Skip to main content
  • Original investigation
  • Open access
  • Published:

log(TG)/HDL-C is related to both residual cardiometabolic risk and β-cell function loss in type 2 diabetes males



T2DM is associated with atherogenic dyslipidemia (AD), defined as decreased HDL-C plus raised triglycerides (TG). AD confers increased risk for CAD, even when LDL-C is at target. AD is rarely assessed due to lack of screening methods consensus.


To establish the prevalence and severity of AD from log(TG)/HDL-C in T2DM males, and to determine how it relates to cardiometabolic phenotype, glucose homeostasis, micro- and macrovascular complications, and 10-year UKPDS CV risk.


585 T2DM males divided according to quintiles (Q) of log(TG)/HDL-C. AD prevalence defined as HDL-C <40 mg.dL-1 plus TG ≥150 mg.dL-1. β-cell function assessed with HOMA.


Mean HDL-C and TG were 44 (13) and 204 (155) mg.dL-1. AD prevalence was 35%. AD correlated with lower β-cell function, with accelerated loss of insulin secretion, and with poorer HbA1c levels. AD was related to a high prevalence of CAD, and also to 10-year absolute CAD risk.


log(TG)/HDL-C is a simple means to estimate AD and the residual CV risk it confers in T2DM. AD closely associates with major cardiometabolic and glucose homeostasis determinants and poorer metabolic control. The ratio also relates to macroangiopathy prevalence and ranks future CAD risk, and is well-suited to capture non-LDL-related macrovascular residual risk and major glycemic determinants.


Current guidelines recommend aggressive treatment of low-density lipoprotein cholesterol (LDL-C) in patients with type 2 diabetes mellitus (T2DM) with statins as preferred agent [1]. Besides hypercholesterolemia, T2DM is associated with a specific non-LDL dyslipidemia, known as atherogenic dyslipidemia (AD). The hallmark of AD consists of decreased high-density lipoprotein cholesterol (HDL-C) together with raised triglycerides (TG). AD is associated with insulin resistance (IR), and confers a marked increase in residual vascular risk, even when LDL-C is low [29].

Screening for AD may provide clinically relevant information for assessing residual vascular risk associated with the common determinants of low HDL-C and high TG. However, this is rarely performed in practice, due to lack of agreement criteria or consensual cut-offs to routinely establish and grade AD based on measurements of HDL-C and fasting TG levels. Low HDL-C and high TG are part of the metabolic syndrome (MetS) definition, either as single AD components or in combination. Cordero et al. proposed gender-specific ratio's thresholds to correlate with MetS in non-diabetics [10]. Yet, despite their usefulness for diagnosing MetS and AD, low HDL-C and high TG are often dealt as separate, second-rank modifiable CV risk factors [5, 6, 11, 12].

A more comprehensive approach defines AD by the combined occurrence of high TG levels and low HDL-C. Yet, such seemingly easy estimation is hampered by (i) lack of consensual cut-off values across gender, ethnicities and underlying conditions (ii) requirement for baseline lipid values prior to lipid-lowering drugs (LLD), (iii) a sine qua non association does not take into account imbalances between the severity of the two non-LDL lipid abnormalities, underestimating the magnitude of AD in groups with spontaneously low (Afro-Americans, sub-Saharan Africans) or with elevated TG levels prior to LLD or insulin [1216].

Computing a ratio from fasting TG (numerator) and fasting HDL-C (denominator) appears as such intuitively more attractive than the combined occurrence to grade AD severity. Prior log transformation of TG levels allows for computing a broad range of TG values, as observed in everyday practice, from normal levels in subjects without AD to markedly hypertriglyceridemic states, such as MetS and many T2DM patients. Assessing AD with a log(TG)/HDL-C ratio also incorporates occurrence of mutually reinforcing or diverging confounders affecting both fraction's components, such as insulin-sensitizers, certain LLDs, or ethanol intake (which often increase both terms), or diet, exercise or menopausal status, which affect numerator and denominator in opposite directions [12, 1720].

The aim of the present study was to establish the prevalence, distribution and severity of AD from log(TG)/HDL-C in a large cohort of T2DM males. We also determined whether the ratio provides predictive information on cardiometabolic phenotype and glucose homeostasis determinants. Finally, we analyzed the association between AD and prevalence of micro- and macrovascular complications, as well as its impact on 10-year CV risk from the T2DM-specific United Kingdom Prospective Diabetes Study (UKPDS) calculator [2125].


The study design was cross-sectional and included 585 consecutive adult male outpatients with T2DM defined according to the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus[26] and followed in a single academic centre in Brussels (Belgium) between January 2008 and June 2010. Exclusion criteria were other forms of diabetes, severely impaired hepatic function, cancer, untreated thyroid conditions associated with hypo- or hyperthyroidism, malignancies and malabsorptive gastro-intestinal disorders. The following variables were recorded: age, ethnicity, socio-educational level, known diabetes duration, family history (cardiovascular diseases, diabetes mellitus), self-reported leisure-time (LT) weekly exertional exercise duration and daily duration spent watching screen(s) such as television, personal computer or visual numeric media, as surrogate for leisure-time sedentarity.

The presence of a metabolic syndrome (MetS score ≥3/5) was defined according to AHA/NHLBI criteria [11, 16]. The prevalence of AD was defined as the combination of MetS criteria for low HDL-C at baseline (< 40 mg.dL-1) and high fasting TG at baseline (≥150 mg.dL-1). The AD ratio log(TG)/HDL-C computed log(fasting TG) to fasting HDL-C. In patients treated with LLD(s), the TG and HDL-C values were the last available ones prior to LLD implementation. Normal values for log(TG)/HDL-C from a control cohort of 80 lean subjects without diabetes or family history for diabetes and untreated with LLD were: 0.036 (mean); 0.034 (median); 0.012 (SD); 0.014 (minimum); 0.067 (maximum); 0.029 (percentile 25) and 0.042 (percentile 75).

As regards bioanthropometry, patients were measured for body weight, height, body mass index (BMI), relative and total body fat (BodyFat Analyzer, Omron BF 500), waist circumference and conicity index (waist circumference (m)/0.109 √[weight (kg)/height (m)]) [2729]. Ultrasonographic evidence for fatty liver was considered in the presence of hyperreflectivity and in the absence of other etiological factors known to be associated with liver steatosis, including excess ethanol intake.

Computer-based Homeostasis Model Assessment (HOMA) of insulin sensitivity and beta-cell function was previously detailed Values of HOMA B (%) were plotted as a function of HOMA S (%), defining a hyperbolic product area (β × S) [unit: %2; normal value: 100%, corresponding to 104 %2], which represents the true, underlying β-cell function. (B × S) loss over a subject's lifetime span was obtained by dividing 100-(B × S) by subjects' age at the time of HOMA-modeling, providing an estimate of annual (B × S) loss rate (%.year-1) [15, 3034].

Medications history inquired upon current use of oral antidiabetic drugs, insulin, blood pressure (BP)-lowering drugs, aspirin, and any LLD(s)). Hypertension history was defined as systolic BP ≥140 mmHg and/or diastolic BP ≥90 mmHg or current treatment with BP-lowering drug(s) prescribed for treating high BP.

As regards macroangiopathy, coronary artery disease (CAD) was retrospectively inferred from medical history (myocardial infarction, angioplasty, stenting, revascularization surgery and/or significant coronary stenosis confirmed by angiography) and systematic review of all procedures, screening (exercise testing, echocardiography) or subclinical disease imaging data in the patient's records. Stroke was defined according to UK Prospective Diabetes Study (UKPDS) criteria, i.e. any neurological deficit lasting ≥1 month, no distinction being made between ischaemic, embolic and haemorrhagic strokes [21]. In patients with multiple CV events, only the first one was considered for prevalence. Peripheral artery disease (PAD) was defined by a well-documented medical history of lower-limb(s) claudication and/or clinical or imaging evidence for ischemic diabetic foot, angioplasty, stenting, revascularization surgery and/or significant lower-limb artery stenosis at Doppler ultrasonography and/or angiography. A diabetic retinopathy (DRP) was diagnosed following dilated fundus examination, with fluoangiography performed when deemed necessary by an in-house ophthalmologist. The presence of a peripheral neuropathy was based on clinical examination (knee and ankle reflexes, Semmes-Weinstein 5.07 monofilament test) and/or 4-limbs electromyography.

The following biologic variables were assessed: glycated haemoglobin (HbA1c), total cholesterol (C), HDL-C, TG, LDL-C (from Friedewald's formula), non-HDL-C (by subtracting HDL-C from total C), apolipoproteins (apo) A-I and B100 . As for non-lipid cardiometabolic markers, levels of high-sensitivity C-reactive protein (hsCRP), uric acid, cystatin C, fibrinogen, leucocytes count, total and free testosterone, sex-hormone-binding globulin (SHBG), ferritin, liver enzymes (aspartate aminotransferase [AST], alanine aminotransferase [ALT], γ-glutamyl transferase [γGT], homocysteine, folic acid and vitamin B12 were determined. Normo, micro- and macro-albuminuria were defined as urinary albumin excretion <30 (normo-), 30-299 (microalbuminuria) and ≥300 μ creatinine-1 (macro-albuminuria) from first-morning urine sample. Glomerular filtration rate was estimated (eGFR) using the Modified Diet in Renal Disease (MDRD) formula [35].

The UKPDS Risk Engine provided 10-year absolute CV risk estimates for T2DM individuals in primary CV prevention related to coronary artery disease (CAD), lethal CAD, stroke and lethal stroke, and was based on the 10 following variables: known T2DM duration, age, gender, ethnicity, smoking status, chronic atrial fibrillation, HbA 1c level, systolic BP, total C and HDL-C[2125]. Each patient gave informed consent, and the protocol was approved by the local Institutional Review Board.

Statistical methods

Results are presented as means (± 1 standard deviation (SD)) or proportions. The significance of differences between means was assessed by one-way analysis of variance for linear trend between means, and by a Chi-squared test trend for differences in proportions. Results were considered significant or non-significant (NS) for p < or > 0.05, respectively.


Patient's characteristics are described in Table 1. There were 585 male patients in the whole cohort, 89% of whom of Caucasian ancestry. Diabetes duration was 14 (9) years. The prevalence of AD, defined as the combination of baseline HDL-C <40 mg.dL-1 and TG ≥150 mg.dL-1 was 35%. When patients were divided according to AD quintiles of log(TG)/HDL-C, patients in the 1st AD quintile (Q I) had a mean log(TG)/HDL-C value similar to that of control, non-diabetic subjects (see Methods). T2DM patients from the 2nd AD quintile (Q II) had a mean log(TG)/HDL-C value corresponding to the 75th percentile of controls.

Table 1 Patients' characteristics

In the whole T2DM cohort, mean age (1 SD) was 65 (12) year, with a significant downward trend across AD quintiles. Both smoking and ethanol intake showed significant decreasing trends across quintiles. The proportion of patients reporting LT physical activity and screen-watching daily duration were significantly different across quintiles, suggestive of both a marked increase in sedentarity and a marked decrease in exercise practice. There were highly-significant trends across AD quintiles for progressively higher values of BMI, waist circumference, conicity index, waist-to-height ratio, fat mass and visceral fat. Liver steatosis was present in 68% of the whole cohort, with significant rising trends across quintiles, from 50% (Q I) to 83% (Q V) (p < 0.0001). There was a stepwise increase across quintiles of log(TG)/HDL-C in fasting insulinemia, from 80 to 132 pmol.l-1 (p < 0.0001). Insulin sensitivity and the hyperbolic product (B × S) were lower than the normal value (100%) in the whole cohort, with a mean value of 53% (HOMA S) and 28.5% (B × S), respectively. There were stepwise decreases across quintiles for both parameters, from 69% (HOMA S) and 35.7% (B × S) (Q I) to 39% (HOMA S) and 22.5% (B × S) (Q V) (both p < 0.0001). The B × S loss rate also showed a highly-significant trend toward worsening loss across AD quintiles, from 1.15%.yr-1 (Q I) to 1.44%.yr-1 (Q V) (p < 0.0001; Table 1).

Statin and fibrate were given to 54 and 21% of the whole cohort, with a significant trend for increased fibrate use across quintiles, from 9% (Q I) to 46% (Q V) (p < 0.0001; Table 1). None of the patients were treated with niacin. In the whole cohort, macroangiopathy was present in 40%, as CAD (29%), TIA/stroke (9%) and/or PAD (12%) (Table 1). There were significant trends for higher prevalence of overall macroangiopathy, CAD and TIA/stroke across AD quintiles, from 29% (overall), 15% (CAD) and 5% (TIA/stroke) in Q I to 48% (overall), 33% (CAD) and 13% (TIA/stroke) in Q V (p 0.060, 0.0097 and 0.0129, respectively). No trend was observed for PAD across AD quintiles. After adjustment for inter-quintile differences in age, overall macroangiopathy, CAD and TIA/stroke prevalence increased even more significantly across quintiles, from 28%, 15% and 6% (Q I) to 52%, 36% and 14% (Q V), respectively (p 0.0010 (overall), 0.0024 (CAD), and 0.0126 (TIA/stroke)).

As regards glucose-lowering therapies, 59, 47, 7, 3, and 42% were receiving metformin, a β-cell stimulant, a glitazone, a DPP4-inhibitor and/or insulin, without relevant trends between quintiles. Mean HbA1c in the whole cohort was 7.53 (1.53)%, with 43% of patients at HbA1c target (< 7.0%). There was a non-significant rise in HbA1c across AD quintiles, from 7.32% (Q I) to 7.85% (Q V), and a significant decrease across AD quintiles in proportion of patients reaching target HbA1c, from 49% (Q I) to 35% (Q V; p 0.0155; Table 2).

Table 2 Laboratory values

The contribution of discrete components to the MetS score for each AD quintile is depicted on Figure 1. A MetS phenotype (score ≥3/5) was present in 76% in the whole cohort. There was a highly-significant stepwise rise in both MetS prevalence and score across AD quintiles, from 36% and 2.3 (Q I) to 97% and 4.4 (Q V) (both p < 0.0001; Table 1 and Figure 1). There were no significant trends for higher prevalence of hypertension or enlarged waist prevalence across quintiles. In the whole cohort, hypertension prevalence was 86% and mean systolo-diastolic blood pressure values were 139 (19) - 80 (11) mmHg, without trends across quintiles.

Figure 1
figure 1

Contribution of individual components of the metabolic syndrome (MetS) phenotype to the MetS cumulative score in 585 male patients with T2DM, divided into quintiles ( n = 117; QI to QV) based on log (TG)/HDL-C ranking distribution. HBP: high blood pressure; HDL-C: high-density lipoprotein cholesterol; TG: triglycerides.

In the whole cohort, mean LDL-C was 98 (35) mg.dL-1, HDL-C was 44 (13) mg.dL-1, and current TG levels were 171 (119) mg.dL-1. Baseline, pre-LLD TG values were 204 (155) mg.dL-1. Non-HDL-C was on average 131 (40) mg.dL-1, and the delta between non-HDL-C and LDL-C was 31 (18) mg.dL-1, suggesting elevated number of both apoB100-carrying particles and TG-rich lipoproteins enrichment. There were significant trends for decreasing values across AD quintiles for LDL-C, apoA-I, and LDL-C/apoB100 ratio as surrogate for LDL size (p 0.0047, < 0.0001, and < 0.0001, respectively). There were significant trends as well for increasing values across AD quintiles for non-HDL-C, non-HDL-C minus LDL-C, and apoB100 (all p < 0.0001; Table 2).

As regards non-lipids markers, there were significant trends for increasing levels across quintiles for hsCRP, uric acid, cystatin C, leucocytes and ALT. Significant decreases across quintiles were observed for SHBG and vitamin B12 levels (p < 0.0001) (Table 2) as well as for total testosterone (11.5 (4.4) [Q I], 11.2 (5.9) [Q II], 10.6 (4.1) [Q III], 9.7 (3.9) [Q IV] and 10.4 (4.9) nmol.L-1 [Q V] (p 0.0331)). Mean values for the non-lipid parameters in the whole cohort were: fibrinogen 316 (76) mg.dL-1, ferritin 206 (189) ng.mL-1, free testosterone 0.19 (0.12) nmol.L-1, homocysteine 14.1 (6.9) μmol.L-1, folic acid 7.7 (5.2) ng.mL-1, AST 26 (11) IU.L-1, and γGT 43 (48) IU.L-1, with no trends across quintiles.

Diabetic retinopathy and peripheral polyneuropathy were diagnosed in 22 and 30% of the whole cohort, with no trend for higher prevalence across AD quintiles. Such lack of difference across quintiles remained after adjustment for inter-quintile differences in mean age. In the whole cohort, eGFR was 80 (27) ml.min-1.1.73 m2, with a significant decrease across AD quintiles, from 85 (Q I) to 73 ml.min-1.1.73 m2 (Q V; p 0.0105). There were significant increases across quintiles in mean albuminuria values and in albuminuria prevalence (both p < 0.0001; Table 2).

In the whole cohort, 331 patients were in primary CV prevention, and therefore eligible for UKPDS risk estimation. Their mean 10-year absolute CV risk prediction was: 22 (16)% (CAD); 15 (14)% (fatal CAD); 14 (19)% (stroke) and 2 (3)% (fatal stroke). Figure 2 depicts UKPDS risk for CAD after adjustment for inter-quintile differences in mean age. There were significant trends for a markedly heightened risk across AD quintiles, from 15% (CAD) and 10% (fatal CAD) in Q I up to 29% (CAD) and 20% (fatal CAD) in Q V (p < 0.0001 and 0.0004). No such trends across quintiles were observed for risk of stroke or lethal stroke (not illustrated).

Figure 2
figure 2

UKPDS Risk Engine 10-year absolute predicted risk of developing non-fatal or fatal coronary artery disease (CAD; grey bars ) or fatal CAD ( solid bars ) in 331 T2DM males according to quintiles of log (TG)/HDL-C ranking distribution. Within each quintile, data were obtained from subsets of patients in primary cardiovascular prevention: QI: n = 83; QII: n = 59; QIII: n = 65; QIV: n = 65; and QV: n = 59. HDL-C: high-density lipoprotein cholesterol; TG: triglycerides. Significance of differences between quintiles: P < 0.0001 (CAD) and P = 0.0004 (fatal CAD).


The main findings of the present study are fourfold. Firstly, we observed a 35% prevalence of AD in T2DM males. Secondly, the AD ratio log(TG)/HDL-C proved an easy means to estimate IR, and was strongly associated with IR and its cardiometabolic phenotype, including fat distribution, liver steatosis and sedentarity markers. Thirdly, the ratio was linked to a marked gradient for residual risk of future CAD. Fourthly, the ratio related to insulin secretion and β-cell function loss rate in T2DM, and as such a higher ratio was associated with non-attainment of target HbA1c. This AD ratio is therefore exquisitely suited to assess both non-LDL residual vascular risk and glucose homeostasis determinants in diabetic patients.

In the common form of T2DM, i.e. associated with MetS and IR, the hallmark of AD is decreased HDL-C level together with raised TG, while LDL-C is often only marginally elevated. Further to these obvious non-LDL abnormalities, there usually coexists an array of quantitative and qualitative lipoproteins abnormalities, not evidenced from standard lipid assessment. These include (i) a decrease in LDL size, often obscuring the absolute increase in LDL particles number; (ii) an increase in fasting and/or postprandial TG-rich apoB48 and/or apoB100-carrying lipoproteins and their remnants; (iii) a relative shift toward hepatic overproduction of apoCIII-carrying VLDL and (iv) detrimental changes in HDL quality and natural history (eg. lesser atheroprotective subclasses, TG enrichment, reduced half-life and decreased nascent HDL production) [2, 5, 12, 13, 3639].

In the present study, current TG levels, non-HDL-C levels, non-HDL-C minus LDL-C, and the ratio of LDL-C to apoB100 confirm the high prevalence of AD-related abnormalities in this mostly North-Caucasian population of T2DM males. AD was also associated with high number of apoB100-carrying particles and with markers of lipoprotein TG-enrichment [2, 12, 36, 37]. Using consensual MetS cutoffs for defining decreased HDL-C and elevated TG [11, 16], we observed a high prevalence of AD in T2DM males, at more than one-third. AD is however more likely to behave as a continuous CV risk variable. This underlies the rationale to use the combined occurrence of low HDL-C and high TG as dichotomic variable in order to assess AD prevalence in a population, while the log(TG)/HDL-C appears more appropriate to rank AD severity as a continuous variable in a given individual [12]. This is confirmed from comparison with values from control, non-diabetic subjects, showing that in T2DM, the AD ratio was well above normal values from the third AD quintile onwards, hinting to the fact that >60% within this population presents with some form of AD. This was not unexpected, TG and HDL-C being used to define the MetS phenotype, the latter being highly-prevalent (76%) of this T2DM population.

Whereas the Atherogenic Index of Plasma (log(TG/HDL-C) uses log transformation of the whole ratio, the present study only applied log transformation to TG numerator, as HDL-C range per se did not justify transformation. Skewness from untransformed TG/HDL-C distribution frequency was expected, with the TG term accounting for most of it. Whereas skewness of HDL-C frequency distribution was statistically significant, it was technically unimportant, and not surprisingly, HDL-C is rarely subject to prior log transformation neither in routine clinical assessment nor in clinical trials [12, 1720]. Dealing with a male-only population offered the advantage of avoiding bimodality in HDL-C and TG distributions due to gender differences (T2DM females exhibiting higher mean HDL-C and lower TG values (by an average 6 and 12 mg/dl, respectively; (data not shown)).

In order to rule out any confounding effect of LLD, we used baseline TG and HDL-C values to compute the ratio, prior to any anti-dyslipidemic drug(s), such as statins, fibrates or niacin, all of which being increasingly used in combination therapy for T2DM. The difference between baseline and current TG levels in this study averaged 33 mg.dL-1, supporting the rationale for such an unyielding approach to obtain log(TG)/HDL-C. This also allowed to establishing precisely the true, underlying magnitude and frequency of AD in T2DM. Regarding lifestyle-related confounders on AD prevalence, it is noteworthy that both smoking and ethanol intake showed significant decreasing trends across quintiles, the inverse association with ethanol being expected, due to its HDL-C-raising effects.

For non-lipid biochemistry the AD ratio was associated, in a stepwise gradation, with a series of emerging non-lipid cardiometabolic markers, all comorbid to T2DM, such as fasting insulinemia, hsCRP, uric acid, cystatin C, leucocytes count and glomerular filtration rate. SHBG and total testosterone, but not free testosterone, significantly decreased in parallel across AD quintiles, an expected finding since stepwise insulin resistance is associated with decreased levels of the androgen transporter [24, 25, 4042].

IR develops in a conditional genetic or acquired environment associated with overall and central body fat, sedentarity, and skeletal sarcopenia. This array of metabolic abnormalities predisposes to CV risk, and is easily captured clinically by (i) the presence of a MetS phenotype and (ii) by the incremental MetS score [11, 12, 16, 43, 44]. Numerous abnormalities defining AD are associated, causally or as markers, with IR and its compensatory portal hyperinsulinemia, and with liver steatosis, the upstream organ responsible of excess VLDL synthesis and export [12]. The present results show that log(TG)/HDL-C is an easy means to rank T2DM patients alongside IR, liver steatosis, leisure-time sedentarity (with screen-watching as surrogate), lack of leisure-time physical activity, and MetS score. This extends previous findings showing that this ratio was a surrogate for IR in healthy African Americans and white Caucasians [45]. In our study, log(TG)/HDL-C also associated with both prevalent CAD, prevalent stroke but neither prevalent eye or nerve microangiopathy, nor stroke risk, nor prevalent PAD, the two latter caused by a combination of micro- and macrovascular end-organ damage. On the other end, albuminuria prevalence or level, both markers of mixed micro/macrovascular end-organ damage in T2DM, were significantly associated with AD.

In the ACCORD Lipid trial, impacting upon AD using a simvastatin-fenofibrate combination therapy reduced macrovascular risk in a subset patients with AD defined by concurrent HDL-C <1st tertile plus TG >3rd tertile of baseline cohort lipids. This was in line with previous beneficial impact of fibrates on macrovascular outcomes in post-hoc subgroup analyses from landmark fibrate trials [12, 13, 44, 4648]. By contrast, in the ACCORD Lipid Eye substudy, microvascular risk of retinopathy progression was markedly reduced by combination simvastatin-fenofibrate therapy in a broader range of baseline non-LDL lipids [49]. Besides fibrates, another promising pharmaceutical means to improve the AD ratio is to raise HDL-C levels with niacin or cholesteryl ester transfer protein inhibitors [5]. Parallel to pharmaceutical reinforcement, therapeutic lifestyle intervention emphasizing on a diet low in carbohydrates and ethanol, on caloric restriction and expenditure through exercise should not be overlooked as a means to improve the AD ratio, which in this study was clearly associated with sedentarity and lack of leisure-time exercise.

The AD ratio was also linked with more severe impairment in endogenous insulin secretion, as well as worsening of hyperbolic product (B × S) loss, heralding earlier and more intensive requirement for stepping-up of glucose-lowering therapies, including insulin [14, 15, 30, 34, 50]. This unexpected finding is all the more relevant, as the ratio simultaneously associates with more severe IR and faster insulin secretory loss. Such a detrimental combination is poised to lead to poorer glycemic control. This association between log(TG)/HDL-C and β-cell impairment does not simply reflect decreased insulin output due to β-cell loss, as the latter would bring about opposing effects on the ratio terms, with reduced liver TG synthesis and VLDL export (portal hypoinsulinemia) and reduced lipolysis of TG-rich particles (systemic hypoinsulinemia). In addition, systemic TG removal from VLDL may also be altered as a result of decreased insulin sensitivity affecting lipoprotein lipase [12]. Regarding glycemic control, T2DM patients with elevated raised AD ratio qualify for more proactive approaches to control hyperglycemia with insulin-sensitizers and/or insulin replacement therapy.

This study has several limitations. Firstly, we analyzed only male patients. Secondly, the cross-sectional design does not allow establishing causality relationships. Third, incorporation of routine measurement of HDL-C as denominator does not allow distinguishing dysfunctional, less atheroprotective or even atherogenic HDL subclasses which may be more prevalent in patients with T2DM. Besides compositional changes in HDL induced by overproduction of TG-rich lipoproteins [12], loss of functionality may also result from oxidative changes in predisposed individuals, such as those with haptoglobin 2-2 genotype, not captured from simple measurement of HDL-C level [51]. Nevertheless, dysfunctional subclasses often coexist alongside decreased HDL-C, so that a low HDL-C level in the ratio may reflect to some extent underlying qualitative defects [52]. Finally, the population under study was mostly of European Caucasian ancestry. The conclusions need to be confirmed in other major racial/ethnic diabetic and nondiabetic subgroups. The validity of the TG/HDL-C ratio as a simple, clinically useful estimator of IR was reported by Li et al. across race/ethnic subgroups in a large, nationwide sample of nondiabetic subjects from three major US populations (non-Hispanic whites, blacks and Mexican Americans) [53].

Based on the present results, we consider log(TG)/HDL-C as a simple means to estimate AD, and the residual vascular risk it confers, in fully-treated T2DM male patients. In addition, AD closely associates with the two major determinants of glucose homeostasis, affecting the natural history of hyperglycemia and its metabolic control. In patients in primary CV prevention, this AD ratio also yielded a marked, stepwise increase, in 10-year CV risk for CAD and fatal CAD. Using the log(TG)/HDL-C therefore adds a wealth of pertinent information on residual macrovascular risk, beyond that provided by individual components of the available lipid profile. In T2DM, this AD ratio is a simple means to assess both non-LDL-related residual vascular risk and glucose homeostasis.


  1. American Diabetes Association: Standards of medical care in diabetes - 2009. Diabetes Care. 2009, 32: S13-61. 10.2337/dc09-S013.

    Article  PubMed Central  Google Scholar 

  2. Davignon J, Cohn JS: Triglycerides: a risk factor for coronary heart disease. Atherosclerosis. 1996, 124: S57-64. 10.1016/0021-9150(96)05858-3.

    Article  CAS  PubMed  Google Scholar 

  3. Brites FD, Bonavita CD, De Geitere C, Cloës M, Delfly B, Yael MJ, Fruchart J, Wikinski RW, Castro GR: Alterations in the main steps of reverse cholesterol transport in male patients with primary hypertriglyceridemia and low HDL-cholesterol levels. Atherosclerosis. 2000, 152: 181-92. 10.1016/S0021-9150(99)00452-9.

    Article  CAS  PubMed  Google Scholar 

  4. da Luz PL, Favarato D, Faria-Neto JR, Lemos P, Chagas AC: High ratio of triglycerides to HDL-cholesterol predicts extensive coronary disease. Clinics (Sao Paulo). 2008, 63: 427-32.

    Google Scholar 

  5. Fruchart J-C, Sacks F, Hermans MP, Assmann G, Brown WV, Ceska R, Chapman MJ, Dodson PM, Fioretto P, Ginsberg HN, Kadowaki T, Lablanche J-M, Marx N, Plutzky J, Reiner Z, Rosenson RS, Staels B, Stock JK, Sy R, Wanner C, Zambon A, Zimmet P, for the Residual Risk Reduction Initiative: The Residual Risk Reduction Initiative: a call to action to reduce residual vascular risk in patients with dyslipidemia. Am J Cardiol. 2008, 102: S1-34. 10.1016/j.amjcard.2008.10.002.

    Article  Google Scholar 

  6. Jones PH: Expert perspective: reducing cardiovascular risk in metabolic syndrome and type 2 diabetes mellitus beyond low-density lipoprotein cholesterol lowering. Am J Cardiol. 2008, 102: 41L-47L. 10.1016/j.amjcard.2008.09.074.

    Article  PubMed  Google Scholar 

  7. Rana JS, Visser ME, Arsenault BJ, Després JP, Stroes ES, Kastelein JJ, Wareham NJ, Boekholdt SM, Khaw KT: Metabolic dyslipidemia and risk of future coronary heart disease in apparently healthy men and women: The EPIC-Norfolk prospective population study. Int J Cardiol. 2010, 143: 399-404. 10.1016/j.ijcard.2009.03.123.

    Article  PubMed  Google Scholar 

  8. Sarwar N, Sandhu MS, Ricketts SL, Butterworth AS, Di Angelantonio E, Boekholdt SM, Ouwehand W, Watkins H, Samani NJ, Saleheen D, Lawlor D, Reilly MP, Hingorani AD, Talmud PJ, Danesh J: Triglyceride Coronary Disease Genetics Consortium and Emerging Risk Factors Collaboration. Triglyceride-mediated pathways and coronary disease: collaborative analysis of 101 studies. Lancet. 2010, 375: 1634-9. 10.1016/S0140-6736(10)60545-4.

    Article  CAS  PubMed  Google Scholar 

  9. Triglyceride Coronary Disease Genetics Consortium and Emerging Risk Factors Collaboration: Triglyceride-mediated pathways and coronary disease: collaborative analysis of 101 studies. Lancet. 2010, 375: 1634-9. 10.1016/S0140-6736(10)60545-4.

    Article  PubMed Central  Google Scholar 

  10. Cordero A, Laclaustra M, León M, Casasnovas JA, Grima A, Luengo E, Ordoñez B, Bergua C, Bes M, Pascual I, Alegría E, MESYAS Registry Investigators: Comparison of serum lipid values in subjects with and without the metabolic syndrome. Am J Cardiol. 2008, 102: 424-8. 10.1016/j.amjcard.2008.03.079.

    Article  CAS  PubMed  Google Scholar 

  11. Grundy SM, Cleeman JI, Daniels SR, Donato KA, Eckel RH, Franklin BA, Gordon DJ, Krauss RM, Savage PJ, Smith SC, Spertus JA, Costa F, American Heart Association; National Heart, Lung, and Blood Institute: Diagnosis and management of the metabolic syndrome: an American Heart Association/National Heart, Lung, and Blood Institute Scientific Statement. Circulation. 2005, 112: 2735-52. 10.1161/CIRCULATIONAHA.105.169404.

    Article  PubMed  Google Scholar 

  12. Hermans MP, Fruchart JC: Reducing residual vascular risk in patients with atherogenic dyslipidaemia: where do we go from here?. Clinical Lipidology Review. 2010.

    Google Scholar 

  13. ACCORD Study Group, Ginsberg HN, Elam MB, Lovato LC, Crouse JR, Leiter LA, Linz P, Friedewald WT, Buse JB, Gerstein HC, Probstfield J, Grimm RH, Ismail-Beigi F, Bigger JT, Goff DC, Cushman WC, Simons-Morton DG, Byington RP: Effects of combination lipid therapy in type 2 diabetes mellitus. N Engl J Med. 2010, 362: 1563-74. 10.1056/NEJMoa1001282.

    Article  Google Scholar 

  14. Dehout F, Haumont S, Gaham N, Amoussou-Guenou KD, Hermans MP: Metabolic syndrome in Bantu subjects with type 2 diabetes from sub-Saharan extraction: prevalence, gender differences and HOMA-hyperbolic product. Diab Metab Syndr. 2008, 2: 5-11.

    Article  Google Scholar 

  15. Munoko Th, Hermans MP: Phenotypic characterization of first generation Maghrebian migrants with type 2 diabetes: a gender-based comparison with a reference North-Caucasian Belgian cohort. Diab Met Syndr. 2008, 2: 115-24. 10.1016/j.dsx.2008.02.004.

    Article  Google Scholar 

  16. Sadikot S, Hermans MP: Here we go again ... The metabolic syndrome revisited!. Diab Metab Syndr. 2010, 4: 111-20. 10.1016/j.dsx.2010.05.011.

    Article  Google Scholar 

  17. Dobiásová 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: 583-8.

    Article  PubMed  Google Scholar 

  18. Kannel WB, Vasan RS, Keyes MJ, Sullivan LM, Robins SJ: Usefulness of the triglyceride-high-density lipoprotein versus the cholesterol-high-density lipoprotein ratio for predicting insulin resistance and cardiometabolic risk (from the Framingham Offspring Cohort). Am J Cardiol. 2008, 101: 497-501. 10.1016/j.amjcard.2007.09.109.

    Article  PubMed Central  CAS  PubMed  Google Scholar 

  19. Bittner V, Johnson BD, Zineh I, Rogers WJ, Vido D, Marroquin OC, Bairey-Merz CN, Sopko G: The triglyceride/high-density lipoprotein cholesterol ratio predicts all-cause mortality in women with suspected myocardial ischemia: a report from the Women's Ischemia Syndrome Evaluation (WISE). Am Heart J. 2009, 157: 548-55. 10.1016/j.ahj.2008.11.014.

    Article  PubMed Central  CAS  PubMed  Google Scholar 

  20. Cordero A, Andrés E, Ordoñez B, León M, Laclaustra M, Grima A, Luengo E, Moreno J, Bes M, Pascual I, Civeira F, Pocoví M, Alegría E, Casasnovas JA, MESYAS (MEtabolic Syndrome Active Subjects) Study Investigators: Usefulness of triglycerides-to-high-density lipoprotein cholesterol ratio for predicting the first coronary event in men. Am J Cardiol. 2009, 104: 1393-7. 10.1016/j.amjcard.2009.07.008.

    Article  CAS  PubMed  Google Scholar 

  21. Kothari V, Stevens RJ, Adler AI, Stratton IM, Manley SE, Neil HA: UKPDS 60: risk of stroke in type 2 diabetes estimated by the UK Prospective Diabetes Study risk engine. Stroke. 2002, 33: 1776-81. 10.1161/01.STR.0000020091.07144.C7.

    Article  PubMed  Google Scholar 

  22. Stevens RJ, Coleman RL, Adler AI, Stratton IM, Matthews DR, Holman RR: UKPDS 66: risk factors for myocardial infarction case fatality and stroke case fatality in type 2 diabetes. Diabetes Care. 2004, 27: 201-7. 10.2337/diacare.27.1.201.

    Article  PubMed  Google Scholar 

  23. Hermans MP, Ahn S, Rousseau MF: Neurohormonal biomarkers and UKPDS stroke risk in type 2 diabetic women on primary cardiovascular prevention. Diabetes Metab Syndr. 2008, 2: 93-8. 10.1016/j.dsx.2008.02.001.

    Article  Google Scholar 

  24. Hermans MP, Ahn SA, Rousseau MF: Erectile dysfunction, microangiopathy and UKPDS risk in type 2 diabetes. Diabetes Metab. 2009, 35: 484-9. 10.1016/j.diabet.2009.06.003.

    Article  PubMed  Google Scholar 

  25. Hermans MP, Ahn SA, Rousseau MF: Cardiometabolic phenotype and UKPDS risk in male type 2 diabetic patients with obstructive sleep apnoea syndrome. Diabetes Metab Syndr. 2009, 3: 50-4. 10.1016/j.dsx.2008.10.011.

    Article  Google Scholar 

  26. Report of the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Diabetes Care. 1997, 20: 1183-97.

  27. Valdez R, Seidell JC, Ahn YI, Weiss KM: A new index of abdominal adiposity as an indicator of risk for cardiovascular disease: a cross-population study. Int J Obes. 1992, 16: 77-82.

    Google Scholar 

  28. Kyle UG, Morabia A, Schutz Y, Pichard C: Sedentarism affects body fat mass index and fat-free mass index in adults aged 18 to 98 years. Nutrition. 2004, 20: 255-60. 10.1016/j.nut.2003.11.019.

    Article  PubMed  Google Scholar 

  29. Burkhauser RV, Cawley J: Beyond BMI: the value of more accurate measures of fatness and obesity in social science research. J Health Econ. 2008, 27: 519-29. 10.1016/j.jhealeco.2007.05.005.

    Article  PubMed  Google Scholar 

  30. Levy J, Atkinson AB, Bell PM, McCance DR, Hadden DR: Beta-cell deterioration determines the onset and rate of progression of secondary dietary failure in type 2 diabetes mellitus: the 10-year follow-up of the Belfast Diet Study. Diabet Med. 1998, 15: 290-6. 10.1002/(SICI)1096-9136(199804)15:4<290::AID-DIA570>3.0.CO;2-M.

    Article  CAS  PubMed  Google Scholar 

  31. Levy JC, Matthews DR, Hermans MP: Correct Homeostasis Model Assessment (HOMA) evaluation uses the computer program. Diabetes Care. 1998, 21: 2191-92. 10.2337/diacare.21.12.2191.

    Article  CAS  PubMed  Google Scholar 

  32. Hermans MP, Levy J, Morris RJ, Turner RC: Comparison of tests of beta-cell function across a range of glucose tolerance from normal to diabetes. Diabetes. 1999, 48: 1779-86. 10.2337/diabetes.48.9.1779.

    Article  CAS  PubMed  Google Scholar 

  33. Hermans MP, Levy J, Morris RJ, Turner RC: Comparison of insulin sensitivity tests across a range of glucose tolerance from normal to diabetes. Diabetologia. 1999, 42: 678-87. 10.1007/s001250051215.

    Article  CAS  PubMed  Google Scholar 

  34. Hermans MP: Diabetic macro- and microvascular disease in type 2 diabetes. Diabetes Vasc Dis Res. 2007, 4: S7-11. 10.3132/dvdr.2007.019.

    Article  Google Scholar 

  35. Levey AS, Bosch JP, Lewis JB, Greene T, Rogers N, Roth D: A more accurate method to estimate glomerular filtration rate from serum creatinine: A new prediction equation. Modification of Diet in Renal Disease Study Group. Ann Intern Med. 1999, 130: 461-70.

    Article  CAS  PubMed  Google Scholar 

  36. Hermans MP, Ahn SA, Rousseau MF: The non-HDL-C/HDL-C ratio provides cardiovascular risk stratification similar to the ApoB/ApoA1 ratio in diabetics: comparison with reference lipid markers. Diab Metab Syndr. 2007, 1: 23-8. 10.1016/j.dsx.2006.11.003.

    Article  Google Scholar 

  37. Brunzell JD, Davidson M, Furberg CD, Goldberg RB, Howard BV, Stein JH, Witztum JL: Lipoprotein management in patients with cardiometabolic risk: consensus conference report from the American Diabetes Association and the American College of Cardiology Foundation. J Am Coll Cardiol. 2008, 51: 1512-24. 10.1016/j.jacc.2008.02.034.

    Article  PubMed  Google Scholar 

  38. Mendivil CO, Zheng C, Furtado J, Lel J, Sacks FM: Metabolism of very-low-density lipoprotein and low-density lipoprotein containing apolipoprotein C-III and not other small apolipoproteins. Arterioscler Thromb Vasc Biol. 2010, 30: 239-45. 10.1161/ATVBAHA.109.197830.

    Article  PubMed Central  CAS  PubMed  Google Scholar 

  39. Rosenson RS, Davidson MH, Pourfarzib R: Underappreciated opportunities for low-density lipoprotein management in patients with cardiometabolic residual risk. Atherosclerosis. 2010, 213: 1-7. 10.1016/j.atherosclerosis.2010.03.038.

    Article  CAS  PubMed  Google Scholar 

  40. Hermans MP, Ahn SA, Rousseau MF: Raised natriuretic peptides, Big-endothelin-1 and improved beta-cell function in type 2 diabetic males with hyperuricaemia. Diab Vasc Dis Res. 2009, 6: 190-3. 10.1177/1479164109336689.

    Article  PubMed  Google Scholar 

  41. Jerums G, Premaratne E, Panagiotopoulos S, Clarke S, Power DA, MacIsaac RJ: New and old markers of progression of diabetic nephropathy. Diabetes Res Clin Pract. 2008, 82: S30-7. 10.1016/j.diabres.2008.09.032.

    Article  CAS  PubMed  Google Scholar 

  42. Hermans MP, Ahn SA, Gruson D, Rousseau MF: The metabolic syndrome phenotype is associated with raised circulating Big endothelin-1 independently of coronary artery disease in type 2 diabetes. Diab Metab Syndr. 2007, 1: 229-37. 10.1016/j.dsx.2007.09.001.

    Article  Google Scholar 

  43. Hermans MP, Ahn SA, Amoussou-Guenou KD, Rousseau MF: Impact of metabolic syndrome on microvascular complications in type 2 diabetes. Diab Metab Syndr. 2010, 4: 150-4. 10.1016/j.dsx.2010.05.021.

    Article  Google Scholar 

  44. Hermans MP: Impact of fenofibrate on type 2 diabetes patients with features of the metabolic syndrome: subgroup analysis from FIELD. Curr Cardiol Rev. 2010, 6: 112-8. 10.2174/157340310791162686.

    Article  PubMed Central  CAS  PubMed  Google Scholar 

  45. Kim-Dorner SJ, Deuster PA, Zeno SA, Remaley AT, Poth M: Should triglycerides and the triglycerides to high-density lipoprotein cholesterol ratio be used as surrogates for insulin resistance?. Metabolism. 2010, 59: 299-304. 10.1016/j.metabol.2009.07.027.

    Article  CAS  PubMed  Google Scholar 

  46. Ducobu J, Scheen A, Van Gaal L, Velkeniers B, Hermans M: Belgian Expert opinion: how to reduce the residual risk in atherogenic dyslipidemic patients: place of fibrates. Acta Cardiol. 2008, 63: 235-48. 10.2143/AC.63.2.2029534.

    Article  PubMed  Google Scholar 

  47. Fruchart J-C, Sacks FM, Hermans MP: Implications of the ACCORD lipid study: perspective from the Residual Risk Reduction Initiative (R3i). Curr Med Res Opin. 2010, 26: 1793-7. 10.1185/03007995.2010.489341.

    Article  CAS  PubMed  Google Scholar 

  48. Jun M, Foote C, Lv J, Neal B, Patel A, Nicholls SJ, Grobbee DE, Cass A, Chalmers J, Perkovic V: Effects of fibrates on cardiovascular outcomes: a systematic review and meta-analysis. Lancet. 2010, 375: 1875-84. 10.1016/S0140-6736(10)60656-3.

    Article  CAS  PubMed  Google Scholar 

  49. The ACCORD Study Group and ACCORD Eye Study Group: ACCORD Eye. Effects of medical therapies on retinopathy progression in type 2 diabetes. N Engl J Med. 2010, 363: 233-44. 10.1056/NEJMoa1001288.

    Article  Google Scholar 

  50. Turner RC, Cull CA, Frighi V, Holman RR: Glycemic control with diet, sulfonylurea, metformin or insulin in patients with type 2 diabetes mellitus: progressive requirement for multiple therapies (UKPDS 49). UK Prospective Diabetes Study (UKPDS) Group. JAMA. 1999, 281: 2005-12. 10.1001/jama.281.21.2005.

    Article  CAS  PubMed  Google Scholar 

  51. Asleh R, Blum S, Kalet-Litman S, Alshiek J, Miller-Lotan R, Asaf R, Rock W, Aviram M, Milman U, Shapira C, Abassi Z, Levy AP: Correction of HDL dysfunction in individuals with diabetes and the haptoglobin 2-2 genotype. Diabetes. 2008, 57: 2794-800. 10.2337/db08-0450.

    Article  PubMed Central  CAS  PubMed  Google Scholar 

  52. Sorrentino SA, Besler C, Rohrer L, Meyer M, Heinrich K, Bahlmann FH, Mueller M, Horváth T, Doerries C, Heinemann M, Flemmer S, Markowski A, Manes C, Bahr MJ, Haller H, von Eckardstein A, Drexler H, Landmesser U: Endothelial-vasoprotective effects of high-density lipoprotein are impaired in patients with type 2 diabetes mellitus but are improved after extended-release niacin therapy. Circulation. 2010, 121: 110-22. 10.1161/CIRCULATIONAHA.108.836346.

    Article  CAS  PubMed  Google Scholar 

  53. Li C, Ford ES, Meng YX, Mokdad AH, Reaven GM: Does the association of the triglyceride to high-density lipoprotein cholesterol ratio with fasting serum insulin differ by race/ethnicity?. Cardiovasc Diabetol. 2008, 7: 4-10.1186/1475-2840-7-4.

    Article  PubMed Central  PubMed  Google Scholar 

Download references

Author information

Authors and Affiliations


Corresponding author

Correspondence to Michel P Hermans.

Additional information

Competing interests

The authors declare that they have no competing interests.

Authors' contributions

All authors read and approved the final manuscript: MPH collected and managed the T2DM patients database; MPH, SAA and MFR contributed equally to the study design, data and statistical analyses and to drafting the manuscript.

Authors’ original submitted files for images

Below are the links to the authors’ original submitted files for images.

Authors’ original file for figure 1

Authors’ original file for figure 2

Rights and permissions

Open Access This article is published under license to BioMed Central Ltd. This is an Open Access article is distributed under the terms of the Creative Commons Attribution 2.0 International License (, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Reprints and permissions

About this article

Cite this article

Hermans, M.P., Ahn, S.A. & Rousseau, M.F. log(TG)/HDL-C is related to both residual cardiometabolic risk and β-cell function loss in type 2 diabetes males. Cardiovasc Diabetol 9, 88 (2010).

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: