Kept in the loop: longitudinal strain–volume relationships for the assessment of left ventricular mechanical performance

In health, the aortic valve facilitates unidirectional forward flow from the left ventricle (LV) to the systemic circulation. During systole, the valve offers little impedance to LV outflow, while during diastole, it prevents regurgitant backflow. Aortic stenosis (AS) and regurgitation (AR) present two classic forms of haemodynamic overload to the LV. AS represents systolic pressure overload and is associated with concentric remodelling and/or hypertrophy. AR results primarily in diastolic volume overload, and is associated with eccentric hypertrophy. Both lesions alter the mechanical loading of the LV and, along with their associated patterns of chamber remodelling, may provoke changes in myocardial mechanics. Recent advances in ultrasonic imaging have allowed for an improved assessment of ventricular mechanics through the use of speckle-tracking strain analysis. Strain is a dimensionless quantity which represents the fractional length change of a deforming tissue segment. As such, strain may be used to assess the shortening and lengthening of myocardial fibres, which reduce chamber volume during systolic ejection, and restore chamber volume during diastolic filling. Relating changes in myocardial strain throughout the cardiac cycle to changes in cavity volume may yield additional insight into the effects of haemodynamic derangement on chamber mechanics. In a recent article in The Journal of Physiology, Hulshof et al. (2017) examined whether strain–volume relationships provide novel information regarding LV mechanics operating under different loading states. The authors hypothesized that strain–volume loop characteristics would provide unique information that could better discern AS and AR, relative to routinely used echocardiographic measures including LV ejection fraction (EF) and peak systolic strain. Participants were retrospectively selected and assigned to study groups based on an absence of documented cardiovascular disease (Control), severe AR, or severe AS. Apical four-chamber echocardiographic images were analysed using speckle-tracking software to assess longitudinal myocardial strain, with cavity area determined by tracing the endocardial border at time points corresponding to 5% intervals of the cardiac cycle to couple strain and area. Chamber volume was then estimated from monoplane area measurements, and used to construct strain–volume relationships. The authors noted that loop characteristics could distinguish patients with valvular disease from controls, and that early systolic strain and linear slope could further distinguish the two patient groups from each other, while LV-EF did not distinguish between groups. The authors concluded that strain–volume loops were better able to discriminate between functional characteristics of aortic valve disease and provide insight into the cardiac mechanical impact of these diseases. The study design has both strengths and weaknesses that merit discussion. Inclusion and exclusion criteria were appropriate and a blinded a priori inspection of potential cases for sonographic image quality prior to inclusion may limit bias in data analysis. LV mechanical performance is known to differ between males and females, and the authors should be commended for considering whether strain–volume relationship indices exhibit sex differences. Although there were no statistically significant differences between sexes within each group, the relatively limited sample size may have precluded meaningful analysis and this could be explored more thoroughly in future work. An abnormal LV-EF was cited as an exclusion criterion. Given the ongoing variability in the clinical definitions of ‘reduced’, ‘preserved’, and ‘borderline’ EF, it would have been helpful to have stated the cut-off value used. Excluding subjects with a reduced EF is rational, since reduced systolic function may present a falsely low aortic valve gradient; further, such patients may be indicated for valve replacement. The selection of patients with ‘normal’ EF is therefore a strength of the study, providing support for the discriminative and/or prognostic valve of the strain–volume loops in health and disease that are not immediately apparent from conventional echocardiographic measures. It would also have been useful to know if patients were in sinus rhythm at the time of their echocardiogram, since atrial fibrillation is common in patients with valvular heart disease. Lastly, apical four-chamber images were analysed for monoplane volume. Other works from this group have reported the strain–area relationship, which involves fewer assumptions of chamber geometry. In this study, LV strain–volume relations were examined within the longitudinal plane. Global longitudinal LV strain represents the fractional shortening (∆L/L0) of the myocardium parallel to the endocardial border, and was determined across the cardiac cycle using speckle-tracking software. LV volume was estimated by tracing the endocardial border to quantify cavity area at corresponding time points. The close association between strain and volume can thus be appreciated on the basis of the geometric relationship between cavity perimeter and area. This is illustrated by the S-slope, which approximates the total strain–∆volume relationship over systole (and more broadly, the entire cardiac cycle). The authors speculated that a lower S-slope in AR may reflect reduced myocardial contractility. Contractility is most accurately described as the maximal shortening velocity of isolated muscle at a given loading state. However, translating this concept to the intact, three-dimensional ventricle is challenging. Longitudinal LV strain does not relate independently to indices of inotropic state, but varies inversely with preload (Mak et al. 2012). Moreover, AR is characterized by the serial addition of sarcomeres and eccentric remodelling in response to volume overload, and contractile performance at the sarcomere level is preserved in volume overload (Ross & McCullagh, 1972). An alternative interpretation of the group differences in S-slope may be that the AR group is able to reduce cavity size (i.e. generate stroke volume) by nearly twice the magnitude of the control and AS groups with slightly less total strain, due to the substantially greater cavity size (i.e. L0 or end-diastolic volume (EDV)). It is possible that if the volume axis was normalized to %EDV (i.e. the %∆L–%∆V relation), the between-group differences in the intercept and/or slope of the strain–volume relationship may be partially attenuated (Fig. 1). However, other loop characteristics such as early systolic strain and early diastolic uncoupling were normalized to %EDV and showed clear differences between groups. Less clear is the mechanism underlying the dissociation of systolic and diastolic strain at a given %EDV in patient groups. Diastolic strain–volume relations were similar in morphology among groups, though AS and AR groups demonstrated a shallower prevailing slope as well as a rightward shift in AR, which may reflect eccentric remodelling. However, during diastole, both patient groups were characterized by a more negative diastolic strain relative to systolic strain at a given volume compared to the control group. Further, the absolute magnitude of the systolic−diastolic strain difference was relatively similar between groups (∼3–5%), although it appeared statistically different due to the opposing directionality between control and patient groups. Uncoupling of the strain–volume relation may relate to differences in cavity shape over the cardiac cycle (i.e. the perimeter–area relation). Due to the bullet shape of the LV, more volume is contained in basal segments than in apical segments (Carlsson et al. 2007). This may be further affected by spatial–temporal differences in active contraction and relaxation patterns, structural remodelling in disease, and regional heterogeneity in strain that are not communicated by global values. In the normal heart, contraction proceeds from apex to base, and an apical-to-basal gradient in the magnitude of strain exists. In contrast, in dilated cardiomyopathy, apical strain is reduced while basal strain is preserved, with a resulting reduction in peak longitudinal strain (Sengupta et al. 2007). However, because the volume stored in basal segments increases with eccentric remodelling, the absolute longitudinal contribution to stroke volume is preserved despite reduced longitudinal function (Carlsson et al. 2007). Thus, the altered relationship between longitudinal function and volume change due to ventricular dilatation may modify the relationship between global LV strain and volume. While the current study provides novel insight into the ability of strain–volume loops to temporally link the mechanical and haemodynamic consequences of severe aortic valvular diseases, the clinical implications could be far-reaching. Early detection of latent systolic and/or diastolic dysfunction may improve clinical decision making in patients with valvular disease. The transition from compensatory LV hypertrophy to pathological disease processes with potentially irreversible consequences exists on a continuum. Currently, no single haemodynamic measurement exhibits adequate sensitivity to indicate severe disease in asymptomatic patients. The temporal characteristics of strain–volume loops may prove useful to link structural remodelling to LV mechanical function in advance of symptom onset. For example, the hypertrophic response to valvular disease is heterogeneous, and the degree of LV fibrosis is correlated with unfavourable postoperative outcomes in aortic valvular disease. This method may allow for discrimination of the transition from compensated to decompensated remodelling to optimize the timing of valve replacement. Further, the authors speculate that LV fibrotic remodelling may account for the functional differences observed between groups in the current study. The presence of extensive LV fibrosis has been linked to reduced longitudinal strain, diastolic dysfunction, and the clinical progression from compensatory hypertrophy to heart failure. Prospectively linking the presence or absence of fibrosis, hypertrophy, or other hallmarks of maladaptive structural remodelling to functional changes and disease severity would significantly enhance the clinical and prognostic value of the strain–volume loop. In conclusion, Hulshof et al. (2017) have developed a novel approach to study the mechanical haemodynamic consequences of severe aortic stenosis and regurgitation on LV function. The discriminative capacity of strain–volume loops offers a potentially promising and advantageous method to study the temporal relationship between structural and functional cardiac remodelling in the context of health and disease. None declared. All authors have approved the final version of the manuscript and agree to be accountable for all aspects of the work. All persons designated as authors qualify for authorship, and all those who qualify for authorship are listed.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.