AUTHOR=Li Ling , Lu Xiuhong , Qin Shuangwen , Huang Debin TITLE=Association between geriatric nutritional risk index and 28 days mortality in elderly patients with sepsis: a retrospective cohort study JOURNAL=Frontiers in Medicine VOLUME=10 YEAR=2023 URL=https://www.frontiersin.org/journals/medicine/articles/10.3389/fmed.2023.1258037 DOI=10.3389/fmed.2023.1258037 ISSN=2296-858X ABSTRACT=Background

There is a significant controversy surrounding the impact of the geriatric nutritional risk index (GNRI) on mortality among elderly septic patients. This retrospective cohort study aimed to investigate the association between GNRI at admission and 28 days mortality in elderly septic patients.

Methods

We retrospectively analyzed data collected from the MIMIC IV database between 2009 and 2019, which included 2,834 septic patients aged 65 years and above. The exposure variable was the GNRI, determined according to albumin levels, height, and weight. The primary outcome was 28 days mortality. We employed multivariable Cox regression analyses and Kaplan–Meier survival curves to examine the association between GNRI and 28 days mortality. We used restricted cubic splines to determine if there was a non-linear relationship between 28 days mortality and GNRI in elderly patients with sepsis and to examine the presence of a threshold saturation effect. In addition, interaction tests were conducted to identify subgroups that exhibited significant differences.

Results

A total of 2,834 elderly patients with sepsis participated in the study. Following adjustment, multivariable Cox regression analyses demonstrated that the GNRI was related to 28 days mortality (HR = 0.97, p < 0.001, 95% CI: 0.97–0.98). An L-shaped connection between GNRI and 28 days mortality was discovered via restricted cubic spline analysis, with an inflection point of 98.1. On the left side of the inflection point, GNRI levels were significantly negatively linked with 28 days mortality (HR = 0.967, 95% CI: 0.959–0.974; p < 0.001), and on the right side, there was no significant correlation (HR = 1.043, 95% CI: 0.984–1.106; p = 0.1549).

Conclusion

In this analysis of data from a large cohort of elderly septic patients, GNRI scores on admission were correlated with a 28 days risk of death from sepsis in the elderly suggesting that GNRI scores could serve as a valuable indicator for evaluating mortality rates among elderly septic patients in the intensive care unit (ICU).