The decreasing tendency within 7 days in SOFA score seemed to favor the T��1 group but with no significant difference in changes between the two groups. However, selleck chemicals U0126 considering the fact that we observed the changes of these indices for only 7 days, there could have been some difference between the two groups if the observation had been extended to 14 or 28 days.The median time from the first organ dysfunction detected to enrollment was more than 24 hrs in both groups, but longer in the T��1 group. We adopted a retrospective method to determine the time window between the onset of the first organ dysfunction detected and study enrollment according to objective data (such as blood gas analysis), many of which were obtained before transferring the severe sepsis patients to the ICU [7].

However, those patients without indicative objective data could also have suffered from severe sepsis and the delay in laboratory tests could substantially underestimate the time after onset. In other words, the time after onset determined by laboratory tests in non-ICU departments was out of our control and subject to errors, especially when the estimation was based on hours instead of days. The precise time window between onset of the first organ dysfunction and enrollment could exceed the recorded time and could possibly be balanced between the two groups. The better way of enrolling severe sepsis patients in immunotherapy research may be through mHLA-DR value, which has been proved to be a good predictor to evaluate patients’ immune status and a good parameter for individualized goal-directed therapy [43].

Reductions in the relative risk of death were observed in all subgroups including those stratified according to age, sex, APACHE II score, SOFA score and levels of mHLA-DR, but without statistical significance. The aim of analyzing different prespecified subgroups in our research was to prepare for our future research in targeted specific groups of severe sepsis patients who might benefit from the T��1 treatment since it is unlikely that thymosin alpha 1 is equally beneficial to all patients in view of the significant heterogeneity in severe sepsis patients. The results of subgroup analysis in our research were inconclusive and whether T��1 is more effective in specific groups of patients with severe sepsis should be explored in trials with a larger sample size.

Types of pathogen and empirical antibiotic therapy are very important factors that affect the outcome of severe sepsis. In our study, there was no difference between groups in these perspectives. It is noted that Entinostat the origins of microorganisms are substantially diverse in different areas and even in different hospitals in the same area. So is empirical therapy. In the present study, there was a high isolation rate of gram-negative bacteria (pseudomonas, acinetobacter) compared with some other epidemiology study of infection in ICU [44].