A quantitative and qualitative comparison of fibrin glue, albumin, and blood as agents to pretreat porous vascular grafts

Abstract

Recent reports suggest that fibrin glue can be used to seal porous vascular grafts prior to insertion, but this ability has not been quantitatively compared to existing methods. We compared blood loss from and handling characteristics of grafts pretreated with either fibrin glue (FG) (Tisseel), albumin autoclaving (AA), or blood preclotting (BP). Five 6-cm segments of 6-mm internal diameter grafts, both knitted and woven double velour Dacron were treated in each group (30 specimens). Human blood was forced through the BP group until clotted; AA segments were soaked in 25% human albumin and autoclaved for 10 min; FG segments were treated with a topical application of Tisseel (0.5 ml/graft) followed by treatment with topical thrombin + CACl (0.5 ml/graft). Graft ends were sealed and attached to a transducer/syringe pump mechanism which pumped heparinized human blood into the graft at 100 mm Hg intraluminal pressure. All blood that leaked through the grafts over 2 min was collected and the amount was averaged for the five grafts in each group. Graft handling was characterized as either pliable or stiff. Blood pretreatment caused 21 ± 2 and 13 ± 4 cc/2 min of leak in knitted and woven grafts, respectively. Albumin autoclaving resulted in 9 ± 2 and 1 ± 0.5 cc of leak ( P < 0.01 compared to blood), while fibrin glue produced 2 ± 2 and 0.4 ± 0.5 cc leaks ( P < 0.01 compared to blood). Both blood and fibrin glue produced soft pliable grafts, while albumin pretreatment resulted in stiff grafts. We conclude that fibrin glue or albumin is superior to blood for pretreatment of woven grafts in limiting blood loss, but that fibrin glue is superior to either albumin or blood in knitted grafts. Fibrin glue imparts superior handling characteristics.