Table_2_The prognostic effects of the geriatric nutritional risk index on elderly acute kidney injury patients in intensive care units.DOCX

IntroductionThe geriatric nutritional risk index (GNRI), a nutritional screening tool specifically for the aging population, has been proven to be associated with worse outcomes in chronic kidney disease patients, especially in the hemodialysis population. However, the predictive validity of GNRI in critically ill elderly patients with acute kidney injury (AKI) is yet to be determined. This analysis sought to examine the prognostic effects of GNRI on elderly AKI patients in intensive care units (ICUs).
MethodsWe collected elderly AKI patient-relevant data from the Medical Information Mart for Intensive Care III database. AKI was diagnosed and staged according to the “Kidney Disease Improving Global Outcomes” criteria. In the study, 1-year mortality was considered the primary outcome, whereas in-hospital, ICU, 28-day and 90-day mortality, and prolonged length of stay in ICU and hospital were selected as the secondary outcomes.
ResultsIn all, 3,501 elderly patients with AKI were selected for this study, with a 1-year mortality rate of 36.4%. We classified the study population into low (≤98) and high (>98) GNRI groups based on the best cutoff value. The incidence of endpoints was remarkably lower in patients with elevated GNRI (p < 0.001). When stratified by the AKI stage, patients with high GNRI at AKI stages 1, 2, and 3 had markedly lower 1-year mortality than those with low GNRI (all p < 0.05). The multivariable regression analysis identified the independent prognostic ability of GNRI on the research outcomes (all p < 0.05). Restricted cubic spline exhibited a linear correlation between GNRI and 1-year death (p for non-linearity = 0.434). The prognostic implication of GNRI on 1-year mortality was still significant in patients with the most subgroups.
ConclusionIn critically ill elderly patients with AKI, elevated GNRI upon admission was strongly correlated with a lower risk of unfavorable outcomes.

引言
老年营养风险指数（Geriatric Nutritional Risk Index, GNRI）是一款专为老年人群设计的营养筛查工具，已有研究证实其与慢性肾脏病患者的不良预后相关，在血液透析人群中这一关联尤为显著。然而，GNRI在合并急性肾损伤（Acute Kidney Injury, AKI）的重症老年患者中的预测效度仍有待明确。本分析旨在探讨GNRI对重症监护病房（Intensive Care Unit, ICU）内老年AKI患者的预后影响。
方法
我们从重症监护医学信息数据库第三版（Medical Information Mart for Intensive Care III, MIMIC-III）中提取了老年AKI患者的相关数据。AKI的诊断与分期依据改善全球肾脏病预后组织（Kidney Disease: Improving Global Outcomes, KDIGO）制定的标准。本研究以1年死亡率为主要结局指标，以院内死亡率、ICU死亡率、28天死亡率、90天死亡率以及ICU和住院时间延长为次要结局指标。
结果
本研究共纳入3501例老年AKI患者，其1年死亡率为36.4%。基于最佳截断值，我们将研究人群分为低GNRI组（≤98）与高GNRI组（>98）。GNRI水平较高的患者，其各终点事件发生率显著更低（p < 0.001）。按AKI分期进行分层分析后发现，AKI 1、2、3期的高GNRI组患者1年死亡率均显著低于低GNRI组（所有p < 0.05）。多变量回归分析证实，GNRI对本研究的结局指标具有独立的预后预测能力（所有p < 0.05）。限制性立方样条分析显示，GNRI与1年死亡率之间呈线性相关（非线性检验p=0.434）。在绝大多数亚组患者中，GNRI对1年死亡率的预后预测价值仍具有统计学意义。
结论
在合并AKI的重症老年患者中，入院时较高的GNRI水平与不良结局风险降低显著相关。