Dyssynchrony in Obese Subjects without a History of Cardiac Disease Using Velocity Vector Imaging Bhaskar Purushottam, MD, Anoop C. Parameswaran, MD, MPH, and Vincent M. Figueredo, MD, Philadelphia, Pennsylvania Background: The aim of this study was to examine the occurrence of intra–left ventricular (LV) dyssynchrony in obese versus nonobese subjects without known cardiac disease using Velocity Vector Imaging (VVI). Methods: One hundred ninety consecutive subjects with no known cardiac disease had their echocardio- grams analyzed using VVI after excluding subjects with QRS durations > 120 msec or LV ejection fractions < 55%. Study subjects were divided into two groups on the basis of body mass index: obese (>30 kg/m 2 ) and nonobese (<30 kg/m 2 ). Results: The final cohort included 136 subjects (74 obese; 32% women; mean age, 55 6 16 years). The occur- rence of intra–LV dyssynchrony was higher in the obese group compared with the nonobese group. Conclusions: There was an increased prevalence of intra–LV dyssynchrony in obese subjects, especially lon- gitudinal and radial dyssynchrony. This dyssynchrony may signal a mechanism by which obesity predisposes to the development of heart failure. (J Am Soc Echocardiogr 2011;24:98-106.) Keywords: Dyssynchrony, Obese, Velocity Vector Imaging Obesity is a modern epidemic, with >60 million adults affected in the United States alone. 1 Obesity is an important risk factor for heart fail- ure in both men and women. Increased body mass index (BMI) has been reported in 11% and 14% of men and women with heart failure, respectively. 2 Left ventricular (LV) hypertrophy and dilatation, 1,3-5 which are known precursors of heart failure, 6,7 are associated with obesity. Also, obesity is associated with altered LV remodeling, possibly due to increased hemodynamic load, neurohormonal activation, and increased cytokine production. 8 Myocardial triglycer- ide content appears to increase progressively with BMI. 9 Recent experimental investigations suggest cardiac steatosis (excessive accu- mulation of cytosolic triglycerides in the myocardial cells), increased myocardial fibrosis, lipoapoptosis, and the activation of certain cardiac genes may underlie obesity cardiomyopathy. 10,11 Recent studies using positron emission tomography have found that in obese young women, insulin resistance and obesity are related to alterations in fatty acid metabolism, which could play a role in decreased cardiac performance. 11-15 Whatever these intricate and complex molecular mechanisms may be, evidence suggests that longstanding obesity results in LV structural and functional alterations, producing volume overload, eccentric LV hypertrophy, systolic and diastolic dysfunction, and heart failure. 16 Marfella et al. 17 demonstrated a higher occurrence of interventric- ular dyssynchrony in obese subjects using two-dimensional (2D) echocardiography with Doppler. To our knowledge, there have been no studies thus far examining the incidence of intra–LV dyssyn- chrony in obese subjects who do not have histories of significant car- diac disease. Myocardial contraction and relaxation are complex processes involving longitudinal, circumferential, radial, and torsional forces. Velocity Vector Imaging (VVI; Siemens Medical Solutions USA, Inc., Mountain View, CA) is a novel technique that uses myocardial speckle tracking to assess myocardial mechanics from 2D echocardi- ography. 18 VVI uses an algorithm that automatically tracks motion of the tissue-cavity border and motion of reference points (mitral annu- lus), displaying tissue motion, direction, and velocity (Figure 1). Unlike Doppler tissue imaging, VVI measures velocities independent of transducer angle. Also, in a recent study, Lim et al. 19 demonstrated that the accuracy of Doppler tissue imaging in assessing LV wall re- gional motion is limited in dilated ventricles and probably affects LV dyssynchrony measurement. Our aim was to examine the prevalence of intra–LV dyssynchrony in obese subjects with no histories of cardiac disease and compare them with nonobese controls using VVI. METHODS Five hundred consecutive subjects who underwent 2D echocardiog- raphy at Albert Einstein Medical Center in Philadelphia between November 2008 and March 2009 on an Acuson Sequoia C512 (Siemens Medical Solutions, Inc.) were screened. Three hundred ten subjects were excluded with the following exclusion criteria: (1) history of coronary artery disease; (2) LV ejection fraction (LVEF) < 55%; (3) diastolic dysfunction greater than grade 1 (mitral early to late diastolic inflow peak velocity ratio > 0.8, deceleration time of the mitral inflow < 200 msec, isovolumetric relaxation time From the Einstein Institute for Heart and Vascular Health, Albert Einstein Medical Center and Jefferson Medical College, Philadelphia, Pennsylvania. Reprint requests: Vincent M. Figueredo, MD, Albert Einstein Medical Center, 5501 Old York Road, Levy 3, Philadelphia, PA 19141 (E-mail: fi[email protected]). 0894-7317/$36.00 Copyright 2011 by the American Society of Echocardiography. doi:10.1016/j.echo.2010.10.003 98
Transcript
From the Eins
Center and Je
Reprint reque
Old York Road
0894-7317/$3
Copyright 201
doi:10.1016/j.
98
Dyssynchrony in Obese Subjects without a Historyof Cardiac Disease Using Velocity Vector Imaging
Bhaskar Purushottam, MD, Anoop C. Parameswaran, MD, MPH, and Vincent M. Figueredo, MD,Philadelphia, Pennsylvania
Background: The aim of this study was to examine the occurrence of intra–left ventricular (LV) dyssynchrony inobese versus nonobese subjects without known cardiac disease using Velocity Vector Imaging (VVI).
Methods: One hundred ninety consecutive subjects with no known cardiac disease had their echocardio-grams analyzed using VVI after excluding subjects with QRS durations > 120 msec or LV ejection fractions< 55%. Study subjects were divided into two groups on the basis of body mass index: obese (>30 kg/m2)and nonobese (<30 kg/m2).
Results: The final cohort included 136 subjects (74 obese; 32%women; mean age, 556 16 years). The occur-rence of intra–LV dyssynchrony was higher in the obese group compared with the nonobese group.
Conclusions: There was an increased prevalence of intra–LV dyssynchrony in obese subjects, especially lon-gitudinal and radial dyssynchrony. This dyssynchrony may signal a mechanism by which obesity predisposesto the development of heart failure. (J Am Soc Echocardiogr 2011;24:98-106.)
Obesity is a modern epidemic, with >60 million adults affected in theUnited States alone.1 Obesity is an important risk factor for heart fail-ure in both men and women. Increased body mass index (BMI) hasbeen reported in 11% and 14% ofmen and womenwith heart failure,respectively.2 Left ventricular (LV) hypertrophy and dilatation,1,3-5
which are known precursors of heart failure,6,7 are associated withobesity. Also, obesity is associated with altered LV remodeling,possibly due to increased hemodynamic load, neurohormonalactivation, and increased cytokine production.8 Myocardial triglycer-ide content appears to increase progressively with BMI.9 Recentexperimental investigations suggest cardiac steatosis (excessive accu-mulation of cytosolic triglycerides in the myocardial cells), increasedmyocardial fibrosis, lipoapoptosis, and the activation of certain cardiacgenes may underlie obesity cardiomyopathy.10,11 Recent studies usingpositron emission tomography have found that in obese youngwomen, insulin resistance and obesity are related to alterations infatty acid metabolism, which could play a role in decreased cardiacperformance.11-15 Whatever these intricate and complex molecularmechanisms may be, evidence suggests that longstanding obesityresults in LV structural and functional alterations, producing volumeoverload, eccentric LV hypertrophy, systolic and diastolicdysfunction, and heart failure.16
Marfella et al.17 demonstrated a higher occurrence of interventric-ular dyssynchrony in obese subjects using two-dimensional (2D)
tein Institute for Heart and Vascular Health, Albert Einstein Medical
fferson Medical College, Philadelphia, Pennsylvania.
sts: Vincent M. Figueredo, MD, Albert Einstein Medical Center, 5501
echocardiography with Doppler. To our knowledge, there havebeen no studies thus far examining the incidence of intra–LV dyssyn-chrony in obese subjects who do not have histories of significant car-diac disease.
Myocardial contraction and relaxation are complex processesinvolving longitudinal, circumferential, radial, and torsional forces.Velocity Vector Imaging (VVI; Siemens Medical Solutions USA, Inc.,Mountain View, CA) is a novel technique that uses myocardialspeckle tracking to assess myocardial mechanics from 2D echocardi-ography.18 VVI uses an algorithm that automatically tracks motion ofthe tissue-cavity border and motion of reference points (mitral annu-lus), displaying tissue motion, direction, and velocity (Figure 1). UnlikeDoppler tissue imaging, VVI measures velocities independent oftransducer angle. Also, in a recent study, Lim et al.19 demonstratedthat the accuracy of Doppler tissue imaging in assessing LV wall re-gional motion is limited in dilated ventricles and probably affects LVdyssynchrony measurement.
Our aim was to examine the prevalence of intra–LV dyssynchronyin obese subjects with no histories of cardiac disease and comparethem with nonobese controls using VVI.
METHODS
Five hundred consecutive subjects who underwent 2D echocardiog-raphy at Albert Einstein Medical Center in Philadelphia betweenNovember 2008 and March 2009 on an Acuson Sequoia C512(Siemens Medical Solutions, Inc.) were screened. Three hundredten subjects were excluded with the following exclusion criteria:(1) history of coronary artery disease; (2) LV ejection fraction(LVEF) < 55%; (3) diastolic dysfunction greater than grade 1 (mitralearly to late diastolic inflow peak velocity ratio > 0.8, decelerationtime of the mitral inflow < 200 msec, isovolumetric relaxation time
Journal of the American Society of EchocardiographyVolume 24 Number 1
Purushottam et al 99
< 60 msec, pulmonary venoussystolic to diastolic peak velocityratio < 1, and mitral early inflowto early diastolic annular septaltissue peak velocity ratio > 9, aslisted in American Society ofEchocardiography criteria20);(4) QRS duration > 120 msec;(5) moderate or severe valvularheart disease (using Dopplerechocardiographic parameters:
central jet > 4 cm2 or jet area > 20% of left atrial area for mitral re-gurgitation, central jet width > 25% or vena contracta > 0.3 cm2 orpressure half-time < 500 msec for aortic regurgitation, central jetarea > 5 cm or proximal isovelocity surface radius > 0.5 cm for tricus-pid regurgitation, jet size by color Doppler > 10 mm for pulmonaryregurgitation for regurgitant lesions21; mean gradient > 20 mm Hgor aortic valve area < 1.5 cm2 or aortic jet velocity > 3m/sec for aorticstenosis, mitral valve area < 1.5 cm2 or mean gradient > 5 mmHg formitral stenosis, tricuspid valve area < 1 cm2 or mean gradient > 5mmHg or inflow time-velocity time integral > 60 cm or pressure half-time> 190 msec for tricuspid stenosis, peak velocity > 3 m/sec or peakgradient > 36 mm Hg for pulmonic stenosis for stenotic lesions22);(6) pacemaker; (7) hypertrophic cardiomyopathy; (8) pericardial effu-sion or disease; (9) poor-quality images in which the myocardiumwasnot visible; and (10) admission to the intensive care unit. Subjectswere divided into two groups on the basis of BMI: (1) obese (BMI$ 30 kg/m2) and (2) nonobese (BMI < 30 kg/m2).We also comparedmorbidly obese subjects (BMI$ 40 kg/m2) with obese subjects (BMI$ 30 and < 40 kg/m2).
VVI was performed using the Acuson Sequoia C512. Images werecaptured using frame rates used for traditional 2D echocardiography(30–60 frames/sec). VVI uses a series of tracking algorithms whosedetails are described elsewhere.18 In brief, the endocardial-myocardial interface is traced manually in a single frame on a digitalcine loop. When the image is processed, a complex algorithm trackseach pixel, and the myocardial velocity vectors are displayed in cineformat. The lengths of the vectors are proportional to the magnitudeof velocity, and the direction of the arrows corresponds to the direc-tion of myocardial motion. One cardiac cycle was analyzed if the RRintervals were regular, and an average of three beats was used if RRintervals were irregular. Apical four-chamber, two-chamber, andshort-axis views at the papillary muscle level were studied offline. Inthe apical four-chamber and two-chamber views, a trace was made(along the endocardial-myocardial interface) from the septal to lateralmitral annulus and from the inferior to anterior mitral annulus, respec-tively. In the short-axis view at the level of papillary muscles, a circum-ferential trace was made starting at the 12 o’clock position and endingat the same point in a clockwise direction, excluding the papillarymuscles. Approximately one point per myocardial segment wasused to draw the trace. A point of reference was placed at the apexin the two-chamber and four-chamber views to calculate longitudinalvelocities and strain. The point of reference was moved to the LVcavity to calculate radial velocities. In the short-axis view, the pointof reference was at the center of the left ventricle to calculate circum-ferential velocities and strain. Longitudinal velocity, longitudinalstrain, and radial velocities weremeasured at the basal septal, basal lat-eral, basal anterior, and basal inferior walls in the apical four-chamberand two-chamber views. The circumferential velocities and strainwere measured in the short-axis view at the papillary muscle level.Time to peak velocities and strain were calculated from the onset
of the QRS complex to the peak systolic velocity or peak strain,respectively, during the ejection phase. We defined mechanical dys-synchrony as longitudinal opposing wall delay > 75 msec by VVIon the basis of a prior study.23 Because there are no published criteriafor circumferential dyssynchrony and because we were looking at themaximum delay between all six segments in the short-axis view, notjust the opposing wall delays, we used a higher number (maximumdelay $ 100 msec) to define circumferential dyssynchrony. Weused a value of 75 msec for septal–to–lateral wall radial delay.Examples of longitudinal and circumferential LV dyssynchrony analy-sis using the above-mentioned VVI technique are illustrated in Figures2 and 3, respectively.
Patient demographics, clinical characteristics, hemodynamic mea-surements, laboratory data, echocardiographic parameters, and elec-trocardiographic data were collected (Tables 1 and 2). Patientdemographics collected were age, gender, and race. Clinicalcharacteristics collected were any history of hypertension (bloodpressure > 140/90 mm Hg as defined by the seventh report of theJoint National Committee on Prevention, Detection, Evaluation,and Treatment of High Blood Pressure24 or use of antihypertensiveagents); diabetes mellitus (fasting plasma glucose > 126 mg/dL perthe American Diabetes Association25 or receipt of antidiabetic treat-ment); history of transient ischemic attack,26 ischemic stroke,27 or in-tracranial hemorrhage; hypercholesterolemia (low-densitylipoprotein level > 130 mg/d per the National CholesterolEducation Program28 or statin use); a diagnosis of obstructive sleepapnea29; a diagnosis of stable chronic obstructive pulmonarydisease30; and hemodialysis status. Hemodynamic measurementsrecorded were heart rate and systolic and diastolic blood pressure,recorded immediately before performing echocardiography.Laboratory data obtained were hemoglobin and creatinine, whichwere assessed closest to the time of echocardiography. QRS durationwas recorded from the electrocardiogram. All echocardiographicmeasurements were based on American Society ofEchocardiography guidelines. LV diastolic dimension, septal wallthickness, posterior wall thickness, left atrial diameter, and LV masswere measured using 2D-guided M-mode echocardiography, assum-ing that the left ventricle is a prolate ellipse in the parasternal long-axisacoustic window.31 LV mass was indexed to body surface area.Diastolic dysfunction and the grade of dysfunction,20 LVEF (calcu-lated using the modified Simpson’s rule31), and pulmonary artery sys-tolic pressures32 were measured.
This study was approved by the institutional research board ofAlbert Einstein Medical Center.
Statistical Analysis
Data were analyzed using SPSS version 10 (SPSS, Inc., Chicago, IL).Continuous data are presented as mean 6 SD. Means were com-pared using two-tailed Student’s t tests. Multivariate analysis was per-formed using the regression model. Chi-square tests were used tocompare categorical variables. P values < .05 were considered signif-icant. Coefficients of variation were used to measure interobserverand intraobserver variability in 20 random obese and nonobesesubjects.
RESULTS
The final cohort consisted of 136 subjects. Thirty subjects were ex-cluded from the obese group and 24 from the nonobese group.
Figure 1 Myocardial VVI. The figure shows the image with myocardial velocity vectors. The direction of the arrows represents thedirection of contraction, and the lengths of the arrows are proportional to myocardial velocity. (Top right)Myocardial velocity curves.(Bottom right) M-mode representation of dyssynchrony, with the red phase representing systole and the blue phase representingdiastole.
Figure 2 An example of longitudinal LV dyssynchrony analysis using the VVI technique. The figure shows a longitudinal VVI analysis.(Top left)Myocardial longitudinal velocity vectors are shown. (Bottom left) Longitudinal velocity curves for basal septal and basal lat-eral myocardial walls are shown. (Top right) The time to peak longitudinal velocities of the basal septal (base left) and basal lateral(base right) walls are measured in this view. In this analysis, the longitudinal septal to lateral wall delay is 34 msec.
100 Purushottam et al Journal of the American Society of EchocardiographyJanuary 2011
Among the 30 excluded obese subjects, 17 were excluded for incom-plete data and 13 for poor-quality echocardiograms. Among the 24excluded nonobese subjects, 15 were excluded for incomplete dataand 9 for poor-quality echocardiograms. In the final cohort, 74were obese subjects and 62 were nonobese controls. The mean agewas 55 6 16 years, 32% were women, the mean QRS durationwas 84 6 9 msec, and the mean LVEF was 60 6 8%. Of note, no
subject had an LVEF < 55%, while others were more hyperdynamic.Demographic and clinical data were well matched between the twogroups (Table 1). Echocardiographic, hemodynamic, and electrocar-diographic parameters are reported in Table 2. Among the 72 obesesubjects, 50 (68%) had BMIs of 30 to 40 kg/m2, and 24 (32%) hadBMIs $ 40 kg/m2 (morbid obesity). Interobserver and intraobservervariability was calculated for all the different VVI measurements done
Figure 3 An example of circumferential LV dyssynchrony analysis using the VVI technique. The figure shows the circumferential VVIanalysis of an obese subject. (Top left) Myocardial circumferential velocity vectors are shown. (Bottom left) Circumferential velocitycurves for the various myocardial segments seen in the short-axis view at the level of the papillary muscles. (Top right) The time topeak circumferential velocities of the anterior, anterolateral, inferolateral, inferior, inferoseptal, and anteroseptal walls (starting fromthe 12 o’clock position and proceeding in a clockwise direction) are shown. In this VVI analysis, the maximal circumferential wall dys-synchrony is 167 ms.
Table 1 Baseline demographic, clinical, and laboratory data
Variable
Obese
(n = 74)
Nonobese
(n = 62) P
Age (y) 53 6 13 57 6 18 .13
Men 22 (29%) 22 (35%) .58
African Americans 64 (86%) 40 (65%) <.01
Journal of the American Society of EchocardiographyVolume 24 Number 1
Purushottam et al 101
in the study for 20 random obese and nonobese subjects. The coeffi-cients of variation for longitudinal, radial, and circumferential time topeak velocities were 5%, 6%, and 6%, respectively, for interobservervariability and 5%, 5%, and 6%, respectively, for intraobserver vari-ability. The coefficients of variation for longitudinal and circumferen-tial strain were 8% and 8%, respectively, for interobserver variabilityand 7% and 8%, respectively, for intraobserver variability.
Data are expressed as mean 6 SD or as number (percentage).COPD, Chronic obstructive pulmonary disease; OSA, obstructive
sleep apnea; TIA, transient ischemic attack.
*Not all study subjects underwent sleep studies to confirm OSA.
VVI Analysis
Longitudinal Velocity. Among the obese subjects, 9.4% (n = 7)had longitudinal septal–to–lateral wall time to peak delay > 75msec, whereas none of the nonobese subjects had evidence of dys-synchrony (P = .01; Table 3). There were no significant differencesin the longitudinal absolute peak velocities of the basal myocardialwalls (measured in the apical two-chamber and four-chamber views)between the two groups.
Radial Velocity. Among the obese subjects, 31.0 % (n = 23) hadradial septal–to–lateral wall time to peak delay > 75 msec, comparedwith 8.0% (n = 5) among nonobese subjects (P < .01). There were nosignificant differences in the radial absolute peak velocities of the basalmyocardial walls (measured in the apical two-chamber and four-chamber views) between the two groups.
Circumferential Velocity. Among the obese subjects, 27.0%(n = 20) had maximum opposing wall time to peak circumferentialdelay > 100 msec, compared with 1.6% (n = 1) among nonobesesubjects (P < .01). There were no significant differences in thecircumferential absolute peak velocities between the obese andnonobese subjects.
Strain Analysis
Longitudinal Strain. Among the obese subjects, 58.1% (n = 43)had maximum opposing wall time to peak delay in longitudinal strain> 100 msec, compared with 33.8% (n = 21) among nonobese sub-jects (P < .01). There were no significant differences in longitudinalstrain between the two groups.
Table 2 Baseline hemodynamic, echocardiographic, andelectrocardiographic data
102 Purushottam et al Journal of the American Society of EchocardiographyJanuary 2011
Circumferential Strain. Among the obese subjects, 10.8% (n = 8)had maximum opposing wall time to peak delay in circumferentialstrain > 100msec, comparedwith 1.6% (n=1) among nonobese sub-jects (P= .03). There were no significant differences in circumferentialpeak strain between obese and nonobese subjects.
Comparison Between Obese and Nonobese Groups
There was significantly increased time to peak delay in longitudinal,radial, and circumferential velocities and delay in time to peak longi-tudinal and circumferential strain in obese subjects compared withnonobese subjects (Table 4). Obese subjects had higher LV diastolicdimensions, LV mass, QRS durations, and systolic and diastolic bloodpressures (but all still within normal reference limits) compared withnonobese subjects. Even after adjusting for these confounding vari-ables, in addition to age, race, gender, LVmass index, and LVEF, obesesubjects had significantly increased LV dyssynchrony compared withnonobese subjects on multivariate analysis (Table 5).
Comparison of Dyssynchrony Between Obese and MorbidlyObese Groups
There were no significant differences in longitudinal, radial, and cir-cumferential time to peak velocities and time to peak longitudinaland circumferential strain or myocardial peak velocities and peakstrain between obese and morbidly obese subjects (Table 6).
Comparison of LVEF Between Obese Subjects With andWithout Dyssynchrony
Obese subjects with time to peak delay in longitudinal septal–to–lateralwall velocity > 75 msec had lower LVEFs (55 6 0%) comparedwith obese subjects with time to peak longitudinal septal–to–lateral
wall delay < 75 msec (60 6 7%) (P < .01; Figure 4). Obese subjectswith time to peak delay in radial septal–to–lateral wall velocity > 75msec, delay in time to peak circumferential velocity > 100msec, or de-lay in time to peak longitudinal and circumferential strain had similarLVEFs compared with nonobese subjects.
Comparison of LVEF Between Obese and Nonobese SubjectsWith Dyssynchrony
LVEFs were lower in the obese group compared with the nonobesegroup among the subjects with longitudinal strain maximum oppos-ing wall delay > 100 msec (596 6% vs 626 8%, P > .05), radial sep-tal–to–lateral wall delay > 75 msec (596 7% vs 636 9%, P > .05),and radial maximum opposing wall delay > 100msec (606 7% vs 616 9%, P > .05). There were very few nonobese subjects with longitu-dinal, circumferential, and circumferential strain dyssynchrony (Table3) to make a statistically appropriate comparison.
Subjects With QRS Durations > 100 msec
Seven study subjects had QRS durations > 100 msec; six were obese.When comparing dyssynchrony between subjects with QRS dura-tions > 100 msec and those with QRS durations < 100 msec, onlylongitudinal strain septal–to–lateral wall delay was increased (86.436 39.97 vs 48.56 6 72.14 msec, P < .05).
DISCUSSION
Multiple parameters of intra–LV dyssynchrony, including radial andlongitudinal dyssynchrony, were more frequent in obese subjects
Table 4 Comparison of dyssynchrony parameters betweenobese and nonobese subjects using univariate analysis
duration, LV mass, LV mass index, and systolic and diastolic bloodpressure.
Journal of the American Society of EchocardiographyVolume 24 Number 1
Purushottam et al 103
compared with nonobese subjects. After multivariate analysis, obesityremained a significant independent predictor of intra–LV dyssyn-chrony. Interestingly, there were no significant differences in thepeak myocardial velocities or peak myocardial strain between obeseand nonobese subjects.
Obesity is associated with LV hypertrophy and dilatation,1,3-5
known precursors of heart failure.6,7 Obesity is associated withaltered LV remodeling, possibly due to increased hemodynamicload, neurohormonal activation, and increased cytokine
production.8 There are very few data examining ventricular dyssyn-chrony in obese subjects. One study by Marfella et al.17 in 2004 de-scribed interventricular dyssynchrony among premenopausal obesewomen, which improved significantly after 10% weight loss. Theyused pulmonary vein flow analysis, E/A ratio (the ratio of mitral earlyand late diastolic flow velocities), and myocardial performance index([isovolumetric relaxation time + isovolumetric contraction time]/LVejection time) to assess interventricular dyssynchrony. Tumukluet al.33 suggested that decreased regional strain rate seen in obesecompared with nonobese subjects was a reflection of subclinicalchanges in LV systolic function. However, they did not study dyssyn-chrony between these two groups.
Ten Harkel et al.34 investigated intra–LV dyssynchrony and LV vol-umes in 73 healthy adolescents (age range, 12–18 years) using real-time three-dimensional echocardiography. In contrast to the presentstudy, they found that dyssynchrony values were independent ofweight, length, and body surface area. However, there were significantdifferences in these study populations. In the present study, subjectswere older than the adolescents (55 vs 15 years). Further, TenHarkel et al. did not report the proportion of adolescents who wereoverweight or obese, therefore making it difficult to assess the associ-ation of obesity and intra–LVdyssynchrony.On similar lines,Ng et al.35
prospectively investigated the impact of age, gender, and other phys-iologic parameters on LV longitudinal and radial synchrony usingDoppler tissue imaging and 2D speckle tracking and found that dys-synchronywas independent of BMI. ThemeanBMIof the study groupwas 25.86 4.9 kg/m2, and the proportion of obese subjects was notreported. Therefore, it is difficult to arrive at any conclusions with re-gard to obesity andLVdyssynchronyon the basis of these prior studies.
Bernheim et al.36 found that patients with normal results on clini-cally indicated exercise echocardiography (LVEF > 50%) and QRSdurations < 120 msec, who had abnormal dyssynchrony parametersat rest, had higher resting heart rates and achieved lower workloads.They felt that this indicated early myocardial impairment. Changet al.37 found that LV systolic and diastolic dyssynchrony in asymp-tomatic patients with hypertension with QRS durations < 120 msecand normal-range LVEFs were significantly associated with LV fillingpressure. In view of these studies and the results of our study, in-tra–LV dyssynchrony may play a role in the mechanisms underlyingheart failure development in obese subjects.
Myocardial triglyceride content appears to increase progressivelywith BMI.9 Experimental investigations suggest that this cardiac stea-tosis (excessive accumulation of cytosolic triglycerides in the myocar-dial cells) increases myocardial fibrosis and lipoapoptosis and mayunderlie obesity cardiomyopathy.10,11 Rijzewijk et al.38 demonstratedthat myocardial steatosis is an independent predictor of diastolic dys-function in patients with type 2 diabetes mellitus, and Kankaanp€a€aet al.39 showed that the free fatty acid levels were significantly corre-lated with LV mass. These studies applied magnetic resonance imag-ing and spectroscopic techniques to quantify myocardial triglyceridecontent. However, the quantification of regional differences of triglyc-eride content in the myocardium is difficult, because the heart is per-petually in motion and is surrounded by a large depot of adipocytes(epicardial fat pad) that interferes with measurements.40 Thus,whether regional variations in myocardial steatosis exist and playa role in the observed dyssynchrony in obese patients requires furtherstudy. Of note, the obese subjects in the present study had increasedLV mass, LV diastolic dimensions, posterior wall thicknesses, bloodpressure, and QRS durations compared with nonobese subjects.However, these confounding variables did not influence our resultsafter multivariate analysis (Table 4).
Table 6 Comparison of dyssynchrony parameters betweenmorbidly obese and obese subjects using univariate analysis
Figure 4 Bar diagram comparing the mean and SD LVEFs ofobese subjects with and without longitudinal septal–to–lateralwall (S-L) delay > 75 ms.
104 Purushottam et al Journal of the American Society of EchocardiographyJanuary 2011
As mentioned earlier, we found no differences in the peak veloci-ties and strain achieved by the different myocardial walls betweenobese and nonobese subjects. Contrary to our study, Tumukluet al.,33 using reconstructed spectral pulsed-wave tissue Doppler,showed significantly decreased myocardial peak velocities and re-gional and global strain among obese subjects compared with nonob-ese controls. The present study was cross-sectional, so we could not
test the possibility of dyssynchrony preceding changes in peak veloc-ities and strain. Furthermore, dyssynchrony looks at the difference inthe time taken to achieve peak velocities or strain between opposingwalls, not at the absolute velocities or strain, and changes in these dif-ferent parameters need not simultaneously occur.
Contrary to the existing literature, there were fewer obese subjectsin our study with obstructive sleep apnea and elevated pulmonary ar-tery pressures. This could be explained by the fact that our study wasretrospective, and many of our subjects had not yet undergone sleepstudies. Thus, the reported number of subjects with obstructive sleepapnea is observational in this population. It is likely that the number ofobese subjects with obstructive sleep apnea is underestimated. Giventhe fact that we excluded subjects who hadmoderate or severe tricus-pid regurgitation, low LVEFs, diastolic dysfunction greater than grade1, intensive care unit admissions, several obese subjects with elevatedpulmonary artery pressures were excluded.
Our study raises multiple questions and possibilities regarding theoccurrence of dyssynchrony in obese subjects and its role in the causa-tion of systolic dysfunction in obese subjects.With obesity being a risingworldwide epidemic, and with its harmful effects on cardiac functionand contribution toheart failure, further studies arewarranted. Itwouldalso be interesting to see if obese individuals are better cardiac resynch-ronization therapy responders than their nonobese counterparts, as wesee an increased occurrence of dyssynchrony among obese subjects.
Limitations
This was a retrospective study limited to a single inner-city medicalcenter. Because a lower frame rate was used for VVI, it is conceivablethat some very rapid velocities may not have been recorded.Nevertheless, comparison between the time to peak velocities be-tween twowalls should remain valid. Somemagnetic resonance imag-ing studies looking at the heterogeneity in LV contraction used framerates ranging from14 to 35msec. This is comparablewith the 30 to 60frames/sec (16–33 ms) used in this study. Unlike speckle-trackingechocardiography, which can measure radial strain in short-axisviews,41 VVI cannot measure these radial velocities in the short-axisview and LV torsion that could have given us more information aboutLV function. Waist-to-hip ratio was not calculated, as these were notstandard measurements for subjects who underwent echocardiogra-phy. Waist-to-hip ratio is a stronger correlate of LV dysfunction andmortality than BMI.42 Because thiswas a retrospective study, we couldnot accurately estimate the duration of obesity for each subject andtherefore assess its effect on dyssynchrony. Because this was not a lon-gitudinal study and clinical effects were not measured, these findingsshould be viewed as thought provoking, with future studies assessingthe potential contribution of intra–LV dyssynchrony in obese subjectswith clinical end points such as heart failure.
CONCLUSIONS
There was increased intra–LV dyssynchrony among obese subjectscompared with nonobese subjects, especially longitudinal and radialdyssynchrony. This dyssynchrony may signal one mechanism bywhich obesity predisposes to the development of heart failure.
REFERENCES
1. American Obesity Association. AOA fact sheets: ‘‘obesity in the U.S.’’ Avail-able at: http://www.obesity.org/subs/fastfacts/obesity_US.shtml. AccessedFebruary 2004.
Journal of the American Society of EchocardiographyVolume 24 Number 1
Purushottam et al 105
2. Kenchaiah S, Evans JC, LevyD,Wilson PW, Benjamin EJ, LarsonMG, et al.Obesity and the risk of heart failure. N Engl J Med 2002;347:305-13.
3. Messerli FH, Sundgaard-Riise K, Reisin ED, Dreslinski GR, Ventura HO,Oigman W, et al. Dimorphic cardiac adaptation to obesity and arterial hy-pertension. Ann Intern Med 1983;9:757-61.
4. Ogden CL, Carroll MD,McDowell MA, Flegal KM. Obesity among adultsin the United States—no change since 2003-2004 (NCHS data brief no 1).Hyattsville, MD: National Center for Health Statistics; 2007.
5. Alpert MA, Lambert CR, Terry BE, Cohen MV, Mukerji V, Massey CV,et al. Influence of left ventricular mass on left ventricular diastolic fillingin normotensive morbid obesity. Am Heart J 1995;130:1068-73.
6. Vasan RS, Larson MG, Benjamin EJ, Evans JC, Levy D. Left ventriculardilatation and the risk of congestive heart failure in people without myo-cardial infarction. N Engl J Med 1997;336:1350-5.
7. Gardin JM, McClelland R, Kitzman D, Lima JA, Bommer W,Klopfenstein HS, et al. M-mode echocardiographic predictors of six- toseven-year incidence of coronary heart disease, stroke, congestive heartfailure, and mortality in an elderly cohort (the Cardiovascular HealthStudy). Am J Cardiol 2001;87:1051-7.
8. Kopelman PG. Obesity as medical problem. Nature 2000;404:635-43.9. Szczepaniak LS, Dobbins RL, Metzger GJ, Sartoni-D’Ambrosia G,
Arbique D, Vongpatanasin W, et al. Myocardial triglycerides and systolicfunction in human: in vivo evaluation by localized proton spectroscopyand cardiac imaging. Magn Reson Med 2003;49:417-23.
10. Cittadini A, Mantzoros CS, Hampton TG, Travers EK, Katz SE, Morgan PJ,et al. Cardiovascular alterations in transgenicmice with reduced brown fat:an animal model of human obesity. Circulation 1999;100:2177-83.
11. Chiu HC, Kovacs A, Ford DA, Hsu FF, Garcia R, Herrero P, et al. A novelmousemodel of lipotoxic cardiomyopathy. J Clin Invest 2001;107:813-22.
12. Peterson LR, Herrero P, Schechtmann KB, Racette SB, Waggoner AD, Kis-rieva-Ware Z, et al. Effect of obesity and insulin resistance on myocardialsubstrate metabolism and efficiency in young women. Circulation 2004;109:2191-6.
13. Marwick TH, Wong CY. Role of exercise and metabolism in heart failurewith normal ejection fraction. Prog Cardiovasc Dis 2007;49:263-74.
14. Peterson LR, Waggoner AD, Schechtmen KB, Meyer T, Gropler RJ,Barzilai B, et al. Alterations in left ventricular structure and function inyoung healthy obese women. J Am Coll Cardiol 2004;43:1399-404.
15. Di Bello V, Santini F, Di Cori A, Pucci A, Palagi C, Delle Donne MG, et al.Obesity cardiomyopathy: is it a reality? An ultrasonic tissue characteriza-tion study. J Am Soc Echocardiogr 2006;19:1063-71.
16. Flegal KM, Carrol MD, Ogden CL, Johnson CL. Prevalence and trends inobesity among US adults, 1999-2000. JAMA 2002;288:1772-3.
17. Marfella R, Esposito K, Siniscalchi M, Cacciapuoti F, Giugliano F,Labriola D, et al. Effect of weight loss on cardiac synchronization andproinflammatory cytokines in premenopausal obese women. DiabetesCare 2004;27:47-52.
18. Jackson J,HouleH. SyngoVelocityVector Imaging.Available at: http://www.medical.siemens.com/siemens/en_US/gg_us_FBAs/files/misc_downloads/Whitepaper_Jackson.pdf. Accessed October 24, 2010.
19. Lim P,Mitchell-Heggs L, Buakhamsri A, Thomas JD, GrimmRA. Impact ofleft ventricular size on tissue Doppler and longitudinal strain by speckletracking for assessing wall motion and mechanical dyssynchrony in candi-dates for cardiac resynchronization therapy. J Am Soc Echocardiogr 2009;22:695-701.
20. Nagueh SF, Appleton CP, Gillebert CT, Marino PN, Oh JK, Smiseth OA,et al. Recommendations for the evaluation of left ventricular diastolicfunction by echocardiography. J Am Soc Echocardiogr 2009;22:107-33.
21. Baumgartner H, Hung J, Bermejo J, Chambers JB, Evangelista A, Griffin BP,et al. Echocardiographic assessment of valve stenosis: EAE/ASE recom-mendations for clinical practice. J Am Soc Echocardiogr 2009;22:1-23.
22. Zoghbi WA, Enriquez-Sarano M, Foster E, Grayburn PA, Kraft CD,Levine RA, et al. Recommendations for evaluation of the severity of nativevalvular regurgitation with two-dimensional and Doppler echocardiogra-phy. J Am Soc Echocardiogr 2003;16:777-802.
23. Cannesson M, Tanabe M, Suffoletto MS, Schwartzman D, Gorcsan J III.Velocity Vector Imaging to quantify ventricular dyssynchrony and predict
response to cardiac resynchronization therapy. Am J Cardiol 2006;98:949-53.
24. Chobanian AV, Bakris GL, Black HR, CushmanWC, Green LA, Izzo JL Jr.,et al. The JNC 7 report. JAMA 2003;289:2560-72.
25. American Diabetes Association. Position statement. Standards of medicalcare in diabetes—2010. Diabetes Care 2010;33(suppl):S11-61.
26. Easton JD, Saver JL, Albers GW, Alberts MJ, Chaturvedi S, Feldmann E,et al. Definition and evaluation of transient ischemic attack: a scientificstatement for healthcare professionals from the American Heart Associa-tion/American Stroke Association Stroke Council; Council on Cardiovas-cular Surgery and Anesthesia; Council on Cardiovascular Radiology andIntervention; Council on Cardiovascular Nursing; and the Interdisciplin-ary Council on Peripheral Vascular Disease. Stroke 2009;40:2276-93.
27. van der Worp HB, van Gijn J. Clinical practice. Acute ischemic stroke. NEngl J Med 2007;357:572-9.
28. Expert Panel onDetection, Evaluation, and Treatment of High BloodCho-lesterol in Adults. Third report of the Expert Panel on Detection, Evalua-tion, and Treatment of High Blood Cholesterol in Adults (Adult TreatmentPanel III): final report. Available at: http://www.nhlbi.nih.gov/guidelines/cholesterol/atp3full.pdf. Accessed October 24, 2010.
29. Patil SP, Schneider H, Schwartz A, Smith P. Adult obstructive sleep apnea:pathophysiology and diagnosis. Chest 2007;132:325-37.
30. Qaseem A, SnowV, Shekelle P, Sherif K, Wilt TJ, Weinberger S, et al. Clin-ical Efficacy Assessment Subcommittee of the American College of Phy-sicians. Diagnosis and management of stable chronic obstructivepulmonary disease: a clinical practice guideline from the American Col-lege of Physicians. Ann Intern Med 2007;147:633-8.
31. Lang RM, Bierig M, Devereux RB, Flachskampf FA, Foster E, Pellikka PA,et al. Chamber Quantification Writing Group; American Society of Echo-cardiography’s Guidelines and Standards Committee; European Associa-tion of Echocardiography. Recommendations for chamberquantification: a report from the American Society of Echocardiography’sGuidelines and Standards Committee and the Chamber QuantificationWriting Group, developed in conjunction with the European Associationof Echocardiography, a branch of the European Society of Cardiology.J Am Soc Echocardiogr 2005;18:1440-63.
32. Rudski LG, Lai WW, Afilalo J, Hua L, Handschumacher M,Chandrasekaran K, et al. Guidelines for the echocardiographic assessmentof the right heart in adults: a report from the American Society of Echocar-diography. Endorsed by the European Association of Echocardiography,a registered branch of the European Society of Cardiology, and theCanadian Society of Echocardiography. J Am Soc Echocardiogr 2010;23:685-713.
33. Tumuklu MM, Etikan I, Kisacik B, Kayikcioglu M. Effect of obesity on leftventricular structure and myocardial systolic function: assessment by tis-sue Doppler imaging and strain/strain rate imaging. Echocardiography2007;24:802-9.
34. Ten Harkel AD, VanOsch-GeversM, HelbingWA. Real-time transthoracicthree dimensional echocardiography: normal reference data for left ven-tricular dyssynchrony in adolescents. J Am Soc Echocardiogr 2009;22:933-8.
35. Ng AC, Tran DT, Newman M, Allman C, Vidaic J, Lo ST, et al. Left ven-tricular longitudinal and radial synchrony and their determinants inhealthy subjects. J Am Soc Echocardiogr 2008;21:1042-8.
36. Bernheim AM, Nakajima Y, Pellikka PA. Left ventricular dyssynchrony inpatients with normal ventricular systolic function referred for exerciseechocardiography. J Am Soc Echocardiogr 2008;21:1145-9.
37. Chang SA, Kim HK, Kim DH, Kim YJ, Sohn DW, Oh BH, et al. Left ven-tricular systolic and diastolic dyssynchrony in asymptomatic hypertensivepatients. J Am Soc Echocardiogr 2009;22:337-42.
38. Rijzewijk LJ, Van der Meer RW, Smit JW, Diamant M, Bax JJ, Hammer S,et al. Myocardial steatosis is an independent predictor of diastolic dysfunc-tion in type 2 diabetes mellitus. J Am Coll Cardiol 2008;52:1793-9.
39. Kankaanp€a€a M, Lehto HR, P€arkk€a JP, Komu M, Viljanen A, Ferrannini E,et al. Myocardial triglyceride content and epicardial fat mass in humanobesity: relationship to left ventricular function and serum free fattyacid levels. J Clin Endocrinol Metab 2006;91:4689-95.
106 Purushottam et al Journal of the American Society of EchocardiographyJanuary 2011
40. McGavock JM, Victor RG, Unger RH, Szczepaniak LS. Adiposity of theheart, revisited. Ann Intern Med 2006;144:517-24.
41. Gorcsan J III, Abraham T, Agler DA, Bax JJ, Derumeaux G, Grimm RA,et al. Echocardiography for cardiac resynchronization therapy: recommen-dations for performance and reporting—a report from the American Soci-
ety of Echocardiography Dyssynchrony Writing Group endorsed by theHeart Rhythm Society. J Am Soc Echocardiogr 2008;21:191-213.
42. Ammar KA, Redfield MM, Mahoney DW, Johnson M, Jacobsen SJ,Rodeheffer RJ. Central obesity: association with left ventricular dysfunc-tion and mortality in the community. Am Heart J 2008;156:975-81.
