Clinical Investigation and Reports

Indications for ACE Inhibitors in the Early Treatment of

Acute Myocardial Infarction
Systematic Overview of Individual Data From 100 000 Patients in
Randomized Trials
ACE Inhibitor Myocardial Infarction Collaborative Group*

Background—Several large-scale trials have demonstrated improved survival with ACE-inhibitor therapy started during
acute myocardial infarction. A systematic overview was conducted to resolve uncertainties regarding time of initiation,
time course of effect, and identification of patients in whom the benefits or the risks may be greater.
Methods and Results—This overview aimed to include individual data from all randomized trials involving more than
1000 patients in which ACE-inhibitor treatment was started in the acute phase (0 to 36 hours) of myocardial infarction
and continued for a short time (4 to 6 weeks). Data were available for 98 496 patients from 4 eligible trials, and the
results were consistent among the trials. Thirty-day mortality was 7.1% among patients allocated to ACE inhibitors and
7.6% among control subjects, corresponding to a 7% (SD, 2%) proportional reduction (95% CI, 2% to 11%; 2P,0.004).
This represented avoidance of '5 (SD, 2) deaths per 1000 patients, with most of the benefit observed within the first
week. The proportional benefit was similar in patients at different underlying risk. The absolute benefit was particularly
large in some high-risk groups (ie, Killip class 2 to 3, heart rate $100 bpm at entry) and in anterior MI. ACE-inhibitor
therapy also reduced the incidence of nonfatal cardiac failure (14.6% versus 15.2%, 2P50.01) but was associated with
an excess of persistent hypotension (17.6% versus 9.3%, 2P,0.01) and renal dysfunction (1.3% versus 0.6%, 2P,0.01).
Conclusions—These results support the use of ACE inhibitors early in the treatment of acute MI, either to a wide range
of patients or selectively in patients with anterior MI and in those at increased risk of death. (Circulation.
1998;97:2202-2212.)
Key Words: ACE inhibitors n myocardial infarction n trials n systematic overview
Downloaded from http://ahajournals.org by on March 4, 2024

ate ACE inhibitors in patients with congestive heart failure,10

A total of more than 120 000 patients have been random-
ized in several large-scale controlled trials to evaluate
the effect of ACE inhibitors during and after acute myocar-
dial infarction (MI).1– 8 In general, mortality and morbidity
were reduced when ACE inhibitors were given during the
acute phase of MI (“early”) in a relatively unselected popu-
lation of patients,1–5 as well as when they were started
sometime after MI (“late”) in patients with evidence of left
ventricular dysfunction.6 – 8 However, uncertainty still exists
regarding a number of clinically relevant questions: for
example, whether there are subgroups of patients in whom the
benefits or the risks are greater.9
See p 2192
Thus, a collaborative group involving the principal inves-
tigators of all the randomized trials was created to collate the
individual patient data from the early and late trials in
systematic overviews. Prespecified end points were analyzed
by use of an approach already successfully applied to evalu-
antiplatelet therapy in occlusive vascular diseases,11 or
thrombolytic therapy in acute MI.12
The results of such systematic overviews should help to
guide clinical practice appropriately and thus optimize the
benefits attainable with therapeutic approaches. In this report,
we present data from an overview of the early trials of ACE
inhibitors in MI.
Methods
Trials to Be Included
The present overview was to be of the randomized trials in which
ACE-inhibitor treatment begun in the acute phase of MI (0 to 36
hours from symptom onset) and continued for a short period of time
(generally 4 to 6 weeks) was compared with no routine ACE-inhib-
itor treatment. Such trials are CONSENSUS-II,1 GISSI-3,2 ISIS-4,3
CCS-1,4 SMILE,5 GISSI-3 Pilot,13 ISIS-4 Pilot,14 and several other
smaller studies.15–21 Long-term trials in which ACE-inhibitor treat-
ment was started after the acute phase of MI in selected patients with
left ventricular dysfunction and continued for a longer period (at
least 6 months)6 – 8,22–26 will be the subject of a separate overview. The

Received September 25, 1997; revision received January 15, 1998; accepted January 30, 1998.
*A list of the ACE Inhibitor Myocardial Infarction Collaborative Group Members appears in the Appendix.
From the Istituto di Ricerche Farmacologiche “Mario Negri,” Milan, Italy.
Correspondence to Dr Maria Grazia Franzosi, ACE Inhibitor Collaborative Group, GISSI Coordinating Centre, Istituto di Ricerche Farmacologiche
“Mario Negri,” Via Eritrea 62, 20157 Milano, Italy.
E-mail depcardio@irfmn.mnegri.it
© 1998 American Heart Association, Inc.

2202
 ACE Inhibitor MI Collaborative Group
TABLE 1. Characteristics of Trials of ACE Inhibitors vs Control in Acute MI Involving >1000 Patients
CONSENSUS-II GISSI-3
Eligibility criteria ,24 h; ST elevation; Q waves, ,24 h; typical symptoms, ECG and
or raised cardiac enzymes enzymatic abnormalities
ACE inhibitor Enalapril Lisinopril
Dose 1 mg IV infusion15220 mg 5 mg110 mg daily
PO daily
Control group Placebo Open
Scheduled duration 6 mo 42 d
of treatment
Recruitment period March 1990 June 1991
April 1991 July 1993
Number of patients 6090 19 394
randomized
MI indicates myocardial infarction.
GISSI Coordinating Center, Milan, Italy, was responsible for data
collection, checking, and analysis of the present overview, and the
Canadian Cardiovascular Collaboration Project Office at McMaster
University, Hamilton, Ontario, was responsible for the overview of
long-term trials.
In the present overview, only those “early short-term” trials that
randomized more than 1000 patients to ACE-inhibitor therapy versus
control1–5 were to be included, because patients from these trials
represent '98% of all randomized patients, and retrieving reliable
individual patient data from small trials is generally more difficult.12
As a result, several smaller trials were excluded.13–21
Individual Patient Data Collection
Data were collected for individual patients because this allows more
Downloaded from http://ahajournals.org by on March 4, 2024
detailed consistency checking, analysis, and life-table calcula-
tions.11,12,27 A common protocol was developed to provide data to the
overview coordinating centers in a standardized format. These
included data recorded before randomization (such as ECG classifi-
cation, supplemented by discharge ECG in the GISSI-3 study; age
and sex; systolic blood pressure and heart rate; history of MI,
diabetes, or hypertension; Killip class; and hours from onset of
symptoms), as well as concomitant medications, clinical events (such
as hypotension, renal dysfunction, cardiogenic shock, second- to
third-degree atrioventricular block, heart failure, ventricular fibrilla-
tion, and stroke), and mortality after randomization. The last date of
follow-up was also collected to allow survival analyses. The original
definitions adopted in each trial for clinical events were used. Data
were checked for completeness and consistency with the published
results; apparent discrepancies were reviewed with the principal
investigators, and any corrections required were included in the main
database.
Available Data From Different Trials
Individual data were available for 98 496 patients in four of the
eligible trials1– 4 but not for the 1556 patients in SMILE,5 representing
availability of 98% of eligible patients. Mortality data and most other
key clinical outcomes were available from each trial, but some other
data items were not systematically collected in each trial. For
example, history of hypertension or diabetes was not recorded in
ISIS-4; dates of some clinical events were not always collected in
CONSENSUS-II and CCS-1; heart failure at entry, rather than Killip
class, was recorded in ISIS-4 and CONSENSUS-II (and considered
equivalent to Killip class .1 in this overview); and the randomiza-
tion date was not available for 13 patients in CCS-1.
All randomized patients were to be included in the analyses, and
follow-up for survival to day 30 was almost complete. Only 2% of
patients were lost before day 30 (474 on day 0, 414 on days 1 to 7,
and 950 on days 8 to 30), and they were well balanced between the
two groups (920 allocated ACE inhibitors versus 918 allocated
control treatment). No significant differences between baseline
June 9, 1998 2203
ISIS-4 CCS-1
,24 h; suspected or ,36 h; suspected or
definite MI definite MI
Captopril Captopril
6.25 mg112.5 mg at 2 h, 6.25 mg112.5 mg TID
then 50 mg BID
Placebo Placebo
28 d 28 d
July 1991 January 1990
August 1993 May 1995
58 050 14 962
characteristics were found, except for a slight imbalance in history of
hypertension (37.7% ACE inhibitors versus 36.7% control,
2P50.04).
Statistical Methods
The main analyses were of mortality and clinical events up to day 30
(with special emphasis on days 0 to 7) and of mortality by subgroup
based on baseline characteristics. Statistical analyses used the mod-
ified Mantel-Haenszel method11,12,28 to calculate stratified estimates
of the proportional treatment effect, which were described as odds
ratios or as percentage reductions in odds. x2 tests for heterogeneity
of the proportional effects were calculated between different trials or
between subgroups. x2 tests for linear trend were also calculated
whenever appropriate. Given the modest size of the overall effect
and the number of subgroups studied, such subgroup analyses need
to be interpreted cautiously; indeed, in many instances, the overall
proportional effect may provide more reliable guidance as to the
proportional effect in some particular subgroup than the effect
observed just within that subgroup.11,12,29
To investigate whether the treatment benefit varied according to
the underlying risk, the effects were evaluated in subgroups accord-
ing to a prognostic index derived from logistic regression analysis
among all study patients (irrespective of allocated treatment). The
following variables adjusted by treatment allocation were included in
the model: age (as a continuous variable), sex, systolic blood
pressure (,100, 100 to 120, 121 to 150, and .150 mm Hg), heart
rate (,80, 80 to 99, and $100 bpm), previous MI, Killip class (1 and
.1), and location of MI (anterior and nonanterior). The patients were
divided into four groups that included approximately equal numbers
of deaths: low risk (30-day mortality of 2% to 6%), medium risk (6%
to 10%), high risk (10% to 16%), and very high risk (16% to 42%).
x2 values for heterogeneity and for linear trend were calculated to
test the treatment effects between these groups.
All P values are two-sided; values of 2P,0.05 and 2P,0.01 were
considered conventionally significant in the overall and subgroup
analyses, respectively. Similarly, 95% CIs were used for overall
analyses and 99% CIs for subgroup analyses to make some allow-
ance for the effects on probability values of multiple comparisons.
For survival analyses, the Kaplan-Meier method was used, and the P
value was determined by the log-rank test.30
Results
Features of Trials and Available Data
The 4 trials of early ACE-inhibitor therapy versus control
treatment, which provided individual patient data for the
overview, recruited a total of 98 496 patients (Table 1).
Captopril was used in 2 of these trials,3,4 enalapril in 1,1 and
lisinopril in 1.2 Three trials were placebo controlled,1,3,4 and 1
 2204 Systematic Overview of ACE Inhibitors in Acute MI
used an open control2; all trials used a 1:1 allocation ratio.
Three trials included patients presenting within 24 hours from
the onset of symptoms,1–3 and 1 up to 36 hours.4 Patients with
definite or suspected acute MI were eligible for 2 trials3,4: 1
included only patients with ST-segment elevation, new patho-
logical Q waves, or raised cardiac enzymes1; and 1 required
two of the following: typical chest pain, abnormal Q waves
with evolutionary ST-T wave changes on serial ECG, or
enzymatic evidence.2
There was no specified upper age limit in any of the trials:
even so, the percentage of patients .75 years old varied from
9% in CCS-1 to 23% in CONSENSUS-II (Table 2). Patients
presenting with cardiogenic shock (ie, Killip class 4) were
generally excluded from the trials, with a larger proportion of
those patients in CCS-1 in Killip class .1. Antiplatelet and
fibrinolytic therapy were explicitly recommended in GISSI-3
and ISIS-4 (Table 3), and they were used more commonly in
those trials. Nitrates were used commonly in the CCS-1 trial
(87% of patients), and oral or transdermal nitrates were given
to approximately half of the patients in the GISSI-3 and
ISIS-4 trials, because such nitrates were also allocated at
random in factorial study designs.
Effects on 30-Day Mortality and on
7-Day Mortality
Overall, the cumulative mortality for patients allocated to
ACE inhibitors and to control showed a significant difference
in survival at 30 days (log-rank test, P50.004). There were
3501 deaths (7.11%) during days 0 to 30 among 49 214
Downloaded from http://ahajournals.org by on March 4, 2024
patients allocated to ACE inhibitor compared with 3740
deaths (7.59%) among 49 269 control patients (Figure 1).
This 7% (SD, 2%) proportional reduction in 30-day mortality
(95% CI, 2% to 11% reduction) corresponds to the avoidance
of 4.8 (SD, 1.7) deaths per 1000 patients. There was no
statistical difference between the effects in the four trials (x2
on 3 df55.8, 2P50.1) (Figure 2).
Among patients allocated ACE inhibitors, there were 239
fewer deaths, and subdivision of these deaths by day from
initiation of treatment indicates that 200 (ie, four fifths) were
avoided during the first week (Figure 3). The treatment was
associated with a significant 8% (SD, 3%) proportional
reduction during days 0 to 7 (95% CI, 3% to 14% reduction),
with similar benefit in days 0 to 1 and 2 to 7. This
corresponded to avoidance of 4.0 deaths (SD, 1.4) per 1000
patients in the first week (Figure 3).
Effects on 30-Day Mortality in
Different Subgroups
Age and Sex
The mortality reductions at 30 days were separately signifi-
cant in the patients 55 to 64 years old (16% [SD, 5%]
proportional reduction; 2P50.001) and 65 to 74 years (10.8%
[SD, 3.7%] proportional reduction; 2P50.004; Figure 4).
There was a trend toward greater proportional mortality
reduction among younger patients (x2 on 1 df56.2; 2P50.01)
(Figure 4). The proportional and absolute reductions in death
were similar for both sexes (4.6 [SD, 1.8] lives saved per
1000 in men versus 5.5 [SD, 3.9] in women) (Figure 4).
Systolic Blood Pressure and Heart Rate
The proportional reductions in mortality were not signifi-
cantly influenced by systolic blood pressure at entry (x2 for
trend52.2, 2P50.1). However, few patients with systolic
blood pressure ,100 mm Hg were studied, because such
patients were often excluded (Figure 4). By contrast, there
was a significant trend toward greater proportional mortality
reductions among patients with higher heart rates at entry (x2
for trend59.9, 2P50.002). Hence, the absolute benefits
observed among patients with higher heart rates were larger:
22.7 (SD, 6.7) fewer deaths per 1000 among those with heart
rates $100 bpm and 8.7 (SD, 3.1) fewer deaths per 1000
among those with heart rates 80 to 99 bpm (Figure 4).
Prior MI, Diabetes, Hypertension
The proportional reductions in mortality in patients with a
history of MI, of diabetes, or of hypertension were nonsig-
nificantly different from those observed in patients without
these conditions (Figure 4). Because such patients are at
higher absolute risk of death after MI, the absolute benefits of
ACE-inhibitor treatment were greater among patients with
prior MI (8.9 [SD, 4.7] versus 4.1 [SD, 1.8] lives saved per
1000), among diabetics (17.3 [SD, 8.9] versus 3.2 [SD, 2.7]
lives saved per 1000), and among hypertensives (9.0 [SD,
4.7] versus 2.1 [SD, 3.1] lives saved per 1000).
Killip Class at Entry
There was no significant difference between the proportional
mortality reduction among patients with heart failure or Killip
class .1 at entry (11% [SD, 4%]) and that among patients in
Killip class 1 (5% [SD, 3%]). Because patients with heart
failure are at greater risk of death, the absolute benefit of
treatment was greater among these patients (14.1 [SD, 5.4]
versus 2.9 [SD, 1.6] lives saved per 1000) (Figure 4).
Delay From Symptom Onset and Concomitant
Fibrinolytic Therapy
The benefits of ACE inhibitors were not significantly influ-
enced by the delay from onset of symptoms to randomization
(within 24 to 36 hours) (Figure 4) or by whether fibrinolytic
therapy was used (7.1% [SD, 3.2%] proportional reduction
with ACE inhibitors in the presence of fibrinolytic therapy
and 6.6% [SD, 3.5%] reduction in its absence; data not
shown).
Site of MI
Among patients with evidence of anterior MI (ie, anterior
ST-segment elevation with or without other changes), the
proportional reduction of 14% (SD, 3.6%) was greater than
that among patients with other MI locations (2% [SD, 3%]; x2
for heterogeneity on 1 df56.3, 2P50.01). This corresponds
to 10.6 (SD, 2.9) deaths avoided per 1000 patients with
anterior MI (Figure 4).
Subgroups at Different Risk
When the effects of ACE-inhibitor treatment were evaluated
in subgroups of patients according to a multivariate prognos-
tic index (see “Methods”), there was no evidence of a
difference in the proportional benefits in patients at different
underlying risk (Figure 4, bottom). Hence, the absolute
benefits were greater in patients at greater risk of death (3.8
 ACE Inhibitor MI Collaborative Group June 9, 1998 2205

TABLE 2. Baseline Characteristics in Relevant Trials

Total
(n598 496)
CONSENSUS-II GISSI-3 ISIS-4 CCS-1
(n56090) (n519 394) (n558 050) (n514 962) No. %
Age, y, %
,55 17 24 26 23 24 301 25
55–64 23 29 29 39 29 292 30
65–74 36 30 30 29 29 818 30
$75 23 18 15 9 15 064 15
NR 0 zzz zzz 0 21 0
Mean age, y 66 63 62 61 63
SD 11 12 12 11 12
Sex, %
Male 73 78 74 74 73 645 75
Female 27 22 26 26 24 834 25
NR zzz zzz 0 zzz 17 0
Systolic blood pressure, mm Hg, %
,100 0 zzz 2 7 2463 3
100–120 34 39 35 45 36 839 37
121–150 49 46 44 34 42 751 43
.150 17 15 18 13 16 339 17
NR 0 zzz 0 1 104 0
Heart rate, bpm, %
,80 61 61 52 42 51 791 53
80–99 31 31 36 23 31 890 32
$100 8 9 13 11 11 130 11
NR 0 zzz 0 24 3685 4
Prior MI, %
Downloaded from http://ahajournals.org by on March 4, 2024

Yes 24 14 17 12 15 686 16
No 76 82 83 88 81 991 83
NR zzz 4 0 zzz 819 1
Diabetes, %
Yes 11 15 * 9 5014 12
No 89 79 * 91 34 332 85
NR zzz 6 * zzz 1100 3
History of hypertension, %
Yes 27 38 * 40 15 059 37
No 73 55 * 60 24 016 59
NR zzz 7 * zzz 1371 3
Killip class at entry,† %
1 82 83 86 59 79 828 81
.1 18 14 14 34 16 984 17
NR zzz 3 0 7 1684 2
Site of infarction at entry, %
Anterior 46 36 34 44 36 317 37
Other 54 64 66 56 62 179 63
Hours from onset of symptoms, %
0–3 1 14 18 8 14 129 14
3–6 10 21 23 13 19 723 20
6–12 25 25 28 22 25 655 26
12–24 63 40 32 35 35 576 36
24–36 2 zzz zzz 22 3378 3
NR zzz zzz 0 0 35 0
Creatinine at entry, mg/dL
Mean 1.1 1.1 * * 1.1
SD 0.3 0.3 * * 0.3
NR indicates not recorded; MI, myocardial infarction.
*Variable not collected.
†CONSENSUS-II and ISIS-4 recorded heart failure at entry that had been classified as Killip class .1.
 2206 Systematic Overview of ACE Inhibitors in Acute MI
TABLE 3. Concomitant Therapies in Hospital
CONSENSUS-II
(n56090)
Fibrinolytics, % 56
Antiplatelets,† % 79
IV b-blockers, % 15
IV nitrates, % 49
PO/TD nitrates,‡ % 31
*Variable not collected.
†CONSENSUS-II, GISSI-3, and ISIS-4 include aspirin and other antiplatelets. CCS-1 includes only aspirin.
‡GISSI-3 and ISIS-4 also introduced randomization between nitrate (TD and PO indicate transdermal and oral,
respectively) versus control, using a factorial design, and use of this allocated study nitrate plus any use of nonstudy
oral/transdermal nitrate is included.
[SD, 1.5] lives saved per 1000 low-risk patients compared
with 13.6 [SD, 9.1] in very-high-risk patients). Mortality after
MI increased steeply with increasing age, whereas the uni-
variate analyses (above) indicated that the proportional re-
ductions in mortality with ACE inhibitors were greater at
younger ages. Prognostic scores were therefore also con-
structed with age excluded, both for all patients and, sepa-
rately, for those ,75 years old and those $75 years old. In
each case, the proportional benefits among patients at differ-
ent absolute risk were not significantly different from each
other (data not shown).
Effects on Nonfatal Heart Failure
ACE-inhibitor therapy significantly reduced the incidence of
Downloaded from http://ahajournals.org by on March 4, 2024
nonfatal cardiac failure: there were 6687 cases of nonfatal
heart failure (14.6%) during days 0 to 30 among patients
allocated to ACE inhibitors compared with 6937 cases
(15.2%) among control subjects. This corresponds to the
avoidance of 6.1 (SD, 2.4) cases of nonfatal heart failure per
1000 patients (2P50.01), which was distributed throughout
the period of hospitalization (Table 4A).
Effect on Other Clinical Events
The incidence rates of reinfarction and stroke were similar in
the two treatment groups (Table 4B). ACE-inhibitor therapy
was associated with a significant excess of 84 (SD, 2) cases
of persistent hypotension per 1000 patients treated (17.6%
versus 9.3%). There were also small but significant increases
in cardiogenic shock (4.6 [SD, 1.2] per 1000) and in second-
to third-degree atrioventricular block (5.4 [SD, 1.2] per
Figure 1. Effect of ACE-inhibitor therapy on cumulative mortality
during days 0 to 30 in all trials combined.
GISSI-3 ISIS-4 CCS-1 Total
(n519 394) (n558 050) (n514 962) (n598 496)
70 68 29 62
86 93 75 88
30 9 * 14
69 47 62 54
48 56 87 57
1000). Most of these excesses occurred during days 0 to 7,
when most of the mortality benefit was also observed. There
was also a significant excess of 6.2 (SD, 0.6) cases of renal
dysfunction per 1000 (1.3% ACE inhibitor versus 0.6%
control), which was distributed throughout the period of
hospitalization (Table 4B).
There was a slight but significant increase with increasing
age in the proportional effect on persistent hypotension after
ACE-inhibitor treatment (test for trend, 2P50.003) (Table 5).
Otherwise, the proportional effects on hypotension and on
renal dysfunction in the subgroups examined were not clearly
different from those observed overall. In some subgroups,
however, the control risks were much higher, and thus the
absolute excesses of complications were greater. For exam-
ple, among patients presenting with systolic blood pressure
,100 mm Hg, the absolute excess of persistent hypotension
was 132 per 1000 compared with an overall excess of 84 per
1000. Similarly, the absolute risks of renal dysfunction were
greater in the elderly, with an absolute excess of 17 per 1000
among those $75 years old compared with 6 per 1000
overall.
Discussion
This systematic overview of trials of early ACE-inhibitor
therapy in acute MI yields four main messages. First, the
benefit on 30-day mortality is consistent among the trials and,
on average, corresponds to '5 lives saved per 1000 patients
treated for '1 month. Second, most of the benefit occurs
during the first few days, when mortality is highest. Third, the
proportional benefit is generally consistent among patients
with differing baseline characteristics, so patients at higher
risk generally benefit to a greater absolute extent. Fourth,
there is no subgroup in which the treatment was shown to be
definitely harmful, although hypotension and renal dysfunc-
tion with therapy were more common in patients $75 years
old, and there was no direct evidence of any survival
advantage in such patients. It should be remembered that
patients presenting in cardiogenic shock or with persistently
low systolic blood pressure (ie, ,100 mm Hg) were generally
excluded from these studies.
Before these points are discussed in detail, some limita-
tions of this overview should be underlined: for example, if
only the larger trials (.1000 randomized patients) were
considered, information from smaller trials was lost; also,
 ACE Inhibitor MI Collaborative Group
individual patient data were not available from the SMILE
study.5 However, these trials would have added only '5% to
the total number of patients, and their overall results are
consistent with the present ones. In particular, adding the
published results for the 1556 patients randomized within 24
hours of the onset of anterior MI in the SMILE study (38
deaths at 42 days among 772 patients allocated to ACE-in-
hibitor therapy versus 51 among 784 control subjects) to
those for the nearly 100 000 patients included in this over-
view would not alter the findings even minimally (data not
shown).
Downloaded from http://ahajournals.org by on March 4, 2024
A second potential limitation concerns heterogeneity be-
tween different studies. Differences in the drug regimens
studied include the study agents (eg, drugs with a short
half-life, such as captopril, and a long half-life, such as
lisinopril), dosages (eg, captopril dose in CCS-1 versus
ISIS-4), and routes of administration (in particular, an initial
intravenous infusion was used in CONSENSUS-II, which
was the only trial to report an adverse trend, albeit nonsig-
nificant, on mortality), and concomitant nonstudy treatments
(Table 3). Moreover, there were some differences in the types
of patients studied in the different trials (Table 2) and in some
of the definitions of variables used (such as “site of MI” or
“persistent hypotension”). In assessments of the differences
between subgroups, variations in definitions or baseline
characteristics would tend to decrease the sensitivity of such
analyses to show interactions. However, such differences are
inherent to all overviews and do not introduce any biases into
ascertainment of the average effects observed among the
patients included.
Benefit of Early ACE-Inhibitor Therapy
An average 7% proportional reduction in mortality with ACE
inhibitors was observed within 30 days of acute MI, corre-
sponding to an average absolute benefit of '5 lives saved per
1000 patients treated. In other words, on average, ACE-in-
hibitor therapy begun early in acute MI and continued for
only 1 month in '200 patients leads to the saving of 1 life. At
first glance, this beneficial effect may appear modest com-
pared with that observed in the trials of long-term ACE-in-
hibitor therapy among high-risk post-MI patients. But those
June 9, 1998 2207
Figure 2. Proportional effects of ACE-inhibitor therapy on
30-day mortality in each trial. “Observed minus expected”
(O2E) number of deaths among ACE inhibitor–assigned
patients (and its variance) is quoted for each trial. This is
used to calculate odds ratio of death among patients
assigned ACE-inhibitor therapy to that among patients
assigned control treatment. Odds ratios (solid squares,
with areas proportional to amount of information contrib-
uted by each trial) are plotted with their 95% CIs (horizon-
tal lines).
trials involved a much longer treatment period (1 to 3 years),
so the number of lives saved per month of treatment per 1000
individuals varied from 1.0 in SAVE to 3 to 4 in AIRE.9 Thus,
the early use of ACE inhibitors in relatively unselected MI
patients leads to at least comparable absolute survival bene-
fits during the first month of treatment.
In addition, nonfatal heart failure, which was not a primary
end point in any of the studies, was also significantly reduced
by early ACE-inhibitor treatment. This corresponds to 6
additional events avoided per 1000 patients on top of the
survival benefit.
Time Course of the Beneficial Effect of
ACE-Inhibitor Therapy on Survival
ACE-inhibitor therapy saved lives early after its initiation,
with 40% of the 30-day survival advantage observed in
days 0 to 1, '45% in days 2 to 7, and '15% subsequently
(Figure 3). Data from the different studies were consistent
in this respect, strongly supporting the strategy of starting
ACE inhibitors early to maximize their potential benefits.
But because the reduction in mortality was observed
irrespective of the interval between symptom onset and
randomization, ACE inhibitors should not be withheld
from patients who present late.
Figure 3. Absolute effect of ACE-inhibitor therapy on deaths in
days 0 to 1, 2 to 7, and 8 to 30.
 2208 Systematic Overview of ACE Inhibitors in Acute MI

Figure 4. Effects of ACE-inhibi-

tor therapy on mortality in days
0 to 30 subdivided by presenta-
tion features. Odds of death
among patients allocated to
ACE-inhibitor therapy to that
among those allocated to control
treatment is derived from
“observed minus expected”
numbers of death (and vari-
ances) calculated within each
subdivision of presentation fea-
tures stratified by trial. Odds
ratios within each presentation
feature are plotted with their
99% CIs, whereas overall result
and 95% CI are represented by
diamond.
Downloaded from http://ahajournals.org by on March 4, 2024

Effects of ACE-Inhibitor Therapy on Survival in Safety of ACE-Inhibitor Therapy During

Different Patient Subgroups
In general, most of the patient subgroups studied benefited
from ACE inhibition, and there was no subgroup in which the
treatment was clearly shown to be harmful. The CIs for the
proportional effects on 30-day mortality in different sub-
groups overlap each other substantially and, in general, did
not differ significantly from the overall proportional reduc-
tion of 7% (Figure 4). Similarly, when the results were
analyzed in groups subdivided with respect to a risk score
based on multivariate logistic regression (even with age
excluded), the proportional reductions in mortality were
similar at different levels of risk (Figure 4). However, certain
characteristics, such as anterior site of MI and high heart rate,
were clearly associated with a higher benefit. But such
patients represent only a minority of those who present with
suspected acute MI, and the beneficial effects of early
ACE-inhibitor therapy in a lower-risk population may still
have some influence on the impact of the treatment.
Acute MI
Hypotension was anticipated, and this was the reason for
titration of the initial doses of the ACE inhibitors. It was
significantly more common than among control subjects
(17.6% versus 9.3%), and although there was an increase in
mortality (2.3% versus 1.6%) among patients who became
hypotensive, this may well indicate differences in their
baseline risk.31 Moreover, the excess of hypotension was seen
primarily during the first week, when most of the survival
advantage was also seen. A significant increase in the
incidence of renal dysfunction and cardiogenic shock was
also observed in the ACE inhibitor–treated patients, perhaps
reflecting the hypotensive effect of treatment during this
acute phase. The higher incidence of second- to third-degree
AV block might have been a consequence of increased
parasympathetic activity and decreased sympathetic tone
induced by the ACE inhibitor32 or, perhaps, of hypotension-
induced ischemia in some patients.
 ACE Inhibitor MI Collaborative Group June 9, 1998 2209

TABLE 4. Clinical Events by Days After AMI by Allocated Treatment

ACE Inhibitors Control
(n549 214) (n549 269)
Absolute Benefit per 1000
n % n % (95% CI)
A. Nonfatal cardiac failure and ventricular fibrillation
Nonfatal cardiac failure, d
0 –7 4972 10.9* 5047 11.1* 2.1; SD 2.1 (22.0 to 6.1)
8–30 358 0.9* 374 0.9* 0.5; SD 0.7 (20.8 to 1.8)
Total 0–30 6687 14.6* 6937 15.2* 6.1; SD 2.4 (1.5 to 10.7)
Ventricular fibrillation, d
0–7 1265 2.6 1318 2.7 1.0; SD 1.0 (23.0 to 0.9)
8–30 216 0.4 234 0.5 0.4; SD 0.4 (21.2 to 0.5)
Total 0–30 1538 3.1 1611 3.3 1.4; SD 1.1 (23.6 to 0.8)
Absolute Excess per 1000
(95% CI)

B. Other clinical events

Persistent hypotension, d
0–7 7431 15.1 3992 8.1 70; SD 2 (66 to 74)
8–30 559 1.3 308 0.7 7; SD 1 (5 to 8)
Total 0–30 8680 17.6 4573 9.3 84; SD 2 (79 to 88)
Renal dysfunction, d
0–7 413 0.8 239 0.5 3.5; SD 0.5 (2.5 to 4.6)
8–30 208 0.4 76 0.1 2.7; SD 0.3 (2.0 to 3.4)
Total 0–30 622 1.3 315 0.6 6.2; SD 0.6 (5.0 to 7.5)
Cardiogenic shock, d
Downloaded from http://ahajournals.org by on March 4, 2024

0–7 1724 3.5 1490 3.0 4.8; SD 1.1 (2.6 to 7.0)

8–30 210 0.4 220 0.4 20.2; SD 0.4 (21.1 to 0.6)
Total 0–30 1934 3.9 1710 3.5 4.6; SD 1.2 (2.2 to 7.0)
2nd- and 3rd-degree AV block, d
0–7 1886 3.8 1672 3.4 4.4; SD 1.2 (2.1 to 6.7)
8–30 100 0.2 79 0.2 0.5; SD 0.3 (20.1 to 1.0)
Total 0–30 2076 4.2 1810 3.7 5.4; SD 1.2 (3.0 to 7.9)
Reinfarction, d
0–7 1484 3.0 1385 2.8 1.5; SD 1.1 (20.7 to 3.6)
8–30 442 0.9 456 0.9 20.3; SD 0.6 (21.5 to 1.0)
Total 0–30 1938 3.9 1849 3.8 1.8; SD 1.2 (20.6 to 4.3)
Stroke, d
0–7 372 0.76 358 0.73 0.3; SD 0.5 (20.8 to 1.4)
8–30 91 0.19 71 0.15 0.4; SD 0.3 (20.1 to 0.9)
Total 0–30 463 0.94 430 0.87 0.7; SD 0.6 (20.5 to 1.9)
The distribution in days 0 –7 and 8 –30 includes only events with a known date of occurrence. Events with missing date are included in the total 0 –30 for patients
discharged within 30 days. AMI indicates acute myocardial infarction.
*Percentages calculated on patients surviving at day 30.

ACE-inhibitor therapy was not generally associated with Clinical Implications

proportionally higher risks of hypotension or of renal dys-
function in specific patient subgroups, except for hypotension
with increasing age. The absolute excesses, however, were
greater in certain subgroups at higher risk (eg, hypotension
among those presenting with systolic blood pressure
,100 mm Hg or renal dysfunction among those $75 years
old), and this should be considered when we decide whether
the likely survival benefit is likely to justify the treatment of
a particular patient.
This overview supports and expands the conclusions of a
previous consensus meeting9 and leads to the following
general considerations. First, ACE-inhibitor treatment may be
started immediately during the acute phase of MI, along with
other routinely recommended treatments (such as
thrombolytics, aspirin, and b-blockers), in the absence of
clear contraindications. The presence of cardiogenic shock or
systolic blood pressure persistently ,100 mm Hg should
generally be considered a contraindication to early treatment
 2210 Systematic Overview of ACE Inhibitors in Acute MI

TABLE 5. Effect on 30 Days Persistent Hypotension and Renal Dysfunction

Persistent Hypotension Renal Dysfunction

ACE Inhibitor Control ACE Inhibitor Control

Odds Ratio Odds Ratio
Characteristics at Entry Events /n % Events /n % (99% CI) Events /n % Events /n % (99% CI)
Age, y
,55 2030/12 159 16.70 1135/12 142 9.35 1.93 (1.74–2.12) 49/12 159 0.40 20/12 142 0.16 2.31 (1.24–4.30)
55–64 2464/14 663 16.80 1311/14 628 8.96 2.03 (1.85–2.22) 91/14 663 0.62 49/14 628 0.33 1.82 (1.18–2.82)
65–74 2652/14 832 17.88 1397/14 986 9.32 2.09 (1.92–2.28) 237/14 832 1.60 127/14 986 0.85 1.87 (1.42–2.45)
$75 1533/7556 20.29 729/7508 9.71 2.31 (2.05–2.60) 245/7556 3.24 119/7508 1.58 2.04 (1.55–2.68)
Sex
Male 6218/36 759 16.92 3200/36 878 8.68 2.11 (1.99–2.23) 460/36 759 1.25 213/36 878 0.58 2.11 (1.73–2.57)
Female 2460/12 449 19.76 1368/12 380 11.05 1.97 (1.80–2.15) 162/12 449 1.30 102/12 380 0.82 1.57 (1.14–2.17)
Systolic blood pressure,
mm Hg
,100 435/1201 36.22 290/1262 22.98 1.89 (1.51–2.38) 6/1201 0.50 7/1262 0.55 0.88 (0.21–3.71)
100–120 4087/18 428 22.18 2167/18 404 11.77 2.12 (1.97–2.28) 187/18 428 1.01 111/18 404 0.60 1.66 (1.23–2.24)
121–150 3217/21 359 15.06 1642/21 389 7.68 2.10 (1.94–2.27) 291/21 359 1.36 136/21 389 0.64 2.10 (1.63–2.70)
.150 936/8177 11.45 464/8159 5.69 2.10 (1.82–2.43) 138/8177 1.69 61/8159 0.75 2.20 (1.52–3.18)
Heart rate, bpm
,80 4323/25 925 16.68 2188/25 859 8.46 2.13 (1.99–2.28) 296/25 925 1.14 136/25 859 0.53 2.11 (1.64–2.70)
80–99 2867/15 915 18.01 1504/15 974 9.42 2.09 (1.92–2.27) 233/15 915 1.46 113/15 974 0.71 2.03 (1.53–2.68)
$100 1212/5566 21.78 693/5564 12.46 1.94 (1.70–2.21) 91/5566 1.63 62/5564 1.11 1.48 (0.97–2.26)
Prior MI
Yes 1437/7840 18.33 762/7844 9.71 2.06 (1.83–2.32) 124/7840 1.58 63/7844 0.80 1.93 (1.32–2.82)
No 7232/40 966 17.65 3803/41 014 9.27 2.07 (1.96–2.18) 496/40 966 1.21 249/41 014 0.61 1.96 (1.62–2.37)
Downloaded from http://ahajournals.org by on March 4, 2024

Diabetes
Yes 254/2491 10.20 146/2523 5.79 1.83 (1.39–2.39) 57/2491 2.29 26/2523 1.03 2.16 (1.22–3.82)
No 2447/17 143 14.27 1286/17 176 7.49 2.04 (1.86–2.23) 238/17 143 1.39 113/17 176 0.66 2.07 (1.57–2.73)
History of hypertension
Yes 772/7614 10.14 409/7440 5.50 1.91 (1.63–2.23) 163/7614 2.14 74/7440 0.99 2.09 (1.49–2.93)
No 1921/11 907 16.13 1018/12 101 8.41 2.08 (1.88–2.31) 127/11 907 1.07 61/12 101 0.50 2.07 (1.42–3.03)
Killip class
1 6811/39 895 17.07 3500/39 926 8.77 2.11 (2.00–2.23) 442/39 895 1.11 192/39 926 0.48 2.21 (1.80–2.71)
.1 1804/8491 21.25 1020/8489 12.02 1.95 (1.75–2.17) 179/8491 2.11 122/8489 1.44 1.52 (1.12–2.05)
Hours from onset of
symptoms
0–3 1250/7030 17.78 701/7099 9.87 1.95 (1.72–2.21) 86/7030 1.22 33/7099 0.46 2.48 (1.54–3.99)
3–6 1800/9891 18.20 925/9830 9.41 2.11 (1.90–2.35) 135/9891 1.36 71/9830 0.72 1.85 (1.29–2.66)
6–12 2289/12 769 17.93 1240/12 885 9.62 2.03 (1.84–2.22) 157/12 769 1.23 91/12 885 0.71 1.74 (1.25–2.42)
12–24 3059/17 828 17.16 1528/17 744 8.61 2.16 (1.99–2.35) 238/17 828 1.33 115/17 744 0.65 2.02 (1.53–2.66)
24–36 276/1666 16.57 175/1662 10.53 1.67 (1.29–2.17) 5/1666 0.30 5/1662 0.30 1.00 (0.20–5.14)
Site of infarction
Anterior 3110/18 060 17.22 1687/18 254 9.24 2.01 (1.86–2.18) 237/18 060 1.31 122/18 254 0.67 1.94 (1.47–2.55)
Other 5570/31 154 17.88 2885/31 015 9.30 2.10 (1.97–2.23) 385/31 154 1.24 193/31 015 0.62 1.95 (1.57–2.42)
MI indicates myocardial infarction.

with an ACE inhibitor. Second, the benefit occurs during the
first few days after MI, suggesting that mechanisms other
than benefits on the remodeling process may play a role.
These mechanisms may include an early effect on infarct
expansion, a reduction of neurohormonal activation, or an
increase in collateral coronary flow. Irrespective of the
mechanisms, however, these findings strongly support early
initiation of treatment.33 Third, the proportional benefit is
generally larger in higher-risk subgroups, such as those with
anterior site of MI or high heart rate. However, elderly
patients, particularly those $75 years old, are at increased
risk of hypotension with ACE inhibitors, and there is no
evidence of a survival advantage among them. Finally, the
data suggest that the early benefits of '1 month of ACE-in-
 ACE Inhibitor MI Collaborative Group
hibitor therapy started early in acute MI patients is observed
largely during the first week, and it seems likely that this
would be complementary to that observed later in trials of
prolonged ACE-inhibitor therapy initiated several days or
weeks after MI in patients with evidence of heart failure or
left ventricular dysfunction (and the absolute benefits per
month of treatment are of similar size).
In translating these results into clinical practice, two
strategies could be adopted. One strategy involves starting
ACE-inhibitor therapy in acute MI in all patients who do not
have clear contraindications. Such treatment should be re-
evaluated at discharge or after a few weeks and should be
continued long-term only in patients considered to be at high
risk (such as those with extensive left ventricular damage or
obvious heart failure). An alternative strategy involves initi-
ating therapy early only in patients presenting with anterior
infarct and in certain higher-risk individuals, such as those
with tachycardia, heart failure, and perhaps diabetes. By
using a number of these risk markers, we may avoid a high
proportion (but not all) of the deaths potentially prevented by
this treatment. Physicians may justifiably choose either strat-
egy, but irrespective of which they choose, the present results
support the benefits of the use of ACE inhibitors early in the
treatment of acute MI.
Appendix
The ACE Inhibitor Myocardial Infarction
Collaborative Group
Downloaded from http://ahajournals.org by on March 4, 2024
Writing Committee: M.G. Franzosi, E. Santoro, G. Zuanetti, C.
Baigent, R. Collins, M. Flather, J. Kjekshus, R. Latini,
L.S. Liu, A.P. Maggioni, P. Sleight, K. Swedberg, G. Tognoni,
S. Yusuf.
Steering Committee (early and late trials): GISSI-3: L. Tavazzi, G.
Tognoni. ISIS-4: R. Collins, C. Baigent, M. Flather, P. Sleight.
CCS-1: L.S. Liu. CONSENSUS II: J. Kjekshus, K. Swedberg. AIRE:
S. Ball. TRACE: L. Køber, C. Torp-Pedersen. SAVE: E. Braunwald,
L. Moyé, M. Pfeffer. SOLVD: S. Yusuf.
Coordinating Centers: Early trials: Gruppo Italiano per lo Studio
della Sopravvivenza nell’Infarto Miocardico (GISSI): M.G.
Franzosi, R. Latini, A.P. Maggioni, E. Santoro, L. Santoro, G.
Zuanetti. Late trials: Canadian Cardiovascular Collaboration (CCC),
McMaster Clinic: M. Flather, J. Pogue, Y. Wang, S. Yusuf.
Acknowledgments
This study was supported in part by a Canadian Medical Research
Council joint industry award with Astra, Bristol-Myers Squibb,
Hoechst Marion Roussel, Merck, and Zeneca. The collection, anal-
ysis, and interpretation of the data were performed independently of
the industrial sponsors.
References
1. Swedberg K, Held P, Kjekshus J, Rasmussen K, Rydén L, Wedel H, on
behalf of the CONSENSUS II Study Group. Effects of the early admin-
istration of enalapril on mortality in patients with acute myocardial
infarction: results of the Cooperative New Scandinavian Enalapril
Survival Study II (CONSENSUS II). N Engl J Med. 1992;327:678 – 684.
2. Gruppo Italiano per lo Studio della Sopravvivenza nell’Infarto Mio-
cardico. GISSI-3: effects of lisinopril and transdermal glyceryl trinitrate
singly and together on 6-week mortality and ventricular function after
acute myocardial infarction. Lancet. 1994;343:1115–1122.
3. ISIS-4 Collaborative Group. ISIS-4: a randomised factorial trial assessing
early oral captopril, oral mononitrate, and intravenous magnesium
sulphate in 58050 patients with suspected acute myocardial infarction.
Lancet. 1995;345:669 – 685.
June 9, 1998 2211
4. Chinese Cardiac Study Collaborative Group. Oral captopril versus
placebo among 13,634 patients with suspected acute myocardial
infarction: interim report from the Chinese Cardiac Study (CCS-1).
Lancet. 1995;345:686 – 687.
5. Ambrosioni E, Borghi C, Magnani B, for the Survival of Myocardial
Infarction Long-term Evaluation (SMILE) Study Investigators. The effect
of the angiotensin converting enzyme inhibitor zofenopril on mortality
and morbidity after anterior myocardial infarction. N Engl J Med. 1995;
332:80 – 85.
6. Pfeffer MA, Braunwald E, Moyé LA, Basta L, Brown EJ, Cuddy TE,
Davis BR, Geltman EM, Goldman S, Flaker GC, Klein M, Lamas GA,
Packer M, Rouleau J, Rouleau JL, Rutherford J, Wertheimer JH, Hawkins
CM, on behalf of the SAVE Investigators. Effect of captopril on mortality
and morbidity in patients with left ventricular dysfunction after myo-
cardial infarction: results of the Survival and Ventricular Enlargement
Trial. N Engl J Med. 1992;327:669 – 677.
7. The Acute Infarction Ramipril Efficacy (AIRE) Study Investigators.
Effect of ramipril on mortality and morbidity of survivors of acute
myocardial infarction with clinical evidence of heart failure. Lancet.
1993;342:821– 828.
8. Køber L, Torp-Pedersen C, Carlsen JE, Bagger H, Eliasen P, Lyngborg K,
Videbek J, Cole DS, Auclert L, Pauly NC, Aliot E, Persson S, Camm A,
for the Trandolapril Cardiac Evaluation (TRACE) Study Group. A
clinical trial of the angiotensin converting enzyme inhibitor trandolapril
in patients with left ventricular dysfunction after myocardial infarction.
N Engl J Med. 1995;333:1670 –1676.
9. Latini R, Maggioni AP, Flather M, Sleight P, Tognoni G, for the meeting
participants. ACE inhibitor use in patients with myocardial infarction:
summary of evidence from clinical trials. Circulation. 1995;92:
3132–3137.
10. Garg R, Yusuf S, for the Collaborative Group on ACE Inhibitor Trials.
Overview of randomized trials of angiotensin-converting enzyme inhib-
itors on mortality and morbidity in patients with heart failure. JAMA.
1995;273:1450 –1456.
11. Antiplatelet Trialists’ Collaboration. Collaborative overview of ran-
domised trials of antiplatelet therapy, I: prevention of death, myocardial
infarction, and stroke by prolonged antiplatelet therapy in various cate-
gories of patients. BMJ. 1994;308:80 –106.
12. Fibrinolytic Therapy Trialists’ Collaborative Group. Indications for
fibrinolytic therapy in suspected acute myocardial infarction: collabo-
rative overview of early mortality and major morbidity results from all
randomised trials of more than 1000 patients. Lancet. 1994;343:311–322.
13. Latini R, Avanzini F, De Nicolao A, Rocchetti M, and the GISSI-3
Investigators. Effects of lisinopril and nitroglycerin on blood pressure
early after myocardial infarction: the GISSI-3 pilot study. Clin Pharmacol
Ther. 1994;56:680 – 692.
14. Flather M, Pipilis A, Collins R, Budaj A, Hargreaves A, Kolettis T, Jacob
A, Millane T, Fitzgerald L, Cedro K. Randomised controlled trial of oral
captopril, of oral isosorbide mononitrate and of intravenous magnesium
sulphate started early in acute myocardial infarction: safety and haemo-
dynamic effects. Eur Heart J. 1994;15:608 – 619.
15. Kingma JH, van Gilst WH, Peels CH, Dambrink J-HE, Verheught FWA,
Wielenga RP, for the CATS Investigators. Acute intervention with cap-
topril during thrombolysis in patients with first anterior myocardial
infarction: results from the Captopril And Thrombolysis Study (CATS).
Eur Heart J. 1994;15:898 –907.
16. Nabel EG, Topol EJ, Galeana A, Ellis SG, Bates ER, Werns SW, Walton
JA, Muller DW, Schwaiger M, Pitt B. A randomized placebo-controlled
trial of combined early intravenous captopril and recombinant tissue-type
plasminogen activator therapy in acute myocardial infarction. J Am Coll
Cardiol. 1991;17:467– 473.
17. Foy SG, Crozier IG, Turner JG, Richards AM, Frampton CM, Nicholls
MG, Ikram H. Comparison of enalapril versus captopril on left ventricular
function and survival three months after acute myocardial infarction (the
‘PRACTICAL’ Study). Am J Cardiol. 1994;73:1180 –1186.
18. Galcera-Tomas J, Nuno de la Rosa JA, Torres-Martinez G, Rodriguez-
Garcia P, Castillo-Soria FJ, Canton-Martinez A, Campos-Peris JV, Pico-
Aracil F, Ruiz-Ros JA, Ruiperez-Abizanda JA. Effect of early use of
captopril on haemodynamics and short-term ventricular remodelling in
acute anterior myocardial infarction. Eur Heart J. 1993;14:259 –266.
19. Hargreaves AD, Kolettis T, Jacob AJ, Flint LL, Turnbull LW, Muir
AL, Boon NA. Early vasodilator treatment in myocardial infarction:
appropriate for the majority or minority? Br Heart J. 1992;68:
369 –373.
 2212 Systematic Overview of ACE Inhibitors in Acute MI
20. Hochman JS, Srichai MB, Picard M, Palmeri S, Deutsch E, Edelstein R,
Sands M, Sonnenblick E, LeJemtel T, for the CAPTIN Investigators.
Very early ACE inhibition reduces progressive left ventricular dilation
when tPA fails to establish early reperfusion during acute anterior myo-
cardial infarction. Circulation. 1994;90(suppl I):I-19. Abstract.
21. Ray SG, Pye M, Oldroyd KG, Christie J, Connelly DT, Northridge DB,
Ford I, Morton JJ, Dargie HJ, Cobbe SM. Early treatment with captopril
after acute myocardial infarction. Br Heart J. 1993;69:215–222.
22. Sharpe N, Smith H, Murphy J, Greaves S, Hart H, Gamble G. Early
prevention of left ventricular dysfunction after myocardial infarction
with angiotensin converting enzyme inhibition. Lancet. 1991;337:
872– 876.
23. Sharpe N, Murphy J, Smith H, Hannan S. Treatment of patients with
symptomless left ventricular dysfunction after myocardial infarction.
Lancet. 1988;1:255–259.
24. Pfeffer MA, Lamas GA, Vaughan DE, Parisi AF, Braunwald E. Effect of
captopril on progressive ventricular dilatation after anterior myocardial
infarction. N Engl J Med. 1988;319:80 – 86.
25. The SOLVD Investigators. Effect of enalapril on survival in patients with
reduced left ventricular ejection fractions and congestive heart failure.
N Engl J Med. 1991;325:293–302.
26. The SOLVD Investigators. Effect of enalapril on mortality and the devel-
opment of heart failure in asymptomatic patients with reduced left ven-
tricular ejection fractions. N Engl J Med. 1992;327:685– 691.
Downloaded from http://ahajournals.org by on March 4, 2024
27. Stewart LA, Clarke MJ. Practical methodology of meta-analyses (over-
views) using updated individual patient data. Stat Med. 1995;14:
2057–2079.
28. Yusuf S, Peto R, Lewis J, Collins R, Sleight P. Beta blockade during and
after myocardial infarction: an overview of the randomized trials. Prog
Cardiovasc Dis. 1985;27:335–371.
29. Efron B, Morris C. Stein’s paradox in statistics. Sci Am. 1977;236:
119 –127.
30. Peto R, Pike MC, Armitage P, Breslow NE, Cox DR, Howard SV, Mantel
N, McPherson K, Peto J, Smith PG. Design and analysis of randomized
clinical trials requiring prolonged observation of each patient, II: analysis
and examples. Br J Cancer. 1977;35:1–39.
31. Kjekshus J, Swedberg K, for the CONSENSUS II Study Group.
Importance of first dose hypotension with enalaprilat in acute myocardial
infarction. J Am Coll Cardiol. 1992;19:206. Abstract.
32. Bonaduce D, Marciano F, Petretta M, Migaux ML, Morgano G, Bianchi
V, Salemme L, Valva G, Condorelli M. Effects of converting enzyme
inhibition on heart period variability in patients with acute myocardial
infarction. Circulation. 1994;90:108 –113.
33. Van Gilst WH, Kingma JH, Peels KH, Dambrink JE, St John Sutton M, on
behalf of the CATS Investigators. Which patient benefits from early angio-
tensin-converting enzyme inhibition after myocardial infarction? Results of
one-year serial echocardiographic follow-up from the Captopril and Throm-
bolysis Study (CATS). J Am Coll Cardiol. 1996;28:114–121.