Echocardiographic assessment of canine degenerative … · Echocardiographic assessment of canine degenerative mitral valve disease Vale´rie Chetboul, DVM, PhD a,b,*, ... Echocardiography;

* CorrespondEcole National

E-mail addr

1760-2734/$ -doi:10.1016/j.

Journal of Veterinary Cardiology (2012) 14, 127e148

www.elsevier.com/locate/jvc

Echocardiographic assessment of caninedegenerative mitral valve disease

Valerie Chetboul, DVM, PhD a,b,*, Renaud Tissier, DVM, PhD b,c

aUniversite Paris-Est, Ecole Nationale Veterinaire d’Alfort, Unite de Cardiologie d’Alfort (UCA),Centre Hospitalier Universitaire Veterinaire d’Alfort (CHUVA), 7 avenue du general de Gaulle,94704 Maisons-Alfort cedex, Franceb INSERM, U955, Equipe 03, 51 avenue du Marechal de Lattre de Tassigny, 94010 Creteil cedex, FrancecUniversite Paris-Est, Ecole Nationale Veterinaire d’Alfort, Unite de Pharmacie-Toxicologie,7 avenue du general de Gaulle, 94704 Maisons-Alfort cedex, France

Received 6 September 2011; received in revised form 30 October 2011; accepted 1 November 2011

KEYWORDS
Canine;Doppler;Echocardiography;Strain;Tissue Doppler
ing author. Unite de Ce Veterinaire d’Alfort,ess: vchetboul@vet-alf

see front matter ª 201jvc.2011.11.005

Abstract Degenerative mitral valve disease (MVD), the most common acquiredheart disease in small-sized dogs, is characterized by valvular degeneration result-ing in systolic mitral valve regurgitation (MR). Worsening of MR leads to severalcombined complications including cardiac remodeling, increased left ventricularfilling pressure, pulmonary arterial hypertension, and myocardial dysfunction.Conventional two-dimensional, M-mode, and Doppler examination plays a criticalrole in the initial and longitudinal assessment of dogs affected by MVD, providinginformation on mitral valve anatomy, MR severity, left ventricular (LV) size andfunction, as well as cardiac and vascular pressures. Several standard echocardio-graphic variables have been shown to be related to clinical outcome. Some ofthese markers (e.g., left atrium to aorta ratio, regurgitation fraction, pulmonaryarterial pressure) may also help in identifying asymptomatic MVD dogs at higherrisk of early decompensation, which remains a major issue in practice. However,both afterload and preload are altered during the disease course. This representsa limitation of conventional techniques to accurately assess myocardial function,as most corresponding variables are load-dependent. Recent ultrasound tech-niques including tissue Doppler imaging, strain and strain rate imaging, andspeckle tracking echocardiography, provide new parameters to assess regionaland global myocardial performance (e.g., myocardial velocities and gradients,deformation and rate of deformation, and mechanical synchrony). As illustration,

ardiologie d’Alfort (UCA), Centre Hospitalier Universitaire Veterinaire d’Alfort (CHUVA),7 avenue du general de Gaulle, 94704 Maisons-Alfort cedex, France.ort.fr (V. Chetboul).

2 Elsevier B.V. All rights reserved.

mailto:[email protected]

http://dx.doi.org/10.1016/j.jvc.2011.11.005

http://www.elsevier.com/locate/jvc

128 V. Chetboul, R. Tissier

Abbreviations

ARJ/LAA regurgitant jet aratrium area

CT chordae tendineaeCTR chordae tendineae rEF% ejection fractionESVI end-systolic volumeFS% fractional shorteninIVRT isovolumic relaxatioLA left atriumLA/Ao left atrium to aortaLV left ventricleMR mitral regurgitationMVD mitral valve diseaseNT-proBNP N-terminal pro-B

peptidePAH pulmonary arterial hPISA proximal isovelocityRF regurgitant fractionSTE speckle tracking echTDI tissue Doppler imag

the authors present new data obtained from a population of 91 dogs (74 MVD dogs,17 age-matched controls) using strain imaging, and showing a significant longitu-dinal systolic alteration at the latest MVD heart failure stage.ª 2012 Elsevier B.V. All rights reserved.

ea signal to left

upture

indexgn time

ratio

-type natriuretic

ypertensionsurface area

ocardiographying

Introduction

Degenerative mitral valve disease (MVD) is themost common acquired canine heart disease.1,2 Itsprevalence can attain 14%e40% in small-sized dogsdepending on the breed, and even reaches highervalues in geriatric canine populations.3e5 Largebreed dogs such as German Shepherds can also beaffected by the disease.6,7 Whichever the caninebreed considered, MVD is characterized by chronicmyxomatous mitral valve degeneration resulting inthickening and incomplete apposition of the valveleaflets during systole with secondary mitral valveregurgitation (MR),8e10 the severity of which isa major determinant of the natural diseaseprogression. Although most dogs with MVD remainasymptomatic for years and even for life,11e15

severe complications can occur concomitantlywith MR worsening, including left- and then right-sided congestive heart failure secondary topulmonary arterial hypertension (PAH).16e20 Thisultimately leads to death or euthanasia due tounresponsive symptoms.

Because of the potential deleterious conse-quences and high prevalence of MVD, its accuratediagnosis and the monitoring of its progressionover time are critical clinical concerns for pre-dicting the risk of decompensation, guiding prog-nosis and adapting medical prescription.

The standard transthoracic echocardiographicexamination is currently considered as the non-invasive diagnostic method of choice for earlydetection of the mitral valve lesions, evaluation ofMR severity, and lastly, for assessing its impact oncardiac remodeling, myocardial function, leftventricular (LV) filling pressures as well as pulmo-nary arterial pressure.18e23 However, owing tovolume overload and complex hemodynamicchanges associated with disease progression, thedetection of myocardial dysfunction in the settingof chronic MR still remains challenging.22e28

Nevertheless, more recent advances in ultra-sound technology, together with the introductionof other imaging modalities such as tissue Dopplerimaging (TDI), strain and strain rate imaging, andtwo-dimensional (2D) speckle tracking echocardi-ography (STE), currently offer new opportunitiesfor assessing and monitoring global and regionalmyocardial function over time.29

The present article provides a review of theconventional echocardiographic alterations asso-ciated with canine MVD, including a criticalapproach to the assessment of LV contractileperformance, followed by an outline of the morerecent ultrasound techniques and their impact onthe understanding of MVD-associated myocardialdysfunction.

Identification of mitral valve lesions

Macroscopically mitral valve lesions associatedwith MVD are firstly characterized by small,smooth nodules on the leaflet tips (Fig. 1) andthickened chordae tendineae (CT), which may beidentified by 2D and M-mode echocardiography(Fig. 2). These nodular deformations are usuallygreater for the anterior leaflet (Fig. 2A and C) andbecome thicker and more irregular during diseaseprogression.2,21 Abnormal systolic flattening of oneor both mitral valve leaflets and then mitral valve

Figure 1 Three-dimensional transthoracic echocardiogram obtained from a dog with mild mitral valve disease (viewfrom the apex) showing the nodular deformation of the 2 mitral valve leaflets (arrows). LA: left atrium. LV: leftventricle. LVOT: left ventricular outflow tract.

Figure 2 (2Ae2C): Echocardiograms obtained from 3 dogs with mitral valve disease (2A and 2B, two-dimensionalright parasternal 5- and 4-chamber views, respectively; 2C, M-mode echocardiogram recorded at the level of themitral valve). Mitral valve lesions are characterized by the presence of nodules at the leaflet tips (arrows, 2A) andthick chordae tendineae (double arrow, 2B). The nodular deformation is greater for the anterior leaflet (2A). The M-mode image (2C) also shows an irregular thickened anterior mitral valve leaflet (arrow) with a short mitral-inter-ventricular septum distance related to a hyperkinetic state. Ao: aorta. IVS: interventricular septum. LA: left atrium.LV: left ventricle. LVFW: left ventricular free wall.

Echocardiography of mitral valve disease 129


prolapse, which is characterized by one or bothleaflets buckling back into the left atrium (LA)during systole (Fig. I e Data available inSupplemental Data on-line), are also commonechocardiographic findings.30,31 In one recent studyinvolving 537 dogs with MVD and mitral valveprolapse, a significant correlation was foundbetween the severity of mitral valve prolapse andMR severity as well as the International Small AnimalCardiac Health Council (ISACHC) class.31 Prolapse ofthe anterior leaflet was present in 48% of therecruited animals, that of the posterior leaflet inonly 7%, and bi-leaflet prolapse was detected in45%.31 Such a distribution is different from thattypically found in humans, where the posteriorleaflet is more commonly involved whereas soleinvolvement of the anterior leaflet is rare.32

Chordae tendineae are major components ofthe atrioventricular valve apparatus, whichdetermine the position of and tension on theleaflets at end-systole, thus contributing to propersystolic closure of the mitral valve.24 Chordaetendineae rupture (CTR) therefore representsa severe MVD-associated lesion, usually respon-sible for marked MR. It can be accurately diag-nosed using 2D rather than M-modeechocardiography (Fig. II e Data available inSupplemental Data on-line).33,34 Rupture ofprimary CT is seen in several 2D imaging planes,whereas that of secondary CT is usually identifiedin only a single view. In one study involving 706dogs with MVD, CTR was diagnosed in 16% of casesand CTR prevalence was shown to increase withheart failure class.34 In the great majority of dogs(96.5%), the ruptured CT had been attached to theanterior mitral valve leaflet (Fig. II e Dataavailable in Supplemental Data on-line). As ex-pected, all dogs with CTR had moderate (7%) tosevere MR (93%), and the prevalence of PAH wasrelatively high (35%) although most recruited dogs(87%) were under medical treatment.34 Interest-ingly, 25% of the dogs with CTR were asymptomaticat the time of diagnosis and belonged to eitherclasses 1a (7%) or 1b (18%) of the ISACHC classifi-cation (see example in Fig. III e Data available inSupplemental Data on-line). All these werereferred for a cardiovascular check-up, because ofa left apical systolic heart murmur detected oncardiac auscultation by the referring veterinarypractitioner. Similarly, in humans, CTR may be anincidental echo finding (in up to 70% of casesaccording to studies), and the detection of CTR inasymptomatic patients with MVD is currently anindication of early surgical repair in order toimprove long-term survival.35e38 Although, in thecanine study, the asymptomatic dogs with CTR

represented only 2% of dogs with MVD class 1a and21% of dogs with MVD class 1b (out of a total of 412and 96 animals, respectively),34 these data illus-trate the potential interest of performing echo-cardiographic examinations on asymptomatic MVDdogs, in order to help identify those with the mostsevere lesions that are at a higher risk ofdecompensation.

Assessment of MR severity

Evaluation of MR severity is of critical importancein dogs with MVD, as MR directly reflects theprimary hemodynamic consequence of incompleteapposition of the mitral valve leaflets duringsystole.

Semi-quantification of MR

One of the methods commonly used to assess MRseverity in dogs with MVD consists of calculatingthe maximal ratio of the regurgitant jet area signalto LA area (ARJ/LAA ratio) using color-flow Dopplermode (Fig. 3).9,10,21 On the basis of this Dopplermethod, MR is usually considered as mild if theARJ/LAA ratio is <20e30% (Fig. 3A), moderate if itis �20e30% but �70%, or severe if it is >70%(Fig. 3B).9,10,21,39 The major advantage of thiscolor Doppler mapping method is the rapidity andease of data acquisition,39 and its good repeat-ability and reproducibility in the awake dog fora trained observer.10 However, this technique onlyallows a semi-quantification of MR, as the ARJ/LAAratio compares 2 areas without any assessment ofthe regurgitant volume. Moreover, the ARJ/LAAratio may be influenced by several factorsincluding systemic arterial blood pressure, LApressure, spatial orientation of the jet (eccentricwall-impinging jets can appear smaller thancentrally-directed jets of similar hemodynamicseverity), pulse repetition frequency and gainsettings.39 Another limitation of the ARJ/LAA ratiois that the maximum of 100% is attained in a rela-tively large percentage of dogs in each heartfailure class, which precludes an accuratediscrimination between these dogs with “signifi-cant” MR.10

The vena contracta is the narrowest portion ofthe MR jet that occurs at or just downstream fromthe regurgitant orifice.39 The width of the venacontracta seen by color-flow imaging is a surrogatefor the regurgitant orifice size and representsanother approach to estimate MR severity.39

However, this method is prone to errors

Figure 3 Semi-quantitative assessment of mitral insufficiency using the color-flow Doppler mapping method in 2dogs suffering from mitral valve disease of various grades (left parasternal 4-chamber views). In Fig. 6A, the color-flowjet of mitral insufficiency (arrow) involves less than 20% of the left atrial chamber, which is consistent with mildvalvular regurgitation. Conversely, in Fig. 6B, the color-flow turbulent jet of mitral insufficiency (arrow) fills up nearlyall the left atrial chamber, and enters into the pulmonary veins (vp), which is consistent with severe valvularregurgitation. LA: left atrium. LV: left ventricle.


(particularly in case of dynamic regurgitantorifices”) and few data are available in the dog.39

Quantification of MR

The PISA (Proximal Isovelocity Surface Area)method, also called the flow convergence method,is another Doppler technique routinely used inhuman medicine to quantify (rather than “semi-quantify”) mitral valve regurgitation (see details inFig. IV e Data available in Supplemental Data on-line).39 Although the PISA method is more time-consuming than color mapping and although manyprecautions need to be taken to ensure optimalacquisition of the flow convergence images, thistechnique has been shown to be repeatable andreproducible in the awake-dog for a trainedobserver.10 Its main advantage is the more reliable“discrimination” of MR severity compared to thecolor Doppler mapping method, providingmeasurement of the flow rate through the regur-gitant orifice, calculation of the regurgitantvolume, and assessment of the regurgitation frac-tion (RF, i.e., percentage of stroke volume ejectedinto the LA during systole).8,10,39 Mitral insuffi-ciency is usually considered as moderate andsevere, for RF values higher than 30e50% and 75%,respectively.8,10,14 In several studies performed on

dogs with MVD, RF was significantly correlated withclinical parameters (ISACHC class, heart murmurgrade), and a significant correlation was demon-strated between RF and several indirect echo-Doppler markers of MVD severity including the LAsize as assessed by the LA/aorta ratio (LA/Ao) andsystolic pulmonary arterial pressure.10,14,40 Asignificant correlation has also been shownbetween N-terminal pro-B-type natriuretic peptideplasma concentrations (NT-proBNP) and RF both inasymptomatic and symptomatic dogs with MVD.14,40

Interestingly, the ARJ/LAA ratio assessed bycolor Doppler mapping has also been shown to besignificantly correlated with RF.10 The formerindex may therefore be used to assess MR severityand may also be useful for longitudinal follow-upof canine MVD despite its previously explainedlimitations. It may also sometimes compensate forseveral PISA method limitations: the PISA methodis more time-consuming and is more accurate forregurgitations with a circular orifice rather thana non-circular orifice.41,42 Moreover, it may some-times be difficult to judge the precise location ofthe orifice and the flow convergence shape.41,42 Inaddition, the PISA method only applies to hol-osystolic MR. Lastly, the presence of multipleregurgitant jets, which can occur in dogs withMVD, also precludes use of the PISA method.41,42


Indirect assessment of MR severity

Extensive left heart dilation, high LV filling orpulmonary arterial pressures, and echocardio-graphic signs of congestive heart failure (e.g.,pleural and pericardial effusion, and ascites) indogs with MVD (see below) are all indirect signs ofsevere MR. Nevertheless, asymptomatic dogs withor even without left heart enlargement (Fig. III eData available in Supplemental Data on-line) mayalso show moderate to severe MR, thus indicatingthat an assessment of MR severity is relevant atthis stage. In one study devoted to MR quantifica-tion in canine MVD using the PISA method,10 dogswithout clinical signs exhibited a wide range of RF,and approximately one-third of them showedmoderate to severe RF. The dispersion of RF valuesat the asymptomatic stage is also well-known inhuman patients with MR, and the PISA method isnow systematically recommended at this stage forthe quantitative grading of mitral incompetenceand identification of high-risk patients that mightbenefit from early medical or surgical treat-ment.38,43 Similarly, in one study performed on 72dogs with asymptomatic MVD, a wide range of RFwas found (4e69%), and NT-proBNP in dogs fromISACHC class 1a with mild RF (i.e., RF �30%) wasnot significantly different from that of healthycontrol dogs. Conversely, dogs from ISACHC class1a with moderate to severe MR (i.e., RF >30%) had

Figure 4 Two-dimensional echocardiograms obtained from(right parasternal transaortic views at the level of the aorticIn this ISACHC class 1a dog, the left atrial diameter is stillreference ranges ([0.52e1.13]).45 4B: Conversely, in this ISdilated with a left atrium/aorta ratio of 2.86. Ao: aorta. Aoutflow tract.

significantly higher NT-proBNP values than healthydogs and dogs from ISACHC class 1a with mild RF.Moreover, RF was also higher in dogs that under-went decompensation (cardiac death or develop-ment of congestive heart failure) one year afterthe initial MVD diagnosis. These results suggestthat MR severity assessed by RF is one of the majordeterminants of MVD severity and reflects thelatter well, even at early stages of the disease.Serial echo-Doppler examinations for the identifi-cation of progressive MR worsening could thereforebe recommended to help predict the risk ofdecompensation in dogs with asymptomatic MVD.

Left heart remodeling, myocardialalteration, and hemodynamic changes

Left atrial overload

Chronic and hemodynamically significant MRresults in volume overload, which is first charac-terized by LA enlargement as assessed by the LA/Ao ratio (Fig. 4).44,45 An abnormal curved form ofthe interatrial septum, increased diameter of thepulmonary veins and the presence of atrialarrhythmias are other indirect signs of elevated LApressure (Fig. V e Data available in SupplementalData on-line). The high prognostic value of the

2 dogs with mitral valve disease of different severitiesvalve at end-diastole as described and validated44). 4A:normal with a left atrium/aorta ratio (¼1) within theACHC class 3 dog, the left atrial diameter is markedlyur: left auricle. LA: left atrium. RVOT: right ventricular


degree of LA dilation has been demonstrated inboth symptomatic and asymptomatic dogs withMVD.14,15,34 In one study involving a large pop-ulation of dogs with MVD (n ¼ 558), LA/Ao >1.7was the only variable significantly associated withsurvival time when cardiac-related death wasconsidered.15

Left atrial rupture related to endocardial split-ting of its caudal wall is an infrequent complica-tion of canine MVD, which is characterized by thepresence of pericardial effusion with a laminatedblood clot in the pericardial space (Fig. 5).46e49

The etiology of spontaneous LA endocardial split-ting includes increased LA pressure and mechan-ical trauma to the atrial endocardium due tosevere MR.50

Acquired atrial septal defect secondary to atrialseptal rupture is another uncommon complicationof severe canine MVD (Fig. VI e Data available inSupplemental Data on-line). Such lesions haverarely been described using echocardiography.50

Atrial septal ruptures usually occur in a thin areaof the atrial septum, i.e., close to the fossa ovalis.The degree of shunting depends mainly on theatrial septal defect size and the transatrial pres-sure gradient.50 Atrial septal rupture providesa path of lower resistance for the dilated LA, whichmay be a hemodynamic advantage.50 However, inthe case of considerable left-to-right shunting it

Figure 5 Left atrial rupture in a Yorkshire Terriersuffering from severe mitral valve disease and referredfor recurrent syncope (right parasternal 4-chamberview). Note the irregular and thickened mitral valveleaflet (ml), the presence of pericardial effusion (PE)and the blot clot appearing like laminar echoes withinthe pericardial space (arrow). LA: left atrium. LV: leftventricle. RA: right atrium. RV: right ventricle.

may also contribute to right-sided volume overloadas well as elevated systemic venous pressure andright-sided congestive heart failure.50

Left ventricular remodeling

As MR worsens over time, the volume overloadcreated by MR results in an increase in end-diastolic LV dimensions.24 This eccentric LVhypertrophy, reflecting a marked increased pre-load, may in turn worsen MR, by annular dilationand papillary muscle malalignment.24

Diastolic LV volumes and diameters can beassessed by M-mode and 2D echocardiography(Figs. VII and VIII e Data available in SupplementalData on-line; Fig. 6). A recent study, examining thechanges in LA and LV dimensions before and atonset of congestive heart failure in Cavalier KingCharles Spaniels, showed that the left heartchambers rapidly increased in size only during thelast year before the onset of congestive heartfailure, thus suggesting that the rate of increase inheart dimensions may be a useful indicator ofimpending decompensation.16

The progression of MVD is also associated withan increased sphericity of the LV: a study per-formed on 77 dogs with MVD demonstrateda significant decrease in the LV sphericity index(defined as the ratio of LV end-diastolic length to

Figure 6 Left ventricular M-mode echocardiogramshowing myocardial dysfunction in a Cavalier KingCharles Spaniel with end-stage mitral valve disease(regurgitation fraction of 79% and high systolic pulmo-nary arterial pressure of estimated 84 mmHg). Althoughthe left ventricular free wall (LVFW) is hypokinetic, thefractional shortening is still “normal-high” (44%, refer-ence ranges [30e49%]45) owing to the great amplitude ofthe interventricular septal (IVS) systolic motion and theincreased left ventricular end-diastolic diameter. Sweepspeed is 100 mm/s. LV: left ventricle. RV: right ventricle.


the M-mode LV end-diastolic diameter) withISACHC heart failure class.23

Systolic myocardial dysfunction

Another potential MVD complication is LV systolicdysfunction (Fig. 6).22,23,51,52 However, thedetection of MR-associated systolic dysfunctionremains challenging in both humans and dogs.22e28

In humans, assessment of systolic dysfunctionassociated with MR may rely on a comparison ofpre- and post-operative values of echocardio-graphic variables in patients undergoing mitralvalve surgery.24e28,38 Unfortunately, such studiesare currently difficult to undertake in veterinarymedicine, as surgical repair or replacement of thenative mitral valve are very rare treatment optionsfor canine MVD.

Assessment of the ejection fraction (EF%)and the fractional shortening (FS%)

Ejection fraction (%) and FS% are the 2 indicesmost commonly used to assess systolic myocardialfunction in the dog by conventional echocardiog-raphy.22 As described in Fig. VIII, EF% representsthe percent of blood volume ejected from the LVduring systole. It is therefore defined by thepercent of change in LV volumes between thediastolic and systolic phases (EF% ¼ (EDV � ESV)*100/EDV, where EDV and ESV are the LV systolicand diastolic volumes), and a low EF% value isconsistent with decreased systolic function.21e23

As shown in Fig. VII, FS% (which corresponds toa 1-dimensional assessment of myocardial systolicfunction) is defined by the percent change inradial LV diameters between the diastolic andsystolic phases (FS% ¼ (LVD � LVS)*100/LVD,where LVD and LVS are the LV systolic and dia-stolic diameters, usually assessed by M-modeechocardiography), and again a low FS% value isconsistent with decreased contractility.21,22 Ina prospective study performed by our group in 101small-breed dogs (body weight <15 kg), EF%values assessed in normal dogs, using the mono-plane Simpson’s derived method of discs (seeFig. VIII for explanation), were 67 � 6% [55e75%],and the normal values for FS% were 38.8 � 4.8%[30e49%].23

Limitation of EF% and FS% in the canine MVDsetting

One important limitation of both FS% and EF% isthat these indices depend on several factors

other than intrinsic myocardial contractility,such as preload and afterload. Evolution ofcanine MVD is characterized by an increasedpreload due to MR (see above). In addition,mitral valve lesions create a new pathologicpathway of low resistance for ejection of bloodfrom the LV.24 This “easy” retrograde bloodejection into the LA begins early, before aorticvalve opening, as soon as the LV pressure beginsto rise in early systole (without any isovolumiccontraction period), thus reducing the peaksystolic LV wall stress (i.e., afterload).22,24

Therefore, canine MVD progression is mainlycharacterized by a LV hyperdynamic state withelevated FS% and EF% due to combined volumeoverload (i.e., LV diastolic dilation), decreasedafterload and increased sympathetic tone(Fig. VII e Data available in Supplemental Dataon-line). Ejection fraction and FS% are thusrarely decreased, even at advanced MVDstages.21e23 In a study performed on 77 dogs withMVD, a significant increase in FS% and EF%(assessed by 3 different methods) was foundfrom ISACHC class 1 to ISACHC class 3. These 2indices were also significantly positively corre-lated with RF in heart failure dogs (ISACHCclasses 2 and 3).23 Such elevated values of FS%and EF% are usually associated with exaggeratedmotions of the interventricular septum and theLV free wall (Fig. VII). Unlike normal dogs, theinterventricular septum may exhibit a greatersystolic excursion than the LV free wall, probablyreflecting changes in LV geometry with deviationof the interventricular septum to the right,associated or not with regional dysfunction ofthe LV free wall (Fig. 6).21,22

The load dependency of EF% and FS% explainswhy, in the case of MVD, impaired myocardialfunction may be associated with normal values ofthese indices. Therefore, most authors agree thatnormal values for FS% or EF% in dogs with severeMR suggest systolic myocardial failure.21,22

Although rarely present, hypokinetic myocardialwalls may also help in identifying severelydecreased contractile performance (Fig. 6). Lastly,progressive (instead of sudden) systolic closure ofthe aortic valve may be observed using M-modeviews at the level of the aortic valve, reflectinga severe decrease in cardiac output due to high RFand/or myocardial systolic dysfunction. Therefore,and to summarize, because of MVD-associated loadchanges, and except for severe myocardial alter-ations, the identification of MR-associatedmyocardial failure remains difficult to evidenceusing traditional indices.


End-systolic LV dimensions

End-systolic LV dimensions, including end-systolicdiameter, end-systolic volume, and end-systolicvolume indexed to body surface area (ESVI), areother conventional echocardiographic variables,which may be used to identify systolic myocardialdysfunction in dogs with MVD.22,23,53 Increasedend-systolic dimensions, despite enhanced systolicLV ejection into the low-pressure LA chamber,suggest impaired systolic function. The ESVI indexhas been shown to be a relatively load-independent predictor of post-operative systolicdysfunction in humans with MR and a relativelyindependent variable of experimental increasedpreload in the dog.54e56 Three ultrasound methodsmay be used to evaluate ESVI: the Teichholzmethod, also called the geometric method, basedon M-mode linear measurement of the LV systolicdiameter,23,53 and 2 planimetric methods includingthe monoplane Simpson’s derived method of discsand the length-area method (see Fig. VIII forexplanation). Unlike planimetric methods, theTeichholz method does not involve directmeasurement of longitudinal LV dimension. Ourgroup has demonstrated that MVD progression isassociated with changes in LV shape (increasedsphericity),23 which was recently confirmed by 3Dechocardiography.57 Because the Teichholzformula does not take into account this progressivechange in LV shape during the time course of thedisease, the geometric method is inaccurate forassessing LV volumes in dogs with MVD. This inad-equacy of the Teichholz method, as compared tothe 2 other “anatomy-based” methods, wasconfirmed by the BlandeAltman results, whichshowed that the Teichholz method overestimatesESVI in a non-linear way.23 This bias is of majorpractical importance because it means that theTeichholz method leads to an overestimation of LVsize (and therefore an overdiagnosis of systolicdysfunction) in canine MVD that becomes moremarked with disease progression. Similar resultswere obtained by Tidholm et al. who showed thatthe Teichholz method overestimates LV volumes indogs by a factor of approximately 2 in comparisonwith real-time 3-dimensional echocardiography,whereas a good agreement exists between thelatter and the Simpson’s method.58 The Teichholzmethod should therefore not be recommended toassess ESVI in dogs with MVD. When planimetricmethods are used, ESVI has been found to besignificantly increased in dogs with advanced MVD,with a negative prognostic value on survival at 5months.23 As breed reference intervals are lacking

for ESVI assessed by planimetric methods and asthe repeatability and reproducibility of thesevariables have been shown to be good for a trainedobserver (coefficients of variation <11%),23

repeated measurements of this variable overtime (along with EF% value) may be recommendedto help identify ongoing myocardial dysfunction ina given MVD dog.

Other variables

The LV Tei index, assessed by pulsed-wave Dopplerexamination of transmitral and aortic flows, isdefined as the sum of the isovolumic contractionand relaxation times divided by the ejectiontime.59 Although rarely used in practice, the LV Teiindex (or index of myocardial performance) hasbeen shown to be significantly correlated with theLV peak þdp/dt in healthy dogs. It has also beenshown to increase with the progression of clinicalsigns in dogs with MVD due to a shortening of theejection time.59 However, in moderate to severeMR, isovolumic contraction and relaxation timescan be extremely short or non-existent due to thepresence of an opening in the mitral valve area.Therefore, in that case, the TEI index seems to beinaccurate for assessing myocardial function.

Information provided by spectral Dopplerrecordings of systolic MR and diastolictransmitral flow

Continuous-wave Doppler trace of systolic MRMitral regurgitation flow profile, assessed bycontinuous-wave Doppler mode, reflects LA pres-sure, systolic LV function, preload as well assystemic arterial pressure. The peak velocity ofa holosystolic MR is usually between 5 and 6 m/s(corresponding to a maximum systolic LVeLApressure gradient of 100 mmHg or slightly higher,according to the Bernoulli equation).21,22 SystolicLV impairment and high LA pressure (as well assystemic arterial hypotension) may result ina decreased peak MR velocity (Figs. IX and X eData available in Supplemental Data on-line;Fig. 7). Other changes include pointed and/orasymmetric flow profiles (Fig. IX e Data availablein Supplemental Data on-line) and the presenceof a cut-off sign in mid to late systole (Fig. 7),reflecting a decrease in MR flow due to high LApressure. Lastly, some authors propose a methodto calculate þdp/dt (which is decreased in case ofsystolic dysfunction) using MR flow profiles (Fig. Xe Data available in Supplemental Data on-line).22

Figure 7 Continuous-wave Doppler trace of a mitralvalve regurgitation jet obtained in a dog with severemitral valve disease showing a low peak velocity (4 m/s),an asymmetric flow profile and a cut-off sign (arrows)indicative of high atrial pressure. LA: left atrium.


Pulsed-wave Doppler trace of transmitral flowThe normal transmitral flow profile, assessed bypulsed-wave Doppler mode and recorded at themitral leaflet tips, is characterized by an early Ewave and a late A diastolic wave related to atrialcontraction, with an E/A ratio >1 and decreasingwith age.60 In one study performed on 100 healthydogs, values for E and A were 0.87 � 0.13 m/s[0.58e1.17] and 0.61 � 0.12 m/s [0.39e0.86],respectively.45 The transmitral flow profilereflects diastolic function (relaxation andcompliance, volume and recoil) as well as LV fillingpressure.61e65 A high velocity E wave (>1.5 m/s)suggests elevated LA pressure (Fig. XI e Dataavailable in Supplemental Data on-line).22 A highvelocity E wave with a shorter E deceleration time(<80 ms in dogs older than 10e12 years) issuggestive of the association of high LA pressureand a non-compliant LV.22 Conversely an E/A ratio<1 and/or a prolonged E deceleration time indi-cates impaired relaxation (Fig. 8A).22,60 Severalstudies demonstrated that high velocity E wave orhigh E/A ratio are associated with higher risk ofdeath or decompensation in dogs with MVD.14,15,40

As elevated LA pressure and impaired relaxationhave opposite effects on the E value, a “pseudo-normal” filling pattern with a normal E/A ratiomay be observed when the 2 abnormalities arecombined (Fig. 8B and C). In other words, anabnormal relaxation may mask high LA pressure.

As myocardial and annular velocities are lessload-dependent than mitral E wave, the E/E0

ratio has been proposed to “correct” this effectand predict filling pressure (with E0 defined asthe early longitudinal mitral annular velocity). E0

decreases in the case of impaired relaxation andis usually assessed using the pulsed-wave TDImode (Fig. XII e Data available in SupplementalData on-line). An E/E0 ratio >9 has been shownto indicate a 95% probability for LA pressure tobe >20 mmHg in a canine model of acute MR,and a ratio >12 along with a high E wave velocitycan predict the presence of congestive heartfailure on thoracic radiographs in spontaneouslydiseased dogs.22,65 Similarly, in another study E/E0 in dogs with MVD and congestive heart failurewas significantly higher than in those withcompensated MVD, and an E/E0 cut-off value of13 could be used to identify congestive heartfailure with a sensitivity and a specificity of 80%and 83%, respectively.66 However, E0 is nottotally load-independent.67 As volume overloadmarkedly increases with severe MR, the E/E0

ratio becomes of limited relevance (except inthe case of impaired relaxation and subsequentreduced E0).

Other conventional Doppler variablesAssessment of the pulmonary venous flow profile isof limited value in dogs with MVD, as MR may enterthe pulmonary veins and contribute to noisyprofiles of poor quality. Conversely, calculation ofthe isovolumic relaxation time (IVRT, defined asthe time interval from the Doppler signal of aorticvalve closure to the beginning of the mitral Ewave) and assessment of peak E/IVRT may be ofinterest as these variables have been shownexperimentally to be correlated with LA pressureand LV end-diastolic pressure in the dog.61,62 Adecrease in IVRT is suggestive of increased fillingpressure. According to Schober et al.,63 an E/IVRTratio >2.5 and an IVRT <45 ms in dogs withmoderate to severe chronic MR are usually asso-ciated with the presence of congestive heartfailure on thoracic radiographs.

Pulmonary arterial hypertension

Pulmonary arterial hypertension is defined as highdiastolic or systolic pulmonary arterial pressure,a condition that may lead to right ventricular andatrial enlargement, followed by right-sided heartfailure.17e19,68 Systolic and diastolic pulmonaryarterial pressure can be evaluated non-invasively

Figure 8 Pulsed-wave Doppler traces of transmitral flow in 2 dogs with mitral valve disease and diastolicdysfunction. 8A: In this dog (with mild mitral regurgitation), the transmitral flow profile shows an alteration of therelaxation phase with an E/A ratio less than 1.63 8B and 8C: This dog has severe mitral regurgitation associated withdiastolic dysfunction, thus explaining that the transmitral flow profile is normal (“pseudonormal” filling pattern,Fig. 8B). However tissue Doppler imaging (here two-dimensional color mode) of the left ventricular free wall confirmsa marked diastolic dysfunction for both radial (here shown) and longitudinal motion with an E0/A0 ratio lower than 1.Fig. 8C displays a simultaneous recording of myocardial velocities in a sub-endocardial (yellow) and sub-epicardial(green) segment of the left ventricular free wall. S0, E0, and A0 are the peak myocardial velocities during systole,early diastole, and late diastole, respectively. AVC: aortic valve closure and AVO: aortic valve opening (assessed fromthe pulsed-wave Doppler trace of the systolic aortic flow).


by Doppler assessment of tricuspid valve regurgi-tation or pulmonic valve regurgitation, respec-tively (see examples in Fig. XIII e Data available inSupplemental Data on-line).8,16,18 In dogs withMVD, both severity and prevalence of Doppler-derived evidence of PAH have been shown toincrease significantly with heart failure class(prevalence of 27% and 72% for ISACHC class II andIII, respectively).18 However, PAH may also bepresent at early stages of the disease even in dogsbelonging to ISACHC classes 1a and 1b if MVD isassociated with moderate to severe MR.18 Thisagain illustrates the potential interest of per-forming echocardiographic examinations on dogswith MVD but no clinical signs. Systolic pulmonaryarterial pressure has also been shown to bea prognostic factor for survival or disease

worsening in dogs with symptomatic MVD and alsoa prognostic factor for decompensation in dogswith asymptomatic MVD.14,23

One limitation of the method used to assess PAHis that it relies on Doppler measurement ofpulmonic and tricuspid regurgitations, which maybe absent or insufficient for determining pulmo-nary arterial pressure. In that case, “surrogate”variables may be used to compensate this draw-back and predict the presence of PAH, includingthe main pulmonary arterial diameter to Aodiameter ratio (MPA/Ao), the pulmonary flowacceleration time (AT), the acceleration to ejec-tion time ratio (AT/ET), and lastly the Tei index ofright ventricular function. A significant correlationbetween systolic pulmonary arterial pressure andall these conventional echo-Doppler variables


(positive correlation for MPA/Ao and the Tei index,and negative for the others), has been demon-strated in the dog.68 In the same study, systolic aswell as diastolic right myocardial TDI alterationswere demonstrated to be associated with PAH,even for mild increases in pulmonary arterialpressure (Fig. XIV e Data available inSupplemental Data on-line).68 Thus any alterationof right myocardial function using TDI may raisethe suspicion of systolic PAH in dogs with MVD andno measurable tricuspid regurgitation, althoughfactors other than pulmonary arterial pressure caninfluence right TDI variables (e.g., primary rightsystolic dysfunction).

Recent echocardiographic techniques(tissue Doppler imaging, strain and strainrate imaging, speckle trackingechocardiography)

Tissue Doppler imaging

Tissue Doppler imaging is a relatively recentlydeveloped ultrasound technique which allows the

Figure 9 Longitudinal myocardial velocity profiles of the lemoderate mitral valve disease using two-dimensional color tSimultaneous recordings of myocardial velocities are obtaineyellow and green curves respectively). A typical severe lonsegments by an inverted E/A ratio. A positive post-systolic conot shown), occurring after aortic valve closure and greater tcolor display of velocity is superimposed on the two-dimemyocardial velocities during systole, early diastole, and late

quantification of both global and regional myocar-dial function from measurements of myocardialvelocities in real time (see reference 29 forreview). Annular velocities can also be quantifiedusing the TDI technique. The pulsed-wave TDI modeprovides information on tissue movements froma single sample volume. As shown in Fig. XII (Dataavailable in Supplemental Data on-line), thepulsed-wave TDI mode is currently used mainly toassess the early diastolic E0 wave at the mitral valveannulus, in order to calculate the mitral E wave/E0

ratio.22,63e67 The 2D color TDI mode is better thanthe pulsed-wave TDI mode for the evaluation ofmyocardial function because of its ability tosimultaneously quantify velocities in severalsegments within a single myocardial wall or withindifferent myocardial walls in order to calculatevelocity gradients (Figs. 9 and 10) and to assessintra- or interventricular synchrony, respectively(Fig. 11). The TDI technique has been shown to bemore sensitive than conventional echocardiographyin detecting systolic and diastolic myocardialalterations in humans as well as in experimental orspontaneous heart diseases in animals, as demon-strated by our group in a dog model of dilatedcardiomyopathy.29,69 One important TDI application

ft ventricular free wall (LVFW) obtained from a dog withissue Doppler imaging (left parasternal 4-chamber view).d in 2 segments of the LVFW (basal and apical segments,gitudinal dysfunction is observed, characterized in bothntraction wave (PSC), confirmed by Strain Imaging (datahan the S wave, is also present for the apical motion. Thensional view (left upper panel). S, E, and A are peakdiastole, respectively. LA: left atrium. LV: left ventricle.

Figure 10 Radial velocity profiles recorded within 2 segments of the left ventricular free wall using the two-dimensional color tissue Doppler imaging (TDI) mode in 2 Cavalier King Charles Spaniels with severe mitral valvedisease but “opposite” systolic myocardial performance (right parasternal transventricular short-axis views). These 2simultaneous recordings of myocardial velocities in a sub-endocardial (yellow) and sub-epicardial (green) segmentboth indicate that the sub-endocardium is moving more rapidly than the sub-epicardium in systole and also in diastole,thus defining systolic and diastolic myocardial velocity gradients (white double arrows). However, for the dog inFig. 10B, the sub-endocardial and sub-epicardial velocity profiles are nearly superimposed in systole, thus indicatinga very low systolic myocardial velocity gradient (0.9 cm/s versus 3.9 cm/s in Fig. 10A; for comparison values obtainedin a population of 100 healthy dogs were 2.5 � 0.8 cm/s).45 Systolic dysfunction of the dog from Fig. 10B was not“overt” using conventional echocardiography (fractional shortening of 35%, i.e., within normal ranges). The dog inFig. 10A is characterized by a radial hyperkinetic state, which was confirmed using Strain and Strain Rate imaging (seeFigs. 12 and 13). AVC: aortic valve closure and AVO: aortic valve opening (assessed from the pulsed-wave Dopplertrace of the systolic aortic flow). See Fig. 9 for remainder of key.


Figure 11 Example of interventricular dyssynchrony assessed by two-dimensional color tissue Doppler imaging fromthe left parasternal 4-chamber view in a dog with mitral valve disease. The longitudinal velocity profiles obtainedfrom 3 basal segments of the left ventricular free wall (LVFW, red), the interventricular septum (green), and the rightmyocardial wall (yellow) show a delayed peak systolic LVFW velocity compared with the 2 others (white arrows),corresponding to the presence of a post-systolic contraction wave. Note that the peak systolic velocity of the leftventricular free wall between AVO and AVC (red stars) is low, suggesting longitudinal systolic alteration, which wasconfirmed by Strain imaging (data not shown). AVC: aortic valve closure and AVO: aortic valve opening (assessed fromthe pulsed-wave Doppler trace of the systolic aortic flow). RA: right atrium. RV: right ventricle. See Fig. 9 forremainder of key.


is the assessment of diastolic function, which maybe impaired in aged dogs with MVD (Figs. 8 and 9).Such diastolic alterations are usually more commonand more pronounced for the longitudinal than forthe radial motion (personal observations).However, in one recent study, no decrease in radialor longitudinal early diastolic velocities or in TDI E/A ratio (suggesting an impaired diastolic function)was observed in dogs with MVD compared tohealthy controls.70 Conversely several diastolic TDIvariables (i.e., early TDI diastolic wave (E) and TDIE/A ratio in different segments) were significantlyincreased in dogs with congestive heart failurecompared with those without any sign of decom-pensation and normal controls. However, theformer had a significantly higher heart rate thanthe latter, which may have contributed to thesediastolic findings to an unknown extent. In addi-tion, although TDI variables are less load-dependent than conventional echocardiographicindices, a preload influence is highly likely, asa significant correlation has been found betweenthe LA/Ao ratio and early diastolic velocities.70

Recent data from human studies have confirmedthat both systolic and diastolic TDI velocities of theLV are preload-dependent.71

Alteration of myocardial synchrony may also beobserved in dogs with MVD, using left parasternallong-axis views (Fig. 11). Tidholm et al. demon-strated that the difference in longitudinal systolicvelocity time-to-peak between the LV and theinterventricular septum was significantly increasedin dogs with MVD, with and without congestiveheart failure, as compared to healthy control dogs,indicating intraventricular dyssynchrony.70

Whether such alteration is primarily related toMVD or represents an age-related finding remainshowever to be determined.

Lastly, as shown in Fig. 10, the TDI techniquemay also be used to help identify systolic changes,revealing increased systolic velocities and gradi-ents in the case of hyperkinesia and the oppositepattern in the case of hypokinesia. Systolic trans-mural gradients should be preferred because theyare more reliable and sensitive indicators ofregional LV dysfunction than isolated systolicvelocity values.29,69 In the study by Tidholm et al.,no decrease in systolic performance (but insteadan increase in several systolic velocities) was evi-denced in dogs with MVD using the TDI tech-nique.70 Again a load effect and an influence ofheart rate may partially explain these results.

Figure 12 Example of radial hyperkinesia confirmed by regional radial strain recorded within the left ventricularfree wall from a dog with moderate mitral valve disease (same dog as in Fig. 10A, right parasternal transventricularshort-axis view). The radial strain profile (expressed in %) is positive and maximal in end-systole (StS), and thendecreases during diastole, thus confirming regional systolic expansion (i.e., thickening) and diastolic compression(i.e., shortening), respectively. The StS value is high (92%) compared to values obtained in a population of healthydogs (62.9 � 10.4%).72 The color display of strain is superimposed on the right parasternal transventricular short-axisview (left upper panel). Strain length ¼ 10 mm. Region of interest size ¼ 5/3 mm. AVC: aortic valve closure and AVO:aortic valve opening (assessed from the pulsed-wave Doppler trace of the systolic aortic flow). LV: left ventricle.


Strain and strain rate imaging

Strain and Strain Rate imaging are 2 quantitativeTDI-derived imaging techniques, which can be usedto measure myocardial segmental deformation(contraction or stretching) and rate of deformation,respectively.29,72 Myocardial strain represents thedeformation of a myocardial segment over time,and is expressed as the % change from its originaldimension. Myocardial strain rate (expressed in s�1)is the temporal derivative of strain, and thereforedescribes the rate of myocardial deformation.Unlike TDI, Strain and Strain Rate imaging providetrue measures of local myocardial deformation,thereby separating active from passive myocardialmotion and offering a more “representative” ordirect evaluation of intrinsic myocardial function.Figs. 12 and 13 show examples of increased systolicfunction assessed by strain and strain rate imagingin a dog with MVD. Very few data are availableregarding strain and strain rate alterations associ-ated with canine MVD.70 This is why preliminaryresults are presented here of an ongoing prospec-tive study being performed at our unit, which isdesigned to assess radial and longitudinal systolic

myocardial function in canine MVD using severalultrasound techniques including Strain imaging. Forthis purpose 74 dogs from different breeds withMVD were prospectively recruited between 2006and 2011. Dogs receiving an inotropic drug (pimo-bendan) were excluded from the study. As ex-pected, this MVD Group was mostly composed ofmales (n ¼ 53, 72%), aged adult small-breed dogs(10.1 � 1.7 years, range: 6.5e12.7 years), with anoverrepresentation (49/74, 66%) of English ToySpaniels (n ¼ 24), Yorkshire Terriers (n ¼ 9),Bichons (n ¼ 8) and Poodles (n ¼ 8). Forty-seven ofthe 74 MVD dogs (64%) received one or moretreatments for heart failure at the time of diag-nosis. Treatments included angiotensin convertingenzyme inhibitors such as benazepril, enalapril,imidapril, and ramipril (44 dogs [59%]), furosemide(20 dogs [27%]), and spironolactone (19 dogs [26%]).Additionally, 17 healthy small-breed dogs matchedfor body weight (9.6 � 4.8 kg, range: 2.0e18.5 kg)and age (9.0 � 1.5 years, range: 7.0e11.5 years)were recruited during the same period. Strainimaging examinations were performed by the sametrained observer (VC), as previously described andvalidated by our group, using an ultrasound unit

Figure 13 Example of radial hyperkinesia confirmed by regional radial strain rate recorded within the leftventricular free wall from a dog with moderate mitral valve disease (same dog as in Figs. 10A and 12, right parasternaltransventricular short-axis view). The strain rate profile (expressed in s�1) is positive during systole (SRS), indicatingregional thickening, and then features 2 negative diastolic peaks during early filling and atrial contraction (SRE andSRA) corresponding to a biphasic thinning phase. The SRS value is high (9.2 s�1) compared to values obtained ina population of healthy dogs (5.8 � 1.1%).72 The color display of strain rate is superimposed on the right parasternaltransventricular short-axis view (left upper panel). Strain length ¼ 8 mm. Region of interest size ¼ 5/3 mm. SeeFig. 12 for remainder of key.

Figure 14 Box plots representing minimal values, 25th, 50th, 75th percentiles, and maximal values of radial (14A)and longitudinal strain (14B) assessed in healthy dogs (n ¼ 17) and in 74 age- and body-weight-matched dogs withmitral valve disease from different heart failure classes (ISACHC class 1, 2 and 3; n ¼ 37, n ¼ 22 and n ¼ 15respectively). Compared to healthy dogs (control group), diseased dogs (MVD group) are characterized by a wide rangeof radial and longitudinal strain values, particularly in ISACHC class 2. Additionally, a significant decrease in absolutevalue of longitudinal systolic strain is observed in dogs from ISACHC class 3 (16.1 � 6.7%) as compared with healthydogs (22.9 � 5.1%) and dogs from the ISACHC class 1 (24.6 � 5.1%) and 2 (21.9 � 6.0%). *, P < 0.01 versus controlgroup; y, P < 0.001 versus ISACHC class 1; z, P < 0.001 versus ISACHC class 2.


Figure 15 Longitudinal strain profiles recorded at the base of the left ventricular free wall using Strain imaging in 2King Charles Spaniels with mitral valve disease and different longitudinal myocardial function (from the left para-sternal 4-chamber view). For both dogs, the longitudinal strain profiles are negative, confirming a regionalcompression (i.e., myocardial shortening) during systole and then a regional diastolic expansion. The dog fromFig. 15A shows a normal longitudinal strain profile, maximal at the time of AVC (StS) with a normal StS value (�23%; forcomparison, values obtained in a population of healthy dogs were �25.5 � 3.6%).72 Conversely, the dog from Fig. 15Bshows an abnormal longitudinal strain profile characterized by a very low absolute value of peak strain (StS) betweenAVO and AVC (¼10%). The maximal longitudinal deformation occurs after AVC indicating the presence of a post-systolic contraction wave (PSC). Moreover, the systolic shortening is preceded by an early systolic lengthening(yellow stars). The color display of strain is superimposed on the left parasternal 4-chamber view (left upper panel ofFig. 15A). Strain length ¼ 12 mm. Region of interest size ¼ 5/3 mm. AVC: aortic valve closure and AVO: aortic valveopening (assessed from the pulsed-wave Doppler trace of the systolic aortic flow). LA: left atrium. LV: left ventricle.RA: right atrium. RV: right ventricle.


Figure 16 Example of abnormal longitudinal strain recorded in 6 myocardial segments of the left ventricle (LV)using Speckle Tracking echocardiography in a dog with mitral valve disease (left parasternal long-axis view). Thesoftware algorithm has automatically defined 6 equidistant myocardial segments within the interventricular septumand the LV free wall. The right upper panel shows the 6 corresponding LV longitudinal strain versus time curves. Fiveout of the 6 LV segments undergo a systolic myocardial shortening during systole (negative strain, maximal before orat end-systole (ES), which is normal). However, the values of peak systolic strain are heterogeneous, as also shown inthe left lower panel (from �14% to �38%). In addition, one segment (at the base of the LV free wall, red curve)undergoes an abnormal positive (instead of a negative) longitudinal systolic strain (þ12%). This dyskinesia is alsoapparent on the two-dimensional (left upper and lower panels) and M-curves (right lower panel) color-coded viewsshowing the positive systolic strain in blue (instead of red). The M-curves color-coded view also shows the hetero-geneous values of strain of the 5 other segments, appearing as red gradation. LA: left atrium.


equipped with 2.2e3.5 and 5.5e7.5 MHz phased-array transducers and a specific software.72,d,e

Radial systolic LV strain was measured betweenthe 2 papillary muscles by using the right para-sternal ventricular short-axis view and longitudinalsystolic LV strain was measured at the base usingthe left apical 4-chamber view. As shown in Fig. 14,no significant difference in radial strain wasobserved between dogs with MVD and healthycontrols. Conversely, a significant decrease in thelongitudinal absolute value of systolic strain wasidentified at the latest heart failure stage (ISACHCclass 3) compared with healthy controls and dogs

d Vivid 7, General Electric Medical System, Waukesha, Wisc,USA.e Echopac Dimension, General Electric Medical System, Wau-

kesha, Wisc, USA.

from other ISACHC classes (Figs. 14 and 15). Theseresults are in accordance with those described inhumans with non-ischemic MR showing that defor-mation imaging techniques are more sensitive thanconventional ones to detect systolic alteration.26,51

Interestingly, our study also shows a marked vari-ability of strain values in the MVD group, andparticularly in heart failure dogs with a wide over-lap of values, suggesting that systolic changes(including both hyper- and hypokinesia) associatedwith canine MVD are highly dependent on theindividual and cannot be predicted solely from theISACHC failure class.

Speckle tracking echocardiography

Two-dimensional STE is the most recently devel-oped ultrasound technique and allows quantitative

Figure 17 Bull’s-eye map of peak longitudinal systolicstrain values recorded in the left ventricle using SpeckleTracking echocardiography from a dog with mitral valvedisease. This bull’s-eye representation of strain datawas obtained from 3 different left apical views (leftparasternal 4-, 5- and 2-chamber views). Similarly toFig. 16, 6 strain profiles were obtained from the 3 views.Segmental peak strain values in each of the 18 segmentswere then used to generate this parametric (“bull’seye”) display of the entire left ventricle (with the apexlocated in the center). Different color codes are used:red and pink colors correspond to normal and reducedstrain values, respectively, and the blue color indicateslongitudinal dyskinesia (positive instead of negativestrain). Segmental hypokinesia and dyskinesia in the leftventricular lateral wall (arrow) is clearly demonstratedin this dog (the bull’s-eye representation should showa homogeneous red color map without any blue color).


assessment of regional myocardial function. Thisnon-Doppler technique is based on frame-by-frametracking of the tiny echo-dense speckles observedwithin the myocardium on gray scale 2D echocar-diographic images and subsequent measurementof myocardial motion.73e75 By analyzing specklemotion, STE offers the opportunity to assessmyocardial tissue velocity, strain and strain rate,as well as the LV rotation independently of cardiactranslation and beam angle, unlike the Doppler-based techniques (TDI or TDI-derived tech-niques). The STE technique may be used to assesssystolic strain and regional synchrony in canineMVD. An example of longitudinal systolic strainalteration, identified by STE in a dog with MVD, ispresented in Fig. 16. Longitudinal systolic strainmay also be assessed by STE for the whole LV in 18different segments using 3 different standardapical planes (Fig. 17). The obtained bull’s-eyerepresentation of segmental systolic strain has

been found to be closely correlated with coronaryanatomy in humans.75

Conclusion

In conclusion, conventional echo-Doppler examina-tion provides an accurate diagnosis of canine MVDand of its consequences on cardiac remodeling, LVfunction and cardiovascular pressures. Serial echo-Doppler examinations can therefore be recom-mended in dogs with MVD in order to identify andtrack these alterations over time and detect anongoing worsening of the disease. Additionally, inasymptomatic MVD dogs, several standard echo-Doppler variables have been shown to be relatedto clinical outcome and may help in predictinganimals at risk for decompensation. As performed inour lab, serial measurements of these combinedvariables over time can also therefore be recom-mended in asymptomatic dogs (at least once a year,according to the case), as they can allow an earlydetection of dogs at risk for rapid progression intocongestive heart failure (which could representa rationale for a closer monitoring and early treat-ment prescription). These measurements mainlyinclude grading of MR severity, evaluation ofpulmonary arterial pressures, assessment of fillingpressures and left heart changes (LA/Ao, FS%, LVdiameters, LV volumes and corresponding ESVI andEF% using a planimetric method). Nevertheless,further studies in large canine populations areneeded to determine breed-by-breed referenceintervals for some these parameters (e.g., theplanimetric ones) and to investigate the effect ofheart rate, age, concomitant treatment or bloodvolume status. Further prospective studies using STEare also needed to confirm our results regarding thedeterioration of systolic longitudinal strain at thelatest stage of the disease, to analyze myocardialsynchrony changes and determine the influence ofsuch alterations on clinical outcome and thera-peutic strategy.

Conflict of interest

None.

Supplementary data

Supplementary data associated with thisarticle can be found, in the online version, atdoi:10.1016/j.jvc.2011.11.005.

http://dx.doi.org/doi:10.1016/j.jvc.2011.11.005


References

1. Thrusfield MV, Aiken CGC, Darke PGG. Observations onbreed and sex in relation to canine heart valve incompe-tence. J Small Anim Pract 1985;26:709e717.

2. Kvart C, Haggstrom J. Acquired valvular heart disease. In:Ettinger SJ, Feldman EC, editors. Textbook of veterinaryinternal medicine. 6th ed. Philadelphia: WB Saunders; 2005.p. 1022e1039.

3. Beardow AW, Buchanan JW. Chronic mitral valve disease inCavalier King Charles spaniels: 95 cases (1987e1991). J AmVet Med Assoc 1993;203:1023e1029.

4. Serfass P, Chetboul V, Carlos Sampedrano C, Nicolle A,Benalloul T, Laforge H, Gau C, Hebert C, Pouchelon JL,Tissier R. Retrospective study of 942 small sized-dogs:prevalence of left apical systolic heart murmur and left-sided heart failure, critical effects of breed, and sex. JVet Cardiol 2006;8:11e18.

5. Chetboul V, Tissier R, Villaret F, Nicolle A, Dean E,Benalloul T, Pouchelon JL. Epidemiological, clinical, echo-Doppler characteristics of mitral valve endocardiosis inCavalier King Charles in France: a retrospective study of 451cases (1995 to 2003). Can Vet J 2004;45:1012e1015.

6. Borgarelli M, Zini E, D’Agnolo G, Tarducci A, Santilli RA,Chiavegato D, Tursi M, Prunotto M, Haggstrom J. Compar-ison of primary mitral valve disease in German shepherddogs and in small breeds. J Vet Cardiol 2004;6:27e34.

7. Borgarelli M, Tarducci A, Zanatta R, Haggstrom J. Decreasedsystolic function and inadequate hypertrophy in large andsmall breed dogs with chronic mitral valve insufficiency. JVet Intern Med 2007;21:61e67.

8. Kittleson MD, Brown WA. Regurgitant fraction measured byusing the proximal isovelocity surface area method in dogswith chronic myxomatous mitral valve disease. J Vet InternMed 2003;17:84e88.

9. Muzzi RAL, de Araujo RB, Muzzi LA, Pena JL, Silva EF.Regurgitant jet area by Doppler color flow mapping: quan-titative assessment of mitral regurgitation severity in dogs.J Vet Cardiol 2003;5:33e38.

10. Gouni V, Serres F, Pouchelon JL, Tissier R, Lefebvre HP,Nicolle AP, Carlos Sampedrano C, Chetboul V. Quantificationofmitral valve regurgitation in dogswith degenerativemitralvalve disease by use of the proximal isovelocity surface areamethod. J Am Vet Med Assoc 2007;231:399e406.

11. Kvart C, Haggstrom J, Pedersen HD, Hansson K, Eriksson A,Jarvinen AK, Tidholm A, Bsenko K, Ahlgren E, Ilves M,Ablad B, Falk T, Bjerkfas E, Gundler S, Lord P, Wegeland G,Adolfsson E, Corfitzen J. Efficacy of enalapril for preventionof congestive heart failure in dogs with myxomatous valvedisease and asymptomatic mitral regurgitation. J Vet InternMed 2002;16:80e88.

12. Atkins CE, Keene BW, Brown WA, Coats JR, Crawford MA,DeFrancesco TC, Edwards NJ, Fox PR, Lehmkuhl LB,Luethy MW, Meurs KM, Petrie JP, Pipers FS, Rosenthal SL,Sidley JA, Straus JH. Results of the veterinary enalapril trialto prove reduction in onset of heart failure in dogs chroni-cally treated with enalapril alone for compensated,naturally-occurring mitral valve insufficiency. J Am Vet MedAssoc 2007;231:1061e1069.

13. Pouchelon JL, Jamet N, Gouni V, Tissier R, Serres F, CarlosSampedrano C, Castaignet M, Lefebvre HP, Chetboul V.Effect of benazepril on survival and cardiac events in dogswith asymptomatic mitral valve disease: a retrospectivestudy of 141 cases. J Vet Intern Med 2008;22:905e914.

14. Chetboul V, Serres F, Tissier R, Lefebvre HP, CarlosSampedrano C, Gouni V, Poujol L, Hawa G, Pouchelon JL.

Association of plasma N-terminal pro-B-type natriureticpeptide concentration with mitral regurgitation severityand outcome in dogs with asymptomatic degenerativemitral valve disease. J Vet Intern Med 2009;23:984e994.

15. Borgarelli M, Savarino P, Crosara S, Santilli RA, Chiavegato D,Poggi M, Bellino C, La Rosa G, Zanatta R, Haggstrom J,Tarducci A. Survival characteristics and prognostic variablesof dogswithmitral regurgitation attributable tomyxomatousvalve disease. J Vet Intern Med 2008;22:120e128.

16. Lord P, Hansson K, Kvart C, Haggstrom J. Rate of change ofheart size before congestive heart failure in dogs withmitral regurgitation. J Small Anim Pract 2010;51:210e218.

17. Carlsson C, Haggstrom J, Eriksson A, Jarvinen AK, Kvart C,Lord P. Size and shape of right heart chambers in mitralvalve regurgitation in small-breed dogs. J Vet Intern Med2009;23:1007e1013.

18. Serres FJ, Chetboul V, Tissier R, Carlos Sampedrano C,Gouni V, Nicolle AP, Pouchelon JL. Dopplerechocardiography-derived evidence of pulmonary arterialhypertension in dogs with degenerative mitral valvedisease: 86 cases (2001e2005). J Am Vet Med Assoc 2006;229:1772e1778.

19. Stepien RL. Pulmonary arterial hypertension secondary tochronic left-sided cardiac dysfunction in dogs. J Small AnimPract 2009;50(Suppl. 1):34e43.

20. Chiavegato D, Borgarelli M, D’Agnolo G, Santilli RA.Pulmonary hypertension in dogs with mitral regurgitationattributable to myxomatous valve disease. Vet RadiolUltrasound 2009;50:253e258.

21. Boon JA. Acquired heart disease: mitral insufficiency. In:Boon JA, editor. Manual of veterinary echocardiography.Baltimore: Williams & Wilkins; 1998. p. 261e286.

22. Bonagura JD, Schober KE. Can ventricular function beassessed by echocardiography in chronic canine mitral valvedisease? J Small Anim Pract 2009 Sep;50(Suppl. 1):12e24.

23. Serres F, Chetboul V, Tissier R, Poujol L, Gouni V, CarlosSampedrano C, Pouchelon JL. Comparison of 3 ultrasoundmethods for quantifying left ventricular systolic function:correlation with disease severity and prognostic value indogs with mitral valve disease. J Vet Intern Med 2008;22:566e577.

24. O’Gara P, Sugeng L, Lang R, Sarano M, Hung J, Raman S,Fischer G, Carabello B, Adams D, Vannan M. The role ofimaging in chronic degenerative mitral regurgitation. JACCCardiovasc Imaging 2008;1:221e237.

25. Song JM, Kang SH, Lee EJ, Shin MJ, Lee JW, Chung CH,Kim DH, Kang DH, Song JK. Echocardiographic predictors ofleft ventricular function and clinical outcomes aftersuccessful mitral valve repair: conventional two-dimensional versus speckle-tracking parameters. AnnThorac Surg 2011;91:1816e1823.

26. Marciniak A, Sutherland GR, Marciniak M, Kourliouros A,Bijnens B, Jahangiri M. Prediction of postoperative leftventricular systolic function in patients with chronic mitralregurgitation undergoing valve surgery e the role ofdeformation imaging. Eur J Cardiothorac Surg 2011;40:1131e1137.

27. Lisi M, Ballo P, Cameli M, Gandolfo F, Galderisi M,Chiavarelli M, Henein MY, Mondillo S. Mitral annular longi-tudinal function preservation after mitral valve repair: theMARTE study. Int J Cardiol 2010 Dec 29 [Epub ahead of print].

28. Rosenhek R, Maurer G. Management of valvular mitralregurgitation: the importance of risk stratification. J Car-diol 2010;56:255e261.

29. Chetboul V. Advanced techniques in echocardiography insmall animals. Vet Clin North Am Small Anim Pract 2010;40:529e543.


30. Pedersen HD, Haggstrom J. Mitral valve prolapse in the dog:a model of mitral valve prolapse in man. Cardiovasc Res2000;47:234e243.

31. Terzo E, Di Marcello M, McAllister H, Glazier B, Lo Coco D,Locatelli C, Palermo V, Brambilla PG. Echocardiographicassessment of 537 dogs with mitral valve prolapse andleaflet involvement. Vet Radiol Ultrasound 2009;50:416e422.

32. David TE, Ivanov J, Armstrong S, Christie D, Rakowski H. Acomparison of outcomes of mitral valve repair for degen-erative disease with posterior, anterior, and bileafletprolapse. J Thorac Cardiovasc Surg 2005;130:1242e1249.

33. Jacobs GJ, Calvert CA, Mahaffey MB, Hall DG. Echocardio-graphic detection of flail left atrioventricular valve cuspfrom ruptured chordae tendinae in 4 dogs. J Vet Intern Med1995;9:341e346.

34. Serres F, Chetboul V, Tissier R, Carlos Sampedrano C,Gouni V, Nicolle AP, Pouchelon JL. Chordae tendinaerupture in dogs with degenerative mitral valve disease:prevalence, survival and prognostic factors (114 cases,2001e2006). J Vet Intern Med 2007;21:258e264.

35. Grenadier E, Alpan G, Keidar S, Palant A. The prevalence ofruptured chordae tendineae in the mitral valve prolapsesyndrome. Am Heart J 1983;105:603e610.

36. Grenadier E, Keidar S, Sahn DJ, Alpan G, Goldberg SJ, Val-dez Cruz LM, Lima CO, Barron JV, Allen HD, Palant A.Ruptured mitral chordae tendineae may be a frequent andinsignificant complication in the mitral valve prolapsesyndrome. Eur Heart J 1985;6:1006e1015.

37. Ling LH, Enriquez-Sarano M, Seward JB, Orszulak TA,Schaff HV, Bailey KR, Tajik AJ, Frye RL. Early surgery inpatients with mitral regurgitation due to flail leaflets:a long-term outcome study. Circulation 1997;96:1819e1825.

38. Amirak E, Chan KM, Zakkar M, Punjabi PP. Current status ofsurgery for degenerative mitral valve disease. Prog Car-diovasc Dis 2009;51:454e459.

39. Zoghbi WA, Enriquez-Sarano M, Foster E, Grayburn PA,Kraft CD, Levine RA, Nihoyannopoulos P, Otto CM,Quinones MA, Rakowski H, Stewart WJ, Waggoner A,Weissman NJ, American Society of Echocardiography.Recommendations for evaluation of the severity of nativevalvular regurgitation with two-dimensional and Dopplerechocardiography. J Am Soc Echocardiogr 2003;16:777e802.

40. Serres F, Pouchelon JL, Poujol L, Lefebvre HP, Trumel C,Daste T, Carlos Sampedrano C, Gouni V, Tissier R, Hawa G,Chetboul V. Plasma N-terminal pro-B-type natriureticpeptide concentration helps to predict survival in dogs withsymptomatic degenerative mitral valve disease regardlessof and in combination with the initial clinical status atadmission. J Vet Cardiol 2009;11:103e121.

41. Utsunomiya T, Ogawa T, Doshi R, Patel D, Quan M,Henry WL, Gardin JM. Doppler color flow “proximal iso-velocity surface area” method for estimating volume flowrate: effects of orifice shape and machine factors. J Am CollCardiol 1991;17:1103e1111.

42. Simpson IA, Shiota T, Gharib M, Sahn DJ. Current status offlow convergence for clinical applications: is it a leaningtower of “PISA”. J Am Coll Cardiol 1996;27:504e509.

43. Enriquez-Sarano M, Avierinos J-F, Messika-Zeitoon D,Detaint D, Capps M, Nkomo V, Scott C, Schaff HV, Tajik AJ.Quantitative determinants of the outcome of asymptomaticmitral regurgitation. N Engl J Med 2005;352:875e883.

44. Chetboul V, Athanassiadis N, Concordet D, Nicolle A,Tessier D, Castagnet M, Pouchelon JL, Lefebvre HP.Observer-dependent variability of quantitative clinical

endpoints: the example of canine echocardiography. J VetPharmacol Ther 2004;27:49e56.

45. Chetboul V, Carlos Sampedrano C, Concordet D, Tissier R,Lamour T, Ginesta J, Gouni V, Nicolle AP, Pouchelon JL,Lefebvre HP. Use of quantitative two-dimensional colortissue Doppler imaging for assessment of left ventricularradial and longitudinal myocardial velocities in dogs. Am JVet Res 2005;66:953e961.

46. Buchanan JW. Spontaneous left atrial rupture in dogs. AdvExp Med Biol 1972;22:315e334.

47. Komitor DA. Left atrial rupture. Infrequent sequel tochronic microvalvular insufficiency. Vet Med Small Anim Clin1976;71:620e621.

48. Sadanaga KK, MacDonald MJ, Buchanan JW. Echocardiog-raphy and surgery in a dog with left atrial rupture andhemopericardium. J Vet Intern Med 1990;4:216e221.

49. Prosek R, Sisson DD, Oyama MA. What is your diagnosis?Pericardial effusion with a clot in the pericardial spacelikely caused by left atrial rupture secondary to mitralregurgitation. J Am Vet Med Assoc 2003;222:441e442.

50. Peddle GD, Buchanan JW. Acquired atrial septal defectssecondary to rupture of the atrial septum in dogs withdegenerative mitral valve disease. J Vet Cardiol 2010;12:129e134.

51. Marciniak A, Claus P, Sutherland GR, Marciniak M, Karu T,Baltabaeva A, Merli E, Bijnens B, Jahangiri M. Changes insystolic left ventricular function in isolated mitral regurgi-tation. A strain rate imaging study. Eur Heart J 2007;28:2627e2636.

52. Falk T, Jonsson L, Olsen LH, Tarnow I, Pedersen HD. Asso-ciations between cardiac pathology and clinical, echocar-diographic and electrocardiographic findings in dogs withchronic congestive heart failure. Vet J 2010;185:68e74.

53. Kittleson MD, Eyster GE, Knowlen GG, Bari Olivier N,Anderson LK. Myocardial function in small dogs with chronicmitral regurgitation and severe congestive heart failure. JAm Vet Med Assoc 1984;184:455e459.

54. Borow KM, Green LH, Mann T, Sloss LJ, Braunwald E,Collins JJ, Cohn L, Grossman W. End-systolic volume asa predictor of postoperative left ventricular performancevolume overload from valvular regurgitation. Am J Med1980;68:655e663.

55. Starling MR, Kirsh MM, Montgomery DG, Gross MD. Impairedleft ventricular contractile function in patients with long-term mitral regurgitation and normal ejection fraction. JAm Coll Cardiol 1993;22:239e250.

56. Suzuki K, Aoyagi S, Koie H, Asano R. The effect of 7.2 hyper-tonic saline solution on M-mode echocardiographic indices innormovolemic dogs. J Vet Med Sci 2006;68:749e751.

57. Ljungvall I, Hoglund K, Carnabuci C, Tidholm A,Haggstrom J. Assessment of global and regional leftventricular volume and shape by real-time 3-dimensionalechocardiography in dogs with myxomatous mitral valvedisease. J Vet Intern Med 2011;25:1036e1043.

58. Tidholm A, Westling AB, Hoglund K, Ljungvall I,Haggstrom J. Comparisons of 3-, 2-dimensional, and M-mode echocardiographical methods for estimation of leftchamber volumes in dogs with and without acquired heartdisease. J Vet Intern Med 2010;24:1414e1420.

59. Teshima K, Asano K, Iwanaga K, Koie H, Uechi M, Kato Y,Kutara K, Kanno N, Seki M, Edamura K, Hasegawa A,Tanaka S. Evaluation of left ventricular Tei index (index ofmyocardial performance) in healthy dogs and dogs withmitral regurgitation. J Vet Med Sci 2007;69:117e123.

60. Schober KE, Fuentes VL. Effects of age, body weight, andheart rate on transmitral and pulmonary venous flow inclinically normal dogs. Am J Vet Res 2001;62:1447e1454.


61. Schober KE, Bonagura JD, Scansen BA, Stern JA, Ponzio NM.Estimation of left ventricular filling pressure by use ofDoppler echocardiography in healthy anesthetized dogssubjected to acute volume loading. Am J Vet Res 2008;69:1034e1049.

62. Schober KE, Stern JA, DaCunha DN, Pedraza-Toscano AM,Shemanski D, Hamlin RL. Estimation of left ventricular fillingpressure by Doppler echocardiography in dogs with pacing-induced heart failure. J Vet Intern Med 2008;22:578e585.

63. Schober KE, Hart TM, Stern JA, Li X, Samii VF, Zekas LJ,Scansen BA, Bonagura JD. Detection of congestive heartfailure in dogs by Doppler echocardiography. J Vet InternMed 2010;24:1358e1368.

64. Schober KE, Hart TM, Stern JA, Li X, Samii VF, Zekas LJ,Scansen BA, Bonagura JD. Effects of treatment on respira-tory rate, serum natriuretic peptide concentration, andDoppler echocardiographic indices of left ventricular fillingpressure in dogs with congestive heart failure secondary todegenerative mitral valve disease and dilated cardiomyop-athy. J Am Vet Med Assoc 2011;239:468e479.

65. Oyama MA, Sisson DD, Bulmer BJ, Constable PD. Echocar-diographic estimation of mean left atrial pressure ina canine model of acute mitral valve insufficiency. J VetIntern Med 2004;18:667e672.

66. Teshima K, Asano K, Sasaki Y, Kato Y, Kutara K, Edamura K,Hasegawa A, Tanaka S. Assessment of left ventricularfunction using pulsed tissue Doppler imaging in healthy dogsand dogs with spontaneous mitral regurgitation. J Vet MedSci 2005;67:1207e1215.

67. Jacques DC, Pinsky MR, Severyn D, Gorcsan 3rd J. Influenceof alterations in loading on mitral annular velocity by tissueDoppler echocardiography and its associated ability topredict filling pressures. Chest 2004;126:1910e1918.

68. Serres F, Chetboul V, Gouni V, Tissier R, CarlosSampedrano C, Pouchelon JL. Diagnostic value of echo-

Doppler and tissue Doppler imaging in dogs with pulmo-nary arterial hypertension. J Vet Intern Med 2007;21:1280e1289.

69. Chetboul V, Escriou C, Tessier D, Richard V, Pouchelon JL,Thibault H, Lallemand F, Thuillez C, Blot S, Derumeaux G.Tissue Doppler imaging detects early asymptomaticmyocardial abnormalities in a dog model of Duchenne’scardiomyopathy. Eur Heart J 2004;25:1934e1939.

70. Tidholm A, Ljungvall I, Hoglund K, Westling AB,Haggstrom J. Tissue Doppler and strain imaging in dogs withmyxomatous mitral valve disease in different stages ofcongestive heart failure. J Vet Intern Med 2009;23:1197e1207.

71. Drighil A, Madias JE, Mathewson JW, El Mosalami H, ElBadaoui N, Ramdani B, Bennis A. Haemodialysis: effects ofacute decrease in preload on tissue Doppler imaging indicesof systolic and diastolic function of the left and rightventricles. Eur J Echocardiogr 2008;9:530e535.

72. Chetboul V, Carlos Sampedrano C, Gouni V, Nicolle AP,Pouchelon JL, Tissier R. Ultrasonographic assessment ofregional radial and longitudinal systolic function in healthyawake dogs. J Vet Intern Med 2006;20:885e893.

73. Chetboul V, Serres F, Gouni V, Tissier R, Pouchelon JL.Radial strain and strain rate by two-dimensional speckletracking echocardiography and the tissue velocity basedtechnique in the dog. J Vet Cardiol 2007;9:69e81.

74. Chetboul V, Gouni V, Carlos Sampedrano C, Tissier R,Serres F, Pouchelon JL. Assessment of regional systolic anddiastolic myocardial function using tissue Doppler and strainimaging in dogs with dilated cardiomyopathy. J Vet InternMed 2007;21:719e730.

75. Perk G, Tunick PA, Kronzon I. Non-Doppler two-dimensionalstrain imaging by echocardiographyefrom technicalconsiderations to clinical applications. J Am Soc Echocar-diography 2007;20:234e243.

Available online at www.sciencedirect.com

www.sciencedirect.com/science/journal/17602734

Documents

Echocardiographic assessment of canine degenerative … · Echocardiographic assessment of canine degenerative mitral valve disease Vale´rie Chetboul, DVM, PhD a,b,*, ... Echocardiography;