From the Copenhagen Male Study, Epidemiological Research Unit (J.J.,
H.O.H., P.S., F.G.), Copenhagen University Hospital, and the Glostrup
Population Studies (H.O.H.), Department of Internal Medicine C, Glostrup
University Hospital, Denmark.
Correspondence to Dr Jørgen Jeppesen, The Copenhagen Male Study, Epidemiological Research Unit, Copenhagen University Hospital, Bispebjerg, Bispebjerg Bakke 23, DK-2400 Copenhagen NV, Denmark.
Methods and Results—Baseline measurements of fasting lipids and
other IHD risk factors were obtained for 2906 white men (age range, 53
to 74 years) who were initially free of overt
cardiovascular disease. During an 8-year follow-up
period, 229 men had a first IHD event. Crude cumulative incidence rates
of IHD were 4.6% for the lowest, 7.7% for the middle, and 11.5% for
the highest third of triglyceride levels (P
for trend <.001). Compared with the lowest third level and adjusted
for age, body mass index, alcohol, smoking, physical activity,
hypertension, non–insulin-dependent diabetes mellitus, social class,
and LDL and HDL cholesterol, relative risks of IHD (95%
confidence interval) were 1.5 (1.0 to 2.3; P=.05) and
2.2 (1.4 to 3.4; P<.001) for the middle and highest
third of triglyceride levels, respectively. When
triglyceride levels were stratified by HDL
cholesterol levels (triglyceride third
multiplied by HDL cholesterol third), a clear gradient of
risk of IHD was found with increasing triglyceride levels
within each level of HDL cholesterol, including high HDL
cholesterol level, which are thought to provide protection
against IHD.
Conclusions—In middle-aged and elderly white men, a high level of
fasting triglycerides is a strong risk factor of IHD
independent of other major risk factors, including HDL
cholesterol.
In 1985 through 1986, a new baseline was established, which was
used for the present prospective study. All survivors from the 1970
study were traced by means of the Danish Central Population Register.
Between June 1985 and June 1986, all survivors (4505 except 34
emigrants) from the original cohort were invited to take part in this
study. Three thousand three hundred eighty-seven men (75%) agreed and
gave informed consent. Their mean age was 63 years (age range, 53 to 74
years). The study took place at The Glostrup Population Study, Glostrup
Hospital, University of Copenhagen (Denmark). Each subject was
interviewed by a physician (H.O.H.) regarding a previously completed
questionnaire and examined, with height, weight, and blood pressure
measurements taken. A venous blood sample was taken after the subjects
had fasted for
Criteria of Exclusion
Measurements
The study population was divided into equal thirds according to TG
levels. Cut points were
Self-reported NIDDM was accepted, provided the diagnosis had previously
been verified by a physician. None of the participants had
insulin-dependent diabetes mellitus. No measurements of plasma glucose
and insulin were performed in the present cohort. The presence or
absence of glucosuria was recorded after the subjects had fasted
for
Total weekly consumption of alcohol was calculated with the use of
questionnaire items regarding average alcohol consumption on weekdays
and weekends. Intakes of beer, wine, and spirits were reported
separately. Most of the alcohol consumed was beer. One drink
corresponded to 10 to 12 g ethanol. The men classified themselves
as never smokers, previous smokers, or current smokers. Current tobacco
consumption was calculated on the basis of information about the number
of cigarettes, cheroots, or cigars or the weight of pipe tobacco smoked
daily. One cigarette was taken as equal to 1 g tobacco, 1 cheroot
as equal to 3 g tobacco, and 1 cigar as equal to 4 g tobacco.
As previously estimated on the basis of serum cotinine level, the
validity of tobacco reporting was high.14
With respect to leisure-time physical activity, the men
classified themselves as sedentary, slightly active (<4 hours of
activity per week), or physically more active based on the following
question: Which description most precisely covers your pattern of
physical activity in leisure time? (1) You are almost entirely
sedentary or perform light physical activity for <2 hours per week.
(2) You perform light physical activity for 2 to 4 hours per week. (3)
You perform light physical activity for >4 hours per week or vigorous
activity for 2 to 4 hours per week. (4) You perform highly vigorous
physical activity for >4 hours per week or regular exercise or
competitive sports several times per week. For analytical purposes,
those responding as being in group 3 or 4 were pooled and are referred
to as the "high physical activity group," and those responding as
being in group 1 or 2 were pooled and are referred to as the "low
physical activity group," with <4 hours of activity per week.
According to the system of Svalastoga,15 the men
were divided into five social classes based on level of education and
job profile. Strata were defined as follows: social class I,
self-employed subjects with
End Points
Statistical Analysis
The study was approved by the Ethics Committee for Medical Research in
the County of Copenhagen.
During the follow-up period, 229 men had a first IHD event; 163 events
were nonfatal, and 66 events were fatal. In total, 426 men died from
all causes. Crude cumulative incidence rates of IHD and all-cause
mortality during the 8-year follow-up period according to thirds of
fasting serum TG concentrations, both overall and stratified on thirds
of HDL-C levels, are summarized in Table 2
The relative risks of IHD during the 8-year follow-up period according
to thirds of fasting serum TG concentrations, both overall and
stratified on thirds of HDL-C levels, are summarized in Table 3
Cardiovascular drugs, especially ß-blockers and
diuretics, are known to affect the lipoprotein
profile.24 In addition, subjects with a history
of cancer may have modified their lifestyles and lifestyle-related risk
factors because of their illness. To eliminate any interference from
these two conditions, 96 men with a history of cancer during the period
of 1943 through December 31, 1986, and 438 men taking any kind of
cardiovascular drugs (for practical purposes, men who
received ß-blockers and diuretics for hypertension) were
excluded from the analysis presented in Table 4
To provide results that could be compared with those obtained by other
investigators, we subjected our data to a standard logistic regression
analysis. To express the relative risk of IHD/1.0 mmol/L
change in TG level, we used the antilogarithm on the primary results
from the standard logistic regression analyses. In
univariate analysis, relative risk of IHD (95%
confidence interval) associated with a 1.0-mmol/L change in TG level
was 1.4 (0.8 to 2.1); in multivariate analysis
with control for age and HDL-C, the relative risk of IHD associated
with a 1.0-mmol/L change in TG level was 1.0 (0.2 to 1.8).
Our Approach
Special Subgroups
In the present study, we found that subjects with TG levels of
>2.5 mmol/L had a lower risk of IHD than subjects with TG
levels of 1.6 to 2.5 mmol/L. We further characterized this
subgroup to determine whether the presence of other risk factors could
account for this observation, but this subgroup in general seemed to
have a higher prevalence of other IHD risk factors. Compared with
subjects with a TG level of 1.6 to 2.5 mmol/L, they had a
higher total cholesterol level (0.39 mmol/L),
lower HDL-C level (0.18 mmol/L), higher body mass index
(1.5 kg/m2), higher systolic blood
pressure (2.5 mm Hg), higher diastolic blood
pressure (2.3 mm Hg), higher prevalence of hypertension (26%
versus 17%), higher prevalence of NIDDM (4.2% versus 2.8%), higher
prevalence of glucosuria (5.1% versus 1.3%), and more physically
inactive subjects (62% versus 52%). The decrease in risk of IHD with
TG levels of >2.5 mmol/L was seen in both younger and
older subjects, and the mortality was the same as that in subjects with
TG levels of 1.6 to 2.5 mmol/L.
Biological Plausibility
In addition to a direct atherogenic effect of TG-rich lipoproteins,
high TG levels appear to be a marker of a series of other potentially
atherogenic and prothrombotic changes. High TG levels have an effect on
LDL particle size, density distribution, and composition, leading to a
smaller, denser, more atherogenic LDL
particle,36 37 and through
plasminogen activator inhibitor–1,
TG are associated with deficient
fibrinolysis.38 39 High TG levels
also are closely associated with insulin resistance and
hyperinsulinemia40 ; although
the independent role of insulin in the pathogenesis of IHD is
controversial,41 a high fasting insulin
concentration was recently identified as an independent risk factor in
the Quebec Cardiovascular
Study.42 Thus, there is substantial biological
support for the association found in our study between fasting
hypertriglyceridemia and risk of IHD.
TG and Glucose Tolerance
TG as a Screening Test
Received August 7, 1997;
revision received November 14, 1997;
accepted November 23, 1997.
© 1998 American Heart Association, Inc.
Clinical Investigation and Reports
Triglyceride Concentration and Ischemic Heart Disease
An Eight-Year Follow-up in the Copenhagen Male Study
Abstract
Top
Abstract
Introduction
Methods
Results
Discussion
References
Background—The role of
triglycerides as a risk factor of ischemic heart
disease (IHD) remains controversial. For the present study, we
examined the relation between fasting triglycerides and
risk of IHD in the Copenhagen Male Study.
Key Words: coronary disease • lipids • lipoproteins • risk factors
Introduction
Top
Abstract
Introduction
Methods
Results
Discussion
References
The role of serum TG
as a screening test and a risk factor of IHD remains
controversial.1 2 3 Although in most
epidemiological studies a positive relationship has been found between
TG level and the risk of IHD, the usefulness of measuring TG in general
screening strategies has been questioned because
multivariate analysis control for HDL-C usually
eliminates or substantially diminishes the role of TG as a predictor of
IHD.1 2 3 However, the interpretation of
multivariate models that include TG and HDL-C is
complex and associated with several problems.1 2 3
TG and HDL-C are closely associated both distinct roles of TG and HDL-C in IHD in standard
multivariate
analysis.1 2 3 In addition, in comparison
with HDL-C, the distribution of TG levels is markedly skewed, requiring
logarithmic transformation for distribution-dependent analyses
such as standard regression analysis, a statistical maneuver
that may not provide an appropriate representation of
underlying biological processes.1 Finally, adding
to the complexity, some individuals with very high TG levels, such as
those with lipoprotein phenotype I or V, appear to have no
increased risk of IHD.4 With these problems kept
in mind, the purpose of the present study was to present an
analysis of data from the CMS to determine the effect of TG
versus that of HDL-C on the risk of IHD.
Methods
Top
Abstract
Introduction
Methods
Results
Discussion
References
Participants
The CMS was started in 1970 as a prospective
cardiovascular study.5 6 The
prospective male participants were derived from 14 workplaces in
Copenhagen: the air force, army, navy, emergency management agency,
postal service, customs service, a railroad company, national bank, a
telephone company, three municipal service centers (for electricity and
engineering and a fire brigade), a pharmaceutical company, and a
building contractor company. All eligible 6125 men were employed and
aged 40 to 59 years (mean age, 48 years); a total of 5249 men
(87%) participated. 
12 hours for the measurement of serum concentrations
of lipids.
Men who at baseline had a history of acute myocardial
infarction, angina pectoris, stroke, or intermittent claudication were
excluded from the follow-up study. For all who reported admission to
hospital because of acute myocardial infarction before the start of the
study, hospital records were checked. The diagnosis of acute
myocardial infarction was accepted if at least two of the following
symptoms or signs were recorded: retrosternal pain lasting >20
minutes, typical serial ECG changes in more than two ECG measurements,
and increases in the serum concentration of relevant enzymes (alanine
aminotransferase, lactate dehydrogenase, or creatinine
phosphokinase–MB). Information regarding angina pectoris, stroke, and
intermittent claudication was established with the questionnaire. Three
hundred forty-two men (10.1%) were excluded due to
cardiovascular diseases, and 139 men (4.1%) were
excluded due to missing data or conflicting answers; therefore, 2906
men were eligible for the prospective study.
Serum concentrations of total cholesterol, TG, and
HDL-C were analyzed with the use of standard
methods.7 8 9 10 Fasting lipids were measured only
once in each subject. Fasting lipids were not subjected to
ultracentrifugation, and the amount of possible
chylomicronemia was not determined. The fasting TG measurements were
performed using the Fully Enzymatic Method (Böehringer-Mannheim
Biochemica) and standardized in accordance with the World Health
Organization Collaborating Center for Blood Lipid Research in
Atherosclerosis and Ischemic Heart Disease at
the Institute for Clinical and Experimental Medicine/IKEM (Prague,
Czechoslovakia). All standard deviations were <0.05 mmol/L,
including between-day and within-day measurements, and all coefficients
of variation were <3.5%. LDL-C was determined indirectly according to
Friedewald's formula.11 Approximately 1.5% of
the study population had a TG level of >4.5 mmol/L, a level at
which the indirect LDL-C calculation becomes unreliable. However, the
exclusion from the study of subjects with a TG level of >4.5
mmol/L did not materially affect the relation found between LDL-C and
IHD, so we continued to use Friedewald's formula in subjects with a TG
level of >4.5 mmol/L and did not directly measure LDL-C. 
1.09 and 1.60 mmol/L. The population
was further divided into nine subgroups when each TG third was
stratified by HDL-C thirds (TG thirds multiplied by HDL-C thirds). Cut
points for HDL-C were 1.18 and 1.48 mmol/L. The reason for
selecting TG thirds as the major cut points in the present study
was to provide data comparable to data from the Framingham Heart Study,
in which basically the TG values shown above were used to subdivide the
Framingham population.12 12 hours. Of the 2906 men, 1.4% had glucosuria. Of subjects with
known NIDDM, 24.5% had glucosuria, whereas only 0.8% of subjects
without known NIDDM had glucosuria. With a manometer developed by
London School of Hygiene,13 blood pressure was
measured on the right arm with the subject seated. The definition of
hypertension was based on questionnaire information and blood pressure
measurements; the criteria were self-reported use of antihypertensive
treatment or systolic blood pressure of 150 mm Hg
and diastolic blood pressure of 100 mm Hg. Body
mass index (kgxm-2) was calculated on the basis
of weight and height measurements. 21 employees and white collar workers
with 51 subordinates or subjects with academic degrees (typical jobs
in the study cohort were officer, civil engineer, office
executive, and department head); social class II, self-employed
administrators with 6 to 20 employees and white collar workers with 11
to 50 subordinates or an intermediate education (typical jobs were head
clerk, engineer, and nonacademic architect); social class III,
self-employed subjects with 1 to 5 employees and white collar workers
with 1 to 10 subordinates (typical jobs were engine driver and train
guard); social class IV, self-employed subjects without employees,
white collar workers without subordinates or without qualified work,
and skilled blue collar workers (typical jobs were machine fitter in a
telephone company and station foreman); and social class V, unskilled
blue collar workers (typical jobs were unskilled laborer, mechanic, and
driver). For analytical purposes, those corresponding to social class
IV or V were pooled and are referred to as the "low social
class."
In 1995, a register follow-up was carried out on morbidity and
mortality rates for 1985/1986 and December 31, 1993. All men who had
taken part in the 1985/1986 examination were traced from registers.
Information was obtained on hospital admissions and death certificate
diagnoses within the follow-up period. We used the diagnoses from
official national registers; ischemic heart disease diagnoses
were codes 410 through 414 from the International Classification
of Diseases, 8th Revision. Previous studies have demonstrated a
high validity for Danish national
registers.16 17 18 19 20
Variables of interest are expressed as mean±SD values or
frequencies (in percentages). Differences between groups were tested
using ANOVA or Kendall's 
B test for trend. The
simultaneous contributions of several factors to the risk
of IHD were analyzed with the use of multiple logistic
regression models and the maximum likelihood ratio method. All
calculations were performed with SPSSPC+ basic and advanced statistical
software, version 3.1.21 22 A probability value
of .05 was taken as significant unless otherwise stated.
Results
Top
Abstract
Introduction
Methods
Results
Discussion
References
Lipid and nonlipid IHD risk factor characteristics according
to thirds of fasting serum TG concentrations are summarized in Table 1
; only 
1.5% of the study population
had a TG level of >4.5 mmol/L. The higher the TG concentration,
the higher were the total cholesterol and LDL-C
concentrations and the lower was the HDL-C concentration. In addition,
individuals with higher TG concentrations were less physically active
and had a higher body mass index, higher systolic and
diastolic blood pressures, and a higher prevalence of
hypertension, NIDDM, and glucosuria than those with lower
concentrations. Thus, a high fasting serum TG concentration appeared to
be a marker of the presence of several IHD risk factors in a group of
middle-aged and elderly men.
View this table:
[in a new window]
Table 1. Characteristics of Men With Different Fasting Serum
TG Concentrations Divided Into Thirds . A clear association was found between
TG levels and the risk of IHD, both overall and within each third of
HDL-C levels. As expected, a trend toward a higher risk of IHD was
found in the lowest third of HDL-C levels, although the protective role
of high HDL-C appeared to be lost in the highest third of TG
levels.23 Although study subjects in the highest
third of TG levels had a higher risk of IHD than those in the lower
thirds, the highest incidence rate of IHD (14%) was found in
subjects with TG levels of 1.6 to 2.5 mmol/L, whereas the risk
decreased with higher TG values to 8.5%. For total serum
cholesterol, the corresponding incidence rates of IHD were
6.7% in the lowest, 7.3% in the middle, and 9.5% in the highest
third of total cholesterol levels, with cut points of 6.0
and 6.9 mmol/L. Finally, Table 2 shows that no clear association
was found between TG levels and all-cause mortality, although
individuals in the lowest third of TG levels tended to have a lower
all-cause mortality rate.
View this table:
[in a new window]
Table 2. Crude Cumulative Incidence, % (n/total), of
IHD and All-Cause Mortality from 1985 through 1986 to December 31,
1993, According to Level of Fasting Serum TG and HDL-C Divided Into
Thirds . The relation between risk of IHD and
TG was assessed by successive adjustment for age only and for age and
other potentially confounding factors and covariates. Overall, a clear
gradient of risk of IHD was found with increasing thirds of TG levels;
compared with the lowest third of TG level, the risk of IHD was 50%
higher (P=.05) in the middle third and 120% higher
(P<.001) in the highest third of TG level after control for
the other major risk factors of IHD, including LDL-C and HDL-C. In
addition, within each third of HDL-C level, a gradient of risk of IHD
was found with increasing thirds of TG levels, and the difference in
the risk of IHD remained significant or borderline significant in the
highest third of TG level after control for the other major risk
factors of IHD. If total cholesterol instead of LDL-C was
included in the regression model, all results given in Table 3 were
basically the same (not shown).
View this table:
[in a new window]
Table 3. Relative Risk With 95% Confidence Limits for IHD
From 1985/1986 Through December 31, 1993, According to Level of Fasting
Serum TG and HDL-Cholesterol Divided into Thirds . Restriction of the analysis to
men not taking cardiovascular drugs and without a
history of cancer did not materially affect the relation between risk
of IHD and TG; if anything, the slope of the gradient, both overall and
within each third of HDL-C levels, appeared to get steeper.
View this table:
[in a new window]
Table 4. Crude Cumulative Incidence, % (n/total), of
IHD and All-Cause Mortality From 1985/1986 Through December 31, 1993,
According to Level of Fasting Serum TG and HDL-C Divided Into Thirds
Discussion
Top
Abstract
Introduction
Methods
Results
Discussion
References
Main Results
The present study provides some major new findings strongly
suggestive of a role of fasting serum TG as a risk factor of IHD. This
study appears to be the first to show in men that a clear gradient of
risk of IHD can be found with increasing TG levels within each level of
HDL-C, also after controlling for the other major risk factors of IHD,
including total cholesterol or LDL-C. This study also
appears to be the first in which the TG risk issue was studied with
control for the potentially confounding effects of antihypertensive
drugs, level of physical activity, alcohol use, and social class.
Finally, our results identified a small subgroup of men with high TG
levels who had a high risk of IHD despite their high HDL-C levels,
which are thought to provide protection against IHD.
As discussed by Austin1 and Garber and
Avins,3 the role of the TG level as a risk factor
of IHD has long been controversial. However, it also is well recognized
that the statistical characteristics of the distribution of TG levels,
variability of TG measurements, and statistical and
metabolic relations between TG and other risk factors (in
particular, HDL-C) may reduce the ability to detect an association
between TG and risk of IHD in standard multivariate
analysis.1 To compensate for the problems
described above, we used categorical transformation and divided the
study population into subgroups stratifying TG levels by HDL-C levels,
and in this context it should be remembered that in accordance with
most1 but not all
previous25 26 27 studies, we could not identify
fasting hypertriglyceridemia as a risk
factor of IHD after adjustment for HDL-C in our conventional
multivariate analysis. By working with
categorically transformed TG data in relatively large subgroups, the
effect of the large variability in TG measurements was smoothed out,
and the effect was minimized of the apparent paradox that the highest
TG levels are not necessarily associated with the highest risk of IHD.
This may explain why in the present study, we were able to identify
fasting hypertriglyceridemia as a strong
independent risk factor of IHD. On the basis of our categorically
transformed data and adjustments for HDL-C, subjects with TG levels of
2.0 mmol/L had a >100% increase in risk of IHD
compared with subjects with TG levels of 1.0 mmol/L, a
substantially higher value than the 14% increase in risk found in a
recent meta-analysis study of TG, HDL-C, and risk of IHD with a
1.0 mmol/L increase in TG level.28
In addition, because in our analysis we stratified for HDL-C
levels, it was possible to demonstrate a clear effect of TG on the risk
of IHD distinct from that of HDL-C.
In the present study, we identified a small subgroup of
hypertriglyceridemic men who had a high
risk of IHD although they had a high HDL-C level, which was thought to
be cardioprotective.29 We further characterized
this subgroup to determine whether the presence of other risk factors
could account for the high incidence of IHD in this subgroup. The most
distinct and consistent difference between the high TG/high
HDL-C subgroup and the other subjects in the highest third of TG levels
was that basically all of the high TG/high HDL-C subjects had a TG
level of 1.6 to 2.5 mmol/L, an interval of TG levels
associated with the highest risk of IHD in our study. This subgroup
also had a higher alcohol intake (27.8 beverages/week), a lower body
mass index (25.8 kg/m2 ), more subjects with low
social class (60%), but no subjects with a diagnosis of NIDDM, but
none of these differences are obvious explanations for the unexpected
high risk in the high TG/high HDL-C subgroup.6
There were no significant differences in the other variables listed
in Table 1
. We did not measure HDL subpopulations, and because
different HDL subpopulations may differ in their ability to protect
against IHD,29 we speculate that this subgroup
had a preponderance of HDL particles without the ability to protect
against an increased risk of IHD. In this context, it is relevant to
point out that other subgroups, such as patients with IDDM, have an
increased risk of IHD despite a raised HDL-C
level.30
Because fasting hypertriglyceridemia
can been identified as an independent risk factor of IHD, what is the
pathogenetic link between
hypertriglyceridemia and IHD? In this
question may be hidden another possible explanation for the TG
controversy that involves the existence of different kinds of TG-rich
lipoproteins, some of which are atherogenic and some of which are not.
Although total cholesterol is a reasonable substitute for
LDL cholesterol, total TG sometimes inadequately
represents the atherogenic TG-containing lipoproteins. This
paradox is underscored by the observation that subjects with
lipoprotein phenotypes I and V have no increased risk of IHD
despite very high TG levels,4 whereas the
hypertriglyceridemic state in
phenotype IIB is associated with a remarkably increased
risk.31 32 As discussed by
Castelli,12 the TG-rich lipoproteins that appear
not to be atherogenic are chylomicrons and very large VLDL particles,
whereas smaller VLDL particles, in particular VLDL remnants and
chylomicron remnants, are highly atherogenic, as shown in vitro with
high uptake into macrophages, leading to foam cell
formation,33 and in vivo with progression of
coronary artery lesions.34 Although we
did not perform ultracentrifugation, our results
clearly suggest that TG-rich lipoproteins with differing atherogenic
potential also were clinically significant in the CMS. Subjects with TG
levels of >2.5 mmol/L appeared to have less atherogenic
TG-rich lipoproteins than subjects with TG levels of 1.6 to 2.5
mmol/L, and they indeed appeared to be relatively protected
from IHD despite the fact that they had a higher prevalence of several
other risk factors of IHD. This interesting finding is supported by the
results of studies in animals: thus, the cholesterol-fed
diabetic rabbit has high cholesterol and high TG levels and
very large TG-rich lipoproteins but no
atherosclerosis.35 In addition,
it could be hypothesized that subjects with TG levels of 1.6 to
2.5 mmol/L had such atherogenic TG-rich lipoproteins that
even high levels of cardioprotective HDL-C are unable to protect
against IHD.
In the CMS, fasting plasma glucose levels were not measured;
thus, it was not possible to present direct evidence that
hypertriglyceridemia was a risk factor of
IHD independent of glucose tolerance and glucose control. This point is
of significant relevance; (1) in the CMS, probands with a high TG level
were characterized by an increased prevalence of NIDDM and glucosuria.
(2) Criqui et al25 did not find TG to be an
independent risk factor when fasting glucose levels were included in
multivariate data analysis. However, on the
basis of our global findings and the medical literature, we do not
think adjustment for undiagnosed or uncontrolled diabetes would have
changed our results. Of the 2906 men, 1.4% had glucosuria. Of subjects
with known NIDDM, 24.5% had glucosuria, whereas only 0.8% of subjects
without known NIDDM had glucosuria. The absence of glucosuria does not
rule out mild or biochemical diabetes (diagnosed only with an oral
glucose tolerance test), but it seems reasonable to conclude that the
vast majority of the present study population did not have
clinically significant diabetes. This conclusion is supported by data
from another Danish study from the greater Copenhagen area, the
Glostrup Population Study,43 in which <5% of
the men aged 60 to 70 years had fasting glucose levels of >6.8
mmol/L. In addition, in the subgroup with high TG/high HDL-C
levels, no subjects had a diagnosis of NIDDM and only 1 of the 104
subjects had glucosuria, yet there was a clear gradient of risk with
increasing TG levels within the subjects with the highest levels of
HDL-C. In the medical literature, TG level has been reported to be a
stronger risk factor of IHD than glucose level and other measures of
glucose control in subjects with impaired glucose tolerance or
diabetes,44 45 and TG-rich lipoproteins have been
related to the progression of coronary
atherosclerosis in nondiabetic
subjects.46 47 Finally, in a 10-year follow-up in
the Glostrup Population Study of elderly men,
hypertriglyceridemia was identified as a
risk factor of cardiovascular events and
cardiovascular mortality after control for the other
major risk factors, including impaired glucose tolerance and
diabetes48 ; in that Danish study, HDL-C was not
measured.
Should the fasting TG level be used as a screening test? So
far, no primary prevention trials have been designed to specifically
evaluate TG lowering in relation to risk of IHD. However, there is
evidence from some cholesterol-lowering trials that
subjects with hypertriglyceridemia will
benefit most from treatment. In the Helsinki Heart Study, most of the
risk of IHD and nearly all the benefits of drug treatment were confined
to persons with high concentrations of both TG and
cholesterol.32 In the West of
Scotland Coronary Prevention Study, when the subjects were
divided into groups according to lipid levels at baseline, men with TG
levels above the median (1.6 mmol/L) experienced a
greater treatment benefit in absolute terms than did men with total
cholesterol levels above the median (7.0
mmol/L).49 In the present study, fasting
hypertriglyceridemia was associated with an
increased risk of IHD at all levels of HDL-C, including high HDL-C
levels thought to provide protection against
IHD,23 and fasting
hypertriglyceridemia was a stronger risk
factor than total cholesterol. It also is interesting that
the TG levels found to be associated with an increased risk in the
present study were within a range generally considered to be
without clinical significance.50 Thus, fasting
serum TG levels appear to be a relevant screening test and should be
included in risk factor profiles; in preventive medicine, more
attention to fasting TG levels of 1.6 to 2.5 mmol/L
appears to be warranted.
Selected Abbreviations and Acronyms
CMS
=
Copenhagen Male Study
HDL-C
=
HDL cholesterol
IHD
=
ischemic heart disease
LDL-C
=
LDL cholesterol
NIDDM
=
non–insulin-dependent diabetes mellitus
TG
=
triglyceride(s)
Acknowledgments
This work was supported by grants from King Christian X
Foundation, The Danish Medical Research Counsil, The Danish Heart
Foundation, and Else & Mogens Wedell-Wedellsborg Foundation.
References
Top
Abstract
Introduction
Methods
Results
Discussion
References
This article has been cited by other articles:
 |
|
 |
|
  S. B. Seidelmann, L. Li, G.-Q. Shen, E. J. Topol, and Q. K. Wang Identification of a novel locus for triglyceride on chromosome 1p31-32 in families with premature CAD and MI J. Lipid Res., May 1, 2008; 49(5): 1034 - 1038. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. M Ridker Fasting versus Nonfasting Triglycerides and the Prediction of Cardiovascular Risk: Do We Need to Revisit the Oral Triglyceride Tolerance Test? Clin. Chem., January 1, 2008; 54(1): 11 - 13. [Full Text] [PDF]
|
|
|
|
|
|
Y. Ruixing, F. Qiming, Y. Dezhai, L. Shuquan, L. Weixiong, P. Shangling, W. Hai, Y. Yongzhong, H. Feng, and Q. Shuming Comparison of demography, diet, lifestyle, and serum lipid levels between the Guangxi Bai Ku Yao and Han populations J. Lipid Res., December 1, 2007; 48(12): 2673 - 2681. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Sarandol, S. Kirli, C. Akkaya, N. Ocak, E. Eroz, and E. Sarandol Coronary artery disease risk factors in patients with schizophrenia: effects of short term antipsychotic treatment J Psychopharmacol, November 1, 2007; 21(8): 857 - 863. [Abstract] [PDF]
|
|
|
|
 |
|
 A. Tirosh, A. Rudich, T. Shochat, D. Tekes-Manova, E. Israeli, Y. Henkin, I. Kochba, and I. Shai Changes in Triglyceride Levels and Risk for Coronary Heart Disease in Young Men Ann Intern Med, September 18, 2007; 147(6): 377 - 385. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. H. Criqui Triglycerides and Coronary Heart Disease Revisited (Again) Ann Intern Med, September 18, 2007; 147(6): 425 - 427. [Full Text] [PDF]
|
|
|
|
 |
|
 A. Garg and V. Simha Update on Dyslipidemia J. Clin. Endocrinol. Metab., May 1, 2007; 92(5): 1581 - 1589. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. de Graaf, G. M. van der Vleuten, E. ter Avest, G. M. Dallinga-Thie, and A. F. H. Stalenhoef High Plasma Level of Remnant-Like Particles Cholesterol in Familial Combined Hyperlipidemia J. Clin. Endocrinol. Metab., April 1, 2007; 92(4): 1269 - 1275. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Sarwar, J. Danesh, G. Eiriksdottir, G. Sigurdsson, N. Wareham, S. Bingham, S. M. Boekholdt, K.-T. Khaw, and V. Gudnason Triglycerides and the Risk of Coronary Heart Disease: 10 158 Incident Cases Among 262 525 Participants in 29 Western Prospective Studies Circulation, January 30, 2007; 115(4): 450 - 458. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. J. Choi, Y. E. Jo, Y. K. Kim, S. M. Ahn, S. H. Jung, H. J. Kim, Y.-S. Chung, K. W. Lee, and D. J. Kim High Plasma Concentration of Remnant Lipoprotein Cholesterol in Obese Children and Adolescents Diabetes Care, October 1, 2006; 29(10): 2305 - 2310. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Paramsothy and R. Knopp Management of dyslipidaemias. Heart, October 1, 2006; 92(10): 1529 - 1534. [Full Text] [PDF]
|
|
|
|
 |
|
 R. N. Pejic and D. T. Lee Hypertriglyceridemia. J Am Board Fam Med, May 1, 2006; 19(3): 310 - 316. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Magkos, D. C. Wright, B. W. Patterson, B. S. Mohammed, and B. Mittendorfer Lipid metabolism response to a single, prolonged bout of endurance exercise in healthy young men Am J Physiol Endocrinol Metab, February 1, 2006; 290(2): E355 - E362. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. Grosskopf, N. Baroukh, S.-J. Lee, Y. Kamari, D. Harats, E. M. Rubin, L. A. Pennacchio, and A. D. Cooper Apolipoprotein A-V Deficiency Results in Marked Hypertriglyceridemia Attributable to Decreased Lipolysis of Triglyceride-Rich Lipoproteins and Removal of Their Remnants Arterioscler. Thromb. Vasc. Biol., December 1, 2005; 25(12): 2573 - 2579. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Okada, T. Saito, H. Yoshimura, Y. Noguchi, T. Ito, H. Sasaki, and H. Hama Surfactant-Based Homogeneous Assay for the Measurement of Triglyceride Concentrations in VLDL and Intermediate-Density Lipoprotein Clin. Chem., October 1, 2005; 51(10): 1804 - 1810. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Imke, B. L. Rodriguez, J. S. Grove, J. R. McNamara, C. Waslien, A. R. Katz, B. Willcox, K. Yano, and J. D. Curb Are Remnant-Like Particles Independent Predictors of Coronary Heart Disease Incidence?: The Honolulu Heart Study Arterioscler. Thromb. Vasc. Biol., August 1, 2005; 25(8): 1718 - 1722. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. N. Hopkins, L. L. Wu, S. C. Hunt, and E. A. Brinton Plasma triglycerides and type III hyperlipidemia are independently associated with premature familial coronary artery disease J. Am. Coll. Cardiol., April 5, 2005; 45(7): 1003 - 1012. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W.-D. Li, C. Dong, D. Li, C. Garrigan, and R. A. Price A genome scan for serum triglyceride in obese nuclear families J. Lipid Res., March 1, 2005; 46(3): 432 - 438. [Abstract] [Full Text] [PDF]
|
|
|
|