Table 19. Summary of transfusion rate according to procedure and predonation policy. Transfusion rate (n/N) Procedure Transfusion rate (%)* Predonation No predonation Predonation No predonation statistical model n/N n/N % % - - ** ** case-reports 35/118 19/408 26,3 4,5 probit-normal Le Fort I single jaw complex 3/54 8/105 5,6 7,6 logistic regression Bimaxillary surgery simple 342/969 187/1264 39 12,5 probit-normal Bimaxillary surgery complex 22/219 20/120 15,3 25,7 probit-normal BSSO single jaw Le Fort I single jaw simple * The % transfusion rate results from the statistical model used. ** Reported transfusions after BSSO-surgery reflect cases with excessive accidental peroperative bleeding. Where n is the number of patient and N is the total number of patient. differences (12.5% vs 25.7% in the probit-normal statistical model). The reasons are wide confidence intervals and small sample sizes. The intra-study correlation in the simple bimaxillary procedure group without predonation was 0.32 (p=0.0047). The intra-study correlation in the complex bimaxillary procedure group without predonation was 0.30 (p=0.0941). The clinical interpretation of the statistical result is that we failed to prove a significantly different transfusion behaviour beween simple and complex bimaxillary procedures. It should not be interpreted as having been proven that there is no difference in transfusion rate. 4. Discussion and Conclusions Transfusion for BSSO surgery is rather independent of the duration of surgery and will be necessary only in the event of a vascular injury. Several techniques are implemented to reduce the blood loss and subsequent blood transfusion in orthognathic surgery. This review showed a ‘very low level of evidence’ that the deliberate hypotensive anesthesia is indeed correlated with less blood loss, but it remains one of the most commonly used techniques during maxillary surgery. 57-59 More than any other factor this review showed that transfusion policy, rather than strict criteria, initiates the decision to transfuse. This trend is endorsed in the paper by Faverani et al. 41 who suggest that the indication of blood replacement should be based not only on laboratory parameters (primarily, reduced Hb and Hct levels) but also on clinical signs indicative of a true need for transfusion, such as tachycardia, tremor, diaphoresis, and malaise. If a liberal reinfusion strategy of autologous blood is avoided, single Le Fort I surgery without additional or complex procedures has historically been accompanied by a blood transfusion need of about 4.5 %, and in the case of additional procedures, about 7.6% or less depending on the criteria for transfusion. There seem 2F&RfVV6PbFFF&6VGW&W2vWFW"B66W&06VvVFF"w&gFrFRG&6gW6VVB6vRRf'B7W&vW'2r27G&7BG&6gW67&FW&&RfvVB&'7W&vW'2W72FWVFVBFRFWFbFV6B&FW"FFFV7W&W0FW76V&B72GW&r7W&vW'6FV&'&6W2r&'7W&vW'vFW@6W"FFF&6VGW&W24FVG0vFG&6gW6VVB2r2( 3"RFWVFpFR7&FW&f"G&6gW6BFRf&ƗG`&VFFVB&BF2G&6gW6&FR6&V6WfV22v23>( 33RR6W&'7W&vW'( Gv62&V6֖p&Rg&WVVB6FV&''FvF07W&vW'FRfVBbV6VBf667VGW&rVF6VvVFRf'B7FVF֖W2B&vP&'fVVG2vF'fW7FrbƖ07&W7B&Rw&gG2VF6ǒ&R6&֗6V@FVG2vF4>( Gv&6fW&&@G&6gW6&FRb&WB#bRB7W'72FRPƖ֗BFBvW&6VFW'2W6RFffW"&VFFƖ7FFV"FVG2E$4eU4TTB%DtD27W&vW'$UdUpWF"6G&'WF05FW6vVBFR7GVGBw&FRFRW67&CB&WfWvVBFRW67&B6vVFvVG0FRWF'2FV6&R6fƖ7BbFW&W7B&VFV@FF27GVGFW&R&R6fƖ7G2bFW&W7B@f6FW&W7G2F&RF666VB&VfW&V6W0"66Ɩ2&W6W"6Ɩ6F2BfW&W0'FvF27W&vW'&Wb&VvRVBFVBB##SrsӃ"V$VEХV&wV"6'FvF&&Wv&6v&6&B72'FvF27W&vW'77FVF2&WfWr&f27W&r#c2RУ"Fb2#rgVFWBƖ5ХV$VEХ7FFwVGRW&2BFG6D"&B72fvr'FvF27W&vW'f&W2vFVǒB6WFW2G&6gW62&RVVFVBWfB&6VBFVB&7B#BRbFbদV&G#BgVFWBƖ5V$VEФ,;gGFvW"2fW7FvF2FRVVBf"G&6gW6'FFF27W&v6FW'fVF2VFW"66FW&FbFfGVFVB&WFW'2vW&F76W'FFvW76VDRW7GW2ƖV&rVfW'6L:GBvW76V#r