Key result
SUA/eGFR ratio improves 3-month risk prediction for hard cardiac events in STEMI.
Why the study?
Early risk stratification is paramount in STEMI, and although various uric acid-based indices exist, their comparative prognostic value and potential non-linear relationships with outcomes remained unclear.
Does the SUA/eGFR ratio improve prediction of three-month hard cardiac events compared to other uric acid-based indices and a baseline clinical model in patients with STEMI?
Observational (n=39,941)
Does the SUA/eGFR ratio improve prediction of three-month hard cardiac events compared to other uric acid-based indices and a baseline clinical model in patients with STEMI?
Answers are grounded in this paper and cite their sources.
In STEMI patients, the SUA/eGFR ratio outperforms other uric acid-based indices in predicting 3-month hard cardiac events and exhibits a J-shaped risk relationship, significantly enhancing prognostic accuracy when added to baseline clinical models.
Background Early risk stratification is paramount for ST-segment elevation myocardial infarction (STEMI). While various uric acid-based indices exist, their comparative prognostic value remains unclear. Purpose We aimed to identify the optimal index and explore potential non-linear relationships. Methods We analyzed 39,941 STEMI patients from a large-scale real-world database. The primary endpoint was three-month hard cardiac events (cardiac death and re-infarction). The secondary endpoint was a composite of cardiac death, re-infarction, and non-fatal stroke (3-point major adverse cardiovascular events [MACE]). Phase one involved a head-to-head comparison of five indices: serum uric acid (SUA), SUA/estimated glomerular filtration rate (eGFR), SUA/albumin (UAR), SUA/high-density lipoprotein (HDL) (UHR), and SUA/creatinine. Phase two assessed the incremental predictive value via net reclassification index (NRI), integrated discrimination improvement (IDI), and decision curve analysis (DCA). Restricted cubic splines (RCS) were employed to examine nonlinear associations. Results Among all tested indices, SUA/eGFR demonstrated the highest discriminatory power for three-month hard endpoints (area under the curve [AUC]: 0.658 vs. 0.550–0.620 for others, Figure 1B) and three-month three-point MACE (AUC: 0.644 vs. 0.550–0.610 for others, Figure 1C). Notably, restricted cubic spline analysis unveiled a significant J-shaped association between SUA/eGFR and hard endpoints (P for non-linearity < 0.05), indicating that both excessively low and high ratios confer elevated risk—a finding that reframes the traditional linear risk paradigm (Figure 1A). Adding SUA/eGFR to the clinical baseline model significantly improved discrimination (AUC increased from 0.747 to 0.752, P < 0.001, Figure 2A) and yielded meaningful risk reclassification (NRI: 0.98%, 95% CI 0.38%–1.56%, P < 0.05; IDI: 0.57%, P < 0.05). Decision Curve Analysis further confirmed that the SUA/eGFR-enhanced model provided a higher net benefit across clinically relevant threshold probabilities (Figure 2B). Conclusion The SUA/eGFR ratio is the superior uric acid-based index for predicting early hard cardiac events in STEMI patients. Its distinctive J-shaped relationship with outcomes—challenges conventional linear risk assessment and identifies a vulnerable window at both extremes. Integrating this simple, readily available ratio into established clinical models significantly enhances prognostic accuracy and supports personalized risk management at the bedside.
No takes yet. Share an insight, caveat, or question.
Ren et al. (2026) conducted an observational in ST-segment elevation myocardial infarction (STEMI) (n=39,941). Serum uric acid to estimated glomerular filtration rate ratio (SUA/eGFR) vs. Other uric acid-based indices (SUA, SUA/albumin, SUA/HDL, SUA/creatinine) and clinical baseline model was evaluated on Three-month hard cardiac events (cardiac death and re-infarction) (AUC 0.658, p=<0.001). The SUA/eGFR ratio showed the highest discrimination for 3-month hard cardiac events in STEMI and improved baseline model prediction.
Synapse has enriched 10 related papers. Consider them for comparative context: