**3. Physiology of left ventricular untwist**

degrees at the subepicardium. Left ventricular twist is supposed to originate from the dynamic interaction between these oppositely wound subepicardial and subendocardial myocardial fibre helices, whereby the direction of left ventricular twist is governed by the subepicardial fibres, mainly owing to their longer arm of movement [19]. Left ventricular twist plays a pivotal role in the mechanical efficiency of the heart, making it possible that only 15% fibre shortening results in a 60% reduction in left ventricular volume [20]. Furthermore, mathematical models have shown that the counterdirectional arrangement of muscle fibres in the heart is energeti‐ cally efficient and important for equal redistribution of stresses and strain in the heart [21]. However, controversy remains present. The group of Buckberg published in 2005 a compre‐ hensive compendium, "Rethinking the cardiac helix; a structure function journey", of the Liverpool meeting: "New concepts of cardiac anatomy & physiology" [22]. Buckberg et al. believe that, based on anatomical studies by Torrent-Guasp [23] the heart is a helix that contains an apex, and that sequential contraction of the basal, descending, and ascending loop of the helix leads to the physiological pattern of myocardial contraction [24]. Although interesting, other anatomical studies have failed to reproduce the findings of Torrent-Guasp, and during the past few years this latter theory seems to gradually lose appreciation as compared to the theory of dynamic interaction between oppositely wound subepicardial and subendocardial myocardial fibres [25]. Taber et al. [19] used a theoretical model to underscore the importance of the arrangement of myocardial fibres for left ventricular function. Peak systolic twist approximately doubled with a change in de epicardial / endocardial fibre angles from +90 degrees / –90 degrees to +60 degrees / –60 degrees. The importance of fibre orientation for left ventricular twist was highlighted in clinical context as well [26]. Left ventricular sphericity index was found to have an independent positive linear relation with peak systolic twist in dilated cardiomyopathy patients. Even in dilated cardiomyopathy patient with similar left ventricular ejection fraction, left ventricular sphericity index remained positively correlated to left ventricular twist. Interestingly, in normal hearts the left ventricular sphericity index had a parabolic relation with apical peak systolic rotation and peak systolic twist. A left ventricular sphericity index of about 2.1 was associated with the highest peak systolic twist, lower and higher sphericity indices were associated with less peak systolic twist. The findings of this study seem to support the hypothesis by Taber et al. that alterations in fibre-orientation influence left ventricular peak systolic twist. Furthermore, the curvature of the left ventricular wall is related to wall tension. Since deformation of myocardial fibres is known to be inversely related to wall tension, changes in cardiac shape may also lead to changes in left ventricular

In 1995, Moon et al. [27] investigated the effects of load and inotropic state on left ventricular twist. They studied 6 cardiac transplant recipients 1 year after heart transplantation. At the time of surgery 12 radiopaque midwall left ventricular myocardial markers were implanted. The authors claimed that pressure and volume loading did not affect left ventricular twist. However, in more recent tagged MRI studies by MacGowan et al. [28] and Dong et al. [29] it has been shown that afterload changes do affect left ventricular twist. Dong et al. also inves‐ tigated the influence of preload and contractility. An isolated increase in preload resulted in an increase in left ventricular twist. From a multiple linear regression analysis, they concluded that the effect of preload on left ventricular twist was about two-thirds as great as that of

twist by means of alterations in wall tension [21].

32 Cardiomyopathies

Untwisting starts after the peak of left ventricular twist, just before the end of systole. The twisting deformation of the left ventricle during systole results not only in ejection of blood but also in storage of potential energy. During the isovolumic relaxation period the twisted fibres behave like a compressed coil that springs open while abruptly releasing the potential energy. This process may be actively supported by still depolarized subendocardial fibres that are – in contrast to the systolic period – now not opposed by active contraction of the subepi‐

cardial fibres [38]. However, the effective force of contraction of myocardial fibres is expected to be minimal during this part of the cardiac cycle. Nevertheless, dissimilarities of apparent stiffness of the endocardium and epicardium caused by differences in breakdown of actinmyosin cross-bridges may be of influence. The group of Shapiro and Rademakers was one of the first to investigate the physiology of left ventricular untwisting in more detail with MRI [39]. They found, in an open-chest canine model, that left ventricular untwisting and filling are dissociated in time. In the normal resting heart about 40% of left ventricular untwisting occurs during isovolumic relaxation. Dobutamine enhanced the extent of left ventricular untwisting before mitral valve opening and further accentuated the dissociation between left ventricular untwisting and filling. The untwisting rate, the mean left ventricular untwisting velocity during the isovolumic relaxation phase, is proportional to the rate of isovolumic pressure decay [40]. In addition, left ventricular untwisting precedes and is a strong predictor of the intraventricular pressure gradient, a marker of diastolic suction during early left ventricular filling. This may be caused by a temporal dispersion between basal and apical derotation, the diastolic reversal of systolic rotation [41]. At the left ventricular apical level there is faster de-rotation, as compared to the basal level, which may be explained by the relatively increased systolic apical rotation, and thus stored potential energy. Interestingly, at the left ventricular basal level there is still a profound de-rotation from mitral valve opening until the peak of early left ventricular filling velocity. This may be explained by the temporal dispersion in basal and apical repolarization. Since the basal endocardial fibres are the latest to be repolarized (repolarization progresses from the apex to the base of the heart and from the epicardium to the endocardium, and takes approximately 150ms), an extra de-rotating force may still be present during this period at the basal level. Furthermore, there is a brief episode of re-rotation at the basal level from the peak to the end of the early left ventricular filling velocity that may partially be explained by the sudden omission of the de-rotational forces of the endocardial fibres, at the moment of complete cardiac repolarization. In contrast, during this period continuing de-rotation is seen at the left ventricular apical level. Since rotation is related to an increase and de-rotation to a decrease in left ventricular pressure, this phenom‐ enon may facilitate blood flow all the way to the apex. Thus, left ventricular untwisting provides a temporal link between two crucial diastolic phenomena, relaxation and diastolic suction.

increased potential energy stored during this augmented systolic twisting deformation may be the cause of preserved peak diastolic untwisting velocity and untwisting rate with aging. A strong age-independent relation between left ventricular peak systolic twist and peak diastolic untwisting velocity and untwisting rate supports this hypothesis. Nevertheless, although peak diastolic untwisting velocity and untwisting rate do not change significantly with advancing age, both parameters are significantly impaired when normalized for the increased extent of left ventricular twist. This results in a progressive delay in relative left ventricular untwisting and in the time-to-peak diastolic untwisting velocity with aging. This may reflect the increased stiffness known to occur in aging. In addition, the same subendo‐ cardial dysfunction that is supposed to lead to increased left ventricular twist with aging, may also lead to loss of the active part of untwisting normally caused by in early diastole still depolarized subendocardial fibres. Relatively reduced and delayed left ventricular untwisting may help to explain the increased duration of isovolumic relaxation in the elderly. Because left ventricular untwisting generates the left ventricular pressure gradient that helps filling the left ventricle [10], impediment of left ventricular untwisting may lead to delayed generation of

Left Ventricular Twist in Cardiomyopathy http://dx.doi.org/10.5772/55281 35

this pressure gradient, and thereby to delayed opening of the mitral valve.

As mentioned before, left ventricular twist originates from the dynamic interaction between oppositely wound subepicardial and subendocardial myocardial fibres. The direction of left ventricular twist is governed by the subepicardial fibres, mainly owing to their longer arm of movement. Subendocardial ischemia with loss of contraction of the counteracting subendo‐ cardial fibres will lead to increased left ventricular twist. Therefore, left ventricular twist, and in particular changes within one patient, may provide an easily assessable marker of suben‐ docardial ischemia. Increased left ventricular twist has been described in aging healthy subjects (as discussed previously), and in patients with hypertrophic cardiomyopathy (HCM), aortic

In HCM patients, left ventricular twist is increased [45,46]. Actually, in particular left ventric‐ ular basal rotation is augmented [46]. The increased basal rotation may be explained by loss of counteraction of the subendocardial fibre helix, caused by endocardial ischemia due to microvascular dysfunction [47,48]. Also, larger radius differences between the subepicardium and subendocardium in hypertrophic muscle may increase the dominant action of the subepicardial fibres and increase basal rotation. Interestingly, left ventricular apical rotation and twist are dependent on the pattern of left ventricular hypertrophy. In patients with a sigmoidal septal curvature, left ventricular apical rotation and twist are increased as compared to patients with a reverse septal curvature. This may be partly explained by the degree of subendocardial ischemia, since patients with a sigmoidal septal curvature more often have left ventricular outflow tract obstruction. The extravascular compressive forces caused by

**4. Left ventricular twist in cardiac disease**

**4.1. Subendocardial dysfunction**

stenosis (AS), or diabetes.

In adolescents and young adults, there may be a marked contribution of active left ventricular relaxation to left ventricular filling, resulting in an accentuated early diastolic filling velocity with a short deceleration time, resembling restrictive left ventricular filling at Doppler echocardiography ('pseudo-restrictive' left ventricular filling pattern). Very rapid left ventric‐ ular untwisting plays a pivotal role in this physiological rapid early diastolic filling [42]. In contrast, in dilated cardiomyopathy patients, untwisting is delayed and this impairment to utilize suction may impair left ventricular filling [42].

Marked changes in left ventricular diastolic function are known to occur in healthy elderly [43,44]. As described before, with advancing age left ventricular twist increases, probably due to both a decrease in rotational deformation delay and subendocardial dysfunction leading to loss of the counteraction of the subendocardial fibre helix. The early diastolic release of increased potential energy stored during this augmented systolic twisting deformation may be the cause of preserved peak diastolic untwisting velocity and untwisting rate with aging. A strong age-independent relation between left ventricular peak systolic twist and peak diastolic untwisting velocity and untwisting rate supports this hypothesis. Nevertheless, although peak diastolic untwisting velocity and untwisting rate do not change significantly with advancing age, both parameters are significantly impaired when normalized for the increased extent of left ventricular twist. This results in a progressive delay in relative left ventricular untwisting and in the time-to-peak diastolic untwisting velocity with aging. This may reflect the increased stiffness known to occur in aging. In addition, the same subendo‐ cardial dysfunction that is supposed to lead to increased left ventricular twist with aging, may also lead to loss of the active part of untwisting normally caused by in early diastole still depolarized subendocardial fibres. Relatively reduced and delayed left ventricular untwisting may help to explain the increased duration of isovolumic relaxation in the elderly. Because left ventricular untwisting generates the left ventricular pressure gradient that helps filling the left ventricle [10], impediment of left ventricular untwisting may lead to delayed generation of this pressure gradient, and thereby to delayed opening of the mitral valve.
