Sequence of Events in Angina at Rest:
Primary Reduction in Coronary Flow
S. CHIERCHIA, M.D., C. BRUNELLI, M.D., I. SIMONETTI, M.D.,
M. LAZZARI, M.D., AND A. MASERI, M.D.
SUMMARY To investigate the events that lead to acute myocardial ischemia we monitored continuously the
ECG, the left ventricular (four patients) or aortic (two patients) pressure and the great cardiac vein oxygen
saturation (CSO2S) by a fiberoptic catheter in six patients with frequent anginal attacks at rest. We recorded
137 transient ischemic episodes (10 with chest pain) characterized by ST-segment elevation in 28 episodes,
depression in three episodes and by pseudonormalization of previously inverted or flat T waves in 106 episodes.
The onset of electrocardiographic and hemodynamic changes was preceded by a large drop in CSO2S in all 135
episodes with ST-T changes in the anterior leads but not in two episodes with ST elevation on inferior leads.
The fall in CSOS2, consistently followed by signs of left ventricular function impairment and never preceded by
any detectable increase in the hemodynamic determinants of myocardial oxygen consumption, probably
reflects a reduction in regional perfusion. Thus, a reduction in coronary flow may cause transient ischemia in
patients with angina at rest. These episodes may be associated with variable, often minor electrocardiographic
changes and occasionally with anginal pain.
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
onset of pain, relative to the changes in regional
myocardial perfusion in the initial phase of the attacks
of angina at rest.
ANGINA AT REST is traditionally ascribed to the
presence of critical atherosclerotic stenosis involving
large coronary arteries and limiting the regional coronary reserve to a very low, fixed level. This
hypothesis has been challenged on the basis of recent
observations:
1) Hemodynamic monitoring in patients with
angina at rest showed no increase in the hemodynamic
determinants of myocardial oxygen consumption
preceding the onset of ST-segment changes.'-' 2)
Thallium-20 1 myocardial scintigraphy obtained during anginal attacks showed a reversible deficit of the
tracer uptake, indicating a reduction in regional
myocardial blood flow.5 6 3) Angiographic studies
performed during anginal attacks consistently showed
a reversible coronary artery spasm.3 7-10
Although these findings suggest that a coronary
vasospasm may cause transient ischemic attacks that
occur at rest, evidence is lacking that a reduction in
regional blood supply actually precedes the onset of
electrocardiographic and hemodynamic changes
typical of ischemic episodes. Therefore, in a group of
patients with frequent angina that occurred at rest, we
continuously monitored the ECG, the left ventricular
or aortic pressure and the coronary sinus oxygen saturation (CSO2S), which was assumed to reflect changes in
myocardial blood flow, provided the arterial 02 content and the myocardial 02 consumption remain constant. The study was aimed at investigating the temporal sequence of electrocardiographic changes, of the
impairment of left ventricular function and of the
Materials and Methods
Patients
Six male patients ages 34-65 years (mean age 53
years) admitted to our coronary care unit because of
transient, recurrent episodes of angina at rest with
typical ST-T changes were selected for the study.
Their main physical and clinical characteristics and
electrocardiographic, angiographic and scintigraphic
data are listed in table 1. Patients 2 and 4 had an old
myocardial infarction, and patient 5 had mild rheumatic aortic incompetence. To document the location
and direction of ST-segment changes, 12-lead ECG
tracings were recorded in each patient during several
anginal attacks. Thallium-201 scintigrams performed
in four patients during an anginal attack at rest
showed, relative to the control conditions, a deficit in
the tracer uptake involving the anteroseptal wall in patients 1 and 4, the lateral wall in patient 3 and the inferior wall in patient 2. Coronary angiography, performed in five patients by Judkins' technique, showed
three-vessel disease in patient 3, two-vessel disease in
patients 1 and 4, and one-vessel disease in patients 2
and 5. Left ventriculography showed normal contractility in patient 5, anteroseptal hypokinesia in patients
1 and 3, apical hypokinesia in patient 4 and a small,
inferior, postnecrotic aneurysm in patient 2. Informed
consent for coronary angiography was not obtained
for patient 6. An ergonovine maleate (EM) injection
(0.02-0.6 mg i.v.) produced typical anginal pain, accompanied by ST-T changes similar to those observed
during spontaneous attacks in all patients. The EM injection repeated during angiography in patients 1-3
resulted in nonocclusive spasm of the left anterior
descending and right coronary arteries and the first
diagonal branch, and was relieved by sublingual
nitroglycerin. In every case ST-T changes comparable
From the Laboratorio di Fisiologia Clinica (CNR) and the
Istituto di Patologia Medica I, University of Pisa, Pisa, Italy.
Supported by the Italian National Research Council (CNR),
Biomedical Technology Project (TBM-Card 2).
Presented at the 50th Annual Scientific Sessions of the American
Heart Association, Miami, Florida, November 1977.
Address for correspondence: S. Chierchia, M.D., Laboratorio di
Fisiologia Clinica CNR, Via Savi, 8, 56100 Pisa, Italy.
Received June 18, 1979; revision accepted September 21, 1979.
Circulation 61, No. 4, 1980.
759
760
VOL 61, No 4, APRIL 1980
CIRCULATION
TABLE 1. Physical, Clinical, Electrocardiographic, Coronary Arteriographic and Myocardial Scintigraphic Characteristics of the
Patients
ECG
History
Exercise
Onset of
During spontaneous
Pt
Age
Exercise
Basal
attacks
tolerance
no.
symptoms
(years)
3 months
Moderate
3 mm STI
Incomplete RBBB
1
34
-STT Vi V5
and LA hemiblock
(rest and effort)
-ST1 V3-V6
V2-V6
2
3
60
58
4
48
5
55
13 years (rest, effort
and emotion)
3 years
(rest and effort)
Poor
3 months
(rest and effort)
7 months (rest)
Moderate
-TT V1-V4
STT II-III
Old anterior and
inferior MI
Normal
Poor
Old anteroseptal MI
Negative T V1-V4
Negative T V4-V5
Moderate
2 mm STI
aVF
STJ 2 mm I -
V4-V6
aVL-V2-V6
V2-V6
2 mm STI
-STT V1-V4
- TT V1-V4
STT V2-V4
-STT V2-V3
-TT V1-V5
-TUl V2-V5
STT V1-V4
Normal
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
Good
Normal
1 month (rest)
Normal
6
65
Abbreviations: LAD = left anterior descending artery; LCA - left circumflex artery; LMB = left marginal branch; LDB =
left diagonal branch; RCA = right coronary artery; RBBB = right bundle branch block; - not performed; t = elevation;
l = depression; TT = pseudonormalization of T wave.
to those accompanying the spontaneous attacks were
observed. A bicycle ergometer exercise test, per-
formed in all patients, was positive in patients 1, 2 and
3 at a load of over two-thirds of their theoretical maximal working capacity and was negative in the others.
Protocol
The study was usually initiated in the evening 12
hours after withdrawal of any antianginal treatment
and was continued for about 12 hours or until a
sufficient number of attacks (at least two) were
collected. The monitoring period ranged from 3-16
hours (mean 8.6 hours). During the study only sublingual nitroglycerin was given to relieve the attacks.
Each patient was premedicated with diazepam, 10 mg
i.m. Each patient gave informed consent to the study.
No patient developed serum enzyme elevation or permanent ECG changes indicative of acute myocardial
infarction before or during the study.
parison between the values obtained by the two
methods are shown in figure 1. A #5F Teflon catheter
was introduced percutaneously via the femoral route
and positioned under fluoroscopic control in the left
ventricle in four patients; in patients 5 and 6 an
arterial polyethylene cannula was inserted percutaneously into the left brachial artery. Left ventricular and arterial catheters were connected to a
Statham P23Db pressure transducer and continuously
flushed under constant pressure with heparinized
saline. The zero reference was positioned 5 cm below
the sternal angle. Coagulation time was checked every
3 hours and was maintained between 20-25 minutes.
Simultaneous arterial and coronary sinus blood
02 SAT. %/
100
Technique and Instrumentation
LUJ
CSO2S monitoring was performed using an in vivo
reflectometer (Schwarzer IVH3) that measures the
percentage of 02-saturated hemoglobin by applying
the reflection spectrophotometry principle.", 12 A #7F
woven Dacron fiberoptic catheter (Schwarzer) was introduced via a left antecubital vein, connected to the
reflectometer, positioned in the coronary sinus outflow
tract and advanced to the left cardiac margin under
fluoroscopic control. Its position was confirmed by the
appearance of a sudden and stable 02 saturation drop
and by the injection of Urographin-75. The calibration
factor was determined before introducing the catheter
by dipping the tip in different cuvettes that contained
blood samples with different percentages of 02 saturation, determined by an in vitro spectrophotometric
method (IL 182 Co-oxymeter). The results of the com-
LUJ
80
0
0
60
00
40
Lii
0-
20
U.)
20
40
80
60
100
O2
SAT. %
FIBRE OPTICS
FIGURE 1. Comparison between the values obtained by
saturation (sat) by the
simultaneous measurements of
fiberoptic and the spectrophotometric method.
0,
761
ANGINA AT REST/Chierchia et al.
TABLE 1. (Continued)
Myocardial scintigraphy
Location of cold areas during
ischemia at rest
Anteroseptal
Coronary arteriography
During angina
(Ergonovine)
Control
LAD
incomplete
90% LAD
90% LDB
Anterior and apical
hypokinesia
75% LCA
RCA incomplete
Inferior hypokinesia
Inferior
75% LAD
75% LDB
LDB incomplete spasm
Anterior hyopkinesia
Lateral
Apical hypokinesia
Anteroseptal
90% LMB
75% LAD
75% LDB
90% LAD
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
samples were obtained during six anginal episodes in
patients 1, 5 and 6 and analyzed for Po2, Pco2 and pH
(IL 213 blood gas analyzer) and for lactate and
potassium content. CSO2S pressure waves and the
ECG lead showing the most evident ST-T changes
during the attack were continuously recorded on a 14channel analog tape recorder (Philips Analog 14) at a
recording speed of 2.5 cm per second. Twelve-lead
ECG tracings were also recorded in each patient during the longer episodes.
Analysis of the Data
The data stored on magnetic tape were played back
Telco-Thompson EP86 photographic recorder at
different paper speeds (0.1-25 mm/sec). The left ventricular dP/dt was differentiated by an R-C analog
circuit. Low-speed playbacks (0.1-1 mm/sec) of all
recording periods were obtained according to the
previously described technique3 4, 1' and were used to
identify the ischemic episodes and to follow the trends
of the different measurements. High-speed recordings
(10-25 mm/sec) were obtained for each episode, including the period preceding, during and after the
onset of ST-segment changes.
CSO2S levels, heart rate, pressure-rate product
(PRP), left ventricular (or aortic) systolic/enddiastolic pressure and left ventricular peak positive
and negative dP/dt were measured during each cardiac cycle and averaged over 10-second periods 60, 30,
20 and 10 seconds before and 10, 20, 30 and 60 seconds after the onset of ST-segment changes. In 19
episodes additional measurements were obtained 5
and 10 minutes before the onset of ST-segment
changes. This was not possible in the remaining
episodes because they followed one another too
rapidly.
Results
on a
Electrocardiographic Findings
We recorded 28 episodes of ST-segment elevation
(STt) and three of ST-segment depression (ST;). Only
Ventriculography
Normal
eight (seven with STt and one with STI) were accompanied by typical anginal pain, which began 68 ± 7
seconds after the onset of the ECG changes. We also
recorded 106 episodes characterized by pseudonormalization of previously inverted T waves or by an increase in amplitude and peaking of positive or flat T
waves. Two of these were accompanied by typical
anginal pain that began 45 and 50 seconds, respectively, after the onset of the ECG changes.
Twelve-lead tracings were obtained in seven episodes with STt, in one with ST; and in 12 with changes
only in T waves. The most pronounced ischemic
changes always appeared in the lead previously
selected for continuous monitoring.
Patients 1, 4, 5 and 6 showed variable electrocardiographic patterns alternating on the same lead in
different episodes. Patient 3 had only episodes of ST;
and patient 2 only episodes of STt. The most evident
ischemic changes occurred on anterior leads in all but
patient 2, who had only two episodes with STt on inferior leads (table 2). All the attacks subsided either
spontaneously or after sublingual nitroglycerin.
TABLE 2. ST-T Changes During 137 Ischemic
Pt
no.
1
2
3
STI
STT
P
1
1
4
1
5
6
3
1
A
16
Episodes
TT
P
A
1
1
P
A
55
1
4
2
31
18
1
2
2
104
1
21
7
Abbreviations: P = painful episodes; A asymptomatic
episodes; STt = ST-segment elevation; STI = ST-segment
depression; Tt= pseudonormalization of T waves.
762
_,ae;&R+8=.\<a>(:,f-N2*9ge;y-sOa<:i *,+.-;z8^Ra<"t\@E>se9.i;m0 s$Sen:
CIRCULATION
b
a
b
a
c
b
a
W
ECG
ts
!~.t
50
CSO2s%v''
18
20
dP/dt
VOL 61, No 4, APRIL 1980
a
c
c
se
b
a
c
XAL\
L
on8
f}tS
-
'S:
,>s..+R.@,;\i >&g:S3t,,
Id
LVP
fA;
50;, u ,.,, W.j,.
EaH\
,
Ni5
,9__a
>,x,
..
_ i<R t0Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
6
*
9
FIGURE 2. Low-speed playback (paper speed 0.3 mm/sec) of ECG, coronary sinus 02 saturation (CSOS2),
left ventricular pressure (L VP) and dP/dt during three successi ve asymptomatic episodes recorded in patient
1 o ver a period of abo ut 15 minutes. At the top are electrocardiographic patterns (lead V2) in resting conditions (a), at the onset (b) and at the peak (c) of the ischemic episode. Vertical lines correspond to the onset
of the ST-T changes. A sharp drop of CSO2S consistently precedes the onset of ECG and hemodynamic
changes.
Hemodynamic Findings
aortic pressure and pressure-rate product). However,
during the ischemic episodes different hemodynamic
patterns were observed in the individual patients.
The transient ischemic changes of the ST segment
and T waves, with or without pain, were never
preceded by a significant increase in the hemodynamic
determinants of myocardial oxygen consumption
(heart rate, left ventricular pressures and peak dP/dt,
1l
ECU
dJ
dldldbJJ
lJle
lJ
In patients 1 and 2, a consistent decrease in left ventricular relaxation and contraction peak dP/dt was
observed preceding or during the onset of ST-T
changes in 69 of the 74 recorded episodes (figs. 2, 3
P4
l
lJ
4 d4J
i,!lJll^
JlJJ2JJJ
\jJ\;/I
1-11, [-- f- [-
V2-
...
.11
50%/o
CORt SIN.
02SAT.
20%
LV
dP/dTTi
+
I Illi
100
? 1
1,
f{{ {{//{{0
{{{i#dGH l{l¢
LVP
FIGURE 3. High-speed playback (paper speed 10 mm/sec) of the "transient in" phase of an ischemic
episode characterized by peaking of T waves, recorded in patient 1. Arrows indicate the onset of change for
each recorded parameter. A drop in CSO2S precedes the onset of ECG and hemodynamic changes.
ANGINA AT REST/Chierchia
et
al.
763
ECG
CpSOisn
oiio'9
CSd2S/d
.
40
.-
"
.wA., i
fm
*:IT '14
ST
iv,1-
7-
.T-
7
x
dP/dt
100;
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
LVP
ts
o ssI
1^kio.ll
v6
f4
t'
"?
1 2 m m
6
IS
^'
9
f
*
-
X
if
S
K
%
S
?
T ifj l 2l
!]
12 'min
FIGURE 4. Low-speed playback of one of the ischemic episodes with ST-segment elevation on inferior
leads recorded in patient 2. No changes in coronary sinus 02 saturation are evident despite the presence of
marked ECG and hemodynamic changes. Abbreviations as in figure 2.
and 4). Nineteen were characterized by ST-segment
elevation and 50 by pseudonormalization of inverted
or flat T waves. In all the episodes with ST-segment
elevation and in 44 with pseudonormalization of T
waves the drop in contractility was shortly followed or
accompanied by a decrease in left ventricular systolic
pressure and by an increase in end-diastolic pressure.
Despite a drop in left ventricular relaxation and contraction peak dP/dt, no significant changes in left ventricular pressures were observed in six episodes with
pseudonormalization of T waves. The percentage of
decrease in left ventricular relaxation and contraction
peak dP/dt relative to the control values was 17.4 3
and 14 ± 2%, respectively, in the episodes with STt
and 11.6
4.2% and 9.2 + 3.7% in those with
pseudonormalization of T waves. The left ventricular
systolic pressure decreased 12 + 3 mm Hg in the
episodes with ST-segment elevation and 7.6 2.8 mm
Hg in those with pseudonormalization of T waves.
The left ventricular end-diastolic pressure increased
10.6 3 mm Hg in the episodes with STt and 5.5 ± 3
mm Hg in those with pseudonormalization of T
waves. Heart rate increased after the onset of ST-T
changes by an average of 10 4 beats/min in 19
3 beats/min in 40
episodes with STt and by 6
episodes with only changes in T waves (table 3).
TABLE 3. Hemodynamic Changes During the Ischemic Episodes
LVSP
LVEDP
LVC dP/dt
(mm Hg)
(mm Hg)
(%f)
+10.6 2.9
-13.9 i 2.2
A)
STT -12.0 3.0
+5.5 = 3.0
-9.2 =4 3.7
TT -7.6 2.8
+13.2 = 3.0
+12.8 f 3.4
B)
STT +17.6 =4 4.1
+12.5
+12.4
STh +22.5
(2 epis)
+8.7 d= 3.2
+7.0 E 3.4
TT +12.1 i 4.4
-
-
LVR dP/dt
HR
(%t)
(beats/min)
17.4 d= 3.1
-11.6 =4 4.2
+10.4 4.0
+6.0 i 2.9
-
+11.2
+9.3
-
3.1
+7.3 i 2.0
+6.0 d 1.9
+9.0
+5.0
2.5
AoSP
C)
STT -19.3 t 4.7
STI -13.4
+ 10.0 i 4.2
+13.0
(1 epis)
+8.0 + 2.9
TT -14.2 3.0
Abbreviations: LVSP left ventricular systolic pressure; LVEDP left ventricular end-diastolic presleft ventricular relaxation; HR - heart rate; AoSP
sure; LVC - left ventricular contraction; LVR
aortic systolic pressure; epis = episode.
=
764
CIRCULATION
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
In patients 3 and 4 no impairment of the left ventricular function was evident in the period preceding
the onset of ST-T changes. A parallel increase in left
ventricular systolic and end-diastolic pressures was
observed during or after the onset of electrocardiographic changes in 27 of the 38 ,recorded episodes.
Five were characterized by STt, two by ST; and 20 by
pseudonormalization of inverted T waves. The left
ventricular systolic and end-diastolic pressures increased 17.6 4 and 12.8 ± 3.4 mm Hg, respectively,
in the episodes with STt; 22.5 and 12.5 mm Hg in
those with ST;; 12.1 i 4.4 and 7 ± 3.4 mm Hg in
those with pseudonormalization of inverted T waves.
Twenty-four episodes, five with STt, two with ST; and
17 with only changes in the T wave, were accompanied
by a simultaneous increase in left ventricular contraction (11.4 ± 2.7%) and relaxation (9.3 ± 3.1 %) peak
dP/dt. In the remaining 11 episodes (all characterized
by pseudonormalization of inverted T waves) no
changes in left ventricular systolic pressure and dP/dt
appeared. End-diastolic pressure increased an average
of 6 ± 2.5 mm Hg. An increase in heart rate,
simultaneous to the onset of the ST-T changes, was
observed in five episodes with STt (6 ± 2 beats/min),
in two episodes with STw (9 beats/min) and in 16
episodes with only changes in T waves (5 ± 2.5 beats/
min) (table 3).
In patients 5 and 6, in whom only aortic pressure
was monitored, a drop in systolic pressure with reduction in pulse pressure accompanied the onset of STsegment changes in 22 of the 25 recorded episodes.
The average drop was 19.3 ± 4.7 in the four episodes
with STt, 13.4 in the episode with STI and 14.2 ± 3 in
VOL 61, No 4, APRIL 1980
the 17 episodes with pseudonormalization of T waves.
Heart rate increased after the onset of ST-T changes
by 10 ± 4 beats/min in the episodes with STt, by 13
beats/min in the episode with STI and by 8 ± 3
beats/min in the episodes with only changes in T
waves. No significant hemodynamic changes were
observed in three episodes, all characterized by peaking of T waves (table 3).
Results of CSO2S Monitoring
A consistent drop in CSO2S preceded the onset
of electrocardiographic and hemodynamic changes in
all the 135 episodes characterized by ST-segment or
T-wave changes occurring on the anterior leads,
regardless of the presence of typical anginal pain (figs.
2, 3 and 5). In contrast, no detectable changes in
CSO2S were observed in the episodes recorded in
patient 2, characterized by a marked STt on inferior
leads (fig. 4). In no instance did an appreciable increase of the hemodynamic determinants of myocardial 02 consumption precede or parallel the fall in
CSOS2 (fig. 5). In all episodes CSO2S began returning
to the control almost simultaneously with the normalization of the ST segment or T wave. In the 26
episodes with STt on anterior leads, CSO2S began
decreasing 27.2 ± 2 seconds before the onset of the
ST-segment changes and reached its minimum level,
48.5 ± 3% lower than the control values, 66 ± 9 seconds later. In the three episodes with ST;, the CSO2S
drop occurred 37 ± 2 seconds before the onset of the
electrocardiographic changes and attained its
minimum, 27.3 ± 2% lower than the control values, 95
± 6 seconds later. In the 106 episodes characterized by
ONSET ST-CHANGES
,Q
1200
|
0
PRP
900 1
6 00
45
*
=
+ =
COR SIN
02 SAT. %
40
35
FE. 2
F.L. 2
o
=
PA.
P.U.
*
=
C.G.
o =
23
E Pl SOD ES
..
.,
36
72
60
0
30
30
60 sec.
FIGURE 5. Behavior of coronary sinus 02 saturation (CSO2S) and pressure-rate product (PRP X 1-2) in the
five patients who showed ECG changes on anterior leads. Each point represents the mean of the values
measured in all the recorded episodes. Time 0 refers to the onset of ST-T changes. A drop in CSO2S
precedes the onset of ST-segment changes in the absence of any detectable changes in myocardial oxygen
consumption.
ANGINA AT REST/Chierchia et al.
only changes in the T-wave, CSO2S began decreasing
33.5 ± 1 seconds before the electrocardiographic
changes and reached its minimum, 31 ± 2.7% relative
to the control values, 26.6 ± 4 seconds later (fig. 6). In
episodes in which a left ventricular functional impairment was evident preceding the onset of the electrocardiographic changes, the CSO2S drop occurred
13 + 1.5 seconds before the fall in contractility in the
episodes with STt and 8 ± 1 seconds in those with only
changes in T wave. No significant differences in
CSO2S were evident between the episodes associated
with typical anginal pain and those that were asymptomatic.
Discussion
Electrocardiographic and Hemodynamic Findings
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
From a clinical, electrocardiographic and hemodynamic point of view our data confirm and extend
previously reported observations.
In our patients typical anginal pain was associated
with only 10 of 137 ischemic episodes and, if present,
always occurred 50-120 seconds after the onset of STT changes. Therefore, pain does not appear to be a
reliable and sensitive marker of transient, acute
myocardial ischemia3, 4, 14, 15
Four patients showed variable electrocardiographic patterns in the same leads in different
ischemic episodes. In particular, patients 1, 4 and 5
had either episodes with STt or with pseudonormalization of inverted or flat T waves, and patient 6
showed STt and STI in the same leads during two consecutive episodes. Since different ST-T changes,
probably reflecting variable degrees of ischemic involvement, may be observed in the same area at
different times, angina with STt does not appear to
765
identify a special subset of anginal patients,"6 but
rather represents only one of the possible clinical
manifestations of the anginal syndrome.3 4,15
In none of the 137 episodes recorded were the ST-T
changes preceded by consistent increases in the
hemodynamic determinants of myocardial oxygen
consumption. In fact, in patients 1 and 2 a marked
decrease in left ventricular peak dP/dt and systolic
pressure, followed by an increase in end-diastolic
pressure, preceded or accompanied the onset of ST-T
changes in 69 out of the 74 recorded episodes
characterized by either STt or by pseudonormalization of inverted T waves.3 4 Furthermore, in patients 5
and 6, in whom only aortic pressure was monitored, a
significant drop in systolic pressure with reduction in
pulse pressure, probably reflecting acute left ventricular functional impairment, accompanied the ST-T
changes in 22 of the 25 episodes. The report of a
similar hemodynamic and electrocardiographic sequence of events, observed in the experimental animal
upon acute coronary ligation,'1719 supports the
hypothesis that a transient reduction in regional blood
supply may have been the initiating event for the transient ischemic episodes recorded in these patients.
However, in 27 of the 38 episodes recorded in
patients 3 and 4, a simultaneous increase in left
ventricular systolic and end-diastolic pressures and
contraction and relaxation peak dP/dt consistently
accompanied or shortly followed the ST-T changes.
We could not find any significant differences between
these episodes and those recorded in the other
patients, concerning the duration, the ECG pattern,
the behavior of left ventricular end-diastolic pressure
or either the presence or absence of anginal pain.
Consequently, the mechanisms that underly the
hemodynamic patterns during the episodes remain un-
ONSET OF
ST-T CHANGES
CS02 saturation
controt level
100 %
FIGURE 6. Average behavior of coronary
sinus 02 saturation (CSO2S) in the three
electrocardiographic classes of ischemic
episodes. The first point, before the onset of
ST-T changes, represents the time at which
50 %0/
CSO,S began to decrease; the second, after
the onset of ST-T changes, represents the
time at which CSO2S reached the minimum
level. Values are mean ± SD.
-20
-40
*
=
o
=
o
=
0
20
40
60
80
100
ST elevation ( 26 episodes )
ST depression ( 3 episodes )
T wave changes ( 106 episodes )
Sec
CIRCULATION
766
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
does not increase as indicated by the absence of rise in
the hemodynamic parameters monitored throughout
the study. It appears reasonable to assume that the
drop in CSO2S, consistently observed in our patients
in the absence of any detectable increase in myocardial 02 consumption, reflects an increase in myocardial 02 extraction caused by reduced regional myocardial blood flow. An increase in the total coronary
resistance as observed during hyperventilation28 31
probably did not occur in our patients. Blood gas
analysis, performed by simultaneous arterial and coronary sinus sampling during four asymptomatic
ischemic episodes, demonstrated a marked metabolic
acidosis, accompanied by lactate and potassium
production in the coronary sinus blood without any
evidence for arterial alkalosis and hypocapnia (fig. 7).
In two additional episodes accompanied by typical
anginal pain, a significant increase in arterial pH and
Po2 and decrease in Pco2 occurred, accompanied by a
decrease in venous pH and Po2 and increase in Pco2.
Furthermore, in two other patients submitted to coronary sinus catheterization using the fiberoptic
catheter, prolonged hyperventilation failed to
precipitate typical anginal pain, ST changes or lactate
and potassium production despite the extremely low
levels of 02 saturation and Pco2 attained in the coronary sinus blood.
The advanced position of the catheter tip in the coronary sinus accounts for the sensitivity of the system
in detecting changes in CSO2S even in the presence of
minimal electrocardiographic changes and in the
absence of anginal pain. It also explains the absence of
significant change of the CSO2S in the episodes
recorded in patient 2, characterized by marked STt in
the inferior leads. This confirms that the CSO2S
changes are localized to the ischemic area rather than
being diffuse, as expected in hyperventilation. The fall
in CSO2S appears related to the severity of electrocar-
clear, although several experimental studies have
demonstrated that both depressor or pressor cardiocardiac reflexes may be evoked by acute myocardial ischemia.20 21 Indeed, in the same experimental
preparation, after an acute coronary ligation, both
pressor and depressor responses may occur.22 A
variable severity of ischemia and a variable reactivity
of the autonomic nervous system may therefore account for the different hemodynamic responses during
ischemia in our patients.
Although the degree of left ventricular functional
impairment was more pronounced in the episodes accompanied by STt, a qualitatively similar hemodynamic pattern was also observed during episodes
with pseudonormalization of negative or flat T waves.
Consequently, according to a hypothesis recently suggested,23' 24 even changes in T-wave polarity,
traditionally considered nonspecific, must be
regarded, at least in some patients with angina at rest,
as indicative of transient acute myocardial ischemia.
Finally, the temporal lag between the electrocardiographic changes and the onset of anginal pain,
when present, and the possibility that a hemodynamic
pressor response accompanies the ST-T changes, may
account for the different conclusions of previous investigators.25 27 In fact, the rise in heart rate and aortic
pressure observed by these authors during anginal
pain was interpreted as the cause of the ischemic attacks.
Results of CSO2S Monitoring
According to Fick's indirect principle, in the experimental conditions of our study the measurement of the
CSO2S should represent a semiquantitative estimate
of regional myocardial blood flow changes. In fact, the
arterial 02 saturation may be assumed as practically
constant and the myocardial oxygen consumption
C
A
C
C
A
VOL 61, No 4, APRIL 1980
A
100
mmHg
mEcq
i
50
6
L 4.5
5o
H
3
0
O
PCO2
P02
C
A
C
K
A
C
A
100
mEq
%4
L
50
0
02 sat.
Lc
pH
*--
ARTERY
o-o
CORONARY SINUS
FIGURE 7. Blood gas analysis, potassium
(K) and lactate (LC) levels in control condition (C) and during four asymptomatic
episodes (A) recorded in patients 1, 5 and 6.
Each point represents mean ± SD.
ANGINA AT REST/Chierchia et al.
767
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
diographic changes during the episodes. The CSO2S
reached the minimum level during the episodes with
STt, but decreased to a lesser extent during those
episodes with ST; or T-wave changes (fig. 6). Consequently, in our patients the severity of the ST-T
ischemic changes appears proportional to the extent of
the reduction in coronary flow.
The observation that in the same patient different
ECG patterns preceded by proportional decreases in
CSO2S may occur in the same leads during different
episodes further supports the hypothesis that the
variant form of angina represents only one aspect of
the pleiomorphic anginal syndrome. The finding of a
similar behavior of CSO2S in patient 3, who had only
episodes with STI, confirms the hypothesis7`15 that a
primary reduction in regional myocardial blood flow
is not peculiar to the variant form of angina.
Our study does not provide direct information
about the cause of the reduced coronary blood supply.
We could not find evidence for a primary role of transient platelet aggregation as suggested by recent experimental data reported by Folts et al.,32 which,
however, are different from those of Uchida et al.33 in
a similar experimental preparation. However, another
study performed on a group of patients with similar
clinical, angiographic and electrocardiographic
characteristics failed to demonstrate any effect of both
high- and low-dose aspirin in preventing anginal attacks occurring at rest (Chierchia S: unpublished
data).
In contrast, the positive response to the EM test obtained in all patients, the angiographic documentation
of a coronary artery spasm in three patients, and the
favorable response to nitrates in all patients support
the hypothesis that the ischemic episodes in these
patients are caused by an altered coronary
vasomotricity.
pathogenetic mechanism of angina at rest more frequent than excessive increase in myocardial demands.
This conclusion should be studied in a larger series of
Conclusions
Our data document the temporal sequence of events
in the initial phase of transient ischemic attacks occurring at rest. They support the hypothesis that a
reduction in regional myocardial perfusion, probably
caused by coronary vasoconstriction or vasospasm,
precedes the onset of electrocardiographic and
hemodynamic changes typical of transient, acute myocardial ischemia. They also demonstrate that even
minor electrocardiographic changes, such as
pseudonormalization of inverted or flat T waves, once
considered nonspecific for transient, acute myocardial
ischemia, are preceded by a reduction in regional
myocardial perfusion and are accompanied by
hemodynamic signs indicative of acute left ventricular
function impairment, indistinguishable from those
observed during episodes accompanied by typical STsegment changes and typical angina.
Although the number of the patients we studied is
small, they were included in the study only because of
the high frequency of angina. They therefore represent
an unselected population.
Thus, a reduction in blood supply may be a
11.
patients.
References
1. Guazzi M, Polese A, Fiorentini C, Magrini F, Bartorelli C: Left
ventricular performance and related hemodynamic changes in
Prinzmetal's variant angina pectoris. Br Heart J 33: 84, 1971
2. Guazzi M, Polese A, Fiorentini C, Magrini F, Olivari MT, Bartorelli C: Left and right heart hemodynamics during spontaneous angina pectoris. Comparison between angina with S-T
segment depression and angina with S-T segment elevation. Br
Heart J 37: 401, 1975
3. Maseri A, Mimmo R, Chierchia S, Marchesi C, Pesola A,
L'Abbate A: Coronary spasm as a cause of acute myocardial
ischemia in man. Chest 68: 625, 1975
4. Chierchia S, Marchesi C, Maseri A: Evidence of angina not
caused by increased myocardial metabolic demand and pattern
5.
6.
7.
8.
9.
10.
12.
13.
14.
15.
16.
17.
18.
19.
of electrocardiographic and hemodynamic alterations during
"primary" angina. In Primary and Secondary Angina Pectoris,
edited by Maseri A, Klassen GA, Lesch M. New York, Grune
& Stratton, 1978, p 145
Maseri A, Parodi 0, Severi S, Pesola A: Transient transmural
reduction of myocardial blood flow, demonstrated by thallium201 scintigraphy, as a cause of variant angina. Circulation 54:
280, 1976
Parodi 0, Severi S, Uthurralt N, Solfanelli S, Maseri A:
Angina pectoris at rest: regional myocardial perfusion during
S-T segment elevation or depression. (abstr) Circulation 56
(suppl III): Il1-229, 1977
Maseri A, L'Abbate A, Pesola A, Ballestra AM, Marzilli M,
Severi S, Maltinti G, Denes DM, Parodi 0, Biagini A: Coronary vasospasm in angina pectoris. Lancet 1: 713, 1977
Higgins CB, Wexler L, Silverman J, Schroeder JS: Clinical and
arteriographic features of Prinzmetal variant angina: documentation of arterial spasm. Am J Cardiol 37: 831, 1976
Wiener L, Kasparian H, Duca PR, Walinsky P, Gottlieb RS,
Hanckel F, Brest AN: Spectrum of coronary arterial spasm.
Clinical, angiographic and myocardial metabolic experience in
29 cases. Am J Cardiol 38: 945, 1976
Amiel M, Delaye J, Delahaye JP, Janin A, Devic J: Prinzmetal
et spasme coronarien concomitants. Etude angiographique chez
13 patients. J Radiol Electrol 58: 27, 1977
Hugenholtz PG, Wagner HR, Gamble WJ, Polanyi ML: Direct
read-out of cardiac output by means of the fiberoptic indicator
dilution method. Am Heart J 77: 178, 1969
Polanyi ML: Recent development in fibre optics oximetry. In
Oxygen Measurements in Biology and Medicine, edited by
Payne JP, Hill DW. London, Butterworths, 1975, p 369
Marchesi C, Chierchia S, Maseri A: Left and right ventricular
pressure monitoring in CCU. Method and significance. In
IEEE Proc Comput Cardiol Rotterdam, 1977, p 579
Guazzi M, Olivari MI, Polese A, Fiorentini C, Magrini F:
Repetitive myocardial ischemia of Prinzmetal type without
angina pectoris. Am J Cardiol 37: 923, 1976
Maseri A, Severi S, De Nes M, L'Abbate A, Chierchia S,
Marzilli M, Ballestra AM, Parodi 0, Biagini A, Distante A:
"Variant angina": one aspect of a continuous spectrum of
vasospastic myocardial ischemia. Am J Cardiol 42: 1019, 1978
Prinzmetal M, Kennamer R, Merliss R, Wada T, Bor N:
Angina pectoris. A variant form of angina pectoris. Am J Med
27: 375, 1959
Bishop VS, Kaspar RL, Barnes GE, Kardon MB: Left ventricular function during acute regional myocardial ischemia in the
conscious dog. J Appl Physiol 37: 785, 1974
Vatner SF, Millard RW, Patrick TA, Heyndrickx GR: Effects
of propranolol on regional myocardial function, electrogram
and blood flow in conscious dog with myocardial ischemia. J
Clin Invest 57: 1261, 1976
Heyndrickx G, Millard WR, McRitchie RJ, Maroko PR,
Vatner SF: Regional myocardial functional and electrophysiological alterations after brief coronary artery occlusion in
768
CIRCULATION
conscious dogs. J Clin Invest 56: 978, 1975
20. Constantini LR: Extracardiac factors contributing to hypotension during coronary occlusion. Am J Cardiol 11: 205, 1963
21. Brown AM: Excitation of afferent cardiac sympathetic nerve
fibres during myocardial ischemia. J Physiol 190: 35, 1967
22. Malliani A, Schwarz PJ, Zanchetti A: A sympathetic reflex
elicited by experimental coronary occlusion. Am J Physiol 217:
703, 1969
23. Noble RJ, Rothbaum DA, Knoebel SB, McHenry PL, Anderson GJ: Normalization of abnormal T waves in ischemia. Arch
Intern Med 136: 391, 1976
24. Haiat R, Halphen C, Derrida JP, Chiche P: Pseudo normalization of the repolarization during transient episodes of myocardial ischemia. Am Heart J 94: 390, 1977
25. Roughgarden JW, Newman EV: Circulatory changes during
the pain of angina pectoris, 1772-1965. A critical review. Am J
Med 41: 935, 1966
26. Robinson BF: Relation of heart rate and systolic blood pressure
to the onset of pain and angina pectoris. Circulation 35: 1073,
1967
VOL 61, No 4, APRIL 1980
27. Littler WA, Honour AJ, Sleight P, Stoff FD: Continuous
recording of direct arterial pressure and electrocardiogram in
unrestricted man. Br Med J 3: 76, 1972
28. Rowe GG, Castillo CA, Crumpton CW: Effects of hyperventilation on systemic and coronary hemodynamics. Am Heart J
63: 67, 1962
29. Neill WA, Hattenaur M: Impairment of myocardial 02 supply
due to hyperventilation. Circulation 52: 854, 1975
30. Case RB, Greenberg H, Moskowitz R: Alterations in coronary
sinus of PO2 and 02 saturation resulting from pCO2 changes.
Cardiovasc Res 9: 167, 1975
31. Case RB, Greenberg H: The response of canine coronary
vascular resistance to local alterations in coronary arterial
PCO2 Circ Res 39: 558, 1976
32. Folts JD, Crowell EB Jr, Rowe GG: Platelet aggregation in
partially obstructed vessels and its elimination by aspirin. Circulation 54: 365, 1976
33. Uchida Y, Yoshimoto N, Murao S: Cyclic fluctuations in coronary blood pressure and flow induced by coronary artery constriction. Jpn Heart J 16: 454, 1975
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
Effect of Acute Coronary Artery Occlusion
on Local Myocardial Extracellular
K+ Activity in Swine
JEFFREY L. HILL, PH.D.,
AND
LEONARD S. GETTES, M.D.
SUMMARY We studied the time course, magnitude and homogeneity of the change in extracellular
myocardial potassium activity after acute ligation of the left anterior descending coronary artery in pigs using
potassium-sensitive electrodes made from a valinomycin-polyvinyl chloride matrix membrane. We also studied
the relationship betweein the changes in potassium activity and the simultaneous changes in ventricular activation using the reference barrel of the K+ electrode to record ventricular electrograms. We found that the K+
rose sooner, more rapidly and to higher levels than previously reported. The K+ changes occurred in three
phases: a phase of rapidly rising K+ that began within seconds of the ligation and lasted 5-15 minutes, a
plateau phase that lasted approximately 15 minutes and a phase of slowly rising K+ that extended throughout
the longest occlusion (60 minutes) used in this study. The K+ changes were reversed by release of the occlusion
during the rapidly rising and plateau phases, but were not reversed by release of the occlusion during the phase
of slowly rising K+. Inhomogeneities in the K+ rise appeared between the center and lateral margins of the
midmyocardial ischemic zone, between the subendocardium and the subepicardium in the center of the
ischemic zone, and between closely spaced electrodes located in the midmyocardial center of the ischemic zone.
Thus, the change in K+ activity, as recorded by our electrodes, can be considered an excellent marker of
ischemia. Changes in ventricular activation paralleled the K+ rise, the inhomogeneities of K+ rise and the
reversal of the K+ rise after release but could not be entirely explained by the change in K+.
THE ACCUMULATION of substances released by
the ischemic myocardium is thought to play an important role in the genesis of ischemia-related
arrhythmias.' Harris and co-workers2 demonstrated
that the ischemic myocardium lost K+ into the exFrom the Division of Cardiology, University of North Carolina
School of Medicine, Chapel Hill, North Carolina.
Supported in part by NHLBI grant HL 23624.
Part of this work was included in a dissertation submitted to the
University of Kentucky by Jeffrey L. Hill in partial fulfillment of
the requirements for a Ph.D. degree.
Address for correspondence: Jeffrey L. Hill, Ph.D., Division of
Cardiology, University of North Carolina School of Medicine,
Chapel Hill, North Carolina 27514.
Received May 12, 1979; revision accepted September 12, 1979.
Circulation 61, No. 4, 1980.
tracellular space after acute coronary artery ligation.
They also showed that the intracoronary injection of
KCI produced electrocardiographic changes and ventricular arrhythmias similar to those induced by coronary artery ligation.2 3 Harris et al. concluded that
the rise in extracellular K+ was a major cause of
ischemia-induced ventricular fibrillation.2 3 These
observations have been confirmed by other investigators.4'10 However, Downar et al.'0 recently
presented data that suggest that factors other than or
in addition to the ischemia-induced K+ rise cause the
associated electrophysiologic changes.
In all prior experiments, K+ was measured in the
vein draining the ischemic zone. Thus, the K+ values
were only an approximation of the actual myocardial
extracellular K+ and the evidence for or against a
Sequence of events in angina at rest: primary reduction in coronary flow.
S Chierchia, C Brunelli, I Simonetti, M Lazzari and A Maseri
Downloaded from http://circ.ahajournals.org/ by guest on June 16, 2017
Circulation. 1980;61:759-768
doi: 10.1161/01.CIR.61.4.759
Circulation is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1980 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/61/4/759
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/