Left ventricular twist dynamics: principles and applications

BASIC SCIENCE

Left ventricular twist dynamics: principles andapplicationsCarmen C Beladan, Andreea Călin, Monica Roşca, Carmen Ginghină,Bogdan A Popescu

▸ Additional references andvideos are published onlineonly. To view please visit thejournal online (http://dx.doi.org/10.1136/heartjnl-2012-302064).

Department of Cardiology,“Carol Davila” University ofMedicine and Pharmacy, “ProfDr C C Iliescu” Institute ofCardiovascular Diseases,Bucharest, Romania

Correspondence toAssociate Professor BogdanA Popescu, “Carol Davila”University of Medicine andPharmacy—Euroecolab, “ProfDr C C Iliescu” Institute ofCardiovascular Diseases, SosFundeni 258, sector 2,Bucharest 022328, Romania;[email protected]

To cite: Beladan CC,Călin A, Roşca M, et al.Heart Published Online First:[please include Day MonthYear] doi:10.1136/heartjnl-2012-302064

PHYSIOLOGICAL FUNDAMENTALS OF LEFTVENTRICULAR TORSIONLeft ventricular anatomyThe normal left ventricular (LV) shape has beenassimilated to a thick walled prolate ellipsoid withits long axis directed from apex to base.1 w1 Thenormal LV morphology is characterised by a highdegree of regional heterogeneity. With the use oftagging magnetic resonance, a wide variation incircumferential and longitudinal radii of LV wallcurvature has been shown, with a more ovalshaped LV cavity in the short axis direction andmore flattened LV wall towards the apex.2 In add-ition, a gradual thinning of the LV wall was notedtoward the apex, whereas around the LV circum-ference, the posterolateral wall was significantlythicker than the septum.2 The law of Laplace andvariations in the amount of longitudinally and cir-cumferentially oriented fibres have been used toexplain the wide variation in the thickness of theLV wall.2 w2 w3

The myocardial architecture of the left ventriclehas been described in most studies as having anoblique helical fibre arrangement with a righthanded helix in the subendocardial region thatgradually changes into a left handed helix in thesubepicardial region.3 4 w4–w7 Myocardial fibres inthe mid LV wall are mainly oriented in the circum-ferential direction, whereas epicardial fibres spiralobliquely toward the apex and endocardial fibresspiral obliquely toward the base (figure 1).w8 w9

This counterdirectional helical arrangement ofmyocardial fibres is energetically efficient andnecessary for uniform redistribution of stresses andstrain in the heart, as demonstrated using mathem-atical models.w5

A helical ventricular myocardial fibre array,very similar across mammalian species, wasrecently confirmed using diffusion tensor cardiacmagnetic resonance (CMR) imaging, which pro-vides a discrete measurement of the three dimen-sional (3D) arrangement of myocytes by theobservation of local anisotropic diffusion ofwater molecules (figure 1C).w9 w10 Using thesimplified tractographic reconstruction method,Poveda et alw10 demonstrated a continuoushelical structure of the ventricular myocardium,tracked from the pulmonary artery to the aorta.The helical structure enclosing the basal ring wasfurther tracked inside the left ventricle towardsthe apex and a counterdirectional helical orien-tation around the ventricles was observed.

LV torsional deformation and its physiologicalsignificanceShortening and lengthening of helically orientedmyocardial fibres result in a systolic twist followedby a diastolic untwisting of the left ventricle.In systole, the LV apex undergoes a counterclock-

wise rotation about its longitudinal axis as viewedfrom the apex (figure 1B), preceded by a brief clock-wise rotation due to the earlier shortening of thesubendocardial fibres during isovolumic contraction.Rotation of the LV base in systole is opposite in dir-ection compared to apical rotation, with a briefcounterclockwise rotation during isovolumic con-traction followed by a clockwise rotation during LVejection.3 5 w11–w13 The net result is the twist of theleft ventricle which helps to eject the blood into theaorta with increased mechanical efficiency. Thismotion has been compared with that used tosqueeze water out of a wet towel. As the systolicrotation of the LV base is lower in magnitude com-pared to apical rotation, the main determinant of LVtwist is apical rotation.w12 The definitions of LVrotational parameters are presented in box 1.It is assumed that the direction of LV twist is due

to the mechanical advantage of the subepicardialfibres relative to the inner layers, because of theirgreater distance from the LV axis, increasing theirtorque relative to the endocardial fibres due to alarger lever arm.6

Systolic twist is a key determinant of an efficientLV systolic performance, allowing the increase ofintraventricular pressure with reduced ventriculartransmural strain and oxygen consumption.7 w14

Systolic LV twist is followed by rapid LV untwistingin early diastole, as the potential energy storedduring systole in the compressed elastic elements ofthe myocardium is released. Most of the LV untwist-ing occurs during isovolumic relaxation, the apex ofthe LV being the main source of LV recoil.8 Rapiddiastolic untwisting is critical for a swift decrease inLV pressure as it creates an early diastolic intraven-tricular pressure gradient (IVPG) which contributesto diastolic suction.8 It has been demonstrated thatthe untwisting rate is correlated with both the timeconstant of isovolumic relaxation (tau) and earlydiastolic IVPG.8 w15 Moreover, a temporal sequenceof diastolic events was demonstrated, peak diastolicuntwisting preceding peak diastolic IVPG and peakearly diastolic filling velocity (figure 2).8 Therefore,any condition affecting the rate of untwisting cansignificantly impair LV filling, LV end diastolicvolume, and stroke volume.w14

Beladan CC, et al. Heart 2013;0:1–10. doi:10.1136/heartjnl-2012-302064 1

Education in Heart Heart Online First, published on May 9, 2013 as 10.1136/heartjnl-2012-302064

Copyright Article author (or their employer) 2013. Produced by BMJ Publishing Group Ltd (& BCS) under licence.

group.bmj.com on May 9, 2013 - Published by heart.bmj.comDownloaded from

http://dx.doi.org/10.1136/heartjnl-2012-302064




http://heart.bmj.com/

http://group.bmj.com/

IMAGING TECHNIQUES TO ASSESS LVTORSIONThe twisting motion of the left ventricle was firstdescribed in the 17th century by Sir WilliamHarvey after observation of open chest animalhearts. Measurements of LV torsion in man werecarried out from 1962.9

Invasive studiesIn early studies, radio-opaque markers were suturedinto the LV epicardium or myocardium of patientsduring cardiac surgery and the relative positions ofthese markers were measured throughout severalcardiac cycles using cineangiography.w16 w17 Ingelset al9 demonstrated that there was a wringingmotion of the LV during contraction.Sonomicrometry is another invasive method for

the assessment of LV rotation and torsion. Thistechnique requires intramyocardial implantation ofsmall crystals that are tracked using specialiseddigital receivers.w17 Given its high accuracy, sono-micrometry is frequently used as the gold standardfor validation of non-invasive methods.The assessment of LV torsion in man using inva-

sive methods has important limitations, these tech-niques being feasible only in patients who undergocardiac surgery. Moreover, the invasive nature ofthese methods may lead to local complications andmay therefore influence LV torsion. Therefore,non-invasive imaging techniques were proposed toquantify LV torsion.10

CMR imagingCMR with tissue tagging is a non-invasive tech-nique currently used as the imaging gold standardto assess LV rotation and torsion. The principle ofthis method relies on its ability to manipulate the

Figure 1 (A) Dissection of the human heart illustrating the progressive change in the orientation of myocardial fibres of the left ventricular (LV)wall. (a) Fibres in the superficial subepicardial layer course obliquely toward the apex in a left handed helix. (b) In the mid LV wall fibres are mainlyoriented in the circumferential direction. (c) An intermediate layer between a and b. (B). The vortex-like orientation of myocardial fibres in the LVapex explains the counterclockwise systolic apical rotation. (C) Tractography of the heart—a graphical representation of the direction and orientationof the vector fields using diffusion tensor cardiac MRI, allowing a description of fibre tracts passing through a region of interest placed in the lateralwall of the left ventricle in normal human, sheep, and rat hearts. The tracts have been limited in length to half the LV circumference. The helix angleis steeper in the mid ventricular region of myofibres and becomes more circumferential as the fibre approaches the apex and base of the leftventricle. Fibre organisation in the three species is extremely similar, with the endocardial and epicardial fibres crossing over each other with anangle of 100–120°. (A and B) Reproduced with permission from Filipoiu.w8 (C) Reproduced with permission from Mekkaoui et al.w9

Box 1 Terms and definitions: parameters ofleft ventricular torsionw30

▸ Left ventricular (LV) rotation = myocardialrotation around the long axis of the leftventricle. It is rotational displacement and isexpressed in degrees.

▸ LV twist = the absolute apex-to-base differencein LV rotation referred to as the net angle (alsoexpressed in degrees).

▸ LV torsion = the base-to-apex gradient in therotation angle along the long axis (=LV twist/LVlength), expressed in degrees per centimetre.

▸ LV twist rate = systolic time derivative of LVtwist.

▸ LV untwisting is usually defined as the peakdiastolic time derivative of twist.

2 Beladan CC, et al. Heart 2013;0:1–10. doi:10.1136/heartjnl-2012-302064

Education in Heart




longitudinal magnetisation of the myocardiumusing a unique pulse sequence and to create non-invasive markers that can be tracked in any anatom-ical plane throughout the cardiac cycle.10 Analysingthe movement of these tissue tags, the dynamic LVrotation can be quantified by acquiring cross-sectional images at both the LV apex and base (atthe level of the mitral valve) (see onlinesupplementary videos 1 and 2).w14 The majoradvantages of CMR derived torsion measurementsare the high spatial resolution and the precision oftissue tagging. The low temporal resolution—themain limitation of this technique—has been par-tially overcome with the development of fasterimaging protocols.w18–w20 However, the high cost,limited availability, and the impossibility of study-ing patients with pacemakers and implantable car-dioverter defibrillators remain limitations of CMRin quantifying LV rotational dynamics in the clinicalsetting.w1

More recently, diffusion tensor CMR has beenshow to provide a 3D characterisation of myocardialarchitecture (figure 1C).w9 CMR tractography is agraphical representation of both direction and orien-tation of vector fields, allowing a description of ven-tricular anatomy.w10 This method may quantify fibrearchitecture in normal and diseased hearts.w9

A multiscale tractographic visualisation approachbased on diffusion tensor CMR to characterise thearchitectural organisation of the normal heart wasused in a canine model.w9 The simplified tracto-graphic reconstruction method supports a helicalventricular myocardial fibre array. Diffusion spec-trum imaging with tractography was also capable ofdistinguishing normal from disordered myoarchi-tecture in a mouse model of congenital hyper-trophic cardiomyopathy (HCM). Wang et alw21

demonstrated that the ablation of myosin binding

protein C was associated with a characteristicchange in the transmural progression of helixangles throughout the myocardial wall.

EchocardiographyTissue Doppler imaging (TDI) has been introducedas a method to quantify LV rotation and torsionwith better temporal resolution.11 Myocardialtissue velocities can be measured at the base andapex of the left ventricle in the short axis plane,allowing for the determination of rotational veloci-ties and torsion. LV rotation is mathematically cal-culated by integrating the rotational velocitydetermined from TDI velocities of the septal andlateral regions and correcting for the LV radius overtime.11 w14 The TDI technique of assessing LV rota-tion has been shown to correlate well with taggingCMR.11 Given its high temporal resolution, TDIallows a reliable assessment of LV twisting anduntwisting over time. However, angle dependencyis a well known limitation. In addition, the meas-urement of myocardial velocities is influenced bycardiac translational artefacts. Moreover, the needfor complex mathematical formulas to derive rota-tion and torsion limits its use in clinical practice.Two dimensional speckle tracking echocardio-

graphy (2D-STE) has emerged as an alternative non-invasive technique to assess LV rotation. Naturalacoustic markers present within 2D ultrasoundtissue images are tracked throughout the cardiaccycle to quantify the rotational displacement of themyocardium.12 w22 w23 After manually tracing theendocardial border, ‘speckles’ are automaticallyselected using computer software and then trackedframe by frame.w14 w24 Similar to CMR, two shortaxis planes at the apex and the base need to beacquired (see online supplementary videos 3 and 4).Care should be taken to ensure that the basal short

Figure 2 Myocardial velocity profile of three left ventricular (LV) motion components during systole and early diastole in normal subjects. AveragedLV twisting and velocity profiles derived from an average of three beats each. Long (red line) and short (orange line) axis myocardial velocities andthe LV torsional velocity (purple line, systolic twisting and diastolic untwisting) profile at rest and during exercise are shown. After the contraction/systolic twisting phase, the diastolic process has started. LV untwisting precedes both long axis lengthening and short axis expansion. Duringexercise, the LV untwisting velocity was notably enhanced, keeping the temporal sequence in early diastole. The time sequence was normalised tosystolic duration, ie, t=0 indicates the onset of the QRS interval of the ECG, and t=100, end systole (AC). Error bars are marked at every 10% ofsystolic duration. AC, aortic valve closure (ie, end systole); AO, aortic valve opening; Ej, peak ejection flow velocity in the outflow tract; En-E, end ofearly filling; MC, mitral valve closure; MO, mitral valve opening; Pk-E, peak early filling velocity; Pk-IVPG, peak intraventricular pressure gradient(the timing is inserted with arrow). Reproduced with permission from Notomi et al.8


Education in Heart


http://heart.bmj.com/lookup/suppl/doi:10.1136/heartjnl-2012-302064EiH0780/-/DC1



axis plane contains the mitral valve, and the apicalplane is acquired distally to the papillary muscles.Rotation angles and rates of rotation are providedby the software for six equidistant regions aroundthe circumference for both LV apex and base(figure 3).w14

Reliable speckle tracking analysis requires highquality second harmonic images as well as sectorsize and imaging depth adjustments for an optimalframe rate. The frame rates suggested for restingstudies are within a range of 50–90 Hz.w17

Speckle tracking has been shown to correlatewell with CMR in humans and with sonomicro-metry in experimental models, during both rest anddobutamine infusion.12 w21 However, lower valuesfor apical rotation and LV twist were reported for2D-STE when compared with tagged CMR. It wassuggested that acquiring the LV apical plane accord-ing to textbook recommendations (by angulatingthe transducer from the standard parasternalwindow) could explain the reported underestima-tion of apical rotation.12 w25 Discrepancies areprobably not related to intrinsic inaccuracies of2D-STE in estimating myocardial rotation, butrather to difficulties in imaging the real LV apex.Thus, when 2D-STE and CMR data acquired atsimilar levels are compared, both techniquesprovide similar values.w26 Therefore, changing thetransducer position from the standard parasternalposition to a more caudal intercostal space isrecommended for a more accurate measurement ofLV apical rotation (figure 4).w27

As images are acquired at different momentsfrom different cardiac cycles, twist computation asthe difference between basal and apical rotation

angles does not reflect true twist if selected cyclesare not of equal duration. The long axis movementof the heart responsible for the through-planemotion may also lead to inaccurate results.13

3D-STE is able to avoid the loss of speckles dueto motion outside the imaging plane.w28 However,despite recent advances in technology, 3D ultra-sound images are still of relatively low spatio-temporal resolution, with attenuation, shadowing,and signal dropout artefacts. These factors result ina higher degree of inconsistency of specklesbetween successive volumes of a 3D image loop.13

Several reference values for LV torsion measuredby non-invasive methods have been reported in theliterature.w29 However, so far these are not readyto be used as a tool to detect myocardial dysfunc-tion since CMR tagging and STE do not providereproducible and comparable measurements of LVtorsion. Moreover, lack of standardisation for theacquisition of short axis sections when using STE,as well as inter-vendor inconsistency in image post-processing, make it difficult to integrate the assess-ment of LV torsion parameters in clinicalpractice.w30

Vendor consistency is an important prerequisitefor a broader clinical use of STE. The reproducibil-ity of LV deformation parameters among differentultrasound machines, software packages, and obser-vers has been reported.w31 w32 To date, there areno studies to compare the inter-vendor variabilityof LV rotational parameters as assessed by STE.Until standardisation is achieved, laboratory specificnormal values with each vendor analysis packageand use of the same software in serial studies arerecommended.w30

Figure 3 Time profiles of basal rotation (A), apical rotation (B), left ventricular (LV) twist (C), and LV twist velocity (D) obtained by speckle trackingechocardiography in a normal subject. Counterclockwise rotation has positive values by agreement. The white dotted curves in the top rowcorrespond to the mean value of the six segments represented by coloured curves. At the basal level, rotation values for the six segments have awider distribution compared to the narrow range of segmental rotation values observed at the apical level. LV twist is calculated as the netdifference between apical and basal rotation at isosynchronous moments in the cardiac cycle.


Education in Heart




The relative advantages and limitations of usingdifferent techniques to assess LV rotation andtorsion are presented in table 1.

INFLUENCE OF SPECIFIC PHYSIOLOGICALCONDITIONS ON LV TORSIONAgeAn increase in LV torsion and peak untwisting vel-ocity from infancy to adulthood have beenreported in studies using tagging CMR, TDI orSTE. Moreover, Notomi et al demonstrated a dif-ferent pattern of rotation for basal and apicalplanes with increasing age.14 While apical rotationis counterclockwise constantly, basal rotationchanges from counterclockwise in infancy toneutral in early childhood and to clockwise inadulthood. As a consequence, twisting motion ofthe heart changes from solid body rotation ininfancy to a distinct wringing motion by middleage. As for LV untwisting, it was demonstrated thatin infancy this occurs simultaneously with long andshort axis lengthening, while by childhood itoccurs earlier during the isovolumic relaxationperiod and this trend continues to middle age,when untwisting is almost completed beforemitral valve opening.14 It was suggested that

subendocardial fibrosis may be a possible explan-ation for the increase in resting LV torsion withaging, with a reduction in subendocardial functionand unopposed increased subepicardial contraction.It may also represent a mechanism to compensatefor reductions in long axis function seen withaging, similar to the observed increases in radialfunction previously reported in aging individuals.14

The effect of exercise on torsion parametersacross a range of ages was evaluated by Burnset al.w33 The augmentation of torsion parametersduring exercise documented in young healthyadults is attenuated with age, and LV untwistingrate also fails to augment with exercise as ageincreases.8 w33 w34 Thus, the age related increase inresting LV torsion is associated with decreased tor-sional reserve during exercise.A senescent myocardial fibrosis has been suggested

as a possible explanation for these findings.w33

ExerciseAn enhancement of LV systolic torsion during exer-cise was reported in healthy volunteers evaluatedduring supine bicycle ergometry.8 Moreover, LVuntwisting also increased with exercise, more sig-nificantly than LV lengthening or expansion

Figure 4 The influence of transducer position on left ventricular (LV) apical rotation measurement by speckle tracking echocardiography.(A) Schematic representation of three transducer positions used for the acquisition of the short axis view of the LV apex: position 1 (parasternalstandard position) and positions 2 and 3 (respectively, one and two intercostal spaces more caudal). RV, right ventricle. Reproduced from van Dalenet al,w27 with permission from Elsevier. (B) Rotational profiles of the LV apex obtained using the three different transducer positions. Increased meanvalues for LV apical rotation and a narrower range of values for segmental rotation are obtained from a more caudal transducer position. Transducermisplacement may cause underestimation of true LV apical rotation and, consequently, of LV twist.


Education in Heart




(figure 2). It was suggested that the exerciseinduced increase in untwisting rate is a key deter-minant of early diastolic LV suction, facilitating LVfilling without an increase in left atrial pressure.8

Other studies demonstrated that long term exercisetraining may, however, reduce LV twist at rest.Thus, professional athletes showed significantlylower resting LV twist values and untwisting veloci-ties than sedentary controls.w35 Differences in heartrate or structural adaptations of the myocardiumwere considered to explain these findings. It wasalso suggested that a lower torsion at rest in athletesmay allow for a greater reserve of LV torsionduring exercise.w1

Loading conditionsPreload, afterload, contractility, heart rate, and sym-pathetic activation are physiological factors studiedfor their possible influence on LV torsion. A directcorrelation between peak LV torsion and LV end-diastolic volume and an inverse relationshipbetween LV torsion and LV end-systolic volumewere demonstrated in early experimental studiesusing canine models.w36 It was also demonstratedthat increasing contractility during inotropic stimu-lation by dobutamine infusion increased LV twistand untwisting rate whereas negative inotropicinterventions notably reduced twist.w1 w17 w36–w38

Increased rotation with chronotropic stimulationwas demonstrated on atrially paced dogs.w39

CLINICAL APPLICATIONS OF LV TORSIONASSESSMENTChanges in the pattern or magnitude of LV rota-tional parameters have been associated with variouscardiovascular diseases. At present, despite thegrowing evidence supporting their clinical implica-tions, routine use of these parameters is not

recommended.w30 However, in selected cases, mea-surements of LV torsion and untwisting can provideinformation that complements standard pumpfunction indices, offering new diagnostic and thera-peutic perspectives.

HypertensionTakeuchi et alw40 reported similar values for LVtwist among patients with hypertension with orwithout LV hypertrophy (LVH), while early dia-stolic LV untwisting was significantly delayed andreduced, in parallel to the severity of LVH.Increased LV torsion was reported in patients withhypertension and preserved LV ejection fraction(LVEF), in the presence of increased concentricityprobably due to the increased mechanical torqueadvantage of subepicardial fibres over subendocar-dial fibres.w41

Heart failure with preserved ejection fractionOne of the proposed explanations for the exer-tional elevation of LV filling pressures in patientswith heart failure with preserved ejection fraction(HFPEF) is a steep diastolic pressure–volume rela-tion in a stiff ventricle that may still keep fillingpressures low at rest, but this reserve is rapidlyexhausted during exercise.w42

Using a mathematical model, MacIver et alsought to explain how patients with HFPEF mayhave LV systolic dysfunction (ie, reduced long axisshortening and stroke volume) despite a normalejection fraction. The normal LVEF in this settingmay be explained by the presence of LVH. Theresulting increase in radial thickening explains thepreservation of LVEF despite reduced long axisshortening.w43

Initial studies reported reduced LV longitudinaland radial strains but preserved circumferential

Table 1 Advantages and limitations of different techniques in assessing left ventricular (LV) torsional dynamics

Established techniques to assess LVrotation and torsion Advantages Limitations

Implanted radiopaque metallic markers orsonomicrometers in the myocardium

High accuracy Invasive, limited to the research settingPossible source of local complications that couldinfluence LV torsion

Tagging cardiac magnetic resonance Non-invasiveHigh spatial resolution andprecision of tissue tagging

Low temporal resolutionTime consuming post-processing proceduresHigh costLimited availabilityImpossibility of studying patients with pacemakersand implantable cardioverter defibrillators

Tissue Doppler imaging (TDI) Non-invasiveHigh temporal resolutionAvailableLow cost

Angle dependentCardiac translational artefactsComplex mathematical formulas needed to deriverotation and torsionInability to image the real LV apex in a significantnumber of subjects

Two dimensional speckle trackingechocardiography

Non-invasiveLess dependent on angle ofinsonationAvailableLow cost

Lower temporal resolution as compared to TDILoss of speckles due to through-plane motionInability to image the real LV apex in a significantnumber of subjectsLack of standardisation in acquisitionInter-vendor variability in image post-processing


Education in Heart




strain, LV twist and untwisting rate in HFPEFpatients compared with normal controls.15 Parket al showed that LV torsion is dependent on thestage of diastolic dysfunction. After an increaseduring the early stage, normalisation and furtherreduction in torsion with more advanced diastolicdysfunction can explain the similar values of LVtorsion in diastolic heart failure compared withnormal controls.16 Hypertensive patients with wellcontrolled blood pressure without LVH, complain-ing of exertional dyspnoea, were evaluated at restand during submaximal supine exercise echocardio-graphy.w44 These authors found that both LV sys-tolic twist and diastolic untwist were significantlyreduced in patients compared with controls, at restand during exercise.

CardiomyopathiesA decreased and delayed systolic LV torsion as wellas depressed, delayed and disorganised LV

untwisting have been previously reported inpatients with dilated cardiomyopathy (DCM).w45

Moreover, paradoxical reversal of LV rotation, withthe base rotating counterclockwise and the apexclockwise, with subsequent reduction or even lossof LV twist were observed in some patients withDCM.17 w46 w47 As more severe LV remodellingwas noted in patients with reversed apical rotation(figure 5), it was suggested that increased LV spher-icity and widening of the apex lead to equalisationof the radii of the subepicardial and subendocardialmyocardial layers. As a result, the mechanicaladvantage of the subepicardial myofibres is reducedand LV twist decreases with increasing cavityvolume.17

The effect of cardiac resynchronisation therapy(CRT) on LV torsional mechanics is still under eva-luation.w48 w49 Preliminary data suggest that CRTmay restore LV twist in patients who showed LVreverse remodelling, possibly by providing a more

Figure 5 Two dimensional apical four chamber views (top), speckle tracking left ventricular (LV) rotation curves (middle), and schematicrepresentations of LV geometry and rotation (bottom) in: a normal subject (column A); a patient with dilated cardiomyopathy (DCM) andcounterclockwise apical rotation (column B); and a patient with DCM and reversed apical rotation (column C). Top row: Progressive increase in LVdimensions and sphericity with widening of the QRS complex, from normal (A) to DCM with severe LV remodelling (C). Middle row: (A) Normal LVapical (blue) and basal (green) rotation curves; (B) normally directed but reduced in amplitude LV rotation curves; (C) reversed apical rotation withclockwise motion of both apex and base in the patient with severely dilated LV. Bottom row: Changes in LV geometry and LV rotation at basal(green arrow) and apical (blue arrow) levels from normal (A) to mildly dilated (B), and severely dilated left ventricle (C). The amplitude of both basaland apical rotation is reduced but normally directed in the patient with DCM and mildly dilated left ventricle (B). Reversed apical rotation is seen inassociation with severe LV remodelling (C). Reproduced with permission from Popescu et al.17


Education in Heart




physiological electrical depolarisation and mechan-ical contraction of the myofibres.18 LV reductionsurgery does not change LV twist, although the rateof early diastolic untwisting may improve.w50

Pronounced variability in LV systolic twist pat-terns may be seen in HCM, depending on the dis-tribution and extent of LVH.19

Delayed untwisting is a uniform characteristic ofpatients with HCM, negatively influenced bydynamic LV outflow tract obstruction and affectingLV filling pressure and exercise capacity.w51

Moreover, during exercise, the HCM ventricle failsto increase torsion, untwisting, and diastolic IVPG,providing a possible explanation for the diastolicdysfunction commonly seen in this disorder.8 InHCM patients subjected to septal reduction LVtwist may exceed baseline values 1 week after theprocedure, and untwisting may occur earlierleading to an improvement in LV filling and exer-cise tolerance.w51 w52

Valvular heart diseaseAortic stenosisPressure overload induced by aortic stenosis is asso-ciated with increased LV twist, mainly due to anincreased apical rotation.w53 It has been suggestedthat coronary flow reduction in the subendocardialregion results in subendocardial dysfunction andunopposed subepicardial layer contraction. LVtorsion increases proportionally to the severity ofaortic stenosis and declines with increasing LV dila-tion.w54 w55 Within the same theoretical framework,Delhaas et al tested in paediatric patients with aorticstenosis the ratio of LV torsion to endocardialcircumferential shortening (torsion-to-shorteningratio), which reflects the transmural distribution ofcontractile myofibre function. Using magnetic reson-ance tagging, an increase of 40% in the LVtorsion-to-shortening ratio was found, probably indi-cating LV subendocardial dysfunction.w56

Moreover, diastolic apical untwisting is pro-longed in comparison with normal subjects and asignificant relationship between delayed LV untwist-ing and increased LV filling pressures was found inpatients with severe aortic stenosis and preservedLVEF.w53 w57 w58 Aortic valve replacement leads tonormalisation of LV torsion by a decrease in bothapical and basal rotation.w55

Mitral regurgitationIn patients with chronic mitral regurgitation, avail-able data suggest that LV torsion might depend onthe stage of the disease, with progressive deterior-ation in patients with severe mitral regurgitationdespite normal LV systolic function.w59 Significantdelays in the onset and peak of LV untwisting werereported by Borg et alw60 in patients with chronicmoderate-severe mitral regurgitation due to mitralvalve prolapse, possibly as a consequence ofincreased preload due to mitral regurgitation.These parameters may become promising indicatorsof subclinical LV dysfunction in this setting.

IschaemiaA few studies have evaluated LV torsional mechanicsin patients with myocardial ischaemia. A reduction inapical rotation with balloon occlusion of the leftanterior descending coronary artery (LAD) wasdemonstrated in patients without previous myocardialinfarction undergoing angioplasty.w61 In patients witha history of myocardial infarction and already com-promised apical rotation, no further reduction follow-ing LAD occlusion was seen. In contrast, Bansalet alw62 reported no effect of stress induced myocar-dial ischaemia on LV torsion. The authors suggestedthat transient ischaemia, predominantly confined tothe subendocardium, provoked by dobutamine infu-sion may not affect LV systolic torsion to a significantextent due to preservation of subepicardial function.A reduced apical rotation and reduced and

delayed diastolic untwisting in comparison with con-trols was reported in patients with anterolateral

Left ventricular twist dynamics: key points

▸ The myocardial architecture of the left ventriclehas an oblique double helical fibrearrangement.

▸ Shortening and lengthening of the helicallyoriented myofibres result in a wringing motionof the left ventricle with systolic twistingfollowed by diastolic untwisting.

▸ Left ventricular (LV) torsion is a keydeterminant of cardiac performance for bothsystolic ejection and diastolic filling.

▸ Both LV systolic torsion and LV untwistingincrease during exercise. The exercise inducedincrease in the untwisting rate is a keydeterminant of early diastolic LV suction,facilitating LV filling at normal pressure.

▸ Preload, afterload, contractility, heart rate, andsympathetic activation are physiological factorsinfluencing LV torsion.

▸ LV torsion and peak untwisting velocityincrease from infancy to adulthood. Moreover,a different pattern of rotation for basal andapical planes has been demonstrated withincreasing age.

▸ Sonomicrometry is an invasive method used asthe gold standard for validation of non-invasivemethods.

▸ Cardiac magnetic resonance imaging withtissue tagging is the current gold standard forthe non-invasive assessment of LV rotation andtorsion.

▸ Two dimensional speckle trackingechocardiography may also measure LVrotation and twist. Lack of standardisation inimage acquisition and inter-vendor variabilityin image post-processing limit the use of thistechnique in clinical practice at present.

▸ Changes in the pattern and/or magnitude of LVrotational parameters have been described invarious cardiovascular diseases.


Education in Heart




myocardial infarction and abnormal LVEF.20 In con-trast, systolic twist is maintained in patients withanterior myocardial infarction and LVEF ≥45%.CMR tagging was used to investigate LV function

in patients suspected of coronary artery disease(CAD).w63 Under low dose dobutamine stress, sys-tolic rotation velocity, time to peak untwist and dia-stolic rotation velocity were all significantlyimpaired in patients with CAD. The authorsreported the time to peak untwist as the mostpromising parameter for identification of patientswith CAD. A significant myofibre reorganisation inremote areas of remodelled hearts after myocardialinfarction was recently described using quantitative3D diffusion CMR tractography.w9

FUTURE DIRECTIONSThe development of imaging techniques willfurther improve our understanding of LV myocar-dial architecture and its relation to LV function,providing easier ways to quantify LV torsionaldynamics in clinical practice.Diffusion tensor CMR may be used to evaluate

the extent to which myofibre architecture can adaptto different stressors. It may also be used to studydifferent disease specific changes in fibre orienta-tion as LV remodelling progresses. Statisticalmodels of cardiac function may provide expectedranges of LV torsion and may assist in quantifyingthe degree of abnormality.Standardisation of acquisition and image post-

processing will increase the accuracy of 2D-STE inmeasuring LV rotation. Improvements in spatialand temporal resolution of 3D echocardiographyare expected to reduce speckle pattern variability.To date, no prognostic information exists on the

role of rotational parameters of LV function.However, quantifying LV rotational parametersduring exercise in different pathologies is likely toprovide important insights into pathophysiological

changes and heart failure mechanisms. The areasmost likely to benefit from further improvements inassessing LV architecture and torsional dynamics areHFPEF, cardiac dyssynchrony/resynchronisationtherapy, and LV reconstructive surgery.

Contributors BAP initiated and designed this article, criticallyrevised the draft paper, and is the guarantor of the final version ofthe manuscript. CCB performed the literature search and draftedthe initial version of the manuscript. AC, MR and CG were involvedin the conception of the review and undertook critical revisions ofthe manuscript. All the authors read and approved the finalmanuscript.

Funding This work was partially supported by Romanian NationalResearch Programme II grants: CNCSIS –UEFISCSU, project numberPNII—IDEI 2007 code ID_222 (contract 199/2007), and CNCSIS –UEFISCSU, project number PNII—IDEI 2008 code ID_447 (contract1208/2009).

Competing interests In compliance with EBAC/EACCMEguidelines, all authors participating in Education in Heart havedisclosed potential conflicts of interest that might cause a bias inthe article. BAP has received research support and lecture honorariafrom General Electric Healthcare. The other authors report noconflicts.

Provenance and peer review Commissioned; externally peerreviewed.

REFERENCES1 Rankin JS, McHale PA, Arentzen CE, et al. The three-dimensional

dynamic geometry of the left ventricle in the conscious dog. CircRes 1976;39:304–13.

▸ This study proposed the prolate ellipsoid model to describe thenormal LV shape.

2 Bogaert J, Rademakers FE. Regional nonuniformity of normaladult human left ventricle. Am J Physiol Heart Circ Physiol2001;280:H610–20.

▸ Study describing the morphological and functional non-uniformityof the normal left ventricle using magnetic resonance myocardialtagging.

3 Sengupta PP, Korinek J, Belohlavek M, et al. Left ventricularstructure and function: basic science for cardiac imaging. J AmColl Cardiol 2006;48:1988–2001.

▸ Well written review summarising the parameters required todescribe LV geometry and deformation. The emerging conceptsregarding the sequence of electromechanical activation and LVintracavitary flow are also presented.

4 Torrent-Guasp FF, Ballester M, Buckberg GD. Spatial orientationof the ventricular muscle band: physiologic contribution andsurgical implications. J Thorac Cardiovasc Surg 2001;122:389–92.

▸ This landmark paper described for the first time the continuum ofmyocardial architecture as a muscle band wrapping into twodistinct helicoids bounding both ventricles, extending from thepulmonary valve to the aortic valve.

5 Sengupta PP, Khandheria BK, Korinek J, et al. Apex-to-basedispersion in regional timing of left ventricular shortening andlengthening. J Am Coll Cardiol 2006;47:163–72.

6 Taber LA, Yang M, Podszus WW. Mechanics of ventricular torsion.J Biomech 1996;29:745–52.

▸ Study showing that the magnitude and direction of ventriculartorsion depend on the competing effects of transmural fibre stressgradients and describing the mechanical advantage of the outermyocardial layers relative to the inner layers.

7 Beyar R, Sideman S. Effect of the twisting motion on thenonuniformities of transmyocardial fiber mechanics and energydemand–a theoretical study. IEEE Trans Biomed Eng1985;32:764–9.

8 Notomi Y, Martin-Miklovic MG, Oryszak SJ, et al. Enhancedventricular untwisting during exercise: a mechanistic manifestationof elastic recoil described by Doppler tissue imaging. Circulation2006;113:2524–33.

▸ This study described the quantitative and temporal relation of LVtorsion to the dynamic events of early diastole, at rest and duringexercise.

You can get CPD/CME credits for Education in Heart

Education in Heart articles are accredited by both the UK Royal College ofPhysicians (London) and the European Board for Accreditation in Cardiology—youneed to answer the accompanying multiple choice questions (MCQs). To accessthe questions, click on BMJ Learning: Take this module on BMJ Learning fromthe content box at the top right and bottom left of the online article. For moreinformation please go to: http://heart.bmj.com/misc/education.dtl▸ RCP credits: Log your activity in your CPD diary online (http://www.

rcplondon.ac.uk/members/CPDdiary/index.asp)—pass mark is 80%.▸ EBAC credits: Print out and retain the BMJ Learning certificate once you

have completed the MCQs—pass mark is 60%. EBAC/ EACCME Credits cannow be converted to AMA PRA Category 1 CME Credits and are recognisedby all National Accreditation Authorities in Europe (http://www.ebac-cme.org/newsite/?hit=men02).

Please note: The MCQs are hosted on BMJ Learning—the best available learningwebsite for medical professionals from the BMJ Group. If prompted, subscribersmust sign into Heart with their journal’s username and password. All users mustalso complete a one-time registration on BMJ Learning and subsequently log in(with a BMJ Learning username and password) on every visit.


Education in Heart




9 Ingels NB Jr, Daughters GT 2nd, Stinson EB, et al. Measurementof midwall myocardial dynamics in intact man by radiography ofsurgically implanted markers. Circulation 1975;52:859–67.

▸ This study demonstrated that tantalum markers can be safelyimplanted in the ventricular midwall in man at the time of surgery,allowing quantitative measurements of segmental dynamics.

10 Buchalter MB, Weiss JL, Rogers WJ, et al. Noninvasivequantification of left ventricular rotational deformation in normalhumans using magnetic resonance imaging myocardial tagging.Circulation 1990;81:1236–44.

▸ This is the first study showing in the normal human heart thatventricular torsion and shear can be demonstrated and quantifiedusing tagging magnetic resonance.

11 Notomi Y, Setser RM, Shiota T, et al. Assessment of leftventricular torsional deformation by Doppler tissue imaging:validation study with tagged magnetic resonance imaging.Circulation 2005;111:1141–7.

12 Helle-Valle T, Crosby J, Edvardsen T, et al. New noninvasivemethod for assessment of left ventricular rotation: speckle trackingechocardiography. Circulation 2005;112:3149–56.

▸ This study demonstrated that LV rotation and torsion could bemeasured accurately by speckle tracking echocardiography,showing good correlation and agreement with MRI andsonomicrometry used as reference methods.

13 Ashraf M, Myronenko A, Nguyen T, et al. Defining left ventricularapex-to-base twist mechanics computed from high-resolution 3Dechocardiography: validation against sonomicrometry. JACCCardiovasc Imaging 2010;3:227–34.

14 Notomi Y, Srinath G, Shiota T, et al. Maturational and adaptivemodulation of left ventricular torsional biomechanics: Dopplertissue imaging observation from infancy to adulthood. Circulation2006;113:2534–41.

▸ Study describing the age related changes in LV torsion in normalsubjects.

15 Wang J, Khoury DS, Yue Y, et al. Preserved left ventricular twistand circumferential deformation, but depressed longitudinal andradial deformation in patients with diastolic heart failure. EurHeart J 2008;29:1283–9.

16 Park SJ, Miyazaki C, Bruce CJ, et al. Left ventricular torsion bytwo-dimensional speckle tracking echocardiography in patientswith diastolic dysfunction and normal ejection fraction. J Am SocEchocardiogr 2008;21:1129–37.

17 Popescu BA, Beladan CC, Calin A, et al. Left ventricularremodelling and torsional dynamics in dilated cardiomyopathy:reversed apical rotation as a marker of disease severity. Eur JHeart Fail 2009;11:945–51.

18 Bertini M, Marsan NA, Delgado V, et al. Effects of cardiacresynchronization therapy on left ventricular twist. J Am CollCardiol 2009;54:1317–25.

19 van Dalen BM, Kauer F, Soliman OI, et al. Influence of thepattern of hypertrophy on left ventricular twist in hypertrophiccardiomyopathy. Heart 2009;95:657–61.

20 Takeuchi M, Nishikage T, Nakai H, et al. The assessment of leftventricular twist in anterior wall myocardial infarction usingtwo-dimensional speckle tracking imaging. J Am Soc Echocardiogr2007;20:36–44.


Education in Heart




doi: 10.1136/heartjnl-2012-302064 published online May 9, 2013Heart

Carmen C Beladan, Andreea Calin, Monica Rosca, et al. and applicationsLeft ventricular twist dynamics: principles

http://heart.bmj.com/content/early/2013/05/08/heartjnl-2012-302064.full.htmlUpdated information and services can be found at:

These include:

References http://heart.bmj.com/content/early/2013/05/08/heartjnl-2012-302064.full.html#ref-list-1

This article cites 20 articles, 12 of which can be accessed free at:

P<P Published online May 9, 2013 in advance of the print journal.

serviceEmail alerting

the box at the top right corner of the online article.Receive free email alerts when new articles cite this article. Sign up in

CollectionsTopic

(3745 articles)Clinical diagnostic tests � (452 articles)Education in Heart �

(4 articles)Basic science � Articles on similar topics can be found in the following collections

Notes

(DOIs) and date of initial publication. publication. Citations to Advance online articles must include the digital object identifier citable and establish publication priority; they are indexed by PubMed from initialtypeset, but have not not yet appeared in the paper journal. Advance online articles are Advance online articles have been peer reviewed, accepted for publication, edited and

http://group.bmj.com/group/rights-licensing/permissionsTo request permissions go to:

http://journals.bmj.com/cgi/reprintformTo order reprints go to:

http://group.bmj.com/subscribe/To subscribe to BMJ go to:


http://heart.bmj.com/content/early/2013/05/08/heartjnl-2012-302064.full.html

http://heart.bmj.com/content/early/2013/05/08/heartjnl-2012-302064.full.html#ref-list-1

http://heart.bmj.com/cgi/collection/basic_science

http://heart.bmj.com/cgi/collection/education_in_heart

http://heart.bmj.com/cgi/collection/clinical_diagnostic_tests

http://group.bmj.com/group/rights-licensing/permissions

http://journals.bmj.com/cgi/reprintform

http://group.bmj.com/subscribe/



Left ventricular twist dynamics: principles and applications

Documents

Transcript of Left ventricular twist dynamics: principles and applications