementary Figure 1, https://links.lww.com/CM9/C296]. Furthermore, linear dose–response relationships were observed for myocardial infarction, stent thrombosis, and cardiac death, whereas a non-linear relationship was observed for stroke [Supplementary Figure 2, https://links.lww.com/CM9/C296]. Male patients had similar dose–response relationships with all the patients. Notably, there was no dose–response relationship for myocardial infarction and a linear relationship for stroke in female patients. The revised China-PAR score had potential discrimination ability for all-cause death with a C-index of 0.661 (95% CI: 0.632–0.689), which was comparable to that of the GRACE score (0.671, 95% CI: 0.641–0.700) and better than that of the PARIS score (0.598, 95% CI: 0.569–0.627). To some extent, it could also predict MACCEs with a C-index of 0.588 (95% CI: 0.571–0.606), which was slightly higher than those of the GRACE (0.572, 95% CI: 0.553–0.590) and PARIS scores (0.570, 95% CI: 0.552–0.587). Moreover, it could potentially discriminate all MACCE subtypes except myocardial infarction [Supplementary Table 4, https://links.lww.com/CM9/C296]. Interestingly, the performance among males was slightly better than that among females. Subgroup analysis indicated that the HRs and C-indices of the revised China-PAR score related to the endpoints in patients without previous CAD were slightly higher than those in patients with previous CAD, although the difference was not statistically significant [Supplementary Tables 5 and 6, https://links.lww.com/CM9/C296]. Sensitivity analysis revealed the robustness of the associations after excluding incident endpoints within the first month or 3 months after PCI [Supplementary Table 7, https://links.lww.com/CM9/C296]. A previous study demonstrated the poor performance of the Framingham risk score for first CAD in predicting long-term recurrent events among patients who underwent PCI.[5] By contrast, our study revealed the potential discrimination ability of the revised China-PAR score for secondary events, which might be explained by its derivation from contemporary cohorts involving more comprehensive CVD events.[4] Moreover, its performance was equal to or better than the existing scores for secondary prevention, which emphasized more on short-term risk within one year. Nevertheless, short- and long-term secondary outcomes exhibit different underlying pathophysiology and lesion sites. Additionally, the free risk assessment platform (www.cvdrisk.com.cn) and smartphone-based tools using more easily obtainable risk factors further improved accessibility of the China-PAR score to clinicians, health staff, and the general population. As it is known, intensive pharmacotherapy for CAD, such as dual antiplatelet and lipid-lowering therapies, has a dramatic impact on coronary events. However, these factors are not considered in the China-PAR score. The C-indices of revised China-PAR for secondary events might not have exceeded 0.7 for this reason. Consequently, an updated China-PAR score is required for more precise secondary prevention by including intensive pharmacotherapy. The clinical implications of the China-PAR score are highlighted by the lack of a published decision support tool to guide self-assessment in patients with CAD. The China-PAR score eliminates major obstacles in the implementation of risk-based secondary prevention between clinicians and patients. Thus, it could be used to direct secondary preventative measures by clinicians, understand their risk, and enhance the adherence to the therapeutic decisions of physicians, especially in resource-constrained settings. Our study extended the utility of the China-PAR score from primary to secondary prevention, provided novel ideas for the risk assessment of secondary prevention of CAD, and further bridged the gap between clinicians and patients using this pragmatic tool. It also provides examples of other primary prevention scores. The major strength of this study is the large sample size and a 5-year follow-up. This study had several potential limitations. First, the China-PAR score had to be revised before evaluating its utility because waist circumference and family history of CVD were unavailable in this study. This might have caused misclassification bias, but the effect should be small because the difference between the original and revised China-PAR scores approached zero. Second, this was a tertiary-hospital single-center study on patients who underwent PCI, and more studies from multiple centers and diverse regions with longer follow-up periods are needed to validate our findings and enhance their generalizability. Finally, the applicability of the China-PAR score in patients with CAD in other countries or for secondary prevention of stroke and other CVDs remains unclear. However, further studies are required to confirm this hypothesis. In conclusion, the China-PAR score has the potential to discriminate long-term secondary events in patients who underwent PCI. It can be used as a pragmatic tool for the secondary prevention of CAD by both clinicians and patients. It can identify patients with the greatest potential benefit, enhance risk-based secondary prevention, and further decrease the CVD burden in China. Acknowledgements We acknowledge the patients and staffs for their important participation and contribution. Funding This work was supported by the Noncommunicable Chronic Diseases-National Science and Technology Major Project (Nos. 2023ZD0504000 and 2023ZD0503500), the National Key Research and Development Program of China (Nos. 2016YFC1301300, 2016YFC1301301, 2017YFC0211706, 2018YFE0115300, and 2023YFC2506703), the National Natural Science Foundation of China (Nos. 82473720 and 81770365), Chinese Academy of Medical Sciences (CAMS) Innovation Fund for Medical Sciences (No. 2021-I2M-1-010), and the National High Level Hospital Clinical Research Funding (No. 2022-GSP-GG-2). Conflicts of interest None.