Association Between Geriatric Nutritional Risk Index and Critically Ill Patients With Pressure Injury: Analysis of the MIMIC-IV Database.

Abstract

To explore the relationship between the Geriatric Nutritional Risk Index (GNRI) and the occurrence of Pressure injury (PI) in elderly Intensive Care Unit (ICU) patients. PI represent a significant health concern within ICU, where the occurrence of such injuries is notably high among critically ill patients. However, few studies have explored the relationship between GNRI and PI. A longitudinal, single-centre, retrospective study. The study utilised GNRI calculation to identify PI occurrences. A comprehensive set of covariates was analysed, including demographic information, severity of illness scores, comorbidities, therapeutic interventions, vital signs and laboratory values. Statistical analysis involved descriptive statistics, logistic regression and receiver operating characteristic curves, supplemented by a doubly robust estimation method and propensity score modelling. This study follows the STROBE-nut checklist. Among 5327 critically ill patients enrolled in the study, with a median age of 76 years, of which 2339 were females, representing 43.91% of the total study population. Binary logistic regression analysis revealed that with each unit increase in the GNRI, the likelihood of PI occurrence decreased by 3.7% in the fully adjusted model. Furthermore, there was a significant decrease in PI occurrence among patients deemed to have no risk compared to those identified as at-risk. GNRI emerges as a significant, independent predictor of PI risk in elderly ICU patients, underscoring the importance of nutritional assessment and management in this population. This study highlights the critical importance of nutritional assessment, specifically through GNRI, in early identification and prevention of pressure injuries among elderly ICU patients, underscoring the need for integrated nutritional strategies in clinical settings. This is a retrospective cohort study, and no patients or the public were involved in the design and conduct of the study.