Association between the serum uric acid-to-creatinine ratio and contemporaneous discordance between visual coronary diameter stenosis and quantitative flow ratio in patients with unstable angina: development and internal validation of an identification model
Year: 2026
Type: article
Study design: Observational Study
Abstract: Background Visual coronary diameter stenosis (DS) and quantitative flow ratio (QFR) can yield discordant contemporaneous classifications for intermediate coronary lesions. We evaluated the serum uric acid (SUA)-to-serum creatinine (SCr) ratio (SUA/SCr) and developed an internally validated model to identify visual DS–QFR classification discordance. Methods This retrospective observational study included 286 consecutive patients with adjudicated unstable angina who underwent coronary angiography and offline QFR analysis from March 2025 to March 2026. Discordance was defined as DS < 50% with QFR≤0.80 or DS≥50% with QFR > 0.80. Associations were estimated using multivariable logistic regression. Model performance was assessed using the area under the receiver operating characteristic curve (AUC), Brier score, calibration, nested patient-level bootstrap validation, and decision-curve analysis. Results Discordance occurred in 80 patients (28.0%). SUA/SCr showed a weak positive correlation with the Gensini score ( r s =0.242, P < 0.001). Each 1-standard-deviation increase in SUA/SCr was associated with discordance after full adjustment (odds ratio, 2.961; 95% confidence interval, 2.024–4.334; P < 0.001). In continuous-QFR analyses, the adjusted coefficient was −0.026 (95% confidence interval, −0.038 to −0.014) in the DS < 50% stratum and 0.008 (95% confidence interval, 0.002–0.014) in the DS≥50% stratum (interaction P < 0.001). Bootstrap-corrected AUCs were 0.756 for the extended model, 0.650 for the base model, 0.750 for the primary component model, and 0.752 for the eGFR-retaining component model. Paired AUC differences versus the SUA-only, primary component, and eGFR-retaining component models were 0.021 (95% confidence interval, −0.006–0.048), 0.006 (95% confidence interval, −0.019–0.031), and 0.004 (95% confidence interval, −0.021–0.029), respectively. At a threshold of 0.30, the extended model would prompt 36.7 additional QFR assessments while missing 9.4 discordant classifications per 100 patients. Conclusions Higher SUA/SCr was associated with contemporaneous visual DS–QFR classification discordance. The extended model had a higher corrected AUC than the base model, but its discrimination advantages over the SUA-only and component models were not statistically significant. These findings do not establish independently verified ischemia, treatment appropriateness, or clinical benefit.
Authors: Hongli Li(Writing – original draft, Writing – review & editing), Ying Zhao(Writing – original draft, Writing – review & editing), Jie Wang(Writing – original draft, Writing – review & editing), Xiao Feng(Writing – original draft, Writing – review & editing), Bintao Cao(Writing – original draft, Writing – review & editing), Dan Kou(Writing – original draft, Writing – review & editing), Nan Gao(Writing – original draft, Writing – review & editing), Wenjuan Zhang(Writing – original draft, Writing – review & editing)
Language: English
Subfield: Cardiology and Cardiovascular Medicine
Field: Medicine
Domain: Health Sciences
Keywords: serum uric acid to creatinine ratio, unstable angina, coronary artery disease, quantitative flow ratio, intermediate coronary lesions, Gensini score, EGFR, identification model, coronary angiography, risk stratification, internal validation, bootstrap validation, decision curve analysis, Brier score
Open Access status: gold
Estimated APC paid: $3,801
OpenAlex ID: W7235935477
