The Wavefront Phenomenon of Ischemic Cell Death
1. Myocardial Infarct Size
Downloaded from http://circ.ahajournals.org/ by guest on September 25, 2017
vs Duration of Coronary Occlusion in Dogs
KEITH A. REIMER, M.D., PH.D., JAMES E. LOWE, M.D.,
MARGARET M. RASMUSSEN, M.D., PH.D., AND ROBERT B. JENNINGS, M.D.
SUMMARY Irreversible ischemic myocardial cell injury develops
in an increasing number of cells as the duration of coronary occlu-
sion is prolonged. The present study quantitates myocardial necrosis
produced by 40 minutes, 3 hours, or 6 hours of temporary circumflex
coronary occlusion (CO) followed by 2 to 4 days of reperfusion, or by
24 or 96 hours of permanent circumflex ligation In pentobarbital
anesthetized open chest dogs. After 40 minutes of ischemia, myocyte
necrosis was subendocardial but with increasing duration of coronary
IN RECENT YEARS, there has been much interest in the
idea that it is possible to salvage acutely ischemic, but still
reversibly injured, myocardium, by either pharmacologic or
surgical intervention.1 S This concept is based on the belief
that not all cells die simultaneously in an area of acute myo-
cardial infarction in man. The evidence supporting this
belief is largely intuitive but is suggested by a variety of per-
manent and temporary coronary artery occlusion ex-
periments in dogs. Although subendocardial necrosis results
from even a brief period (20-40 min) of temporary coronary
occlusion, much of the subepicardial myocardium survives if
blood flow is restored.10' 11 Temporary occlusion studies
employing a variety of models and techniques, have gener-
ally supported the idea that salvageable myocardium is pres-
ent for 2-3 hours after coronary artery occlusion.3' 10, 12-19
On the other hand, permanent occlusion of the vessel
for 24 or more hours causes necrosis which is essentially
transmural.3 These observations indicate that not all cells in
the ischemic myocardium die simultaneously and that, in the
dog, ischemic cells in the mid- or subepicardial myo-
cardium are available for salvage for a period of hours.
The present study was done to assess the progression of
cell death during acute myocardial ischemic injury in dogs
and to determine the amount of ischemic but viable myo-
cardium available for salvage at various times following
occlusion of a major coronary artery. The results show that
irreversible injury develops at variable times. A wave of cell
death begins in the subendocardial myocardium and pro-
gresses toward the subepicardial myocardium. The sub-
endocardial myocardium dies early, while many cells in the
subepicardial myocardium survive for up to six hours
following coronary occlusion. These results provide an
anatomic baseline in dogs for evaluating pharmacologic or
surgical attempts to modify infarct size.
Materials and Methods
Mongrel dogs weighing 9-20 kg were assigned to one of
five groups as described below. Each dog was anesthetized
From the Department of Pathology, Duke University Medical Center,
Durham, North Carolina, and Northwestern University Medical School,
Chicago, Illinois.
Supported in part by NIH contracts NOI-HV-52999 and 72-2984 and by
grants HL-18833 and HE-08729.
Address for reprints: Keith A. Reimer, M.D., Department of Pathology,
Duke University Medical Center, Durham, North Carolina 27710.
Received February 24, 1977; revision accepted June 2, 1977.
786
occlusion, irreversible injury progressed as a wavefront toward the
subepicardium. Transmural necrosis was 38 4% after 40 mi,
57 7% after 3 hours, 71 7% after 6 hours and 85 2% after 24
hours of ischemic injury. These results document the presence
of a subepicardial zone of ischemic but viable myocardium
which is available for pharmacologic or surgical salvage for at
least three and perhaps six hours following circumflex occlusion
in the dog.
with intravenous sodium pentobarbital (30 mg/kg), un-
derwent endotracheal intubation, and was ventilated with a
Harvard 1063 respirator. Sterile instruments, sponges and
drapes were used during the surgical procedure. A catheter
was placed in the right femoral artery via its saphenous
branch to monitor blood pressure. Lead II of the standard
ECG and peripheral blood pressure were recorded on a
Brush model 440 recorder at standard intervals throughout
the experimental procedure. The left chest was opened
through the fourth intercostal space. The lung was retracted
and the pericardium was opened. The circumflex artery was
isolated 10-15 mm from the aorta and one #I silk suture was
placed around it.
Three groups of dogs underwent temporary left cir-
cumflex coronary artery occlusion followed by two-four
days of reperfusion: Group 1: 40 minute occlusions; Group
2: three hour occlusions; and Group 3: six hour occlusions.
In these dogs, occlusion was produced by pulling both ends
of the suture through a small glass tube and fastening the
suture ends with the tips of a mosquito hemostat, thereby
kinking the artery within the end of the tube. In two groups,
occlusion was permanent (Group 4: 24 hours; Group 5: 96
hours) and was produced by double ligation of the artery.
Subsequent to reperfusion in groups 1-3 and subsequent
to ligation in groups 4 and 5, incisions were closed, air was
removed from the chest and dogs were allowed to recover
from anesthesia.
Determination of Extent of Necrosis
After the appropriate period of occlusion with or without
reperfusion, each dog was re-anesthetized, the chest
reopened and the heart excised. The circumflex artery was
opened to establish the point of occlusion and ensure that
arterial thrombi had not developed in the reperfused groups.
The left ventricle was opened and three transmural slices
were cut longitudinally through the posterior papillary mus-
cle which is centrally located within the area made ischemic
by circumflex coronary artery occlusion.20 The slices were
photographed and fixed in 10% phosphate buffered formalin
for preparation of histologic sections. Myocardial necrosis
was identified in sections stained with hematoxylin and eosin
(H&E) and was quantitated from photographic enlarge-
ments of the sections stained with Heidenhain's variant of
Mallory's connective tissue stain (CT) (fig. 1). (The latter
stain accentuates contraction bands and cytoplasmic
 WAVEFRONT OF MYOCARDIAL NECROSIS/Reimer," Lowe, Rasmussen, Jennings 787

FiGuRE 2. A transmural Slice throt4h the poster-wr papillary

FiGu;tE 1. ArOotive prinit qf a hi'stolbgic seciion stained with musek jivrn a dog sacrificed thme dM qfter a 40 min occlusion.
Heidenhain's wrtoant Qf Udibry's connective t0sue-stain (CT-) to
i 77he yetlbw ama in the subendocaidial zone (top qf phologrqph) U*
demonstrate the technique qfquantitating nec?MU. no section was necrotic. Focal hypererni and emormW can be seen in e center
Qf ihe necrotic area. ne outer two thir& qfthie wall is Wable m3v-
preparedjivm a transmura! sike iqf heart cut 16"Itudinally ihmugh cardium which was isehemic bui was sWvoged 4 reperfitsion,
the posten'or papillary mtack. ne nogatiw print %w made by Prow
jeciin the histologic sectibn ftom a photogrVhk enlarvr onto
black and whiie photographic pqper. ne neerotic arem out. (especially 3) subepicardial myocardium. Siiben-
Downloaded from http://circ.ahajournals.org/ by guest on September 25, 2017

were in group
Uhed, withjrequent mkrosc QPe ver.ftcatjon, One tmcedphotogtqph do cardial hemorrhage was prominent. in addifim, there
I i

jivm each qf three serial sikeslheari %w phoiocVied and the were often small, pale gray foci, usually in the center of the
necrotic and viable areas on the cppi'es were cut out and we
papiollary
I muscle, which grossly appeared to contain no
0 ve mfl nec Pwis vms calculated ias I h e p emen t qf myorardiiim ih th e blood (fig. 4). These pale gray fbc'i, sur un
three immmuml sections qf liqpiltaq muscle. 7he ittlitret M this y
section, ohlainedjivm a dog sacrificeiijbur dq)s qftep a three hour
hemorfhage. were present in two of eight dogs with three
hour occlusions and in six of eight dogs with six hour
ternporary occlusion, sho%i three zones: a morphologically weflw
preserved ceniral core f C),, a hemorrhqg1c midibhe (Ikht siatning in occlusions.
this negatlw print) ^ and a pemhemi necrotic zone with t'htael Micro opj ttle hemorrhage or inflammatory
vusilukiturv (P.). Ae laiter zone showw extmtve phagocytosis of
necrotic cdh and eariy rqiair. nt's is a relatively krge three hour
imant but vtabk musek is present in a patehy dbtribution within
the outer vmU qf ihe sectibn.

granularity which are prominent features of necrosis es4.

pecially fbllowing temporary coronary occlusion.) Necrosis
4

was expressed as the overall percentage fbr the three

transtriural slices cut from each hearL Statist'ical corn.
parison between groups was done Using the Student's ncn.
paired r4est.

Restdts
Gross and Microscopic Fiestures 6f iscbemic ligury
70.nvorary Oceiiistiolu
Necrosis was observed 'in every dog fbilowing either tetn-i
porary or permanent coronary; occlusion, F flowing 40 mih
of temporary tircumflex occlusion, necrosis occurred
throughout the region supplied by .the circum ex artery but
typically was restricted to the subendocardial myo ard'
Longitudinal sections through the posterior papillary muscle
(which is in the approximate center of the rcgion supplied by
the circumflex artery) always demonstrated subendocardial
necrosis i(fig. 2). MicrosCo 4cally necrotic myocytes ShOWM
loss of nutlei and ftequently showed promment myofibr'illar
contraction bands (fig. 3)i By two days an acute inflamma"
tory response was present and by three to fbur days; .the
nrocess of 6haeocvtogis and rienair had beeun around the
 788 CIRCULATION VOL 56, No 5q NOVEMB3ER 1977

| |~~~~~~~~~~~cular
|S dam;age and resultant hemorr:hag aro0uid these cen:
i2| M _ 2 ~~~~~~~~~~tr4l
~~~~~~~~~~in}pai-red
ggE pale fiocl suggested1 t;hat mVicrovasc:ular perftsion was
and thle myocrardi um Nvithin these smal:l foci mlost
t | | | |--
g ct~~~~~~~dosely reEsembled 0myocEardium underg oing autoyi:35s 1:n the
||g ~~~~~~~~~~~~~absenc'*
| | of pert'ilsion.-8io Myo4cyte>s isho 0x Wely spate
"9

ss sStiailons indica:ting Myria;r:axatXia:Dluc

Q | s 8 B l 2 el~~~~~~~~w ch were very Nintly sataine but 8:til apipeare to bie::in-
; | |i g i l _ i1n the he;morthagitc miadz:one (fig8: 5b;) necrofil: myo
~~~~~~~~tacCw
gZ i
4| ~~~~~~cytes showewd nt(3flb-ri l'ar contractionlbands, increased
cytoplasmic tgran:ularity, and n:ucletar karyol:ysiis.
FiGuRt 4. 7)nsnuml sike throt4 ihe pwitenior pqpft:mw- in the4 s:ubepicardial mysicardi:um, necrosis wAas occa
CiEeobtained t" d@ fer a tihre hour ;temora occls4b ne Sionally conflu:ent bu:tt- pically 1island necrWotie myocytes
inenertwothiris f he 4fs necmii; bt tth;e outrthWqthe"U wereintersperse W:th slands oftVible eci ofe sen4d
(bottofm0 ptb m4Z0:sZ4 ytetnnt to small penetrating ariteSn's HemortbaS w0as A t@t
vta
:Honorhge s m mnent in ihe subeVndoaa regiton b tdos nt but te p;rocesses ofiS nflSaffm:fion a:nd repai'r haLd begsun,
etenint themidnWo<tiuStXAsmipakSctentmicoreipreset 4
:::: ::
jiibewan o h'rgt fth hade edfa. Vdb"ol at s:uggesting thaSt the vasSculaturle wasS pree:rver 'in this; zone.Ll
~~~~(
g thetpxfhheartrgt 0theitiu At0he nontoelded0

InN:rcts produced by permanent coronary o Wlusions w;ere

Downloaded from http://circ.ahajournals.org/ by guest on September 25, 2017

response w;as observed w.ithin th central pa;le aredas buSt th?ey simiElar In many re:spect2s to tose reul0tin em sl h:rS of
were always surrtounded by hetmorrhage aLnd an infltrate of Itempo0rary cortonary ofccluslion but ditred acording to their
segmetedi nuttrophiis (figt 5a). Necrtic arterioles ob- age at sacrifice.0 At bot;h 24 (fig. 60) ahd 96 ;(fig. 7) hour0s; t;hree
structed by swoellenl endothelial cells Were ft^equent. Thie vas distinct zonles coul be: identified, The Sube: ocardi:al myo-lO
 WAVEFRONT OF MYOCARDIAL
FIGURE 6A Dwwmurdl slice ihmWh ih- e posrehbr pqpilkiry mta.
ek jivm dbg sacrificed 4fter a 24 hour pomanetit 14ation q/* the
a
etnumflex anerv,, Necro! 4w muvck iy not as cleat.y deniareated
gr=4 iivm vta bk mwck a! iht's time. Rowver &is tologic
anabdis sho" that necmsts iv pmeni throooout the trunsmuml
4
seciibn eweptjbr a thin sub.pkardiW Hm of vidble niusek .arwhw).
ne iql4ret Los chamctert'zed Oy a sti6endocardial cenlwl core, a cenw
tmi hemomhogic zone and a per#theral zone.
Downloaded from http://circ.ahajournals.org/ by guest on September 25, 2017
cardium contained a central zohe whiieh Showed wucturally
relatively well preserved myocvtts Without hemorrhage 6r
inflammation, By 96 hours; cells 1'n this central zone were
still relatively well preserved although they showed nuclear
and glassy cvtoplasm (fig. 84
A hemorrhagic midzone was invariably present at 24
hours but hemorrhage was relativdly mild, By 96 hours of
permanent coronary occlusion (fig., 7), hemorthage was
prominent and appeared equivalent to hemorr age in in.
t
Lrcts produced by 3 or 6 hours of temporary coronary
Occlusion with 96 hours of reperfasion.
The peripheral zone at 24 hours was characterized by
coagulation necrosis and act.te inflammation. Myocytes
showed nuclear pallbr oi karyolysts and glassy hyper.
eosinophilic cytoolasm. Coittraction bands, whicb were abw
sent ftom subendocardial areas of necrosis; were present in a
few scattered cells on the periphery of the area of necrosis..
However, they were much less ftequent fbilowing permanent
than fbilowing temporary coronary occlusion. By 96 hours,
pbagoeytosis of dead cells and replacement by grantilation
tissue had bqun on the periphery of the areas of necrosis
(fig, 8b)4
NECROSISIReimer, Lowe, Rasmiusen, Jennings 789
Traosinural Progression of ischemic Cell Deaffi
Summartzing the morphologic observations, m oca ial
infArcts showed three distinct zones of necrosis I
devdloped uentially wi"th increasing durAtion of coronary
occlustom These were 1) a central zone 'in whl&h hemorrhage
and ammation were absento 2) an hemorrhagic midzone&
and 3) a peri eral nonhemorrhagic zone in which inflam-
mation, hagocytosis and infhret repair occurred rtiat' 1*
quickly,
These zones and the transmural progression of ischernic
imury w'joth respect to dur a t on of occlus'ton are summarized
I
diagrammatically in figure 9. Irreversi le hijury of ruyo.
cytes was typically subendoiardial at 40 min whereas sub.
epicardial myocardium was presaved by re-establishing
blood flow at this time. As the ischernic period was
prolonged, irreversi'ble im'ury moved progress'ively toward
J
the subepicardial mvocardiurn and b 24 hours necrosis
d y
tvplcaliy was transmaral. Microvascular injury as
evidenced by hemorr age att repa t ol
y 0 ic
areas, also progressed ftom subendocardial to subepicardial
myocardium. However, the time scale of mj'icrovascular in.
jury appeared to he slower since hemorthage did not extend
out to the edge of necrotic areas. Rather, there was always a
peripheral zone of myo yte necrosis in the presence of an ap.
*
a
parently intact
vasculature.
Infard Size vss Duradon of OWuw'on
The -transmural extent of necrosis, calculated as a percent
of transmurat sections througb the posterior papillary mus-
cle i is summarized in table I and figiure 10. The progression
of irreversible irv-ury from subendocardium to 8tibepicar-
dium with increasing duration of 6CCWsion describei above
restilted in more necrosis at three
siginificantly (p A. 0.05)
hours (57 .b 7%) compared with 40 nim (38 4%) and at 24
hours (85 .-i 3%) compared with three hours (P .< 0,01).
Mean mf4rct size at six hours (71 .i 7%) was intermediate
between the three hour and 24 hour means but did not di&r
significantly ftom tither.
The average rdlative proport'ion of the transmura myo.
cardium at risk but still salvageable after various durations
of coronary occlusion can be estimated (fig. I as follows:
t1)
Since permanent ligation resulted in necrosis of 85
myocardium in the transmurtil sections studi min
of TCO cau ed 38% rosis; 38/85 X 100 45% of the =
m y ocardium 4t iisk was already irreversibly irij*ured by 40
min, Nevertheless; an average of 55% of the myocardium at
risk was reversibly inJured at 40 min and wa salvaged bv
reperffision. By three boars, about 100 (57/85 X I -
33% of myocardium at risk was still potentially salvageable
and bv six bours only 16% remained salvagi Able.
Hemodynamicsl Eleiitroc"ographic Oanges and Survival
Acute occlusion of 'the circumfl" coronary arterv in the
open chest dog, while producing a large area of ischernia,
caused little acute chahee in heart rate and a mild reduct'ion
 790 CI RCULATION VOL 56, No 5, NOVEMBER 1977

> st ,Jt .;
Downloaded from http://circ.ahajournals.org/ by guest on September 25, 2017

"' -P'v:sS s& #gr * + 6 t

', 4 ..,

FIGUJRE 8. Four day permanent infarct. a) Central core: Myocytes are relatively well preserved but show pale or absent
nuclei and pale cytoplasm with loss of sarcomere cross striations. Capillaries appear obstructed by swollen endothelial
cells. There is a prominent neutrophilic infiltrate around the edge of this core. H & E X 325. b) Peripheral edge: Necrotic
myocytes have been removedfrom the edge of the infarct and have been replaced by capillaries andfibroblasts which have
begun to grow into the infarct from surrounding viable muscle. Viable muscle is present in the subepicardial myocardium
around a small penetrating artery (low left corner).

tricular contractions often were frequent. Between 30 and fused but occurred in 10 of 29 dogs that acutely survived per-
120 min, arrhythmias temporarily subsided, but with manent coronary ligation. These late deaths were not
occlusions continued beyond two hours, intermittent ven- monitored but most likely resulted from ventricular fibrilla-
tricular tachycardia usually developed. Reperfusion after 40 tion. The overall survival was 56% and did not differ
min of coronary occlusion was associated with a high in- statistically between groups.
cidence of ventricular fibrillation (41 % of dogs surviving
occlusion). However, six of 14 of these dogs were
resuscitated by direct current countershock. In contrast to
the 40 min period of ischemia, reperfusion after three or six
hours of occlusion did not precipitate VF. Deaths occurred
during the first few minutes after occlusion in all groups, and
at reflow in the 40 min group. Deaths subsequent to the
acute experiment were infrequent in dogs that were reper-
40 Minutes

FIc;URE 9. Diagrammatic summary of the progression of the

wavefront of ischemic cell death with respect to duration of cor-
onary occlusion. Necrosis (all shaded areas) occurs first in the sub-
endocardial myocardium and with longer durations of coronary
occlusion involves progressively more ofthe transmural thickness of
the ischemic zone. (Dashed line indicates the anatomic boundary
between ischemic circumflex and nonischemic LA D coronary beds.) 3 Hours
Microvascular injury, evidenced by interstitial hemorrhage
(horizontal cross hatching), also progresses from subendocardial to
subepicardial zones but the time scale is slower than that for
L~~
myocyte necrosis. Complete cessation of microvascular perfusion
may result in a central core (dotted areas) of necrotic muscle devoid
of either hemorrhage or inflammatory response. Thus following
either temporary or permanent coronary occlusions, the infarct may
demonstrate, grossly and by light microscopy, a central core of rela-
tively preserved necrotic muscle, a hemorrhagic midzone, and a
peripheral zone of organizing necrosis. At three hours after occlu- -~~----- ~ XJ5 24 Hours
sion there is typically a significant amount of subepicardial myo-
cardium which is viable and can be salvaged by reperfusion.
 WAVEFRONT OF MYOCARDIAL NECROSIS/Reimer, Lowe, Rasmussen, Jennings 791

TABLE 1. Percent Necrosis in the Posterior Papillary Muscle* wavefront which gradually moves toward the subepicardium
Group I Group 2 Group 3 Grour 4 Group 5 during a period of six hours (figs. 9-11). Previous studies
(40 min (3 hr (6 hr (24 Br (96 hr have shown that cell death does not develop immediately or
TCO) TCO) TCO) PCO) PCO)
14 49 83 96 75
simultaneously even in areas of severe ischemia.'0 Cir-
42 56 44 91 70 cumflex occlusion for 15 min is completely reversible in that
22 64 47 71 90 reperfusion completely prevents necrosis, but occasional
43 46 96 83 small foci of necrosis result from 20 min occlusions.'0 Reper-
41 18 67 72 fusion after 40 min of circumflex occlusion always results in
47 74 58 87 focal or confluent subendocardial necrosis. Three hours of
81 75 93 78
6 77 79 83 occlusion produces, on the average, significantly more
32 72 necrosis which is typically confluent in the subendocardial
52 92 myocardium with focal involvement of the mid- or sub-
26 92 epicardial myocardium. By 6 to 24 hours, necrosis becomes
27 98
35 71 nearly transmural. Residual viable myocardium in these sec-
62 83 tions is observed primarily in subepicardial foci, which are
36 93 often adjacent to blood vessels, and at the base of the
21 99 papillary muscle near the apex of the heart, i.e., at the junc-
23 tion of the beds perfused by the circumflex and the anterior
31
59 descending coronary arteries. A transmural progression of
Downloaded from http://circ.ahajournals.org/ by guest on September 25, 2017

38 cell death has previously been suggested by histochemical

59 and electrophysiologic techniques. Cox has shown irrevers-
27 ible loss of dehydrogenase enzyme activity2l and of elec-
44 tromotive force22 gradually progressing from subendocardial
Mean 38 57 71 85 79 to subepicardial myocardium following coronary occlusion.
SE 4 = 7 7 = 2 6 The mechanism of this subendocardial to subepicardial
Number of dogs 23 8 8 16 3 progression of injury probably is related primarily to the
P values (based on Student's nonpaired t-test) transmural distribution of coronary collateral flow. In dogs,
40 min vs 3 hr: <0.05 collateral flow occurs largely via epicardial anastomoses and
40 min vs 6 hr: <0.001 reaches the myocardium via penetrating muscular arteries
40 min vs 24 hr: <0.001 which are functionally end arteries. During systole, in-
3 hr vs 6 hr: NS
3hrvs24hr: <0.01 tramyocardial tension is greatest in,2' and limits perfusion of
6 hr vs 24 hr: NS the subendocardial myocardium.'4 Increased subendocardial
*Overall percent necrosis in three transmural sections of the PP (see flow must occur during diastole in order to attain the uni-
methods and fig. 1).
TCO = temporary coronary occlusion; PCO = permanent coronary
form transmural distribution of flow normally observed in
artery occlusion. nonischemic myocardium20' 25. 2B This autoregulatory mech-
anism cannot compensate for total coronary artery occlu-
Discussion sion and a transmural gradient of collateral flow develops.
Flow in the central subendocardial zone is reduced to less
This study demonstrates that in dogs with circumflex cor- than 15% of flow in nonischemic areas (severe ischemia). In
onary artery occlusions, irreversible ischemic injury occurs the mid- and subepicardial myocardium, flow may be up to
first in the subendocardial myocardium and extends as a 15-30% of normal (moderate ischemia).'7
In addition, recent studies by Dunn and Griggs2' suggest
that, in vivo, there may be a transmural gradient of meta-
bolic rate with the greatest energy utilization occurring in
-I-- the subendocardial myocardium. Thus transmural differ-

100

80

n n 60-
16 3

40 mins. 3 hrs. 6 hrs. 24 h rs. 96 hrs.

40X
DURATION OF OCCLUSION
20
FIGURE 10. Transmural necrosis, calculated as described in the
text and legend to figure 1, is plotted with respect to duration of
occlusion. Brackets indicate the SEM. Infarct size increased 0 3 6 24
HOURS
progressively with time. Reperfusion at 40 min and 3 hours resulted
in significantly less necrosis compared with 24 hour infarcts. Six FIGURE 11. Proportion of ischemic muscle which is viable and
hour infarcts were intermediate between but not significantly potentially salvageable as a function of time after coronary occlu-
different from either three or 24 hour infarcts. sion. Data are plotted as a percent of 24 hour infarct size.
 792 CIRCULATION
ences in metabolic demand may also contribute to the
wavefront phenomenon of cell death.
Effects of Reperfusion
Reperfusion of the area of ischemia had one or more of
several effects in the present study which can be summarized
as follows:
1) Reperfusion preserved viable myocytes.
2) Reperfusion, where not impeded by vascular damage
(peripheral zone), accelerated the disruption of irreversibly
injured myocytes and permitted the process of inflamma-
tion, phagocytosis and infarct repair to begin quickly.
3) Reperfusion permitted interstitial hemorrhage from
vessels which had been injured by ischemia but were still per-
fusable at the time of reflow.
4) Reperfusion had no effect on areas which apparently
had become nonperfusable (central pale zone).
Necrosis following reperfusion of irreversibly injured
myocardium has been extensively studied at both the light
Downloaded from http://circ.ahajournals.org/ by guest on September 25, 2017
and electron microscopic level. Reperfusion after 40 min of
coronary occlusion results in the rapid development of
prominent myofibrillar contraction bands (fig. 3) and tissue
edema (by light microscopy).29 The ultrastructural features
of accelerated necrosis following reperfusion of irreversibly
injured ischemic myocardium include explosive cell swell-
ing with formation of large subsarcolemmal blebs, breaks in
the sarcolemma, and the development of myofibrillar con-
traction bands and mitochondrial calcification.30 32
Interstitial hemorrhage was prominent after reperfusion
of 3 and 6 hour occlusions. It seems likely that this
hemorrhage reflects ischemic injury of the vasculature and
that hemorrhage may have contributed to increased in-
terstitial pressure resulting in production of the no reflow
phenomenon in the central core33 and of the decreased com-
pliance and the further deterioration of local contractile
function which has been observed by Lang et al.34 and
others. It is unlikely, however, that vascular damage and the
resultant hemorrhage following reperfusion, caused infarct
extension"' because a) reperfusion resulted in a smaller
average transmural extent of necrosis at 3 hours and in no
change at 6 hours compared with permanent infarcts, and b)
the transmural progression of myocyte necrosis appeared to
precede the development of vascular damage such that a
midzone of hemorrhage was always surrounded by a
peripheral zone of necrosis with an intact vasculature.
The relative morphologic preservation of the central core
in permanent occlusions has been emphasized by Bishop et
al.,35 who implied that myocytes in this core might still be
viable. The latter hypothesis is untenable in light of the
severe early ultrastructural damage which has been demon-
strated in this central zone.36' 37 Rather, the slow progres-
sion of histopathologic changes in the central zone is
equivalent to the changes associated with postmortem
autolysis in the complete absence of blood perfusion.38' 3
The presence of hemorrhage and vascular obstruction
around these central core areas suggest that little or no per-
fusion is penetrating such foci.
Estimation of Salvageable Myocardium with Respect to
Duration of Ischemia
Although much research effort has been devoted to the
concept of limiting infarct size by pharmacologic or sur-
VOL 56, No 5, NOVEMBER 1977
gical means, the rate of progression of ischemic injury and
the size of the ischemic but still viable "border zone" with
respect to the duration of ischemia has received relatively lit-
tle attention in experimental animals. Also, at the present
time, it is impossible to estimate, during life, the amount of
salvageable myocardium following onset of myocardial in-
farction in man.
In the present study, necrosis within transmural sections
of the posterior papillary muscle was used as a quantitative
estimate of the transmural extent of cell death. Although
total left ventricular infarct size was not measured in this
study, the transmural progression of injury described oc-
curred throughout the circumflex bed. Necrosis within
transmural sections of the posterior papillary muscle has
been shown to be a reliable index of overall left ventricular
necrosis in this model (fig. 12).
The present study re-emphasizes the fact that cell death
occurs early in the subendocardial zone of severe ischemia
and that about half of the ischemic myocardium that is
necrotic at 24 hours has already died by 40 min. However,
cell death usually occurs more slowly in the mid- and subepi-
cardial myocardium such that on the average, about a third
of the ischemic myocardium at risk is still salvageable at
three hours. By six hours, there is little salvageable myocar-
dium in most dogs.
These conclusions are in general agreement with other
reperfusion studies. Estimated or anatomically measured in-
farct size has been reduced by reperfusion after 30 min to
three hours3 10, 12-18 but not after four or more hours of tem-
porary coronary occlusion.14 17 Experimental reduction of
ST segment elevation or CPK depletion has been observed
with a variety of pharmacologic or mechanical inter-
ventions begun 30 min or more after coronary occlusion40-44
and some salvage of ischemic myocardium has been
documented anatomically with propranolol therapy in-
stituted three hours after circumflex occlusion in our
model.45 Functional studies have shown that reperfusion as
early as one hour after coronary occlusion does not result in
immediate hemodynamic improvement or in reversal of con-
tractile abnormalities.16' 46 However, Constantinil6 has
shown that reperfusion after three hours of coronary occlu-
sion, while providing no immediate improvement in cardiac
75
50.
PP%
Necrosis
25
0
1lo 20 30
LV % Necrosis
FIGURE 12. Necrosis as a percent of the myocardium in three
transm ural sections of the posterior papillary muscle was calculated
from 17 dogs as described in the Materials and Methods section.
Necrosis as a percent of the left ventricle was calculated from
histologic sections of serial slices cut at 2-3 mm intervals through
the entire infarcts of the same 17 dogs. Each point represents one
dog. With two exceptions, transmural papillary muscle necrosis
provided a reliable index of total left ventricular necrosis. (From
previously unpublished data of K.A. Reimer & R.B. Jennings).
 WAVEFRONT OF MYOCARDIAL NECROSIS/Reimer, Lowe, Rasmussen, Jennings
function, decreased mortality and improved cardiac func-
tion measured seven days later. Bresnahan et al.19 observed
increased CPK release following reperfusion and concluded
that reperfusion causes infarct extension which they related
to the development of hemorrhage. It has subsequently been
shown, however, that reperfusion facilitates washout of
CPK from ischemic areas and results in overestimation of
true infarct size.'8
Conclusions and Clinical Implications
In conclusion, this study demonstrates that irreversible in-
jury of ischemic myocardium develops as a transmural
wavefront which begins in the subendocardial myocardium
and moves progressively toward the subepicardial myo-
cardium (fig. 9). In dogs with circumflex occlusions there is a
subendocardial zone of severe ischemia which dies relatively
quickly. Subepicardial myocardium, which is moderately
ischemic, dies more slowly and survives if the ischemic insult
is terminated by reperfusion. The timing of the transmural
Downloaded from http://circ.ahajournals.org/ by guest on September 25, 2017
progression of cell death (figs. 10, 11) provides a basis for
evaluating pharmacologic interventions in this model.
A similar wavefront of cell death may occur in man. If we
assume that the ischemic cells do not die simultaneously,
and that a similar transmural wavefront of necrosis develops
in man, the rate of progression of cell death in man may
proceed at a rate different from that we have observed in the
open chest anesthetized dog. The transmural progression of
injury in man may be influenced by the fact that the chest is
closed and the patient is conscious. Several important ad-
ditional variables in man include a) degree of coronary
stenosis or occlusion, b) anatomic size of the ischemic bed,
c) size and number of pre-existing collateral vessels, and d)
presence or absence of hemodynamic complications. In
man, as in the dog, it seems likely that severe ischemia
results in relatively rapid cell death. Such severely ischemic
cells would be difficult to salvage both because of the short
time available and because blood flow would be inadequate
to sustain delivery of therapeutic doses of drugs. On the
other hand, areas of moderate ischemia die relatively slowly.
Such areas are potentially salvageable for up to 3-6 hours in
the dog and possibly for longer periods in man. Since the
amount of potentially salvageable myocardium pro-
gressively decreases as the period of ischemia is prolonged,
time is the most critical variable involved in delaying or
preventing ischemic myocardial cell death.
Acknowledgment
We are grateful to Nancy J. Kramer for surgical and technical assistance,
to Betty Schlitz for preparation of histologic sections and to William
Boyarsky for preparation of photographs.
References
1. Maroko PR, Kjekshus JK, Sobel BE, Watanabe T, Conell JW, Ross J,
Braunwald E: Factors influencing infarct size following experimental cor-
onary artery occlusions. Circulation 43: 67, 1971
2. Maroko PR, Braunwald E: Modification of myocardial infarction size
after coronary occlusion. Ann Intern Med 79: 720, 1973
3. Jennings RB, Reimer KA: Salvage of ischemic myocardium. Mod
Concepts Cardiovasc Dis 43: 125, 1974
4. Corday E: Interventions that might influence viability of ischemicjeopar-
dized myocardium. Introduction to Symposium on the Management of
Jeopardized Ischemic Myocardium. Am J Cardiol 37: 461, 1976
5. Scanlon PJ, Nemickas R, Tobin JR Jr, Anderson W, Montoya A,
Pifarre R: Myocardial revascularization during acute phase of myo-
cardial infarction. JAMA 218: 207, 1971
793
6. Smullens SN, Wiener L, Kasparian H, Brest AN, Bacharach B, Noble
PH, Templeton JY: Evaluation and surgical management of acute evolv-
ing myocardial infarction. J Thorac Cardiovasc Surg 64: 495, 1972
7. Ross J: Early revascularization after coronary occlusion. Circulation 50:
1061, 1974
8. Loop FD, Cheanvechai C, Sheldon WC, Taylor PC, Effler DB: Early
myocardial revascularization during acute myocardial infarction. Chest
66: 478, 1974
9. Bolooki H, Kotler MD, Lottenberg L, Dresnick S, Anrews RC, Kipnis S,
Ellis RM: Myocardial revascularization after acute infarction. Am J
Cardiol 36: 395, 1975
10. Jennings RB, Sommers HM, Smyth GA, Flack HA, Linn H: Myocardial
necrosis induced by temporary occlusion of a coronary artery in the dog.
Arch Pathol 70: 68, 1960
11. Reimer RA, Rasmussen MM, Jennings RB: Reduction by propranolol of
myocardial necrosis following temporary coronary artery occlusion in
dogs. Circ Res 33: 353, 1973
12. Maroko PR, Libby P, Ginks WR, Bloor CM, Shell WE, Sobel BE, Ross
J: Coronary artery reperfusion. I. Early effects on local myocardial func-
tion and the extent of myocardial necrosis. J Clin Invest 51: 2710, 1972
13. Ginks WR, Sybers HD, Maroko PR, Conell JW, Sobel BE, Ross J:
Coronary artery reperfusion. II. Reduction of myocardial infarct size at 1
week after the coronary occlusion. J Clin Invest 51: 2717, 1972
14. Symes JF, Arnold IMF, Blundell PE: Early revascularization of the
acute myocardial infarction: The critical time factor. Canad J Surg 16:
275, 1973
15. Ginks W, Ross J, Sybers H: Prevention of gross myocardial infarction in
the canine heart. Arch Pathol 97: 380, 1974
16. Constantini C, Corday E, Lang T-W, Meerbaum S, Brasch J, Kaplan L,
Rubins S, Gold H, Osher J: Revascularization after 3 hours of coronary
arterial occlusion: Effects on regional cardiac metabolic function and in-
farct size. Am J Cardiol 36: 368, 1975
17. Mathur VS, Guinn GA, Burris WH III: Maximal revascularization
(reperfusion) in intact conscious dogs after 2 to 5 hours of coronary occlu-
sion. Am J Cardiol 36: 252, 1975
18. Jarmakani JM, Limbird L, Graham TC, Marks RA: Effect of reperfu-
sion on myocardial infarct, and the accuracy of estimating infarct size
from serum creatine phosphokinase in the dog. Cardiovasc Res 10: 245,
1976
19. Bresnahan GF, Roberts R, Shell WE, Ross J Jr, Sobel BE: Deleterious
effects due to hemorrhage after myocardial reperfusion. Am J Cardiol
33: 82, 1974
20. Kloner RA, Reimer KA, Jennings RB: Distribution of collateral flow in
acute myocardial ischemic injury - Effect of Propranolol Therapy. Car-
diovasc Res 10: 81, 1976
21. Cox JL, McLaughlin VW, Flowers NC, Horan LG: The ischemic zone
surrounding acute myocardial infarction. Its morphology as detected by
dehydrogenase staining. Am Heart J 76: 650, 1968
22. Cox JL, Daniel TM, Boineau JP: The electrophysiologic time-course of
acute myocardial ischemia and the effects of early coronary artery reper-
fusion. Circulation 48: 971, 1973
23. Kirk ES, Honig CR: An experimental and theoretical analysis of myo-
cardial tissue pressure. Am J Physiol 207: 361, 1964
24. Downey JM, Downey HF, Kirk ES: Effects of myocardial strains on cor-
onary blood flow. Circ Res 34: 286, 1974
25. Becker LC, Ferreira R, Thomas M: Mapping of left ventricular blood
flow with radioactive microspheres in experimental coronary artery
occlusion. Cardiovasc Res 7: 391, 1973
26. Wiusten B, Flameng W, Schaper W, Carl M: The distribution of myo-
cardial flow. Part I: Effects of experimental coronary occlusion. Basic
Res Cardiol 69: 422, 1974
27. Jennings RB, Ganote CE, Reimer KA: Ischemic tissue injury. Am J
Pathol 81: 179, 1975
28. Dunn RB, Griggs DM Jr: Transmural gradients in ventricular tissue
metabolites produced by stopping coronary blood flow in the dog. Circ
Res 37: 438, 1975
29. Sommers HM, Jennings RB: Experimental acute myocardial infarction.
Histologic and histochemical studies of early myocardial infarcts induced
by temporary or permanent occlusion of a coronary artery. Lab Invest
13: 1491, 1964
30. Herdson PB, Sommers HM, Jennings RB: A comparative study of the
fine structure of normal and ischemic dog myocardium with special
reference to early changes following temporary occlusion of a coronary
artery. Am J Pathol 46: 369, 1965
31. Jennings TB, Ganote CE, Kloner RA, Whalen DA, Hamilton DG: Ex-
plosive cell swelling of myocardial cells irreversibly injured by transient
ischemia. In Recent Advances in Studies on Cardiac Structure and
Metabolism, vol. 6, edited by Fleckenstein A, Rona G. Baltimore,
University Park Press, 1975, p 405
32. Kloner RA, Ganote CE, Whalen D, Jennings RB: Effect of a transient
period of ischemia on myocardial cells: II. Fine structure during the first
few minutes of reflow. Am J Pathol 74: 399, 1974
33. Kloner RA, Ganote CE, Jennings RB: The "no-reflow" phenomenon
after temporary coronary occlusion in the dog. J Clin Invest 54: 1496,
1974
34. Lang T-W, Corday E, Gold H, Meerbaum S, Rubins S, Constantini C,
 794 CIRCULATION
Hirose S, Osher J, Rosen V: Consequences of reperfusion after coronary
occlusion. Effects on hemodynamic and regional myocardial metabolic
function. Am J Cardiol 33: 69, 1974
35. Bishop SP, White FC, Bloor CM: Regional myocardial blood flow dur-
ing acute myocardial infarction in the conscious dog. Circ Res 38: 429,
1976
36. Jennings RB, Baum JH, Herdson PB: Fine structural changes in myo-
cardial ischemic injury. Arch Pathol 79: 135, 1965
37. Jennings RB, Ganote CE: Structural changes in myocardium during
acute ischemia. Circ Res 34 & 35 (suppl III): III-156, 1974
38. Yokoyama HO, Jennings RB, Clabaugh GF: Histochemical studies of
early experimental myocardial infarction. Arch Pathol 59: 347, 1955
39. Herdson PB, Kaltenbach JP, Jennings RB: Fine structural and bio-
chemical changes in dog myocardium during autolysis. Am J Pathol 57:
539, 1969
40. Libby P, Maroko PR, Bloor CM, Sobel BE, Braunwald E: Reduction of
experimental myocardial infarct size by corticosteroid administration. J
Clin Invest 52: 599, 1973
Infarct Size Reduction by Propranolol
Downloaded from http://circ.ahajournals.org/ by guest on September 25, 2017
before and after Coronary Ligation in Dogs
MARGARET M. RASMUSSEN, M.D., PH.D., KEITH A. REIMER, M.D., PH.D.,
ROBERT A. KLONER, M.D., PH.D., AND ROBERT B. JENNINGS, M.D.
SUMMARY Coronary occlusion in the dog results in irreversible
myocardial cell injury which develops first in subendocardial areas of
severe ischemia and subsequently spreads into mid and subepicardial
areas of moderate ischemia. The effect of propranolol on this
progression of ischemic injury was evaluated. Three groups of dogs
were studied: 1) untreated, 2) treated with propranolol before and
throughout coronary ligation, and 3) treated with propranolol begin-
ning three hours after ligation. Dogs were sacrificed 24 hours after
coronary ligation and necrosis was quantitated from histologic sec-
MYOCARDIAL CELL DEATH following acute coronary
occlusion, develops first in severely ischemic subendocardial
myocardium, and progressively spreads to involve areas of
moderately ischemic mid- and subepicardial myo-
cardium. ' 2 Many studies have supported the idea that this
process can potentially be limited by early pharmacologic,
mechanical or surgical intervention.1
Previous studies have demonstrated that , blockade with
propranolol or practolol reduces epicardial or precordial ST
segment elevation following experimental coronary occlu-
sion or following myocardial infarction in man.8-10
Histologic studies of necrosis following temporary coronary
occlusions have provided direct demonstration that pro-
pranolol delays cell death in areas of severe ischemia.1l-l3
The present study was done to determine whether necrosis
within the peripheral, moderately ischemic zone of infarcts
produced by permanent coronary ligation in dogs could be
From the Departments of Pathology, Northwestern University Medical
School, Chicago, Illinois, and Duke University Medical Center, Durham,
North Carolina.
Supported in part by NIH contracts NOl-HV-52999 and 72-2984 and by
grants HL 18833 and HE08729.
Address for reprints: Keith A. Reimer, M.D., Department of Pathology,
Duke University Medical Center, Durham, North Carolina 27710.
Received February 24, 1977; revision accepted June 2, 1977.
VOL 56, No 5, NOVEMBER 1977
41. Maroko PR, Libby P, Bloor CM, Sobel BE, Braunwald E: Reduction by
hyaluronidase of myocardial necrosis following coronary artery occlu-
sion. Circulation 46: 430, 1972
42. Maroko PR, Libby P, Sobel BE, Bloor CM, Sybers HD, Shell WE,
Covell JW, Braunwald E: Effect of glucose-insulin-potassium infusion on
myocardial infarction following experimental coronary artery occlusion.
Circulation 45: 1160, 1972
43. Sugg WL, Webb WR, Ecker: Counterpulsation. Ann Thorac Surg 7:
310, 1969
44. Meerbaum S, Lang T-W, Osher JV, Hashimoto K, Lewis GW, Feldstein
C, Corday E: Diastolic retroperfusion of acutely ischemic myocardium.
Am J Cardiol 37: 588, 1976
45. Rasmussen MM, Reimer KA, Kloner RA, Jennings RB: Infarct size re-
duction by propranolol before and after coronary ligation in dogs. Cir-
culation 56: 794, 1977
46. Banka VS, Chadda KD, Helfant RH: Limitations of myocardial re-
vascularization in restoration of regional contraction abnormalities
produced by coronary occlusion. Am J Cardiol 34: 164, 1974
tions of transmural slices through the posterior papillary muscle.
Propranolol reduced infarct size by preventing necrosis in peripheral
(subepicardial) areas of moderately ischemic myocardium. Pre-
treatment with propranolol reduced necrosis from 85 3% (un-
treated) to 52 4% (P < 0.05). Delayed propranolol therapy was
about half as effective as pre-treatment and reduced necrosis to
71 3% (P < 0.05). Propranolol also limited microvascular injury
so that perfusion defects, detected with the dye thioflavin S, were
smaller in treated dogs.
limited by propranolol. Infarct size was significantly reduced
by propranolol therapy instituted prior to, and maintained
throughout coronary occlusion. Intervention with pro-
pranolol three hours after coronary ligation was less effec-
tive but still produced a significant reduction in infarct size.
Materials and Methods
Mongrel dogs of both sexes weighing 9-28 kg were
anesthetized with sodium pentobarbital (30 mg/kg), in-
tubated, and ventilated with a Harvard (Model 1063)
respirator. Aseptic surgical techniques were used. Lead II of
the standard electrocardiogram (ECG) and peripheral blood
pressure, via a catheter in the right femoral artery, attached
to a Statham (P23AC) pressure transducer, were monitored
on a Grass (Model 5) polygraph. The right saphenous vein
was catheterized for propranolol or saline infusion and both
catheters were kept open with dilute heparinized saline.
The chest was opened in the fourth left intercostal space,
the pericardium was incised, and the circumflex coronary
artery was isolated 1-2 cm from the aorta. Dogs in a
pretreated group (group B) were given 5 mg/kg propranolol
(intra-arterially) 10 min prior to coronary occlusion and an
additional 1.0 mg/kg/hr in heparinized saline was infused
intravenously at 0.7 cc/min beginning 30 min after occlusion
 The wavefront phenomenon of ischemic cell death. 1. Myocardial infarct size vs duration
of coronary occlusion in dogs.
K A Reimer, J E Lowe, M M Rasmussen and R B Jennings
Downloaded from http://circ.ahajournals.org/ by guest on September 25, 2017

Circulation. 1977;56:786-794
doi: 10.1161/01.CIR.56.5.786
Circulation is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright 1977 American Heart Association, Inc. All rights reserved.
Print ISSN: 0009-7322. Online ISSN: 1524-4539

The online version of this article, along with updated information and services, is located on
the World Wide Web at:
http://circ.ahajournals.org/content/56/5/786

Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally
published in Circulation can be obtained via RightsLink, a service of the Copyright Clearance Center, not the
Editorial Office. Once the online version of the published article for which permission is being requested is
located, click Request Permissions in the middle column of the Web page under Services. Further
information about this process is available in the Permissions and Rights Question and Answer document.

Reprints: Information about reprints can be found online at:

http://www.lww.com/reprints

Subscriptions: Information about subscribing to Circulation is online at:

http://circ.ahajournals.org//subscriptions/