Kim Hye Jin, Shim Jae Chan, Oh Ju Hyun, Choi Sang Bong, Lee Hyuk Pyo, Chang Youjin
Department of Clinical Nutrition, Inje University Sanggye Paik Hospital, Seoul, Republic of Korea.
Department of Radiology, Inje University College of Medicine, Sanggye Paik Hospital, Seoul, Republic of Korea.
Am J Clin Nutr. 2025 Jul;122(1):306-314. doi: 10.1016/j.ajcnut.2025.05.005. Epub 2025 May 10.
There is no gold standard tool for nutritional assessment in critically ill patients. The modified Nutrition Risk in Critically Ill (mNUTRIC) score assesses risk of future malnutrition, whereas the Global Leadership Initiative on Malnutrition (GLIM) criteria diagnose only current malnutrition.
We aimed to evaluate the prognostic performance of a novel nutritional assessment tool that combines the mNUTRIC score and GLIM criteria for hospital outcomes among intensive care unit (ICU) patients.
A single-center, retrospective observational cohort study was conducted on patients admitted to the ICU between 1 August, 2018, and 7 June, 2021. The novel tool was categorized into 4 groups: group 1: low risk (0-4 points) by mNUTRIC plus no malnutrition from the GLIM; group 2: low risk plus moderate malnutrition or high risk (5-9 points) plus no malnutrition; group 3: high risk plus moderate malnutrition or low risk plus severe malnutrition; and group 4: high risk plus severe malnutrition.
A total of 724 patients were enrolled. The in-hospital mortality rates for groups 1-4 were 3%, 16%, 32%, and 52%, respectively. A poor nutritional status by the novel nutritional assessment tool was an independent risk factor for in-hospital mortality when compared with group 1, particularly as patients progressed to group 4 (odds ratio: 2.32; 95% confidence interval: 1.84, 2.93). In the receiver operating characteristic (ROC) analysis, the novel tool demonstrated the strongest prognostic performance for in-hospital mortality (area under the ROC curve: 0.759; 95% CI: 0.723, 0.795) compared with validated severity scoring systems and other significant risk factors such as use of mechanical ventilation.
This novel nutritional assessment tool has strong prognostic value in being associated with in-hospital mortality. Prospective validation with an external dataset is warranted. A comprehensive nutritional assessment of the current status and future risk at ICU admission could be the most critical factor in assessing the prognosis of critically ill patients.
在危重症患者中，尚无用于营养评估的金标准工具。改良的危重症营养风险（mNUTRIC）评分评估未来发生营养不良的风险，而全球营养不良领导倡议（GLIM）标准仅用于诊断当前的营养不良。
我们旨在评估一种结合mNUTRIC评分和GLIM标准的新型营养评估工具对重症监护病房（ICU）患者医院结局的预后评估性能。
对2018年8月1日至2021年6月7日期间入住ICU的患者进行了一项单中心回顾性观察队列研究。该新型工具分为4组：第1组：mNUTRIC评分为低风险（0 - 4分）且GLIM标准判定无营养不良；第2组：低风险且存在中度营养不良，或高风险（5 - 9分）且无营养不良；第3组：高风险且存在中度营养不良，或低风险且存在重度营养不良；第4组：高风险且存在重度营养不良。
共纳入724例患者。第1 - 4组的院内死亡率分别为3%、16%、32%和52%。与第1组相比，新型营养评估工具提示的不良营养状况是院内死亡的独立危险因素，尤其是当患者进展到第4组时（比值比：2.32；95%置信区间：1.84，2.93）。在受试者工作特征（ROC）分析中，与已验证的严重程度评分系统及其他重要风险因素（如机械通气使用情况）相比，该新型工具对院内死亡的预后评估性能最强（ROC曲线下面积：0.759；95% CI：0.723，0.795）。
这种新型营养评估工具与院内死亡相关，具有很强的预后价值。有必要使用外部数据集进行前瞻性验证。在ICU入院时对当前状况和未来风险进行全面的营养评估可能是评估危重症患者预后的最关键因素。
