The clinical importance of viscoelastic testing in patient blood management when performing cardiovascular surgery is increasing. We aimed to examine the effect of a blood transfusion protocol including an assessment of fibrin-based rotational thromboelastometry on transfusion volume, mortality, and bleeding complications in patients undergoing cardiac or thoracic aortic surgery. We retrospectively studied a cohort of 376 consecutive patients who underwent cardiopulmonary bypass before (control group: 150 cardiac and 35 thoracic aortic surgeries) and after (assessment group: 154 cardiac and 37 thoracic aortic surgeries) introducing the fibrin polymerization assessment with thromboelastometry in the blood transfusion protocol. The transfusion volume and clinical outcomes were compared between the control and assessment groups, and the standardized (mean) difference (S[M]D) was calculated as an indicator of statistical effect size. Compared with the control group, the assessment group had a lower total blood transfusion volume (mL) in cardiac (2720 ± 1282 vs. 2034 ± 1330, p < 0.0001, [SMD] = 0.68) and thoracic aortic surgeries (5236 ± 2732 vs. 3714 ± 1768, p < 0.0001, SMD = 0.67). The 1-year mortality rates were 1.9 % and 2.7 % in cardiac and thoracic aortic surgeries, respectively. Significant differences were not observed in the 1-year mortality (3.2 % vs. 1.0 %, p = 0.16, relative risk [RR] = 0.32 with 95 % confidence intervals [CI] = 0.06–1.57, SD = 0.15), re-exploration for bleeding (4.8 % vs. 2.6 %, p = 0.28, RR = 0.53 with 95 % CI = 0.18–1.57, SD = 0.12), and major bleeding (17.3 % vs. 13.0 %, p = 0.31, RR = 0.75 with 95 % CI = 0.46–1.22, SD = 0.12) rates between the control and assessment groups. The assessment of fibrin polymerization with thromboelastometry using the blood transfusion protocol reduced the blood transfusion volume in cardiovascular surgery.