Purpose: Propensity score methods have increasingly been used to reduce or minimize the confounding that frequently occurs in observational studies. We investigated risk factors after curative resection for hepatocellular carcinoma (HCC) and the association between blood transfusion and outcome using inverse probability of treatment weighting (IPTW) and the propensity score to control for confounding factors.
Methods: We analyzed prognostic factors in prospectively collected clinical data from 256 consecutive patients undergoing surgical resection for HCC after adjusting for different covariate distributions in patients who did and did not receive a transfusion.
Results: The crude 5-year disease-free survival (DFS) for patients who received a transfusion was 29%, whereas the 5-year DFS for those who did not receive a transfusion was 45% (p < 0.001). The crude 5-year overall survival (OS) for patients who received a transfusion was 66%, whereas the 5-year OS for patients who did not receive a transfusion was 81% (p = 0.003). After IPTW with 13 variables, the weighted DFS for patients who received a transfusion was significantly worse than the weighted DFS for patients who did not receive a transfusion (p = 0.049), whereas the weighted OS for patients who received a transfusion was not significantly worse than the weighted OS for patients who did not receive a transfusion (p = 0.221).
Conclusions: Blood transfusion was associated with a recurrence of HCC after hepatectomy. Surgeons could decrease the recurrence of tumors by minimizing bleeding and blood transfusions.