Correlation of triglyceride (TG)-to-high density lipoprotein cholesterol (HDL-C) ratio (TG/HDL-C) and the survival of gastric cancer (GC) remain unclear. The purpose of this study was to explore the precise effect of preoperative TG/HDL-C on clinical outcomes in GC patients. Patients with GC were enrolled from 2006 to 2014. A total of 957 individuals from a single center were divided into prospective training and retrospective test cohorts. The optimal cutoff value of TG/HDL-C was determined using X-tile software to separate the training cohort into low and high survival groups according to TG/HDL-C levels. Survival analyses were performed using Kaplan-Meier curves and a Cox proportional hazards regression model. Preoperative TG/HDL-C and clinical outcomes were obtained to determine the prognostic significance of serum lipids in the training and test cohorts. We observed that high TG and TG/HDL-C were significantly correlated with poor outcome in GC patients, and high TG/HDL-C harbored the highest area under curve to independently predict 5-year overall survival in two cohorts. Furthermore, c-index of the prognostic nomogram including TG/HDL-C was significantly higher than that without it. In summary, TG/HDL-C was an efficient and independent prognostic factor to predict 5-year case fatality of GC patients and to improve the efficacy of its prognostic nomogram.
Keywords: clinical outcome; gastric cancer; high density lipoprotein cholesterol; nomogram; triglyceride.
© The author(s).