Significance of the ratio of plasma vascular endothelial growth factor level to platelet count in the prognosis of patients with sepsis

Abstract

To investigate the clinical value of the ratio of plasma vascular endothelial growth factor level to platelet count (VEGF/PLT) in predicting 28-day prognosis in patients with sepsis. A prospective cohort study was conducted. From September 2009 to March 2013, 164 sepsis patients in Intensive Care Unit (ICU) of Guangdong General Hospital were included for study. Patients with age younger than 18 years old, the illness already reaching final stage of chronic diseases, suffering from two or more organs dysfunction within 3 days, acute pancreatitis without infection, or less than 28 days of expected survival time were excluded. Finally, 135 patients were included in the further analysis. Peripheral blood samples were collected at admission. Routine blood tests were done, and then VEGF levels in plasma were measured by enzyme linked immunosorbent assay (ELISA). Acute physiology and chronic health evaluation II (APACHEII) scores were recorded every day for 7 days. Patients' prognosis was assessed during the following 28 days. The patients were divided into 28-day survival group and non-survival group. Comparison between two groups was done by single factor analysis. Spearman rank correlation was used to analyze the correlation between VEGF levels and PLT. Mutivariate logistic regression analysis was performed to identify the independent risk factor for 28-day prognosis. Receiver operating characteristic curve (ROC curve) was plotted, and the effect of related indexes on predicting 28-day survival was evaluated by area under ROC curve (AUC). There were no significant differences in VEGF (471.73±198.34 ng/L vs. 383.49±266.54 ng/L, t=-1.918, P=0.057), PLT (220.40±127.60×10(9)/L vs. 246.42±100.72×10(9)/L, t=1.275, P=0.204), leucocyte counts (12.48±4.62×10(9)/L vs. 13.70±5.97×10(9)/L, t=1.063, P=0.292), mean arterial pressure (86.50±12.04 mmHg vs. 91.03±13.10 mmHg, t=1.557, P=0.123) and blood lactic acid (1.79±1.30 mmol/L vs. 1.50±0.60 mmol/L, t=-1.768, P=0.079) at admission between the non-survival group (n=42) and survival group (n=93). VEGF/PLT (2.59±1.44 vs. 1.73±1.13, t=-3.756, P=0.000) as well as APACHEII scores (15.50±4.50 vs. 13.28±4.61, t=-2.022, P=0.045) of the non-survival group were significantly higher than those of survival group, and oxygenation index (PaO2/FiO2) of the non-survival group was significantly lower than that of survival group (32.38±11.12 kPa vs. 37.04±10.97 kPa, t=2.278, P=0.024). Correlation analysis showed that the concentration of VEGF was positively correlated with PLT (r=0.271, P=0.001). It was shown by multivariate logistic regression analysis that only VEGF/PLT was the independent risk factor in predicting 28-day prognosis in patients with sepsis [odds ratio (OR) was 1.591, 95% confidence interval (95%CI) 1.164-2.175, P=0.004]. AUC of VEGF/PLT was 0.704±0.047 (P=0.000, 95%CI: 0.611-0.797) for predicting 28-day survival. The optimal cut-off point was 1.32, and the sensitivity and specificity were 81.0% and 48.4 %,respectively. VEGF/PLT can be used as one of the indicators to predict 28-day survival in patients with sepsis.