Transfusions are one of the most overused treatments in modern medicine, and saving blood is one important issue all around the world. Cardiac surgery makes up a large percentage of the overall blood components consumption in surgery.

Acute normovolemic hemo-dilution (ANH) is a well-known strategy which has been used for years without the support of high quality evidence based medicine to improve post-cardiopulmonary bypass coagulation and reduce red blood cells (RBC) transfusion. We designed a multicenter randomized controlled trial to investigate the effect of ANH in reducing the number of cardiac surgery patients receiving RBC transfusions during hospital stay. We will randomize 2000 patients to have sufficient power to demonstrate a 20% relative and 7% absolute risk reduction in the number of patients' RBC transfusion. If the results of the study will confirm our hypothesis, this will have a great impact on blood management in cardiac operating room.

Inclusion Criteria:

• Signed informed consent
• Any cardiac surgical intervention on CPB
• Elective surgery

Exclusion Criteria:

• Medical decision (e.g.: planned pre-CPB ANH considered undeniable for ethical reasons or not applicable for safety issues)
• Unstable Coronary Artery Disease: Recent (< 6 weeks) myocardial infarction, unstable angina, severe (> 70%) left main coronary artery stenosis
• Critical preoperative state (ventricular tachycardia or ventricular fibrillation or aborted sudden death, preoperative cardiac massage, preoperative ventilation before anesthetic room, hemodynamic instability, preoperative inotropes or IABP, preoperative severe acute renal failure (anuria or oliguria <10ml/hr.)
• Emergency surgery
• Pregnancy
• Unfeasibility to withdraw ≥ 650 ml without inducing hemodynamic instability
• Unfeasibility to withdraw ≥ 650 ml without inducing pre-CPB anemia (Htc <30%)
• Unfeasibility to withdraw ≥ 650 ml without inducing low Htc during CPB (Htc <24%)