Background <p>Metabolic dysregulation in critical illness may involve early changes in early morning fasting blood glucose (FBG) and high-density lipoprotein cholesterol (HDL-C), but the prognostic value of combining these measures is unclear.</p> Methods <p>We analyzed adult ICU patients from MIMIC-IV (v3.1) with FBG (03:00–07:59) and HDL-C measured within 24&#xa0;h of admission. Using multivariable Cox models, we assessed the association between the FBG/HDL-C ratio and 28-day all-cause mortality, examining nonlinear effects via restricted cubic splines and exploratory mediation by white blood cell count (WBC) and blood–urea–nitrogen (BUN). Findings were externally validated in the eICU database.</p> Results <p>Elevated FBG/HDL-C ratios were independently associated with 28-day all-cause mortality, with a nonlinear threshold identified at 5.934 and an AUC of 0.773. This association may be partially mediated by systemic inflammation and renal impairment.</p> Conclusions <p>Among critically ill patients with available FBG and HDL-C measurements, the FBG/HDL-C ratio was independently associated with short-term mortality, suggesting its potential as a supplementary risk stratification tool—though generalizability is limited by nonroutine HDL-C testing.</p>

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Prognostic value of the FBG/HDL-C ratio for 28-day all-cause mortality in critically ill patients: a retrospective cohort study with external validation

  • Xiao-jun Xiang,
  • Li Ma,
  • Xiao-long Sun,
  • Dong-yu Ma,
  • Meng-yang Liu,
  • Jian-feng Li,
  • Wen-bo Zhang,
  • Ping Xie

摘要

Background

Metabolic dysregulation in critical illness may involve early changes in early morning fasting blood glucose (FBG) and high-density lipoprotein cholesterol (HDL-C), but the prognostic value of combining these measures is unclear.

Methods

We analyzed adult ICU patients from MIMIC-IV (v3.1) with FBG (03:00–07:59) and HDL-C measured within 24 h of admission. Using multivariable Cox models, we assessed the association between the FBG/HDL-C ratio and 28-day all-cause mortality, examining nonlinear effects via restricted cubic splines and exploratory mediation by white blood cell count (WBC) and blood–urea–nitrogen (BUN). Findings were externally validated in the eICU database.

Results

Elevated FBG/HDL-C ratios were independently associated with 28-day all-cause mortality, with a nonlinear threshold identified at 5.934 and an AUC of 0.773. This association may be partially mediated by systemic inflammation and renal impairment.

Conclusions

Among critically ill patients with available FBG and HDL-C measurements, the FBG/HDL-C ratio was independently associated with short-term mortality, suggesting its potential as a supplementary risk stratification tool—though generalizability is limited by nonroutine HDL-C testing.