Cox regression based on serum TGmax groups.

BackgroundSepsis is characterized by upregulated lipolysis in adipose tissue and a high blood triglyceride (TG) level. It is still debated whether serum TG level is related to mortality in septic patients. The aim of this study is to investigate the association between serum TG level and mortality in septic patients admitted to the intensive care unit (ICU).MethodsData from adult septic patients (≥18 years) admitted to the ICU for the first time were obtained from the Multiparameter Intelligent Monitoring in Intensive Care IV (MIMIC-IV) database. The patients’ serum TG levels that were measured within the first week after ICU admission were extracted for statistical analysis. The endpoints were 28-day, ICU and in-hospital mortality.ResultsA total of 2,782 septic patients were included. Univariate analysis indicated that the relationship between serum TG levels and the risk of mortality was significantly nonlinear. Both the Lowess smoothing technique and restricted cubic spline analyses revealed a U-shaped association between serum TG levels and mortality among septic patients. The lowest mortality rate was associated with a serum TG level of 300–500 mg/dL. Using 300∼500 mg/dL as the reference range, we found that both hypo-TG (300 mg/dL) and hyper-TG (≥500 mg/dL) were associated with increased mortality. The result was further adjusted by Cox regression with and without the inclusion of some differential covariates.ConclusionsThere was a U-shaped association between serum TG and mortality in septic ICU patients. The optimal concentration of serum TG levels in septic ICU patients is 300–500 mg/dL.