Abstract
BACKGROUND/AIMS: Variceal bleeds occur due to portal hypertension and not primary hemostatic dysfunction. However, these patients frequently have low platelet counts and elevated international normalized ratio (INR) values, for which they often receive transfusions upon admission despite guidance from multiple liver societies. METHODS: We performed a retrospective analysis using the TriNetX United States Collaborative Network. Cohorts were defined based on transfusion of fresh frozen plasma (FFP), platelets, or cryoprecipitate within 72 h of variceal bleeding diagnosis. Propensity score matching was used to balance demographic and clinical characteristics including variables representative of the model for end-stage liver disease score, Child-Pugh class, baseline hemoglobin, and receipt of packed red blood cells. Outcomes included mortality, intubation, intensive care unit (ICU) admission, hospital readmission, and transjugular intrahepatic portosystemic shunt (TIPS) placement at 6 weeks and 6 months. RESULTS: Of 63,140 patients with variceal bleeding, 5350 received blood products. After matching, each cohort contained 5246 patients. At 6 weeks, transfusion was associated with higher odds of intubation (adjusted odds ratio [aOR]: 1.83; 95% confidence interval [CI]: 1.54-2.16), ICU admission (aOR: 1.49; 95% CI: 1.27-1.74), hospital readmission (aOR: 1.74; 95% CI: 1.52-1.99), TIPS (aOR: 1.92; 95% CI: 1.59-2.33), and mortality (aOR: 3.36; 95% CI: 3.03-3.73). At 6 months, these associations persisted for intubation (aOR: 1.54; 95% CI: 1.33-1.78), hospital readmission (aOR: 1.51; 95% CI: 1.37-1.68), TIPS (aOR: 1.79; 95% CI: 1.50-2.13), and mortality (aOR: 2.36; 95% CI: 2.13-2.62), while ICU admission was not significantly different between groups (aOR: 1.07; 95% CI: 0.93-1.23). CONCLUSION: In this large, real-world cohort, transfusion of platelets, cryoprecipitate, or FFP in the setting of variceal bleeding was associated with higher odds of adverse outcomes including intubation, hospital readmission, TIPS procedures, and mortality at both 6 weeks and 6 months. These associations underscore the need for careful, individualized consideration of hemostatic product use in cirrhotic patients and highlight the importance of further prospective studies to clarify optimal transfusion practices.