Leptin/adiponectin ratio in patients with coronary heart disease: comparing subjects with and without metabolic syndrome

Abstract

Background

Adiponectin and leptin are adipose tissue-derived hormones, shown to have opposing associations with the metabolic syndrome and coronary heart disease (CHD). This study evaluated the association between the leptin/adiponectin ratio and the components of the metabolic syndrome in a cohort with CHD.

Methods and results

This cross-sectional study included data from 105 subjects (men = 91), undergoing first-time elective coronary artery bypass grafting (CABG). Leptin and adiponectin concentrations were determined by enzyme-linked immunosorbent assay (ELISA). Association was found between the leptin/adiponectin ratio and homeostatic model assessment (HOMA) (r _s = 0.34, P = 0.0006), fasting insulin concentrations (r _s = 0.37, P = 0.0001), fasting glucose concentrations (r _s = 0.24, P = 0.01), systolic blood pressure (r _s = 0.20, P = 0.05), diastolic blood pressure (r _s = 0.24, P = 0.02), waist circumference (r _s = 0.55, P < 0.0001), body mass index (BMI) (r _s = 0.55, P < 0.0001) and waist/hip ratio (r _s = 0.38, P = 0.0001). A significant difference was found in ratios between those with and without insulin resistance (HOMA > 3 and HOMA ≤ 3) (P = 0.029) and those with and without metabolic syndrome, defined by the International Diabetes Federation, (P < 0.001). However, using receiver operating characteristic (ROC) analysis and assessment of area under curve (AUC), the leptin/adiponectin ratio did not perform significantly better than its components.

Conclusion

In patients with severe CHD, the leptin/adiponectin ratio was not found to be a robust tool to distinguish patients with and without insulin resistance and those with and without the metabolic syndrome.

Introduction

Leptin and adiponectin are adipose tissue-derived hormones involved in energy homeostasis. In 1994, Zhang et al. ¹ identified the first of the adipose tissue-derived hormones, leptin, a product of the ob gene. Leptin is considered a satiety hormone. Lack of the functional gene in knockout mice leads to hyperphagia and severe obesity.¹ Congenital lack of the functional gene in humans results in severe childhood obesity, delayed maturation of the reproductive system and immune dysfunction.^2,3 Adiponectin was first identified in 1995.⁴ It is a protein coded for by a gene on chromosome 3q27. Evidence suggests adiponectin enhances the action of insulin,⁵ stimulates insulin secretion in vitro and in vivo ⁶ and possibly has a central effect at the arcuate hypothalamus stimulating food intake and decreasing energy expenditure during fasting.⁷ Both have been found to be cross-sectionally and prospectively associated with the metabolic syndrome^8,9 and coronary heart disease (CHD), with leptin increasing and adiponectin decreasing. It has been proposed that leptin stimulates the development of atherosclerosis and adiponectin inhibits the process. However, this relationship is more controversial with some reports suggesting there has been overestimation of the association of the adipokines with CHD.^10,11

Assuming that the proposed associations are true, we sought to determine whether combining these two variables as a ratio, leptin/adiponectin, resulted in an accurate predictor of the metabolic status of a patient in a cohort of patients with severe CHD. Strong correlations between the adipokines could improve understanding of the pathogenesis of the metabolic syndrome and atherosclerosis and offer a novel potential treatment pathway.

Method

A total of 105 unselected patients referred for first-time elective coronary artery bypass surgery (CABG) at King's College Hospital (London, UK) between January 2003 and December 2003 participated in the study. The study population was 87% men (n = 91), 90.5% Caucasian (n = 95), 8.5% South Asian (n = 9) and 0.95% Arab (n = 1) with a mean age of 66.3 (range 42–81) years. Patients were excluded if they had symptomatic heart failure or an acute coronary syndrome in the previous six months. Patients were recruited from a preadmission clinic one month prior to surgery when smoking status (current or within the last six months) was ascertained and anthropometric measurements taken. The characteristics of the patients are described in Table 1. All patients were taking HMG-CoA reductase inhibitors and antiplatelet agents, 42% were taking angiotensin-converting enzyme inhibitors, 76% β-adrenoceptor blockers, 44% calcium channel antagonists and 22% a diuretic. Drugs were omitted for 24 h prior to surgery. On the morning of surgery, blood was obtained for measurement of lipids, fasting blood glucose, insulin, leptin and adiponectin concentrations. The study protocol and collection of samples were approved by the local research ethics committee and all subjects provided written informed consent.

Table 1

The characteristics of the patients studied

Characteristic	Mean + SEM	Range
Age (y)	66.3 + 0.8	42–81
BMI (kg/m²)	27.6 + 0.4	19–39.7
Waist (cm)	99.9 + 1.1	76–127
Waist:hip	1.00 + 0.007	0.8–1.18
SBP (mmHg)	141 + 2.3	80–200
DBP (mmHg)	74 + 1.3	46–120
Cholesterol (mmol/L)	3.2 + 0.1	0–5.9
LDL (mmol/L)	1.68 + 0.07	0–3.7
HDL (mmol/L)	0.94 + 0.03	0–1.7
Triglycerides (mmol/L)	1.27 + 0.08	0–4.7
Fasting glucose (mmol/L)	6.5 + 0.26	0.9–16.6
Insulin (mIU/L)	13.0 + 2.5	0.3–144.5
HOMA	4.96 + 1.2	0.1–89.1

BMI, body mass index; SBP, systolic blood pressure; DBP, diastolic blood pressure; LDL, low-density lipoprotein; HDL, high-density lipoprotein, HOMA, homeostatic model assessment

Metabolic syndrome

Ninety-nine patients were assessed for metabolic syndrome using the International Diabetes Federation (IDF) criteria. Of those, 66% (n = 65) were assessed as having the syndrome.

Plasma assays

Assays for serum total cholesterol, high-density lipoprotein (HDL) cholesterol, low-density lipoprotein (LDL) cholesterol, triglycerides and glucose were carried out in the hospital biochemistry laboratory (Advia 2400, Siemens Healthcare, Camberley, UK). Leptin and adiponectin were measured using a commercially available enzyme-linked immunosorbent assays (R&D Systems, Minneapolis, MN, USA). The intra-assay coefficient of variation (CV) for leptin was 5.9–6.9% (3.15–24.6 μg/L) and for adiponectin 2.5–4.7% (1980–14,300 μg/L). The inter-assay CV for leptin was 11.6–8.7% (2.7–26.1 μg/L) and for adiponectin 5.8–6.9% (2050–15,700 μg/L). Homeostatic model assessment (HOMA) greater than 3.0 was used as a marker of insulin resistance; the validity of this cut-off has been previously determined.¹²

Statistical analysis

Continuous variables are presented as mean ± standard error of the mean (SEM). The data were tested for a normal distribution using the Shapiro–Wilk test. Spearman's rank correlation was used to examine the relationship between variables. The Mann–Whitney U test was used to assess associations between non-parametric data-sets. The adiponectin and leptin concentrations were log transformed and stepwise multiple regression analysis performed to identify independent predictors of adiponectin and leptin concentrations. A value of P < 0.05 was considered statistically significant. Analysis was performed using Analyse-it General 1.65 and SPSS version 14.0.

Results

Leptin and adiponectin

To compare the strength of associations of the leptin/adiponectin ratio, the differences were first determined using the components of the ratio.

In this cohort, significantly lower concentrations of adiponectin were detected in patients with metabolic syndrome (mean adiponectin concentration 5061 ± 527 μg/L) compared with those without (mean adiponectin concentration 7111 ± 911 μg/L) where P = 0.04. However, no significant difference was observed in adiponectin concentrations between those with or without insulin resistance as defined as a HOMA > 3, P = 0.33. Similar findings were seen when assessing the associations of leptin alone. Significantly higher concentrations of leptin were found in patients with metabolic syndrome (mean ± SD leptin concentration 19.35 ± 2.04 μg/L) compared with those without (mean leptin concentration 11.15 ± 1.44) where P = 0.004. However, no significant difference was observed in leptin concentrations between those with or without insulin resistance as defined as a HOMA > 3, P = 0.0724.

Leptin/adiponectin ratio and its correlates

We went on to examine the associations between the components of the metabolic syndrome and the leptin/adiponectin ratio in this large unselected group (Table 2).

Table 2

Associations of the leptin/adiponectin ratio

	BMI	Waist	WHR	SBP	DBP	HDL	LDL	TGs	Cholesterol	Insulin	Glucose	HOMA
r _s	0.53	0.56	0.5	0.20	0.24	−0.15	−0.21	0.17	−0.19	0.37	0.24	0.34
P	<0.0001	<0.0001	<0.0001	0.05	0.02	0.15	0.036	0.09	0.07	0.0001	0.02	0.0006

BMI, body mass index; waist, waist circumference; WHR, waist/hip ratio; SBP, systolic blood pressure; DBP, diastolic blood pressure; HDL, high-density lipoprotein; LDL, low-density lipoprotein; TGs, triglycerides; HOMA, homeostatic model assessment

Association was found between the leptin/adiponectin ratio and a number of the components. Positive correlations were identified with: HOMA, used as a surrogate measure of insulin resistance (r _s = 0.34, P = 0.0006); fasting insulin concentrations (r _s = 0.37, P = 0.0001) and fasting glucose concentrations (r _s = 0.24, P = 0.01); systolic blood pressure (r _s = 0.20, P = 0.05) and diastolic blood pressure (r _s = 0.24, P = 0.02); waist circumference (r _s = 0.56, P < 0.0001) and waist/hip ratio (r _s = 0.5, P = 0.0001) and body mass index (BMI) (r _s = 0.53, P < 0.0001). A negative correlation was identified when examining the association between the leptin/adiponectin ratio and LDL concentrations (r _s = −0.21, P = 0.036).

Variation in HDL (r _s = −0.15, P = 0.15), triglycerides (r _s = 0.17, P = 0.09) and cholesterol (r _s = −0.19, P = 0.07) were not found to be associated with the leptin/adiponectin ratio.

Leptin/adiponectin ratio and the metabolic syndrome

As described earlier, metabolic syndrome has been found to be cross-sectionally and prospectively associated with changes in circulating leptin and adiponectin concentrations.^8,9 This study aimed at determining whether alterations in the leptin/adiponectin ratio could be used to distinguish patients with insulin resistance from those without. Insulin resistance was defined as a HOMA index >3, which equates to the upper quartile of HOMA results in this study. Significantly higher ratios of leptin/adiponectin were found in those with a HOMA <3 compared with those with a HOMA >3 (P = 0.029).

We also assessed the ability of the ratio to distinguish patients with the metabolic syndrome from those without the syndrome. We detected significantly lower ratios of leptin/adiponectin in those who had metabolic syndrome, as defined by the IDF, and those who did not (P < 0.001).

Stepwise multiple regression analysis of logarithmically transformed leptin/adiponectin ratios was performed to identify components that were independently associated with the leptin/adiponectin ratio. Waist circumference (r = 0.51, P < 0.001) and fasting glucose concentrations (r = 0.56, P = 0.01) were found to be independent predictors of the ratio.

Receiver operating characteristic curves

The test performance was assessed using receiver operating characteristic (ROC) curves. The sensitivity and specificity of the leptin/adiponectin ratio at determining the presence of metabolic syndrome or HOMA >3 were assessed and compared with the individual components of the ratio (Table 3).

Table 3

ROC analysis of leptin, adiponectin and the leptin/adiponectin ratio for the identification of subjects with metabolic syndrome or a HOMA > 3.0

	AUC (95% confidence interval)	P
L/A by metabolic syndrome	0.73 (0.63–0.84)	0.0001
Leptin by metabolic syndrome	0.68 (0.57–0.79)	0.0009
Adiponectin by metabolic syndrome	0.63 (0.51–0.75)	0.0167
L/A by HOMA	0.62 (0.49–0.75)	0.0341
Leptin by HOMA	0.60 (0.46–0.74)	0.0755
Adiponectin by HOMA	0.56 (0.43–0.69)	0.1986

ROC, receiver operating characteristic; AUC, area under curve; L/A, leptin/adiponectin ratio; HOMA, homeostatic model assessment

When the leptin/adiponectin ratio was identifying patients with metabolic syndrome, with a sensitivity of 80%, the specificity was 59% compared with specificities of 38% with leptin and 35% with adiponectin alone. The area under curve (AUC) for the leptin/adiponectin ratio and metabolic syndrome was found to be 0.73 (0.63–0.84), only marginally better than leptin, 0.68 (0.57–0.74), or adiponectin, 0.63 (0.51–0.75), alone. There was no significant difference between these ROCs.

When the leptin/adiponectin ratio was identifying patients with a HOMA > 3, with a sensitivity of 75%, the specificity was 38% compared with specificities of 36% with leptin and 32% with adiponectin alone. The AUC for the leptin/adiponectin ratio and insulin resistance was found to be 0.62 (0.62–0.75), only marginally better than leptin, 0.60 (0.46–0.74), or adiponectin, 0.56 (0.43–0.69), alone. There was no significant difference between these ROCs.

Discussion

This report identifies the leptin/adiponectin ratio to be a more useful marker of metabolic syndrome and insulin resistance (using HOMA as a surrogate) than each of its individual components alone. However, the advantage was small, particularly when assessing the data on insulin resistance.

Leptin/adiponectin ratio as a metabolic marker

Studies of the association of the leptin/adiponectin ratio are limited. The leptin/adiponectin ratio has been previously proposed as an indicator of atherosclerosis, insulin resistance and the metabolic syndrome.^13–17

In 2004, Satoh et al. ¹³ were the first group to demonstrate the greater sensitivity of the leptin/adiponectin ratio than its components in predicting atherosclerosis in obese type 2 diabetic patients and proposed its use as a potential atherogenic index in this population. Support was lent to this finding when the leptin/adiponectin ratio was found to have close correlation with carotid intima-media thickness, a surrogate marker of atherosclerosis, again in a population with type 2 diabetes.¹⁴ This finding was promptly replicated in healthy subjects.¹⁵ The same group also demonstrated the leptin/adiponectin ratio to have stronger correlations with several anthropometric, metabolic and clinical parameters than the individual adipokines. Correlations in this healthy population were detected between the leptin/adiponectin ratio and BMI, waist, hip, waist-to-hip ratio, systolic blood pressure, intima-media thickness, HDL, apolipoprotein A-I, glucose and HOMA.¹⁵ Our study supports these findings, showing the ratio to be associated with many of the components of the metabolic syndrome. However, we did not find any association between the leptin/adiponectin ratio and the patients' lipid profiles, apart from LDL.

Recently, the leptin/adiponectin ratio was shown to be better able to classify subjects with or without metabolic syndrome and have a stronger association with the components of the syndrome than adiponectin or leptin alone in older Chinese adults, using the IDF definition of the syndrome,¹⁶ a finding supported by our study, but in a predominantly Caucasian population. Most recently, the leptin/adiponectin ratio has been identified as a useful measure of insulin sensitivity in non-diabetic individuals; it was demonstrated to be at least as strongly associated with the gold standard measure of insulin resistance (clamp M/I value) as other frequently used methods such as fasting insulin or HOMA.¹⁷ Our study suggests that the leptin/adiponectin ratio may be a less useful predictor of insulin resistance than suggested by Finucaine et al. ¹⁷ in our mixed population of treated diabetic and non-diabetic patients with a significant co-morbidity. Unlike previous studies, notably the MONICA/KORA study, we did not find either adipokine to be significantly independently associated with insulin resistance.¹⁸

The subjects in our study were patients with severe CHD; we supposed that all would have deranged ratios^13–15 and further derangements caused by the metabolic syndrome would enable us to distinguish between the groups. However, it has recently been demonstrated that this logic may be flawed. In 2010, the MONICA/KORA study of 333 case subjects with incident CHD and 1728 non-case subjects concluded that the ratio did not improve prediction of CHD compared with either adipokine alone.¹⁹ They examined carotid intima thickening as a surrogate marker of atherosclerosis and found that after controlling for metabolic syndrome, carotid intima thickness is associated with plasma adiponectin levels but not with the leptin/adiponectin ratio.²⁰

Limitations

This study is cross sectional and therefore does not indicate a cause and effect relationship between the leptin/adiponectin ratio and insulin resistance or the metabolic syndrome.

The study examined an older, predominantly male, predominantly Caucasian population. Although this is representative of the population within the UK with CHD, it may make the findings of this study less applicable to other groups.

HOMA was used as a surrogate marker of insulin sensitivity as only blood samples were available for analysis.²¹ The gold standard remains the hyperinsulinaemic euglycaemic clamp.²²

Conclusions

We demonstrated the leptin/adiponectin ratio to have good correlation with HOMA and the components of the metabolic syndrome, a constellation of conditions associated with insulin resistance. However, on examining the sensitivity and specificity of the marker we found it to be a less reliable tool for distinguishing those with and without insulin resistance and the metabolic syndrome. Our study places doubt on the utility of the ratio as a potential surrogate marker. Its strong ability to distinguish those with and without insulin resistance in the healthy Caucasian population¹⁷ is not supported by our data from a mixed unhealthy population, indicating a possible instability in this potential marker. Further studies to evaluate the correlations between leptin/adiponectin ratios and the quantitative insulin sensitivity check index (QUICKI), and most importantly, the hyperinsulinaemic euglycaemic clamp in populations with co-morbidities would help determine this for certain.

DECLARATIONS

Competing interests: None.

Funding: Local R&D Funds (08/H0808/173).

Ethical approval: Code KCH1036: 08/H0808/173.

Guarantor: CWlR.

Contributorship: JIH and NV carried out the research as a project contributing to a BSc qualification. RL and SS assisted with lab techniques. AM collected the patient data (anthropometric measurements and blood samples). RAS assisted with statistical analysis. CWlR and JA-Z oversaw the project.

Acknowledgements: None.

References

Zhang

, Proenca

, Maffei

, Barone

, Leopold

, Friedman

. Positional cloning of the mouse obese gene and its human homologue. Nature 1994;372:425–32

Farooqi

, Jebb

, Langmack

, Effects of recombinant leptin therapy in a child with congenital leptin deficiency. N Engl J Med 1999;341:879–84

Farooqi

, Matarese

, Lord

, Beneficial effects of leptin on obesity, T cell hyporesponsiveness, and neuroendocrine/metabolic dysfunction of human congenital leptin deficiency. J Clin Invest 2002;110:1093–103

Scherer

, Williams

, Fogliano

, Baldini

, Lodish

. A novel serum protein similar to C1q, produced exclusively in adipocytes. J Biol Chem 1995;270:26746–9

Wang

, Mao

, Wang

, Adiponectin sensitizes insulin signalling by reducing p70 S6 kinase-mediated serine phosphorylation of IRS-1. J Biol Chem 2007;282:7991–6

Okamoto

, Ohara-Imaizumi

, Kubota

, Adiponectin induces insulin secretion in vivo and in vitro at low glucose concentration. Diabetologia 2008;51:827–35

Kubota

, Yano

, Kubota

, Adiponectin stimulates AMP-activated protein kinase in the hypothalamus and increases food intake. Cell Metab 2007;6:55–68

Franks

, Brage

, Luan

, Leptin predicts a worsening of the features of the metabolic syndrome independently of obesity. Obes Res 2005;13:1476–84

Salas-Salvado

, Granada

, Bulló

, Corominas

, Casas

, Foz

. Plasma adiponectin distribution in a Mediterranean population and its association with cardiovascular risk factors and metabolic syndrome. Metabolism 2007;56:1486–92

10.

Sattar

, Wannamethee

, Sarwar

, Adiponectin and coronary heart disease: a prospective study and meta-analysis. Circulation 2006;114:623–9

11.

Sattar

, Wannamethee

, Sarwar

, Leptin and coronary heart disease: prospective study and systematic review. J Am Coll Cardiol 2009;53:167–75

12.

Kuwana

, Urayama

, Kawai

. Reference value and cut-off value for diagnosis of insulin resistance in type 2 diabetes mellitus. Rinsho Byori 2002;50:398–403

13.

Satoh

, Naruse

, Usui

, Leptin-to-adiponectin ratio as a potential atherogenic index in obese type 2 diabetic patients. Diabetes Care 2004;27:2488–90

14.

Kotani

, Sakane

, Saiga

, Kurozawa

. Leptin:adiponectin ratio as an atherosclerotic index in patients with type 2 diabetes: relationship of the index to carotid intima-media thickness. Diabetologia 2005;48:2684–6

15.

Norata

, Raselli

, Grigore

, Leptin:adiponectin ratio is an independent predictor of intima media thickness of the common carotid artery. Stroke 2007;38:2844

16.

Zhuo

, Wang

, Fu

, Comparison of adiponectin, leptin and leptin to adiponectin ratio as diagnostic marker for metabolic syndrome in older adults of major cities. Diabetes Res Clin Pract 2009;84:27–33

17.

Finucane

, Luan

, Wareham

, (on behalf of the European Group for the Study of Insulin Resistance: relationship between insulin sensitivity and cardiovascular disease risk study group). Correlation of the leptin: adiponectin ratio with measures of insulin resistance in non-diabetic individuals. Diabetologia 2009;52:2345–9

18.

Thorand

, Zierer

, Baumert

, Associations between leptin and the leptin/adiponectin ratio and incident Type 2 diabetes in middle-aged men and women: results from the MONICA//KORA Augsburg study 1984–2002. Diabet Med 2010;27:1004–11

19.

Karakas

, Zierer

, Herder

, Leptin, adiponectin, their ratio and risk of coronary heart disease: results from the MONICA/KORA Augsburg Study 1984–2002. Atherosclerosis 2010;209:220–5

20.

Dullart

, Kappelle

, Dallinga-Thie

. Carotid intima media thickness is associated with plasma adiponectin but not with the leptin:adiponectin ratio independently of metabolic syndrome. Atherosclerosis 2010;211:393–6

21.

Bonora

, Targher

, Alberiche

, Homeostasis model assessment closely mirrors the glucose clamp technique in the assessment of insulin sensitivity: studies in subjects with various degrees of glucose tolerance and insulin sensitivity. Diabetes Care 2000;23:57–63

22.

American Diabetes Association: Consensus. Development conference on insulin resistance. Diabetes Care 1998;21:310–4