Objective: This research aimed to investigate the predictive value of the uric acid-to-serum albumin ratio (UAR) in establishing the severity and extent of coronary artery disease (CAD) with non-ST segment elevation myocardial infarction (NSTEMI) patients. Methods: A total of 402 patients (mean age 63.5 ± 11.6 years) were included in this retrospectively designed study. We compared Synergy between Percutaneous Coronary Intervention with Taxus and Cardiac Surgery Scores (SS) between low (≤22) and intermediate-high (>22) groups. The UAR, the neutrophil-to-lymphocyte ratio (NLR), and the C-reactive protein-to-albumin ratio (CAR) were evaluated and compared. Results: SS >22 were observed in 30.8% (n = 124) of the patients, and their UAR, NLR, and CAR were significantly higher. Three separate multivariate analysis models performed as the outcome of a reliable correlation between UAR, NLR, CAR, and consequently UAR (OR = 2.08; 95% CI 1.21–3.58; p = 0.008) and CAR (OR = 3.33; 95% CI 1.85–5.9; p < 0.001) reached significance but NLR (OR = 1.26; 95% CI 0.86–1.84; p = 0.20) clinically trended significance (not statistically). Model performance comparisons demonstrated that UAR is a better predictor regarding likelihood ratios (UAR, 60.95; NLR, 57.8; and CAR, 59.0). Conclusion: As a novel inflammatory marker, UAR independently predicted better outcomes than CAR and might be used reliably in prediction of the extent of CAD in NSTEMI patients.