Forty-one RCTs (n=3,434) were included.
In 13 studies, treatment groups were comparable for 4 or more of 8 prognostic criteria for blood loss. Thirteen studies were double-blinded. Seven studies used low-dose heparin in the HBC group. The median Jadad score was 2.7 (range: 2 to 5). Sixteen studies were classified as high quality.
There was no statistically significant difference in the amount of post-operative blood loss at 24 hours for patients allocated to HBCs compared with NHBCs; the WMD was -83.7 mL (95% CI: -217.6, 50.2), based on 7 studies (n=537). The amount of blood loss until drain removal was significantly less for patients allocated to HBCs than for those allocated to NHBCs; the WMD was -164.2 mL (95% CI: -262.1, -66.3; p=0.001), based on 13 studies (n=562).
There was no statistically significant difference between treatments in the amount of post-operative blood product transfused (packed red blood cells, fresh frozen plasma and platelets analysed separately; results were reported). The proportion of patients who received a blood product transfusion was significantly less for patients allocated to HBCs compared with NHBCs; the OR was 0.8 (95% CI: 0.6, 0.9, p=0.004), based on 22 studies (n=2,415).
There was no statistically significant difference between groups in specific adverse events, apart from re-sternotomy which was significantly reduced in patients allocated to HBCs (OR 0.6, 95% CI: 0.4, 0.8, p=0.002).
HBCs were associated with a significant reduction in the duration of ventilation, ICU LOS and hospital LOS (data were reported).
Statistically significant heterogeneity was detected for blood loss, blood product transfusion and ICU LOS. There was no significant difference in blood loss up to drain removal in the double-blind studies.
Funnel plots showed no clear evidence of publication bias.