**Background:** Coronary artery disease (CAD) accounts for 32.7% of global heart disease and remains a leading cause of morbidity and mortality. Despite advances in revascularization (PCI, CABG) and medical therapy, many patients remain at risk for mortality and MACE. Malnutrition has emerged as a potentially modifiable prognostic factor, affecting >40% of heart failure patients and >50% of CAD patients. The Prognostic Nutritional Index (PNI), calculated as serum albumin (g/L) + 5 × total lymphocyte count (×10⁹/L), is an objective marker of nutritional and immune status. While PNI has been validated in gastrointestinal surgery and various cancers, no prior systematic review had assessed its prognostic value specifically in CAD patients.
**Methods:** A comprehensive literature search of PubMed, Web of Science, Scopus, Embase, Open Gray, and Google Scholar was conducted from database inception to 1 November 2022. Search terms included 'myocardial infarction,' 'STEMI,' 'NSTEMI,' 'coronary artery disease,' 'acute coronary syndrome,' 'PCI,' 'CABG,' and 'prognostic nutritional index.' All study types reporting multivariable-adjusted hazard ratios for the association between PNI and mortality or MACE in CAD patients were eligible. Two reviewers independently screened 3,474 records; after removing duplicates, 1,372 articles underwent title/abstract screening, 20 full texts were assessed, and 15 studies met inclusion criteria. Data were pooled using random-effects meta-analysis. PNI was analyzed both as a categorical variable (low vs. high PNI based on study-specific cut-offs) and as a continuous variable (per-unit increase). Heterogeneity was assessed using the I² statistic. Subgroup analyses were conducted by region (Asia vs. Europe), sample size (>1,000 vs. <1,000), diagnosis (MI only vs. mixed), treatment (PCI only vs. mixed), and PNI cut-off (38, 44–47.1, 50–51). Risk of bias was assessed using the Newcastle-Ottawa Scale.
**Key Results:** Fifteen retrospective observational studies with a cumulative sample of 22,521 patients were included. Studies originated from China, Korea, Japan, Turkey, Spain, and Italy. Mean age exceeded 58 years across studies, with male predominance. Diabetes prevalence ranged from 19% to 54.9%; hypertension from 46.7% to 81.9%. PNI cut-offs varied from 38 to 50.7. Most studies scored 8 on the NOS (maximum 9). For mortality with PNI as a categorical variable (11 studies), low PNI was associated with significantly higher mortality (HR 1.67; 95% CI 1.39–2.00; I²=95%; p<0.00001). For MACE with PNI as a categorical variable (6 studies), low PNI predicted higher MACE (HR 1.57; 95% CI 1.08–2.28; I²=94%; p=0.02). For PNI as a continuous variable, each unit increase was associated with lower mortality (5 studies; HR 0.94; 95% CI 0.91–0.97; I²=89%; p=0.0003) and lower MACE (4 studies; HR 0.84; 95% CI 0.72–0.92; I²=97%; p=0.0007). Sensitivity analysis confirmed robustness for all outcomes except MACE (categorical), which became non-significant after excluding Keskin et al. (HR 1.37; 95% CI 0.97–1.94; p=0.08) or Raposeiras Roubín et al. (HR 1.54; 95% CI 0.98–2.42; p=0.06). Subgroup analyses for mortality (categorical PNI) showed significant results across Asian (HR 1.78; 95% CI 1.23–2.61) and European (HR 1.62; 95% CI 1.20–2.19) studies, sample sizes >1,000 (HR 2.92; 95% CI 1.68–5.09) and <1,000 (HR 1.07; 95% CI 1.00–1.13), MI-only (HR 1.38; 95% CI 1.08–1.76) and mixed diagnosis (HR 2.23; 95% CI 1.46–3.40), and PCI-only treatment (HR 1.85; 95% CI 1.46–2.35). Results were non-significant for mixed treatment groups (HR 1.41; 95% CI 0.86–2.31; p=0.17) and for PNI cut-off 44–47.1 (HR 1.88; 95% CI 0.87–4.09; p=0.11). For MACE (categorical PNI), subgroup results were significant only for sample size >1,000 (HR 1.77; 95% CI 1.30–2.42), mixed diagnosis (HR 1.55; 95% CI 1.20–2.00), and PNI cut-off 44–47.1 (HR 1.85; 95% CI 1.07–3.19). Funnel plots showed no major asymmetry.
**Clinical Implications:** This meta-analysis demonstrates that PNI independently predicts mortality and MACE in CAD patients, with low PNI conferring a 67% increase in mortality risk and a 57% increase in MACE risk. PNI offers advantages over other malnutrition tools: unlike CONUT (which uses cholesterol, potentially confounded by statin therapy) and GNRI (which uses BMI, potentially underestimating malnutrition in normal-weight or obese patients), PNI relies only on albumin and lymphocyte count — both routinely available at minimal cost. These findings support incorporating PNI into early risk stratification to identify high-risk CAD patients who may benefit from individualized, priority treatment. However, high inter-study heterogeneity (I² 89–97%) and variable PNI cut-offs (38–50.7) are major limitations. The authors note that albumin levels may be affected by congestion during MI or cardiogenic shock, and lymphocyte counts may be influenced by inflammatory status — both factors that could confound PNI assessment timing. Further research is needed to establish standardized PNI cut-offs, evaluate PNI in specific CAD subgroups (e.g., CABG-only, CCS-only), assess the impact of serial PNI changes during MI, and determine whether nutritional intervention strategies in patients with low PNI can improve outcomes.