Haematological effects of reinfused mediastinal blood after cardiac surgery

To explore the risk of inducing a coagulation defect in cardiac surgery patients by the reinfusion of mediastinal blood.

Ten patients who underwent coronary artery surgery were prospectively studied for the haematological effects of reinfusion of blood drained from the chest drain tubes after the operation by a Sorenson autologous transfusion system.

Surgery was performed at a private hospital and patient selection was made at the time of reinfusion.

Nine patients had primary coronary artery surgery and one had a reoperation.

Blood samples were taken from the patients before reinfusion, one hour after reinfusion, and 24 hours later as well as from the collected blood. Measurements were made of the haemoglobin content, white cell and platelet counts, fibrinogen and fibrinogen degradation products, D-dimer, antithrombin III and plasma haemoglobin content. Estimations were also made of the prothrombin time, the thrombin clotting time and the activated partial thromboplastin time. The hypothesis to be tested by this study was that the reinfusion of mediastinal blood after the operation did not cause any significant disturbance of the patient's blood clotting profile.

An average of 535 mL (range, 400-950 mL) was reinfused after a period of three hours drainage (range, 45 minutes to 5 hours). While the initial patient samples contained a raised plasma haemoglobin level (0.19 g/L) significantly related to the cardiopulmonary bypass time (P less than 0.001), these were free of fibrinogen degradation products except for a sample from one patient who had a reoperation. The blood drained by the Sorenson system was lower in haemoglobin content (7.7 g/dL), and had a significant content of fibrinogen degradation products (147 mg/L) and D-dimer (6.4 mg/L) together with reduced clotting factors when compared with the patients' blood. After reinfusion, the patient sample showed evidence of altered coagulation with mildly increased clotting times (activated partial thromboplastin time 57 s, thrombin clotting time 123 s), the extent of which was related to the volume reinfused (P less than 0.001), but 24 hours later, these effects had all disappeared. All samples were sterile in aerobic and anaerobic culture media.

We conclude that the Sorenson system of retrieval of mediastinal blood after cardiac surgery provides a safe and simple method of blood conservation provided that the volume of reinfusion is not excessive.