乳酸脱氢酶与白蛋白比值与危重阻塞性睡眠呼吸暂停患者全因死亡率之间的关联：一项基于MIMIC-IV数据库的回顾性队列研究
Association between lactic dehydrogenase-to-albumin ratio and all-cause mortality in critical obstructive sleep apnea patients: a retrospective cohort study from the MIMIC-IV database.
文献信息
| PMID | 42958154 |
|---|---|
| 原文 | 在 PubMed 查看原文 ↗ |
| 发表日期 | 2026 |
| 作者 | Enhui Zhou |
| 作者单位 | Department of Otorhinolaryngology Head and Neck Surgery, Shanghai Sixth People's Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, No. 600 Yishan Road, Shanghai, 200233, China. |
| 期刊 | BMC pulmonary medicine |
| SCI 分区 | Q2 |
| IF | 3.2 |
| 研究类型 | 临床研究 · 临床 |
| 所属专科 | 鼻科 |
中文摘要
目的： 本研究旨在阐明乳酸脱氢酶与白蛋白比值（LDAR）与危重阻塞性睡眠呼吸暂停（OSA）人群全因死亡率（ACM）之间的联系。
方法： 利用MIMIC-IV 3.1数据进行了一项回顾性观察性队列研究。本研究纳入了首次进入重症监护病房（ICU）的成年OSA患者。患者按LDAR四分位数进行分层。主要结局为进入ICU后30天、90天、180天和1年的ACM。通过Kaplan-Meier（KM）生存分析比较各LDAR四分位数间的ACM。利用多变量Cox比例风险回归和限制性立方样条（RCS）模型检验LDAR与ACM的关系。通过亚组、交互和敏感性分析评估研究结果的一致性和稳健性。
结果： 共纳入1118例危重OSA患者。749例（66.99%）为男性，中位年龄为66岁。30天、90天、180天和1年ACM率分别为14.58%、19.59%、22.00%和26.74%。KM曲线显示，随着LDAR四分位数升高，死亡率显著增高（log-rank P < 0.001）。在调整可能的混杂因素后，LDAR作为连续变量与90天ACM风险升高独立相关（HR，1.318；95%CI，1.142-1.520）。与最低LDAR四分位数相比，最高四分位数的90天ACM可能性高出139.8%（HR，2.398；95%CI，1.512-3.803；Ptrend < 0.001）。30天、180天和1年ACM也观察到类似关系。RCS分析显示LDAR与各时间点死亡呈正线性关系（Pnonlinear > 0.05）。亚组和交互分析未显示显著的效应修饰，证实了该关联在不同临床分层中的稳健性。
结论： 在危重OSA人群中，LDAR升高与短期和长期死亡率中ACM风险升高独立且线性相关。上述研究结果提示LDAR可作为该高危人群风险分层和临床管理的有价值预后生物标志物。
英文摘要
OBJECTIVES: Our study endeavors to elucidate the link of the lactic dehydrogenase-to-albumin ratio (LDAR) to all-cause mortality (ACM) among the critically ill obstructive sleep apnea (OSA) population.
METHODS: A retrospective observational cohort study was performed leveraging the MIMIC-IV 3.1 data. This study encompassed adult OSA patients first entering the intensive care unit (ICU). Patients were stratified into quartiles by LDAR. The primary outcomes were ACM at 30, 90, 180 days, and one year after ICU entry. ACM across LDAR quartiles was compared via the Kaplan-Meier (KM) survival analysis. The relation of LDAR to ACM was examined utilizing multivariable Cox proportional hazards regression and restricted cubic spline (RCS) models. The consistency and robustness of our findings were rated through subgroup, interaction, and sensitivity analyses.
RESULTS: One thousand one hundred eighteen critically ill OSA patients were encompassed. 749 (66.99%) were male, with a median age of 66. The 30-day, 90-day, 180-day, and one-year ACM rates were 14.58%, 19.59%, 22.00%, and 26.74%. KM curves demonstrated significantly higher mortality rates across increasing LDAR quartiles (log-rank P < 0.001). After possible confounders were adjusted, LDAR was independently related to elevated 90-day ACM risk (HR, 1.318; 95%CI, 1.142-1.520) as a continuous variable. In contrast to the lowest LDAR quartile, the highest quartile displayed a 139.8% greater likelihood of 90-day ACM (HR, 2.398; 95%CI, 1.512-3.803; Ptrend < 0.001). Similar relations were noted for 30-day, 180-day, and one-year ACM. RCS analysis revealed a positive linear relation of LDAR to death at all time points (Pnonlinear > 0.05). Subgroup and interaction analyses revealed insignificant effect modification, confirming the robustness of the association across various clinical strata.
CONCLUSION: Among the critically ill OSA population, elevated LDAR is independently and linearly related to elevated ACM risk in short-term and long-term mortality. The foregoing findings suggest LDAR as a valuable prognostic biomarker for risk classification and clinical management among this high-risk population.