乳酸与白蛋白比值与危重患儿28天院内全因死亡风险的队列研究
该回顾性队列研究基于儿科重症数据库(PIC)2010至2018年收入ICU的7 523例年龄≥28天且≤18岁的危重患儿,评估乳酸与白蛋白比值(LAR)与28天院内全因死亡的关系。按LAR分为<0.03组(2 055例)、0.03≤LAR<0.06组(2 409例)和LAR≥0.06组(3 059例)。多因素Cox回归显示LAR每增加0.1单位,28天死亡风险增加30%(HR=1.30,95%CI 1.23-1.38,P<0.001);以LAR<0.03组为参照,LAR≥0.06组死亡风险显著升高(HR=2.45,95%CI 1.73-3.46,P<0.001)。ROC曲线显示LAR预测28天死亡的AUC为0.705(95%CI 0.673-0.737),最佳截断值0.045,敏感度62.1%、特异度68.4%。Kaplan-Meier曲线显示LAR≥0.06组28天累积生存率更低(Log-rank χ²=182.60,P<0.001),提示LAR升高与危重患儿不良预后独立相关。
要点
- 危重患儿LAR每升高0.1单位,28天院内全因死亡风险增加30%(HR=1.30,95%CI 1.23-1.38,P<0.001)。
- 以LAR<0.03组为参照,LAR≥0.06组28天死亡风险显著升高(HR=2.45,95%CI 1.73-3.46,P<0.001)。
- LAR预测28天死亡的AUC为0.705(95%CI 0.673-0.737),最佳截断值0.045,敏感度62.1%、特异度68.4%。
- Kaplan-Meier分析显示LAR≥0.06组28天累积生存率更低(Log-rank χ²=182.60,P<0.001)。
- 亚组分析中除血糖、性别与28天死亡存在交互作用(交互P<0.05)外,年龄、脓毒症、肺炎、脑炎、恶性肿瘤等未见交互作用。
适用人群与场景
- 纳入2010至2018年儿科重症数据库(PIC)中收入各ICU、年龄≥28天且≤18岁的危重患儿,共7 523例,中位年龄1.4(0.4,4.5)岁,女性占44.2%。
- 敏感性分析纳入年龄<28天患儿及排除ICU住院时间<48小时者后,LAR与不良结局的关联保持一致。
诊断标准与关键阈值
- 按LAR分为三组:LAR<0.03组(2 055例)、0.03≤LAR<0.06组(2 409例)、LAR≥0.06组(3 059例)。
- LAR预测28天死亡的ROC曲线最佳截断值为0.045,AUC为0.705(95%CI 0.673-0.737),敏感度62.1%、特异度68.4%。
研究设计与关键结果
- 研究类型为回顾性队列研究,样本量7 523例,人群为PIC数据库中年龄≥28天且≤18岁的危重患儿,主要暴露变量为LAR,主要结局为28天院内全因死亡。
- 28天院内全因死亡率为4.5%;LAR越高,28天全因死亡率越高、住院时间越长。
- 校正潜在混杂因素后,LAR每增加0.1单位,28天死亡风险增加30%(HR=1.30,95%CI 1.23-1.38,P<0.001)。
- 以LAR<0.03组为参照,LAR≥0.06组28天死亡风险显著升高(HR=2.45,95%CI 1.73-3.46,P<0.001)。
- Kaplan-Meier曲线显示LAR≥0.06组28天累积生存率更低(Log-rank χ²=182.60,P<0.001)。
- 亚组分析显示除血糖、性别与28天死亡存在交互作用(交互P<0.05)外,年龄、脓毒症、肺炎、脑炎、恶性肿瘤等指标未见交互作用。
- 两项敏感性分析(纳入年龄<28天患儿、排除ICU住院时间<48小时者)结果保持一致,提示LAR与不良结局的关联稳定。
证据类型、等级与局限
- 证据类型:回顾性队列研究。
- 局限:原文未报告局限性。
对床旁实践的改变
- 作者认为LAR与危重患儿28天院内全因死亡独立相关,LAR升高的危重患儿临床预后更差。
提炼说明
- 本条依据论文摘要整理;摘要未涉及「处置流程与药物剂量」「红旗征与不能漏的情况」「指南或共识出处」,故未列出这些栏目;全文请点「阅读原文」。
英文原文(摘要)· 点击展开
OBJECTIVE: To assess the association between the lactate to albumin ratio (LAR) and 28-day in-hospital all-cause mortality in critically ill children.
METHODS: A retrospective cohort study was conducted. Children admitted to various intensive care units (ICU), aged≥28 days and ≤18 years in the Pediatric Intensive Care Database (PIC) from 2010 to 2018 were enrolled. Demographic data, ICU type, vital signs, comorbidities, the first laboratory parameters within 24 hours of admission, and outcomes were collected. The primary exposure variable was LAR. The primary outcome was 28-day in-hospital all-cause mortality. Children were divided into three groups based on LAR: LAR<0.03 group, 0.03≤LAR<0.06 group, and LAR≥0.06 group. Variables with significant differences in univariate analysis and potential risk factors were incorporated into multivariate Cox regression analysis. Risk models were constructed by adjusting for relevant covariates to evaluate the association between LAR and mortality. Kaplan-Meier survival curve for 28-day mortality and receiver operator characteristic curve (ROC curve) were plotted. Subgroup analyses were performed based on relevant covariates. In addition, sensitivity analyses were conducted by including children aged<28 days and excluding those with the length of ICU stay<48 hours to evaluate model robustness.
RESULTS: A total of 7 523 children were included, with a median age of 1.4 (0.4, 4.5) years old, and 44.2% were female, 28-day in-hospital all-cause mortality was 4.5%. Among them, there were 2 055 cases in the group with LAR<0.03, 2 409 cases in the group with 0.03≤LAR<0.06, and 3 059 cases in the group with LAR≥0.06. Higher LAR was associated with increased 28-day all-cause mortality and prolonged hospital stay. After adjusting for potential confounders, LAR was independently associated with mortality risk. For every 0.1-unit increase in LAR, the risk of 28-day mortality increased by 30% [hazard ratio (HR)=1.30, 95% confidence interval (95%CI) was 1.23-1.38, P<0.001]. When analyzed as a categorical variable using the LAR<0.03 group as reference, children with LAR≥0.06 had a significantly higher mortality risk (HR=2.45, 95%CI was 1.73-3.46, P<0.001). Kaplan-Meier curve analysis showed lower 28-day cumulative survival in the LAR≥0.06 group (Log-rank test: χ 2=182.60, P<0.001). ROC curve analysis indicated that the area under the curve (AUC) for LAR prediction of 28-day mortality was 0.705 (95%CI was 0.673-0.737), with a optimal cut-off value of 0.045. The sensitivity was 62.1% and the specificity was 68.4%. Subgroup analysis of relevant factors showed no interactions were observed in other indicators such as age, sepsis, pneumonia, encephalitis, and malignant tumors, except for interaction between glucose, gender, and 28-day mortality (P for interaction <0.05). Results from both sensitivity analyses remained consistent, indicating a stable association between LAR and adverse outcomes.
CONCLUSIONS: LAR is independently associated with 28-day in-hospital all-cause mortality in critically ill children. Critically ill children with elevated LAR levels have a poorer clinical prognosis.
参考来源
- He W, Jiang R, Yang X, et al. [A cohort study on the association between lactate to albumin ratio and 28-day in-hospital all-cause mortality in critically ill children]. Zhonghua Wei Zhong Bing Ji Jiu Yi Xue. 2026. PMID: 41876246. doi:10.3760/cma.j.cn121430-20250107-00015
供执业医师抢救与诊疗参考,不替代临床判断;剂量请结合体重、肝肾功能和本院药品规格复核,以指南原文为准。