The objective of this study was to evaluate the effectiveness of acute normovolemic hemodilution (ANH) as a sole method of reducing allogenic blood requirement in patients undergoing primary elective valve surgery. One hundred eighty eight patients undergoing primary elective valve surgery were prospectively randomized into two groups: Group I (n=100) acted as control and in Group II (n=88) autologous blood was removed (10% of estimated blood volume in patients with hemoglobin (Hb) > 12g% and 7% when the Hb was < 12g%) in the pre-cardiopulmonary bypass (CPB) period for subsequent re-transfusion after protamine administration. The autologous blood withdrawn was replaced simultaneously with an equal volume of hydroxyl-ethyl starch solution. Banked blood was transfused in both the groups when Hb was < or = 6g% on CPB and < or = 8g% after CPB. Platelets were transfused when the count fell to < 100 x 10(9)/L and fresh frozen plasma (FFP) was transfused whenever there was diffuse bleeding with laboratory evidence of coagulopathy. The two groups were comparable as regards demographic data, type of surgical procedures performed, duration of CPB and ischemia, duration of elective ventilation and re-exploration for excessive bleeding. The autologous blood withdrawn in patients with Hb > or = 12g% was 288.3+/-69.4 mL and 244.4+/-41.3 mL with Hb < 12g% (P=NS). The Hb concentration (g%) was comparable pre-operatively (Group I = 12.1+/-1.6, Group II = 12.4+/-1.4), on postoperative day 1 (Group I = 10.3+/-1.1, Group II = 10.6+/-1.2) and day 7 (Group I = 10.9+/-1.5, Group II = 10.4+/-1.5). However, the lowest Hb recorded on CPB was significantly lower in Group II (Group I = 7.7+/-1.2, Group II = 6.7+/-0.9, P < 0.05). There was no difference in the chest tube drainage (Group I = 747.2+/-276.5 mL, Group II = 527.6+/-399.5 mL), blood transfusion (Group I = 1.1+/-1.0 units vs. Group II = 1.3+/-1.0 units intra-operatively and Group I = 1.7+/-1.2 units vs. Group II = 1.7+/-1.4 units post-operatively) and FFP transfusion (Group I = 581.4+/-263.4 mL, Group II = 546.5+/-267.8 mL) in the two groups. We conclude that low volume autologous blood pre-donation does not seem to provide any added advantage as a sole method of reducing allogenic blood requirement in primary elective valve surgery.