Universal changes in biomarkers of coagulation and inflammation occur in patients with severe sepsis, regardless of causative micro-organism [ISRCTN74215569]

Abstract

Introduction: PROWESS (Recombinant Human Activated Protein C Worldwide Evaluation in Severe Sepsis) was a phase III, randomized, double blind, placebo controlled, multicenter trial conducted in patients with severe sepsis from 164 medical centers. Here we report data collected at study entry for 1690 patients and over the following 7 days for the 840 patients who received placebo (in addition to usual standard of care).

Methods: Nineteen biomarkers of coagulation activation, anticoagulation, fibrinolysis, endothelial injury, and inflammation were analyzed to determine the relationships between baseline values and their change over time, with 28-day survival, and type of infecting causative micro-organism.

Results: Levels of 13 of the 19 biomarkers at baseline correlated with Acute Physiology and Chronic Health Evaluation II scores, and nearly all patients exhibited coagulopathy, endothelial injury, and inflammation at baseline. At study entry, elevated D-dimer, thrombin-antithrombin complexes, IL-6, and prolonged prothrombin time were present in 99.7%, 95.5%, 98.5%, and 93.4% of patients, respectively. Markers of endothelial injury (soluble thrombomodulin) and deficient protein C, protein S, and antithrombin were apparent in 72%, 87.6%, 77.8%, and 81.7%, respectively. Impaired fibrinolysis (elevated plasminogen activator inhibitor-1) was observed in 44% of patients. During the first 7 days, increased prothrombin time (which is readily measurable in most clinical settings) was highly evident among patients who were not alive at 28 days.

Conclusion: Abnormalities in biomarkers of inflammation and coagulation were related to disease severity and mortality outcome in patients with severe sepsis. Coagulopathy and inflammation were universal host responses to infection in patients with severe sepsis, which were similar across causative micro-organism groups.

Figure 1

The time courses of biomarkers of coagulation in placebo-treated patients with severe sepsis in the PROWESS study are shown here as means and 95% confidence intervals, using standard error of the mean and repeated measures analysis without imputing for missing data. The number of observations for each time point appears below the x-axis for (●) survivors and (○) nonsurvivors. *P < 0.05 versus nonsurvivors. APTT, activated partial thromboplastin time; PT, prothrombin time.

Figure 2

Time courses of biomarkers of anticoagulants in placebo-treated patients with severe sepsis in the PROWESS study are shown here as means and 95% confidence intervals, using standard error of the mean and repeated measures analysis without imputing for missing data. The number of observations for each time point appears below the x-axis for (●) survivors and (○) nonsurvivors. *P < 0.05 versus nonsurvivors.

Figure 3

The time course of biomarkers of thrombin generation and fibrinolysis in placebo-treated patients with severe sepsis in the PROWESS study are shown here as means and 95% confidence intervals, using standard error of the mean and repeated measures analysis without imputing for missing data. The number of observations for each time point appears below the x-axis for (●) survivors and (○) nonsurvivors. *P < 0.05 versus nonsurvivors. F1.2, prothrombin fragment 1.2; PAI, plasminogen activator inhibitor; TAFI, thrombin-activatable fibinolysis inhibitor; TAT, thrombin–antithrombin complex.

Figure 4

Time courses of biomarkers of inflammation and endothelial damage in placebo-treated patients with severe sepsis in the PROWESS study are shown here as means and 95% confidence intervals, using standard error of the mean and repeated measures analysis without imputing for missing data. The number of observations for each time point appears below the x-axis for (●) survivors and (○) nonsurvivors. *P < 0.05 versus nonsurvivors. sTM, soluble thrombomodulin.

Figure 5

Time courses of biomarkers of coagulation and inflammation in placebo-treated patients with severe sepsis for selected causative micro-organisms are shown here as mean and 95% confidence intervals, using standard error of the mean and repeated measures analysis without imputing for missing data. The number of observations at each time point appears below the x-axis by causative micro-organism class. *P < 0.05 (based on repeated measures analysis).