Thirty (0.46%) patients required one or more reoperations to control b
leeding after 6499 elective surgical procedures. A specific bleeding s
ite was identified in 10 (0.15%), and bleeding was diffuse in 20 (0.31
%), Bleeding occurred despite normal prothrombin and partial thrombopl
astin times and adequate platelet counts in all 30 cases. Diffuse blee
ding was associated with. the preoperative use of aspirin alone or in
combination with other nonsteroidal anti-inflammatory drugs (NSAIDS) i
n 19 of 20 patients (95%). None of the patients with a discrete bleedi
ng point identified at reoperation were taking these products (P < 0.0
01). More than one reoperation was more often required in the patients
with diffuse postoperative bleeding than those with a specific site o
f hemorrhage. Many of these subsequent procedures were required to man
age infections that developed after the first reoperation to control b
leeding. Intensive care unit use, length of intensive care unit stay,
total hospital stay, and hospital charges were all significantly incre
ased when diffuse postoperative bleeding was noted. Postoperative blee
ding, especially when it is diffuse rather than from a specific bleedi
ng point, significantly prolongs hospital stay and increases costs. Bl
eeding is not prevented by obtaining routine screening coagulation pro
files. A medication history with special attention to the recent use o
f aspirin and NSAIDS is advised before elective operations. Delaying s
urgery until the antiplatelet effects of these drugs have worn off may
be advisable.