Prognostic nutritional index and geriatric nutritional risk index as predictors of in-hospital major adverse cardiovascular events after percutaneous coronary intervention in acute coronary syndrome patients: a multicenter retrospective cohort study.

OBJECTIVE: To evaluate the association between PNI and GNRI with in-hospital MACE after PCI in patients with ACS, and to compare the predictive value of these two nutritional assessment tools. The optimal cutoff values of PNI and GNRI were determined by calculating the maximum Youden's index based on ROC curve analysis in the training cohort. METHODS: This multicenter retrospective cohort study included ACS patients who underwent PCI at three hospitals between January 2021 and December 2024. The training cohort consisted of 638 patients from Anhui Provincial Hospital, while the validation cohort included 212 patients from Hefei Second People's Hospital and 238 patients from Ma'anshan People's Hospital. PNI was calculated based on serum albumin and lymphocyte count at admission, while GNRI was calculated using serum albumin and body mass index. The primary endpoint was in-hospital MACE, including all-cause mortality, myocardial infarction, stroke, and target vessel revascularization. Multivariate logistic regression analysis was used to assess the association of PNI and GNRI with in-hospital MACE, and the predictive abilities of the two indices were compared using ROC curve analysis. RESULTS: In the training cohort, 63 patients (9.9%) experienced in-hospital MACE. Multivariate analysis showed that low PNI (<45.2) and low GNRI (<98.0) were both independent predictors of in-hospital MACE (PNI: adjusted OR = 2.37, 95%CI: 1.42-3.96, p = 0.001; GNRI: adjusted OR = 2.18, 95%CI: 1.31-3.64, p = 0.003). ROC analysis revealed that PNI (AUC = 0.721, 95%CI: 0.653-0.789) and GNRI (AUC = 0.703, 95%CI: 0.632-0.774) had similar predictive values (p = 0.247). Similar results were observed in the validation cohort, and DeLong test confirmed no significant difference in AUC values of the two indices between training and validation cohorts, confirming the reliability of these findings. The combination of both nutritional indices (AUC = 0.753, 95%CI: 0.687-0.819) showed superior predictive ability compared to either index alone. CONCLUSION: Both PNI and GNRI serve as effective predictive tools for in-hospital MACE after PCI in ACS patients. The two nutritional assessment tools have similar predictive capabilities, and their combination may provide more accurate risk assessment. Routine nutritional status evaluation in ACS patients helps identify high-risk individuals, providing a basis for early intervention.