Glaser R, Herrmann HC, Murphy SA, Demopoulos LA, DiBattiste PM, Cannon CP, Braunwald E. Benefit of an Early Invasive Management Strategy in Women With Acute Coronary Syndromes. JAMA. 2002;288(24):3124-3129. doi:10.1001/jama.288.24.3124
Author Affiliations: Department of Medicine, University of Pennsylvania, Philadelphia (Drs Glaser and Herrmann); Department of Medicine, Brigham and Women's Hospital, Boston, Mass (Ms Murphy, and Drs Cannon and Braunwald); and Merck Research Laboratories, Blue Bell, Pa (Drs Demopoulos and DiBattiste).
Context Women who present with acute coronary syndromes (ACSs) have different
characteristics than men. Reports have conflicted about whether different
outcomes exist for women with use of a routine invasive management strategy.
However, these studies were performed prior to the widespread use of platelet
glycoprotein IIb/IIIa inhibitors and intracoronary stents.
Objective To determine sex differences in baseline characteristics and outcomes
in ACS and whether women benefit from a contemporary early invasive management
Design and Setting Prospective analysis of women and men enrolled in the TACTICS-TIMI 18
randomized trial, conducted December 1997 to December 1999 in 169 centers
in 9 countries in North America and Europe, with follow-up at 1 and 6 months.
Participants A total of 2220 patients (757 women and 1463 men) with ACS.
Interventions All patients received aspirin, 325 mg/d; intravenous unfractionated
heparin; and tirofiban for 48 hours or until revascularization, with tirofiban
administered for at least 12 hours after percutaneous coronary revascularization.
Patients assigned to the early invasive strategy (n = 1114) underwent coronary
angiography 4 to 48 hours after randomization and revascularization when appropriate.
Patients assigned to the early conservative strategy (n = 1106) were treated
medically and underwent coronary angiography and appropriate revascularization
only if they met specified criteria.
Main Outcome Measures Baseline characteristics and the primary composite end point of death,
myocardial infarction, or rehospitalization for ACS at 6 months in women and
men assigned to early invasive vs conservative management.
Results Women were older and more frequently had hypertension (P<.001 for both). Women less frequently had previous myocardial
infarction, coronary artery bypass grafting, and elevations in cardiac markers
(P<.001 for all), but there was no difference
in distribution of TIMI risk scores (P = .76). Angiography
and intervention rates were similar, but women had less severe coronary artery
disease, including no critical lesions in 17% of women vs 9% of men (P<.001). Women had a 28% odds reduction in the primary
end point with an early invasive strategy (adjusted odds ratio [OR], 0.72;
95% confidence interval [CI], 0.47-1.11), similar to the benefit in men (adjusted
OR, 0.64; 95% CI, 0.47-0.88; P = .60 for sex interaction).
When adjusted for baseline characteristics, the benefit of invasive therapy
in women with elevated troponin T levels was further enhanced (adjusted OR,
0.47; 95% CI, 0.26-0.83).
Conclusions Despite differences between women and men in baseline characteristics,
the benefit of an early invasive strategy incorporating tirofiban and intracoronary
stents was similar in women and men and was enhanced in women presenting with
markers of increased risk.
Studies of acute coronary syndromes (ACSs) have suggested that the sex
of patients influences both patient outcomes and the response to recent advances
in medical and invasive therapies for ACS.1- 8 In
particular, no benefit of platelet glycoprotein IIb/IIIa (Gp IIb/IIIa) inhibition
in women with ACS was seen in a meta-analysis, while invasive management of
ACS led to worsened outcomes for women in the FRISC II (Fragmin and Revascularization
during Instability in Coronary artery disease II) and RITA 3 (Randomized Intervention
Trial of unstable Angina 3) randomized controlled trials6,9,10 but
to improved outcomes in a prospective observational study.11 These
conflicting data have led to controversy about the role that the sex of patients
should play in determining risk and in selecting optimal management strategies.
The TACTICS-TIMI 18 (Treat angina with Aggrastat and determine Cost
of Therapy with an Invasive or Conservative Strategy–Thromolysis In
Myocardial Infarction 18) trial demonstrated a 22% odds reduction at 6 months
in the combined end point of death, myocardial infarction (MI), and rehospitalization
in patients with ACS treated with a Gp IIb/IIIa inhibitor and randomized to
an early invasive strategy, compared with those patients treated more conservatively.12 We prospectively assessed the clinical characteristics
and outcomes in women enrolled in this trial. Our primary goal was to determine
whether, in the setting of a contemporary strategy, treatment should be influenced
by patients' sex, or by objective measures of risk such as the TIMI risk score,
ST-segment changes, and levels of troponin T.
The study design and details of the TACTICS-TIMI 18 trial have been
described previously.12,13 Briefly,
between December 18, 1997, and December 22, 1999, 2220 patients in 169 centers
in 9 countries in North America and Europe underwent randomization to an early
invasive or conservative strategy, with follow-up at 1 and 6 months (Figure 1). The study enrolled women and men
who had experienced an episode of unstable angina within the preceding 24
hours, were candidates for coronary revascularization, and had at least 1
of the following: (1) ST-segment depression, transient ST-segment elevation,
or T-wave inversion in at least 2 leads; (2) elevated levels of cardiac biomarkers;
or (3) previously documented coronary disease.
All patients received 325 mg of aspirin daily (unless contraindicated),
intravenous unfractionated heparin, and tirofiban (Aggrastat, Merck, West
Point, Pa)14 for 48 hours or until revascularization,
with tirofiban administered for at least 12 hours after percutaneous coronary
revascularization procedures. Patients assigned to the early invasive strategy
were to undergo coronary angiography between 4 and 48 hours after randomization,
and revascularization when appropriate. Patients assigned to the early conservative
strategy were treated medically and underwent coronary angiography and appropriate
revascularization only if they met specified criteria, including prolonged
angina at rest associated with electrocardiographic evidence of ischemia or
changes in cardiac biomarker levels, hemodynamic instability, significant
ischemia during stress testing, unstable angina requiring hospitalization,
Canadian Cardiovascular Society class III or IV angina with an abnormal exercise
tolerance test, or a new MI.12,13
All patients randomized were included in the analyses by the intention-to-treat
principle. Multivariable logistic regression was performed to evaluate the
effect of the sex of patients as an independent prognostic factor relating
to the primary combined end point (death, MI, and rehospitalization for ACS
at 6 months). Myocardial infarction was defined as new Q waves in 2 contiguous
electrocardiogram leads or a creatine kinase-MB (CK-MB) fraction higher than
the upper limit of normal. Within 48 hours following percutaneous coronary
intervention (PCI), a new MI was defined as new Q waves or a CK-MB level elevated
greater than 3 times the upper limit of normal. After coronary artery bypass
graft (CABG) surgery, only new Q waves were used to define MI. All primary
end points were adjudicated by members of an independent clinical end points
committee who were unaware of patients' treatment assignments. Age, hypertension,
smoking status, prior CABG surgery, prior MI, an index event of non–ST-segment
elevation MI, troponin T level elevation, PCI or CABG surgery during hospitalization,
and an interaction variable for patients' sex were included in the multivariable
model. Separate analyses were performed with the additional inclusion of the
presence of significant coronary artery disease (CAD) or primary aspirin use
in the model. Statistical significance was defined as P≤.05. Survival curves using the Kaplan-Meier method were constructed
to evaluate invasive vs conservative therapy in women and in men. All analyses
were performed using STATA v7.0 (Stata Corp, College Station, Tex).
One third of patients enrolled in the trial were women (Figure 1 and Table 1).
Compared with men, women were older and more frequently had hypertension.
Women less often had prior cardiac disease and a presentation with elevated
levels of troponin T. The distribution of TIMI risk scores and ST-segment
changes of women was not different compared with that of men.
The rates of angiography were similar in women and men randomized to
either the invasive or conservative strategy (Table 2). In both the invasive and conservative groups, women who
underwent angiography had no significant CAD more often (17% vs 9%, P<.001), and had disease of the left main coronary artery
less often (7% vs 10%, P = .049). Similar differences
between women and men in the extent of CAD were present in patients randomized
to invasive or conservative therapy.
Pharmacological therapy was similar in women and men, except that women
received calcium channel blockers more frequently. This difference persisted
after adjustment for angiographic presence of CAD (Table 1).
Fewer women than men underwent CABG surgery in both the invasive and
conservative groups (13% vs 18%, P = .001). This
difference persisted when adjusted for the presence of 3-vessel CAD or disease
of the left main coronary artery (for CABG surgery in women vs men: adjusted
odds ratio [OR], 0.60; 95% confidence interval [CI], 0.41-0.86; P = .007). Rates of PCI and stent use were similar in women and men
in both strategies (32% of women underwent PCI vs 33% of men; 84% of women
and men underwent PCI with stent placement).
In women, the rate of the primary end point of death, MI, or rehospitalization
for ACS at 6 months was 17% with the early invasive strategy and 19.6% with
the conservative strategy. Adjustment for differences in baseline characteristics
enhanced the benefit of invasive management (women: adjusted OR, 0.72; 95%
CI, 0.47-1.11; men: adjusted OR, 0.64; 95% CI, 0.47-0.88). In multivariable
analysis, the sex of patients was not an independent risk factor for outcome
by strategy (P = .60 for interaction) (Table 3). The likelihood of death or nonfatal MI in women was also
lower with use of the early invasive strategy, particularly when adjusted
for baseline characteristics (6.6% in the early invasive strategy vs 9.7%
in the conservative strategy, adjusted OR, 0.45; 95% CI, 0.24-0.88; P = .02). In addition, female sex was not an independent
risk factor for outcome even after adjustment for the presence or absence
of significant stenosis at angiography.
Those women who underwent PCI had similar rates of death and MI at 6
months, compared with men (10.7% in women undergoing PCI vs 10.9% in men undergoing
PCI, P>.99). Furthermore, CABG surgery mortality
at 6 months was not increased in women compared with men (5.3% vs 4.5%, respectively; P = .78).
Rates of major bleeding were higher in women undergoing PCI compared
with men (8.3% vs 2.9%; adjusted OR, 3.6; 95% CI, 1.6-8.3; P = .001), and this difference persisted after adjustment for baseline
characteristics and a higher mean activated clotting time measured in women.
Rates of bleeding (12.6% vs 15%) and stroke (2.1% vs 1.5% at 30 days) were
similar in women and men undergoing CABG surgery.
While there was a trend toward improved outcomes in those women with
intermediate (3-4) and high (5-7) TIMI risk scores who received the invasive
when compared with the conservative strategy, this was not statistically significant
(P = .20 for trend) (Table 4). Women with ST-segment changes had a similar trend toward
improvement in death, MI, and rehospitalization for ACS with invasive management
(OR, 0.66; 95% CI, 0.38-1.15; P = .14), and a significant
reduction in death and MI (OR, 0.41; 95% CI, 0.19-0.89; P = .02). Women with elevated levels of troponin T had marked benefit
with an invasive strategy; the rate of the primary end point was 19% in the
invasive group and 29% in the conservative group (OR, 0.56; 95% CI, 0.32-0.97; P = .02). When adjusted for baseline characteristics, the
benefit of invasive therapy in women with elevated troponin T levels was further
enhanced (adjusted OR, 0.47; 95% CI, 0.26-0.83) (Figure 2).
This study demonstrates important differences in the baseline clinical
and angiographic characteristics between women and men presenting with unstable
angina and non–ST-segment elevation MI. Importantly, these differences
do not translate into significant differences in major outcomes between women
and men who both benefit from a contemporary management strategy for ACS incorporating
early invasive treatment and platelet Gp IIb/IIIa inhibitor use.
Previous analyses of women in trials of ACS have included those of the
TIMI IIIB trial and the FRISC II trial.1,6 In
TIMI IIIB, outcomes for women, adjusted for comorbidity, were similar to those
for men, with no difference between invasive and conservative therapy. In
the present analysis of the TACTICS-TIMI 18 trial there were similar differences
in baseline characteristics, but there was additionally a benefit for an invasive
strategy for both women and men. The improved outcome for women receiving
invasive therapy may reflect the routine use of a platelet Gp IIb/IIIa inhibitor
and the high use of intracoronary stents.15- 17
In a recent analysis of the FRISC II trial, women who were treated with
invasive therapy had significantly worse outcomes compared with women treated
with a conservative strategy (for invasive strategy, including adjustment
for presence of coronary disease: OR for death or MI, 1.72; 95% CI, 1.11-2.65; P = .01). The FRISC II investigators cite the lower prevalence
in women of CAD at angiography as a potential explanation for the worse outcomes
with invasive management in that trial, although the difference persisted
after adjustment for CAD. In our cohort as well, there was a lower prevalence
of severe CAD in women compared with men, but women still derived the overall
benefit of invasive management found in the trial.
Several important differences between these randomized controlled trials
may explain the discordant findings. First, the women undergoing CABG surgery
in the FRISC II trial had significantly higher mortality rates at 12 months
(9.9%) than did the men (1.2%), and higher mortality rates than did women
undergoing CABG surgery in our trial at 6 months (5.3%). Although it has been
demonstrated that women may have increased mortality with CABG surgery when
compared with men, recent studies have shown that most of this difference
is related to comorbid conditions.18- 20 Second,
the patients undergoing invasive management in TACTICS-TIMI 18 did so within
the first 48 hours of presentation, whereas patients in the FRISC II trial
did so on average during the fifth day after presentation. It is possible
that earlier therapy may have enhanced the benefits seen in women, as well
The RITA 3 trial found that women did not retain the benefits of invasive
management found in the overall trial.10 However,
the women in this trial may represent a cohort at lower risk, with lower rates
of death and MI at 1 year in women in both the invasive (8.6%) and conservative
groups (5.1%) than those of patients enrolled in the FRISC II and TACTICS-TIMI
18 trials (12.4% vs 10.5% in FRISC II at 1 year; 6.6% vs 9.7% in TACTICS at
6 months). Additionally, these lower event rates were observed despite a more
sensitive definition of MI used in the RITA 3 trial, implying a cohort at
even lower risk. The finding that the benefits of invasive therapy are significantly
mitigated in women at lower risk are not dissimilar from our findings, where
the benefit of invasive management was primarily confined to women with markers
of increased risk.
An important finding of our study was the predictive value of the serum
marker, troponin T, in women with ACS. Although troponin T had greater predictive
value for benefit than ST-segment changes and the TIMI risk score, the trends
were similar for all 3 of these measures of risk. A test that can predict
enhanced benefit with early invasive management may be especially useful in
women with ACS, since the proportion of women presenting with chest pain and
criteria for ACS who ultimately do not have significant CAD is higher than
in men.1,21- 23
There are important limitations to this analysis. The women enrolled
in this study are part of a randomized trial and may not be representative
of all women who present with ACS. However, the women in this trial had baseline
characteristics similar to those in registry data.1,4 Also,
subanalyses may not be adequately powered to detect differences among women.
The optimal management of women with ACS has been unclear. This study
shows that important differences exist in both baseline characteristics and
presentation findings between women and men. The hypothesis that higher procedural
complications, comorbidities, and less severe disease at angiography in women
may favor conservative therapy is not supported by our data. In fact, our
prospectively defined, randomized data show that women with unstable angina
and non–ST-segment elevation MI benefit, as do men, from broader inclusion
in contemporary management strategies incorporating Gp IIb/IIIa inhibitors
and early invasive therapy with stents. This is especially true for women
with objective evidence of ischemia including ST-segment changes and elevations
in the levels of the serum marker troponin T. In this regard, the choice of
an invasive vs conservative strategy for ACS should be based on objective
risk stratification measures, and not be influenced by the sex of patients.