[The diagnostic and prognostic values of serum total RNA concentration in patients with sepsis]

To explore the diagnostic and prognostic roles of serum total RNA in patients with sepsis.

Seventy-six patients with sepsis admitted to intensive care unit (ICU) of Chinese PLA General Hospital from October 2009 to November 2010 were enrolled. Blood samples were drawn from these patients and total RNAs of these blood samples were isolated. The concentration of total RNA in the samples was detected. C-reactive protein (CRP) was determined by scattering turbidimetry, and procalcitonin (PCT) was determined using enzyme-linked fluorescence analysis (ELFA). Acute physiology and chronic health evaluation II (APACHE II ) score and sequential organ failure assessment (SOFA) score were recorded. Twenty-four healthy individuals with matched age and sex were enrolled as normal controls. Receiver operating characteristic curve (ROC curve) was used to determine the optimal concentration of serum total RNA for the diagnosis of sepsis. Binary logistic regression analysis was used to analyze the risk factors of sepsis patients.

The concentration of serum total RNA (ng/μl) in sepsis group was significantly higher than healthy controls [4.233 (0.204, 37.733) vs. 1.593 (0.319, 5.913), P<0.01]. The area under curve (AUC) of the diagnosis of sepsis was 0.820. When the cut-off point was set at 2.027 ng/μl, the concentration of serum total RNA yielded a sensitivity of 95.9% and a specificity of 65.2%. Total RNA concentration in the healthy controls, the survival group (n=42) and non-survival group (n=34) were step wisely increased [1.593 (0.319, 5.913), 4.178 (0.204, 10.776), 5.006 (0.997, 37.733), respectively, F=8.061, P=0.001]. Binary multivariate logistic regression analysis showed that APACHE II score and SOFA score were independent risk factors of septic patients [APACHE II score: odds ratio (OR)=1.265, 95% confidence interval (95%CI) 1.031 to 1.553, P=0.025; SOFA score: OR=1.151, 95%CI 1.031 to 2.284, P=0.012]. But PCT was not independent risk factor of sepsis patients (OR=1.075, 95%CI 0.974 to 1.187, P=0.151). In the binary univariate logistic regression analysis model, the OR of total RNAs was 1.149, but P value was 0.061, because of the small sample size.

The concentration of total RNA in sepsis patients' serum was significantly increased and correlated to the severity of sepsis, which can be used as biomarkers for sepsis patients.