Age–D‐Dimer–Albumin (ADA) Score as a Predictor of Long‐Term Mortality in Coronary Artery Disease: A Prospective Cohort Study

Objectives

The Age–D‐dimer–Albumin (ADA) score, which integrates age, coagulation activity, and nutritional status, was originally designed to predict thrombotic risk but has not been systematically validated in coronary artery disease (CAD). This study aimed to assess its prognostic value for long‐term mortality in a large prospective CAD cohort.

Methods

We conducted a secondary analysis of the PRACTICE cohort, including 10,613 CAD patients enrolled from 2016 to 2021. The ADA score was calculated as −0.5 × Alb (g/L) + 0.4 × D‐dimer/100 (ng/mL) + 0.3 × age (years). Receiver operating characteristic analysis identified an optimal cutoff of −13.5, and patients were categorized into low‐ADA (< −13.5, n  = 6733) and high‐ADA (≥ −13.5, n  = 3880) groups. The primary endpoints were all‐cause mortality (ACM) and cardiac mortality (CM).

Results

During a median follow‐up of 24 months, 340 ACM and 258 CM events occurred. High‐ADA patients showed significantly higher risks of ACM (6.0% vs. 1.6%) and CM (4.5% vs. 1.2%) compared with low‐ADA patients (both p < 0.001). After adjustment for conventional risk factors, ADA remained independently associated with ACM (HR 1.58, 95% CI 1.13–2.21; p = 0.008) and CM (HR 1.62, 95% CI 1.10–2.38; p = 0.015) but not with ischemic events (MACEs or MACCEs). Restricted cubic spline (RCS) analysis demonstrated an approximately linear positive dose–response relationship between the ADA score and the risks of ACM and CM. Subgroup analysis revealed a distinct ACS subtype‐specific prognostic value of the ADA score, with its independent predictive efficacy for long‐term mortality being significantly better in the ACS subgroup than in the stable coronary artery disease (SCAD) subgroup. ROC curve comparison between the ADA score and its individual components demonstrated significantly better prognostic performance only over Alb, with no significant differences observed for age or D‐dimer.

Conclusions

The ADA score is independently associated with long‐term ACM and CM in CAD patients, with an approximately linear positive dose–response relationship by restricted cubic spline analysis. Its prognostic efficacy is significantly superior in the ACS subgroup than in the SCAD subgroup, serving as a simple, accessible tool for risk stratification. Further multicenter studies are needed to validate its clinical utility.

Trial Registration: ClinicalTrials.gov identifier: NCT05174143