Download Abnormal left ventricular diastolic function during

Clinical Science (1995) 89, 461-465 (Printed in Great Britain)
461
Abnormal left ventricular diastolic function during
cold pressor test in uncomplicated insulin-dependent
diabetes mellitus
Ole Gq>TZSCHE, Ahmed DARWISH, Lars Peter HANSEN* and Liv G(j>TZSCHE
University Department of Cardiology, A.rhus Amtsslygehus and *Department of Pediatrics, A.rhus
Kommunehospital, Cardiovascular Research Center, A.rhus University Hospital, A.rhus, Denmark
(Received 23 March/16 June 1995; accepted 4 July 1995)
1. Insulin-dependent diabetes mellitus is a known risk
factor for congestive heart failure and an early
diastolic dysfunction has been described. In order to
see if diabetes itself and not complications like
hypertension, nephropathy or ischaemic heart disease
can be considered responsible for the abnormal diastolic function of the left ventricle, 17 young patients
with uncomplicated insulin-dependent diabetes mellitus and 12. control subjects were exposed to a cold
pressor test.
2. Blinded echo-Doppler examination was performed
before and during the test. During basal conditions,
left ventricular dimensions and volumes were smaller
in diabetes and atrial contributions to left ventricular
filling were increased.
3. During the cold pressor test, isovolumic relaxation
time increased, peak early filling velocity (E) decreased, E deceleration time decreased and atrial
contribution ( A ) increased significantly in diabetes,
while only A increased in the control group. A
marked increase in left atrial ejection force was seen
in diabetes only (P < 0.002). This difference was seen
in spite of comparable reductions in mitral area and
atrioventricular compliance in the two groups.
4. The hyperfunction of the left atrium in diabetes is
hypothesized to be due to reduced size of the left
ventricle combined
with incipient autonomic
neuropathy.
INTRODUCTION
Insulin-dependent diabetes mellitus is characterized by an increased tendency for heart failure after
myocardial infarction [1, 2]. In the presence of
preserved systolic function, this heart failure is considered mainly diastolic in nature [3]. The existence
of a specific diabetic heart disease unrelated to
ischaemic heart disease has been suggested [4-6],
and even before the appearance of ischaemic heart
disease, a myocardial dysfunction mainly localized
to diastole has been described [7-10]. Coexisting
ischaemic heart disease, hypertension or late
diabetic complications like neuropathy or nephropathy have, in many reports, obscured the pathogenic mechanism. If diabetes itself can be considered
responsible, investigations in patients with early
diabetes and in age groups without possible latent
ischaemic heart disease must be performed.
In the present study, the cold pressor test was
used to disclose subclinical changes in left ventricle
diastolic function in uncomplicated diabetes
mellitus.
MATERIAL AND METHODS
Seventeen patients with insulin-dependent diabetes
mellitus were recruited from our outpatient clinic
(aged 18.3 ± 2.8 years, diabetes duration 7-17 years).
The patients did not have hypertension, albumin
excretion was less than 200 mg/l and all had normal levels of serum creatinine. Four had microalbuminuria with excretion in the range 20-200 mg/l,
None of the patients had proliferative retinopathy,
but one had retinopathia simplex. Resting ECGs
were normal and none of the patients had cardiac
symptoms.
The patients were all on a multiple insulin rejection regime and received no other drugs. Six of the
patients smoked. The control group (n= 12) were
recruited from relatives and friends of the patients
as well as medical students. Two of the control
subjects smoked.
The work was carried out in accordance with the
Declaration of Helsinki and the protocol was
approved by the local ethics committee. Informed
consent was obtained from all participants.
Long-term metabolic status was judged from the
HbAlc concentration, while the short-term status
was evaluated by estimation of serum fructosamine
(normal range 2OQ-285Ilmoljl). The acute metabolic
status was estimated by blood glucose measurements before and after the examination. Autonomic
nervous system function was evaluated by ECG
recording during six cycles of deep breathing over
1min. The difference between maximum and mini-
Key words: cold pressor test, diabetes mellitus, diastolic function, echo-Dopplercardiography, heart failure.
Abbreviations: A, peak filling velocity during atrial contraction; AEF. atrial ejection force; E, peak early ventricular filling velocity; IVRT, isovolumic relaxation time.
Correspondence: Dr Ole G;tzsche, University Department of Cardiology, Arhus University Hospital, Tage Hansensgade, DK-8000 Arhus, Denmark.
O. G~tzsche et al.
462
mum heart rate during each breath was measured
and the mean was calculated.
Echocardiographic investigation was performed
by the same echocardiographer, with the patient in
a left lateral recumbent position, after 15 min of rest
using standard projections. Standard M-mode echocardiographic techniques and projections were used
according to the American Society of Echocardiography recommendations [II]. AH estimations were
based on the mean of at least four measurements in
end-expiration. Relevant parameters were indexed
by body surface area.
Left ventricular volumes were calculated from the
M-mode recordings by the method of Teichholz et
al. [12]. Left ventricular geometry was measured by
dividing the average posterior wall plus septal wall
thickness by left ventricular end-diastolic dimension
(T-index). Transmitral flow was measured from an
apical 4-chamber view with pulsed Doppler, with a
2 mm sample size positioned between the tips of the
mitral leaflets. Isovolumic relaxation time (IVRT)
was measured from pulsed Doppler curves with the
sample volume in an intermediate position between
left ventricular inflow and outflow tract using angle
correction. Estimation of left atrial size was performed by planimetry fronm an apical 4-chamber
view in end-diastole. Blood pressure was measured
by the cuff method (Dinamap).
Left atrial ejection force (AEF) was calculated as
the product of mass and acceleration of the blood
ejected from the left atrium into the ventricle. The
method was modified from earlier reports [13],
assuming the acceleration during initial transmitral
A-wave to be linear and time to peak to be half the
A-wave duration. The following equation was used:
Table I. Clinical data
Age (years)
Weight (kg)
Height (em)
Body surface area (m 2)
Diabetes duration (years)
Systolic blood pressure (mmHg)
Diastolic blood pressure (mmHg)
Heart rate (beats/min)
Hearl rate variation (beats/min)
Sex (F/M)
Serum HbAIc (%)
Serum fructosamine (Ilmol/l)
Diabetic patients
(n= 17)
Control subjects
(n= 12)
18.3 ±2.8
6U±6.6
175.0H.0
1.79±0.11
11.S±10
127 ± 12
18.8±19
64.2± 13.3
174.1 ±9.6
1.77 ±0.22
72±8
74±8*
13.3±5.1*
6/11
9.0± 1.0
412±79
122± 14
69±7
65± II
H.0±9.3
5/7
*p<0.01 compared with control group.
velocity, A-wave duration, left atrial area) were
measured with the patient lying supine. A cold
pressor test was then performed in which the left
hand was lowered into an ice-bath. The echoDoppler examination was performed 2 min after
cold exposure. The examiner was unaware of the
identity of the patients and control subjects. AH
echo-examinations were recorded on video and later
read blindly without knowledge of the diabetic
status of the heart using the software in the echocardiograph (Toshiba SSH-140A HG).
Statistics
where c is the density of blood (1.06 g/cm"), The
unit of AEF is indicated in kdyn. Atrioventricular
compliance was calculated from: mitral orifice area x
deceleration time of E-wave/c x peak E-velocity
[14]. The unit is cm 4s 2g- 1 .
Comparison between groups was performed by
unpaired r-test while evaluation of the effect of
intervention was made by paired t-test. In case of
inhomogeneity of variance, non-parametric statistics
were used (Mann-Whitney). Correlations were performed by standard regression analysis using
Pearson's analysis preceded by a test for equality of
variation and normal distribution. Alternatively,
Wilcoxon's signed-rank test was applied. A level of
P = 0.05, was considered significant. Values are given
as means ± SO.
Intraobserver variability
RESULTS
The variability of the estimations was calculated
by dividing the difference between two measurements 1h apart by the first measurement in the
same subject. Expressed as a percentage, the variability in left ventricular dimension was 1.6%, left
ventricular volume 3.7%, peak E-wave 5.4%, peak
A-wave 4.3%, E/A ratio 1.9% and deceleration
time 6.4%.
Clinical data are shown in Table I. The two
groups did not differ in terms of age, weight, height
or blood pressure. Heart rate was slightly but
significantly increased in diabetic patients compared
with control subjects and heart rate variation during
deep breathing decreased significantly, with six of
the 17 patients having values below 10 beats/min.
Left ventricular dimensions and stroke volume were
significantly decreased in diabetes and there was no
evidence for left ventricular hypertrophy in diabetes
(Table 2). None of the patients had a left ventricular
mass index above 134 g/m 2 • Doppler flow showed a
significantly increased flow during atrial contraction
(0.51 ±0.08 m/s compared with 0.43 ±0.07 m/s, P =
AEF = c x mitral orifice area
(2 x peak A-velocity/A-wave duration)
Intervention
After basal examination in a 45" left lateral
recumbent position, the diastolic parameters (IVRT,
peak E-velocity, E-deceleration time, peak A-
Left ventricular diastolic function in diabetes
Table2. Basal M-mode and Doppler echocardiography: estimations
in left lateral recumbent position. Abbreviations: A, peak filling velocity
during atrial contraction; E. peak early ventricular filling velocity; E dec.
duration of deceleration of early filling velocity. NS. not significant.
End-systolic dimension (mm)
End-systolic dimension index
(mm/m')
End-diastolic dimension (mm)
End-diastolic dimension index
(mm/m')
Shortening fraction (%)
Stroke volume (ml)
Posterior wall thickness (mm)
Thickness index
Left atrial area (em')
Left atrial area index (cm'/m')
IVRT (ms)
E (m/s)
A(m/s)
Edec (ms)
E/A
Diabetic patients
(n= 17)
Control subjects
(n= 12)
P
29.4±15
32.H2.6
0.02
16.H1.9
46.3±4.5
18.H 2.0
50.7 ±4.0
0.01
0.01
26.0 ± 2.4
36.7 ±4.0
66.6± 15.6
9.3± 1.1
0.19 ±0.03
15.7±2.1
8.79± 1.17
64±7
0.92±0.14
0.51 ±0.08
130±22
1.9±0.3
28.9±12
35.9±4.1
80.5 ± 17.9
9.3 ± 1.3
0.18±0.02
14.H11
8.22±1.32
67±5
0.8HO.13
0.43 ± 0.07
142±28
2.1 ±0.4
0.008
NS
0.04
NS
NS
NS
NS
NS
NS
0.01
NS
NS
0.01) in patients with diabetes, while IVRT and
peak early filling velocity, indices of early diastolic
function, were normal.
Cold pressor test induced the same increase in
systolic and diastolic blood pressure in both groups
(Table 3). The increase in heart rate was also
comparable. Significant changes with prolongation
of IVRT (59±5ms to 69±5ms, P=O.OOI), decrease
in peak early filling (0.91 ± 0.1 m/s to 0.87 ± 0.1 mis,
P = 0.03) and reduction in deceleration time of the
early filling (131 ± 19ms to 113± 29 ms, P= 0.003)
were seen only in patients with diabetes. In both
groups, A-wave duration shortened and peak Awave increased significantly. During the cold pressor
test, AEF increased in patients with diabetes
(5.79± 1.38kdyn to 7.17± 1.42kdyn, P = 0.002),
while no change was registered in the control group
(P=0.67) (Fig. 1). Sixteen out of 17 diabetic patients
increased their atrial function during this stress test
compared with only seven out of 12 in the control
group. Mitral ring area and atrial ventricular compliance were decreased to the same extent in the
two groups (Table 3 and Fig. 2). The change in
AEF showed no relation to autonomic neuropathy,
as judged by heart rate variation; to left ventricular
size or to metabolic status (results not shown).
DISCUSSION
Transmitral pulsed Doppler flow as a measure of
diastolic function is affected by many factors including preload and afterload, heart rate, left ventricular
compliance, mitral ring area, left atrial pressure and
atrial contractility. An early sign of diastolic dysfunction as seen in pressure overload and ischaemic
heart disease is a decreased early filling rate and an
increased atrial contribution to left ventricular filling, .c~lled. the relaxation abnormality [IS]. This
condition IS also characterized by a prolonged
IVRT and an increased deceleration time of the
early filling phase. In the presence of an increased
end-diastolic pressure in the left ventricle deceler~tion ti~e of the E-wave can decrease, 'together
With an increased peak E-velocity and a reduced
atrial contribution to left ventricular filling (restrictive filling pattern) [15].
This study has shown that in young patients with
insulin-dependent diabetes without hypertension or
nephropathy and whose age disfavours any possible
influence of ischaemic heart disease, such an abnormal left ventricular diastolic function can be
detected during a cold pressor test. A reduced size
of the left ventricle and an increased flow velocity
across the mitral valve during atrial contraction was
seen in the diabetic group during basal conditions.
Such a reduction in left ventricular size in diabetes
mellitus has previously been described in invasive
and non-invasive studies [16,17]. An inverted EIA
ratio has been described in adults with various
diabetic complications and in whom the presence of
ischaemic heart disease seems likely [7-10].
In the present study, the cold pressor test induced
the same increase in blood pressure and heart rate
in the two groups. Moreover, the non-invasive
estimate of atrioventricular compliance showed the
same reduction in diabetic patients and control
subjects during the cold pressor test. This suggests
that the left ventricle was subjected to the same
strain in the two groups. Differences consisted of
prolonged IVRT, reduced E-wave and decreased
deceleration time of the E-wave in patients with
diabetes. Such changes point to an inhibition of the
relaxation process during this stress test, but also
serve as an indirect indication of increased enddiastolic pressure in the left ventricle in diabetes.
Defects in the relaxation process in diabetes have
been described in experimental studies, with a diminished calcium-clearing capacity in the cytosol of
myocardial cells from streptozotocin diabetic rats
[18]. Other explanations might include scattered
differences in relaxation due to localized hypoxia
throughout the myocardium. This hypoxia could be
due to a reduced vasodilatory capacity of the
coronary arteries. The restrictive filling pattern
depicted by the reduced deceleration time could
reflect interstitial changes in the diabetic myocardium [19].
The increased AEF was seen in the patients with
diabetes and not in the control subjects. The mitral
ring area did not differ before or during the cold
pressor test in the two groups. The smaller size of
the left ventricle in the diabetic patients may still
have contributed to the excessive AEF because a
small ventricle with a relaxation abnormality
becomes critically dependent on the atrial contribution to filling. The precise mechanism behind the
regulation of ventricular filling and atrial contrac-
O. G~tzsche et al.
Table 3. Diastolic Doppler echocardiographic parameters before and during cold pressor test in diabetic patients and
control subjects. Abbreviations: A, peak filling velocity during atrial contraction; E. peak early ventricular filling velocity; E dec,
duration of deceleration of early filling velocity.
Diabetic patients
Mitral area (em')
A-wave (rn.s)
Aduration (ms)
Left atrial area (em')
E-wave (m/s)
E dec (ms)
IVRT (ms)
Heart rate (beats/min)
Systolic blood pressure (mmHg)
Diastolic blood pressure (mmHg)
Diabetic patients
(P = 0.002)
Before test
During test
9.08± 1.43
0.47 ± 0.07
ISH 17
15.6±2.1
0.91 ±O.IO
131 ± 19
59±5
70±9
124± 12
73±7
783± 1.16
0.55 ±0.09
127 ± 16
15.2±2.1
0.87 ±OIO
113±29
69± 5
77 ±9
147 ± 16
89± 12
Control subjects
(NS)
l /t
l
Basal
Cold pressor
test
Basal Cold pressor
test
Fig. I. Atrial ejection force (AEF) before and after cold pressor test
in diabetic and control groups
2.5
Diabetic patients
(P =0.001)
Control subjects
(P =0.001)
~ 2.0
T
I
<:>
e
1.5
.~
Q.
E
8 1.0
M
B
~
.
.g
0.001
0.001
0.001
0.26
0.03
0.003
0.001
0.001
0.001
0.001
During test
9.27 ± 1.70
0.43 ± 0.07
153± 14
16.0 ± 2.7
0.88± 1.13
134±22
59±6
61 ±9
119± 12
70±7
7.44± 1.26
0.48 ± 0.07
133± 17
15.9±2.2
0.8HO.09
121 ±28
61 ±6
66± 10
139± 15
84± 10
0.001
0.04
0.003
0.44
059
0.11
0.48
0.001
0.001
0.001
tion is not known but the autonomic nervous
system may be involved. The function test comprising beat-to-beat variation during deep breathing did
show early signs of autonomic dysfunction in the
diabetic group, suggesting a compromised vagal
tone.
Neither long-term nor short-term metabolic
status, judged from the HbAlc concentration and
the serum level of fructosamine, showed any correlation to the described changes. Nonetheless, a slight
reduction in the E-wave has previously been
reported in diabetic children in a non-blind study
[20]. The blood glucose level was considerably
higher in that study compared with our patients.
In conclusion, this study has shown signs of
relaxation abnormality as wel1 as restriction in the
left ventricle of young patients with uncomplicated
insulin-dependent diabetes exposed to stress induced
by the cold pressor test. The smal1er size of the
ventricle and incipient diabetic autonomic neuropathy may explain this phenomenon, which is considered to reflect metabolically induced dysfunction
of the myocardial cel1s as wel1 as structural changes
in the myocardial insterstitium and vasculature in
diabetes mel1itus.
REFERENCES
g
~
Control subjects
Before test
tt
0.5
.0:
Basal Cold pressor
test
Basal Cold pressor
test
Fig. 2. Atrioventricular compliance before and after cold pressor
test in diabetic and control groups
I. Stone PH, Muller JE. Hartwell T, et al. The effect of diabetes mellitus on
prognosis and serial left ventricular function after acute myocardial infarction:
contribution of both coronary disease and diastolic left ventricular dysfunction
to the adverse prognosis. J Am Coli Cardiol 1989: 14! 49--57
2. Jaffe AS. Spadaro JJ. Schechtman K, Roberts R, Geltman EM. Sobel BE.
Increased congestive heart failure after myocardial infarction of modest extent
in patients with diabetes mellitus. Am Hear J 1988; lOB! 31-7.
3. Savage I MP, Krolewski AS, Kenien GG, Lebeis MP, Christlieb R, Lewis SM.
Acute myocardial infarction in diabetes mellitus and significance of congestive
heart failure as a prognostic factor. Am J Cardiol 1988; 62! 665-9.
4. Ledet T, Neubauer B, Christensen NJ, Lundbaek K. Diabetic cardiography.
Diabetologia 1979; 16: 207-9.
S. Fein FS, Sonnenblick EH. Diabetic cardiomyopathy. Prog Cardiovasc Dis 1985;
27: 255-70.
6. G~tzsche O. Myocardial cell dysfunction in diabetes mellitus. A review of
clinical and experimental studies. Diabetes 1986; 35: 1158-62.
7. Kahn JK, Zola B, Juni JE, Vinik AI. Radionuclide assessment of left ventricular
diastolic filling in diabetes mellitus with and without cardiac autonomic
neuropathy. J Am Call Cardiol 1986; 7: 1303-9.
Left ventricular diastolic function in diabetes
8. Sampson MJ, Chambers JB, Sprigings DC, Drury PL. Abnormal diastolic
function in patients with type I diabetes and early nephropathy. Br Heart J
1990; 64: 266-71.
9. Paillole C, Dahan M, Paycha F, Solal AC, Passa P, Gourgon R. Prevalence and
significance of left ventricular filling abnormalities determined by Doppler
echocardiography in young type I (insulin dependent) diabetic patients.
Am J Cardiol 1989; 64: 101l}-16.
10. Perez JE, McGill JB, Santiago JV, et al. Abnormal myocardial acoustic
properties in diabetic patients and their correlations with the severity of the
disease. J Am Coli Cardiol 1992; 19: 115+62.
II. Feigenbaum H. Echocardiography, 4th edn. Philadelphia: Lea and Febiger, 1986:
621-39.
12. Teichholz LE, Kruelen TH, Herman MV, Gorlin R. Problems in
echocardiographic volume determinations: echo-angiographic correlations in the
presence and absence of asynergy. Am J Cardiol 1976; 37: 7-12.
13. Manning WJ, Silverman DI, Katz SE, Douglas PS. Atrial ejection force: a
non-invasive assessment of atrial systolic function. J Am Coli Cardiol 1993; 22:
221-5.
14. Flaschkampf FA, Weyman AE, Guerrero JL, Thomas JD. Calculation of
atrioventricular compliance from mitral flow profile: analytic and in vitro study.
465
J Am Coli Cardiol 1992; 19: 99S-1004.
15. Appleton CP, Hatle LK, Popp RL. Relation of transmitral flow velocity
patterns to left ventricular diastolic function: new insights from a combined
hemodynamic and Doppler echocardiographic study. J Am Coli Cardiol 1988;
12: 426-40.
16. Regan TJ, Lyons MM, Ahmed S, et al. Evidence for cardiomyopathy in familial
diabetes mellitus. J Clin Invest 1977; 60: 885-99.
17. G,tzsche 0, S,rensen K, Mcintyre B, Henningsen P. Reduced left
ventricular afterload and increased contractility in children with
insulin dependent diabetes mellitus. Pediatr Cardiol 1991; 12:
69-73.
18. Penpargkul S, Fein FS, Sonnenblick EH, Scheuer J. Depressed cardiac
sarcoplasmic reticular function from diabetic rats. J Mol Cell Cardiol 1981; 13:
303-9.
19. Regan TJ, Ettinger PO, Khan MI, et aI. Altered myocardial function and
metabolism in chronic diabetes mellitus without ischemia in dogs. Circ Res
1974; 35: 222-37.
20. Riggs TW, Transue D. Doppler echocardiographic evaluation of left ventricular
diastolic function in adolescents with diabetes mellitus. Am J Cardiol 1990; 65:
899-902.