Seroma formation is a difficult problem to treat and prevent. Its sequ
elae include wound infection, dehiscence, and skin-flap necrosis. The
purpose of this study was to determine the effects of fibrin sealant o
n seroma formation and wound healing. Seromas were created in a rat mo
del by harvesting the latissimus dorsi muscle. In group I (n = 20), on
ly the latissimus dorsi muscle was harvested. In group II (n = 20), th
e latissimus dorsi muscle was harvested and fibrin sealant applied. Se
romas were routinely aspirated. In group III (n = 20), the latissimus
dorsi muscle was harvested, and once a seroma was evident clinically,
it was aspirated and injected with fibrin sealant. Fibrin sealant was
created by combining virally deactivated fibrinogen and thrombin (Amer
ican Red Cross, Rockville, Md.). In group I, 90 percent of the animals
formed seromas compared with only 20 percent in group II. The average
total fluid aspirated in group I was 21 cc versus 6 cc in group II. S
ixty percent of the animals in group I and 5 percent in group II requi
red serial drainage for chronic seromas. Skin-flap necrosis occurred i
n 80 percent of the animals in group I, in 10 percent of group II, and
in 40 percent of group III. Histologic evaluation confirmed that grou
p II underwent better wound healing. At necropsy, group I animals with
seromas had gross capsular formation; this was not readily apparent i
n the fibrin sealant groups. We conclude that (1) the harvesting of th
e rat latissimus dorsi muscle is a reliable model for creating seromas
, (2) fibrin sealant effectively prevents seroma formation when applie
d intraoperatively, (3) wound healing in the seroma rat model is impro
ved with intraoperative fibrin sealant application, (4) closed injecti
on of fibrin sealant for existing seromas cannot be recommended at thi
s time, (5) virally deactivated fibrin sealant retains its hemostatic
and adhesive properties, and (6) current clinical trials of virally de
activated fibrin sealant may facilitate its use in the United States.