Nutrition, Metabolism & Cardiovascular Diseases (2009) 19, 401e408

available at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/nmcd

Triglyceride/HDL-cholesterol ratio is an
independent predictor for coronary heart disease in
a population of Iranian men*
F. Hadaegh a,*, D. Khalili a, A. Ghasemi b, M. Tohidi a,
F. Sheikholeslami a, F. Azizi b

a
Prevention of Metabolic Disorders Research Center, Research Institute for Endocrine Sciences, Shahid Beheshti University
(M.C.), Tehran, Iran
b
Endocrine Research Center, Research Institute for Endocrine Sciences, Shahid Beheshti University (M.C.), Tehran, Iran

Received 2 June 2008; received in revised form 24 August 2008; accepted 5 September 2008

KEYWORDS Abstract Background and aims: To determine whether triglyceride/high-density lipoprotein

Coronary heart disease; cholesterol ratio (TG/HDL-C), which has been shown to be an indicator of the metabolic
Triglycerides; syndrome (MetS) and insulin resistance, can predict coronary heart disease (CHD) indepen-
HDL-cholesterol; dently of total cholesterol (TC) and other risk factors in an Iranian population with a high prev-
Triglycerides/ alence of MetS and low HDL-C.
HDL-cholesterol ratio Methods and results: Between February 1999 and August 2001, 1824 men 40 years old, free of
clinical cardiovascular diseases at baseline, were followed. Baseline measurements included
serum level of TC, HDL-C, TG and risk factors for CHD including age, systolic and diastolic blood
pressure, body mass index, waist circumference, diabetes, smoking and a family history of
premature cardiovascular diseases. During a median follow up of 6.5 years until March 2007
(11,316 person-years at risk), a total of 163 new CHD events (27 fatal and 136 nonfatal)
occurred. The prevalence of MetS in subjects with TG/HDL-C 6.9 (top quartile) reached
63.6% versus 3.0% in those with TG/HDL-C <2.8 (low quartile). According to a stepwise Cox
proportional hazard model, including TG and TG/HDL-C quartiles, with TC and other risk
factors, men in the top quartile of TG/HDL-C relative to the first quartile had a significant

Abbreviations: ATP III, Adult Treatment Panel ___; BMI, Body mass index; CVD, Cardiovascular diseases; CHD, Coronary heart disease; CV,
Coefficient of variation; FPG, Fasting plasma glucose; HR, Hazard ratio; HDL-C, High-density lipoprotein cholesterol; LDL-C, Low-density
lipoprotein cholesterol; MetS, Metabolic syndrome; MI, Myocardial infarction; TLGS, Tehran Lipid and Glucose Study; TC, Total cholesterol;
TG, Triglyceride; VLDL, Very low density lipoprotein; WC, Waist circumference.
*
Funding source: This study was supported by grant No. 121 from National Research Council of the Islamic Republic of Iran.
* Corresponding author. Prevention of Metabolic Disorders Research Center, Research Institute for Endocrine Sciences, Shahid Beheshti
University (M.C.), P.O. Box. 19395-4763 Tehran, Iran. Tel.: þ98 21 22432503; fax: þ98 21 22402463.
E-mail address: fzhadaegh@endocrine.ac.ir (F. Hadaegh).

0939-4753/$ - see front matter ª 2008 Elsevier B.V. All rights reserved.
doi:10.1016/j.numecd.2008.09.003
402
hazard ratio (HR) of 1.75 (95% CI, 1.02e3.00), while TG did not remain in the model.
Conclusion: The evaluation of TG/HDL-C ratio should be considered for CHD risk prediction in
our male population with a high prevalence of MetS.
ª 2008 Elsevier B.V. All rights reserved.
Introduction
The association of serum total cholesterol (TC) and low-
density lipoprotein cholesterol (LDL-C) with developing
coronary heart disease (CHD) has been well established,
and low serum high-density lipoprotein cholesterol (HDL-C)
is considered a major risk factor for CHD [1]. There is some
evidences that serum triglycerides (TG) may be an inde-
pendent risk factor for cardiovascular diseases (CVD) [2].
Elevated TG and low HDL-C are basic characteristics of
insulin resistance and the metabolic syndrome (MetS),
which are strongly associated with CHD [1,3]. TG/HDL-C
ratio, as a relatively novel lipoprotein index, indicates the
presence of small dense LDL particles and could serve as
a good predictor of CHD [4,5]. As a result of this interre-
lation between TG and HDL-C, recent focus on high TG-low
HDL-C abnormality has grown considering risk assessment
and drug therapy for CHD [5e7]. Formerly, in the Tehran
Lipid and Glucose Study (TLGS) the association of TC, LDL-
C, non-HDL-cholesterol and TC/HDL-C, but not for HDL-C
and TG, with CVD has been reported [8]. To the best of our
knowledge, no data from Middle East populations have yet
been reported to show associations between TG/HDL-C and
CHD events. In the present study, we examine the TG/HDL-
C index as a predictive factor for CHD in a male population
with a high prevalence of MetS and low HDL-C [9e11].
Methods
Study population
TLGS is a prospective population-based study to determine
the risk factors for non-communicable diseases among
a representative Tehran urban population [12]. In the TLGS,
15,005 people aged 3 years and over, living in district 13 of
Tehran were selected by a multistage cluster random
sampling method; among them there were 2398 men aged
40 years and over evaluated in the cross-sectional phase of
TLGS (February 1999eAugust 2001). Subjects with a history
of CVD (n Z 259) at the baseline of the study and those with
missing data (n Z 94) were excluded, leaving 2045 subjects
40 years old. We followed 1824 subjects (89%) until March
2007 with median follow up of 6.5 years. The ethical
committee of our institute approved the proposal for this
study and informed written consent was obtained from
each subject.
Clinical and laboratory measurements
A trained interviewer collected information using a pre-
tested questionnaire. The obtained information included
demographics, anthropometrics, a family history of CVD
and a past medical history of CVD, cancer, diabetes mellitus
F. Hadaegh et al.
and smoking. Weight was measured while subjects were
minimally clothed without shoes using digital scales (Seca
707: range 0.1e150 kg) and recorded to the nearest 100 g.
Height was measured in a standing position without shoes,
using a tape meter, with the shoulders in a normal align-
ment. Body mass index (BMI) was calculated as weight (kg)
divided by the square of height (m2). Waist circumference
(WC) was measured at the umbilical level over light
clothing, using an unstretched tape meter, without any
pressure to the body surface. Two measurements of systolic
and diastolic blood pressure were performed using a stan-
dardized mercury sphygmomanometer (calibrated by the
Iranian Institute of Standards and Industrial Researches) on
the right arm after a 15 min rest in a sitting position; the
mean of the two measurements was considered to be
the subject’s blood pressure.
A blood sample was drawn between 7:00 and 9:00 AM
from all study participants after 12e14 h overnight fasting.
All the blood analyses were done at the TLGS research
laboratory on the day of blood collection. Fasting plasma
glucose (FPG) was measured using an enzymatic colori-
metric method with glucose oxidase. TC was assayed using
enzymatic colorimetric method with cholesterol esterase
and cholesterol oxidase. HDL-C was measured after
precipitation of the apolipoprotein B-containing lipopro-
teins with phosphotungstic acid. TG was assayed using an
enzymatic colorimetric method with glycerol phosphate
oxidase. These analyses were performed using Pars Azmon
kits (Pars Azmon Inc., Tehran, Iran) and a Selectra 2 auto-
analyzer (Vital Scientific, Spankeren, The Netherlands).
The intra and inter-assay coefficients of variation (CV) were
2.2% for glucose. For both total and HDL-Cholesterol, intra
and inter-assay CVs were 0.5 and 2% respectively. Intra and
inter-assay CVs were 0.6 and 1.6% for TG respectively.
CHD outcome
Details of the cardiovascular outcome collection have
recently been published [8]. To summarize, participants
were followed up for any medical event annually by phone
calls. A trained nurse asked them for any medical condi-
tions and then, a trained physician collected complemen-
tary data regarding that event during a home visit and by
acquisition of data from medical files. The collected data
were then evaluated by an outcome committee consisting
of an internist, an endocrinologist, a cardiologist, an
epidemiologist and other experts, when needed, to assign
a specific outcome for every event. In the present study,
our desired events were the first CHD events, including
definite myocardial infarction (MI) (with diagnostic ECG and
biomarkers), probable MI [positive ECG findings plus cardiac
symptoms or signs plus missing biomarkers or positive ECG
findings plus equivocal biomarkers] [13] or angiographic
proven CHD in subjects with new onset angina and death
Triglyceride/HDL-C ratio and coronary heart disease 403

from CHD. Death from CHD or other reasons was confirmed
by reviewing the death certificate or medical records.
Definition of terms
Diabetes mellitus was defined as FPG 7.0 mmol/l (126 mg/
dl) or current use of anti-diabetic drugs [14]. A family
history of premature CVD reflected any prior diagnosis of
CVD by a physician in any female first-degree relative under
65 years old, and any male first-degree relative under 55
years old. Smoking includes past or current smoking of any
kind of tobacco. MetS was defined according to the Adult
Treatment Panel ___ (ATP___) [1].
Statistics
Mean (SD) values or frequencies (percentage) of the base-
line characteristics are expressed for individuals with and
without CHD events. Because of the skewed distribution of
TG and TG/HDL-C index, they are shown as median (inter-
quartile range). Comparison of baseline characteristics
between the two groups was made by student’s t-test for
continuous variables, c2 test for categorical variables or
ManneWhitney U-test for TG and TG/HDL-C. We calculated
quartiles for TC, TG, HDL-C and TG/HDL-C according to
their distribution in the study population. Linear trend for
CHD risk factors across the TG/HDL-C quartiles was calcu-
lated by the linear regression and c2 test for continuous
and categorical variables respectively. To determine the
linear trend of the CHD event rate among lipid quartiles, c2
test was used.
Adjusted Hazard Ratios (HRs) of the CHD for each lipid
marker and TG/HDL-C index were calculated by stepwise
Cox proportional hazards models. Covariates (age, systolic
and diastolic blood pressure, BMI, WC, diabetes, smoking
and a family history of premature CVD) were entered into
the model if the probability value was <0.2 and retained
when the probability value was <0.1. Lipid markers, which
had a significant HR after full adjustment, were adjusted
for TC as well. For lipid markers and TG/HDL-C index
quartiles, HRs were calculated relative to the lowest
quartile. To examine the linear trend of CHD events across
quartiles, we used the median of the quartiles as a contin-
uous variable in the model. The proportional hazards
assumption in the Cox model was assessed graphically and
with a time-dependent covariate test. All proportionality
assumptions were generally appropriate.
SPSS for Windows (version 15; SPSS Inc., Chicago, IL.)
was used for data analyses and p-values <0.05 were
considered statistically significant.
Results
The study sample consisted of 1824 men aged 40e86 years
(mean age 55  10.6). There was no significant difference
between followed and non-followed persons in age, lipid
markers and major risk factors for CHD. During 11,316
person-years of follow up, 163 new CHD events, 27 fatal
and 136 nonfatal, occurred.
The mean and prevalence of baseline characteristics in
men with and without CHD events are summarized in Table
1. Compared to men who had no CHD, those who experi-
enced CHD events were significantly older, more likely to
be diabetic and had greater systolic and diastolic blood
pressure, BMI, WC, and higher TC, TG and TG/HDL-C.
The prevalence of ATPIII-defined MetS in our male pop-
ulation was 32.0% and the prevalence of its components
was: low HDL-C 62.9%, high TG 56.9%, high blood pressure
45.0%, high FPG 19.0%, and abdominal obesity 13.3%. MetS
was more prevalent in men with CHD than those without
(55.2 versus 29.7, p < 0.001).
As highlighted in Table 2 all CHD risk factors had
a significant trend across TG/HDL-C quartiles except
smoking and a family history of premature CVD. In addition,
the prevalence of MetS in men with TG/HDL-C 6.9
reached 63.6% versus 3.0% in those with TG/HDL-C <2.8.
Table 3 shows the incidence rate of CHD according to
quartiles of TC, TG, HDL-C, and TG/HDL-C. A trend toward
higher risk of CHD was seen among quartiles of TC, TG and
TG/HDL-C but not HDL-C.
Hazard Ratios of CHD for quartiles of TC, TG, HDL-C and
TG/HDL-C, after adjustment for age per se and age with the

Table 1 Comparison of baseline characteristics between participants with and without Coronary Heart Disease (CHD)
Variablea With CHD (n Z 163) Without CHD (n Z 1661) p value
Age (years) 58.7 (9.5) 54.9 (10.6) <0.001
Systolic blood pressure (mm Hg) 136 (22.5) 125 (20.2) <0.001
Diastolic blood pressure (mm Hg) 83.6 (13.2) 79.4 (11.7) <0.001
Body mass index (kg/m2) 27.4 (3.9) 26.2 (3.8) <0.001
Waist circumference (cm) 95.2 (10.1) 90.8 (10.7) <0.001
Diabetes mellitus (%) 31.3 11.0 <0.001
Smoking (past or current) (%) 50.3 46.4 0.35
Family history of premature cardiovascular disease (%) 16.0 12.5 0.21
Metabolic syndrome (%) 55.2 29.7 <0.001
Total cholesterol (mmol/l) 6.0 (1.2) 5.5 (1.1) <0.001
Triglyceride (mmol/l) 2.18 (1.59e3.26) 1.82 (1.24e2.59) <0.001
HDL-cholesterol (mmol/l) 0.96 (0.22) 1.00 (0.25) 0.07
Triglyceride/HDL-cholesterol 5.14 (3.43e8.39) 4.34 (2.69e6.75) <0.001
a
Continuous variables shown as mean (SD) with p according to t-test, Categorical variables as % with p according to c2, TG and
TG/HDL-C as median (interquartile range) with p according to ManneWhitney U-test.
404 F. Hadaegh et al.

Table 2 Comparison of baseline characteristics among Triglyceride /HDL-cholesterol quartiles

a
Variable Triglyceride /HDL-cholesterol quartiles p for trendb
1 (<2.8) 2 (2.8e4.4) 3 (4.4e6.9) 4 (6.9)
Age (years) 56.8 (11.4) 56.3 (10.8) 54.1 (10.0) 53.7 (9.7) <0.001
Systolic blood pressure (mm Hg) 124.3 (21.8) 125.7 (21.9) 126.7 (19.6) 127.1 (19.3) 0.032
Diastolic blood pressure (mm Hg) 77.3 (12.6) 79.1 (11.6) 81.4 (11.7) 81.3 (11.1) <0.001
Body mass index (kg/m2) 24.4 (4.1) 26.1 (3.7) 27.1 (3.6) 27.5 (3.3) <0.001
Waist circumference (cm) 85.5 (11.5) 91.2 (10.2) 93.3 (9.5) 94.9 (9.0) <0.001
Diabetes mellitus (%) 6.9 8.9 13.4 21.8 <0.001
Smoking (past or current) (%) 46.8 47.8 45.4 47.0 0.879
Family history of premature 11.2 13.8 14.3 12.1 0.669
cardiovascular disease (%)
Metabolic syndrome (%) 3.0 15.4 45.8 63.6 <0.001
Total cholesterol (mmol/l) 5.09 (0.96) 5.47 (0.99) 5.63 (1.04) 5.88 (1.22) <0.001
HDL-cholesterol (mmol/l) 1.23 (0.26) 1.03 (0.18) 0.94 (0.18) 0.80 (0.16) <0.001
Triglyceride (mmol/l) 1.00 (0.81e1.16) 1.59 (1.38e1.79) 2.18 (1.91e2.49) 3.44 (2.83e4.32) <0.001
a
Continuous and categorical variables shown as mean (SD) and percent respectively, except TG reported as median (interquartile
range).
b
Test for a linear trend according to linear regression and c2 for continuous and categorical variables respectively, TG log-transformed
for analysis.

other risk factors, are shown in Fig. 1. Age-adjusted HRs
(Fig. 1A) for the top quartile versus the lowest one for TC,
TG, TG/HDL-C and for HDL-C lowest versus highest quartile
were significant; however after more adjustment with the
other risk factors (Fig. 1B), the HRs showed relatively no
change for TC, attenuated for TG and TG/HDL-C and
became non-significant for HDL-C.
Table 4 shows HRs of CHD for TG and TG/HDL-C quartiles
after further adjustment with TC besides other risk factors.
After adjustment for TC, just TG/HDL-C predicted CHD
events (HR for top versus low quartile was 1.75, 95% CI
1.02e3.00, p Z 0.044).
Finally, on stepwise multivariate analysis including TG
and TG/HDL-C quartiles, with TC and other risk factors, the
risk associated with TG was no longer evident (data not
shown) but TG/HDL-C, TC, age, systolic blood pressure, WC
and diabetes remained significant independent predictors.
Discussion
The present population-based study with a follow up of
11,316 person-years, indicates that TC, TG and TG/HDL-C
are each important risk factors for CHD in males with an HR
range from 1.99 to 2.87 (comparing highest to lowest
quartile) after adjustment for age and traditional risk
factors. However, in the presence of TC only TG/HDL-C but
not TG was an independent predictor of CHD risk with an HR
of 1.75 (1.02e3.00).

Table 3 Coronary heart disease events among quartiles of total cholesterol, triglyceride, HDL-cholesterol and
TG/HDL-cholesterol
Variables Quartiles p for trenda
1 2 3 4
Total cholesterol, <4.76 (4.32) 4.76e5.43 (5.07) 5.43e6.13 (5.74) 6.13 (6.69)
mmol/l (median)
Number of events (%) 22 (4.9) 25 (5.6) 43 (9.3) 73 (15.5) <0.001
Person-years of follow-up 2752 2830 2875 2857
Triglyceride, mmol/l (median) <1.29 (0.99) 1.29e1.84 (1.57) 1.84e2.63 (2.18) 2.63 (3.49)
Number of events (%) 21 (4.5) 38 (8.6) 46 (10.1) 58 (12.4) <0.001
Person-years of follow-up 2857 2766 2834 2858
HDL-cholesterol, mmol/l (median) <0.85 (0.72) 0.85e1.00 (0.90) 1.00e1.11 (1.03) 1.11 (1.27)
Number of events (%) 60 (10.2) 25 (8.7) 44 (9.0) 34 (7.4) 0.131
Person-years of follow-up 3551 1807 3063 2897
Triglyceride/HDL-cholesterol (median) <2.8 (1.95) 2.8e4.4 (3.55) 4.4e6.9 (5.40) 6.9 (9.6)
Number of events (%) 21 (4.5) 42 (9.4) 39 (8.7) 61 (13.1) <0.001
Person-years of follow-up 2905 2780 2804 2830
a
p-values are tests of trend by increasing quartile level.
Triglyceride/HDL-C ratio and coronary heart disease 405

Figure 1 (A).Age-adjusted hazard ratios (95% CI) of coronary heart disease (CHD) based on Cox proportional hazard models for total
cholesterol (TC), triglyceride (TG), HDL-cholesterol (HDL-C) and TG/HDL-C quartiles. (B). Hazard ratios (95% CI) of CHD based on
stepwise Cox proportional hazard models for TC, TG, HDL-C and TG/HDL-C quartiles. Age, systolic blood pressure, waist circumference
and diabetes remained in the models. Body mass index, diastolic blood pressure, smoking (past or current) and family history of
premature cardiovascular diseases that did not remain in the model because of p > 0.10. * p < 0.05, y p < 0.01, z p < 0.001.
406 F. Hadaegh et al.

Table 4 Multivariate hazard ratios (HR) and 95% CI of coronary heart disease based on stepwise Cox proportional hazard
models for triglyceride (TG) and triglyceride/HDL-cholesterol (TG/HDL-C) quartiles*
HRa 95% CI p value HRb 95% CI p value
Age (years) 1.02 1.00e1.04 0.016 1.02 1.01e1.04 0.015
Total cholesterol (mmol/l) 1.36 1.19e1.55 <0.001 1.35 1.19e1.53 <0.001
Systolic blood 1.01 1.01e1.02 <0.001 1.01 1.01e1.02 <0.001
pressure (mm Hg)
Waist circumference (cm) 1.02 1.00e1.03 0.051 1.02 0.99e1.03 0.068
Diabetes mellitus 2.32 1.62e3.33 <0.001 2.25 1.57e3.23 <0.001
TG (mmol/l) 0.296c
<1.29 1
1.29e1.83 1.25 0.72e2.18 0.437
1.84e2.62 1.52 0.88e2.62 0.134
>2.62 1.36 0.77e2.38 0.291
TG/HDL-C 0.086c
<2.78 1 _ _
2.78e4.40 1.54 0.90e2.63 0.117
4.41e6.87 1.31 0.75e2.29 0.344
>6.87 1.75 1.02e3.00 0.044
*Diastolic blood pressure, body mass index, smoking (past or current) and family history of premature cardiovascular diseases did not
remain in the model because of p > 0.10.
a
Model includes triglyceride.
b
Model includes triglyceride/HDL-cholesterol.
c
P for trend (median of the quartiles as a continuous variable in the model).

However, our results should be considered with specific
reference to the absence or presence of MetS as shown in
other studies [6,15]. The MetS is a cluster of risk factors
that has been shown to be highly predictive of CHD and
diabetes [1,16,17]. The prevalence of MetS and its char-
acteristic lipid abnormality including high TG and low HDL-
C in our male population were high (32.0%, 56.9% and 62.9%
respectively). However, the prevalence of abdominal
obesity according to the ATPIII definition was about 13%.
The discrepancy between the prevalence of dyslipidemia
and abdominal obesity might be due to the underestimation
of the latter according to the ATPIII cutoff point for the high
WC in Iranian population [18]. The high prevalence of MetS
in our study is comparable to those reported for the US and
Turkish adult men in a similar age group [15,19]. Our study
sample is unique in the sense that the prevalence of low
HDL-C is relatively high, causing only a marginally signifi-
cant difference in the baseline level of HDL-C between
those who developed and did not develop CHD (Table 1),
because of lack of variation [10,11].
Both insulin resistance and central obesity have been
suggested as being essential factors for the MetS and are
associated with high TG and low HDL-C, which are the
characteristic lipid abnormalities of this syndrome
[1,16,17]. It has been reported that the TG/HDL-C ratio
(like fasting plasma insulin concentration) is a good and
practical surrogate estimate of insulin resistance, and is as
effective as the ATPIII MetS criteria in identifying insulin-
resistant-individuals [3]; however this was not shown in
African Americans [20]. McLaughlin et al. showed that the
ability to identify insulin-resistant-individuals with TG
1.5 mmol/l and TG/HDL-C 3.0 was similar to the criteria
proposed by the ATPZ for diagnosis of the MetS [3].
Recently Alhassan et al. also showed that the TG/HDL-C
ratio 3.0 moderately agreed with the clinical criteria for
MetS definitions in premenopausal women (k range, 0.36e
0.59) [21]. In our data analyses TG/HDL-C 6.9 (accompa-
nied by 64% prevalence of MetS) had a HR z 1.8 for incident
CHD events versus TG/HDL-C <2.8 (accompanied by 3%
prevalence of MetS).
The TG/HDL-C ratio indicates the relative size of LDL
particles and thus, their resulting atherogenic potential.
A low TG/HDL-C ratio is indicative of primarily large, non-
atherogenic LDL particles, whereas a high TG/HDL-C ratio
indicates a larger population of small, dense pro-atherogenic
LDL particles [22,23]. Dobiasova et al., showed that log TG/
HDL-C as an atherogenic index indicates a balance between
the actual concentration of plasma TG and HDL-C which
may predetermine the direction of cholesterol transport in
the intravascular volume pool, [23]. With elevation of the
TG/HDL-C ratio, HDL particles shifted toward smaller sizes
(indicating that the maturation of HDL-C might be impeded)
leading to increased risk of CHD [24].
It has been reported that the most common dyslipidemia
for the Iranian male population is characterized by low
HDL-C and hypertriglyceridemia [10,25]. The traditional
Iranian diet is wheat-based and between 1985 and 1995,
urban households increased their bread consumption by
almost 25% [26]; thus the hypertriglyceridemia in our pop-
ulation may be induced by high carbohydrate diets which
are different from standard Western high-fat diets because
high carbohydrate diet stimulates the production and
secretion of TG and very low density lipoprotein (VLDL)
through lipogenesis in the liver [27,28]. Additionally, the
high prevalence of dysglycemia in the Iranian population is
associated with high TG levels [29]. The important deter-
minants of low HDL-C in our population were male gender,
hypertriglyceridemia, obesity and smoking [11]; thus
Triglyceride/HDL-C ratio and coronary heart disease
although it can be argued that either the concentration of
plasma TG or HDL-C can be used to predict coronary risk,
considering the complex metabolic interaction between TG
and cholesterol-rich lipoproteins, it would appear that
concurrent use of these parameters provides more accurate
information [23].
Our study, like the Asia Pacific Cohort Studies Collabo-
ration, showed that TG/HDL-C is better than TG per se, and
that it enhanced risk prediction more than TG did [4].
As for strengths of our study, although the independent
prediction of CHD risk by the relatively novel marker of TG/
HDL-C has previously been reported, our study differs in
that this prediction is shown for the first time in Middle East
Caucasians after multivariate adjustment. In our data
analyses the last quartile of TG, after adjustment for risk
factors, was a significant predictor for CHD (HR Z 1.99,
Fig. 1B); however, this decreased to a non-significant level
after TC adjustment. It has been argued that this adjust-
ment is not appropriate, because this method adjusts for
triglyceride-rich VLDL component of TC, which rises
disproportionately as serum TG increases [5]. Recently
Onat et al. showed that in Turkish adult men, with a high
prevalence of MetS and low levels of HDL-C, fasting TGs are
predictive of future CVD events with an HR of 1.4 [16].
Information in other populations with such characteristics
has been limited.
The limitations of our study need mentioning; first, we
measured TC, HDL-C and TG only once at the baseline, thus
the potential bias resulting from regression dilution of TG
and HDL-C measurement could not be excluded [30].
Second, the study sample is from Middle East Caucasians
with a high prevalence of MetS and low HDL-C levels, and
the ability of the same metabolic marker to predict CHD in
the other ethnicities with such characteristics is unproven.
In conclusion, our results suggest that measurement of
TG/HDL-C ratio should be considered in the risk assessment
of CHD in our population with a high prevalence of MetS.
Acknowledgements
This study was supported by grant No. 121 from the
National Research Council of the Islamic Republic of Iran.
We would like to thank the participants of district 13,
Tehran, for their enthusiastic support in this study; the
linguistic help given by Ms N. Shiva is much appreciated.
References
[1] Executive summary of the Third report of the National
Cholesterol Education Program (NCEP) expert panel on
detection, evaluation, and treatment of high blood choles-
terol in adults (Adult Treatment Panel III). JAMA 2001;285:
2486e97.
[2] Hokanson JE, Austin MA. Plasma triglyceride level is a risk
factor for cardiovascular disease independent of high-density
lipoprotein cholesterol level: a meta-analysis of population-
based prospective studies. J Cardiovasc Risk 1996;3:213e9.
[3] McLaughlin T, Abbasi F, Cheal K, Chu J, Lamendola C,
Reaven G. Use of metabolic markers to identify overweight
individuals who are insulin resistant. Ann Intern Med 2003;139:
802e9.
407
[4] Barzi F, Patel A, Woodward M, Lawes CM, Ohkubo T, Gu D,
et al. A comparison of lipid variables as predictors of cardio-
vascular disease in the Asia Pacific region. Ann Epidemiol
2005;15:405e13.
[5] Gaziano JM, Hennekens CH, O’Donnell CJ, Breslow JL,
Buring JE. Fasting triglycerides, high-density lipoprotein, and
risk of myocardial infarction. Circulation 1997;96:2520e5.
[6] Jeppesen J, Hein HO, Suadicani P, Gyntelberg F. Low triglyc-
erides-high high-density lipoprotein cholesterol and risk of
ischemic heart disease. Arch Intern Med 2001;161:361e6.
[7] Rizos E, Mikhailidis DP. Are high-density lipoprotein and
triglyceride levels important in secondary prevention:
impressions from the BIP and VA-HIT trials. Int J Cardiol 2002;
82:199e207 [discussion-8].
[8] Hadaegh F, Harati H, Ghanbarian A, Azizi F. Association of
total cholesterol versus other serum lipid parameters with the
short-term prediction of cardiovascular outcomes: Tehran
Lipid and Glucose Study. Eur J Cardiovasc Prev Rehabil 2006;
13:571e7.
[9] Azizi F, Rahmani M, Emami H, Mirmiran P, Hajipour R,
Madjid M, et al. Cardiovascular risk factors in an Iranian urban
population: Tehran Lipid and Glucose Study (phase 1). Soz
Praventivmed 2002;47:408e26.
[10] Azizi F, Salehi P, Etemadi A, Zahedi-Asl S. Prevalence of
metabolic syndrome in an urban population: Tehran Lipid and
Glucose Study. Diabetes Res Clin Pract 2003;61:29e37.
[11] Azizi F, Raiszadeh F, Salehi P, Rahmani M, Emami H,
Ghanbarian A, et al. Determinants of serum HDL-C level in
a Tehran urban population: the Tehran Lipid and Glucose
Study. Nutr Metab Cardiovasc Dis 2002;12:80e9.
[12] Azizi F, Rahmani M, Emami H, Madjid M. Tehran Lipid and
Glucose Study: rationale and design. CVD Prevention 2000;3:
242e7.
[13] Luepker RV, Apple FS, Christenson RH, Crow RS, Fortmann SP,
Goff D, et al. Case definitions for acute coronary heart disease
in epidemiology and clinical research studies: a statement
from the AHA Council on epidemiology and prevention; AHA
Statistics Committee; World Heart Federation Council on
epidemiology and prevention; the European Society of
Cardiology Working Group on epidemiology and prevention;
Centers for Disease Control and Prevention; and the National
Heart, Lung, and Blood Institute. Circulation 2003;108:
2543e9.
[14] Report of the expert committee on the diagnosis and classi-
fication of diabetes mellitus. Diabetes Care 2003;26(Suppl. 1):
S5e20.
[15] Onat A, Sari I, Yazici M, Can G, Hergenc G, Avci GS. Plasma
triglycerides, an independent predictor of cardiovascular
disease in men: a prospective study based on a population
with prevalent metabolic syndrome. Int J Cardiol 2006;108:
89e95.
[16] Grundy SM, Brewer Jr HB, Cleeman JI, Smith Jr SC, Lenfant C.
Definition of metabolic syndrome: report of the National
Heart, Lung, and Blood Institute/American Heart Association
conference on scientific issues related to definition. Circula-
tion 2004;109:433e8.
[17] Ford ES. The metabolic syndrome and mortality from cardio-
vascular disease and all-causes: findings from the National
Health and Nutrition Examination Survey II Mortality Study.
Atherosclerosis 2004;173:309e14.
[18] Esmaillzadeh A, Mirmiran P, Azizi F. Waist-to-hip ratio is
a better screening measure for cardiovascular risk factors
than other anthropometric indicators in Tehranian adult men.
Int J Obes Relat Metab Disord 2004;28:1325e32.
[19] Patel A, Huang KC, Janus ED, Gill T, Neal B,
Suriyawongpaisal P, et al. Is a single definition of the meta-
bolic syndrome appropriate? A comparative study of the USA
and Asia. Atherosclerosis 2006;184:225e32.
408
[20] Sumner AE, Finley KB, Genovese DJ, Criqui MH, Boston RC.
Fasting triglyceride and the triglyceride-HDL cholesterol ratio
are not markers of insulin resistance in African Americans.
Arch Intern Med 2005;165:1395e400.
[21] Alhassan S, Kiazand A, Balise RR, King AC, Reaven GM,
Gardner CD. Metabolic syndrome: do clinical criteria identify
similar individuals among overweight premenopausal women?
Metabolism 2008;57:49e56.
[22] Maruyama C, Imamura K, Teramoto T. Assessment of LDL
particle size by triglyceride/HDL-cholesterol ratio in non-
diabetic, healthy subjects without prominent hyperlipidemia.
J Atheroscler Thromb 2003;10:186e91.
[23] Dobiasova M, Frohlich J. The plasma parameter log (TG/HDL-
C) as an atherogenic index: correlation with lipoprotein
particle size and esterification rate in apoB-lipoprotein-
depleted plasma (FERHDL). Clin Biochem 2001;34:583e8.
[24] Jia L, Long S, Fu M, Yan B, Tian Y, Xu Y, et al. Relationship
between total cholesterol/high-density lipoprotein choles-
terol ratio, triglyceride/high-density lipoprotein cholesterol
ratio, and high-density lipoprotein subclasses. Metabolism
2006;55:1141e8.
F. Hadaegh et al.
[25] Azizi F, Rahmani M, Ghanbarian A, Emami H, Salehi P,
Mirmiran P, et al. Serum lipid levels in an Iranian adults pop-
ulation: Tehran Lipid and Glucose Study. Eur J Epidemiol 2003;
18:311e9.
[26] Ghassemi H, Harrison G, Mohammad K. An accelerated nutri-
tion transition in Iran. Public Health Nutr 2002;5:149e55.
[27] Parks EJ. Effect of dietary carbohydrate on triglyceride
metabolism in humans. J Nutr 2001;131:2772Se4.
[28] Ezenwaka CE, Kalloo R. Carbohydrate-induced hyper-
triglyceridaemia among West Indian diabetic and non-diabetic
subjects after ingestion of three local carbohydrate foods.
Indian J Med Res 2005;121:23e31.
[29] Hadaegh F, Bozorgmanesh MR, Ghasemi A, Harati H, Saadat N,
Azizi F. High prevalence of undiagnosed diabetes
and abnormal glucose tolerance in the Iranian urban pop-
ulation: Tehran Lipid and Glucose Study. BMC Public Health
2008;8:176.
[30] Patel A, Barzi F, Jamrozik K, Lam TH, Ueshima H, Whitlock G,
et al. Serum triglycerides as a risk factor for cardiovascular
diseases in the Asia-Pacific region. Circulation 2004;110:
2678e86.