检验医学 ›› 2026, Vol. 41 ›› Issue (4): 362-368.DOI: 10.3969/j.issn.1673-8640.2026.04.008

• 论著 • 上一篇    下一篇

基于MIMIC-Ⅳ数据库数据分析红细胞分布宽度/血小板计数比值预测ARDS患者短期预后的价值

王源1, 陈丹2, 黄娜3()   

  1. 1 眉山市第二人民医院呼吸与危重症医学科四川 眉山 620500
    2 眉山市仁寿县运长医院呼吸内科四川 眉山 620500
    3 成都医学院第一附属医院呼吸与危重症医学科四川 成都 610500
  • 收稿日期:2024-12-08 修回日期:2025-06-16 出版日期:2026-04-30 发布日期:2026-05-07
  • 通讯作者: 黄 娜,E-mail:717308813@qq.com
  • 作者简介:王 源,男,1993年生,硕士,主治医师,主要从事慢性阻塞性肺疾病的基础和临床研究。
  • 基金资助:
    国家临床重点专科建设培育科室专项科研项目(CYFY2018GLPHX02);四川省医学科研创新课题(Q18041)

Role of red blood cell distribution width-to-platelet count ratio in predicting the short-term prognosis of patients with ARDS based on the datum analysis of MIMIC-Ⅳ database

WANG Yuan1, CHEN Dan2, HUANG Na3()   

  1. 1 Department of Respiratory and Critical Care MedicineMeishan Second People's HospitalMeishan 620500,Sichuan, China
    2 Department of Respiratory MedicineMeishan City Renshou County Yunchang HospitalMeishan 620500,Sichuan, China
    3 Department of Respiratory and Critical Care Medicinethe First Affiliated Hospital of Chengdu Medical CollegeChengdu 610500,Sichuan, China
  • Received:2024-12-08 Revised:2025-06-16 Online:2026-04-30 Published:2026-05-07

摘要:

目的 分析红细胞分布宽度/血小板计数比值(RPR)预测急性呼吸窘迫综合征(ARDS)患者短期预后的价值。方法 从MIMIC-Ⅳ数据库中提取重症监护病房(ICU)的ARDS患者相关临床资料。根据患者28 d预后情况分为生存组和死亡组,比较2组RPR差异。根据RPR最佳临界值将ARDS患者细分为高RPR组(RPR≥6.76%)和低RPR组(RPR<6.76%)。采用受试者工作特征(ROC)曲线评估RPR和相关疾病评分预测ARDS患者28 d死亡的效能。通过生存曲线分析高RPR组和低RPR组预后差异。采用Cox风险回归模型分析ARDS患者28 d死亡的影响因素。基于性别、年龄和相关合并症进行亚组分析,评价RPR和ARDS患者28 d死亡相关性的影响因素。结果 死亡组RPR高于生存组(P<0.001)。RPR升高是ARDS患者28 d预后的独立影响因素,预测ARDS患者28 d预后的曲线下面积(AUC)为0.776。高RPR组28 d存活率低于低RPR组(P<0.05)。RPR亚组分析结果表明,RPR和ARDS患者28 d预后的相关性与性别、年龄、相关合并症间均无交互作用(P交互均>0.05)。结论 RPR升高可用于预测ARDS患者28 d预后。

关键词: 红细胞分布宽度/血小板计数比值, MIMIC-Ⅳ数据库, 急性呼吸窘迫综合征, 短期预后

Abstract:

Objective To analyze the role of red blood cell distribution width-to-platelet count ratio(RPR) in predicting the short-term prognosis of patients with acute respiratory distress syndrome(ARDS). Methods The clinical data of ARDS patients in the intensive care unit(ICU) were extracted from the MIMIC-Ⅳ database. The patients were classified into survival group and death group based on their 28 d prognosis. The differences in RPR between the 2 groups were compared. ARDS patients were further classified into high RPR group(RPR≥6.76%) and low RPR group(RPR<6.76%) according to RPR optimal cut-off values. Receiver operating characteristic(ROC) curve was used to evaluate the efficacy of RPR and related disease scores in predicting 28 d mortality in ARDS patients. The differences in prognosis between high RPR group and low RPR group were analyzed through survival curves. Cox risk regression model was used to analyze the influencing factors of 28 d mortality in ARDS patients. Subgroup analysis was conducted based on gender,age and related comorbidities to evaluate the influencing factors related to 28 d mortality in patients with RPR and ARDS. Results The RPR of death group was higher than that of survival group(P<0.001). Elevated RPR was an independent predictor of the 28 d prognosis in patients with ARDS [hazard ratio (HR)=4.05,P<0.001],and the area under curve(AUC) for predicting the 28 d prognosis of ARDS patients was 0.776. The 28 d survival rate in high RPR group was lower than that in low RPR group(P<0.05). The results of subgroup analysis indicated that the correlation between RPR and the 28 d prognosis of ARDS patients had no interaction with gender,age and comorbidities(Pinteraction>0.05). Conclusions Elevated RPR level can be used to predict the 28 d prognosis of patients with ARDS.

Key words: Red blood cell distribution width-to-platelet count ratio, MIMIC-IV database, Acute respiratory distress syndrome, Short-term prognosis

中图分类号: