Coagulation profile in hyperglycemic septic shock patients

Introduction

A hypercoagulable state is frequently found in sepsis patients as well as hyperglycemia [1]. The objective of this study was to determine the coagulation profile in hyperglycemic septic shock patients and the impact of glycemia control.

Methods

Septic shock patients with less than 48 hours of organ dysfunction with normoglycemia (<150 mg/dl, Group 1) or hyperglycemia (>200 mg/dl, Group 2) were included. Exclusion criteria were previous diabetes, coagulation diseases, use of heparin or fresh frozen plasma/platelets, sepsis in the past 30 days or insulin use in the past 24 hours. The coagulation and inflammation profile was determined at inclusion and after 24 hours, only if the patients remained normoglycemic or achieved glycemic control with insulin infusion (Group 2). Results were expressed in median and interquartiles, P < 0.05 was considered significant.

Results

Forty-one patients were included, 59.7% male; mean APACHE II score: 16.9 ± 8.6, P = 0.21; mean SOFA score: 8.1 ± 2.0; mean number of organ dysfunctions: 3.6 ± 0.9, with a significant difference between the groups only regarding age (Group 1 (n = 21): 53.8 ± 16.7, Group 2 (n = 20): 68.8 ± 16.4, P = 0.006). In both groups, a hypercoagulation profile was detected, with high levels of fragment 1+2 (F1+2), thrombin/antithrombin complex (TAT), D-dimer and tissue factor and with low levels of factor VII. Consumption of coagulation inhibitors (protein C (PC), antithrombin (AT), tissue factor pathway inhibitor) and increased fibrinolysis (plasminogen activator inhibitor 1 (PAI-1), tissue plasminogen activator (TPA) and plasminogen) were also found. However, at baseline no significant difference between groups was found except for a trend towards high levels of F1+2 (P = 0.08) and TAT (P = 0.058) in Group 2. After 24 hours higher levels of AT (P = 0.02), PC (P = 0.02), plasminogen (P = 0.04) and a reduction in PAI-1 (P = 0.01) and TPA (P = 0.03) were detected in Group 2. Inflammatory parameters (IL-6, IL-8, IL-10, TNF) were similar at baseline, with decreased levels in both groups after 24 hours but the IL-6 (P = 0.02) and IL-8 (P = 0.0004) reduction was more pronounced in Group 2.

Conclusions

Septic shock patients have a hypercoagulation pattern regardless of their glycemic status. However, glycemia control seems to be associated with reduction in inflammation and hypercoagulation parameters, possibly related to insulin infusion as glycemia levels did not differ between groups after 24 hours.