PurposeHepatic angiosarcoma(HAS) is rare and is the primary vascular-derived malignancy of the liver. Its clinical characteristics, therapeutic strategy management, and the outcome are unclear. MethodsThis is a retrospective study of patients from the SEER database. Survival analysis was performed by the Kaplan-Meier method. Univariate and multivariate Cox models were used to identify risk factors. Propensity Score Matching(PSM) analysis was used to remove confounding bias. The nomogram was constructed, and the performance was measured using the C-index. ResultsA total of 300 HAS patients diagnosed between 1975 and 2016 were identified for this study, with an estimated median cancer-specific survival(CSS) of 1 month. The median CSS was 6 months in both the surgery and chemotherapy groups. Age(HR = 1.5206, p = 0.0058), sex(HR = 1.3906, p = 0.0391), SEER stage(HR = 1.4426, p < 0.0001), surgery(HR = 0.4493, p = 0.0001) and chemotherapy(HR = 0.28161, p < 0.0001) are potential independent prognostic factors. Of these HAS patients, 29 received surgical treatment without chemotherapy, and 63 received chemotherapy without surgery. A 1:1 PSM was performed to select candidates from the surgery-only group and the chemotherapy-only group. The survival analysis showed that the median survival time was 3 months in the surgery-only cohort and 5 months in the chemotherapy-only cohort, and there was no statistical difference between the two groups. Finally, a nomogram was constructed, with a C-index of 0.754. ConclusionsHAS is uncommon and has a poor prognosis. It was found that age, sex, SEER stage, surgery, and chemotherapy were independent prognostic factors for patients. Both surgery and chemotherapy could significantly prolong the survival of patients, and there was no statistical difference between the prognosis of patients treated with chemotherapy alone and those treated with surgery alone.