International Journal of Cardiology xxx (2012) xxx–xxx

IJCA-14968; No of Pages 6

Contents lists available at SciVerse ScienceDirect

International Journal of Cardiology

j ourna l homepage: www.e lsev ie r .com/ locate / i j ca rd

Association of epicardial fat thickness with the severity of obstructive sleep apnea inobese patients☆

Stefania Mariani a, Daniela Fiore a, Giuseppe Barbaro a, Sabrina Basciani b, Maurizio Saponara c,Enzo D'Arcangelo d, Salvatore Ulisse a, Costanzo Moretti e, Andrea Fabbri e, Lucio Gnessi a,⁎a Department of Experimental Medicine, Section of Medical Physiopathology and Endocrinology, Sapienza University of Rome, Italyb Cardiovascular Research Unit, Istituto di Ricovero e Cura a Carattere Scientifico San Raffaele, Roma, Tosinvest Sanità, Rome, Italyc Department of Otolaryngology, Audiology and Phonation, Sapienza University of Rome, Italyd Department of Statistics, Probability and Applied Statistics, Sapienza University of Rome, Italye Department of Internal Medicine, University of Rome Tor Vergata, Rome, Italy

☆ Funding source: This study was funded by SapienProgetti di Ricerca di Ateneo.⁎ Corresponding author at: Department of Experiment

Physiopathology and Endocrinology, Sapienza UniversitTel.: +39 6 49970509; fax: +39 6 4461450.

E-mail address: lucio.gnessi@uniroma1.it (L. Gnessi)

0167-5273/$ – see front matter © 2012 Elsevier Irelanddoi:10.1016/j.ijcard.2012.06.011

Please cite this article as: Mariani S, et. al, AsInt J Cardiol (2012), doi:10.1016/j.ijcard.20

a b s t r a c t

a r t i c l e i n f o

Article history:

Received 16 September 2011Received in revised form 6 June 2012Accepted 7 June 2012Available online xxxx

Keywords:Obstructive sleep apneaObesityEpicardial fatCardiovascular risk factorBody composition

Background: The correlation between obesity and severity of obstructive sleep apnea (OSA) is controversial.Although fat excess is a predisposing factor for the development of OSA, it has not been determined whetherfat distribution rather than obesity per se is associated with OSA severity. Epicardial fat thickness (EFT) is anindependent index of visceral adiposity and cardiometabolic risk. We investigated the relation between fatdistribution and cardiometabolic risk factors, including EFT and common carotid intima-media thickness(cIMT), with the severity of OSA in obese patients.Methods: One hundred and fifteen obese patients (56 males, 59 females) with polysomnographic evidence ofOSA (≥5 apnea/hypopnea events per hour) of various degrees, without significant differences in grade ofobesity as defined by body mass index (BMI), were evaluated. The following parameters were measured:BMI, body composition by dual energy X-ray absorptiometry, EFT, right ventricular end-diastolic diameter(RVEDD) and cIMT by ultrasound, and parameters of metabolic syndrome (waist circumference, arterial blood

pressure, fasting glucose, HDL-cholesterol and triglycerides).Results: EFT, RVEDD, cIMT and trunk/leg fat mass ratio showed a positive correlation with OSA severity in uni-variate analysis (r=0.536, pb0.001; r=0.480, pb0.001; r=0.345, pb0.001; r=0.330, pb0.001, respectively).However, multiple linear regression analysis showed that EFT was themost significant independent correlate ofthe severity of OSA (R2=0.376, p=0.022).Conclusions: The present study suggests that, in obese patients, EFT may be included among the clinical param-eters associating with OSA severity. The association of EFT with OSA, both cardiovascular risk factors, is inde-pendent of obesity as defined by classical measures.

© 2012 Elsevier Ireland Ltd. All rights reserved.

1. Introduction

Epicardial fat is the visceral fat depot of the heart [1] and reflectsvisceral adiposity rather than general obesity. It is considered a met-abolically active tissue, being a local source of pro-inflammatoryfactors associated with the development of metabolic syndrome andcardiovascular diseases. Obstructive sleep apnea (OSA) is a sleep dis-order characterized by repeated partial or complete collapse of theupper airways, which leads to oxygen desaturation, fragmentation

za University of Rome, Italy,

al Medicine, Section ofMedicaly of Rome, 00161‐Rome, Italy.

.

Ltd. All rights reserved.

sociation of epicardial fat thi12.06.011

of sleep, and daytime sleepiness, with associated features includingcardiovascular and metabolic consequences [2–4]. Obesity, a majorcause of morbidity and mortality through the development of its co-morbidities, is a predisposing factor in the development of OSA [5]. Itis well established that the prevalence of OSA is higher among obesepatients than in general population, and it is under diagnosed in over-weight individuals [5]. OSA worsens with weight gain and improveswithweight reduction [6] andmay itself predispose toworsening obe-sity because of sleep deprivation and disrupted metabolism [7,8].

There are complex associations between OSA, clinical and subclin-ical cardiovascular and metabolic disorders and obesity, and earlycardiovascular abnormalities in newly diagnosed OSA patients areobserved [9]. Studies on the relationship between body mass index(BMI) and the severity of OSA are contradictory [10–13], and the ques-tion remains whether in obese individuals regional fat distributionmore than BMI correlates with the severity of OSA.

ckness with the severity of obstructive sleep apnea in obese patients,

2 S. Mariani et al. / International Journal of Cardiology xxx (2012) xxx–xxx

The aim of this study was to investigate, in an homogeneous pop-ulation of obese patients, the association of OSA severity withmarkersof cardiometabolic disease risk and with estimated regional fat mass(FM), measured through traditional anthropometric measures, bodycomposition and epicardial fat thickness (EFT).

2. Methods and procedures

2.1. Study population

We studied 115 obese patients with diagnosis of OSA (56 men and 59 women)recruited from the Department of Otolaryngology, Audiology and Phonation, SapienzaUniversity of Rome from June 2010 to June 2011. Exclusion criteria comprised hypo-thyroidism, acromegaly, acute illness, current use of hypnotics, heart diseases, lungdiseases, any other respiratory disorder, uncontrolled hypertension, craniofacial abnor-malities, history of smoking, or use of continuous positive airway pressure (CPAP). Pa-tients with a previous diagnosis of diabetes mellitus, dyslipidemia, or hypertension,were receiving drugs for each of these conditions. Metformin was used as glucose-lowering drug, fenofibrate as lipid-lowering treatment, and diuretics or sartans asblood pressure (BP)-lowering agents. Patients receiving other medications were ex-cluded. All subjects were enrolled after written consent and approval by the InstitutionEthic Committee.

2.2. Clinical measurements

Weight and height were measured to the nearest 0.1 kg and 0.1 cm, respectively.BMI was expressed as weight (kilograms)/height (m2). Obesity was defined for anyBMI ≥30 kg/m2. The waist circumference (WC) was measured just above the bonylandmark of the iliac crest and expressed in centimeters. Sitting BP was measuredtwice at 5-min intervals, and the average of two measurements was used for analysis.

2.3. Dual energy X-ray absorptiometry (DXA)

DXA analysis was performed by one single experienced technician using a DXA scan(Hologic Inc., Bedford, MA, USA, QDR 4500W). Coefficient of variation for fat mass wasb1.5%. Body composition was measured in the whole body and in specific body regions.Delimiters for regional analysis were determined by a standard software (Hologic Inc.,S/N 47168 VER. 11.2). With the use of specific anatomic landmarks, regions of the neck,trunk, arms and legs were distinguished. Scans were performed according to the man-ufacturer's instructions.

2.4. Polysomnography

In-hospital polysomnography (SomnoscreenTM PSG, Somnomedix GmbH Non-nengarten, 8 D-97270 Kist, Germany) was performed overnight between 10:00 pmand 6:00 am in the same laboratory. Recordings included electroencephalogram, elec-trooculogram, electromyogram, electrocardiogram, thermistors for nasal and oral air-flow, thoracic and abdominal impedance belts for respiratory effort, pulse oximetryfor oxyhemoglobin level, tracheal microphone for snoring and sensor to assess chang-ing of the position during sleep. Polysomnography records were scored according tostandard criteria [14] by a single experienced operator. OSA degree was determinedby the apnea/hypopnea index (AHI) defined as the total number of obstructive apneas(cessation of airflow for at least 10 s) and hypopneas (decrease of the airflow signalamplitude by at least 50% associated by oxyhemoglobin desaturation of at least 4% orby an arousal) per hour of sleep. Mild, moderate and severe OSA were defined by anAHI of 5–14, 15–29, and ≥30 events/h, respectively.

2.5. EFT, right ventricular end-diastolic diameter (RVEDD) and carotid intima-mediathickness (cIMT) measurements

EFT was measured through a validated echocardiographic procedure [15]. Partici-pants underwent high-resolution M-B-mode transthoracic echocardiography using a2.5-MHz probe, and spectral Doppler exam of the common carotid artery using a 7.5-MHz probe (Esaote MyLab40, Esaote Europe B.V., The Netherlands). The EFT was iden-tified as the echo-free space between the outer wall of themyocardium and the viscerallayer of the pericardium, and its thickness was measured perpendicularly on the freewall of the right ventricle (RV) at end-systole in three cardiac cycles. The averagevalue of three cardiac cycles from each echocardiographic view was considered. TheRVEDD was measured according to the recommendations of the American Society ofEchocardiography. The cIMT was measured in the anterior wall of the common carotidartery as the distance from the trailing edge of the adventitia to the leading edge of theintima-media; in the posterior wall of the vessel, it was measured as the distance fromthe leading edge of the intima-media to the trailing edge of the adventitia. The cIMTvalues for any given subject were the mean value for the two common carotid arteries.All echocardiograms and carotid ultrasonography were recorded by the same experi-enced operator who was blinded to the other study data.

Please cite this article as: Mariani S, et. al, Association of epicardial fat thiInt J Cardiol (2012), doi:10.1016/j.ijcard.2012.06.011

2.6. Laboratory

Blood samples were collected in the morning after an overnight fast. HDL cholesterol(HDL-C) and triglyceride concentrations weremeasured on fresh serum by an automatedenzymatic method (Dimension Analyzer, Dade Behring S.p.A. Milano, Italy). All sampleswere assayed in duplicate with intraassay and interassay mean coefficients of variationof 2% and 4%, respectively. Fasting plasma glucose (FPG) determinations were performedusing the hexokinase method (Aeroset, Abbott Park, IL, USA).

2.7. Statistical analysis

Data were analyzed with the use of STATISTICA software, version 6.1 (Stat Soft,Inc., Tulsa, Oklahoma). Results are expressed as mean±SD. Differences betweengroups were analyzed using ANOVA for continuous variables. Pearson correlation testwas used to measure a linear association between variables. The roles of sex, age,BMI, body fat distribution, cIMT, EFT, RVEDD, glucose and serum lipids as associatedvariables with AHI were tested by linear regression with the use of univariate and mul-tivariate models. All p values presented are two-tailed, and values less than 0.05 areconsidered to indicate statistical significance.

3. Results

The demographic, anthropometric, clinical and sleep characteristicsof the patients, as well as the comparisons of patients stratifiedaccording to AHI categories are shown in Table 1. Mean patient agewas 53.64±9.49 yr (range 30–65 yr), 48.7% of the participants weremale and 51.3% were female. Their BMI was ≥30.0 kg/m2 (mean,36.30±4.57 kg/m2)with amean value of 37.14±4.76 kg/m2 in femalesand 35.41±4.22 kg/m2 in males. TheWC was constantly >90 cm in fe-males (mean, 109.49±12.97 cm) and >100 cm in males (mean,111.78±12.96 cm). Fifty patients (43.48%) presented with mild OSA.Moderate OSA was observed in 33 patients (28.70%), and severe OSAwas found in 32 patients (27.82%). Although the women had a lowerAHI than men (18.80±17.55 vs. 29.70±20.67, respectively; p=0.002)the differences between categories of OSA severity were not statisticallysignificant for confounding caused by gender (p=0.796). There wereno significant differences in age, height, weight, FPG, HDL-C, triglycerides,BMI, and WC between patients with mild, moderate or severe OSA. Thesystolic and diastolic BP mean values were significantly higher in thegroup with severe OSA compared with the mild OSA group (p=0.042and p=0.036, respectively).

Body composition, EFT, RVEDD and cIMT of all the patients andacross AHI categories are shown in Table 1. The ranges of EFT,RVEDD and cIMT were 6–11.20 mm, 19–33 mm, and 0.6–1.2 mm, re-spectively. Trunk FM, arms FM and neck FM mean values were notsignificantly different among the three groups. There was a significantincrease in EFT, RVEDD and cIMT with increasing severity of OSA. Onthe contrary, AHI higher values were most frequently detected in pa-tients with the lowest leg FM.

We then evaluated the association between AHI and each single co-variate by linear regression analysis. Therewas a positive association be-tween AHI and EFT (r=0.536; pb0.001), RVEDD (r=0.480; pb0.001),cIMT (r=0.345; pb0.001) and trunk/leg FM ratio (r=0.330; pb0.001)(Fig. 1). In sex-stratified analyses, the likelihood of detection of higherEFT values was higher in men (r=0.571; pb0.001) than in women(r=0.359; p=0.005), with a progressive increase of sex difference athigher AHI values. Similar results were observed for RVEDD (menr=0.550; pb0.001 vs. women r=0.325; p=0.012). The r values forthe cIMT-AHI correlation were almost superimposable for men (r=0.389; p=0.003) and women (r=0.326; p=0.012). The significanceof linear regression between AHI and trunk/leg FM ratio was conservedin men (r=0.298; p=0.026) and was lost in women (r=0.192;p=0.145). There was no correlation between AHI and all the othervariables evaluated except for WC (r=0.299; p=0.025) and neck FM(r=0.267; p=0.046) in men. A positive association, without genderdifferences, between EFT and RVEDD (r=0.759; pb0.001) was alsoobserved.

Sex and age adjusted multivariate linear regression analyses forthe associations between AHI and selected covariates are presented

ckness with the severity of obstructive sleep apnea in obese patients,

Table 1Clinical characteristics, body composition, EFT and cIMT of the patients with mild, moderate and severe OSA.

All Mild OSA(AHI 5–14/h)

Moderate OSA(AHI 15–29/h)

Severe OSA(AHI ≥30/h)

Number 115 50 33 32Gender (male/female) 56/59 16/34 19/14 21/11Age (years) 53.64±9.49 51.64±9.82 55.36±9.72 55.00±8.35AHI (events/h) 24.11±19.82 8.37±2.74 21.88±4.62a 51.00±16.12a,b

Height (cm) 167.50±9.15 166.22±9.60 167.90±9.04 169.09±8.50Weight (kg) 101.97±15.95 100.35±17.26 99.91±13.47 106.62±15.72BMI (kg/m2) 36.30±4.57 36.19±4.49 35.45±4.20 37.34±4.98WC (cm) 110.6±13.50 109.49±12.97 108.22±11.44 114.80±15.58FPG (mg/100 ml) 101.56±17.91 99.44±15.57 104.37±22.47 101.97±16.11Systolic BP (mm Hg) 135.59±12.59 133.00±13.19 136.33±12.12 138.87±11.57a

Diastolic BP (mm Hg) 86.13±8.30 84.48±8.94 86.27±7.38 88.56±7.75a

HDL-C (mg/100 ml) 46.69±12.49 49.13±11.55 45.66±12.78 43.95±13.26Triglycerides (mg/100 ml) 151.88±77.35 154.04±90.11 148.15±67.31 152.33±66.86Metabolic syndrome (%) 55.65 52.00 57.57 59.37Trunk FM (kg) 17.66±4.99 17.44±5.61 17.56±4.59 18.11±4.46Arms FM (kg) 4.85±1.92 4.95±2.32 4.74±1.32 4.83±1.80Leg FM (kg) 11.54±4.57 12.72±5.21 11.02±3.81 10.22±3.81a

Trunk/leg FM ratio 1.63±0.41 1.44±0.30 1.69±0.47a 1.87±0.37a

Neck FM (kg) 0.70±0.17 0.68±0.19 0.69±0.15 0.73±0.18EFT (mm) 7.86±0.98 7.19±0.59 8.14±0.76a 8.62±0.99a,b

RVEDD (mm) 26.87±2.79 25.38±2.44 27.27±2.45a 28.81±2.32a,b

cIMT (mm) 0.87±0.13 0.83±0.09 0.86±0.12 0.94±0.14a,b

Abbreviations: AHI, apnea–hypopnea index; BMI, body mass index; WC, waist circumference; FPG, fasting plasma glucose; BP, blood pressure; HDL-C, HDL cholesterol; FM, fatmass; EFT, epicardial fat thickness; RVEDD, right ventricular end-diastolic diameter; cIMT, carotid intima-media thickness. Values are expressed as means±SD.

a Significantly different from mild OSA group (pb0.05).b Significantly different from moderate OSA group (pb0.05).

3S. Mariani et al. / International Journal of Cardiology xxx (2012) xxx–xxx

in Table 2. In a model for indices of fat content (model 1), there was asignificant association of AHI with EFT (R2=0.306; pb0.0001). In amodel for cardiometabolic risk factors (model 2) there was a signifi-cant association of AHI with cIMT and RVEDD (R2=0.324;p=0.042 and pb0.0001, respectively).

5

6

7

8

9

10

11

12

0 10 20 30 40 50 60 70 80 90 100

EF

T (

mm

)

0 10 20 30 40 50 60 70 80 90 100

Tru

nk /L

egs

FM

Rat

io r = 0.330

P < 0.001

r = 0.298 P = 0.026

r = 0.192 P = 0.145

r = 0.571 P < 0.001

r = 0.536 P < 0.001

r = 0.359 P = 0.005

AHI (events/hour)

0.5

1.0

1.5

2.5

2.0

3.0

3.5

A

C

Fig. 1. Correlation between AHI and EFT (A), cIMT (B), trunk/leg FM ratio (C) and RVEDD (Dthe line of best fit for all data points. The dotted lines represent the line of best fit in men andlevels of significance (p) for each line fit are provided.

Please cite this article as: Mariani S, et. al, Association of epicardial fat thiInt J Cardiol (2012), doi:10.1016/j.ijcard.2012.06.011

After adjustment for both indices of fat depot and cardiometabolicrisk factors (model 3), cIMT shifted to a p value of 0.069, RVEDDshifted to a p value of 0.199, while EFT remained a significant inde-pendent correlate with the severity of OSA, with a predicted 37% ofthe variance in AHI (R2=0.376, p=0.022).

18

21

23

26

28

31

33

0 10 20 30 40 50 60 70 80 90 100 AHI (events/hour)

0 10 20 30 40 50 60 70 80 90 100

cIM

T (

mm

) R

VE

DD

(m

m)

r = 0.345 P < 0.001

r = 0.389 P = 0.003

r = 0.326 P = 0.012

r = 0.550 P < 0.001

r = 0.325 P = 0.012

r = 0.480 P < 0.001

1.3

1.1

0.9

0.7

0.5

B

D

) in obese patients (men, solid circles; women, open circles). The solid lines representthe broken lines represent the line of best fit in women. Correlation coefficients (r) and

ckness with the severity of obstructive sleep apnea in obese patients,

Table 2Multiple regression analysis, adjusted for age and sex, of associations of AHI with indices of fat content (model 1), cardiometabolic risk factors (model 2) and all covariates (model 3).

Model 1 Model 2 Model 3

β SE p β SE p β SE p

Systolic BP – – – −0.177 0.152 0.247 −0.212 0.149 0.159Diastolic BP – – – 0.124 0.141 0.382 0.145 0.139 0.298HDL-C – – – −0.078 0.089 0.380 −0.076 0.088 0.391Triglycerides – – – 0.020 0.086 0.811 0.030 0.085 0.720FPG – – – −0.076 0.094 0.416 −0.071 0.093 0.444cIMT – – – 0.191 0.093 0.042 0.172 0.094 0.069RVEDD – – – 0.384 0.092 b0.0001 0.159 0.123 0.199WC 0.049 0.125 0.691 0.077 0.091 0.400 0.066 0.123 0.590BMI −0.064 0.122 0.598 – – – −0.021 0.123 0.865EFT 0.457 0.102 b0.0001 – – – 0.320 0.138 0.022Trunk/leg FM 0.088 0.095 0.355 – – – 0.066 0.096 0.491Neck FM −0.054 0.087 0.532 – – – −0.102 0.085 0.233

Abbreviations: BMI, body mass index; WC, waist circumference; FPG, fasting plasma glucose; BP, blood pressure; HDL-C, HDL cholesterol; FM, fat mass; EFT, epicardial fat thickness;cIMT, carotid intima-media thickness; RVEDD, right ventricular end-diastolic diameter.

4 S. Mariani et al. / International Journal of Cardiology xxx (2012) xxx–xxx

4. Discussion

Obesity and OSA share many pathophysiological pathways andboth are independent risk factors for all cause cardiovascular events[16–20]. Although patients with OSA are often overweight or obese,BMI is not a good predictor of OSA severity [10,21–25]. Previous stud-ies showed contrasting results also concerning the relationship be-tween body fat distribution and OSA. Echocardiographic EFT is anindependent predictor of visceral adiposity that weakly reflects thedegree of obesity as measured by BMI, but strongly correlates withcardiometabolic risk factors independent of overall adiposity. Thefindings of this study suggest that in obese individuals, specific distri-bution patterns of adiposity associate with OSA severity and that,among them, EFT is the most obvious correlate of the severity ofsleep-disordered breathing in this cohort.

According to previous results [10], we did not find a relationshipbetween BMI and severity of OSA. The comparison between gender re-vealed that women had a lower AHI despite they had a higher BMI,suggesting that men require a lower general obesity index to developapnea and confirming that BMI is not a good predictor of the severityof OSA.

We did not find a significant association between WC, trunk FM orneck FM and OSA severity although, in line with other studies [13,26],when adjusting for sex, the relative contribution of WC and neck fatcontent to the severity of OSA could be more pronounced in men thanin women.

Instead, we observed a higher trunk/leg FM ratio in patients suf-fering from severe OSA compared with patients with mild OSA. Thiscould derive from the lower leg FM measured in the group with se-vere OSA in view of the steady distribution of the trunk FM in the sub-groups. When the analysis was restricted to women the significantassociation between trunk/leg FM ratio and AHI was lost. A possibleexplanation for this may reside, once more, in the fat distribution,since male patients had on average a lower leg FM compared to fe-males. Therefore, trunk fat and leg fat appear to have opposite associ-ations with the severity of OSA. The higher risk of complications dueto an unfavorable fat distribution is well established. For example, theprevalence of subcutaneous adipose tissue compared to visceral fatmay protect against diabetes and CV diseases [27]. On the contrary,low subcutaneous thigh fat is a risk factor for glucose and lipid disor-ders, independently of abdominal fat [28], considering that subcuta-neous fat has a higher potential of expandability than visceral fat,allowing a more effective storage of lipids [29–31].

This is the first study to report a positive association of the trunk/leg FM ratio with the severity of OSA.

A large body of evidence demonstrates that a number of OSA pa-tients display subclinical signs of atherosclerosis and increased cIMT

Please cite this article as: Mariani S, et. al, Association of epicardial fat thiInt J Cardiol (2012), doi:10.1016/j.ijcard.2012.06.011

[32], and that treatmentswith CPAP improve some of these subclinicalconditions [33]. cIMT is a recognized independent predictor of cardio-vascular events [34]. OSA, together with hypertension and metabolicsyndrome, has additive effects on carotid atherosclerosis [35,36].Therefore, obesity and metabolic derangements may represent con-founding factors when studying subclinical atherosclerosis in patientsaffected by OSA [32,37]. We found a significant correlation betweencIMT and OSA severity, and higher cIMT in the group of patientswith severe OSA compared with mild and moderate OSA. Further-more, the direct association between cIMT and AHI remained statisti-cally significant after adjustment for cardiometabolic risk factors.These results confirm previous data showing that subjects with moreserious grade of OSA have greater cIMT than those suffering for mildsleep-disordered breathing [38,39]. Although the correlation betweencIMT and AHI failed to reach statistical significance after correction forall the covariates, a relationship between AHI and cIMT is difficult torule out, given that the p value was very close to statistical signifi-cance. This aspect of the study could be clarified with a larger cohort.

The increased AHI is combined with diastolic dysfunction, espe-cially the right ventricular [40]. These data are consistent with ourstudy in which we found an association between the severity of OSAand RVEDD. The association remained significant after adjustmentfor cardiometabolic risk factors. The statistical significance was lostafter correction for all the covariates. The presence of indices of fatcontent including EFT in the multiple regression analysis, and thefact that obesity, a condition closely related to OSA, can alter RV struc-ture and function [41], might explain this finding.

We explored the visceral epicardial fat depot of the heart in OSAsubjects through the measurement of the EFT. The range of EFT wasconsistent with previously published data in obese subjects [42]. Al-though there is a relationship between EFT and general body adipos-ity, EFT reflects visceral adiposity rather than general obesity [1,43].We found a significant positive correlation between EFT and AHI,and EFT values were significantly higher in patients with moderateand severe OSA compared with the mild OSA group. The regressionplot graphic seems to indicate that the correlation between EFT andAHI is stronger in subjects with milder degree of OSA, however, thistendency needs to be confirmed increasing the number of patientsin the severe OSA group. Interestingly, the strong correlation betweenEFT and AHI kept significant also after adjustment for all other vari-ables, indicating that EFT correlates with OSA severity better thanthe classical indices of obesity in general and of visceral obesity inparticular. Thus, all our findings are consistently supportive of the su-perior value of EFT among the indices for estimating adiposity-relatedOSA burden in both men and women.

Although the present study was purely correlative, and no causa-tive conclusions can be drawn, the enlarged RV cavity size associated

ckness with the severity of obstructive sleep apnea in obese patients,

5S. Mariani et al. / International Journal of Cardiology xxx (2012) xxx–xxx

with both EFT [44] and OSA [40]might represent the link between EFTand AHI. Indeed, the epicardial fat is a metabolically active fat depotwhose anatomical proximity to themyocardium and coronary arteriesimplies close interactions [45]; therefore, its pathological increasemight affect the cardiovascular function, including RV. The exactmechanism for such an association is still unclear, however, potentialmechanisms have been proposed. The increased EFT constitutes a me-chanical load on the heart that could lead to compensatory remodel-ing [46]; the release of adipokines and cytokines, through paracrineand vasocrine interactions, may directly induce deleterious changesin myocardial tissue and coronary arteries [46–49]; epicardial fat mayexacerbate myocardial steatosis leading to adverse structural and func-tional cardiac adaptations [48]; finally, epicardial fat could also affectthe heart through its systemic effects being closely associated withcomponents of the metabolic syndrome like insulin resistance and glu-cose intolerance, conditions that predispose to myocardial derange-ments [50–53]. As a whole, these mechanisms underline the novelrelevance of epicardial fat in heart disease and metabolic syndrome.

Several limitations of the current study need to be considered. Thecohort includes patients with differences in metabolic features that,anyway, were equally distributed in the study population within thesubgroups. We knowingly did not exclude individuals with metabolicsyndrome because we intended to study a representative cohort ofobese OSA patients with all of their associated co-morbidities. Thiswas a cross-sectional study, so we could not assess the causal relation-ship between the covariates evaluated and OSA. To clarify this point, aprospective study involving clinical interventionwill be needed. Thereare clearly other contributing factors such as ethnicity andmenopausestatus in women that were not considered by the scope of this study.Finally, we are aware that our results should be confirmed in largercohorts.

In conclusion, although the question remains whether EFT and OSAare simply common co-travelers due to both being a result of obesity orwhether their co-occurrence reflects a causal effect of visceral fat masson OSA pathogenesis (or vice versa), our study demonstrates that EFTis independently associated with OSA severity in obese individuals.

Acknowledgment

The authors of the manuscript have certified that they complywith the Principles of Ethical Publishing in the International Journalof Cardiology [54].

References

[1] Iacobellis G, Willens HJ. Echocardiographic epicardial fat: a review of research andclinical applications. J Am Soc Echocardiogr 2009;22:1311–9.

[2] Butt M, Dwivedi G, Khair O, Lip GY. Obstructive sleep apnea and cardiovasculardisease. Int J Cardiol 2010;139:7–16.

[3] Zamarron C, García Paz V, Riveiro A. Obstructive sleep apnea syndrome is a sys-temic disease. Current evidence. Eur J Intern Med 2008;19:390–8.

[4] Romero-Corral A, Caples SM, Lopez-Jimenez F, Somers VK. Interactions betweenobesity and obstructive sleep apnea: implications for treatment. Chest 2010;137:711–9.

[5] Young T, Peppard PE, Gottlieb DJ. Epidemiology of obstructive sleep apnea: a pop-ulation health perspective. Am J Respir Crit Care Med 2002;165:1217–39.

[6] Newman AB, Foster G, Givelber R, Nieto FJ, Redline S, Young T. Progression and re-gression of sleep-disordered breathing with changes in weight: the Sleep HeartHealth Study. Arch Intern Med 2005;165:2408–13.

[7] Bonsignore MR, Zito A. Metabolic effects of the obstructive sleep apnea syndromeand cardiovascular risk. Arch Physiol Biochem 2008;114:255–60.

[8] Van Cauter E, Spiegel K, Tasali E, Leproult R. Metabolic consequences of sleep andsleep loss. Sleep Med 2008;9:23–8.

[9] Baguet JP, Nadra M, Barone-Rochette G, Ormezzano O, Pierre H, Pépin JL. Earlycardiovascular abnormalities in newly diagnosed obstructive sleep apnea. VascHealth Risk Manag 2009;5:1063–73.

[10] Martinez-Rivera C, Abad J, Fiz JA, Rios J, Morera J. Usefulness of truncal obesity in-dices as predictive factors for obstructive sleep apnea syndrome. Obesity 2008;16:113–8.

[11] Vgontzas AN. Does obesity play a major role in the pathogenesis of sleep apnoeaand its associated manifestations via inflammation, visceral adiposity, and insulinresistance? Arch Physiol Biochem 2008;114:211–23.

Please cite this article as: Mariani S, et. al, Association of epicardial fat thiInt J Cardiol (2012), doi:10.1016/j.ijcard.2012.06.011

[12] Pillar G, Shehadeh N. Abdominal fat and sleep apnea: the chicken or the egg? Di-abetes Care 2008;31:303–9.

[13] Bruno E, Alessandrini M, Napolitano B, De Padova A, Di Daniele N, De Lorenzo A.Dual-energy X-ray absorptiometry analysis of body composition in patients af-fected by OSAS. Eur Arch Otorhinolaryngol 2009;266:1285–90.

[14] The Report of an American Academy of Sleep Medicine task force. Sleep-relatedbreathing disorders in adults: recommendations for syndrome definition andmeasurement techniques in clinical research. Sleep 1999;22:667–89.

[15] Malavazos AE, Di Leo G, Secchi F, et al. Relation of echocardiographic epicardial fatthickness and myocardial fat. Am J Cardiol 2010;105:1831–5.

[16] Selim B, Won C, Yaggi HK. Cardiovascular consequences of sleep apnea. Clin ChestMed 2010;31:203–20.

[17] Bradley TD, Floras JS. Obstructive sleep apnea and its cardiovascular conse-quences. Lancet 2009;373:82–93.

[18] Marin JM, Carrizo SJ, Vicente E, Agusti AG. Long-term cardiovascular outcomes inmen with obstructive sleep apnoea–hypopnoea with or without treatment withcontinuous positive airway pressure: an observational study. Lancet 2005;365:1046–105.

[19] Yaggi HK, Concato J, Kernan WN, Lichtman JH, Brass LM, Mohsenin V. Obstructivesleep apnea as a risk factor for stroke and death. N Engl J Med 2005;353:2034–41.

[20] Gami AS, Howard DE, Olson EJ, Somers VK. Day–night pattern of sudden death inobstructive sleep apnea. N Engl J Med 2005;352:1206–14.

[21] Levinson PD, McGarvey ST, Carlisle CC, Eveloff SE, Herbert PN, Millman RP. Adipos-ity and cardiovascular risk factors in men with obstructive sleep apnea. Chest1993;103:1336–42.

[22] Newman AB, Nieto FJ, Guidry U, et al. Relation of sleep-disordered breathing tocardiovascular disease risk factors: the Sleep Heart Health Study. Am J Epidemiol2001;154:50–9.

[23] Schäfer H, Pauleit D, Sudhop T, Gouni-Berthold I, Ewig S, Berthold HK. Body fatdistribution, serum leptin, and cardiovascular risk factors in men with obstructivesleep apnea. Chest 2002;122:829–39.

[24] O'Keeffe T, Patterson EJ. Evidence supporting routine polysomnography beforebariatric surgery. Obes Surg 2004;14:23–6.

[25] Namysłowski G, Scierski W, Mrówka-Kata K, Kawecka I, Kawecki D, Czecior E. Sleepstudy in patients with overweight and obesity. J Physiol Pharmacol 2005;56:59–65.

[26] Mohsenin V. Gender differences in the expression of sleep-disordered breathing:role of upper airway dimensions. Chest 2001;120(5):1442–7.

[27] Aasen G, Fagertun H, Halse J. Regional fat mass by DXA: high leg fat mass attenu-ates the relative risk of insulin resistance and dyslipidaemia in obese but not inoverweight postmenopausal women. Scand J Clin Lab Invest 2008;68(3):204–11.

[28] Snijder MB, Visser M, Dekker JM, et al. Low subcutaneous thigh fat is a risk factorfor unfavourable glucose and lipid levels, independently of high abdominal fat.The Health ABC Study. Diabetologia 2005;48:301–8.

[29] Virtue S, Vidal-Puig A. Adipose tissue expandability, lipotoxicity and the metabolicsyndrome—an allostatic perspective. Biochim Biophys Acta 2010;1801(3):338–49.

[30] Stefan N, Kantartzis K, Machann J, et al. Identification and characterization of met-abolically benign obesity in humans. Arch Intern Med 2008;168:1609–16.

[31] Simpson L, Mukherjee S, Cooper MN, et al. Sex differences in the association of re-gional fat distribution with the severity of obstructive sleep apnea. Sleep 2010;33:467–74.

[32] Lévy P, Pépin JL, Arnaud C, Baguet JP, Dematteis M, Mach F. Obstructive sleepapnea and atherosclerosis. Prog Cardiovasc Dis 2009;51:400–10.

[33] Drager LF, Bortolotto LA, Figueiredo AC, Krieger EM, Lorenzi GF. Effects of contin-uous positive airway pressure on early signs of atherosclerosis in obstructivesleep apnea. Am J Respir Crit Care Med 2007;176:706–12.

[34] Manios E, Tsivgoulis G, Koroboki E, et al. Impact of prehypertension on commoncarotid artery intima-media thickness and left ventricular mass. Stroke 2009;40:1515–8.

[35] Drager LF, Bortolotto LA, Krieger EM, Lorenzi-Filho G. Additive effects of obstruc-tive sleep apnea and hypertension on early markers of carotid atherosclerosis.Hypertension 2009;53:64–9.

[36] Drager LF, Bortolotto LA, Maki-Nunes C, et al. The incremental role of obstructivesleep apnea on markers of atherosclerosis in patients with metabolic syndrome.Atherosclerosis 2010;208:490–5.

[37] Lavie L, Polotsky V. Cardiovascular aspects in obstructive sleep apnea syndrome—molecular issue, hypoxia and cytokine profiles. Respiration 2009;78:361–70.

[38] Kaynak D, Goksan B, Kaynak H, Degirmenci N, Daglioglu S. Is there a link betweenthe severity of sleep-disordered breathing and atherosclerotic disease of the ca-rotid arteries? Eur J Neurol 2003;10:487–93.

[39] Suzuki T, Nakano H, Maekawa J, et al. Obstructive sleep apnea and carotid-arteryintima-media thickness. Sleep 2004;27:129–33.

[40] Dursunoğlu N, Dursunoğlu D, Kiliç M. Impact of obstructive sleep apnea on rightventricular global function: sleep apnea and myocardial performance index. Res-piration 2005;72(3):278–84.

[41] Chahal H, McClelland RL, Tandri H, et al. Obesity and right ventricular structureand function: the MESA-Right Ventricle Study. Chest 2012;141:388–95.

[42] Iacobellis G, Willens HJ, Barbaro G, Sharma AM. Threshold values of high-riskechocardiographic epicardial fat thickness. Obesity 2008;16:887–92.

[43] Aydın H, Toprak A, Deyneli O, et al. Epicardial fat tissue thickness correlates withendothelial dysfunction and other cardiovascular risk factors in patients withmetabolic syndrome. Metab Syndr Relat Disord 2010;8:229–34.

[44] Iacobellis G. Relation of epicardial fat thickness to right ventricular cavity size inobese subjects. Am J Cardiol 2009;104(11):1601–2.

[45] Company JM, Booth FW, Laughlin MH, et al. Epicardial fat gene expression afteraerobic exercise training in pigs with coronary atherosclerosis: relationship tovisceral and subcutaneous fat. J Appl Physiol 2010;109(6):1904–12.

ckness with the severity of obstructive sleep apnea in obese patients,

6 S. Mariani et al. / International Journal of Cardiology xxx (2012) xxx–xxx

[46] Iacobellis G, Sharma AM. Epicardial adipose tissue as new cardio-metabolic riskmarker and potential therapeutic target in the metabolic syndrome. Curr PharmDes 2007;13:2180–4.

[47] Mazurek T, Zhang L, Zalewski A, et al. Human epicardial adipose tissue is a sourceof inflammatory mediators. Circulation 2003;108:2460–6.

[48] Sacks HS, Fain JN. Human epicardial adipose tissue: a review. AmHeart J 2007;153:907–17.

[49] Iacobellis G, Malavazos AE, Corsi MM. Epicardial fat: from the biomolecular as-pects to the clinical practice. Int J Biochem Cell Biol 2011;43:1651–4.

[50] Iacobellis G, Leonetti F. Epicardial adipose tissue and insulin resistance in obesesubjects. J Clin Endocrinol Metab 2005;90:6300–2.

Please cite this article as: Mariani S, et. al, Association of epicardial fat thiInt J Cardiol (2012), doi:10.1016/j.ijcard.2012.06.011

[51] Govindarajan G, Hayden MR, Cooper SA, et al. Metabolic derangements in theinsulin-resistant heart. J Cardiometab Syndr 2006;1:102–6.

[52] Iacobellis G, Leonetti F, Singh N, Sharma AM. Relationship of epicardial adiposetissue with atrial dimensions and diastolic function in morbidly obese subjects.Int J Cardiol 2007;115:272–3.

[53] Iacobellis G. Is obesity a risk factor for atrial fibrillation? Nat Clin Pract CardiovascMed 2005;2:134–5.

[54] Coats AJ, Shewan LG. Statement on authorship and publishing ethics in the inter-national journal of cardiology. Int J Cardiol 2011;153:239–40.

ckness with the severity of obstructive sleep apnea in obese patients,