Download Myocardial Function and Coronary Blood Flow Response to Acute

MYOCARDIAL FUNCTION AND CORONARY FLOW IN DIABETES/Haider el al.
distribution. J Physiol (Lond) 257: 699-712, 1976
32. Judy WV, Watanabe AM, Henry DP, Besch HR, Murphy WR,
Hockel GM: Sympathetic nerve activity; role in regulation of blood
pressure in the spontaneously hypertensive rat. Circ Res 38: (suply II):
21-29,1976
33. Heymans C Neil E: Reflexogenic areas of the cardiovascular system.
Boston, Little, Brown, 1958
34. Bilgutay AM, Bilgutay I, Lillehei CW: Baropacing; a new concept in
the treatment of hypertension. In Baroreceptors and Hypertension,
577
edited by P Kezdi. Oxford, Pergamon, 1967, pp 425-437
35. Schwartz SI, Griffiths LSC, Neistadt A, Hagfors N: Chronic carotid
sinus nerve stimulation in the treatment of essential hypertension. Am
J Surg. 114: 5-15, 1969
36. Tuckman J, Reich T, Lyon AF, Medlowitz M, Jacobson JH: Electrical
stimulation of the sinus nerves in hypertension patients; clinical evaluation of physiological studies. In Neural Control of Arterial Pressure,
edited by JE Wood. New York, American Heart Association, 1968, p
23
Myocardial Function and Coronary Blood Flow
Response to Acute Ischemia in Chronic
Canine Diabetes
Downloaded from http://circres.ahajournals.org/ by guest on October 13, 2016
BUNYAD HAIDER, S. SULTAN AHMED, CHRISTOS B. MOSCHOS, HENRY A. OLDEWURTEL,
AND TIMOTHY J. REGAN
SUMMARY To examine the influence of preexistent diabetes
mellitus on left ventricular performance and coronary blood
flow responses to acute ischemia, mild normoglycemic diabetes
was induced in nine mongrel dogs after three doses of alloxan,
(20 mg/kg, iv), at monthly intervals. Hemodynamic measurements and coronary blood flow (85Kr clearance) were obtained
before and after the onset of ischemia. This was produced by occlusion of the proximal left anterior descending coronary artery
via a balloon-type catheter in nine intact anesthetized diabetic
dogs and 10 nondiabetic dogs. During the 1st hour of ischemia in
the diabetic group, the end-diastolic pressure rose from 7 ± 1.1
(mean ± SE) mm Hg to 23.8 ± 2 . 3 without a significant increase
of end-diastolic volume. In controls end-diastolic pressure rose
from 8.6 ± 1.1 mm Hg to 15.3 ± 1.4, and end-diastolic volume
was significantly increased, so that the ratio of end-diastolic
pressure and volume was significantly higher in the diabetic group
(/•< 0.005). Although indices of contractility did not differ, stroke
volume and work reductions were significantly greater in diabetics, despite the fact that coronary blood flow was reduced to
a similar extent. Size of the ischemic areas appeared comparable
as judged by distribution of dye injected distal to the occlusion.
Since potassium loss and sodium gain in the inner and outer layers
of ischemic tissue did not differ between the two groups, the intensity of ischemia seemed similar. Glycogenolysis was unimpaired in the diabetic ischemic muscle but triglyceride levels remained elevated. Morphologically the diabetic myocardium was
characterized by a diffuse accumulation of periodic acid-Schiffpositive glycoprotein in the interstitium, which was thought to
limit diastolic filling of the ischemic ventricle and to contribute to
the substantial reduction of ventricular performance.
ALTHOUGH the influence of acute regional ischemia on
left ventricular function has been well defined in the previously normal animal. lt2 the response of the ventricle affected by a chronic metabolic or structural abnormality has
not been described. Acute myocardial infarction has been
reportedly associated with a greater incidence of pump
failure and higher mortality in diabetes mellitus.3 Although the increased mortality from cardiac disease complicating diabetes mellitus has been traditionally attributed
to accelerated atherosclerosis of the coronary arteries.4
this is a disputed issue since recent evidence in studies
using more quantitative methods and age-matched controls has shown that the complicated lesions of atherosclerosis may occur to only a slightly greater extent in diabetics.5
In a previous study from this laboratory.6 we observed
altered myocardial function in chronic diabetes mellitus in
dogs, associated with accumulation of periodic acid-Schiff
(PAS)-positive glycoprotein in the myocardial interstitium
without coronary obstructive lesions; this morphological
abnormality also has been observed in man ?•8 To examine
the response of the diabetic myocardium during acute
regional ischemia as compared to normal controls the
following study was undertaken.
From the Department of Medicine, College of Medicine and Dentistry
of New Jersey —New Jersey Medical School, Newark, New Jersey. Supported in part by Research Grant HL 09914 and Postgraduate
Training Grant HL 05510 from the National Heart and Lung Institute.
Received July 7. 1976; accepted for publication December 9, 1976.
Address for reprints: Bunyad Haider, M.D., CMDNJ-New Jersey
Medical School, Department of Medicine, 100 Bergen Street, Newark,
New Jersey 07103.
Methods
Two groups of healthy male mongrel dogs 2-4 years old
and weighing 21-28 kg were studied. The dogs had no
clinical evidence of disease for 6-8 weeks before admission to the study groups. Hematocrit and serum albumin
were initially normal and both groups received the same
diet consisting of 8% fat. 22% protein. 58% carbohydrate. 9% ash. and 3% crude fiber. One group (n = 10)
served as controls with normal glucose tolerance by intravenous testing. The other group (n = 9) was made diabetic with low doses of alloxan at monthly intervals. To
produce mild normoglycemic diabetes, alloxan monohy-
578
CIRCULATION RESEARCH
Downloaded from http://circres.ahajournals.org/ by guest on October 13, 2016
drate in sterile saline was administered intravenously in a
dose of 20 mg/kg over a 1-minute period. Two additional
doses were given at monthly intervals to maintain a relatively steady state of glucose intolerance. Larger doses of
alloxan were avoided to prevent ketoacidosis. The diabetic
dogs and the controls were observed for an average period
of 9 months after the initial alloxan dose.
Glucose tolerance was measured before and every 3
months after the initial dose of alloxan. Glucose was
infused over 1 minute (1.05 g/kg. iv) through catheter
tubing in the relatively relaxed, unanesthetized dog. Blood
samples were taken at 1. 2. 4. 6. 10. 20. 30. 45. 60. and
120 minutes, and plasma glucose was analyzed by the
glucose oxidase method.9 The glucose clearance constant
was calculated to estimate the disappearance rate from the
vascular compartment and was derived by a semilogarithmic plot of glucose concentrations beginning with the 1minute sample for calculation of slope. In the diabetic
dogs venous blood samples were obtained in the fasting
state at the onset of the study and at 3-month intervals for
determination of plasma lipids. Blood was placed in
chilled tubes containing ethylenediaminetetraacetic acid
(EDTA); after separation in a refrigerated centrifuge, the
plasma was stored at — 20°C until assay. Duplicate determinations of free fatty acid.10 triglyceride." and phospholipid12 were made.
HEMODYNAMIC STUDIES
Dogs were anesthetized with morphine sulfate (2 mg/
kg) and sodium pentobarbital (12 mg/kg. iv) and studies
were performed with the chest intact. Ventilation was
regulated by a Harvard pump via a cuffed endotracheal
tube to maintain pH and Po2 within normal range. Catheters were placed in the pulmonary artery, left ventricle,
and ascending aorta and maintained patent with infusion
or intermittent flushing with small volumes of saline. The
50-cm Goodale-Lubin catheters were connected directly
to a Statham strain gauge transducer (P23Gb) and recorded on a multichannel oscilloscope recorder (Electronics for Medicine). The first derivative of left ventricular
pressure pulse (dP/dt) was computed continuously by a
resistance-capacitance differentiating circuit and converted to mm Hg per second. Left ventricular enddiastolic pressure was recorded at high sensitivity and the
average of 4-5 end-expiratory pressures were calculated.
To evaluate accuracy of the fluid-filled system used in this
study for measuring left ventricular pressure, four separate
dogs were studied. Left ventricular pressure recordings
were obtained with a no. 8 catheter-tip transducer (MicroTip, Millar Instrument) and compared with pressure recorded via a 50-cm-long Goodale-Lubin catheter connected directly to a P23Gb Statham strain gauge transducer positioned at the midthoracic level. Thirty-two simultaneous determinations of left ventricular end-diastolic
pressure during a wide range of hemodynamic interventions showed close agreement, with a correlation coefficient of 0.98 (y = 1.019x -0.409) in accord with a prior
report.13
Cardiac output and left ventricular volumes were determined in duplicate by the thermal indicator-dilution
VOL. 40, No. 6, JUNE 1977
method. Cardiac output by this technique has been found
to correlate well with the dye-dilution technique,14-l5 and
end-diastolic volume has been correlated with the angiographic method.16'17 A Swan-Ganz thermodilution catheter tip was placed in the main pulmonary artery and 10 ml
of normal saline at room temperature was injected as a
bolus into the right atrium through the proximal lumen.
Change of temperature was detected by the thermistor in
the pulmonary artery and displayed on the recorder
through a Wheatstone bridge; the area of the curve was
computed manually for cardiac output calculation. Left
ventricular injection fraction was obtained in duplicate
from the left ventricular washout curves obtained by thermodilution method. Cold saline (5 ml) was injected as a
bolus at the inflow site of the ventricle and the step change
of temperature was determined from a thermistor-tip catheter placed just above the aortic valve. Adequate mixing
appears to be present at this injection site, since prior
studies showed good correlation of ejection fractions derived sequentially from left atrial and ventricular injection
sites at the level of the inflow tract or apex. The total
amount of saline infused for the measurement of cardiac
output and ejection fraction was approximatey 150 ml in
each dog during the course of study. We have previously
observed that 180 ml/hour infused in dogs of equivalent
size produced no significant hemodynamic effects.18
End-diastolic volume was calculated from the ratio of
stroke volume and ejection fraction and expressed per
kilogram of body weight. The ejection fraction values
were lower than the mean of 38% observed by the indicator-dilution technique in conscious sedated dogs.19 However, this is within the range observed after the administration of sodium pentobarbital anesthesia,20 apparently
related to the myocardial depressant effects of the anesthetic agent.
Stroke work index in gram-meters per kilogram was
calculated from the product of stroke volume per kilogram
and the mean left ventricular systolic pressure minus enddiastolic pressure times 1.36. The ratio of left ventricular
end-diastolic pressure and volume was used as a simple
index of wall stiffness at end-diastole. To evaluate the
contractile state of the ventricle in vivo we have used an
index that normalizes left ventricular dP/dtm^, for (1) the
maximal isovolumic pressure (MIP); (2) circumferential
fiber length (27rr), assuming a spherical shape at the end of
the systolic isovolumic period and deriving the radius from
the end-diastolic volume. The formula21 is (dP/dtn^/MIP)/
277T. Using the same end-diastolic fiber length and pressure, the end-diastolic tension in dynes x 106/beats was
calculated as end-diastolic pressure x r2 x 4,188; the
latter is derived from n x 1.36 (cm H2O/mm Hg) x 980
(cm/sec2) x 1 (g/cm3).
MODEL FOR ISCHEMIA
A double-lumen, 5F catheter with a distal lumen was
positioned in the proximal 1.5 cm of the left anterior
descending coronary artery under fluoroscopic control.
The balloon was inflated gradually over a period of 60
seconds. Aortic pressure, peripheral coronary pressure.
MYOCARDIAL FUNCTION AND CORONARY FLOW IN DIABETES/Haider et al.
Downloaded from http://circres.ahajournals.org/ by guest on October 13, 2016
and electrocardiogram (ECG) lead I were continuously
monitored. Complete coronary occlusion was evidenced
by a sustained reduction of mean coronary pressure to
approximately 25 mm Hg and appearance of an injury
potential on standard lead I in all the dogs studied.
To reduce mortality due to the high incidence of arrhythmias in the initial 15 minutes after ischemia, procainamide (10 mg/kg) was administered intravenously to dogs
developing ventricular tachycardia (four in the nondiabetic group and three of the diabetics). No antiarrhythmic agent was administered after the initial 15 minutes. Since the circulatory effects of procainamide so administered are considered to last a matter of minutes,22 it is
improbable that the antiarrhythmic drug contributed to
the hemodynamic differences observed in these two
groups. Measurement of coronary blood flow to the ischemic area was obtained by injections of 85Kr distal to the
occlusion site. This inert gas method appears to give valid
flow measurements over a wide range of tissue perfusion.23'24 Approximately 100 /u.Ci of 85Kr was injected at
6- to 10-minute intervals in duplicate before ischemia and
12- to 20-minute intervals thereafter. Blood flow was
calculated from the decay slopes obtained by precordial
scintillation counting.
At 60 minutes of ischemia, the chest was opened and
the heart was arrested with iced Ringer's solution. To
delineate the area profused by the left anterior descending
coronary artery (LAD) distal to the occlusion, Evans blue
dye was injected via the LAD catheter just prior to arrest.
A transmural section was rapidly excised fromthe central
ischemic area and the nonischemic posterior wall at least 1
cm from the former. Both were frozen in liquid nitrogen
for glycogen assay.25 The remainder of the dyed area was
excised, weighed, and related to the total left ventricle and
septum. Sections were taken from ischemic and nonischemic areas of left ventricle for electrolyte and lipid
analysis as well as histochemical examination. The ventricle was divided into inner and outer layers; the latter was
carefully trimmed of epicardial adipose tissue. Samples
were homogenized in phosphate buffer and the lipids were
extracted in chloroform-methanol to determine free fatty
acid,10 triglyceride," and phospholipid.12 Separate samples were homogenized and extracted for 48 hours in
distilled water. Potassium and sodium were determined on
an AutoAnalyzer system (Technicon) with flame attachment. Water content was obtained by drying samples in an
oven at 100°C to constant weight. A group of normal
intact anesthetized dogs without ischemia underwent similar tissue studies for comparison with the two experimental
groups.
Histochemical examination included PAS staining after
treating twice with diastase to exclude staining of glycogen.26 Statistical data were expressed as means ± standard
errors; the paired or nonpaired Student's/-test was applied
as appropriate.27
Results
Both groups remaind healthy over the approximate 9month period prior to the induction of ischemia. Body
weight was maintained; the initial hematocrit of 45 ± 1 . 7
579
in the control group and 44 ± 2.0 in the diabetics was not
significantly changed during this period. Prior to the terminal study in the controls, the mean fasting blood sugar
was 83 ± 3 mg/100 ml and the mean glucose clearance
constant was 3.7 ± 0.1. In the alloxan-diabetic dogs the
mean fasting plasma glucose was increased significantly
from a control of 84 ± 4 mg/100 ml to 104 ± 2 (P <
0.001) prior to the final study, although the mean value of
plasma glucose for the group was within normal limits.
The glucose clearance constant was reduced from 3.3 ± 1
prior to alloxan to a level of 2.1 ± 0.1 (P < 0.001), before
the terminal study.
HEMODYNAMIC FINDINGS DURING ISCHEMIA
The hemodynamic measurements were made in duplicate in the control state and during the course of ischemia
in both groups. Baseline hemodynamic values prior to
ischemia were not significantly different in the control and
diabetic groups (Tables 1 and 2). Between 30 and 60
minutes of ischemia in the 10 dogs of the control group
(Table 1), the stroke volume declined an average of 4%,
ejection fraction by 21%, and stroke work by 16%. The
end-diastolic pressure and volume increased by 80%
(P < 0.01) and 25% (P < 0.02), respectively. Changes in
heart rate and arterial pressure were not significant. In this
nondiabetic group, three dogs developed ventricular fibrillation between 30 and 60 minutes of ischemia and the
remaining seven survived the observation period. Since
the seven surviving dogs exhibited ventricular responses at
30 minutes similar to those at 60 minutes, the three that
succumbed early have been included.
The hemodynamic changes during ischemia in the diabetic group are indicated in Table 2. There were two
patterns of hemodynamic response in this group. The two
dogs that exhibited marked hypotension (group B, Table 2) progressed to shock and developed cardiac arrest
between 30 and 60 minutes of ischemia. Group A was
characterized by a modest but significant decline of arterial pressure and left ventricular failure and only one of the
seven dogs developed late ventricular fibrillation.
In the seven dogs of group A, the end-diastolic pressure
rose more than 3-fold while end-diastolic volume did not
change significantly. Stroke volume and stroke work declined by 35% (P < 0.02) and 55% (P < 0.01), respectively, a significantly greater decrease than in the nondiabetics (Table 2). In the normal control dogs, the cardiac
index was maintained during ischemia and reduced insignificantly from a control of 119.9 ± 6.9 ml/kg to 112 ±
9.0. In contrast, during ischemia in the diabetic group, the
cardiac index was reduced by almost one-third, (control,
114 ± 7.9 ml/kg to 78.5 ± 4.2; P < 0.01), a significantly
different response from that of controls (P < 0.02).
Calculation of the normalized index of contractility to
compare the relative performance of the contractile elements revealed no significant difference between normal
dogs (1.10 ± 0.11) and diabetic dogs (1.26 ± 0.09) before
ischemia; there was a similar small decline in both groups
during ischemia to 0.86 ± 0.07 and 1.10 ± 0.19, respectively. Since the index of contractility did not differ in the
CIRCULATION RESEARCH
580
TABLE
VOL. 40. No. 6, JUNE 1977
1 Left Ventricular Response to Acute Ischemia in Normal Dogs
Left ventricular
end-diastolic
Svstolic/diastolic
aortic pressure
(mmHg)
Heart rate
(beats/min)
No.
1
2
3
4*
5
6
1*
8
9*
10
Mean
Downloaded from http://circres.ahajournals.org/ by guest on October 13, 2016
± SEM
Pt
Pressure
Volume
(ml/kg)
(mm Hg)
Stroke
volume
(ml/kg)
EDP/EVI
Stroke
work
(g-m/kg)
Ejection
fraction
C
E
C
E
C
E
C
E
C
E
C
E
C
E
C
E
123
115
112
140
144
150
110
126
137
144
161
102
143
130
149
152
130
123
133
134
137/118
123/105
174/135
186/146
122/64
172/132
139/110
150/122
150/112
143/113
163/135
128/100
170/138
149/118
104/40
148/111
134/110
159/130
140/108
135/106
6
8
10
12
28
23
17
13
13
17
11
9
2.80
4.08
5.34
4.67
4.95
3.29
4.48
6.20
3.96
3.94
2.36
4.51
5.50
4.78
7.80
5.12
8.30
6.46
5.06
4.90
2.15
1.97
2.63
3.43
2.03
2.13
1.79
0.81
1.52
1.43
4.23
3.08
5.30
4.82
2.18
2.55
1.57
2.64
2.18
1.84
0.83
0.71
1.23
1.00
1.08
0.68
1.04
0.92
0.87
0.91
0.50
0.79
0.94
1.03
1.08
1.00
1.15
0.90
0.81
0.82
1.49
1.11
2.68
2.30
1.64
1.52
1.83
1.81
1.56
1.69
0.88
1.26
1.82
1.76
1.29
1.42
1.90
1.73
1.38
1.43
0.294
0.174
0.230
0.214
0.218
0.206
0.232
0.148
0.214
0.230
0.213
0.175
0.171
0.215
0.138
0.197
0.138
0.139
0.160
0.167
129 134
4.9 5.6
NS
150/116
6.8/7.0
143/110
6.3/8.8
4.37
0.31
5.47
0.53
1.98 3.03
0.22 0.37
<0 .01
0.92
0.05
0.89
0.06
1.76 1.48
0.14 0.09
<0.2
0.216
0.012
0.171
0.009
NS
14
16
10
7
8
5
6
6
15.3
8.6
1.4
1.1
< 0 .01
< 0 .02
NS
< 0 .01
C = control data; E = value obtained at 60 minutes of ischemia; EDP/EDVI = ratio of end-diastolic pressure and end-diastolic
volume index.
* E values obtained between 30 and 60 minutes of ischemia in nonsurvivors.
fP = paired f-test comparing C vs. E.
Left ventricular function was further assessed by plotting stroke work index against end-diastolic pressure (Fig.
1 A). In the control group during ischemia, the rising enddiastolic pressure was associated with a small decline of
stroke work. In contrast, the diabetic group at 60 minutes
of ischemia exhibited a substantial reduction of stroke
work associated with a 3-fold increase of end-diastolic
pressure. To determine whether the heart rate response
may have affected this relationship, dogs with relatively
small rate changes, within 16 beats/min of the control
state, were compared in the two groups (Fig. IB). The
two groups, the reduced stroke volume appears related at
least in part to the lack of end-diastolic volume increment.
Calculated end-diastolic tension increased from a control
of 0.29 ± 0.04 dynes x 106 to 0.54 ± 0.08 (P < 0.01) in
the normals, while in diabetics the rise of tension from a
control of 0.22 ± 0.03 dynes x 106 to 0.65 ± 0.06 during
ischemia, was significantly greater than in the normals (P
< 0.05). The ratio of end-diastolic pressure and volume
was unaltered during ischemia in both groups but was
significantly higher in the diabetic group (P < 0.005),
suggesting enhanced wall stiffness in the diabetics.
TABLE
2 Left Ventricular Response to Acute Ischemia in Diabetic Dogs
Left'ventricular end-diastolic
Heart rate
(beats/min)
Systolic/diastolicic
aortic pressure
(mm Hg)
C
E
C
E
72
126
168
144
150
150
81
88
132
138
128
112
132
128
168/118
149/112
165/136
140/105
156/130
149/126
188/143
140/110
119/95
117/109
133/84
137/106
107/87
160/135
Mean
127
± SEM
13.9
122
6.5
159/124
6.5/5.0
130/103
6.6/6.5
Stroke volPressure
(mm Hg)
ume
1EDP/EDVI
(ml/kg)
Stroke work
(g-m/kg)
Ejection fraction
C
E
C
E
C
E
C
E
C
E
4.47
2.37
4.31
3.75
3.09
3.33
4.52
4.40
4.87
3.90
2.17
3.13
2.93
4.50
2 .68
3 .37
1 .39
1 .06
1.94
1 .20
2 .21
5.90
4.92
8.20
9.21
5.64
5.80
6.88
1.70
0.69
0.80
0.80
0.96
0.62
1.30
1.10
0.59
0.57
0.61
0.57
0.51
0.68
3.71
1.34
1.70
1.60
2.00
1.22
3.22
1.66
0.78
1.01
0.71
0.89
0.70
0.92
0.380
0.290
0.185
0.213
0.310
0.186
0.287
0.250
0.121
0.146
0.280
0.182
0.174
0.151
7.1 23.8
2.3
1.1
< 0 .001
< 0 .005
3.69
0.30
3.71
0.37
0 .31
6.65
0.58
< 0 .001
< 0 .005
0.98
0.14
0.66
0.07
< 0 .01
2.11 0.95
0.36 0.12
<0.01
<0.02
< 0 i.Ol
8
4.20
2.80
C
E
Volume
(ml/kg)
Group A
1
2
3
4
5*
6
7
NS
NS
Pt
Pt
< 0 . 001
NS
12
8
6
4
6
4
10
26
24
32
20
17
17
31
1 .97
NS
< 0 .02
0.264 0.186
0.020 0.020
< 0 . 05
NS
Group B
1*
2*
132
156
78
90
132/100
178/150
89/64
90/76
7
14
31
4.40
3.80
1 .91
2.50
C = control data; E = value obtained at 60 minutes of ischemia.
* E values obtained between 30 and 60 minutes of ischemia in nonsurvivors.
t P = paired Mest comparing C vs. E.
%P = unpaired t-test comparing the responses in normals vs. diabetics.
3.19
8.16
1.27
0.50
0.60
0.50
2.14
1.15
0.70
0.85
0.302 0.136
0.178 0.130
MYOCARDIAL FUNCTION AND CORONARY FLOW IN DIABETES/Haider et al.
•
O
*
I
• CONTROL
O DIABETIC
p < 0.01
581
I NORMALS <N=7)
t i l l NORMALS + ISCHEMIA (N=7)
K7Z\ DIABETICS + ISCHEMIA (N=6)
OUTER
INNER
70
60
E
3 1.5
_
50
o- 40
uj
3
+
30
*
20
10
0
3
6
9
12
15
18
21 24 27
0
3
6
9
12
15
18
21
o
24 27
LV END-DIASTOLIC PRESSURE (mmHg)
Downloaded from http://circres.ahajournals.org/ by guest on October 13, 2016
FIGURE 1 Lefi ventricular (LV) function before and during ischemia. The stroke work index is plotted against end-diastolic
pressure in both the nondiabetic and the diabetic group. The
changes from the preischemic control are depicted by arrowhead.
Panel A, which represents the data (mean ± SEM) for all the dogs in
each group, illustrates that the impaired ventricular function in
response to ischemia is significantly greater in the diabetic group
(P < 0.01). Panel B represents the data for six dogs of the control
group and four dogs of the diabetic group with heart rate change
within 16 beats/min of the control state. The responses were qualitatively similar to those observed in the groups as a whole, indicating that these changes were independent of heart rate.
responses were qualitatively similar to those observed in
the groups as a whole, and the same relationship of enddiatolic pressure and stroke work was evident.
Coronary blood flow to the ischemic site, measured by
the 85Kr clearance technique, was similarly reduced in
both groups (Fig. 2). The ischemic area, as ascertained by
injection of Evans blue dye, was 31.0 ± 2.4% of the total
left ventricle in the control group and 32.7 ± 1.6% in the
diabetic group.
To examine the tissue response to ischemia, the transmural concentrations of potassium and sodium in the outer
and inner layers of ischemic myocardium were assayed
(Fig. 3). In the control group, a significant reduction of
CONTROL
DIABETIC
90
O
9
60
30
PRE-ISCHEMIA
30
60
MINUTES OF ISCHEMIA
FIGURE 2 Coronary blood flow before and during ischemia.
The mean coronary blood flow was comparable in the two groups
during the control state. At 30 minutes of ischemia the reduction in blood flow was about 33% of the control value and similar
in the two groups. At 60 minutes of ischemia the coronary blood
flow in the diabetics (six dogs) and the control group (seven dogs)
remained essentially unaltered.
i
i
60
50
™ 40
5
30
+
20
o
10
0
FIGURE 3 Transmural
distribution of cations in ischemic left
ventricle. The nondiabetic and diabetic groups exhibited similar
reduction of K* and increase ofNa+ content in the inner and outer
layers of ischemic myocardium, which were significantly different
from the normal group.
potassium and an increase of sodium concentrations were
present in the inner and outer layers of ischemic tissue.
Diabetic dogs exhibited a similar K+ reduction and sodium
gain in the two layers of ischemic myocardium. Tissue
water was 78.1 ± 0.73% in normals, and 82.9 ± 0.38%
and 81.7 ± 0.44% in the inner and outer layers, respectively, of nondiabetics during ischemia; the respective values were 83.0 ± 0.63% and 82.5 ± 0.85% in the diabetic
group. The nonischemic muscle exhibited small but nonsignificant changes of K+ and Na+ concentration in both
groups with ischemia. However, tissue water was elevated
in the nonischemic posterior wall of diabetics to 81.3 ±
0.42% and 81.4 ± 0.42% in the inner and outer layers,
respectively, compared to 79.7 ± 0.32% (P < 0.01) and
79.0 ± 0.52 (P < 0.01) in nondiabetics. The increment of
tissue water is more apparent when expressed in terms of
grams of H2O per gram of dry weight. In the inner layers,
where the largest changes were observed during ischemia,
water increased to 4.85 ± 0.09 g/g dry weight in nondiabetics, and 4.88 ± 0.19 for diabetics, compared to 3.56
± 0.21 in normals. The nonischemic posterior wall was
increased to 4.35 ± 0.09 in diabetics and 3.92 ± 0.9 in
nondiabetics (P < 0.01).
The expected decrease of glycogen levels in the ischemic
tissue of the anterior wall in normals occurred to a similar
extent in the diabetics (Table 3), supporting the view that
the severity of ischemia was comparable in the two groups.
Myocardial triglycerides were elevated in the inner and
outer layers of both ischemic and nonischemic regions in
the diabetics as compared to the corresponding area in the
controls. Myocardial phospholipid and free fatty acid were
not significantly different in the two groups. In the diabetics the plasma lipid classes in the fasting conscious state
were within normal limits through the 9-month observa-
CIRCULATION RESEARCH
582
TABLE
VOL. 40, No. 6, JUNE 1977
3 Substrate Composition of Left Ventricle
Glycogen
g/g)
Normals (n = 8)
Nondiabetic ischemia (n = 10)
Ant
Post
768
±44
775
±51
197*
±58
163*
Diabetic ischemia {n = 7)
±64
602t
±40
538t
±49
Triglyceride
(nmol/g)
Ant
Post
1.9
2.2
±0.26
±0.28
2.1
1.8
±0.35
3.38*
±0.64
±0.30
2.97*
±0.72
Phospholipid
(fimol/g)
Ant
17.2
±1.3
16.0
±1.17
18.7
±1.1
Free fatty acid
(/imol/g)
Post
Ant
Post
16.9
±1.6
15.7
±1.8
16.5
±1.2
10.8
±1.0
11.9
±1.4
10.4
±1.3
11.3
±1.1
11.8
±1.5
10.1
±1.2
Values are expressed as mean ± SEM. Ant = anterior wall, which was ischemic in the experimental groups; Post = the posterior wall
perfused by normal circumflex artery.
* P > 0.005 (nonpaired Mest vs. corresponding area in normals).
tf > 0.01 (nonpaired f-test vs. corresponding area in normals).
t P > 0.05 (nonpaired Mest vs. corresponding area in nondiabetics with ischemia).
Downloaded from http://circres.ahajournals.org/ by guest on October 13, 2016
tion period, except for free fatty acids, which increased
from a control of 486 ± 48 /xmol/liter to 985 ± 101 (P <
0.001) prior to the terminal study.
As previously observed,6 the diabetic dogs exhibited
PAS-positive, diastase-resistant material in the interstitium of both the ischemic and nonischemic areas. This was
absent in the nondiabetic controls.
Discussion
This diabetic dog model exhibits reduced glucose tolerance due to insulin deficiency6 and the mild degree of
diabetes enables long-term maintenance without exogenous hormone administration. Since abnormalities of myocardial function and composition were not observed in
dogs in which the pancreatic effects of alloxan and resultant diabetes were prevented, a direct cardiotoxic effect of
this compound appears unlikely.6 Moreover, similar morphological and functional abnormalities of left ventricular
myocardium were observed in dogs with spontaneous diabetes.
Occlusion of the proximal left anterior descending artery produces ischemia of approximately one-third of the
left ventricle as judged from distribution of dye injected
distal to the obstructed artery. Despite this sizable area of
ischemia, the normal dog maintained stroke volume near
control levels associated with moderate increments of ventricular end-diastolic pressure. Acute regional ischemia in
the diabetic dogs produced greater impairment of left
ventricular function despite a comparable area of ischemia. The most prominent hemodynamic difference in
the two groups was the 3-fold increase of end-diastolic
pressure without a corresponding end-diastolic volume
increment in the diabetics. Although the large increase in
left ventricular end-diastolic pressure might be due to
myocardial failure, the latter process during acute global
ischemia is associated with an increase of end-diastolic
diameter and presumably volume,28 in contrast to the
diabetic dogs in which end-diastolic volume did not
change.
Although the pressure-volume relationship of the left
ventricle during ischemia was not determined over a wide
range of pressure and volume values in these in vivo
experiments, the significantly higher ratio of end-diastolic
pressure to end-diastolic volume in diabetics is consistent
with an interpretation of reduced compliance compared to
the nondiabetic group. A prior study of ischemia in dogs
with a scar in the nonischemic myocardium2 indicated a
response similar to that of these diabetic dogs. This was
characterized by a 4-fold rise of filling pressure and a more
profound reduction of stroke volume and work. Hypotension analogous to that in the diabetics of group B occurred
in some dogs. Thus, alteration of a portion of the nonischemic myocardium may substantially affect the response
to ischemia even if hypertrophy does not occur. Since the
interstitial accumulation of PAS-positive glycoprotein is a
diffuse process in the diabetic myocardium (cf. Fig. 4 of
Regan et al.8), the relative contribution of ischemic and
nonischemic ventricle cannot be defined with certainty.
However, the behavior of the myocardium in dogs with
scar strongly supports the view that chronic compositional
changes in the nonischemic zone of the left ventricle may
be the basis for the observed differences in response to
ischemia compared to controls.
The nonischemic area is known to undergo mechanical
and metabolic alterations,29 associated with those of the
ischemic tissue. A modest reduction in glycogen levels was
similar in diabetics and nondiabetics (Table 3). Slight
reductions of potassium and gain of sodium in this area of
muscle previously have been shown to be related to an
enlarged extracellular space.30 However, our diabetic dogs
exhibited a significant increase in water content of the
nonischemic segment. This acute change may have contributed, with the accumulated glycoprotein, to the enhanced diastolic stiffness manifested as a larger rise of enddiastolic pressure in relation to end-diastolic volume.
Although the declining blood pressure initially may
have contributed to reduced end-diastolic volume by virtue of the smaller afterload, the significantly greater rise
of left ventricular end-diastolic pressure suggests that this
was not a major factor in altered pressure-volume relationships during ischemia. A pressure decline approximating that which occurred in diabetics has been shown to
effect a relatively small (less than 4 ml) reduction of enddiastolic volume.31 The greater reduction of stroke output
in diabetics is at least partially related to reduced diastolic
filling of the ventricle rather than impaired systolic performance, since neither ejection fraction nor the index of
contractility were significantly lower than in normals sub-
MYOCARDIAL FUNCTION AND CORONARY FLOW IN DIABETES/Haider et al.
Downloaded from http://circres.ahajournals.org/ by guest on October 13, 2016
ject to ischemia. This was also manifested in the stroke
work to end-diastolic pressure relationship, which was
more abnormal than in nondiabetics and was independent
of heart rate change (Fig. 1).
The modest reduction of aortic pressure in the diabetics
was largely due to the reduction in stroke volume, since
there was no significant difference in the calculated mean
peripheral resistance when this group was compared to the
nondiabetics. In two of nine diabetic dogs that exhibited
marked hypotension, the drop in arterial pressure was in
excess of the decline in stroke volume, so that the vasomotor response of the peripheral vasculature appeared to be
inappropriate.
Coronary blood flow to the ischemic myocardium when
the coronary artery is completely occluded is presumed to
represent collateral flow. In this study the perfusion level
in ischemic tisue appeared to be comparable in both normal and diabetic groups undergoing ischemia. Although
morphological abnormalities have been observed in the
media of intramural vessels in this animal model,6 this did
not appear to effect luminal narrowing. The observation
of collateral flow levels which were comparable to the
nondiabetic group at this stage of diabetes implies a functionally normal microvasculature in response to the vasodilator stimulus of ischemia. The fact that the transmural
distribution of potassium and sodium ions in the ischemic
area was similarly altered in both groups supports the view
that blood perfusion of the inner and outer wall was quantitatively similar.
In addition to the anatomical alterations of the ventricle
in the diabetic dogs, the altered response to ischemia may
be related to changes in energy production and utilization.
Enhanced glycogenolysis is a feature of the metabolic
response in ischemic muscle.29 In view of the similarly
reduced glycogen levels in diabetics and in nondiabetics
after 60 minutes of ischemia, the glycogenolytic response
does not appear to be impaired by diabetes of this degree
and duration. On the other hand, endogenous lipolysis
does not seem to contribute significantly to substrate availability for energy production, since the elevated levels of
triglyceride characteristic of the diabetic heart6 were not
altered by ischemia. Because acute ischemia effected by a
complete coronary occlusion in the normal dog results in
severe depletion of high energy phosphate29 and inhibits
the formation of calcium-myosin complexes,32 it remains
to be shown that the diabetic state significantly intensifies
these biochemical abnormalities.
Acknowledgments
We gratefully acknowledge the expert technical assistance of B. Jenkins
and A. K. Gandhi and the secretarial assistance of A. Brown and A.
Binetti in the preparation of this manuscript.
References
1. Kumar R, Hood WR Jr, Joison J: Experimental myocardial infarction.
VI. Efficacy and toxicity of digitalis in acute and healing phase in intact
conscious dogs. J Clin Invest 49: 358-364, 1970
2. Regan TJ, Passannante AJ, Khan MI, Oldewurtel HA, Jesrani MU:
Influence of scar on left ventricular performance at the onset of
myocardial ischemia; shock versus heart failure. J Clin Invest 50: 534543, 1971
583
3. Partamian JO, Bradley RF: Acute myocardial infarction in 258 cases
of diabetes. N Engl J Med 273: 455-461, 1965
4. Bell ET: A post morten study of vascular disease in diabetes. Arch
Pathol 53: 444-455, 1952
5. Vihert AM, Zahdanov VS, Matova EE: Atherosclerosis of the aorta
and coronary vessels of the heart in cases of various disease. J Atheroscler Res 9: 170-192, 1969
6. Regan TJ, Ettinger PO, Khan MI, Jesrani MU, Lyons M, Oldewurtel
HA, Weber M: Altered myocardial function and metabolism in
chronic diabetes mellitus without ischemia in dogs. Circ Res 3: 222237, 1974
7. Sohar E, Ravid M, Benshaul Y, Reshef T, Gafni J: Diabetic fibrillosis.
Am J Med 49: 64-69, 1970
8. Regan TJ, Ahmed SS, Levinson GE, Oldewurtel HA, Ahmad MR,
Haider B, Lyons MM: Cardiomyopathy and regional scar in diabetes
mellitus. Trans Assoc Am Phys 83: 217-223, 1975
9. Hill JB, Kessler G: Automated determination of glucose utilizing a
glucose oxidase-peroxidase system. J Lab Clin Med 57: 970-980,
1961
10. Kelley TF: Improved method of microtitration of fatty acid. Anal
Chem 37: 1078-1079, 1965
11. Kessler G, Lederer H: Flurometric measurement of triglycerides. In
Automation in Analytical Chemistry, vol 1, Technicon Symposia.
White Plains, N.Y., Mediad Inc., 1967, pp 341-344
12. Barlett G: Phosphorous assay in column chromatography. J Biol
Chem 234: 466-468, 1959
13. Falsetti HL, Mates RE, Grant C, Greene DG, Bunnell IL: Left
ventricular wall stress calculated from one-plane cineangiography.
Circ Res 26: 71-83, 1970
14. Ganz W, Donoso R, Marcus HS, Forrester JS, Swan HJC: A new
technique for measurement of cardiac output by thermodilution in
man. Am J Cardiol 27: 392-396, 1971
15. Weisel RD, Berger RL, Hechtman HB: Measurement of cardiac
output by thermodilution. N Engl J Med 202: 682-684, 1975
16. Swan HJC, Ganz W, Wallace JC, Tamura K: Left ventricular enddiastolic volume (EDV) by angiographic and thermal methods in a
single diastole (abstr). Circulation 38 (suppl VI): 193, 1968
17. Weigand KH, Jacob K: Zur Frage der Restvolumenbestimmug des
linken Ventrikels in malurlichen Kreislanf. Arch Kreislaufforsch
Beih z Kreislaufforsch 46: 97-114, 1965
18. Regan TJ, Koroxenidis G, Moschos CB, Oldewurtel HA, Lehan PH,
Hellems HK: The acute metabolic and hemodynamic responses of the
left ventricle to ethanol. J Clin Invest 45: 270-280, 1966
19. Bristow JC, Farrehi C, Ueland K: Left ventricular volume studies in
tranquilized dogs with local anesthesia. Circ Res 18: 457-463, 1966
20. Rappaport E, Wiegand BD, Bristow JD: Estimation of left ventricular
residual volume in dogs by thermodilution method. Circ Res 11: 803810, 1962
21. Frank MJ, Levinson GE: An index of contractile state of the myocardium in man. J Clin Invest 47: 1615-1626, 1968
22. Austin WG, Moran JM: Cardiac and peripheral vascular effects of
lidocaine and procainamide. Am J Cardiol 16: 701-707, 165
23. Ross RS, ueda K, Lichtlen PR, Rees JR: Measurement of myocardial
blood flow in animals and man by selective injection of radioactive
inert gas into the coronary arteries. Circ Res 15: 28-41, 1964
24. Herd JA, Hollenberg M, Thorburn GD, Kopald HH, Barger AC:
Myocardial blood flow determined with krypton85 in unanesthetized
dogs. Am J Physiol 203: 112-124, 1962
25. Roe J, Dailey R: Determination of glycogen with the anthrone reagent. Anal Biochem 15: 245-250, 1966
26. Luna LG:Manual of Histologic Staining Methods of the Armed Forces
Institute of Pathology, ed 3. New York, McGraw-Hill, 1968, p 158
27. Batson HC: An Introduction to Statistics in the Medical Sciences.
Minneapolis, Burgess, 1961, p 14
28. Vatner SF, McRitchie RJ, Maroko PR, Patrick TA, Braunwald E:
Effects of catecholamines, exercise and nitroglycerin on the normal
and ischemic myocardium in conscious dogs. J Clin Inest 54: 563-575,
1974
29. Opie LH, Owen P: Effect of glucose-insulin-potassium infusions on
arteriovenous differences of glucose and of free fatty acids and on
tissue metabolic changes in dogs with developing myocardial infarction. Am J Cardiol 38: 310-321, 1976
30. Polimeni PI, Al-Sadir J: Expansion of extracellular space in the nonischemic zone of the infarcted heart and concomitant changes in tissue
electrolyte contents in the rat. Circ Res 37: 725-732, 1975
31. Tsakiris AG, Vanderberg RA, Banchero N, Strum RE, Wood EH:
Variations of left ventricular end-diastolic pressure, volume, and ejection fraction with changes in outflow resistance in anesthetized intact
dogs. Circ Res 23: 213,22, 1968
32. Katz AM, Hecht HH: The early "pump" failure of the ischemic heart.
Am J Med 47: 497-502, 1969
Myocardial function and coronary blood flow response to acute ischemia in chronic canine
diabetes.
B Haider, S S Ahmed, C B Moschos, H A Oldewurtel and T J Regan
Downloaded from http://circres.ahajournals.org/ by guest on October 13, 2016
Circ Res. 1977;40:577-583
doi: 10.1161/01.RES.40.6.577
Circulation Research is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1977 American Heart Association, Inc. All rights reserved.
Print ISSN: 0009-7330. Online ISSN: 1524-4571
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://circres.ahajournals.org/content/40/6/577
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in
Circulation Research 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 Research is online at:
http://circres.ahajournals.org//subscriptions/