JACC Vol, 16, No . 2
Auauet 1990:359-67
339
Cardiac Rupture, Mortality and the Timing of Thrombolytic Therapy :
A Meta-Analysis
MICHAEL B . HONAN, MD, FRANK E. HARRELL . JR., PHD, KEITH A. REIMER, MD, PHD,
ROBERT M. CALIFF, MD, FACC, DANIEL B. MARK, MD, MPH, FACC,
DAVID B- PRYOR, MD, FACC, MARK A- HLATKY, MD, FACC
Durham. North Carolina
This study examined the relation between the risk of
cardiac rupture and the timing of throsbolytic therapy for
acute myocardial infarction. To test the hypothesis that
cardiac rupture is prnented by early thromfaltle therapy
but is promoted by late treatment, randomized controlled
trials; of thrombolylc agents for myocardial infarction were
patient. A logistic regression model including 58 cases of
cardiac rupture among 1,638 patients front four trials
showed that the odds ratio (trestedlcontra0 of cardiac
rupture was directly correlated with time to treatment (p =
0.01) ; at 7 b, the odds ratio was 0 .4 (95% Confidence limits
0.17 to 0.93); at 11 h, it was 0.93 10.53 to 1 .60) and at 17 h,
it was 3.21 (1.10 to 10.1).
Analysis of data from the Grapple Itagnno per t Studio
Cardiac rupture is a catastrophic complication that occurs in
3% to 4% of patients admitted to the hospital with acute
myocardial infarction (1,2) and has been found in 38% of
patients at autopsy in clinical trials of thrombolytic agents
(3-18) . Cardiac rupture occurs only after acute tansmural
myocardial necrosis (2,19-21), and because reperfusion limits transmural necrosis (22,23), we postulated that early
thrombolytic therapy would decrease the risk of cardiac
rupture (Fig . 1). In addition, however, thrombolytic therapy
promotes hemorrhage into the fresh infarction (10.24-33).
From the Division of Cardiology . Department of Medicine . Division of
Bionelty. Deportment of Community and Family Medicine and Department
of Pathology, Duke University Medical Cuter, Durham, Noes Carolina .
This work was wpponcd in past by Rerearch Grants HS-05635 and HS44873
fmm the National Center for Health Services Research, HyHbsvilk . Maryland . Research Grand HL-17670 and HL-36567 f the National Head .
Lung, and Blood Institute . Bethesda, Maryland and trusts from the Andrew
W. Mell n Foundation, New Yak, New York and the Robert Wood Johnson
Foundation, Ftioceton, New Jersey. It was prcsemed in pan at the List
Annual Scientific Session of the American Heart Association. Washington,
D.C ., November 1n.
Manuuript received May 22.1989: revised manuscript received D,,,..
be, 27.1989. accepted March 14, 19911 .
Address for aadma: Michael B . Hunan, MD . 880 Montclair Road, Ist
Floor, Birmingham, Alabama 35213.
a31990 by
the American College of Cardiology
dells Streptoebl ad .eWlahrlu Mfoardlco (G1SS1) trial
independently coWred the relation between time to
thrombolyric therapy aad risk of crdiac rapture (p =
0.03). Analysis of 4,02 deaths In 44,366 Patients deareastrated that the odds ratio of death was also directly,
correlated with that le treatment (p = 0.000) ; at 3 b, the
odds ratio for death was 0 .72 (6.67 to 0.77); at 14 it, it wait
0 .88 (0 .77 to 1 .00) and at 21 b, it was 1 (0.82 to 1,37) .
Thrombollle therapy early other Kate myocardial Infarction improves survival and decreases the risk of caratrc
capture . We admialstrataa of Irembdytlc therapy also
appears to Improve survlrat but my Increase, the risk of
cardiac rupture.
(j Am Cog Csudiol 1990;16 359-67)
and because cardiac rupture is caused by dissection of blood
through regions of transmural necrosis (19,20,34-36), we
also hypothesized that [ate thrombolytic therapy would
increase the risk of cardiac rupture (Fig. I).
Thrombolytic therapy during the early hours of myocardial infarction improves survival (3,37-41), and recent data
(38) suggest that later treatment may also improve survival .
The effects of late therapy have been less well defined,
however, and excessive delay may actually be detrimental if
the benefit of therapy is reduced while the risk of therapy
remains the same oreven increases. To evaluate the effect of
the time to treatment on the risk of cardiac rupture and
all-cause mortality, we performed a meta-analysis of randomized controlled trials of thrombolytic therapy for acute
myocardial infarction .
Methods
Reported raadombmd control trials of thrombetytle therapy. The Bibliographic Retrieval Service (BRS) program
accessed reports pub!ished before March l, 1989 and is,
dexed in the MEDLINE Information System (42,43) under
the terms "fibrinolytic agents," "myocardial infarction."
0735-1097l%l$3.50
360
JACC Vd. 16, No.2
., 1990 :359-67
Aug
HUNAN ET AL.
CARDIAC RUPTURE AND THROMBOLY TIC THERAPY
®noon, myo<ord,um
paw of sand
h+mcnCn
g r..n.Or mrooordmm
M
m.o, nnmonna9~o
outcome group indicators were duplicated as many times as
there were actual patients in the corresponding class in each
study . By adjusting for treatment, each study served as its
own control, and studies with different patient groups could
be combined in a meaningful way, provided that the randomized groups within each trial were comparable . Trials were
also combined by the Mantel-Haenszel method (46,47) into
time interval subgroups for display of odds ratios and their
associated 95% confidence intervals .
Results
Figure 1 . Proposed pathogenesis of cardiac rupture as affected by
thrombolytic therapy .
"random," "trials" and "clinical trials." This search was
supplemented with known pertinent references and their
bibliographies . Articles were selected if they reported 1)
randomized controlled trials of thrombolytic agents for acute
myocardial infarction, 2) cardiac rupture or mortality data,
3) the mean time to treatment, and 4) five or more outcome
events. An additional criterion for analysis of the effect of
thrombolysis on cardiac rupture was that autopsies be reported on m50% of the fatalities in the trial because previous
studies (3,44) have shown a poor correlation between clinical
suspicion of cardiac rupwre and autopsy findings .
Statistical methods. The results of these trials were
pooled to estimate the odds ratios of events (cardiac rupture
or death) in treated versus control patients with respect to
time from symptom onset to treatment :
Odds unto =
Treated patients
Untreated patients
with event
without event
Treatedpatients X Untreated patients'
without event
with event
An odds ratio <1.0 implies a reduced risk of an event
after treatment, whereas a value >1 .0 implies greater risk.
The mean time to treatment was defined as either the mean
time to treatment for an entire trial or the midpoint of a time
interval for a subgroup of patients in a trial .
Logistic regression models (45) were used to assess the
relation between the risk of an event and 1) the treatment
group, 2) the mean time to treatment, and 3) the interaction
between treatment effect and time to treatment . The latter
term assesses, the degree to which the treated/control odds
ratio varies with time . All groups were weighted by the study
sample size so that, in effect, the mean time, treatment and
Trial selection. From 503 articles reviewed, 42 trials
(4,8-17,37-41,48-76) involving 44,346 patients and 4,692
deaths met criteria for analyzing the effect of the timing of
thrombolytic therapy on mortality . Thirty-two of these trials
(38-41,49-76) were excluded from the analysis of cardiac
rupture because either cause o : death or autopsy data were
not reported; four trials (10,11,14-16,37,77) were excluded
because autopsies were reported on <50% of patients who
died and two trials (17,48) were excluded because only one
or two events were observed . Thus, four trials (4,8,9,12,13)
involving 1,638 patients and 58 episodes of cardiac rupture
met criteria for analysis of the effect of the timing of
thrombolysis on cardiac rupture (Table 1) .
Timing of thrombalytic therapy and cardiac rupture. Intravenous streptokinase was the thombolytic agent in all four
trials used to assess the effect of thrombolysis on cardiac
rupture, although the loading dose, total dose and duration
of the infusion differed (Table i) . The mean time to treatment
for the trials ranged free, 9 to 21 h . Baseline characteristics
thought to be risk factors for cardiac rupture, including age
(1,19-21), female gender (1,19,78), lack of previous angina
(19-21,79) or myocardial infarction (2,19-21,79), hypertension during the acute phase of infarction (19,20,78) and
electrocardiographic evidence of transmural infarction
(21,79,80), were all similarly distributed between treated and
control groups within each trial and among trials where data
were available .
The odds ratio of cardiac rupture increased significantly
(' = 6 .2, p = 0 .013) with delay in the mean time to
treatment (Fig . 2). The regression line crossed an odds ratio
of 1 .0 at 11 h after symptom onset and the 95% confidence
limits associated with this odds ratio were < l .Ofor treatment
begun within the first 7 h, suggesting a beneficial effect of
early treatment on cardiac rupture . Conversely, the 95%
confidence limits for treatment initiated >17 h after symptom onset were > 1 .0, suggesting a detrimental effect of late
therapy on cardiac rupture .
The Gruppo Italiano per to Studio della Streptochinasi
nell'lnfarto Miocardico (GISSI) trial (37,77) did not meet
entry criteria for the primary meta-analysis because autopsy
was performed in only 16% of the 1,386 patients who died in
the hospital, yet the results of this trial merit further study
JACC Vol. 16. o .2
. 1990359-67
A
HONAN ET AL.
CARDIAC RUPTURE AND THROMBOLYT IC THERAPY
36 1
Table 1. Few Trials Used to Assess the Effect of Thrombolytic Agents on Cardiac Rupture
Time to Treatment
(hl
Study (rereretee)
ECSG (12,13)
EWP(9)
Heikinheimo .[a]. (s(
Amery et al . (4)
Agent
Range
IVSK
512
Mean
9
024
9
<72
la
<2
21
Glue
IVSK
Rep
IVSK
Gluc
IVSK
Hip
0
Autopsy
.,
Pa
.
No. of
Cardiac
Rupture9
No. Treated
56
58
61
52
64
65
60
67
71156
11(159
10373
141357
51219
2/207
703
204
Odds
Ratio
0.63
0.67
2.4
3.8
ECSG = European Cooperative Study Group ; EWP = European Working Parry ; Glue = glucose; Hip =
heparin ; IVSK = intravenous streptokinese.
because of the Isrge .^.ember of patients randomized . Cardiac
rupture was the cause of death in 58% of autopsy patients in
the GISSI trial . Cardiac rupture was documented in 1 .1% of
control patients at a fairly constant rate regardless of time to
entry into the study . However, among those treated with 1 .5
million U of intravenous streptokinase, the incidence rate of
cardiac rupture increased progressively from 0 .7% at 0 to 3
hto1 .2%at3to6h,1 .3%at6to9hand2 .0%at9to12h
from symptom onset (Fig. 3A). The odds ratio of cardiac
rupture (treatment/contrul) increased significantly as time to
Fig- 1. A. Odds ratio of cardiac rupture (CR) in treatedkontrol
patients versus mean time to treatment in the G1SSI trial (77). Error
bars show odds ratios and 95% confidence intervals (Cl) for each
time interval . The logistic regression model estimates of this association ()e = 4.6, p = 0.032) and its 95% confidence intervals are
displayed. B, Odds ratio of death from cardiac rupture or electromechanical dissociation (EMD) in treated/control patients versus
mean time to treatment (AZ = 6 .7, p = 0.010).
FIgwe 2. Odds ratio of cardiac rupture in treated/control patients
versus meantime to treatment . Error bars show odds ratios and95%
confidence intervals (Cl) for the four trials with autopsies on >50%
of patients who died . The logistic regression model estimates of this
association (yr = 6.2, p = 0 .013) and its 95% confidence intervals are
displayed .
o emd- d.1 .
..d.1 96 .
m
0 ."
cr sat.
Modal eon CI
- R,aM..lon i.odN
362
HUNAN ET AL
CARDIAC RUPTURE AND THROMBOLYTIC THERAPY
treatment was delayed (XI = 4.6, p = 0.032), with a bensficial effect estimated for treatment within 5 h .
It has been suggested (81) that deaths from electromechanical dissociation in the setting of acute myocardial
infarction should be attributed to cardiac rupture . If fatalities
from electromechanical dissociation in the GISSI trial
(37,77) were attributed to cardiac rupture, then the detrimental effect of delay in treatment would persist (32 = 6.7, p =
0.010) (Fig. 3B). Deaths from cardiac rupture or electromechanical dissociation were reported (77) in 3 .0% of control
patients and in .18%, 3 .0%, 3.9% and 4 .7%, respectively, of
patients treated with streptokinase from 0 to 3, 3 to 6, 6 to 9
and 9 to 12 h after symptom onset.
Timing of thrombolytic therapy and sill-cause mortality. In
the 42 trials (4,8-17,37-41,48-76) pooled to assess this
effect, intravenous streptokinase was used in 19 trials
(34,896 patients), intracoronary streptokinase in 6 (568 patients), intravenous plus intracoronary streptokinase in 1
(533 patients), intravenous tissue-type plasminogen activator
in 6 (6,211 patients), intravenous anisoylated plasminogenstreptokinase activator complex in 8 (1,960 patients) and
intravenous urokinase in 2 (178 patients) . Results from nine
of the trials could be further divided into 32 subgroups by
time to treatment . The mean time to treatment for the 65
trials or trial subgroups ranged from 0.7 to 22 h (Table 2) .
The odds ratio for all-cause mortality (Fig . 4) was directly
correlated with time to treatment in the pooled studies ()? =
7 .6, p = 0.006). The odds ratio did not reach 1 .0 until 21 h
after symptom onset, and the associated 95% confidence
limits were <1 .0 for treatment given before 14 h. Whether
the mortality rate is actually increased by treatment given
_•21 h after symptom onset could not be assessed because
there were no data beyond this point . When this analysis was
repeated for each of the thrombolytic agents individually,
the time dependence of their effect on mortality was not
si fnificantly different from that noted for the overall group.
Discussion
This study suggests that the risk of cardiac rupture after
myocardial infarction is directly related to the timing of
thrombolytic therapy and that early treatment decreases the
risk of cardiac rupture, whereas late treatment increases the
risk of this complication . The analysis confirms that the
beneficial effect of thrombolytic therapy on survival diminishes as treatment is delayed (3,37-40), but it also suggests
that this benefit persists much longer than 6 h after symptom
onset (38,47).
Cardiac rupture . Myocardial rupture occurs in regions of
completely transmural myocardial necrosis (2,19-21), ust ;
aly after extensive hemorrhagic transformation of the acute
infarct (21) . The hemorrhagic transformation is usually contained within the central region of necrosis where ischemia
has been severe enough to cause necrosis of the microvas-
JACC Vol. 16, No. 2
A,gun 1990:359-67
culature (22,23,27,82-84). Myocardial hemorrhage that results in cardiac rupture begins either at the endocardial
surface or within the myocardium and dissects through the
infarct zone to the epicardial surface (21,34,36,44). Myocardial hemorrhage has been more severe among patients given
thrombolytic therapy (10,25-33), especially if treatment has
been delayed (10,33) or if the infarct has become transmural
(84) . Pathologic observations (10,24-27) from the 254 reported cases (3-18,24-27,37,38,48-50,77,85-105) of cardiac
rupture after thrombolytic therapy describe transmural,
massively hemorrhagic infarcts .
The pathophysiology of cardiac rupture after myocardial
infarction suggests that eat :y thombotytic therapy might
prevent cardiac rupture by limiting the proportion of patients
whose condition progresses to transmural infarction (Fig . 1) .
Conversely, late thrombolytic therapy might allow extensive
necrosis of the myocardium and its sustaining microvasculature, such that initiation of a thrombolytic state could then
lead to transmural hemorrhagic dissection and cardiac rupture.
Both the mete-analysis (Fig . 2) and data from the GISSI
trial (37,77) (Fig. 3) independently demonstrate a direct
relation of the risk of cardiac rupture to the timing of
thrombolytic therapy . Both suggest that very early treatment
is beneficial and that late treatment promotes cardiac rupture, but the estimates of the time course of these effects
remain somewhat imprecise.
Previous studies. Previous reviews (47,90) of the effect of
thrombolytic therapy on cardiac rupture did not examine the
effect of time to treatment and failed to account for the poor
and variable autopsy rates in most of the studies . Vasitomanolakis et al. (90) reviewed the results of two placebocontrolled trials (8,11), noted that rupture occurred in 10 of
41 deaths in the streptokinase-treated groups compared with
6 of 37 deaths in placebo-treated patients and suggested that
streptokinase predisposes to cardiac rupture . Yusuf et al .
(47) pooled the results of 33 randomized controlled trials of
thrombolytic agents for acute myocardial infarction and
found that the cases of cardiac rupture were equally divided
between the treated and control groups . These investigators
(47) concluded that thrombolytic therapy had no effect on
the risk of cardiac rupture.
Mortality. Recent large studies (15,16,37,38) have varied
estimates of how long after symptom onset thrombolytic
therapy may be beneficial . Mesa-analysis has been proposed
(43) as a method for resolving uncertainties in the estimation
of treatment effects when similarly designed clinical trials
are available for data pooling . Our analysis of the results of
42 clinical trials of thrombolytic therapy according to the
time to treatment suggests that the mortality rate may be
diminished by treatment later than the conventional time
window of 4 to 6 It after symptom onset, Although these
estimates are imprecise, the 95% confidence limits of the
odds ratio are <l .0 for treatment within 14 h of sympi m
1ACC Vol. 16. No. 2
August 1990:359-67
RONAN ET AL.
CARDIAC RUPTURE AND THROMBOLVTIC THERAPY
3163
T1hk 2. Odds Ratio of Death in Patients Receiving Thrombolytic Therapy Versus Control Suhdects in 42 Trials (n = 44,346)
study ffG&00400)
171151(37)
1513 .2(781
15153138)
Tim 117t
EWP49)
1t5S113A
ECSG (40)
ASSET 157)
A1 MS (531
Da10114 et 11 . (541
,Si .21331
SeinnOSet4 .1551
H .... ('61
110601,111 .07)
Andaslo, ct IL 1581
Croydon e1 d . (59)
. M . (51(
14019n2
Bachdlcrctl.(6W
]SAM its;
WMm0101 .9431
M0dblal8 ti d . (613
31-1 4, 139)
G9elci et 11 . 1621
NHF10411330(30)
.assand914 .1411
luliOlIR10 .1631
Sdnd3ra .1 (m)
Bell ,( .3 .(611
PARmr06L((6)
WWIVSK1491
10(3 .208)
001,1160101.1071
Topol et 01 . (69)
1al,alf 16.(69)
11,19 1401
Renlop e' 41 (70)
ASSET is')
ECSG (121
EWP(9)
17ISSI1371
WWLCSK(141
Oh- 0142. (717
Kho a a 91. (721
ISIS-2 (317
AIMS (51)
De- 0001 .(731
641(,[ .3.(651
ISIS P110( (745
Aber 01 e1
. D 1)
ECSG (121
GISS1137)
00 e1aC1731
N(9)
PIe10hereaal .(76)
1SIS2 (395
Hcikilhaimo0,0,L(4)
2CSD (12)
01551(37)
Bet, er 01. (65)
EWP(9)
ISIS (191
Be0,111 .(65)
Ammo71L(*1
EWP(9)
Dkili8BNnoeo61 .(0)
Agcn,
(95K
(VSK
IVSK
-PA
(V5K
IVSK
rt-PA
(a-PA
APSAC
LVSK
1VSK
IVSK
APSAC
APSAC
11,531
APSAC
APSAC
APSAC
LVSK
IVSK
APSAC
IV S ICSK
ft-PA
R-PA
IV3K
APSAC
IVSK
IVSK,
IC9K
1VSK
IVSK
IVUK
rt-PA
ICSK
0-PA
ICSK
n-PA
IVSK
IY5K
IVSK
LCSK
IVSK
[CSK
IVSK
APSAC
IVSK
IVSK
LVSK
1VSK
IVSK
1V5K
1VUK
IV5K
IVSK
1 3)
V5K
IV SK
IV SK
IVSK
IVSK
VSK
IVSK
IVSK
IV9K
IV.
IVSK
Mean Time to
Tresemem (h)
0 .7
11.7
1 .5
1 .9
2 .0
30
2 .0
2.i
2.3
2.3
2.5
2,5
2.5
2.6
2.7
2.7
2 .8
2 .9
3 .0
3 .0
3 .1
32
3 .2
3.2
3.1
3.3
3.4
3.5
3 .5
3.5
3.5
3.5
3.6
4 .0
4 .0
4 .1
4 .1
4 .5
4.5
4 .5
4 .5
4 .9
5 .0
5 .0
5 .0
6.2
6.5
6.5
7 .0
7.7
7.5
8.5
9.0
9.0
9.0
9.5
10.5
10.3
( 2_.5
17.0
18.0
20 .5
'_I .0
21 .0
22 .0
No. et
D..h61G140P
Tmamd
521635
29057
77/551
4(74
(21106
220)2 .381
9179
90992
181334
3135
Po(/1,247
3 ,50
4148
7160
:124
3135
91162
0121
541859
41(07
71112
161269
4172
7773
4)51
1)45
I/19
1/34
729
121191
100/1 .178
3)74
375
4/22
91176
(3165
991(304
9149
24'99
21711,849
5!134
5128
120
164121621
141168
421
4184
511413
571302
&47
871693
7114
18185
(/12
2142,0(0
14)145
5144
46282
9107
7)30
(00/1,224
W53
27213
513)
&74
C901r91
951642
4&79
1111957
4171
Ig82
270l7,436
137207
10-0979
310326
4729
157/1,247
9451
939
1967
526
4/30
19(151
522
631829.'
14(1(2
71119
311764
5166
7171
7150
7145
4/19
5)33
Z35
171177
147/1,101
8126
225
2/48
8/199
6'61
129/1.488
34/48
2695
254/(040
131116
524
420
190'1.622
31)176
721
IM6
20206
551293
15150
93559
1114
(61en
oil
2492098
171(76
11(34
401302
12130
to"
132(1,757
2/57
15144
13'29
5/71
&
195% e91R4
1130
iavavah
04910
.
.34-4701
0.57(0.35-0.93)
0.671346_3 .959
0.96(0.)3-0,46)
0 .90 (0.37-2120)
0 .8410
.70-1,014
0.17 40-04-0 .761
0,2310.34-0.98)
0.5610.31-1 .03)
059 (6.12-1.116)
0.67(0.52-0,37)
048(0.11-2.031
1 .09(0,294.71)
0.61(0.22-1.661
0.3810 .07-2.191
D.6110.13-2 .97)
.4) (0.18-0
0
.93)
.08(001-0 .63)
0
0 .07(0
.60-1 .27)
0 .27(0-09-0.851
1 .0710,36-3.14)
OAS 10.25-0 .991
0.72 (0.18-2 .80)
3 .66(0
.7}18.3)
0.5E (411-1 .91)
0.12(0,01-t,02)
6.21(0.92-10 .17)
0 .1810.02-1 .651
2 .6410 .4.5011 .6)
0.63(0.29-21361
0.6510,76-0.85)
0.16(0.09-1.41)
0 .18(0.118-3 .05)
1.76(0 .20-11.0)
0.900 .33-2.451
2 .3610,11-6.751
8.7410 .570.98)
0.5510.21-.42)
0,674035-1311
0 .81 0 .67-0.981
0 .3110 .11-0.891
0.83 (0.21-9
.291
0.21 (0.02-3 08)
0.83(068-106)
0.4210.72-0.83)
0.47(0.11-1911
0.3310.10-1 .10)
0.7610,12-1361
1 .0110.67-132)
0.4810.10-1 .37)
0-87(a. -L'01
.6191.09-4.37)
0
1 .8111.51-7 .293
.14)0
0
.02-1 .73)
D.a9-1 .02)
309-248)
.27 (0
0
.08-0
.87)
1 .1910
.76--88)
.67 (0
0
.26-1 .679
0,66(822-1.951
0.79(0
.60-I .(6)
2.25(0.39-12 .K
1.46(0,69-3.10)
0.16(0.22-2 .01)
ISO(0.SO_S.IS)
M
AIMS = APSAC InlweoOon Monahly Study ; APSAC - 4nooyla(cd p(a0mlmg0n,hcplekioa,e scdralur co1Pka : ASSET = Aag(o-Se4nn/I0ri01 Study
ct Eetty, i o41960lyas: ECSG = gmopean Cooperative study Group ; 3WP - Europe- Working Party ; GI55I = Gmppu lmlurm yet (8 5AIdio de41
Slaeptoohiwsi ERB'[hrto Mioc4rdhl:ICSK - iniracoronaay sleeptukinase : ISAM - (, 6319uuu65(rep(9kilale al AW(C My0011di61nf4K00n Trig; ISIS=
a hl"IOUmati0Dal Study of Infarct Survival ; NHF = 56((onal Heart Foundation ; IV - intavenoos; IYSK = ildrerrwul 0taeptolaei of (VDK
; 0-pA = Deco vbinanl dasuc-- :ypc pla0ndwgen aaivmor, TWO - Thmnlboly, o in Acute Corwury ukduo/on : W WICSK = W8R9n Waahin4l0
.
-k6
In4rawronary Soep(okinoae in Acua, My-dial Infarction ROdomiond Trial . WWIVSK = Western Washil9(nn ln3ayenmls Smepmk[n99e in Aam
Myooardal lnfardion Randomioed'frill .
364
HONAN ET AL.
CARDIAC RUPTURE AND THROMBOLYTIC THERAPY
0
Crude data
mm. ass
-
Manages
-00
Cl
Figure 4. Odds ratio of death in treated/control patients versus
mean time to treatment . Error bars show odds ratios and 95%
confidence intervals (CI) for the 65 trials as combined by MantelHaenseel statistics to form I I subgroups by time to treatment . The
logistic regression model estimates of this association (dz = 7.6, p =
0.006) and its 95% confidence intervals are displayed,
onset, and the regression line did not reach an odds ratio of
1 .0 until 21 h. This analysis suggests that randomized trials
of late thrmbolytis would be worthwhile to establish directly the time limit for benefit of treatment on survival .
Our analysis suggests the existence of a time window
daring which all-cause mortality is decreased by thrombotytic therapy, but in which the risk of cardiac rupture is
increased. The regression line of the odds ratio for cardiac
rupture was >1 .0 for treatment given >1I h after symptom
onset (Fig. 2), whereas the regression line of the odds ratio
for mortality did not reach 1 .0 until 21 h after symptom onset
(Fig. 4). Although these exact time limits are somewhat
imprecise, our analysis suggests that extending the time
window for ahrombolytic therapy may increase the number
of deaths due to cardiac rupture and still decrease the overall
mortality rate.
Limitations . Meta-analysis continues to grow in popularity as a way to resolve uncertainty about clinical questions
addressed in published reports, but the method has an
inherent weakness of using only summary data from pooled
studies. Our analysis of both cardiac rupture and mortality
used the average time to treatment for groups of patients
rather than for individual patients . This grouping impairs the
precision with which the effect of time on risk can be
estimated. However, data available from individual trials
grouped by time intervals from symptom onset to treatment
IACC Vol. 16. No. 2
August 19a:359-67
are concordant with the results of both the cardiac rupture
and mortality analyses (3,37-41,77) .
The similar autopsy rates between treated and control
groups and among four selected trials (Table 1) suggest that
a reporting bias in the cardiac rupture rates in these studies
was not likely to have been responsible for the demonstrated
effects. These data suggest that late thrombolysis increased
the risk of cardiac rupture ; however, as shown in Table 1,
each of these trials used aa smaller initial loading dose and
longer Infusion of streptokinase than are generally used now .
The low initial doses may have been insufficient to halt the
transmuted progression of infarction by failing to achieve
reperfusion, and the prolonged infusions may have increased
the risk of intramural hemorrhage and, secondarily, cardiac
rupture . Data from the GISSI trial (37,77) (Fig. 3) suggest,
however, that even the briefer high dose infusions of strep
tokinase now used may promote cardiac rupture when used
late after symptom onset .
Finally, these findings may not be applicable to other
means of coronary reperfusion such as angioplasty, which
has not been associated with the same degree of myocardial
hemorrhage as occurs after thrombolytic therapy (27) .
Clinical implications . Use of thrmbolytic agents in the
United States has largely been limited to patients in whom
therapy can be initiated within the first 4 to 6 h of the onset
of symptoms of myocardial infarction (106) . Some of the
benefit of early treatment appears to be mediated through
prevention of cardiac rupture . Our analysis suggests that a
decrease in the overall mortality rate may be achievable
even if therapy is administered considerably later than 6 h
after onset of pain. Forthcoming data from several ongoing
randomized controlled trials should help to further define the
time window for benefit from "late" thrombolytic therapy .
Our analysis also suggests, however, that the overall survival benefits of later therapy may be associated with an
increased risk of cardiac rupture . Vigilance for this complicatico should be heightened if thrombolytic therapy is given
late after myocardial infarction because successful treatment
of cardiac rupture is possible with prompt pericardiocentesis
and surgical repair (79,97,105,107-109) .
References
1.
Sievers J . Cardiac rupture in mum mymamiri int sonar . Genauics
1966:21 :125-30.
2.
Rasmussen S . Loth A. Kjoller E . Pedersea A. Cardiac rupture in acute
myocardial infarction. Acts Med Send 1979:205:11-6.
3 . Breddin K, Billy AM, Feebler L, e1 al . Die Kurezeilfibtinolyse beim
akutee Myokardintarkt . Drub Mid Wochenschr 197395:861-73 .
4. Amery A, amber C. Vermeulen HI, Vemtraete M . Sinalo-bliad iondomized multicemre trial comparing heparin and streptokinase lrutmers in recent myocardial infarction
. Acta Mod Scand Suppl 19e9 ;503:
1-35.
5.
Praetaius F. Slreptokinase in myocardial inf ution: results of Girman
studies . Poslgrad Med) 1973 :49lsuppb:114-22 .
JACC Vol. 16, No . 2
August 199B.359-67
6. Hale GS . Streptokinase in acme myocardial infarction: rationale and
preliminary results of a combined therapeutic trial . Australas Ann Med
1970;19lsupp1Y63-5.
7 . Dioguardi N . Lotto A, Levi OF, et al . Controlled trial of streptokinase
and heparin in acute myocardial infarction. lancet 1971 :2:891-5.
8. Heikinbeimo R. Ahrenber5 P, Honkapohja H, et a. Fbrieolytic treatment in acute myocardial infarction. Acla Med Sand 1971:189 :7-13 .
9. European Working Pang. Streptokinase in recent myocardla infarction:
a eontrdkd multicemre trial. er Med 1 1971 :3:3'3-31.
10. Berry CL. Thrombolyric ihanpy and myocardial infarction . J Can
Pathol 197518
.333-6.
I I . Aber CP, Bass NM, Berry CL . el a. Streptokinase in acute myocardial
infarction : a controlled multicentre study in the United Kingdom . Br
Mod 1 1976:2 :1100-4.
12, European C.opcmtieo Study Group for Sureptokinaae Tremment in
Acute Myocardial Infacton . Streptokinase in acute myocardial infclion. N Eogl J Med 1979 ;301 :797-802.
13. Velalmete M, van do Loo J, Jcsdimky NJ . Streptokinase in acute
myocardial infrction . Arta Med Sound SuppI 1981 :684:7-57 .
14. Kennedy JW, Ritchie JL, Davis KB . Fritz JK. Western Washington
Randomized Trial of Inlracoronary Slrrplokinase in Acute Myocardial
Infarction. N EngI J Med 1983 ;309:1477-82.
15 . The I .S.A.M. Study Group. A prospective trial of Inttoven mu Soeptokinase in Acute Myocardial Infarction (I.S .A .M .I. N Engl J Med
1956014 :1465-71.
16. Schroder R, Neuhaus K, Leizorovicz A, Linderer T . Tebbe U . A
prospective placebo-controlled dmbk-blind multicenter trial of intravenous Streptokinase in Acute Myocardial Infarction (I .S.A .M.) : longterm atonality and mmhidity . I Am Coll CaMiol 1987 ;9:197-203 .
17. O'Rourke M, Baron D, Keogh A, el a1 . Limitation of myocardial
infarction by early infusion of recombinant Iiss .e-type plasminogen
activator. Circulabn 1911:77:1311-5 .
18. White HD, Rivers 1T, Maslownki AN, a a . Effect of intravenous
streptokitase as compared with that of tissue plasminogen activoraf an
left ventricular faneti, n after first myocardial infreiue . N Enol 1 Mod
1989320:817-21 .
19. Wessler S. tall PM, Schlesinger MI. The pathogenesis of spontaneous
cardiac rupture. Circulation 1952;6:334-51 .
20. Schuster Elf, Bulkley BH . Expansion of Intramural myocardial infarction : a pathophysiolugic factor in cardiac .
rupture Circulation 1979 ;60:
1532-8.
21 . Lewis Al, Burchell NO. Titus JL. Clinical and pathologic features of
post infarction cardiac rapture . Am J Cardiol 1969:23:43-53 .
22 . ReimerKA,JenningsRB .The"wavefrmtphenomean - ofmyocardial
iachemie cell death. II . Trammuel progression of necrosis :vithin a
framework ofischemic bed size (myocardiumat risk) andcollateral lbw.
Jab memo 1979 ;40:633_44.
23 . Rime, KA, Loon JO, Rasmussen MM, Jennings R11 . Th, -front
phenomenon of ischemic cell death. 1 . Myocardial infarct size vs
duration of coronary occlusion in cup . Circulation 1977;b:786-94.
24. Masuda M, Fujiwan H, Onodera T. It a. Quantitative analysis of
infant sift, contraction band necrosis in human aumpsied hearts with
acute myocardial infarction after treatment with selective intracoonary
lhrombulysis. Circulation 1987;76:981-9.
25, Schachennayr W, Haferkatep O . Dee haenorrhagische Herzinfmkt .
Dlsch Med Wechenschr 1972 :97:1172-4 .
26. HoilanderG, Ozick H . Anselmo M. Sanders M, Lichstein E . Greengut
A. Myocardial rupture following intracoronary Ihrombolysis . NY Sister
Med 1984;84 :129-31.
27. Walter BF . Rothbaum DA, Pinkerton CA, 1 0 . Status of the myowdium and infarct-relined coronary artery in 19 eecropsy patients with
&cult recanalization using pharmacologic Istrepldsinase, r4issue pfs-
HONAN ET AL.
CARDIAC RUPTURE AND THROMBOLYTIC THERAPY
365
ino0en activmonl. mechanical (pereumnrnncs teansluminal coronary
angioplasty) or combined types of reperfrtion therapy . I Am Cob
Candid 1987 ;9:785-801 .
28. Matbey DE, Sehofer J. Kick K, et a . Traruonna4 henwrrhagie rny cardial „darction after intrxoroaary streploklnase : clinical . angbgraphic- and necropsy OOdirgs . Br Heart 119B2M8:516-51.
29 . Kao KJ . Hackel DB. Kong Y . Hemorthagk myocardial infamies after
srreprokiuae Ireoment
.
mss c«.nay tamr-6osis. Arch Pathal lab
Med 1984;109:121-4 .
30. Mahfeldt T, Schwarz F. Schuler G. Hermann M, KONer W . Necropsy
evaluation in seven patients with evolving acute myocardial infarction
uemcd with thrmnb.lylie therapy. Am I Cannot 19&;54:530-4.
31 . Yaum M. End r S, Takahashi M. et a. Aogiogmphic and pm8olcgic
evidence of hemorrhage into the myocardium after coronary repeefa.
non. Atgiobgy, 10043fi797-8ot.
32. Topol EJ, Herskowitz A, Hutchhu GM. Massive hemorrhagic myocardial infarction after coronary thranbolysis . Am 1 Med 1916;81:339-43 .
33. Fuji-am H, Onoden T, Tanaka M, et aL A diniaopat ologic study of
patients with hemorrhagic myocardial infarcliml treated with selective
,,rRmary thrmnbnlysis with umkinase. Circulation 190h;73 :749-57.
34 . Datta BN, Bowes VF, Silver MD. Incomplete raptor: of the heart with
divcnicrdum formation. Pathology 1975;7 :179-g5 .
35 . Kamil M. Miller M . Messelof CR . Lepow IN. Cardiac ruplufe in
myocardial infarction . NY Slate J Med 196262 :2327-35.
36. Laulsch BY, Larks KW, Pathogenesis of cadiac rupture. Arch Pathol
1967:94264-71 .
37. Gmppo Raliara per to Studio delta Streptochinasi eelPlnfarto Miocardico IGISSD . EBedivenessof intravenous thrmnholytic treatment in
acute myocardial infaction, Lancet 1986 ;1 :397-402 .
38. la1S-2 Collaborative Group. Randomised trial of intravenous soeplokisn . mat aspirin, both, or reitreranwor i7187 casesofsaspected route
myocardial inlkroion: IS1S-2. Lartcef 1918 :2:349-60.
39. Simoons ML. Ser rays PW, van den Brand M, et a. Improved survival
after early Ihromhaysis in acute myocamia iofamim : a randomised
trial by the Imeruniversity Cardiology Institute in The Netherlands .
Lancer 19432:578-82.
40. Van de Wee° F. Arnold AER, Eumyean Cooperative Study Group for
Recombinant Tissue-Type Plaminogen Activator . Intravenous tissue
pl.-M . adivaoc and size of infant, kit ventricular function, and
survival in acute myocardial infarction. Or Med J 1918397:1374-9.
41 . BasandJP,FaivreR,BecqueO .eta.Intravenous streplohieaseversus
heparin in recent acute myocardial hdarction : multiannue randomized
trial in ire French Cmne area. Haemostats 1966;16(suppi 3) :130-9.
42 . HaynesRR,McKibbenKA,Wa&crCJ.eta.Commrtersearcdrgofthe
medical literature: an evaluation of MEDLINE searching systems. Aim
Intent Med 1985 :103:812-6.
43. Sack" HS, Bonnier 1, Reitnum D, Antonio-Berk VA, Chamers TC .
Mera-analyses of naadorsioed controlled trials . N Earl I Med 1987816 :
450-5 .
44. Bates RJ, Beutkr S, Remekov L, Anageostopoulos CE . Cardiac rap
iurecha etgn in diagnosis and management . Am J Cardbl 1977;40 :42937.
45. 1-1- ca
3 FE Jr. The LOGlST procedure . In : SUGI Supplemental Library
User's Guide. 5th ed. Cary. NC: SAS Institute. 1986:26" .
46. Mantel N, Nanterre] W, Sfaislical aspects of the aolysis of data from
retro"pectire studies of disrmc. J Nod Cancer lost 1939;22:719-4g.
47. Yusuf S. Collins R, Per. R, et a. Imravecau and intracmmury
fibricolyrk therapy in acute myocardial infantion : overview of eee •dls
on mortality, rcildarchon and sideelfectsfrma 33 randomizedconuolkd
teals, Ear Heart J 1945 ;6:556-h5.
48. While HO, Norris RM, Brown MA, 11 a. Effect d' intravenous smep-
366
HONAN ET AL .
CARDIAC RUPTURE AND THROMBOLYTIC THERAPY
toanase on left ventricular function and catty Survival after acute
myocardial infarction . N East J Med 1937317,0- .
49 . Kennedy 3W . Martin GY, Davis KB. a W. The Wastes Washington
Intravenous Slreptokinase in Acute Myocardial lafatctton Randaadzed
Trial. Circulation 1988 ;77 :1311-5.
30. National Heart Foundation of Australia Coronary Thombatysis Group,
Coronary thrombolysis and myocardial savage by sisvm plasminogen
aclivatorgiven up to4 hours afleronset of myocardial infarction . Lancet
1986 ;1203-6 .
51. Meinerla T, Kasper W, Schumacher M, Just H, APSAC Mallicenter
Trial Group. Th . German muhicmns trial of anisoylated plasmiesse r
atreplokinase activator complex versus heparin far acde myocardial
infarction . Am J Candid 1908:62347-51 .
52. Wilcox RG, van der Lippe G, Olsson OG, Jensen 0, Skeet AM,
Hampton JR. Trial of tissue plasminogen activator for monalily reduction in acute myocardial infarction
: Ang!o-Scandinavian Study of Early
Ttnombolyaia(ASSET).lancet 19n8,2:525-31.
53. AIMS Trial Study Group. Effect of intravenous APSAC on mortality
after acute myocardial infarction : preliminary report of a placebocontrolled clinical trial. Lancet 1988;1:545-9.
54. Durand P . Asseman P. Provost P. Bertrand ME . Lablanche JM, Thery
C . Eifidivrnern of inlravennm st .ptokinanc on if-t nine and ten
vcntricalar function in acute myocardial infarction: prospective and
randomized study. Chn Cardio] 1967 ;98383-92 .
55. Sainsous J, Bonnet JL, Serradimigni A, SIVIM firoup . Intravenous
Streptokinase versus heparin in the acute state ofenyecardlal infarction :
a prospective randomized trial in south-east France . Hamostmis 1966;
16(supp131:140-7.
56. Bossaert LL . Safety and tolerance data from The Belgian muttrcenta
study of anisoylated plasminogen slreptokinase aetivmor complex versus heparin in acute myocardial infarction . Drugs 1987;33)suppl3L28792.
57. Ikram S, Lewis S, Bucknall C, al al. Treatment of acute myocardial
infarction with anisoylated plasminogen strclnoki se activates complex. Br Med 11986293 :786-9 .
58. Anderson IL, Marshall HW, Bray BE, al al . A modernized trial of
inoocamnary stoeptokirtanc in the treatment of acid : ropocondial dare.
tion. N Engl 1 Med 1983308 :1312-8 .
09. Croydon EAP. Preliminary safay and tolerance data oblaoead in a
comparative study of anisayialed plasminogen ntrepmkimse activator
complex versus heparin . Drugs 198735(supp131:293-6.
60. BuchalrerME,BourkelP,JenningsK.aal.Theeffect ofdranbulylic
therapy
adsoylate_ plasminogen streptokitase activator eumplex
on the indicators
will
of myocardial salvage . Drugs 1987:33(suppl 31209-15.
61 . Macbant JC, Peycelon P, Cardot IC, Vcadeal J, Fawn! D, Cannel M .
Value of myocardial defect size measured by thallium-201 SPELT :
results of a multin :nler trial comparing Mparin and a new thromholytic
agent . I Nucl Med 1988129 :1486-91.
62 . Guerct AD, Gerstenhlith G, Drinker JA, el al . A randomized trial of
smnveoous tissue plasminugen activator far -1c myocardial inthmtion
with subsequent rrodomiration to elective coronary angioplasty. N Ed,]
J Med 1987;317:1613-8.
63. Julian DO, Borlhwick LS, Reid D . Anisdylated pasminogen streplokiouse activator complex versus placebo: a preliminary muhicenare study
of safety and early mortality in acute myocardid infarction, Drugs
1987 ;33(suppl 3) :261-7.
JACC Vet. 16, No. 3
August 1990:339-67
holytic therapy in acute myocardial infarction : a prospective, ratedneticed, rontrolkd trial . Am J Cardio1195553 301-8 .
67 . Samdli C, Spallarnssa P . GhigJinui G, et al . Peaking time of ereatine.
kinase MB in patients treated with urokinase or conventionally during
mute myacandial tnfaretlon : is It really a clue to rperfuaion7CaNtobpaa
19gSa3 :~9-7d.
68 . loyal EJ. Moms DC, Smalling RW, et al. A aatticenter, rmdmniaed,
plaeehacentroaed trial of a new farm of Intravenous reenmslrdnt
tissue-type plasminogen activator (Activate) in acute myocardial if-line. J An Coo Cardiul 1987:9:1205-13.
69- I<OroffRH, KM RJ, Wassmman AG . A randomized, angiographirally
controlled trial or inlracoronary stteptohinase in 00011 myocerdiul infarclion. Am J Card!.] 19M;53:404-7 .
70 . Rentrop KP, Feit F. Blanke H . Effects of intacoronary slreptokinase
and inlmeoronery nitroglycerin infusion on coronary aegtographic patterns and mortality in patients with acme myocardial infarction . N Rout
1 Mad 1964311 :1457-63.
71- Dlna Hl. Butmaa SM, Piters KM, et al. A sodomized controlled trial
of 'mtravenaas Streptokinase in evolving acute myocardial intarclion.
Am Heart) 1966 ;H 13021-9 .
72 . Khaja F, Wailon M Jr, Brymer JF, a al . Intracoeunary fibrinolytic
therapy in acute mV canliat infarction: repnd of .pmnedivral
teed trial. N Engl 3 Med 1983308 :1305-11 .
73 . Dewar HA, Stephenson P, Horler AR, et al . Pihrinolyllc Nelapy of
coronary rleombosis controlled trial of 75 cases. Br Med J 1963 ;191520.
74 . ISIS Pilot Study lmestigamrs . Randomized factorial trial of high-time
ml raverous slltpmkoase, orolal aspirin, and of intravenous heparin in
wale eoy000rfiel infarction . Bar Heart J 1987;8:634-42.
75 . Gormsen 1 . Tid in,ou B, Feddersen C, Planar J . Binchemird oval udmr
of low dose of -hi- in acute myocadial infarction: a double-Wind
study. Acre Med &and 1973 ;194:191-B.
76. Fletcher AP, Sherry S, AIIBaersig N, Symooiatis FE, lick S . The
maietecerme of a sustained drrombolytic state in man . In. Clinical
Observations er patients with myocardial infarction nod other thromboembolc disorders, 1 Clin Invest 1959;38:1111-9.
77. Maud F. DeBimc And, Franzmi MG, Pampallona S, Foesnti A, Gasparird M. Anent deft cause di mate intnospedakda- G Ital Cordial
1987 :17:17-44 .
76. Edmansml HA, Movie NJ. Hypertension and cardiac rupture
: a clinical
and psdtdogic study of seventy-two cases, in thtne :n of which reps roc
of rile imerventrcdar seplem occurred . Am New ! 1942a4:719-33.
79. PdddaSimanen S, Mullor JE, Stone PH, et al . Ventricular seplal and
free wall rupture complicating acute myocardial :
infarction
experience in
the Mahiaena Invesliptien of Limitation of Infarct Size . Am Heart J
1995917:609-18 .
80 . Gni1Rm GC, Heade B. Math RW . Pasture in myocardial rupture:
analysis d two hundred and four cases al Los Angeles County Haptal
between 1924 and 1959. Am J Candid 1961 :8:792-n.
91 . IBIS-1 (First International Study of Infarct Survival) Collaborative
Group
. Mechanisms for the early mortality reduction produced by
bela-0Iockade stoned early in acme myocardial infamtion : ISIS-I .
lanes 1918;1 :921-3.
62 . Cent FR, Lotus TZ, Monrs M, Siegel Al . Hemorrhagic infarction
:
a
eaperfttnion injury following prolonged myocardial ischemic anoxia .
surgery 1975;76 :95-104.
64. ScmetterTL,MillcrDH,SilvaeiDA.Moses JW.Borcrid .Randamized
doubleblind trial of intravenous slreptokinase for curse myocUdia
infarction . Am 1 Card it! 1986'.38 :47-52.
6J. Kloser R%, Rudc BE, Cudson N, Maroko PR, DeBoer LWY, Bmunweld K Ultraslmdnrel evidence of ntcrovascuaar planate and myocardial call ipory after Coronary artery occlusxan : which comes firm)
Circulation 1996 ;6J947-52.
65, Bell HIM, Castaldi PA, Hale GS. Australian mullicenlre trial of stroplukinase in acute myocardial infarction, Lancer 1973;1:57-60.
66 . Raizner AB, Tortaledo PA, VCrupi MS, et al . lnloacaalary thrum.
84K fdsbbein MC, Y-Rio J, Laura U, Kamaatsuse K, Mercler IC, Gang W.
The relationship of vascular irdury and myocardial bcmmrhaae to
necrosis Bier reprdiuslon. Circulation 1980 :62 :1274-9.
JACC Vol. 16. No. 2
August 1990:359-61
85. Leas AS . Myles L. Cercek B. Marshal, H. Ganz W. Hemorrhagic
myocardial infarction after thrombolytic therapy: intreoperalive ohnation s (abslr). J Am Coll Cardio! 1987 :9(suppl Ai:62A .
06. Moonier P, Sigwan U. Vincent A . et al. Anisoylated plasminogen
slrepokinase activator complex Venus sloeptokinase in acute my .that infarction: pre5minary results of a randomise-1 study . Drugs 1957:
33lsuppl31'.175-9.
87 . Poliwoda VA . Schroder R . Heckner F. Erste Erfahmngen mil der
fibdnolytischen Therapie helm akuten Myokard nfarkt . Dluh Med
Wochenschr 1963:88 :218-24 .
99. Schmutzkr VR. Heckner F- Kong, P, et al. ZLr thrumbolylischen
Thempie den frischen Herzinfarkies. 1. Einfuhrnng. Behandlungsp)ure.
allgemeineklinische Ergehrisse . DIsch Med Wooteruhr 196691 :581-7.
89. ZeF RH, Phillips SJ . The case for erly direct myocardial reperfusion .
lot J Candiol 1983:4 :99-103 .
90. VenilomaralahivEC .FamutarMA.Koatuwski1 .SchmgerB .LieMJR .
Elleslad MH . Cardiac rupture following intracordnary, streptokinase .
Collier Cardiovasc Diagn 1903;9:291-6.
91 . Miyam oa H. Yazawa M. Yamazaki Y . et al . Successful surgical repair
d left ventricular mpturc after acute myocardial infarclion. Jpn Circ J
1984:48 :1215-21.
92. Von Essen R. Uebis R, Schmidt W, cI al . Inlrokrrotmre streptokinase
behn akutea Herzinfarkt: Erfahmngen brim 461 Patienlen. Ouch Med
Wochenschr 1985 :110:570-5.
93. Meyer 1. Erbel R. Pop T. Perculaneous tmnslumiral corcrary angioolasty (PTCAI following thrombolysis . 2 Kardiot 1985 :74(suppl61:135,gl .
94. Williams DO. Borer J. Bruunwald E, et ai . [atecombinant
tissue-type pirminagen aYlvalor in patients with acute myocardial
infarclion: a report from the NHLBI thrombolysis in myocardial infarc .
tion trial. Circulation 1986:73:338-46.
95 . EthelR,PopT,HendchsK-J,etoI .Percutaneoustransluminaluoronary
angioplasty after thombolytic therapy : a prospective conlmlled randomized trial. J Am Coll Cardiol 1986 ;9:485-95.
95. Mailer VI, Rothbad RL, Fitzpatrick PG . Francis CW. Rapid lysis of
coronary artery thrombi with anisaylaled plasmino8en simptokirose
activator complex . Mn Intern Med 1986:104:304-10.
97. Tanimoto Y . Oh. H . Kobayashi V . Hayashi K- Matsuda Y . Fake
HONAN ET AL.
CARDIAC RUPTURE AND THROMBOLYTICTHERAPY
367
aneurysm fornauon after ventricular rapture associated with n:inlasw
.ion following successfal thomhdysis. Hers Vessels 1987:3:170-3.
98. Sliegel M. Zimnem SH. Rohicsek F. Left venleimdar rapture f.11mving
coronary occlusion treated by Hmptdtinaoeinflation: successful surgical
repair. Ann Theme Sung 1987 .44 :413-5.
99 . Verstraetc M, Arnold AE, &ewer RW, et al. Acute emaeasy tlrarnbalysis with recombinant human tissae-type plasmkroten rlivalor- initial
potency a d influence of maintahted infusion on reoaWsion nit . Am J
Carding 1987:60:231-7.
109. TapolEl.Cal~RM.GeorgnOS.etal.Arandamizedlrintefknnerdiale
versus delayed elective angioplasty after hanvenoas living pbassirsogn
activator in acme myocardial infarction . N Ergl J Med 1987917:591-R
101. Erbe] R, Pap T, Meinenz T, et al . Combination of calcium cb-I
bbcker and thrombdyric therapy in acme myocardial idaction . Am
Heart J 1988:115:529-38.
102. Kambar4 H. Rascal C. Kajiwasa N. et al. Randomized. double-bhaded
center star: camprison of imracwaeary sinakehain umkinase
typo plasminogen activator, far urokinase (GE-9903) and intracormHry
urokmare in patients with acute myocardial infuction . Circulation
1988:78 :899-905.
103. Stammer F . Van de Werf F. Vanleaecke J . Piessens 1. DeGeost H.
Inlmverous strepokinasedlningacule myocarrdiat infarction : apmspec
.
live open study. Ama Cadial 19118:43(supp1):141-7.
104. Loukinen KL.O-Neill WW, Lauren N, Law A. Tinmis GC. Myocardial
rupture complicating tissue pum®ogen activator therapy of acute
myocardial idarclion (abslrl . J Am Cog Candid 1989 ;13(sapp A):94A.
:!05. Carry IS, Cukirgnan RA, Eugene 1. Myocardial rupture in expanded
iofarcls: repair using periwdial psa :h. Cue Cadio11989 :11 :157-60 .
106 . HlatkyMA,CmugaoH,D'ConnurCM,Mark DO,Pry . DO.CsIiffRM.
Adoption o thrombolytic therapy in the management of acute myocer,
dial infrdion. Am J Carded 1988:61:510-4.
107 . Bashour T, Kabbani SS, Eiknson DG. Crew J . Hanna ES . Surgical
salvage orhean euprurc : repent of- -. .ad mirwdshe lilemmre.
An, Thorac Sung 1983 :36:209-13.
109. McMutler MH, Kilgore TL. Dear D, Hindnwn JH . Sudden blowout
rupture of the myocardium after infarction : urgent manaeemen .
J Tharac Cardio- Surg 1985 :89 :259-63.
109. Pugliese P. Tmnmassini G. Macri R . Moscheni R. Erfule, S. Successful
repair of postinfanclion hear rapture .I Cadiovasc Sung 1946;27:332-6