Coagulation disturbances in the critically ill patient with special reference to prediction of outcome

Abstract: Critically ill patients in intensive care units (ICU) often have coagulation disturbances. Clinical manifestations such as thromboembolism and/or bleeding may result from activation and consumption of coagulation factors and inhibitors. Disseminated intravascular coagulation (DIC) is a severe condition with an often-poor prognosis. Global haemostatic tests are inexpensive, widely available, and may be performed 24 hours a day. In contrast, more sophisticated coagulation tests, such as protein C, antithrombin, and activated protein C-protein C inhibitor (APC-PCI) complex, are expensive, can be performed only in a limited number of healthcare facilities specializing in coagulation disorders, and often only during office hours. In the present study 92 patients admitted to a general ICU constituted the study group. The inclusion criteria consisted of one or more of the following: platelet count <100×10e9 /L; International Normalized Ratio (INR) >1.36; or activated partial thromboplastin time (APTT) >45 seconds. A control group with a comparable age and sex distribution, but not fulfilling the laboratory criteria, was established. Groups were followed up to 180 days. Survival upon discharge from the ICU and hospital was significantly lower in the study group, especially in patients with disorders that can be related to internal medicine. Odds ratios showed that prolonged APTT, in particular, predicted a lower survival rate at the time points evaluated. The coagulation inhibitors protein C and antithrombin were also analysed. Univariate analysis of variance showed that INR and APTT were independent predictors of protein C and, to some extent, antithrombin. Using a Cox regression model, decreased protein C, but not antithrombin, predicted lower survival at 5 years. Using the same patient material in a theoretical model assuming best-case scenario, we hypothesized that protein C treatment of patients with low levels would increase survival to the same level as a cohort with higher protein C. Total costs per life saved at 30 days (upper limit of 95 % CI) were calculated at SEK 408,000 [recombinant human activated protein C; drotrecogin alfa (activated)] and 435,000 (protein C zymogen), which may be compared to the value of a statistical life at the same age (SEK 7.66 million). We also studied 38 patients who underwent aortic surgery. The level of the coagulation test APC-PCI complex was very high immediately following surgery and remained so, although declining, during the first two days. Higher complex levels at 6-12 hours and 12-18 hours after admission to the ICU were observed in patients who did not survive their stay in the unit.