Patients with a (GNRI) ≤ 98 exhibited a 21.8% in-hospital mortality rate compared to 10.4% for those with GNRI > 98.
GNRI is associated with higher mortality rates in critically ill patients with .
An inverse correlation exists between GNRI scores and both in-hospital and 30-day mortality rates.
GNRI demonstrated a predictive performance with an area under the curve (AUC) of 0.64.
The integration of GNRI with existing scoring systems enhances their predictive accuracy for mortality risk.
Subgroup analysis indicates GNRI's effectiveness, especially in patients with a kidney function (eGFR) of 90 or greater.
Simplified
BACKGROUND: Nutritional status is a key factor influencing outcomes in critically ill patients with (AMI). This study investigated the association between the (GNRI) and mortality among ICU-admitted AMI patients, as well as GNRI's potential to improve the predictive accuracy of current scoring systems.
METHODS: In this retrospective cohort study, data from 5,506 ICU-admitted AMI patients were sourced from three open-access critical care databases. Based on GNRI scores, patients were grouped into two categories: GNRI ≤ 98 and GNRI > 98. Statistical tools such as logistic regression and Cox proportional hazards models assessed in-hospital and 30-day mortality. Kaplan-Meier survival curves and restricted cubic splines analyzed survival trends and dose-response relationships. Sensitivity analyses, including propensity score matching (PSM), inverse probability weighting (IPW), and dropping missing data analysis validated the robustness of findings. The receiver operating characteristic (ROC) curve compared GNRI's predictive ability with and scores. A sensitivity analysis was performed using a four-tier GNRI classification: no risk (> 98), low risk (92-98), moderate risk (82-<92), and major risk (< 82) to further explore its gradient relationship with mortality.
RESULTS: Patients with GNRI ≤ 98 showed higher mortality rates for in-hospital (21.8% vs. 10.4%) and 30-day (22.5% vs. 10.7%) outcomes. GNRI displayed an inverse correlation with in-hospital mortality (OR 0.51, 95% CI 0.43-0.60) and 30-day mortality (HR 0.57, 95% CI 0.50-0.66), even after adjusting for confounders. Subgroup analysis emphasized GNRI's reliability as a predictive marker, particularly in patients with eGFR ≥ 90. ROC analysis confirmed GNRI's predictive performance (AUC = 0.64) and its enhancement of SOFA (AUC = 0.72) and APSIII (AUC = 0.66) scores (all p < 0.001). Sensitivity analyses reinforced GNRI's link to mortality.
CONCLUSION: GNRI serves as a robust predictor of in-hospital and 30-day mortality among critically ill AMI patients. Its integration with existing scoring systems improves risk stratification in this high-risk population.
Key numbers
21.8%
In-hospital Mortality Increase
In-hospital mortality for ≤ 98 vs. > 98
22.5%
30-day Mortality Increase
30-day mortality for ≤ 98 vs. > 98
0.51
for In-hospital Mortality
Adjusted for in-hospital mortality with
Full Text
We can’t show the full text here under this license.
Declarations. Ethics approval and consent to participate: The study was conducted in accordance with the guidelines of the Helsinki Declaration. The MIMIC database was supported by grants from the National Institute of Biomedical Imaging and Bioengineering (NIBIB) of the National Institutes of Health (NIH) under award numbers R01-EB001659 (2003–2013) and R01-EB017205 (2014–2022) and approved by the Institutional Review Boards of Beth Israel Deaconess Medical Center (Boston, MA) and the Massachusetts Institute of Technology (Cambridge, MA). As the data are publicly available, the ethical approval and requirement for informed consent were waived for this study. The database used in this study was approved for research use by the Institutional Review Boards of the Massachusetts Institute of Technology and Beth Israel Deaconess Medical Center. All patient information in the database is anonymized, and therefore, informed consent is not required. We completed online courses and exams and gained access to the database (record ID: 52219361). Competing interests: The authors declare no competing interests.
