Abstract
BACKGROUND: The use of thromboelastography (TEG) has demonstrated decreased blood product utilization in patients with specific etiologies of major gastrointestinal bleeding (GIB), such as variceal and non-variceal bleeding in cirrhosis patients; however, in a non-cirrhosis patient with GIB, there is far less evidence in the literature. Our retrospective study compares the effect of TEG-guided blood product utilization in patients with major GIB with all etiologies, including cirrhosis, admitted to medical intensive care unit (MICU). METHODS: A retrospective chart review was conducted on patients admitted to the MICU of a tertiary academic medical center diagnosed with GIB using ICD-9/10 codes from 2014 to 2018. A total of 1,889 patients were identified, and validation criteria such as "GI or hepatology consult note", type and screen, pantoprazole, or octreotide drip" were used, which resulted in 997 patients, out of which 369 had a diagnosis of cirrhosis. Propensity score matching was done for baseline variables (age, sex, and race), ICU length of stay, hospital length of stay, ventilator days, and vasopressor use. As a result, 88 patients were included in the final analysis, with 44 in TEG and 44 in non-TEG group. A sub-group analysis was done in 46 patients with cirrhosis, 23 in TEG group and 23 in non-TEG group after propensity score matching. RESULTS: There was significantly higher total blood volume (4,207 mL vs. 2,568 mL, P = 0.04) in the TEG group as compared to the non-TEG group, including total volume of cryoprecipitate (80 mL vs. 55 mL, P = 0.03) and total volume of platelet (543 mL vs. 327 mL, P = 0.03). In the cirrhosis sub-group, there was no significant difference in the amount of blood products transfused between the two groups. CONCLUSION: This study revealed that TEG is not superior to conventional coagulation parameters in limiting the volume of blood product transfusion in major GIB patients in ICU settings.