Objective To analyze the potential factors affecting the prognosis of patients with sepsis and evaluate their values in predicting the disease outcome. Methods A clinical prospective study was conducted. Fifty-three septic patients admitted to intensive care unit (ICU) of the First Affiliated Hospital of Guangzhou Medical University from October 17th, 2012 to August 8th, 2013 were enrolled, and in the same term 35 volunteers having passedphysical check-up were assigned in the healthy control group. According to the severity of the patients, they were divided into sepsis, severe sepsis and septic shock groups. Furthermore, based on the difference in scores of acute physiology and chronic health evaluation Ⅱ (APACHE Ⅱ), the patients were divided into low-risk (APACHE Ⅱ scores < 10), middle-risk (APACHE Ⅱ scores 10 - 19) and high-risk groups (APACHE Ⅱ scores≥ 20 ). According to whether the coagulation dysfunction occurred or not and whether the dysfunction was adjusted or not at the end of observation, the septic patients were divided into non-coagulation defect group, and adjusted and non-adjusted coagulation defect groups. After entrance of groups, the clinical data, including gender, age, body temperature, blood pressure, prothrombin time (PT), prothrombin time activity (PTA), activated partial thromboplastin time (APTT), international normalized ratio (INR), levels of D-dimer, fibrinogen (Fib), blood lactate, and serum procalcitonin (PCT) were recorded, and APACHEⅡscore was calculated. In 24 hours after admission, blood samples were collected, and the levels of interleukins (IL-6, IL-8) were tested by Bio-Plex 200 System. The prognostic factors related to sepsis were screened by binary multivariate logistic regression analysis. The receiver operating characteristic curve (ROC) were used to evaluate the prognostic values of blood lactate, PCT, IL-6, IL-8 and APACHE Ⅱ scores for patients with sepsis. Results Fifty-three patients were enrolled, including 17 patients in sepsis group in which blood coagulation dysfunction occurred in 8 cases, of them 7 being adjusted, and 5 died; 15 patients in severe sepsis group, in which blood coagulation dysfunction occurred in 7 cases, of them 2 being adjusted, and 7 died;21 patients in septic shock group in which blood coagulation dysfunction occurred in 18 cases, of them 4 being adjusted, and 18 died. Both IL-6 and IL-8 levels of sepsis group were significantly higher than those in healthy control group (both P<0.01). Univariate logistic regression analysis showed that the grade of sepsis severity, APACHEⅡscore, IL-6, IL-8, blood lactate, PCT and coagulation dysfunction were related to the prognosis of septic patients. Multivariate logistic regression analysis confirmed that blood lactate concentration [β=0.891,χ2 = 5.872, P = 0.015, odds ratio (OR) = 2.438, 95% confidence intervals (95%CI) was 1.186 - 5.013] and coagulation function status (non-adjusted coagulation defect group:β=3.563,χ2=9.980, P=0.002, OR=35.286, 95%CI was 3.868-3.563) were independent prognostic factors of septic patients. The ROC curve analysis showed:for the level of blood lactate, the area value under ROC curve (AUC) was 0.767, the best cutoff value was 2.15 mmol/L with the Youden index of 0.386;for PCT, the AUC was 0.698, the best cutoff value was 9.39μg/L with the Youden index of 0.406;for IL-8, AUC was 0.686, the best cutoff value was 20.06 ng/L with the Youden index of 0.312;and for IL-6, AUC was 0.681, the best cutoff value was 45.44 ng/L with the Youden index of 0.406. Multivariate logistic regression analysis confirmed that the plasma levels of IL-6 and IL-8 were the independent risk factors of septic patients' coagulation function. The IL-6 concentration of non-coagulation defect group was significantly lower than that in adjusted coagulation defect group (ng/L:29.26 vs. 67.98, P<0.05) and non-adjusted coagulation defect group (ng/L:29.26 vs. 128.00 P<0.05). There was no significant difference in IL-6 level between the adjusted and non-adjusted coagulation defect groups (P>0.05). The IL-8 level of non-coagulation defect group was significantly lower than that of adjusted (ng/L:24.67 vs. 27.23, P<0.05) and non-adjusted coagulation defect groups (ng/L:24.67 vs. 60.14, P<0.05). There was no statistically significant difference in IL-8 concentration between adjusted and non-adjusted coagulation defect groups (P>0.05). Conclusions The grade of sepsis severity, APACHEⅡscore, whether existence of coagulation dysfunction being present or not and whether its presence being adjusted or not during the septic patients' stay in ICU, the levels of blood lactate, PCT, IL-6 and IL-8 on the first day in ICU are significantly correlated to the prognosis of septic patients. Whether the existence of coagulation dysfunction being present or not, whether coagulation dysfunction being adjusted or not and the blood lactate level are the independent prognostic factors of septic patients, and the plasma concentrations of IL-6 and IL-8 are the independent affecting factors of whether coagulation dysfunction occurring or not, therefore they have predicting value concerning the occurrence of coagulation dysfunction in septic patients.

Objective To investigate the effect of hydrochloric acid (HCl) stimulation and mechanical stretch on epithelial-mesenchymal transition (EMT) and hyaluronan (HA) production in human lung epithelial cells. Methods Human lung epithelial cell line BEAS-2B was cultured in vitro, which was divided into phosphate-buffer saline (PBS) + static group, HCl + static group, PBS + stretch group, and HCl + stretch group respectively in the logarithmic phase. The BEAS-2B cells in two stretching groups were challenged by cyclic stretch with 20% amplitude, frequency of 0.33 Hz, sine wave of the FX-5000T system for 48 hours. The morphology changes in cells before and after stretch were observed with inverted microscope. The protein expressions of epithelial markers E-cadherin and cytokeratin-8 (CK-8) as well as mesenchymal markers vimentin and α-smooth muscle actin (α-SMA) were determined by Western Blot. The secretion of HA was determined by enzyme linked immunosorbent assay (ELISA). Results ① It was shown by microscopic observation that BEAS-2B cells displayed cobblestone morphology, linked closely and cell polarity in PBS + static group, which did not change obviously after HCl stimulation alone. Given purely mechanical stretch after 48 hours, the cells morphology changed from cobblestone shape into long spindle, and increased intercellular space obviously. Double hit of HCl and stretch changed the cells morphology more significantly. ② It was shown by Western Blot that compared with the PBS + static group, HCl alone or combined with purely mechanical stretch after 48 hours, the expressions of E-cadherin and CK-8 were decreased, while those of vimentin and α-SMA were increased, and it was more pronounced in HCl + stretch group [the expression quantity (gray value) as base 1 in PBS + static group, E-cadherin: 0.16±0.08 vs. 1, CK-8: 0.10±0.03 vs. 1, vimentin: 3.35±0.38 vs. 1, α-SMA: 3.10±0.45 vs. 1, all P < 0.01]. ③ It was shown by ELISA that both HCl stimulation and stretch could induce BEAS-2B cells secreting HA as compared with PBS + static group (μg/L: 55.763±0.687, 63.005±0.493 vs. 49.876±1.867), and the production of HA increased more remarkably after double hit (μg/L: 78.220±1.085 vs. 49.876±1.867, P < 0.01). Conclusions Both HCl and mechanical stretch could induce EMT and increase HA secretion in human lung epithelial cells in vitro. Double hit of HCl stimulation and mechanical stretch induced EMT apparently, and further increased the production of HA.