European Heart Journal (1999) 20, 789795

Article No. euhj.1998.1472, available online at http://www.idealibrary.com on

Clinical Perspective

Diabetes mellitus and congestive heart failure

Further knowledge needed

Introduction The objective of this review is to summarize epi-

demiological, prognostic and therapeutic knowledge
Diabetes mellitus, and in particular the type 2[1], has related to congestive heart failure and diabetes.
progressively become more common. Factors increas-
ing the prevalence include an ageing population, an
increasing body mass and decreased demands of
Epidemiology
physical activity[2]. In type 2 diabetes manifestations
of atherosclerosis are frequently present at the time of The Framingham study was the first epidemiological
diagnosis. Approximately 20% of patients admitted study to demonstrate an increased risk of congestive
to Swedish coronary care units for myocardial infarc- heart failure in patients with diabetes mellitus. Com-
tion have diabetes. In a recent health survey, 22% of pared with non-diabetic males and females, the esti-
diabetic patients reported that they needed to see a mated increase in the incidences of heart failure for
cardiologist during the previous 12 months and up to young diabetic males and females were fourfold and
50% had cardiovascular disease[3]. As type 2 diabetes, eightfold, respectively[7].
including the pre-diabetic period, is an important risk In a retrospective survey of patients in need of
factor for atherosclerosis the increasing prevalence hospitalization for congestive heart failure in western
suggests that there will be a considerable increase in Sweden 10% had diabetes mellitus. Since individuals
diabetes-related cardiovascular disease in the near above the age of 65 years were excluded, and only
future[2]. patients on insulin were classified as having diabetes,
Ischaemic heart disease is still the leading cause of this number is an underestimate of the true pro-
congestive heart failure despite advances in its pre- portion[8]. The large ACE inhibitor clinical heart
vention and treatment. Heart failure has increased failure trials offer somewhat less age-restricted in-
in prevalence while there has been a decrease in the formation. The proportion of subjects with diabetes
age-adjusted morbidity and mortality of coronary was 23% in CONSENSUS (Cooperative North
artery disease. An ageing population and increased Scandinavian Enalapril Survival Study)[9], 25% in
survival of patients with myocardial infarction are SOLVD (Studies of Left Ventricular Dysfunction)[10],
likely explanations[4]. Thus, diabetes mellitus is 20% in V-HeFT II (Vasodilator Heart Failure Trial
closely linked to congestive heart failure. Proper II)[11] and 20% in ATLAS (Assessment of Treat-
treatment, including meticulous metabolic control of ment with Lisinopril and Survival)[12]. As studies are
the diabetes, may considerably improve the progno- always carried out on selected populations these
sis for diabetic patients with myocardial infarction[5] figures also need to be interpreted with caution. Ten
and also possibly prevent its occurrence[6]. Never- percent of the patients in NETWORK, a trial
theless, even with the best preventive strategies and planned to be representative of a complete heart
expert treatment of established cardiovascular failure population, had diabetes[13]. In the
disease, a considerable proportion of heart failure RESOLVD (Randomized Evaluation of Strategies
patients will have diabetes. Further research is for Left Ventricular Dysfunction pilot trial) study[14]
needed for the proper handling of these subjects. the prevalence of diabetes was 27% at the time of
This may contribute not only to patient relief but randomization. In RESOLVD blood glucose was
also to a decrease of future demands on health care measured at baseline. Application of the most recent
resources. diagnostic criteria for diabetes[1] resulted in a
prevalence of 35%. The observed variability between
Received 8 December 1998, and accepted 16 December 1998. different studies is due to several factors of which age,
Correspondence: Lars Ryden, MD, FESC, Department of aetiology, severity of heart failure and the definition
Cardiology, Karolinska Sjukhuset, S-171 76 Stockholm, Sweden. of diabetes mellitus are the most important.

0195-668X/99/110789+07 $18.00/0  1999 The European Society of Cardiology

790 Clinical Perspective

Congestive heart failure is a frequent reason for Heart failure mechanisms in diabetics
hospital admission[15]. Reis et al.[16], evaluated
specialty-related differences in the care and outcome Diabetic cardiomyopathy
of patients admitted to hospital for heart failure.
They noted that as many as 38% of all patients Following the original proposal by the Danish
had diabetes mellitus requiring pharmacological internist Lundback[25] the increased susceptibility of
treatment. Recent Italian cross-sectional data show a diabetic patients to heart failure has often been
30% prevalence of diabetes in an elderly heart attributed to a diabetes-specific myocardial disease
failure population. The association with diabetes was referred to as diabetic cardiomyopathy. Although
independent of age, sex, blood pressure, body mass the most common cause of death in diabetic patients
index or waist/hip ratio and also of a family history of is not cardiomyopathy but coronary artery disease,
diabetes. The incidence of diabetes was 29% during heart failure is more frequent in diabetic than in
3 years of follow-up among heart failure patients non-diabetic patients with myocardial ischaemic
initially free from this disease compared with 18% in injury. This does not seem to be due to more
a group of matched controls. On a basis of multi- extensive myocardial damage because many reports
variate statistics, congestive heart failure indepen- on the subject show that infarct size is no larger in
dently predicted subsequent type 2 diabetes. A poss- diabetic than non-diabetic patients[20].
ible explanation is that an increased adrenergic drive,
caused by heart failure, increases free fatty acid oxi- Morphology
dation and insulin resistance thereby decreasing Numerous investigations have been devoted to
glucose oxidation and precipitating type 2 diabetes[17]. morphological alterations in the diabetic heart, as
In a recent, population-based study of elderly recently discussed by Hardin in an extensive
patients it was concluded that diabetes mellitus is an review[26]. It may be concluded that the most
independent risk factor for heart failure and that the consistent findings are myocyte hypertrophy, inter-
risk increases with disease severity. Furthermore, stitial fibrosis, increased PAS-positive material and
after multivariate adjustment, a 1% increase in intramyocardial microangiopathy. There are no
the baseline level of HbA1c increased the risk of lesions characteristic for diabetes only, indicating that
developing heart failure by 15% in patients with and the reason for a diabetic cardiomyopathy may be
without known diabetes. This indicates that the found at a functional or biochemical level. There
independent risk for developing heart failure in dia- appears to be a synergism with structural changes
betic patients may to some extent be mediated by usually attributed to hypertension. This may have
poor metabolic control[18]. important treatment implications in the light of the
In summary, there is consistent epidemiological favourable effect of antihypertensive therapy in
evidence that diabetes mellitus is common in a heart diabetic patients noted in the HOT[27] and UKPDS[28]
failure population. Furthermore, diabetes and heart trials.
failure may be interrelated.
Diastolic dysfunction
The common response in a non-infarcted myocardial
Prognosis
area subjected to acute ischaemia is a compensatory
Diabetes is an independent risk factor for the devel- hyperkinaesia that may almost normalize the ejection
opment of congestive heart failure and heart failure fraction[29,30]. In GUSTO 1 (Global Utilization of
patients with diabetes have a worse prognosis than Streptokinase and t-PA for Occluded coronary
those without diabetes. In the SOLVD study diabetes arteries-1), with more than 300 diabetic subjects,
was an independent predictor of morbidity and mor- coronary angiograms performed 90 min after throm-
tality both in symptomatic and asymptomatic heart bolysis did not reveal any difference in the global
failure[19]. This relationship was confirmed by the ejection fraction between diabetic and non-diabetic
RESOLVD trial[14]. Diabetic patients also experience patients. The normal compensatory hyperkinetic
a higher incidence of heart failure and increased response in the non-infarcted area was, however,
mortality after acute myocardial infarction than non- blunted among diabetic patients[31]. A decreased re-
diabetic patients[2023]. In the DIGAMI (Diabetes gional ejection fraction in non-infarcted myocardial
mellitus Insulin-Glucose infusion in Acute Myocar- areas of diabetic patients has been reported[32,33].
dial Infarction) study, cohort heart failure was the During follow-up in the GUSTO trial, congestive
most common reason for morbidity and mortality heart failure was almost twice as common in the
accounting for 66% of the total mortality during the diabetic as in the non-diabetic cohort[31]. This is
first year of follow-up[24]. congruent with a report by Stone et al.[20], who noted

Eur Heart J, Vol. 20, issue 11, June 1999

 Clinical Perspective 791

a higher incidence of heart failure among diabetic with adenosine triphosphate dependent ion pumps,
patients despite smaller infarct sizes and ejection and an increased alpha-1-response causing mobiliz-
fractions which were similar to those in subjects ation of intracellular calcium creating calcium over-
without diabetes. These findings probably reflect load and contractile dysfunction. In addition to the
impaired diastolic function, a finding that seems to be effects of insulin deficiency, the increased free fatty
the most characteristic feature of diabetes-associated acids inhibit both glucose transport and metabolism.
myocardial disease[3436]. Increased levels of citrate, produced by free fatty
Several studies verify that coronary artery disease, acid oxidation, inhibit phosphofructokinase. This
even in its asymptomatic form, is more common and leads to decreased glycolysis and promotes glycogen
widespread among diabetics than non-diabetics. This synthesis. Impaired glucose oxidation also leads to
may explain the lack of a compensatory hyperkinetic lactic acid accumulation which further promotes the
response to ischaemia and the diastolic impairment. degradation of free fatty acids.
Diastolic dysfunction is an early sign of myocardial In summary, a diabetes-related myocardial
ischaemia[37]. Most studies that suggested a diabetic dysfunction a diabetic cardiomyopathy exists
cardiomyopathy did not angiographically exclude co- and is of clinical significance. It is characterized by
existent coronary artery disease. This differentiation a lack of compensatory response to myocardial
is a requirement in future studies. ischaemia or injury and includes early impairment of
the diastolic function. The pathophysiological mech-
Myocardial blood flow anisms, although not fully understood, are multi-
Another reason for a compromised myocardial blood factorial and include metabolic and vascular com-
flow, or an inability to increase this flow when ponents. This suggests that interventions against
demanded, relates to impaired endothelial dependent hyperglycaemia and increased free fatty acid oxi-
vasodilatation. The mechanisms behind endothelial dation, e.g. strict use of insulin, may be beneficial.
dysfunction in the diabetic patient are not fully Moreover, there seems to be a synergism between
understood. This dysfunction has been verified both diabetes and hypertension in the development of
in type 1[38,39] and type 2 diabetes[40,41]. Diabetic structural myocardial changes. This may explain why
patients have a reduced myocardial flow reserve com- vigorous treatment of hypertension is of particular
pared with matched controls even in the absence of value in the diabetic patient.
obvious heart disease. Acute hyperglycaemia may
impair endothelial-derived vasodilatation in healthy
humans[42]. The inability to increase myocardial Autonomic dysfunction
blood flow is independently related to long-term
Cardiac autonomic imbalance is a common
control of blood glucose, however, and not to age,
consequence of diabetes. One effect is a decreased or
blood pressure or blood lipids[43]. Accordingly it may
even an eliminated perception of ischaemic pain.
be assumed that elevated blood glucose by itself is of
Silent ischaemia may cause myocardial injury without
considerable importance for the impaired vascular
clinical signs promoting future heart failure[47,48].
response. This may contribute to the lack of hyper-
Even more important may be the effects of a
kinetic response and the diastolic dysfunction seen in
decreased vagal tone. Diabetic patients with dis-
diabetes mellitus. It may serve as a rationale for
turbed autonomic function have a higher heart
treatment aimed at strict glucose control to reduce
rate than non-diabetic patients. This relates to pre-
cardiovascular events in a diabetic population.
dominant parasympathetic dysfunction preceding
the involvement of the sympathetic system[49]. Tachy-
Metabolic aspects
cardia increases myocardial oxygen demand con-
Metabolic factors may be of fundamental importance
comitantly with a decrease in the time for myocardial
in the development of myocardial dysfunction un-
blood flow due to the shortened diastole. Decreased
related to macrovascular disease in patients with
heart rate variability due to impaired vagal tone is
diabetes, as has been excellently reviewed by
another factor of prognostic importance. When
Rodrigues et al.[4446]. In addition to incidence of
present it is linked to an increased risk for sudden
hyperglycaemia, diabetes is characterized by an
cardiac death[50].
increased turnover of free fatty acids. The increased
free fatty acids provoke increased use of myocardial
oxygen and enhance intracellular accumulation of Treatment of heart failure in diabetics
intermediates leading to several untoward effects.
These effects include promotion of intracardiac con- Treatment of symptomatic heart failure in the dia-
duction disturbances and arrhythmias, interference betic patient follows principles extensively outlined in

Eur Heart J, Vol. 20, issue 11, June 1999

792 Clinical Perspective

the European Society of Cardiology guidelines on the The ATLAS trial[12] compared high (32535 mg)
treatment of congestive heart failure[51]. Few, if any, and low (2550 mg) doses of the ACE inhibitor
studies have addressed in any detail the therapeutic lisinopril over a median period of 45 months in heart
efficacy of conventional treatment in a diabetic failure patients of NYHA classes IIIV. The total
population. patient cohort was 3164 patients of whom 611 were
diabetics. The ATLAS trial therefore reports on the
Diuretics largest diabetic heart failure subgroup presently avail-
able. Mortality was considerably higher within the
Diuretics are mandatory for symptomatic treatment diabetic subgroup than in the non-diabetics. The
of heart failure. Whether the use of diuretics will mortality risk reduction in studying the efficacy of the
improve or worsen the prognosis of diabetic patients high and low ACE inhibitor dose strategy was 6% for
is not known. In the absence of studies in diabetic non-diabetics compared with 14% for the diabetic
subjects with heart failure the only available informa- subgroup and emphasizes the need for appropriate
tion is derived from antihypertensive trials. Warram doses of ACE inhibitors when treating diabetic as
et al.[52] claimed that diuretic-based antihypertensive well as non-diabetic patients.
treatment of diabetic patients was associated with Post-myocardial infarction patients with com-
excess mortality; the treatment only related to years promised left ventricular function, defined as left
of treatment, however, not to the severity of con- ventricular ejection fraction c40%, were studied in
comitant nephropathy. The type of diuretic was not SAVE (Survival and Ventricular Enlargement)[55,56].
specified. In contrast, the SHEP (Systolic Hyper- According to a subgroup analysis the diabetic cohort
tension in the Elderly Program) study reported that had a higher morbidity and total mortality than the
low-dose diuretic (saluretics) based antihypertensive non-diabetic group. Treatment with the ACE inhibi-
therapy effectively prevented major cardiovascular tor captopril improved this unfavourable outcome to
events, including mortality, in patients with type 2 an extent similar to that among non-diabetic patients.
diabetes[53]. Although no studies have specifically GISSI 3 (Gruppo Italiano per lo Studio della Sopra-
looked into the outcome of the use of diuretics in a vvivenza nell Infarto miocardio[57]) reported on the
heart failure population, loop diuretics are recom- effect of an ACE inhibitor administered as soon as
mended rather than diuretics that may further impair possible after onset of symptoms indicating myo-
the gluco-metabolic state. cardial infarction. The patients were randomized to
treatment with oral lisinopril or to serve as open
ACE inhibitors controls. After 6 weeks the mortality in the high-risk
diabetic population was 9% in the lisinopril group
The use of ACE inhibitors is a cornerstone in the compared with 12% in the control group. Non-
treatment of congestive heart failure since the land- diabetic subjects were at lower overall risk and did
mark CONSENSUS study[9]. As already emphasized, not further benefit from lisinopril (mortality 6% in
diabetic patients represent a fairly large subgroup in both groups). The beneficial effect of lisinopril
the patient cohorts in long-term trials of several ACE remained at the 6 months follow-up. The reason for
inhibitors, from 10% in NETWORK[13] to 25% in the discrepancy in outcome in this trial compared
SOLVD[10]. Subgroup analysis of diabetic mortality with other ACE inhibitor heart failure trials, in
data from those studies reveal that mortality, as could showing a greater similarity in outcome among non
be expected, is higher within the diabetic cohort than diabetic and diabetic patients, is not clear.
among non-diabetic patients. Hypoglycaemic events have been reported to
In the SOLVD prevention trial[54], placebo or increase following the institution of ACE inhibitors in
enalapril (2520 mg; mean 127 mg) was given to diabetic patients. As some ACE inhibitors have been
patients with compromised left ventricular function shown to decrease insulin resistance, it is recom-
(EF c35%) but without signs of heart failure during mended that blood glucose is carefully monitored in
an average of 374 months. The SOLVD treatment the early phase following the institution of an ACE
study recruited patients with a left ventricular ejection inhibitor in patients on antihyperglycaemic
fraction c35% and signs of heart failure. They were drugs[58,59].
either given placebo or enalapril (2520 mg; mean In conclusion, ACE inhibitors are of value in the
118 mg) over an average follow-up period of 41 treatment of diabetic patients with congestive heart
months. The efficacy of the ACE inhibitor was some- failure. Perhaps the relative efficacy is more apparent
what more marked for diabetic than non-diabetic in this subgroup than among non-diabetic patients,
patients in SOLVD prevention and of approximately fitting into the general knowledge that patients at
similar efficacy in the treatment arm of SOLVD[10]. high risk benefit the most. There are data supporting

Eur Heart J, Vol. 20, issue 11, June 1999

 Clinical Perspective 793

the use of a high-dose ACE inhibitor strategy. Precise decreased. In the DIGAMI study, an insulin-glucose
knowledge is still lacking on the effect of ACE infusion followed by multidose subcutaneous insulin
inhibitors in diabetic patients. Data presented are treatment considerably improved the long-term prog-
derived from subgroup analyses of clinical trials in nosis of diabetics with myocardial infarction[5,24].
which the diabetic population and their anti-diabetic Since a similar concept has not yet been tested in
therapy were poorly defined. The latter may very well heart failure patients, it is not known whether
influence the effect of other therapeutic measures. improved metabolic care would favourably influence
the efficacy of conventional therapy in diabetics with
heart failure or whether it would be of preventative
Beta-blockers value against the development of heart failure in
diabetics with cardiac disease. Studies of the impact
It has been speculated that the particularly beneficial of strict metabolic care in patients with congestive
effect of early administration of beta-blockers heart failure are urgently needed.
following myocardial infarction relates to a reduced
accumulation of free fatty acids and improved myo-
cardial glucose utilization[60]. This would be of par- Future perspectives
ticular value for the diabetic patient in whom glucose
oxidation is already impaired. Another possible This review is an attempt to summarize the available
beneficial effect of beta-blockade is a reduction of the information on heart failure in the diabetic subject
increased heart rate that is caused by a disturbed and to identify areas in need of further studies. Many
autonomic tone. Both these mechanism would offer overviews have been published regarding cardio-
relief not only in the post-myocardial infarction phase vascular disease in diabetics. Few if any have
but in subjects with heart failure. Evidence support- addressed heart failure despite the fact that diabetic
ing this hypothesis has been presented as subgroup patients are at a particularly high risk of congestive
analysis of the effect of the beta-blocker carvedilol in heart failure, and that the prognosis is worse than the
diabetics with congestive heart failure[61]. already unfavourable outlook for non-diabetic heart
There are no studies that specifically address the failure patients. Identified reasons for this dismal
subject of beta-blockade in diabetic subjects with prognosis include the poor outcome of myocardial
heart failure. Such a use is currently advocated. infarction, the possible existence of a specific diabetic
Prospective stratification of diabetic patients in large cardiomyopathy, the severity and distribution of
clinical trials and documentation of their anti- coronary artery disease and disturbed autonomic
diabetic treatment would yield important insight into function. To date a comprehensive approach has been
the value of beta-blockade in this group of heart lacking. It is therefore difficult to know whether there
failure patients. are any strategies that may prevent or at least post-
pone the development of heart failure among diabetic
patients. It seems that such a strategy should com-
Metabolic control prise a multitude of activities including aggressive
normalization of blood glucose with insulin, beta-
There are several reasons to assume that the prog- blockade and ACE inhibition. Since there is an
nosis of patients with the combination of heart failure apparent lack of information on quality of life in the
and diabetes mellitus would improve with a meticu- diabetic patient with congestive heart failure, future
lous metabolic control. For example, the proposed trials should include such aspects. Revascularization
harmful effect of an increased free fatty acid oxi- strategies and studies of ways to improve the outcome
dation rate and decreased glucose utilization could after coronary interventions in diabetic patients are
both be attenuated by intense insulin treatment. also needed. The diabetic subject in need of coronary
Bersin et al.[62] applied a metabolic concept in non- angioplasty or a bypass procedure has a worse prog-
diabetic patients with severe heart failure in using the nosis than non-diabetic patients. This includes a
drug dichloroacetate. This compound stimulates higher risk for restenosis after coronary angioplasty
pyruvate dehydrogenase activity thereby facilitating and of myocardial injury after any of these
glucose oxidation and, in parallel, inhibiting free fatty interventions[63,64].
acid metabolism. Myocardial extraction of lactate Finally, it is clear that compromised left ventricular
increased during dichloroacetate infusion. At the function may precede the development of congestive
same time there was an improved forward stroke heart failure by a considerable period of time. The
volume and the left ventricular minute work implication is that more aggressive metabolic care at
increased, although myocardial oxygen consumption an early stage of the disease may delay restructuring

Eur Heart J, Vol. 20, issue 11, June 1999

794 Clinical Perspective

of the myocardium and atheromatosis. Improved [10] The SOLVD investigators. Effect of enalapril on survival in
patients with reduced left ventricular ejection fraction and
metabolic care may also decrease the thrombo- congestive heart failure. N Engl J Med 1991; 325: 293302.
genicity that characterizes diabetes. Currently, [11] Cohn JN, Johnson G, Ziesche S et al. A comparison of
therapeutic efforts are frequently initiated at a late enalapril with hydralazine-isosorbide dinitrate in the treat-
stage of the disease, a stage during which only a ment of chronic congestive heart failure. N Engl J Med 1991;
325: 30310.
modest symptomatic improvement and prolongation [12] Packer M, Poole-Wilson PA, Armstrong PW et al. on behalf
of longevity may be obtained. The patient should be of the ATLAS Study Group. Comparative effects of low doses
identified at an earlier stage of the disease. Combined and high doses of the angiotensin converting enzyme inhibitor
lisinopril on morbidity and mortality in chronic heart failure.
efforts of cardiologists and diabetologists are needed Submitted, 1998.
in the screening of risk patients and in preventative [13] The NETWORK investigators. Clinical outcome with
measures. Improved knowledge among diabetologists enalapril in symptomatic chronic heart failure; a dose
comparison. Eur Heart J 1998; 19: 4819.
regarding treatment and prevention of cardiovascular [14] Suskin N, McKelvie RS, Roteaus J, Sigouin C, Wiece KE,
complications and among cardiologists regarding Yususf S. Increased insulin and glucose levels in heart failure.
diabetology is a prerequisite for progress. Consider- J Am Coll Cardiol 1998; 3 (Suppl): 249A.
[15] Ho KK, Pinsky JL, Kannel WB, Levy D. The epidemiology of
ing the large and rapidly increasing number of heart failure: the Framingham study. J Am Coll Cardiol 1993;
patients at risk it is certainly urgent to get started. 22 (Suppl A): 613.
The high costs associated with the management of [16] Reis SE, Holubkov R, Edmundowicz D et al. Treatment of
patients admitted to hospital with congestive heart failure:
diabetic patients with heart failure suggest that speciality-related discrepancies in practice patterns and
improved therapy is likely to be cost effective as was outcomes. J Am Coll Cardiol 1997; 30: 7338.
demonstrated, for example, in the use of aggressive [17] Amato L, Paolisso G, Cacciatore F et al. on behalf of the
metabolic care for diabetic patients with myocardial osservatorio geriatrico regione campania group. Congestive
heart failure predicts the development of non-insulin depen-
infarction[65]. dent diabetes mellitus in the elderly. Diabetes Metab 1997; 23:
L. SOLA} NG 2138.
K. MALMBERG [18] Chae CU, Glynn RJ, Manson JE, Guralnik JM, Taylor JO,
Pfeffer MA. Diabetes predicts congestive heart failure risk in
L. RYDE uN the elderly. Circulation 1998; 98 (Suppl I): 721.
Department of Cardiology, [19] Schindler DM, Kostis JB, Yusuf S et al. for the SOLVD
Karolinska Sjukhuset, investigators. Diabetes mellitus: a predictor of morbidity and
mortality in the studies of left ventricular dysfunction
Stockholm, Sweden (SOLVD) trials and registry. Am J Cardiol 1996; 77: 101720.
[20] Stone P, Muller J, Hartwell T et al. and the MILIS study
group. The effect of diabetes mellitus on prognosis and serial
left ventricular function after acute myocardial infarction:
References contribution of both coronary disease and diastolic left ven-
tricular dysfunction to the adverse prognosis. J Am Coll
[1] American Diabetes Association: Report of the expert com- Cardiol 1989; 14: 4957.
mittee on the diagnosis and classification of diabetes mellitus. [21] Gwilt DJ, Petri M, Lewis PW, Nattrass M, Pentecost BL.
Diabetes Care 1997; 20: 118397. Myocardial infarct size and mortality in diabetic patients. Br
[2] Clark CM, Perry RC. Type 2 diabetes and macro vascular Heart J 1985; 54: 46672.
disease. Epidemiology and etiology. Am Heart J 1999 (In
[22] Singer DE, Moulton AW, Nathan DM. Diabetic myocardial
press). infarction. Interaction of diabetes with other preinfarction
[3] Wingard DL, Barett-Connor E. Heart disease in diabetes. In: risk factors. Diabetes 1989; 38: 3507.
Harris MI, Cowie CC, Stern M, Boyko E, Reiber G, Bennet
[23] Melchior T, Gadsboll P, Hildebrandt P, Kober L,
PH, eds. Diabetes in America, 2nd edn. Washington DC: U.S.
Torp-Pedersen C. Clinical characteristics, left and right
Government Printing Office, 1995: 42948.
ventricular ejection fraction, and longterm prognosis in
[4] The Task Force of the European Society of Cardiology.
patients with non-insulin-dependent diabetes surviving and
Guidelines for the diagnosis of heart failure. Eur Heart J 1995;
acute myocardial infarction. Diabetic Med 1996; 13: 4506.
16: 74151.
[5] Malmberg K, Ryden L, Efendic S et al. on behalf of the [24] Malmberg K, Ryden L, Hamsten A, Herlitz J, Waldenstrom
DIGAMI study group. A randomised trial of insulin-glucose A, Wedel H on behalf of the DIGAMI study group. Effects of
infusion followed by subcutaneous insulin treatment in insulin treatment on cause-specific one-year mortality and
morbidity in diabetic patients with acute myocardial
diabetic patients with acute myocardial infarction (DIGAMI
study): effects on mortality at 1 year. J Am Coll Cardiol 1995; infarction. Eur Heart J 1996; 17: 133744.
26: 5765. [25] Lundback K. Diabetic angiopathy: a specific vascular disease.
[6] U.K. Prospective Diabetes Study Group. Intense blood- Lancet 1954; 2: 3779.
glucose control with sulphonylureas or insulin compared with [26] Hardin N. The myocardial and vascular pathology of diabetic
conventional treatment and risk of complications in patients cardiomyopathy. Coron Art Disease 1996; 7: 99108.
with type 2 diabetes (UKPDS 33). Lancet 1998; 352: 83753. [27] Hansson L, Zanchetti A, Carruthers SG, Dahlof B, Elmfeldt
[7] Kannel WB, McGee DL. Diabetes and cardiovascular disease. D, Julius S.Effect of intensive blood-pressure lowering and
The Framingham study. J Am Med Assoc 1979; 214: 20358. low-dose aspirin in patients with hypertension: principal
[8] Andersson B, Waagstein F. Spectrum and outcome of conges- results of the hypertension optimal treatment (HOT)
tive heart failure in a hospitalised population. Am Heart J randomised trial. Lancet 1998; 351: 175562.
1993; 126: 63240. [28] U.K. Prospective Diabetes Study Group. Tight blood pressure
[9] The CONSENSUS trial study group. Effect of enalapril on control and risk of macrovascular and microvascular compli-
mortality in severe congestive heart failure. N Engl J Med cations in type 2 diabetes: UKPDS 38. Br Med J 1998; 317:
1987; 316: 142935. 70313.

Eur Heart J, Vol. 20, issue 11, June 1999

 Clinical Perspective 795

[29] Grines CL, Topol EJ, Califf RM et al. Prognostic implications [50] Klieger R, Miller J, Bigger J, Moss A. Decreased heart rate
and predictors of enhanced regional wall motion of the variability and its association with increased mortality after
noninfarct zone after thrombolysis and angioplasty therapy acute myocardial infarction. Am J Cardiol 1987; 59: 25662.
of acute myocardial infarction. The Tama study group. [51] The Task Force of the European Society of Cardiology.
Circulation 1989; 80: 24553. Guidelines for the treatment of heart failure. Eur Heart J
[30] Jaarsma W, Visser CA, van Eenige MJ et al. Prognostic 1997; 18: 73653.
implications of regionl hyperkinesia and remote asynergi of [52] Warram JH, Laffel LMB, Valsania P, Christlieb AR,
noninfarcted myocardium. Am J Cardiol 1986; 58: 3948. Krolewski AS. Excess mortality associated with diuretic
[31] Woodfield SL, Lundergran CF, Reiner JS et al. Angiographic therapy in diabetes mellitus. Arch Intern Med 1991; 151:
findings and outcome in diabetic patients treated with 13506.
thrombolytic therapy for acute myocardial infarction: the [53] Curb JD, Pressel SL, Cutler JA et al. for the Systolic in the
GUSTO-I experience. J Am Coll Cardiol 1996; 28: 16619. Elderly Program Cooperative Research Group. Effect of
[32] Iwasaka T, Takahashi N, Nakamura S et al. Residual left diuretic-based antihypertensive treatment on cardiovascular
ventricular pump function after acute myocardial infarction in disease risk in older diabetic patients with isolated systolic
NIDDM patients. Diabetes Care 1992; 15: 15226. hypertension. J Am Med Assoc 1996; 276: 188692.
[33] Takahashi N, Iwasaka T. Left ventricular regional function
[54] The SOLVD investigators. Effect of enalapril on mortality
after acute anterior myocardial infarction in diabetic patients. and the development of heart failure in asymptomatic patients
Diabetes Care 1989; 12: 6305. with reduced left ventricular ejection fractions. N Engl J Med
[34] Fisher BM, Frier BM. Evidence for a specific heart disease of 1992; 327: 68591.
diabetes in humans. Diabetic Med 1990; 7: 47889.
[35] Nathan DM. Long-term complications of diabetes mellitus. [55] Pfeffer M, Braunwald E, Moye LA et al. on the behalf of the
N Engl J Med 1993; 328: 167685. SAVE investigators. Effect of captopril on mortality and
morbidity in patients with left ventricular dysfunction after
[36] Fein FS, Sonnenblick EH. Diabetic cardiomyopathy. Cardio-
vasc Drugs Ther 1994; 8: 6573. myocardial infarction. N Engl J Med 1992; 327: 66977.
[37] Detry JM. The pathophysiology of myocardial ischaemia. Eur [56] Moye LA, Pfeffer MA, Wun CC et al. Uniformity of captopril
Heart J 1996; 17 (Suppl G): 4852. benefit in the SAVE study: subgroup analysis. Survival and
[38] Johnstone MT, Creager SJ, Scales KM, Cusco JA, Lee BK, ventricular enlargement study. Eur Heart J 1994; 15 (Suppl B):
Creager MA. Impaired endothelium-dependent vaso- 28.
dilation in patients with insulin-dependent diabetes mellitus. [57] Zuanetti G, Latini R, Maggiono AP, Franzosi MG, Santoro
Circulation 1993; 88: 25106. L, Tognoni G for the GISSI 3 investigators. Effect of the ACE
[39] Calver A, Collier J, Vallance P. Inhibition and stimulation of inhibitor lisinopril on mortality in diabetic patients with
nitric oxide synthesis in the human forearm arterial bed of acute myocardial infarction. Data from the GISSI 3 study.
patients with insulin-dependent diabetes. J Clin Invest 1992; Circulation 1997; 96: 423945.
90: 254854. [58] Herings RMC, deBoer A, Stricker BHC, Leufkens HGM,
[40] Williams SB, Cusco JA, Roddy MA, Johnstone MT, Creager Porsius A. Hypoglycemia associated with the use of inhibitors
MA. Impaired nitric oxide-mediated vasodilation in patients of angiotensin-converting enzyme. Lancet 1996; 345: 11958.
with non-insulin dependent diabetes mellitus. J Am Coll [59] Morris AD, Boyle DIR, McMahon AD et al. ACE inhibitor
Cardiol 1996; 27: 56774. use is associated with hospitalisation for severe hypoglycemia
[41] Ting HH, Timimi FK, Boles KS, Creager SJ, Ganz P, Creager in patients with diabetes. Diabetes Care 1997; 20: 136379.
MA. Vitamin C improves endothelium-dependent vaso- [60] Malmberg K, Herlitz J, Hjalmarsson A r , Ryden L. Effects on
dilation in patients with non-insulin-dependent diabetes metoprolol on mortality and late infarction in diabetics with
mellitus. J Clin Invest 1996; 97: 228. suspected acute myocardial infarction. Retrospective data
[42] Williams SB, Goldfine AB, Timimi FK, Ting HH, Roddy from two large scale studies. Eur Heart J 1989; 10: 4238.
MA, Creafer MA. Acute hyperglycemia attentuates [61] Bristow M, Gilbert EM, Abraham WT et al. Effect of
endothelium-dependent vasodilation in humans in vivo. carvedilol on LV function and mortality in diabetic versus
Circulation 1998; 97: 1695701. non-diabetic patients with ischemic or nonischemic dilated
[43] Youkoyama I, Momomura S-I, Ohtake T et al. Reduced cardiomyopathy. Circulation 1996; 84 (Suppl): AI:664.
myocardial flow reserve in non-insulin-dependent diabetes [62] Bersin RM, Wolfe C, Kwasman M et al. Improved hemo-
mellitus. J Am Coll Cardiol 1997; 30: 14727. dynamic function and mechanical efficiency in congestive
[44] Rodrigues B, McNeill JH. The diabetic Heart: metabolic heart failure with sodium dichloroacetate. J Am Coll Cardiol
causes for the development of a cardiomyopathy. Cardiovasc 1994; 23: 161724.
Res 1992; 26: 91322. [63] BARI investigators. Influence of diabetes on 5-year mortality
[45] Rodrigues B, Cam MC, McNeill JH. Myocardial substrate and morbidity in a randomised trial comparing CABG and
metabolism: implications for diabetic cardiomyopathy. J Mol PTCA in patients with multivessel disease: the Bypass Angio-
Cell Cardiol 1995; 27: 16979.
plasty Revascularisation Investigation (BARI). Circulation
[46] Rodrigues B, Cam MC, McNeill JH. Metabolic disturbances 1997; 96: 17619.
in diabetic cardiomyopathy. Mol Cell Biol Biochem 1998; 180:
537. [64] Aronson D, Bloomgarden Z, Rayfield EJ. Potential mech-
[47] Margolis J, Kannel W, Feinlieb M, Dawber T, McNamara P. anisms promoting restenosis in diabetic patients. J Am Coll
Clinical features of unrecognized myocardial infarctions Cardiol 1996; 27: 52835.
silent and symptomatic. Am J Cardiol 1973; 32: 17. [65] Almbrand B, Johannesson B, Sjostrand K, Malmberg K,
[48] Nikan E, Harati Y, Comstock J, Rhey R. Silent myocardial Ryden L. Costs of improved survival among diabetics with
infarction and diabetic cardiovascular autonomic neuropathy. myocardial infarction: experiences from the DIGAMI study.
Arch Intern Med 1986; 146: 222930. Eur Heart J 1998; 19 (Suppl): 241.
[49] Ewing D. Cardiac autonomic neuropathy. In: Jarret R, ed.
Diabetes and heart disease. Amsterdam: Elsevier, 1984:
99132.

Eur Heart J, Vol. 20, issue 11, June 1999