A. Laupacis et D. Fergusson, DRUGS TO MINIMIZE PERIOPERATIVE BLOOD-LOSS IN CARDIAC-SURGERY - METAANALYSES USING PERIOPERATIVE BLOOD-TRANSFUSION AS THE OUTCOME, Anesthesia and analgesia, 85(6), 1997, pp. 1258-1267
Concern about the side effects of allogeneic red blood cell transfusio
n has increased interest in methods of minimizing perioperative transf
usion. We performed meta-analyses of randomized trials evaluating the
efficacy and safety of aprotinin, desmopressin, tranexamic acid, and e
psilon-aminocaproic acid in cardiac surgery. All identified randomized
trials in cardiac surgery were included in the meta-analyses. The pri
mary outcome was the proportion of patients who received at least one
perioperative allogeneic red cell transfusion. Sixty studies were incl
uded in the meta-analyses. The largest number of patients (5808) was a
vailable for the meta-analysis of aprotinin, which significantly decre
ased exposure to allogeneic blood (odds ratio [OR] 0.31, 95% confidenc
e interval [CI] 0.25-0.39; P < 0.0001). The efficacy of aprotinin was
not significantly different regardless of the type of surgery (primary
or reoperation), aspirin use, or reported transfusion threshold. The
use of aprotinin was associated with a significant decrease in the nee
d for reoperation because of bleeding (OR 0.44, 95% CI 0.27-0.73; P =
0.001). Desmopressin was not effective, with an OR of 0.98 (95% CI 0.6
4-1.50; P = 0.92). Tranexamic acid significantly decreased the proport
ion of patients transfused (OR 0.50, 95% CI 0.34-0.76; P = 0.0009). ep
silon-Aminocaproic acid did not have a statistically significant effec
t on the proportion of patients transfused (OR 0.20, 95% CI 0.04-1.12;
P = 0.07). There were not enough patients to exclude a small but clin
ically important increase in myocardial infarction or other side effec
ts for any of the medications. We conclude that aprotinin and tranexam
ic acid, but not desmopressin, decrease the number of patients exposed
to perioperative allogeneic transfusions in association with cardiac
surgery. Implications: Aprotinin, desmopressin, tranexamic acid, and e
psilon-aminocaprole acid are used in cardiac surgery in an attempt to
decrease the proportion of patients requiring blood transfusion. This
meta-analysis of all published randomized trials provides a good estim
ate of the efficacy of these medications and is useful in guiding clin
ical practice. We conclude that aprotinin and tranexamic acid, but not
desmopressin, decrease the exposure of patients to allogeneic blood t
ransfusion perioperatively in relationship to cardiac surgery.