[Prognostic significance of plasma gelsolin in severe burn patients with sepsis]

To investigate the changes in plasma gelsolin (pGSN) levels in severe burn patients with sepsis, and to evaluate the prognosis of patients when combined with other related clinical indexes.

Sixty-five severe burn patients with sepsis hospitalized from June 2013 to June 2015 conforming to the study criteria were divided into death group (n=24) and survival group (n=41) according to the clinical outcome on post sepsis diagnosis day (PSD) 28. The pGSN levels of patients were determined on PSD 1, 3, 7, and 14 with double antibody sandwich enzyme-linked immunosorbent assay. The serum level of C-reactive protein (CRP), serum level of procalcitonin, lactate level of arterial blood, Acute Physiology and Chronic Health Evaluation (APACHE) II score, and Sequential Organ Failure Assessment (SOFA) score were determined or recorded on PSD 1. Data were processed with repeated measurement analysis of variance, t test, and chi-square test. On PSD 1, the pGSN level, serum level of CRP, serum level of procalcitonin, lactate level of arterial blood, APACHE II score, and SOFA score of 65 patients were collected to screen the independent risk factors related to death with single factor and multi-factor Logistic regression analysis. Receiver operating characteristic (ROC) curves of the independent risk factors related to death were plotted to evaluate the predictive power for death in 65 patients.

(1) The pGSN levels of patients in death group on PSD 1, 3, 7, and 14 were respectively (146±44), (85±24), (28±7), and (19±4) mg/L, obviously lower than those in survival group [(287±82), (179±51), (196±56), and (249±67) mg/L, with t values from 1.735 to 4.304, P<0.05 or P<0.01]. (2) The serum level of CRP, serum level of procalcitonin, lactate level of arterial blood, APACHE II score, and SOFA score of patients in death group on PSD 1 were respectively (56±7) mg/L, (12.54±0.82) μg/L, (2.74±0.27) mmol/L, (24.3±2.4) points, and (11.43±0.57) points, significantly higher than those in survival group [(35±4) mg/L, (2.38±0.16) μg/L, (1.83±0.12) mmol/L, (15.0±1.5) points, and (7.22±0.23) points, with t values from 1.902 to 3.883, P<0.05 or P<0.01]. (3) Multi-factor Logistic regression analysis showed that the pGSN level (odds ratio: 6.83, 95% confidence interval: 4.33-10.25, P<0.01) and APACHE II score (odds ratio: 5.27, 95% confidence interval: 2.28-9.16, P<0.01) were the independent risk factors related to death in 65 patients on PSD 1. (4) The total areas under the ROC curves of pGSN level and APACHE II score for predicting death of 65 patients on PSD 1 were respectively 0.89 and 0.86, and 142 mg/L and 21 points were respectively chosen as the optimal threshold values, with sensitivity of 87% and 83% and specificity of 86% and 89%.

For severe burn patients with sepsis, lowering of pGSN level and elevation of APACHE II score are obviously correlated with increase in case fatality rates. Monitoring the dynamic changes in pGSN level and APACHE II score during the early stage may be useful to predict the prognosis of severe burn patients with sepsis.