From the Chi

(C.F.S, A.C.,

Pittsburgh of

St. Louis, Mis

the Universit

University of

of Pittsburgh,

Affairs Medica

This study w

04S5) to the

Pittsburgh (5U

of this report,

Affairs or the

Texas Childre

Research Ce

M01-RR0018

Washington U

RR024992 fr

component of

Feasibility and Reproducibility of EchocardiographicMeasures in Children with Muscular Dystrophies

Christopher F. Spurney, MD, Francis M. McCaffrey, MD, Avital Cnaan, PhD, Lauren P. Morgenroth, MS,Sunil J. Ghelani, MD, Heather Gordish-Dressman, MD, PhD, Adrienne Arrieta, MS, Anne M. Connolly, MD,

Timothy E. Lotze, MD, Craig M. McDonald, MD, Robert T. Leshner, MD,and Paula R. Clemens, MD, Washington, District of Columbia; Pittsburgh, Pennsylvania; St. Louis, Missouri;

Houston, Texas; Sacramento and San Diego, California

Background: Cardiac disease is a major cause of death in patients with muscular dystrophies. The use offeasible and reproducible echocardiographic measures of cardiac function is critical to advance the field oftherapeutics for dystrophic cardiomyopathy.

Methods: Participants aged 8 to 18 years with genetically confirmed Duchenne muscular dystrophy (DMD),Becker muscular dystrophy, or limb-girdle muscular dystrophy were enrolled at five centers, and standardizedechocardiographic examinations were performed. Measures of systolic and diastolic function and speckle-tracking echocardiography–derived cardiac strain were reviewed independently by two central readers.Furthermore, echocardiographic measures from participants with DMDwere compared with those from retro-spective age-matched control subjects from a single site to assess measures of myocardial function.

Results: Forty-eight participants (mean age, 13.36 2.7 years) were enrolled. Shortening fraction had a greaterinterobserver correlation (intraclass correlation coefficient [ICC] = 0.63) compared with ejection fraction(ICC = 0.49). One reader could measure ejection fraction in only 53% of participants. Myocardial performanceindex measured by pulse-wave Doppler and Doppler tissue imaging showed similar ICCs (0.55 and 0.54).Speckle-tracking echocardiography showed a high ICC (0.96). Focusing on participants with DMD (n = 33),significantly increased mitral A-wave velocities, lower E/A ratios, and lower Doppler tissue imaging mitrallateral E0 velocities were observed compared with age-matched control subjects. Speckle-tracking echocar-diography demonstrated subclinical myocardial dysfunction with decreased average circumferential and lon-gitudinal strain in three distinct subgroups: participants with DMD with normal shortening fractions,participants with DMD aged < 13 years, and participants with DMDwith myocardial performance index scores< 0.40 compared with control subjects.

Conclusions: In a muscular dystrophy cohort, assessment of cardiac function is feasible and reproducible us-ing shortening fraction, diastolic measures, andmyocardial performance index. Cardiac strainmeasures iden-tified early myocardial disease in patients with DMD. (J Am Soc Echocardiogr 2015;-:---.)

Keywords: Muscular dystrophy, Echocardiography, Cardiac strain, Cardiomyopathy

ldren’s National Health System, Washington, District of Columbia

L.P.M., S.J.G., H.G.-D., A.A., R.T.L.); Children’s Hospital of

UPMC, Pittsburgh, Pennsylvania (F.M.M.); Washington University,

souri (A.M.C.); Texas Children’s Hospital, Houston, Texas (T.E.L.);

y of California, Davis, Sacramento, California (C.M.M.); the

California, San Diego, San Diego, California (R.T.L.); the University

Pittsburgh, Pennsylvania (P.R.C.); and the Department of Veterans

l Center, Pittsburgh, Pennsylvania (P.R.C.).

as supported by an administrative supplement (3UL1RR024153-

Clinical and Translational Science Award held by the University of

LRR024153-03). The authors take full responsibility for the contents

which do not represent the views of the US Department of Veterans

US government.

n’s Hospital: The Baylor College of Medicine General Clinical

nter grant was funded by the National Institutes of Health (NIH),

8, General Clinical Research Center.

niversity: This publication was also made possible by grant UL1

om the National Center for Research Resources (NCRR), a

the NIH, and the NIH Roadmap for Medical Research. Its contents

are solely the responsibility of the authors and do not necessarily represent the offi-

cial view of the NCRR or the NIH.

University of California, Davis: The project described was supported by the Na-

tional Center for Advancing Translational Sciences, NIH, through grant UL1

TR000002. The content is solely the responsibility of the authors and does not

necessarily represent the official views of the NIH.

Children’s National Health System: This publication was supported by award

UL1TR000075 from the NIH National Center for Advancing Translational Sciences.

Its contents are solely the responsibility of the authors and do not necessarily

represent the official views of the National Center for Advancing Translational Sci-

ences or the NIH.

Reprint requests: Christopher F. Spurney, MD, Children’s National Health System,

Division of Cardiology, 111Michigan Avenue, NW,Washington, DC 20010 (E-mail:

cspurney@cnmc.org).

0894-7317/$36.00

Copyright 2015 by the American Society of Echocardiography.

http://dx.doi.org/10.1016/j.echo.2015.03.003

1

Abbreviations

BMD = Becker musculardystrophy

CINRG = CooperativeInternational Neuromuscular

Research Group

CS = Circumferential strain

DMD = Duchenne muscular

dystrophy

DTI = Doppler tissue imaging

EF% = Percentage ejection

fraction

ET = Ejection time

ICC = Intraclass correlation

coefficient

IVCT = Isovolumic

contraction time

IVRT = Isovolumic relaxation

time

LGMD = Limb-girdle

muscular dystrophy

LS = Longitudinal strain

MD = Muscular dystrophy

MPI = Myocardial

performance index

PWD = Pulse-wave Doppler

SF% = Percentage

shortening fraction

STE = Speckle-tracking

echocardiography

2 Spurney et al Journal of the American Society of Echocardiography- 2015

Cardiomyopathy causes signifi-cant morbidity and mortality inmultiple forms of muscular dys-trophy (MD) affecting children,including Duchenne MD(DMD), Becker MD (BMD),and subtypes of autosomal-recessive limb-girdle MD(LGMD).1 The prevalence ofcardiomyopathy increases withage and is currently underdiag-nosed in patients with DMD.2,3

Therefore, it is increasinglyimportant to study cardiacdisease in muscular dystrophiesto determine the best diagnosticand treatment modalities.

Adequately powered pharma-ceutical studies for the treatmentof MD require the collection ofreproducible data across multi-ple clinical sites because of therarity of these diseases. Severalstudies have researched cardiacend points, but comparisonsacross studies are hindered bydifferent study designs and func-tional measures, which includepercentage shortening fraction(SF%), percentage ejection frac-tion (EF%), mitral inflow veloc-ities, myocardial performanceindex (MPI), and myocardialstrain.4-25 These measures arealso affected by technicallimitations in patients with MD,including scoliosis, barrel chestdeformities with lung

hyperinflation, and limited mobility.The purpose of this study was to assess the feasibility and reproduc-

ibility of noninvasive echocardiography-based functional cardiacmeasures in a multicenter cohort of participants with MD and deter-mine which measures can detect early subclinical changes in myocar-dial function. The Cooperative International NeuromuscularResearch Group (CINRG) is a coalition of academic clinical centersdedicated to MD research. CINRG has validated skeletal musclefunctional measures for multisite DMD studies.26 The results of thisstudy will help direct the selection of cardiac measurements for futureclinical trials and enhance consistency within the field of cardiomyop-athies in patients with MD.

METHODS

This was a multicenter, prospective study investigating differentechocardiographic parameters with participants enrolling at five insti-tutions in the CINRG network, all also part of the Clinical andTranslational Science Award network (http://www.ctsacentral.org).The study was approved by the institutional review board at eachinstitution. Written informed consent and assent were obtainedfrom all participants and their parents or legal guardians. The study

was registered with ClinicalTrials.gov (identifier NCT01066455).Participants had confirmed genetic diagnoses of DMD, BMD, or anautosomal-recessive subtype of LGMD (LGMD2C, LGMD2D,LGMD2E, LGMD2F, or LGMD2I). Participants were excluded ifthey had histories of congenital cardiac defects or other cardiac dis-eases unrelated to MD. Retrospective control subjects for participantswith DMDwere patients referred to the cardiology clinic at one of thefive participating CINRG clinical sites (Children’s National HealthSystem) for cardiac murmur or chest pain and were determined tohave normal results on cardiac evaluation with normal conventionalechocardiographic parameters. Race and ethnicity were self-reported and not required for control subjects.

Methods to Determine the Feasibility and Reproducibilityof Noninvasive Echocardiography-Based FunctionalCardiac Measures

A central sonographer traveled to each participating CINRG site toreview and train the site personnel on the use of a centralized protocolusing standard imaging planes and recording three-beat loops saved inDigital Imaging and Communications in Medicine format.Echocardiographic studies were performed with participants in thesupine position on an examination table or seated in a power wheel-chair if transfer to an examination table was not possible. These digitalloops were interpreted by two pediatric cardiologists and one pediat-ric cardiology fellow. There were two readers at the Washington site(C.F.S. and S.J.G. [fellow]) and one reader at the Pittsburgh site(F.M.M.).All measurements and calculations, including SF%, EF% using the

single-plane modified Simpson protocol, wall stress, velocity ofcircumferential shortening, rate-corrected velocity of circumferentialshortening, MPI (which includes measures of ejection time [ET], iso-volumic contraction time [IVCT], and isovolumic relaxation time[IVRT]) using both pulse-wave Doppler (PWD) and Doppler tissueimaging (DTI), mitral inflow, and left ventricular peak E-wave veloc-ities, were made according to standards of the American Society ofEchocardiography.27 Each reader (C.F.S. and F.M.M.) measuredtwo-dimensional and Doppler values over three cardiac cycles (orthe maximum feasible when imaging was of limited quality), andthe average was used for analysis. To assess intraobserver variability,47 echocardiograms were reassessed by the same observer (C.F.S.) af-ter a period of $2 weeks. Interobserver variability was assessed byhaving a different reader (F.M.M.) perform all the measures on theconventional echocardiographic images independently.

Methods to Determine Early Subclinical MyocardialChanges

Cardiac strain was measured using speckle-tracking echocardi-ography (STE) on the subset of participants with DMD from allfive participating CINRG centers and historical control subjectsfrom a single institution (Children’s National Health System), us-ing proprietary software (Syngo Velocity Vector Imaging; SiemensMedical Solutions USA Inc, Mountain View, CA). Speckle-tracking echocardiographic analysis was performed by tworeaders (C.F.S. and S.J.G.). For STE, endocardial tracings of theleft ventricle were manually performed in the apical four-chamber view (for longitudinal measurements) and the paraster-nal short-axis view at the level of the midpapillary muscles (forcircumferential measurements). A single cardiac beat with thebest-appearing image quality was used. Tracking was

Journal of the American Society of EchocardiographyVolume - Number -

Spurney et al 3

automatically performed by the software, and the analysis wasaccepted as satisfactory only after visual inspection. Endocardialtracing and automatic tracking were performed twice on eachbeat, and the average of the two measurements was recordedfor each variable. If tracking was suboptimal, the endocardialborder was retraced until either satisfactory tracking was accom-plished or 5 min had passed, in which case the view wasexcluded from analysis. Measured deformation parametersincluded average peak systolic longitudinal strain (LS), averagepeak systolic longitudinal strain rate, average peak systoliccircumferential strain (CS), and average peak systolic circumfer-ential strain rate (CSR). Segmental data were not analyzed.Mathematically, all of these parameters are negative (indicatingshortening), and a smaller absolute value indicates worse ventric-ular function. To assess intraobserver variability for STE, arandomly selected set of 11 echocardiograms (six from patientswith DMD, five from control subjects) were reassessed by thesame observer (S.J.G.) after a period of $2 weeks.Interobserver variability was assessed by having a differentobserver (C.F.S.) perform speckle-tracking echocardiographicmeasurements on all possible DMD echocardiograms using thesame beat as the original observer.

Statistical Analysis

Descriptive measurements, including demographic characteristics(age, ethnicity, and race), blood pressure, heart rate, and bodymass index, are summarized as mean 6 SD or as ranges or fre-quencies appropriate for each data type. Age, blood pressure, andheart rate were compared between the overall MD group and theDMD control subjects using t tests, while body mass index requiredlog-transformed data for t tests. Both the number of participants andthe distribution of diagnosis (DMD, BMD, and LGMD) werecompared by medication use (glucocorticoids, cardiac medications,coenzyme Q10) with exact c2 analysis. Mean age in the differentmedications use group was compared with a t test. Race andethnicity were not consistently captured for the control subjectsand hence were not compared with these variables among patientswith MD.Nineteen continuous echocardiographic parameters are summa-

rized as mean 6 SD and range for each reader. Agreement betweenechocardiographic measurements was assessed by calculation of theintraclass correlation coefficient (ICC) both between the tworeaders (interobserver) and between repeated measurements fromreader 1 (intraobserver). Bland-Altman plots were producedshowing the mean bias and 95% limits of agreement. EF% and SF% were compared using t tests, while MPI required log-transformed data for the t test. Left ventricular EF%, SF%, andMPI measurements were also dichotomized between normal andabnormal on the basis of the following criteria: EF% (normal,$55; abnormal, <55), SF% (normal, $28; abnormal, <28), andDTI and PWD MPI (normal, <0.40; abnormal, $0.40).28,29 Thesedichotomized variables were then compared pairwise to evaluateagreement in normal or abnormal designation using a McNemartest.Continuous measurements from STE, along with basic demo-

graphic characteristics, are summarized as mean 6 SD and werecompared between patients with DMD and control subjects using ttests. Subgroup analyses were performed in the same manner onthose with normal SF% or EF%, those aged < 13 years, and thosewith normal MPI values (<0.40).

All analyses used Stata version 13 (StataCorp LP, College Station,TX), and P values < .05 were considered to indicate statistical signif-icance. No adjustments for multiple tests were done, because of theobservational focus of this study.

RESULTS

Patient Characteristics

Information for all participants (those withMD and retrospective con-trol subjects) is included in Table 1. Fifty-two participants with MDprovided consent at five sites, 48 of whom completed the study.Thirty-three control subjects without MD were used for subset anal-ysis of participants with DMD. Participants with MD had geneticallyconfirmed diagnoses of DMD (73%), BMD (23%), or LGMD (4%).Themean age in theMD cohort was 13.36 2.7 years, with a range of8.4 to 17.7 years, and the mean age in the control cohort was12.8 6 2.8 years, with a range of 8.7 to 17.9 years. Information onother patient characteristics, including demographics, vital signs,body mass index, and past and current medications (glucocorticoids,cardiac medications, and coenzyme Q10), for the MD cohort areshown in Table 1.

Feasibility

On the basis of image quality, only one study could not be read byboth readers (C.F.S. and F.M.M.), and some measures were not ob-tained in all studies by both readers. For EF%, reader 1 was able toconfidently trace the endocardial border in only 25 of 48 studies(52%) and reader 2 in 43 of 48 studies (90%). For reader 1, theaverage age of participants in whom EF% could not be measuredwas 14.0 6 2.7 years, compared with 12.7 6 2.6 years for studieswith measured EF% (P < .08). Six participants could not transferout of their wheelchairs, and reader 1 was unable to measure EF%in four of these participants.

Reproducibility

The means, ranges, and interobserver ICCs for all echocardio-graphic measures from readers 1 and 2 are presented in Table 2.Using SF%, 40 participants (85%) showed normal systolic functionand seven participants (15%) had decreased systolic function.M-mode measures of left ventricular internal diameters in diastoleand in systole used for SF% had two of the three highest ICClevels for all conventional echocardiographic measures. Althoughthere was no single case in which EF% was normal and SF%was abnormal, there were six cases with decreased EF% valuesand normal SF% values. SF% measurements also showed highagreement between repeated measurements by the same reader(intraobserver ICC = 0.73). We did not observe any pattern of dif-ferences in SF% values between two readers (Figure 1A) or bydiagnosis (Figure 1B). The variability of SF% remained consistentover the range of observed magnitudes (Figure 2A) and over theage range of participants in this study (Figure 2B), although thetwo readers showed some differences in range for the youngestage group.

Early Subclinical Myocardial Involvement

Both DTI and PWD methods were used to determine MPI as a mea-sure of subclinical myocardial disease. Between the two readers, theMPI measured via PWD had a similar ICC compared with the MPI

Table 1 Participant characteristics

Characteristic

Muscular dystrophy cohort

Controls for DMD cohort PDMD BMD LGMD All

Number 35 (73%) 11 (23%) 2 (4%) 48 (100%) 33 (100%)

Age (y) 13 6 3 14 6 3 14 6 2 13 6 3 13 6 3 .43

Ethnicity

Hispanic 5 (14%) 0 (0%) 1 (50%) 6 (13%) NA

Non-Hispanic 30 (86%) 11 (100%) 1 (50%) 42 (88%) NA

Race

Caucasian 29 (83%) 10 (91%) 1 (50%) 40 (83%) NA

Asian 2 (6%) 0 (0%) 0 (0%) 2 (4%) NA

African American 3 (9%) 1 (9%) 1 (50%) 5 (10%) NA

Other 1 (3%) 0 (0%) 0 (0%) 1 (2%) NA

Vital signs

SBP (mm Hg) 109 6 13 111 6 9 95 6 3 109 6 12 115 6 10 .02

DBP (mm Hg) 65 6 8 64 6 9 60 6 5 65 6 7 64 6 7 .76

Heart rate (beats/min) 92 6 14 79 6 12 93 6 1 89 6 15 67 6 13 <.001

BMI (kg/m2) 24 6 7 20 6 3 18 6 2 23 6 7 22 6 6 .68

Glucocorticoid use .003

Past users 6 (17%) 0 (0%) 1 (50%) 7 (15%) NA

Current users 24 (69%) 4 (36%) 1 (50%) 29 (61%) NA

Never used 5 (14%) 7 (64%) 0 (0%) 12 (25%) NA

Cardiac medication use .76

Past users 1 (3%) 0 (0%) 0 (0%) 1 (2%) NA

Current users 13 (37%) 4 (36%) 0 (0%) 17 (36%) NA

Never used 21 (60%) 7 (64%) 2 (100%) 30 (63%) NA

CoQ10 use .99

Past users 3 (9%) 2 (18%) 0 (0%) 5 (10%) NA

Current users 8 (23%) 1 (9%) 0 (0%) 9 (19%) NA

Never used 24 (69%) 8 (73%) 2 (100%) 34 (71%) NA

BMI, Body mass index; CoQ10, coenzyme Q10; DBP, diastolic blood pressure; NA, not applicable; SBP, systolic blood pressure.

Data are expressed as number (percentage) or as mean6 SD. P values are from comparisons between the control subjects (n = 33) and the entire

MD cohort (n = 48) for characteristics available in the entire cohort and are between MD diagnoses for characteristics applicable only to the MD

group.

4 Spurney et al Journal of the American Society of Echocardiography- 2015

measured via DTI (Table 2). Looking at specific components of theMPI measure (IVCT, IVRT, and ET), the PWD MPI measures consis-tently had lower correlations compared with the DTI MPI measures,and the ICC for ET measured in DTI tracings showed the highest cor-relation of all measures. In our MD cohort, PWD- and DTI-derivedMPI values demonstrated agreement in only 29 of 46 of the studies(63%) when using an MPI value of 0.40 as a cutoff between normaland abnormal (Figure 3). In cases in which SF% was$28, PWD MPIwas$0.40 in 23 of 38 (61%), and DTI MPI was$0.40 in only 14 of38 (37%). Thus, DTI MPI showed a lower rate of defining subclinicaldisease compared with PWD MPI when SF% was normal. To betterunderstand the differences between PWD and DTI measures ofMPI, we performed a subanalysis on the 15 cases in the upper leftquadrant of Figure 3 with DTI MPI values <0.40 and PWD MPIvalues $0.40. In these cases, DTI measured significantly longer ETs

(267 6 19 msec via DTI vs 239 6 17 msec via PWD, P < .001)and shorter IVRTs (48 6 8 msec via DTI vs 52 6 9 msec via PWD,P < .001), leading to a lower calcula-ted DTI MPI. There were no sig-nificant differences in IVCT measurements. For ET and IVRT mea-sures, a higher ICC was observed for each using DTI comparedwith PWD (see Table 2).

STE was a second technique used to determine subclinicalmyocardial disease. Only participants with DMD were includedin this analysis, because of the often earlier onset of cardiomyop-athy in DMD compared with BMD and LGMD. In this subset of33 participants with DMD (two participant echocardiogramsdid not upload into Syngo), satisfactory longitudinal tracking wasobtained in 30 participants (91%), and satisfactory circumferentialtracking was obtained in 28 participants (85%). The meanframe rates for the apical four-chamber view in DMD and control

Table 2 Echocardiographic measures for readers 1 and 2 and interobserver correlations between the two readers

Cardiac measure

Reader 1 Reader 2 Readers 1 and 2

n Mean 6 SD Range n Mean 6 SD Range

Interobserver ICC

(95% CI)*

Conventional

LVIDd (mm) 47 4.5 6 0.6 3.3–5.9 47 4.4 6 0.6 3.2–5.8 0.88 (0.80–0.93)

LVIDs (mm) 47 3.0 6 0.6 2.2–4.9 47 3.0 6 0.6 2.1–4.6 0.89 (0.82–0.94)

SF% 47 33 6 6 17–46 47 32 6 5 18–43 0.63 (0.42–0.77)

EF% 25 61 6 8 44–79 43 54 6 8 24–64 0.49 (0.19–0.70)

WS (g/cm2) 46 56 6 16 20–112 47 61 6 19 27–116 0.38 (0.12–0.59)

VCF (circ/s) 46 1.3 6 0.3 0.7–2.1 46 1.2 6 0.3 0.6–1.8 0.31 (0.05–0.54)

VCFc (circ/s) 46 1.1 6 0.2 0.6–1.5 46 1.0 6 0.2 0.5–1.4 0.35 (0.07–0.58)

PWD

ET (msec) 47 256 6 26 200–310 46 250 6 20 210–290 0.71 (0.50–0.83)

IVRT (msec) 47 51 6 11 30–80 46 53 6 10 33–77 0.43 (0.16–0.63)

IVCT (msec) 47 49 6 10 30–80 46 49 6 11 30–75 0.41 (0.14–0.62)

MPI 46 0.39 6 0.09 0.26–0.67 46 0.41 6 0.08 0.28–0.60 0.55 (0.27–0.73)

MV E wave (cm/sec) 46 86 6 15 52–124 46 91 6 14 63–123 0.82 (0.53–0.92)

MV A wave (cm/sec) 46 49 6 10 27–78 46 54 6 10 36–75 0.63 (0.17–0.82)

DTI

ET (msec) 46 266 6 25 210–330 46 266 6 22 210–313 0.92 (0.85–0.95)

IVRT (msec) 46 49 6 9 30–70 46 47 6 12 30–70 0.61 (0.40–0.76)

IVCT (msec) 46 52 6 11 30–80 46 48 6 11 20–73 0.57 (0.30–0.74)

MPI 46 0.39 6 0.09 0.26–0.63 46 0.36 6 0.09 0.23–0.60 0.54 (0.28–0.83)

Septal LV peak E0

velocity (cm/sec)

46 12 6 3 3–17 46 14 6 3 7–19 0.61 (0.02–0.83)

Lateral LV peak E0

velocity (cm/sec)

46 15 6 5 0.9–32 46 18 6 5 7–34 0.72 (0.12–0.89)

STE

LS (%) 30 �14 6 4 �21 6 7 30 �15 6 4 �23 6 7 0.90 (0.77–0.95)

CS (%) 28 �20 6 5 �33 6 11 28 �20 6 5 �31 6 12 0.96 (0.92–0.98)

LSR (sec�1) 30 �1.1 6 0.4 �2.4 6 0.5 30 �1.3 6 0.5 �2.7 6 0.5 0.64 (0.35–0.82)

CSR (sec�1) 28 �1.9 6 0.5 �2.7 6 0.9 28 �2.1 6 0.7 �3.4 6 0.8 0.78 (0.47–0.91)

LSR, Longitudinal strain rate; LV, left ventricular; LVIDd, left ventricular internal diameter in diastole; LVIDs, left ventricular internal diameter in sys-

tole; MV, mitral valve; VCF, velocity of circumferential fiber shortening; VCFc, velocity of circumferential fiber shortening corrected for heart rate;

WS, wall stress.*ICC uses lesser value in cases of discrepant n values.

Journal of the American Society of EchocardiographyVolume - Number -

Spurney et al 5

echocardiograms were 64 6 13 and 53 6 12 frames/sec, respec-tively. The mean frame rates for the parasternal short-axis viewin DMD and control echocardiograms were 72 6 18and 56 6 10 frames/sec, respectively. Comparisons of conven-tional and speckle-tracking echocardiographic data between partic-ipants with DMD and control subjects are shown in Table 3.Although both groups demonstrated mean SF% and EF% withinthe normal ranges, these values in control subjects were signifi-cantly higher, as would be expected (P < .001). Participants withDMD had significantly decreased E/A ratios compared with con-trols (P < .001). Average LS, longitudinal strain rate, CS, andCSR were significantly decreased (less negative) in participantswith DMD compared with controls (P < .001). Other measure-ments, in particular the DTI MPI, were not significantly differentbetween the DMD group and control subjects. Conventionalechocardiographic measurements demonstrated normal cardiac

systolic function (SF% $ 28) in 26 of the 33 participantswith DMD (79%). This subgroup with SF% $ 28 showed signifi-cantly decreased average LS, CS, and CSR compared with controls(Table 4). Seventeen of the 33 participants with DMD (52%) were<13 years of age (range, 7.7–12.8 years). These younger partici-pants also showed decreased average LS, CS, and CSR comparedwith controls (17 of 33 [52%]) in the same age group (Table 4).Twenty-two of the 33 participants (67%) had a DTI MPI values<0.40 and also demonstrated significantly decreased average LS,CS, and CSR compared with control subjects (Table 4). AverageLS and CS demonstrated high interobserver correlations, withICCs of 0.94 and 0.96, respectively.

Bland-Altman analysis was performed on SF%, EF%, MPI, and STE(Figure 4). Overall, these plots demonstrated no remarkable differ-ences in fixed or proportional biases between the two readers forthe analyzed measures.

Figure 2 Assessment of variability in SF% measurements across SF% magnitudes (A; a rolling average with SD bars across therange of observed SF% values) and across different patient ages (B; average with SD bars shown by 2-year intervals) with reader1 in blue and reader 2 in red.

Figure 3 Agreement between MPI via PWD and DTI. Blue dotsrepresent participants deemed normal by both PWD and TDI;red dots represent participants deemed abnormal by bothPWD and TDI; black dots represent participants for whomPWD and TDI MPI disagreed.

Figure 1 Agreement of SF%measures within and between readers: (A) agreement within the same reader (blue is measure 1 and redis measure 2) and (B) agreement between readers separated by diagnosis (blue, DMD; red, BMD; black, LGMD. For (A) and (B), thecloser the values are to the identity line, the smaller the discrepancy between the two reads.

6 Spurney et al Journal of the American Society of Echocardiography- 2015

DISCUSSION

As clinical therapies for skeletal and respiratory dysfunction in pa-tients with MD improve, cardiac disease is becoming an increasinglyimportant factor in morbidity and mortality. The investigation of

innovative therapies that may have their greatest benefit if imple-mented early in the disease course will depend on cardiac measuresthat can be compared across sites and studies. The purpose of thisstudy was twofold: (1) to determine the feasibility and reproducibilityof echocardiographic functional cardiac measures using independentcentral readers and a central sonographer who trained site sonogra-phers and (2) to identify echocardiographic measures of early subclin-ical changes in myocardial function.

Systolic function is a commonly used measure clinically and inresearch studies, but methods of measurement can vary greatly. Inthis study, we showed that SF% was the preferred measure in ourMD cohort. SF% could be reproducibly measured by two indepen-dent readers. Using M-mode imaging, the measures of left ventricularinternal diameters in systole and in diastole had two of the three high-est correlation values (ICCs of 0.88 and 0.89, respectively). The ICCfor SF% was greater than that for EF% (0.63 vs 0.49). SF% measuresalso did not significantly vary with age or magnitude of measure. SF%was also more feasible in the study population. One reader could onlyconfidently trace the endocardial border for calculation of EF% in53% of participants, which reflected difficulties in performing two-dimensional imaging in the apical four-chamber position in patientswith MD due to body positioning and lung interference. This diffi-culty was also identified in a previous study by Markham et al.6 Ourdata provide strong support for the use of SF% as the best measureof cardiac systolic function in patients with MD. In contrast, EF%was difficult to obtain in many participants, especially older oneswith more advanced disease who were unable to transfer from chairs,and was less reproducible across readers. A single-plane EF%

Table 4 Comparison of speckle-tracking-derived deformation parameters between control subjects and three subgroups ofparticipants with DMD: (1) those with SF% >28, (2) those <13 years of age, and (3) those with DTI MPI values <0.40

Measure

SF% >28 Age <13 y MPI <0.40

DMD

(n = 26)

Control subjects

(n = 28) P

DMD

(n = 17)

Control subjects

(n = 17) P

DMD

(n = 22)

Control subjects

(n = 25) P

SF% 34 6 4 38 6 4 <.001 34 6 4 38 6 4 .015 32 6 5 38 6 4 <.001

Average LV LS (%) �14.3 6 3.6 �17.8 6 3.0 <.001 �15.3 6 3.6 �17.6 6 2.3 .033 �14.5 6 3.4 �17.4 6 3.0 .003

Average LV CS (%) �20.2 6 4.6 �25.5 6 3.5 <.001 �19.3 6 3.9 �25.9 6 3.5 <.001 �20.7 6 5.1 �25.7 6 3.2 <.001

Average LV LSR (sec�1) �1.2 6 0.4 �1.3 6 0.3 .16 �1.2 6 0.3 �1.3 6 0.2 .07 �1.1 6 0.3 �1.3 6 0.3 .045

Average LV CSR (sec�1) �1.9 6 0.5 �2.4 6 0.5 .001 �1.9 6 0.5 �2.3 6 0.5 .013 �1.9 6 0.5 �2.4 6 0.5 .004

Data are expressed as mean 6 SD.

Table 3 Comparison of conventional and speckle-tracking echocardiographic data in participants with DMD and control subjects

Measure DMD (n = 33) Control subjects (n = 33) P

SF% 33 6 5 38 6 4 <.001

EF% 58 6 6 63 6 5 .001

MPI (DTI) 0.36 6 0.07 0.35 6 0.06 .49

MV E-wave velocity (cm/sec) 94 6 13 100 6 14 .15

MV A-wave velocity (cm/sec) 58 6 12 48 6 14 .005

MV E/A ratio 1.7 6 .3 2.2 6 .6 <.001

Septal E0 velocity (cm/sec) 14 6 3 14 6 3 .67

Lateral E0 velocity (cm/sec) 17 6 5 20 6 3 .014

Septal E/E0 ratio 7 6 2 7 6 2 .79

Lateral E/E0’ ratio 6 6 2 5 6 1 .09

Average LV LS (%) �14.1 6 3.5 (n = 30) �17.7 6 3.4 <.001

Average LV CS (%) �19.8 6 4.8 (n = 27) �25.7 6 3.5 <.001

Average LV LSR (sec�1) �1.1 6 0.4 (n = 30) �1.3 6 0.3 .033

Average LV CSR (sec�1) �1.8 6 0.6 (n = 27) �2.4 6 0.5 <.001

Data are expressed as mean 6 SD.

Journal of the American Society of EchocardiographyVolume - Number -

Spurney et al 7

measurement was also chosen to improve the likelihood of obtainingsufficient images, but this has known limitations and did not improvethe measures.30,31

We also looked for early subclinical markers of myocardial dysfunc-tion, because most young children with MD have normal systolicfunction. A decrease in diastolic function often predates the declinein systolic function and may provide a useful early marker of cardiacdisease in children.32 We found significantly increased mitral A-wavevelocities and decreased mitral DTI lateral peak E-wave velocities inparticipants with DMD compared with control subjects. Shabanianet al.22 also found increased mitral A-wave velocities, decreased mitralE-wave velocities, and decreased DTI lateral peak E-wave velocities inpatients with DMD. Furthermore, we observed a good correlationbetween readers for PWD mitral E-wave measures (ICC = 0.82),PWD mitral A-wave velocities (ICC = 0.63), and DTI lateral LVpeak E0 velocities (ICC = 0.72). These results demonstrate that con-ventional measures of diastolic function are easy to obtain and candetect early changes in myocardial function during longitudinalstudies in young children with DMD.

Another echocardiographic measure designed to identify earlymyocardial disease is the MPI. MPI incorporates measures of systolicfunction (IVCT and ET) and diastolic function (IVRT) and is

independent of heart rate, ventricular geometry, and preload andafterload.33 Previous studies showed that MPI was increased in par-ticipants with MD who demonstrated normal cardiac function bysystolic measures.11,22,34,35 We therefore looked at the utility ofMPI as an early marker of cardiac dysfunction in our DMDcohort. PWD MPI detected subclinical disease in more subjectsthan DTI MPI. However, when comparing the measurements ofMPI between PWD and DTI, it was clear that DTI measurementcomponents had higher correlations between readers. On thebasis of our ICC measures, the DTI MPI was more reproduciblebetween readers and was the preferred method. Also, Bland-Altman analysis of PWD versus DTI MPI measures showed thatoverall, PWD measures were approximately 0.05 units greaterthan DTI measures (Figure 4D). Shabanian et al.22 also studiedMPI measures in participants with DMD using both conventionalPWD and DTI and found higher MPI values using PWD comparedwith DTI. Thus, it is worth considering using a lower threshold forDTI MPI than for PWD MPI to diagnose subclinical cardiac diseasein patients with DMD.

Myocardial strain has emerged as a promising echocardiographicmeasure of subclinical myocardial function.Myocardial strain assessesthe change inmyocardial fiber length comparedwith its original length

Figure 4 Bland-Altman plots demonstrate variability of interobserver measures for SF% (A), EF% (B), MPI (interobserver differencesin C and intermethod [PWD vs DTI] inD), CS% (E), and LS%, (F). The red dashed line represents the mean bias, and the shaded areaindicates the area between the limits of agreement.

8 Spurney et al Journal of the American Society of Echocardiography- 2015

in the plane inwhich it is measured.36 Unlike EF% and SF%,which arelimited to one plane, strain can bemeasured in three planes, providinga more in-depth determination of cardiac function. Although strainimaging can be measured using DTI or STE, STE is superior becauseit is independent of angle andmeasures only active contraction, avoid-ing the tethering effect of noncontractile tissue. In recent years, severalstudies have supported the utility of STE to measure cardiac strain inclinical practice, with excellent reproducibility.37-40

We analyzed myocardial strain measured by STE in a subset of par-ticipants with DMD because this diagnosis is usually associated withearlier onset of cardiomyopathy. We found that average LS and CSmeasurements were significantly decreased in participants withDMD compared with normal control subjects. STE also had veryhigh interobserver ICC values; this correlation may have beenaugmented by measuring the exact same cardiac beat by the two in-dependent observers. When limiting the analyses to participants with

normal systolic function, those <13 years of age, or those with normalDTI MPI values, each time STE demonstrated significantly decreasedaverage LS and CS values. Therefore, the measurement of strain bySTE detected loss of myocardial function, while other conventionalechocardiographic measures remained normal, permitting the studyof cardiac involvement in MD before the development of systolicdysfunction.

Our findings are consistent with other recent studies of cardiacstrain in patients with DMD. Ryan et al.24 compared participantswith DMD<8 years of age with age-matched normal control subjectsand found that despite normal SF% values, average CS at the mid-chamber level was significantly decreased in the participants withDMD (�19.86 4.2% vs �21.76 3.8%, P < .01). Inter- and intraob-server ICCs were 0.88 and 0.91, respectively. Our present studyconfirmed these findings by demonstrating an average CS of �19.36 4% in participants with DMD <13 years of age. These results

Journal of the American Society of EchocardiographyVolume - Number -

Spurney et al 9

demonstrate the usefulness of strain measures to detect subclinicaldisease in the DMD population.

This study had some limitations that reflect constraints of currentechocardiographic practice and the disease process. Echocar-diograms were obtained at different centers with varying imagequality. Six patients were unable to transfer to a supine position forechocardiographic imaging and remained in their power chairs forthe study, contributing to poor acoustic windows often seen in MD.In apical four-chamber imaging, it could be difficult to accurately traceendocardial borders for EF. Some imaging had relatively low framerates, which decreased resolution even further. A primary goal ofthis study was to assess the effects of these limitations as they influ-ence therapeutic research protocols to discover treatment optionsfor cardiac involvement in MD. The study can therefore aid in iden-tifying which echocardiographic measures are most robust for clinicalresearch.

CONCLUSIONS

This study supports the utility of SF% as the most feasible andreproducible conventional echocardiographic measure of systolicfunction in MD. Diastolic measures of mitral A velocity, lowerE/A ratio, lower mitral lateral E0 velocity DTI and MPI valuesprovided assessment of preclinical myocardial disease in patientswith MD. STE-derived myocardial strain detected the presenceof subclinical myocardial dysfunction in participants with DMDwith normal systolic function on conventional echocardiography.Lower average strain values were also seen in subgroupsof younger participants and those with normal DTI MPI. Use ofthese feasible and reproducible cardiac measures for future clinicaltrials in MD will facilitate the assessment of potential therapeuticagents.

ACKNOWLEDGMENTS

University of Pittsburgh: The authors acknowledge the support pro-vided by the director of the University of Pittsburgh Clinical andTranslational Science Institute, Steven Reis, MD, and his staff. The au-thors thank the patients and families who volunteered their time totake part in this project. We also thank the dedicated CINRG mem-bers who continue to commit countless hours to this effort. For thisproject the CINRG members are comprised of the following institu-tions and individuals: Children’s National Health System: MargaretLasota, Mohammad Ahmed, Fengming Hu, Sarah Kaminski,Donovan Stock, and Angela Zimmerman; University of Pittsburgh:Kate Hughes and Reiko Shimizu; Washington University: ReneeRenna; Texas Children’s Hospital: Robert McNeil; and University ofCalifornia Davis: Erica Goude.

REFERENCES

1. Spurney CF. Cardiac complications in neuromuscular disorders. In:Bertorini TE, editor. Neuromuscular Disorders: Treatment and Manage-ment. Philadelphia, PA: Elsevier; 2011. pp. 33-50.

2. Nigro G, Comi LI, Politano L, Bain RJ. The incidence and evolution of car-diomyopathy in Duchenne muscular dystrophy. Int J Cardiol 1990;26:271-7.

3. Spurney C, Shimizu R, Morgenroth LP, Kolski H, Gordish-Dressman H,Clemens PR. Cooperative International Neuromuscular Research GroupDuchenne natural history study demonstrates insufficient diagnosis andtreatment of cardiomyopathy in Duchenne muscular dystrophy. MuscleNerve 2014;50:250-6.

4. Kajimoto H, Ishigaki K, Okumura K, Tomimatsu H, Nakazawa M, Saito K,et al. Beta-blocker therapy for cardiac dysfunction in patients withmuscular dystrophy. Circ J 2006;70:991-4.

5. Kirchmann C, Kececioglu D, Korinthenberg R, Dittrich S. Echocardio-graphic and electrocardiographic findings of cardiomyopathy inDuchenne and Becker-Kiener muscular dystrophies. Pediatr Cardiol2005;26:66-72.

6. Markham LW, Kinnett K, Wong BL, Woodrow Benson D, Cripe LH.Corticosteroid treatment retards development of ventricular dysfunc-tion in Duchenne muscular dystrophy. Neuromuscul Disord 2008;18:365-70.

7. Markham LW, Spicer RL, Khoury PR, Wong BL, Mathews KD, Cripe LH.Steroid therapy and cardiac function in Duchenne muscular dystrophy.-Pediatr Cardiol 2005;26:768-71.

8. Ramaciotti C, Heistein LC, Coursey M, Lemler MS, Eapen RS,Iannaccone ST, et al. Left ventricular function and response to enalaprilin patients with Duchenne muscular dystrophy during the second decadeof life. Am J Cardiol 2006;98:825-7.

9. Thomas TO, Morgan TM, Burnette WB, Markham LW. Correlation ofheart rate and cardiac dysfunction in Duchenne muscular dystrophy.Pediatr Cardiol 2012;33:1175-9.

10. Gulati S, Saxena A, Kumar V, Kalra V. Duchenne muscular dystrophy:prevalence and patterns of cardiac involvement. Indian J Pediatr 2005;72:389-93.

11. Jefferies JL, Eidem BW, Belmont JW, Craigen WJ, Ware SM, Fernbach SD,et al. Genetic predictors and remodeling of dilated cardiomyopathy inmuscular dystrophy. Circulation 2005;112:2799-804.

12. Kwon SW, Kang SW, Kim JY, Choi EY, Yoon YW, Park YM, et al. Out-comes of cardiac involvement in patients with late-stage Duchennemuscular dystrophy under management in the pulmonary rehabilitationcenter of a tertiary referral hospital. Cardiology 2012;121:186-93.

13. Ogata H, Ishikawa Y, Minami R. Beneficial effects of beta-blockers andangiotensin-converting enzyme inhibitors in Duchenne muscular dystro-phy. J Cardiol 2009;53:72-8.

14. Viollet L, Thrush PT, Flanigan KM, Mendell JR, Allen HD. Effects ofangiotensin-converting enzyme inhibitors and/or beta blockers on the car-diomyopathy in Duchenne muscular dystrophy. Am J Cardiol 2012;110:98-102.

15. Houde S, Filiatrault M, Fournier A, Dube J, D’Arcy S, Berube D, et al.Deflazacort use in Duchenne muscular dystrophy: an 8-year follow-up.Pediatr Neurol 2008;38:200-6.

16. Kaspar RW, Allen HD, Ray WC, Alvarez CE, Kissel JT, Pestronk A, et al.Analysis of dystrophin deletion mutations predicts age of cardiomyopathyonset in Becker muscular dystrophy. Circ Cardiovasc Genet 2009;2:544-51.

17. Silversides CK, Webb GD, Harris VA, Biggar DW. Effects of deflazacort onleft ventricular function in patients with Duchenne muscular dystrophy.Am J Cardiol 2003;91:769-72.

18. Duboc D, Meune C, Lerebours G, Devaux JY, Vaksmann G, Becane HM.Effect of perindopril on the onset and progression of left ventriculardysfunction in Duchenne muscular dystrophy. J Am Coll Cardiol 2005;45:855-7.

19. Duboc D, Meune C, Pierre B, Wahbi K, Eymard B, Toutain A, et al. Peri-ndopril preventive treatment on mortality in Duchenne muscular dystro-phy: 10 years’ follow-up. Am Heart J 2007;154:596-602.

20. Rhodes J, Margossian R, Darras BT, Colan SD, Jenkins KJ, Geva T, et al.Safety and efficacy of carvedilol therapy for patients with dilated cardio-myopathy secondary to muscular dystrophy. Pediatr Cardiol 2008;29:343-51.

21. van Bockel EA, Lind JS, Zijlstra JG, Wijkstra PJ, Meijer PM, van denBerg MP, et al. Cardiac assessment of patients with late stage Duchennemuscular dystrophy. Neth Heart J 2009;17:232-7.

10 Spurney et al Journal of the American Society of Echocardiography- 2015

22. Shabanian R, Aboozari M, Kiani A, Seifirad S, Zamani G,NahalimoghaddamA, et al. Myocardial performance index and atrial ejec-tion force in patients with Duchenne’s muscular dystrophy. Echocardiog-raphy 2011;28:1088-94.

23. Kwon HW, Kwon BS, Kim GB, Chae JH, Park JD, Bae EJ, et al. The effectof enalapril and carvedilol on left ventricular dysfunction in middle child-hood and adolescent patients with muscular dystrophy. Korean Circ J2012;42:184-91.

24. Ryan TD, Taylor MD, Mazur W, Cripe LH, Pratt J, King EC, et al.Abnormal circumferential strain is present in young Duchenne musculardystrophy patients. Pediatr Cardiol 2013;34:1159-65.

25. Yamamoto T, TanakaH,Matsumoto K, Lee T, AwanoH, YagiM, et al. Util-ity of transmural myocardial strain profile for prediction of early left ven-tricular dysfunction in patients with Duchenne muscular dystrophy. Am JCardiol 2013;111:902-7.

26. Mayhew JE, Florence JM, Mayhew TP, Henricson EK, Leshner RT,McCarter RJ, et al. Reliable surrogate outcome measures in multicenterclinical trials of Duchenne muscular dystrophy. Muscle Nerve 2007;35:36-42.

27. Lopez L, Colan SD, Frommelt PC, Ensing GJ, Kendall K, Younoszai AK,et al. Recommendations for quantification methods during the perfor-mance of a pediatric echocardiogram: a report from the Pediatric Mea-surements Writing Group of the American Society of EchocardiographyPediatric and Congenital Heart Disease Council. J Am Soc Echocardiogr2010;23:465-95. quiz 576–7.

28. Eidem BW, Tei C, O’Leary PW, Cetta F, Seward JB. Nongeometric quan-titative assessment of right and left ventricular function:myocardial perfor-mance index in normal children and patients with Ebstein anomaly. J AmSoc Echocardiogr 1998;11:849-56.

29. Eidem BW, Sapp BG, Suarez CR, Cetta F. Usefulness of the myocardialperformance index for early detection of anthracycline-induced cardio-toxicity in children. Am J Cardiol 2001;87:1120-2. A9.

30. Schiller NB, Shah PM, Crawford M, DeMaria A, Devereux R,Feigenbaum H, et al. Recommendations for quantitation of the leftventricle by two-dimensional echocardiography. American Society ofEchocardiography Committee on Standards, Subcommittee on

Quantitation of Two-Dimensional Echocardiograms. J Am Soc Echocar-diogr 1989;2:358-67.

31. St John Sutton M, Otterstat JE, Plappert T, Parker A, Sekarski D,Keane MG, et al. Quantitation of left ventricular volumes and ejectionfraction in post-infarction patients from biplane and single planetwo-dimensional echocardiograms. A prospective longitudinal study of371 patients. Eur Heart J 1998;19:808-16.

32. Frommelt PC. Echocardiographic measures of diastolic function in pediat-ric heart disease. Curr Opin Cardiol 2006;21:194-9.

33. Tei C, Ling LH, Hodge DO, Bailey KR, Oh JK, Rodeheffer RJ, et al. Newindex of combined systolic and diastolic myocardial performance: asimple and reproducible measure of cardiac function—a study in normalsand dilated cardiomyopathy. J Cardiol 1995;26:357-66.

34. Gurrala RR, Alla VM, Aronow WS, Shankar JS, Angamutta MK,Lanka K Sr., et al. Occult left ventricular dysfunction diagnosed bymyocardial performance index in patients with limb girdle muscle dystro-phy: a case control study. Int J Angiol 2007;16:139-42.

35. Mertens L, Ganame J, Claus P, Goemans N, Thijs D, Eyskens B, et al. Earlyregional myocardial dysfunction in young patients with Duchennemuscular dystrophy. J Am Soc Echocardiogr 2008;21:1049-54.

36. Gorcsan J 3rd, Tanaka H. Echocardiographic assessment of myocardialstrain. J Am Coll Cardiol 2011;58:1401-13.

37. Basu S, Frank LH, Fenton KE, Sable CA, Levy RJ, Berger JT. Two-dimen-sional speckle tracking imaging detects impaired myocardial performancein childrenwith septic shock, not recognized by conventional echocardiog-raphy. Pediatr Crit Care Med 2012;13:259-64.

38. Koopman LP, McCrindle BW, Slorach C, Chahal N, Hui W, Sarkola T,et al. Interaction between myocardial and vascular changes in obese chil-dren: a pilot study. J Am Soc Echocardiogr 2012;25:401-4101.

39. Blanc J, Stos B, de Montalembert M, Bonnet D, Boudjemline Y. Right ven-tricular systolic strain is altered in children with sickle cell disease. J Am SocEchocardiogr 2012;25:511-7.

40. Cheung YF, HongWJ, Chan GC,Wong SJ, Ha SY. Left ventricular myocar-dial deformation and mechanical dyssynchrony in children with normalventricular shortening fraction after anthracycline therapy. Heart 2010;96:1137-41.