Prognostic value of the geriatric nutritional risk index for all-cause mortality in patients with non-ST-segment elevation myocardial infarction
J Gao, J X Wang, Y H Wang, Y Q Li, J X Chen, L Jiang, W W Tian, R Chu, Y LiuAbstract
Background
The Geriatric Nutritional Risk Index (GNRI), derived from serum albumin and body weight, has demonstrated prognostic value in cardiovascular diseases. However, in patients with non–ST-segment elevation myocardial infarction (NSTEMI), its joint stratification effect with established clinical risk markers and the mechanisms linking GNRI to adverse outcomes remain incompletely understood.
Purpose
To evaluate the prognostic value of GNRI for long-term all-cause mortality in NSTEMI, to assess its combined stratification effect with established clinical risk markers, and to explore potential mechanisms underlying this association.
Methods
This single-center prospective cohort study enrolled 1,686 consecutive patients with NSTEMI admitted to our hospital from March 2016 to December 2020. GNRI was calculated at admission and categorized as ≤98 or >98. The median follow-up was 5.36 years (IQR 4.17–6.57). The primary endpoint was all-cause mortality. Cox proportional hazards models assessed the association between GNRI and mortality. Restricted cubic spline analysis, joint stratification, and mediation analyses were performed to evaluate dose–response patterns, combined risk stratification, and potential mediating pathways.
Results
During follow-up, 249 deaths (14.8%) occurred. Restricted cubic spline analysis demonstrated a linear inverse association between GNRI and mortality risk. Compared with GNRI >98, GNRI ≤98 was associated with higher mortality (HR 1.32, 95% CI 1.01–1.72; P=0.039). In fully adjusted models, lower GNRI remained independently associated with mortality (per 1 SD decrease: HR 1.21, 95% CI 1.06–1.37; P=0.005). The association was consistent across subgroups without significant interactions (all P for interaction >0.05). Patients with low GNRI in combination with higher Killip class, elevated NT-proBNP, renal dysfunction, increased D-dimer, or obesity had the lowest long-term survival, indicating an additive risk pattern. Mediation analysis indicated partial mediation by NT-proBNP (28.26%), D-dimer (18.93%), and hs-CRP (15.66%).
Conclusions
Lower GNRI independently predicts long-term all-cause mortality in NSTEMI. The coexistence of low GNRI and high-risk clinical features identifies patients at particularly elevated risk. This association is partly explained by pathways related to cardiac dysfunction, inflammation, and coagulation, suggesting that nutritional risk reflects an integrated biological vulnerability.