Comparison of cardiovascular risk between patients with type 2 diabetes and those who had had a myocardial infarction: cross sectional and cohort studies
BMJ 2002; 324 doi: https://doi.org/10.1136/bmj.324.7343.939 (Published 20 April 2002) Cite this as: BMJ 2002;324:939
All rapid responses
We would like to congratulate JMM Evans et al 1 on their study.
However, their results deserve further discussion, and in particular the
comparison with Haffner’s study.2 Haffner compared tablet or insulin-
treated subjects with type 2 diabetes (DM), with non-DM subjects with any
history of prior myocardial infarction (MI). On the other hand, Evans
included lower risk diet-treated subjects with DM, and selection of MI
subjects in their cross-sectional study was restricted to those reporting
an event within the preceding 8 years (mean time from MI: 3.5 years).
Evans does not report the proportion of diet-treated DM subjects, but
since this is population based data, the proportion could have been
substantial. Evans’ cohort study results clearly show the important
relationship between ‘time from MI’ and admission with subsequent MI (Fig
3). Within the first 2.5 years of initial MI, when many of Evans’ cross
sectional MI subjects were recruited, admission rates were higher in the
MI group, but thereafter Kaplan-Meier curves are nearly parallel - in
keeping with the results of Haffner. Patient selection may be sufficient
to explain the discrepant results, but an analysis of Evans’ cross-
sectional data stratifying by ‘DM treatment’ and ‘time from MI’, would be
informative.
Comparison of studies has obvious limitations, but it is interesting
to compare CHD event rates in placebo-treated MI subjects without DM in
the CARE study (10 months after MI)3 with conventionally treated newly
diagnosed DM subjects in the UKPDS.4 Although age and sex distributions of
these studies, and exclusion of recent CHD at baseline in UKPDS, favour a
worse outcome in the CARE study population, annual rates of fatal and non-
fatal myocardial infarction are similar in these studies (CARE vs UKPDS,
1.8 % vs 1.7 %). In Evans’ MI groups, there was a high annual rate of
hospitalisation for MI in both cross-sectional and cohort studies (2.5 %
and 1.5 %) and also a high 28-day mortality (28 %) in the cohort study
(excluded patients). In Evans’ DM groups, annual rates of hospitalisation
for MI were 1.2 % and 0.5 % for cross sectional and cohort studies. These
data may be less than the UKPDS data because they do not include out of
hospital CHD deaths and MI.
In conclusion, patient selection, differences in CHD management, and
differences in baseline risk factors are likely to explain the discrepant
results of Haffner’s and Evans’ studies.
References
1. Evans JM, Wang J, Morris AD. Comparison of cardiovascular risk between
patients with type 2 diabetes and those who had had a myocardial
infarction: cross sectional and cohort studies. Br Med J 2002;324:939-42.
2. Haffner SM, Lehto S, Ronnemaa T, Pyorala K, Laakso M. Mortality from
coronary heart disease in subjects with type 2 diabetes and in nondiabetic
subjects with and without prior myocardial infarction. N Engl J Med
1998;339:229-34.
3. Goldberg RB, Mellies MJ, Sacks FM, Moye LA, Howard BV, Howard WJ, et
al. Cardiovascular events and their reduction with pravastatin in diabetic
and glucose-intolerant myocardial infarction survivors with average
cholesterol levels: subgroup analyses in the cholesterol and recurrent
events (CARE) trial. The Care Investigators. Circulation 1998;98:2513-9.
4. Intensive blood-glucose control with sulphonylureas or insulin compared
with conventional treatment and risk of complications in patients with
type 2 diabetes (UKPDS 33). UK Prospective Diabetes Study (UKPDS) Group
[published erratum appears in Lancet 1999 Aug 14;354:602]. Lancet
1998;352:837-53.
Competing interests: No competing interests
To the Editor:
The recent article by Evans et al documenting a significant
difference in the cardiovascular risk in diabetics and post myocardial
infarction patients is of profound significance. The numerous clinical
guidelines which have been based on the erroneous assumption that their
risks are identical will undoubtedly need revising.
One of the observations derived from their calculated survival rates
puzzles me. The authors note that the steep gradient of the survival
curve in post myocardial infarction patients levels off at about 1000 days
post infarct. That would seem to be a logical consequence of the healing
process occurring gradually after the acute event. But, strangely, the
same thing seems to be happening in the diabetic patients.
I have estimated the cardiac death rate among the diabetics for the
first 2000 day interval and the second 2000 day interval. From figure 2
(in bmj.com) the death rate in the first 2000 days is about 5%. At that
point about 72% of the diabetics are still alive (figure 1). At the end of
4000 days an additional 2.5% have died of cardiac causes, for a death rate
in those patients alive at 2000 days of about 3.5%.
Thus the cardiac death rate among diabetics seems dramatically higher
shortly after the diagnosis of diabetes than it is several years later.
How can that be explained? One would expect that a condition notorious for
causing premature vascular disease would demonstrate a cardiac death rate
that was gradually increasing rather than decreasing.
John I. Levitt, M.D.
Park Nicollet Clinic, Minneapolis, MN 54404, U.S.A.
Competing interests: No competing interests
Dear Editor:
In their well designed observational study, Josie Evans et. al.
conclude that patients with type 2 diabetes (DM2) were at lower risk of
cardiovascular outcomes than patients with established coronary heart
disease (CHD).1.- That contradicts the actual consensus to treat DM2 as
CHD equivalent, as officially proposed by the National Cholesterol
Education Program-Adults Treatment Panel III (NCEP-ATPIII). It is widely
known that DM2 have extensive macrovascular and microvascular disease, a
cluster of obesity, insulin resistance, hypertension, nephropathy,
dyslipidemia, glucose toxicity, oxidative stress, endothelial dysfunction,
calcification of the media, pro-coagulating and pro-inflammatory factors,
and a higher mortality during coronary events.
A first analysis would show 29% less mortality in DM2, and an
Absolute Risk Increase of 14% (“NNH”of 7) for CHD. If they were
treatments, and not diseases, DM2 would be clearly preferable. The fact is
that, despite appearing to have lower risk, all the groups studied (DM2
and CHD) have an annual mortality rate over 3%, and belong to the patients
with high risk of death ( similar to those with a left ventricular
ejection fraction under 0.35 or with a high risk exercise ECG score, or
Duke Score less or equal to minus 10), according to the consensus
guidelines of the American College of Cardiology, the American Heart
Association and the American College of Physicians-American Society of
Internal Medicine.2.- Thus, at the bottom line, the study of Evans et. al.
indeed confirm that the patients with DM2 are at high risk and should be
treated as having CHD.
Prof. Enrique Sánchez-Delgado, MD.
Member of the Executive Committee of the International Society of Internal
Medicine (ISIM), and of the Nicaraguan Association of Internal Medicine
(ANMI).
1.- Evans JMM, Wang J and Morris AD. Comparison of cardiovascular
risk between patients with type 2 diabetes and those who had had a
myocardial infarction: cross sectional and cohort studies. BMJ 2002; 324:
939-942 (20 April).
2.- Williams SV, Fihn SD and Gibbons RJ. Guidelines for the
Management of Patients with Chronic Stable Angina: Diagnosis and Risk
Stratification. Ann Intern Med. 2001; 135: 530-547.
Competing interests: No competing interests
Sirs,
Evans JMM et al. conclude their research stating that “patients with type
2 diabetes were at lower risk of cardiovascular outcomes than patients
with established coronary heart disease” (1). In my opinion, apart from
the interesting view-points of such a research, like the lower risk of
death from all causes or cardiovascular causes and of hospital admission
for myocardial infarction among patients with type 2 diabetes in
comparison with patients involved by establshed coronary heart disease,
this paper must be appreciated because it underscores the overlooked
“fact” of prime importance in primary prevention, in general, i.e. the
existence of different constitutions, I described for the first time, from
the “clinical” view-point (2, 3) (Fur further information, See the site
HONCode ID. N. 233736 http://digilander.iol.it/semeioticabiofisica,
Biophysical-Semeiotic Constitutions: diabetic, oncological, dyslipidemic,
hypertensive, arteriosclerotic, a.s.o.).
In a few words, after
differentiating in a clear-cut way between “Glycemology”, the fashionable
“science” based particularly on glycemic blood-levels, evaluated by a
large variety of methods, applied “easily” even by patients at their home,
and Diabetology, nowadays scarsely practised around the world, we have to
know that years or decades before diabetes onset, its asymtomatic
complications initiate and develop, but not in all diabetic patients,
since “not all diabetic patients are equal”. As a matter of fact, we are
now fortunately able to recognize and “quantify” at the bed-side, i.e.
“clinically” with the aid of a simple stethoscope, the real situations of
individual’s endothelium (constitutions) in all biological systems, even
over the two first life decades, including coronary arteries. Certainly,
doctors must unavoidably know firstly the structure and function of the tissue-
microcirculatory unit (wrongly defined by Anglo-American Authors as
“microcirculatory bed” : See the site
http://digilander.iol.it/microangiologia). It is really different in diverse
organs as well as in different zones of the same organs. Secondly they must
recognize physiological as well as pathological tissue-microcirculatory
units conditions, even very localized in a small area, with the aid of
Biophysical Semeiotics.
Sergio Stagnaro MD., Active Member NYAS.
1) Evans JMM, Wang J., Morris AD. Comparison of cardiovascular risk
between patients with type 2 diabetes and those who had had a myocardial
infarction: cross sectional and cohort studies. BMJ 2002;324:939 ( 20
April ).
2) Stagnaro-Neri M., Stagnaro S., Deterministic Chaos, Preconditioning and
Myocardial Oxygenation evaluated clinically with the aid of Biophysical
Semeiotics in the Diagnosis of ischaemic Heart Disease even silent. Acta
Med. Medit. 13, 109, 1997.
3) Stagnaro S.-Neri M., Stagnaro S., Sindrome di Reaven, classica e
variante, in evoluzione diabetica. Il ruolo della Carnitina nella
prevenzione del diabete mellito. Il Cuore. 6, 617, 1993.
Competing interests: No competing interests
Underestimation of risk of death from cardiovascular causes in diabetic patients
EDITOR----Evans et al's estimation of risk of death from all causes
among patients with type 2 diabetes is robust.[1] Nevertheless, their
estimation of the risk of death from cardiovascular causes may be
misleading.
In defining cardiovascular mortality, Evans et al used the underlying
cause of death (UCD), which is determined from the causal sequence among
conditions listed in the death certificate. Even when multiple causes of
death are present, only one UCD can be listed. This situation can cause
problems when interpreting mortality statistics in patients with diabetes.
At the time of death, diabetic patients commonly have other associated
conditions. In our previous study of 407 death certificates mentioning
diabetes, almost all listed more than one diagnosis: 6% had one, 24% had
two, 29% had 3, 22% had four, and 20% had five or more diagnoses.2 On more
than half (208/407) of the death certificates, the physicians entered both
diabetes and cardiovacular disease (CVD) (ICD-9 401-459).2 However, CVD
was assigned as the UCD in only 24% (50/208).[2] In another study of
mortality in people with diabetes in the United States, CVD was listed as
a cause of death in 69.4% of death certificates but was selected as the
UCD in just 49.5%.[3] Moreover, physicians show great variation in
determining the causal sequence when both diabetes and CVD are present,
even when given the same information.[4,5] These variations affect the
assignment of UCD according to ICD selection rules.
In patients with type 2 diabetes, the risk of cardiovascular death may be
underestimated by limiting cause of death analysis to the UCD. Because CVD
is often not listed as the UCD in patients with type 2 diabetes, and
because variations in death certification can affect selection of the UCD,
the multiple causes of death approach would provide a more realistic view
of the impact of CVD on mortality in such patients.
1.Evans JMM, Wang J, Morris AD. Comparison of cardiovascular risk
between patients with type 2 diabetes and those who had had a myocardial
infarction: cross sectional and cohort studies. BMJ 2002; 324; 939-942.
2.Lu TH, Lee MC, Chou MC. Accuracy of cause of death coding in Taiwan:
types of miscoding and effects on mortality statistics. Int J Epidemiol
2000; 29: 336-343.
3.Gu K, Cowie CC, Harris MI. Mortality in adults with and without diabetes
in a national cohort of the U.S. population, 1971-1993. Diabetes Care
1998; 21: 1138-1145.
4.Balkau B, Papoz L, EURODIAB Subarea C Study Group. European study of the
certification and coding of causes of death of six clinical cause
histories of diabetic patients. Int J Epidemiol 1993; 22: 116-126.
5.Lu TH, Shih TP, Lee MC, Chou MC, Lin CK. Diversity in death
certification: a case vignette approach. J Clin Epidemiol 2001; 54: 1086-
1093.
Competing interests: No competing interests