(Circulation. 1999;99:81-89.)
© 1999 American Heart Association, Inc.
Clinical Investigation and Reports |
-Aminocaproic Acid as Effective as Aprotinin in Reducing Bleeding With Cardiac Surgery?
From the Departments of Surgery (J.J.M., N.J.O.B., J.D.B., L.J.D.) and Medicine (G.T.O.), Center for the Evaluative Clinical Sciences (G.T.O., J.D.B.), Dartmouth Medical School, Hanover, NH, and the VA Outcomes Group (J.J.M., J.D.B.), Veterans Affairs Medical Center, White River Junction, Vt.
| Abstract |
|---|
|
|
|---|
-aminocaproic acid, an alternative
antifibrinolytic, is considerably less expensive. Because the results
of 3 small randomized clinical trials comparing these 2 agents directly
were inconclusive, we performed a meta-analysis to compare the
relative effectiveness and adverse-effect profile of these 2 agents
against placebo.
Methods and ResultsData from 52 randomized clinical trials
published between 1985 and 1998 involving the use of
-aminocaproic
acid (n=9) or aprotinin (n=46) in patients undergoing cardiac surgery
were abstracted. Our primary outcomes were total blood loss, red blood
cell transfusion rates and amounts, reexploration, stroke, myocardial
infarction, and mortality. The meta-analysis revealed
substantial reductions in total blood loss with
-aminocaproic acid
and low-dose aprotinin (each with a 35% reduction versus placebo,
P<0.001) and high-dose aprotinin (53% reduction,
P<0.001). There were identical reductions in total
postoperative transfusions with
-aminocaproic acid (61% reduction
versus placebo, P<0.010) and high-dose aprotinin (62%
reduction, P<0.001). The proportion of patients
transfused was similarly reduced with
-aminocaproic acid (OR, 0.32;
95% CI, 0.15 to 0.69) and high-dose aprotinin (OR, 0.28; 0.22 to
0.37). Although both drugs reduced rates of reexploration to similar
degrees, this effect was statistically significant only with high-dose
aprotinin (OR, 0.39; 0.24 to 0.61).
-Aminocaproic acid and aprotinin
had no effect on risks of postoperative myocardial infarction or
overall mortality.
ConclusionsBecause the 2 antifibrinolytic agents appear to have
similar efficacies, the considerably less-expensive
-aminocaproic
acid may be preferred over aprotinin for reducing hemorrhage
with cardiac surgery.
Key Words:
-aminocaproic acid aprotinin hemorrhage meta-analysis surgery
| Introduction |
|---|
|
|
|---|
60% to
75% of patients undergoing open heart surgery receive
transfusions.2 3 Re-exploration for hemorrhage is
required in 2% to 6% of patients, with case-fatality rates ranging
from 10% to 22%.4 5 In addition to its impact on
patients' morbidity and mortality, bleeding and reexploration consume
considerable resources as a result of increased operative time, blood
product use, and prolonged intensive care unit and total hospital
stays.
Prophylactic antifibrinolytic therapy is one approach to
reducing postoperative hemorrhage. The most thoroughly
evaluated antifibrinolytic agent is aprotinin. Its effectiveness in
reducing postoperative hemorrhage has been established in >40
randomized clinical trials and 2 meta-analyses.2 6
However, aprotinin is expensive:
$1000 for the conventional
high-dose regimen (6 million KIU [kallikrein inhibiting units]) and
$500 for the low-dose regimen (3 million KIU). In addition, there
are concerns about potential thrombosis-related side effects and
allergic reactions with antifibrinolytic agents. For these reasons,
prophylactic use of aprotinin is often limited to patients
at high risk for bleeding (eg, reoperations) or those for whom
transfusions are not acceptable because of religious beliefs (eg,
Jehovah's Witnesses).
An alternative to aprotinin is
-aminocaproic acid, a synthetic
antifibrinolytic. Although it shares some of the potential side effects
of aprotinin,
-aminocaproic acid is considerably cheaper
(
$40/average dose). Three randomized clinical trials have directly
compared
-aminocaproic acid and aprotinin in cardiac surgery.
However, these studies were inconclusive owing to limited sample
size.7 8 9 For this reason, we performed a
meta-analysis to assess the relative effectiveness and
side-effect profiles of the 2 antifibrinolytic agents compared with
placebo.
| Methods |
|---|
|
|
|---|
-aminocaproic acid," "aprotinin,"
"cardiac surgery," and "randomized." In addition to inspecting
bibliographies of published review articles, we performed a
bibliographic citation search using the SCI-EXPANDED database of the
Web-based Institute for Scientific Information Citation Database. The
search spanned the time interval between 1985 and 1998. Only
English-language published articles were evaluated.
Trials included in this meta-analysis were selected for the
following characteristics: (1) random allocation to drug treatment
(high-dose aprotinin, low-dose aprotinin, and/or
-aminocaproic acid)
or placebo/control groups; (2) based exclusively on patients undergoing
routine cardiac procedures (ie, primary or repeat CABG, valve
replacement or repair, or both); and (3) explicit evaluation of
1 of
the following outcomes: total blood loss, amounts and/or rates of
transfusion, reexploration, postoperative stroke, myocardial
infarction, renal impairment, allergic hypersensitivity reactions, and
mortality.
Outcomes Extraction
Outcomes data were extracted from each trial and entered into a
computer file for analysis. Only published data were included.
Data presented ambiguously in 4 studies were clarified by
direct discussion with the primary authors.3 9 10 11
We evaluated blood loss by assessing total (intraoperative and postoperative) amounts in milliliters. To examine transfusion requirements, we considered only homologous packed red blood cell (pRBC) transfusions. Few trials provided sufficient information to allow us to examine use of autologous blood, platelets, cryoprecipitate, and plasma. For each trial, we abstracted the total number of pRBC units transfused in treatment and control groups and the proportion of patients transfused in each group. Transfusion requirements recorded in milliliters were transformed to pRBC units by a conversion factor of 350 mL/U. All instances of reexploration in the trials were included, regardless of the specified indication for surgery (eg, cardiac tamponade versus chest tube bleeding versus other) or the operative findings (eg, surgical versus nonsurgical bleeding). Stroke, myocardial infarction, renal impairment, and allergic reaction were categorized as defined by the individual trials.
Statistical and Stratified Analyses
To calculate intervention effects on bleeding and transfusion
requirements, the mean and SDs of blood loss and pRBC units transfused
were recorded for treatment and placebo/control groups in each
trial. Using the methods described by Cochrane, we then calculated
summary measures of absolute (mL and U) and relative (%) reductions in
mean total blood loss and pRBC transfusions in patients receiving the
active drug.12 For dichotomous outcomes (transfusion,
reexploration, postoperative myocardial infarction, stroke, renal
impairment, allergic reactions, and mortality), the total number of
patient outcomes in the drug and placebo groups was recorded in
2x2 tables. We added 0.5 to every cell in any table containing 0 to
improve bias and precision properties.13 The summary
measure of effect used for the dichotomous outcomes was the Peto OR
described by Yusuf et al.14 To account for variance
between studies, random-effects models were used for continuous and
dichotomous summary estimates of effect.15
Stratified analysis was performed to assess whether treatment effects varied according to whether or not trials were restricted to primary CABG or included patients undergoing "mixed" procedures (including CABG, valve, reoperation, and other cardiac procedures). We also performed stratified analysis according to specific study-design criteria, such as whether trials were placebo controlled or double-blinded.
| Results |
|---|
|
|
|---|
-aminocaproic acid. Overall,
71% were double-blind, 71% were placebo-controlled, and 8% were
multicenter studies.
|
|
|
|
Hemorrhage-Related Outcomes
Compared with their placebo/control counterparts, patients
receiving antifibrinolytic therapy had substantially lower total blood
loss (Figure 1A
).
-Aminocaproic acid
and low-dose aprotinin each reduced total blood loss by 35%
(P<0.001), whereas high-dose aprotinin reduced blood loss
by 53% (P<0.001).
|
-Aminocaproic acid was as effective in reducing transfusion
requirements after cardiac surgery as aprotinin. Compared with their
placebo counterparts, patients given
-aminocaproic acid received
61% fewer pRBC transfusions (mean reduction, 0.74 U; Figure 1B
). Patients receiving
-aminocaproic acid were only 32% as
likely to receive any pRBC transfusions (OR, 0.32; 95% CI, 0.15 to
0.69; P=0.004). These reductions in transfusion requirements
were nearly identical to those observed with the high-dose aprotinin
groups (the Table
and Figure 2
).
|
|
Patients receiving prophylactic antifibrinolytic
therapy also were less likely to undergo reexploration for bleeding
after cardiac surgery. This reduction was only statistically
significant for high-dose aprotinin (OR, 0.39; 95% CI, 0.24 to
0.61; P<0.001). Because of smaller sample sizes with
-aminocaproic acid and low-dose aprotinin, effects of approximately
similar magnitude did not reach statistical significance (OR, 0.54;
95% CI, 0.22 to 1.33; P=0.179; and OR, 0.56; 95% CI, 0.29
to 1.08, P=0.085, respectively; Figure 2
).
Other Outcomes
Although there were consistent trends toward reduced
stroke rates in patients receiving antifibrinolytics, these trends did
not reach statistical significance (
-aminocaproic acid: OR, 0.47;
95% CI, 0.15 to 1.56; P=0.22, high-dose aprotinin: OR,
0.56; 95% CI, 0.26 to 1.21; P=0.14; and low-dose aprotinin:
OR, 0.54; 95% CI, 0.20 to 1.45; P=0.22; Figure 2
).
Prophylactic use of
-aminocaproic acid or high- or
low-dose aprotinin was not associated with significant differences in
the incidence of postoperative myocardial infarction or overall
mortality (the Table
and Figure 2
). There were
insufficient data to evaluate the effect of
-aminocaproic acid on
risks of renal impairment and allergic reactions. Although information
pertaining to these outcomes was also limited in the aprotinin trials,
there was a statistically insignificant trend toward increased rates of
renal dysfunction with high-dose aprotinin (OR, 1.46; 95% CI, 0.92 to
2.33; P=0.11) but not low-dose aprotinin (OR, 1.01; 95% CI,
0.65 to 1.57; P=0.96). High-dose aprotinin was associated
with a similar, insignificant trend toward increased risk of allergic
reactions (OR, 1.58; 95% CI, 0.59 to 4.23; P=0.36).
Stratified Analysis
Stratified analysis did not reveal important
interactions between drug effectiveness and characteristics of the
trials. Estimates of the effectiveness of high- and low-dose aprotinin
and
-aminocaproic acid in reducing transfusion requirements were
similar when analysis was stratified according to whether
studies were placebo-controlled or double-blind (data not shown).
Furthermore, drug effectiveness in reducing transfusions was similar
between trials restricted to primary CABG and those enrolling patients
undergoing "mixed" procedures. Stratified analysis of less
frequent outcomes, including reexploration, could not be performed
(meaningfully) because of sample-size limitations.
| Discussion |
|---|
|
|
|---|
-aminocaproic acid have only recently become
available. When we incorporated several new trials involving this drug,
our meta-analysis suggested that
-aminocaproic acid had
similar effectiveness in reducing hemorrhage after cardiac
surgery.
Although our meta-analysis examined the relative
effects of
-aminocaproic acid and aprotinin against placebo, 3
randomized clinical trials have compared the 2 drugs directly. Trials
conducted by Penta de Peppo et al7 and Menichetti et
al8 compared
-aminocaproic acid to high-dose aprotinin
and determined the latter to be more effective in reducing
postoperative blood loss. However, differences in postoperative
transfusion requirements were not identified. In a larger, more recent
trial,9 patients receiving
-aminocaproic acid or
low-dose aprotinin had lower transfusion rates than patients in the
control group. However, no statistically or clinically significant
differences were noted between the 2 drug groups.
-Aminocaproic acid and aprotinin were associated with trends
toward large reductions in risk of postoperative stroke. However,
unlike the findings of several aprotinin studies,3 16
associations between antifibrinolytic use and reduced stroke incidence
were statistically insignificant in our analysis. Although the
relationship remains unproven, some authors hypothesize that the
kallikrein-inhibiting properties of aprotinin may reduce cerebral
vasospasm.17 The biological explanation for the potential
cerebrovascular protective effects of
-aminocaproic acid is still
unclear.
There remains little evidence that antifibrinolytic therapy
increases the risk of thrombosis-related complications with cardiac
surgery. In our analysis, neither
-aminocaproic acid nor
aprotinin was associated with appreciable differences in risk of
myocardial infarction. Two randomized controlled trials showed slightly
higher rates of coronary bypass graft closure with aprotinin;
however, these findings were not statistically
significant.18 19 Another randomized controlled trial
conducted by Havel et al20 revealed no differences in
bypass graft patency rates among patients receiving high-dose
aprotinin, low-dose aprotinin, or placebo. To the best of our
knowledge, the association between
-aminocaproic acid and bypass
graft patency has not been studied.
Limitations
Several potential limitations should be considered in
assessing the validity of our findings. First, it is impossible to rule
out publication bias. Studies showing large benefits with drug
intervention may be more likely to be published than studies showing
little or no benefit. Second, despite the large number of trials, the
meta-analysis lacked sufficient statistical power to detect
clinically important effects of treatment on relatively infrequent
outcomes. For example, large apparent reductions in reexploration risk
with
-aminocaproic acid (OR, 0.54) and in stroke rates with both
antifibrinolytic agents (OR, 0.47 to 0.56) were not statistically
significant. For similar reasons, we cannot exclude clinically
meaningful effects of
-aminocaproic acid or aprotinin on risks of
myocardial infarction and mortality.
Third, drug dosing and administration protocols varied across
trials. Although all but 1 of the high-dose aprotinin trials were
consistent in using the standard dose (6 million
KIU),21 the low-dose aprotinin studies used dosing
protocols ranging from 1 to 3 million KIU. Similarly, trials involving
-aminocaproic acid used doses ranging from 7 to 30 g. Whether
the effectiveness of these agents is dose related remains
unanswered.
Finally, variable outcome definitions across trials may
have limited our analysis in several ways. Renal toxicity
remains among the major concerns with antifibrinolytic therapy, despite
the inconclusive evidence supporting this
relationship.22 23 Because trials used variable
creatinine levels to define renal impairment, our ability
to examine this potential adverse effect of aprotinin and
-aminocaproic acid was severely limited. Other outcomes are also
subject to misclassification. Stroke and myocardial infarction are both
discretionary clinical diagnoses, and transfusion and reexploration
frequently reflect discretionary treatment decisions. In double-blind
trials (when surgeon is unaware of treatment allocation), random
outcomes misclassification in drug and control groups would be expected
to bias studies toward the null hypothesis (no drug effect). In
nonblinded trials, however, outcomes misclassification could bias
results toward larger drug effects. Despite these potential problems,
the main effects of aprotinin and
-aminocaproic acid were
essentially unchanged in stratified analysis of blinded and
nonblinded trials.
Conclusions
Aprotinin and
-aminocaproic acid substantially reduce
hemorrhage in patients undergoing cardiac surgery, without
appreciable adverse effects. For these reasons,
prophylactic antifibrinolytic therapy should be considered
for routine use. Because the 2 drugs appear to have similar efficacies,
the considerably less-expensive
-aminocaproic acid may be preferred
over
aprotinin.
| Footnotes |
|---|
Address correspondence to Dr John J. Munoz, VA Outcomes Group (111B), Department of Veterans Affairs Hospital, White River Junction, VT 05009.
Received May 28, 1998; revision received September 2, 1998; accepted September 17, 1998.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
J. Dunning, M. Versteegh, A. Fabbri, A. Pavie, P. Kolh, U. Lockowandt, S. A.M. Nashef, and on behalf of the EACTS Audit and Guidelines Commit Guideline on antiplatelet and anticoagulation management in cardiac surgery. Eur. J. Cardiothorac. Surg., July 1, 2008; 34(1): 73 - 92. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. P. Eaton Antifibrinolytic Therapy in Surgery for Congenital Heart Disease Anesth. Analg., April 1, 2008; 106(4): 1087 - 1100. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Schneeweiss, J. D. Seeger, J. Landon, and A. M. Walker Aprotinin during Coronary-Artery Bypass Grafting and Risk of Death N. Engl. J. Med., February 21, 2008; 358(8): 771 - 783. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. R. Bridges Valid Comparisons of Antifibrinolytic Agents Used in Cardiac Surgery Circulation, June 5, 2007; 115(22): 2790 - 2792. [Full Text] [PDF] |
||||
![]() |
J. R. Brown, N. J.O. Birkmeyer, and G. T. O'Connor Meta-Analysis Comparing the Effectiveness and Adverse Outcomes of Antifibrinolytic Agents in Cardiac Surgery Circulation, June 5, 2007; 115(22): 2801 - 2813. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. M. Mannucci and M. Levi Prevention and Treatment of Major Blood Loss N. Engl. J. Med., May 31, 2007; 356(22): 2301 - 2311. [Full Text] [PDF] |
||||
![]() |
V. A. Ferraris, C. R. Bridges, R. P. Anderson, for the Blood Conservation Guideline, J. R. Brown, N. J.O. Birkmeyer, G. T. O'Connor, J. H. Levy, J. G. Ramsay, R. A. Guyton, et al. Aprotinin in cardiac surgery. N. Engl. J. Med., May 4, 2006; 354(18): 1953 - 1957. [Full Text] [PDF] |
||||
![]() |
J. R. Cooper Jr, J. Abrams, O.H. Frazier, R. Radovancevic, B. Radovancevic, A. W. Bracey, M. J. Kindo, and I. D. Gregoric Fatal pulmonary microthrombi during surgical therapy for end-stage heart failure: Possible association with antifibrinolytic therapy J. Thorac. Cardiovasc. Surg., May 1, 2006; 131(5): 963 - 968. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. A. Camarasa, G. Olle, M. Serra-Prat, A. Martin, M. Sanchez, P. Ricos, A. Perez, and L. Opisso Efficacy of aminocaproic, tranexamic acids in the control of bleeding during total knee replacement: a randomized clinical trial Br. J. Anaesth., May 1, 2006; 96(5): 576 - 582. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Dial, E. Delabays, M. Albert, A. Gonzalez, J. Camarda, A. Law, and D. Menzies Hemodilution and surgical hemostasis contribute significantly to transfusion requirements in patients undergoing coronary artery bypass J. Thorac. Cardiovasc. Surg., September 1, 2005; 130(3): 654 - 654. [Abstract] [Full Text] [PDF] |
||||
![]() |
D Belway, F D Rubens, D Wozny, B Henley, and H J Nathan Are we doing everything we can to conserve blood during bypass? A national survey Perfusion, September 1, 2005; 20(5): 237 - 241. [Abstract] [PDF] |
||||
![]() |
A. Shander, D. Moskowitz, and T. S. Rijhwani The Safety and Efficacy of "Bloodless" Cardiac Surgery Seminars in Cardiothoracic and Vascular Anesthesia, March 1, 2005; 9(1): 53 - 63. [Abstract] [PDF] |
||||
![]() |
A. M. Mahdy and N. R. Webster Perioperative systemic haemostatic agents Br. J. Anaesth., December 1, 2004; 93(6): 842 - 858. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. M. Moskowitz, J. J. Klein, A. Shander, K. M. Cousineau, R. S. Goldweit, C. Bodian, S. I. Perelman, H. Kang, D. A. Fink, H. C. Rothman, et al. Predictors of transfusion requirements for cardiac surgical procedures at a blood conservation center Ann. Thorac. Surg., February 1, 2004; 77(2): 626 - 634. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. K. Smith, S. K. Datta, L. H. Muhlbaier, G. Samsa, A. Nadel, and J. Lipscomb Cost analysis of aprotinin for coronary artery bypass patients: analysis of the randomized trials Ann. Thorac. Surg., February 1, 2004; 77(2): 635 - 642. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. Shore-Lesserson, D. L. Reich, and D. H. Adams Invited commentary Ann. Thorac. Surg., February 1, 2004; 77(2): 642 - 643. [Full Text] [PDF] |
||||
![]() |
A. Koster, S. Huebler, F. Merkle, T. Hentschel, M. Grundel, T. Krabatsch, L. Tambeur, M. Praus, H. Habazettl, W. M. Kuebler, et al. Heparin-Level-Based Anticoagulation Management During Cardiopulmonary Bypass: A Pilot Investigation on the Effects of a Half-Dose Aprotinin Protocol on Postoperative Blood Loss and Hemostatic Activation and Inflammatory Response Anesth. Analg., February 1, 2004; 98(2): 285 - 290. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. E. Greilich, C. F. Brouse, C. W. Whitten, L. Chi, J. M. DiMaio, and M. E. Jessen Antifibrinolytic therapy during cardiopulmonary bypass reduces proinflammatory cytokine levels: a randomized, double-blind, placebo-controlled study of {epsilon}-aminocaproic acid and aprotinin J. Thorac. Cardiovasc. Surg., November 1, 2003; 126(5): 1498 - 1503. [Abstract] [Full Text] [PDF] |
||||