Newer echocardiographic techniques for aortic-valve imaging: Clinical aids today, clinical practice tomorrow.

Increasing life expectancy is expected to lead to a corresponding increase in the prevalence of aortic valve disease (AVD). Further, the number of indications for transcatheter aortic valve replacement (TAVR) as a treatment option for AVD is expanding, with a growing role for echocardiography in its management. In this review we summarize the current literature on some newer echocardiographic modalities and the parameters they generate, with a particular focus on their prognostic and clinical value beyond conventional methods in the management of aortic stenosis, TAVR, and aortic regurgitation. Speckle tracking and 3D echocardiography are now increasingly being used in the management of AVD. For instance, global longitudinal strain, the best-studied speckle tracking echocardiographic parameter, can detect subtle subclinical cardiac dysfunction in patients with AVD that is not apparent using traditional echocardiographic techniques. The emerging technique of 3D full volume color Doppler echocardiography provides more accurate measurement of the severity of aortic regurgitation than 2D-proximal isovelocity surface area. These novel techniques are promising for evaluating and risk stratifying patients to optimize surgical interventions, predict recovery, and improve clinical outcomes.

Core tip: Reduced strain is now established for early diagnosis, prognosis, and risk stratification, predicting post-op recovery and showing associations with mortality. Decreased global longitudinal strain (GLS) is a robust parameter to diagnose subclinical left ventricular dysfunction before the left ventricular ejection fraction deteriorates and the patient develops symptoms. GLS also correlates with disease severity and helps to identify patients with excess risk of cardiovascular events and death who are likely to benefit from earlier surgical intervention. The high accuracy and reproducibly of 3D echocardiography has made the precise assessment of volume and AR possible and, therefore, the early recognition of its severity.

INTRODUCTION

The prevalence of aortic valve disease (AVD) continues to rise in line with increasing life expectancy. Aortic stenosis (AS) is the most common valvular heart disease, affecting 12.4% of patients over 75 years in North America and Europe[1], while the incidence of aortic regurgitation (AR) increases with age and affects 4%-5% of the population overall[2]. The introduction of minimally invasive techniques such as transcatheter aortic valve replacement (TAVR) has revolutionized the management of AVD. Current AVD management mainly represents a “horse has bolted” approach since symptoms have already developed or the left ventricle is damaged, and consequently the medical and socio-economic impact is greater than if the disease was managed earlier in its course or prevented. Here we discuss the role of newer echocardiographic techniques in the management of AS and AR. For each, we first discuss the current guidelines before outlining the modalities that are expected to change clinical practice in the future.

AORTIC STENOSIS

In developed countries, AS is most commonly caused by calcific degenerative valve disease, with congenitally abnormal valves (commonly bicuspid) with superimposed calcification and rheumatic valve disease being less common causes[3]. With advances in technologies, AS is increasingly recognized as a complex disease with different patient subgroups and pathophysiologies. In order to individualize treatment, accurate and timely sub classification is important.

Current guidelines and classification

According to American College of Cardiology (ACC)/American Heart Association (AHA) 2014 guidelines[3], severe AS is defined as a peak aortic velocity of > 4 m/s, which corresponds to a mean aortic valve gradient of > 40 mmHg and calculated valve area of 1.0 cm2 or less[3]. Valvular heart disease guidelines from major societies like the ACC/AHA (2014)[3], European Society of Cardiology (ESC; 2012)[4], and Canadian Cardiovascular Society (2004)[5] recommend aortic valve replacement (AVR) as a class I indication in symptomatic patients with severe AS, asymptomatic patients with severe AS and left ventricular ejection fraction (LVEF) < 50%, and asymptomatic patients with severe AS who are undergoing other cardiac surgery.

Aortic stenosis leads to pressure overload in the left ventricle (LV). To maintain cardiac output, there is compensatory LV hypertrophy, which, although seemingly beneficial, is in fact maladaptive and results in cardiac fibrosis, heart failure, and ultimately increased mortality[6]. Over the last decade, new perspectives on the pathophysiology of AS have resulted in a new, four flow gradient pattern classification system that challenges the previous misconception that patients with AS and normal ejection fraction (EF) inevitably have normal flow. Patients are first divided based on left ventricular flow state-normal flow (NF) vs low flow (LF), where low flow is defined as LV stroke volume of < 35 mL/m2[7], and then based on pressure gradient-low gradient (LG) vs high gradient (HG), where low gradient is defined as mean trans-aortic pressure gradient of < 40 mmHg. This makes the four categories NF/LG, NF/HG, LF/LG, and LF/HG[7-10], as shown in Table 1.

Table 1

Prevalence[8,9], prognosis[10] and percentage of patients undergoing surgery[8] for severe aortic stenosis and preserved ejection fraction

	High gradient (mean > 40 mmHg)	Low gradient (mean < 40 mmHg)
Normal flow (SV > 35 mL/m2)
Prevalence	30%-62.7%	15.3%-38%
Prognosis (2 yr survival rates)	44% ± 6%	83% ± 6% (best prognosis)
% Undergoing surgery	80% (highest rates of surgery)	53%
Low flow (SV < 35 mL/m2)
Prevalence	8%-13.2%	8.8%-24%
Prognosis (2 yr survival rates)	30% ± 12%	27% ± 13%-worst prognosis
% Undergoing surgery	68%	36%-lowest rates of surgery

The LF/LG subgroup occurs in patients with low forward stroke volume and is associated with a worse prognosis and higher mortality than patients with high gradient severe AS[11]. Despite this, this group was noted to undergo the lowest rates of surgical intervention of the four categories[8]. Prognostic stratification using the above categories is therefore essential to prevent delay in timely intervention.

It is also important to distinguish pseudo-AS from true AS in the LF/LG subgroup. Pseudo-AS is seen in patients with mild to moderate AS due to incorrect calculation of lower aortic valve area (AVA) from poor forward flow causing incomplete valve opening, an inherent pitfall in the continuity equation. Therefore, ACC/AHA guidelines for assessing valvular heart disease recommend low dose (up to 20 mg/kg per minute) dobutamine stress echocardiography (DSE) to distinguish true from pseudo-AS in patients with LF/LG AS and decreased LVEF (< 40%)[3]. In pseudo-AS, gradient increases will correspond to proportionate increases in AVA, whereas in true AS, the AVA will remain diminished (AVA ≤ 1 cm2) even with progressive dobutamine dosage. An absence of contractile reserve, i.e., a failure to increase stroke volume by 20%, also indicates a worse prognosis. However, DSE’s high dependence on flow response results in high inter-patient variability, especially in those with restrictive physiology. Mal-alignment of the Doppler signal, erroneous measurement of LVOT diameter, poor quality images due to severe calcification in and around the AV, and extremes of body size (i.e., very low body surface area or morbidly obese patients) challenge the reliability of DSE.

Impaired LVEF in AS is associated with higher operative mortality[12], worse long-term prognosis[13], and less than 50% of patients recover to normal LVEF following AV replacement[12]. Even in the presence of symptoms, the majority of patients with severe AS have a normal EF[14]. Therefore, more sensitive parameters have been developed to identify patients likely to benefit from surgical treatment despite a preserved EF.

Speckle tracking echocardiography

Impaired myocardial contractility often precedes the observable decrease in LVEF[6]. Early markers and predictors of myocardial dysfunction would therefore be highly desirable to recognize the subset of patients that would benefit from early intervention. A newer technique for strain imaging called speckle-tracking echocardiography (STE) is highly reproducible, angle-independent, and more sensitive for detecting global and regional LV wall function than routine echocardiographic indices like tissue Doppler imaging and conventional imaging techniques like multi-slice computed tomography (MSCT) and cardiac MRI (cMR)[15,16].

LV strain essentially assesses changes in myocardial fiber length relative to their resting phase. Strain rate indicates the speed at which this deformation occurs. An echocardiographic image is a grayscale image derived from the speckles produced by ultrasound waves scattered by body tissue. STE measures myocardial deformation (strain) based on speckle displacement and assesses the parameters of strain, strain rate, twist/torsion, and rotation (see definitions, Table 2 and Figure 1). We summarize current clinical data supporting the use of STE parameters in the prognostication and management of AS below.

Table 2

Definitions of basic strain parameters

Parameter	Definition
Strain	Change in myocardial fiber length relative to its resting phase
Global longitudinal strain	Percentage change in LV fiber length in the longitudinal axis
Global circumferential strain	Percentage change in LV circumference in the short axis view
Global radial strain	Percentage change in LV wall thickness in the short axis view
Strain rate	Rate at which change in myocardial fiber length relative to its resting phase occurs
Twist/Torsion/Rotation	Myofiber geometry in the LV myocardium changes from a right-handed helix in the subendocardium to a left-handed helix in the subepicardium, and this results in twisting during systole with the apex rotating counterclockwise and the base in a clockwise direction

LV: Left ventricle.

Figure 1

The different planes of strain.

Longitudinal strain: Longitudinal strain refers to the percentage change in LV fiber length in the longitudinal axis. Global longitudinal strain (GLS) is the most studied parameter to date, and it reflects contraction of longitudinally arranged subendocardial fibers. Severe AS hinders myocardial perfusion, particularly in the sub-endocardium, due to higher wall stress and impaired coronary blood flow[17]. Hence, decreased longitudinal shortening is the first impairment seen in patients with AS[18]. As the severity of AS increases, GLS decreases[19].

As the ventricle is non-spheroidal, pressure increases lead to the basal and mid LV segments being exposed to higher wall stress than the apical segments[20] according to Laplace’s law, as the heart base has a larger curvature and flatter contour compared to the middle and apical segments[21,22]. As would be expected, these segments respond better after unloading with AVR compared to the apical segments.

In a retrospective study of 395 patients with moderate to severe AS and preserved EF, GLS was an independent predictor of mortality[23]; this was subsequently confirmed in a prospective study of 142 patients[24]. GLS also shows incremental prognostic benefit over traditional parameters like AV gradient, stroke volume index[8,25], and valvuloarterial impedance[26,27]. Therefore, GLS can be used to detect subtle subclinical dysfunction that may not be apparent using current standard echocardiographic parameters. Lee et al[28] reported a higher percentage of two-year cardiac events (re-admission for heart failure or death) in asymptomatic severe AS patients with impaired GLS (-16.5%; 77% sensitivity and 67% specificity) who were managed conservatively. This suggests that incorporating GLS into current risk models might further optimize timing for AVR. In two prospective studies with severe AS and preserved EF (103 patients[29] and 340 patients[9], respectively), GLS was noted to be particularly worse in patients with LF AS compared to patients with NF AS, again supporting the hypothesis that the low flow state represents a more advanced form of the disease.

In asymptomatic moderate and severe AS patients with preserved LVEF, reduced GLS worse than -15% was a significant risk factor for cardiac hospitalization, AVR, cardiovascular death[30,31], all-cause mortality at one year[24], and, interestingly, increased risk of post-operative atrial fibrillation independent of left atrial size and age[32]. Other studies have established that regional deformation analysis is also important, with an additional parameter called basal longitudinal strain (BLS) less than -13% associated with adverse outcomes including heart failure, MI, all-cause mortality, and symptomatic status in asymptomatic severe AS. In fact, BLS has the strongest association with symptomatic status of all the longitudinal strain parameters[33]. Kempny et al[34] studied 101 patients undergoing TAVR and showed that pre-echocardiographic assessment of GLS is associated with post-operative symptom improvement.

Circumferential strain: The percentage change in LV circumference in the short axis view is called circumferential strain, which reflects contraction of the circumferentially arranged mid-layer fibers. Following a decrease in longitudinal strain, circumferential fibers compensate for the loss in longitudinal function. Hence, impairment of both longitudinal and circumferential strain suggests more extensive myocardial damage. Global circumferential strain (GCS) is considered an important prognostic factor in patients with symptomatic AS, and conservatively treated patients with impaired GCS have higher two-year all cause mortality and re-admission for heart failure than those who do not[28].

Radial strain: The percentage change in wall thickness in the short axis view is known as the radial strain. As noted above, impaired subendocardial perfusion is an early feature of AS due to increased wall stress and impaired coronary blood flow. Impaired endocardial radial strain is observed in patients with AS and preserved LVEF and correlates with AS severity; epicardial radial strain is, however, preserved[35]. Hyodo et al[36] suggested the use of a novel parameter called the bilayer ratio (subendocardial to subepicardial strain ratio), the value of which decreases as the severity of AS increases given the decrease in subendocardial thickness due to ischemia and compensatory increase in subepicardial thickness. This was deemed to be superior to longitudinal strain, which only takes the subendocardium into account.

Twist/torsion and rotation: Myofiber geometry in the LV myocardium changes from a right-handed helix in the subendocardium to a left-handed helix in the subepicardium. This configuration results in twisting during systole, with the apex rotating counterclockwise and the base in a clockwise direction. When both layers contract simultaneously, a larger radius of rotation for the outer epicardial myofiber layer results in mechanical predominance of the epicardial fibers in the overall direction of rotation. LV twist is proportionate to the severity of AS[37], which is thought to be a consequence of subendocardial dysfunction reducing inhibition of longitudinal muscle fibers. Conversely, untwisting rate is delayed and decreased[38].

Santoro et al[39] also observed an increase in LV twist in LV hypertrophy with preserved EF and suggested it as a marker of early systolic dysfunction. The LV twist to circumferential shortening ratio (TSR) is also considered a reliable marker of subendocardial dysfunction[40], and van Dalen et al[37] demonstrated an increase in TSR in AS. Apical rotation (ApRot) is also known to increase in severe AS[38,39]. Some studies have shown that ApRot is positively associated with the presence of symptoms[41], while others[42] have found that patients with low ApRot have higher rates of syncopal events, a similar rate of overall symptoms, and that increased ApRot is an independent predictor of mortality in severe AS with preserved EF. Asymptomatic patients with severe AS and increased ApRot have similar survival to symptomatic patients, but the measurement is useful for identifying patients who might benefit from early evaluation for aortic valve replacement (AVR)[42].

3D strain

The widespread emergence of 3D echocardiography has also introduced the possibility of measuring 3D strain (3D-STE). 2D-STE assumes geometric LV morphology, needs multiple, high-quality image acquisition, and there is the possibility of mistracking speckles that move out of the scanning plane. 3D-STE, using the block-matching method in full-volume datasets, has been developed to overcome these shortcomings, and it tracks the 3D motion of the acoustic speckles but is independent of speckles moving out of the scanning plane and free of geometric presumptions. 3D-STE represents a more accurate model than 2D-STE because all strain parameters are simultaneously obtained from one volume image[43] in contrast to 2D-STE, in which the long and short axes are measured at different points in time.

As expected, compared to LVEF, 3D-STE is a more sensitive and accurate assessment of early LV dysfunction[44]. In a study of 104 asymptomatic patients with severe AS and preserved EF, 3D-GLS, 3D-GRS, and 2D-GLS were all found to be useful predictors of major adverse cardiac events (MACE). 3D-GLS was also an independent predictor of MACE after correcting for LV mass index and mean pressure gradient[45]. However, poor temporal resolution and the need for a high frame rate hamper its current widespread use. Further prospective studies utilizing a 3D-GLS-guided approach are imperative to assess its role in predicting adverse cardiovascular outcomes.

Treatment of AS: The role for newer imaging modalities

In addition to playing a role in early diagnostics and patient stratification, newer imaging modalities like MSCT, cMR, and 3D trans-esophageal echocardiography are increasingly being used as part of the treatment of AS. AS is only treatable by valve replacement, traditionally surgically (SAVR) but now also with balloon-expandable or self-expanding valves via TAVR. With the recent United States Food and Drug Administration (FDA) approval of the Sapien XT and Sapien 3 transcatheter heart valves in patients at intermediate risk for open-heart surgery, TAVR will increasingly be used beyond only high-risk and inoperable cases. Pre-procedural, intra-procedural, and post-procedural echocardiography are recommended as part of TAVR evaluation[46]. These are described in detail below (Figure 2).

Figure 2

Ratio of left ventricle twist to circumferential shortening. GLS: Global longitudinal strain; GCS: Global circumferential strain; TSR: The LV twist to circumferential shortening ratio; AS: Aortic stenosis.

Pre-TAVR, transthoracic echocardiography (TTE) helps to determine the degree of stenosis, valvular cusp size, motion, the location and amount of calcification, the aortic root and annular anatomy, mitral valve pathology, the presence of a bulging inter-ventricular septum at the level of the aortic root, baseline LV and RV function, and pulmonary artery pressure[6]. With modern echocardiograms, Doppler echocardiographic interrogation of aortic valve gradients is superior and preferred to other imaging modalities, although misalignment of the Doppler signal with the AS jet of over 20 degrees and the pressure recovery phenomenon in the ascending aorta diameter < 30 mm are known limitations.

The effective orifice area is determined using the left ventricular outflow tract (LVOT) diameter on TTE or transesophageal echocardiography (TEE). It is important to measure this 1-2 mm apical to the aortic annulus, and a difference of > 2 mm between the aortic annulus and LVOT should prompt repeat measurement[6]. A basal inter-ventricular septal bulge may also lead to inaccurate measurements and superior displacement of the deployed valve, so this also needs pre-operative determination[6]. On the other hand, a thin septum is also important to recognize pre-operatively to avoid post-operative ventricular septal rupture.

It is well established that multi-slice computer tomography (MSCT) overestimates the LVOT by 20% and 3D echocardiogram underestimates by 20% given the elliptical shape of the LVOT. Aortic valve area (AVA) measured by CT for severe AS needs a higher cut-off (1.2 cm2) compared to AVA measured by echocardiography (1 cm2)[47]. It is challenging to obtain the appropriate cross-sectional view by 2D-TEE due to the movement of the aortic annulus along the long axis and its tilting movement during the cardiac cycle. Volumetric 3D-TEE overcomes this problem by incorporating the entire AV valve. As anticipated, in a prospective study of 60 patients, Nakai et al[48] showed that 3D-TEE is more accurate that 2D-TEE for calculating the AVA. Further, 3D-TEE provides a more accurate reconstruction of the aortic root and better measurement of the distance between the annulus and coronary ostia, most importantly the left coronary ostium.

One of the most important roles for echocardiography is to assist in choosing the appropriate size of prosthetic heart valve. Undersizing of the prosthetic causes paravalvular aortic regurgitation (PVR), while oversizing of artificial valves, especially in the setting of a calcified LVOT, is associated with higher risk of aortic rupture and periaortic hematoma as the native calcific valves are retained behind the artificial valve. Measurement of LVOT diameter by TTE has long been used and validated for calculation of the aortic valve area and AS severity, but TTE measurements should not be relied upon to decide the prosthetic valve size. There is an ongoing discussion about which method is most accurate and reproducible for valve sizing; Tsang et al[49] compared cMR, MSCT, and 3D-TEE in vivo and in vitro and showed that cMR was most accurate and reproducible. Several studies have suggested that MSCT is superior to 3D-TEE[49,50], while others[51] have noted good agreement between annulus perimeter and the area measured by 3D-TEE and MSCT. Altiok et al[52] also showed highly consistent measurements of sagittal and coronal diameters on 3D-TEE and MSCT.

The distribution and extent of calcification should also be determined preoperatively by echocardiography as it is a useful predictor of procedural success. It is also crucial to identify patients with obliteration of the sinus of Valsalva (SOV) or smaller or shorter SOV height, as these patients require shorter prosthetic valves. 3D-TEE has also emerged as a valuable tool (comparable to MSCT) for accurately measuring left main coronary artery to annulus distance and length of coronary cusp[53].

Intra-TAVR, fluoroscopy is advocated by the 2012 ACCF/AATS/SCAI/STS expert consensus document[54] on TAVR regardless of type of access. However, in a prospective study of 100 patients undergoing transapical transcatheter aortic valve implantation, Bagur et al[55] noted similar acute and 30-d outcomes for patients managed intra-operatively with angiography and TEE. Current American and European guidelines advocate the use of 2D and 3D-TEE support during TAVR.

The first step of the TAVR procedure is to place the pacing wire in the right ventricle, with echocardiography used to confirm its position and exclude perforation of the ventricle or pericardial effusion. Next, a stiff wire is placed into the left ventricle, with echocardiography ensuring its stability at the apex and lack of entanglement with mitral apparatus and again excluding perforation and pericardial effusion. The aortic root must be continually visualized during balloon aortic valvuloplasty, after which coronary artery patency is established, LV wall motion assessed, and position of the calcified coronary leaflets noted. The TAVR valve is then introduced. It is important to ensure that the native leaflets are covered by the TAVR valve. In trans-apical valve placement, the puncture site is visualized. It is also important to ensure angulation of the valve away from the inter-ventricular septum and the right ventricle.

TEE during TAVR is usually performed under general anesthesia, which requires intubation and hence increases the patient’s risk profile. TTE seems to be a reasonable alternative to TEE but factors against TTE include relatively poor image quality, inability to position the patient, and potential compromise of operative field sterility with the trans-apical or trans-aortic approach. Intracardiac echocardiography (ICE) can overcome some of these issues, and ICE has been used in the closure of inter-atrial defects and in some electrophysiological interventions. For TAVR monitoring, the ICE catheter is introduced through the femoral vein and advanced to the superior cavo-right atrial junction. Apart from TR jet velocity, ICE allows placement of guide wires and catheters carrying the valve, valve deployment, and continuous monitoring for pericardial effusion. However, it has yet to be optimized for accurately assessing paravalvular regurgitation (PVR) given its narrow sector angle and poor resolution. Further, ICE requires venous access, which carries a risk of hematoma and bleeding complications. Since the ICE catheter shares the same RV space, there is a risk of dislodging the pacemaker leads used for rapid pacing during valve deployment[56]. 3D-ICE imaging currently has a limited field of view of only 22°, making the measurement of annular size difficult.

Post-TAVR, valve deployment, position, shape, and leaflet motion must be confirmed. Then, hemodynamic measurements need to be performed to assess valve function. Effective orifice area, Doppler velocity index, and mean and peak trans-valvular gradients should be measured. Effective orifice area is determined using the LVOT diameter, which is measured from the outer to outer stent diameter at the lower edge or inner to inner stent diameter at the upper edge if the valve is low. LV and RV function and pulmonary artery pressures are also noted as a part of the post-operative evaluation.

TEE and TTE are routinely used in clinical practice to detect immediate post-operative complications like malposition, valvular regurgitation or PVR, mitral valve damage, aorta to right atrium fistulae, cardiac shunts secondary to inter-ventricular septal damage, pericardial effusion, and cardiac tamponade following free wall or annular rupture, coronary artery patency, and left ventricular wall akinesis due to inadvertent coronary ostial closure. More than a moderate degree of PVR is clearly associated with increased short- and long-term mortality[57]; however, data on mild PVR and outcomes are conflicting. The exact incidence of PVR varies widely across studies due to the differences in the parameters and criteria used to grade PVR; different schemes used to classify PVR severity; and a lack of standard assessments for PVR. To address this issue, a classification of PVR has recently been proposed that divides severity into five categories: trace, mild, mild-to-moderate; moderate; moderate-to-severe, and severe[25].

Doppler is the gold standard for evaluating PVR. Both TTE and TEE may be required, as the PVR jets located posteriorly and anteriorly are often shadowed in TTE and TEE views, respectively, as a result of the shadowing caused by native aortic valve calcifications and the prosthetic stent. It is always important to use color Doppler with echocardiography in both the long and short axis views. Further, jets must be quantified in terms of number, width, path, and convergence. The 2012 Valve Academic Research Consortium (VARC) 2 defines moderate PVR as the circumferential extent of the PVR estimated in the parasternal short axis at 10%-30% and severe PVR at > 30%[58]. Tiny para-valvular jets usually regress spontaneously over a period of 10-15 min[59], as do those appearing with self-expandable balloon aortic valves as the frame expands[59]. Therefore, it is important to wait for a while before intervening. With multiple jets, eccentricity, presence of calcification, and changing loading conditions, the accurate assessment of PVR severity can be difficult using conventional echo methods. 3D-TTE has been shown to be superior for assessing PVR utilizing 3D vena contracta and 3D regurgitant volumes[60].

TTE can be used to determine prosthesis location in the long axis view compared to LVOT location. In the short axis view, TTE can determine if the proper circular shape has been assumed. There is a now a trend toward using only TTE in appropriately selected patients. In a retrospective study of 111 patients, Sengupta et al[61] demonstrated a significant difference in the procedural time with non-inferiority in terms of procedural success, extent of PVR, additional valve implantation, and complications such as peri-procedural stroke rate or death. However, prospective data supporting TTE use during TAVR are still lacking.

Echocardiography should be performed prior to discharge (and after 30 d) to establish a new baseline of replaced valve function including mean transaortic gradient, valve area, and PVR. As data on the long-term functioning of TAVR are not robust, annual TTE follow up to assess valvular and ventricular function should be undertaken.

Post TAVR cardiac remodeling – the role for imaging

Several studies have reported reverse remodeling (improvement in strain parameters) in the minutes, 72 h, and month following TAVR[36,62-64]. In a prospective study, Swan et al[65] demonstrated immediate improvement (within minutes) in circumferential and radial strain following TAVR. Kim et al[66] conducted a multilayer strain study and demonstrated a significant improvement in longitudinal but not circumferential strain following TAVR as early as one week. A study of 68 LF LG severe AS patients showed significant improvements in GLS at 6 and 12 months after TAVR[67]. Interestingly, post TAVR changes in strain pattern do not appear to be influenced by pacemaker-induced rhythm or post-procedure new left bundle branch block[64]. Kempny et al[34] reported a correlation between improvement in longitudinal strain and symptomatic improvement following TAVR, while Løgstrup et al[68] (n = 100, mean EuroScore: 10.5 ± 2.8) noted a correlation between improvement in GLS and decrease in mortality rate following TAVR. Poulin et al[69] (n = 102 patients) reported that improvements in longitudinal systolic and diastolic deformation were significantly lower in patients with prosthesis-patient mismatch at follow-up.

AORTIC REGURGITATION

The most common causes of AR in developed countries are aortic root dilation, calcific valve disease, and bicuspid aortic valve, with rheumatic AR the less common etiology. Less common causes include infective endocarditis and aortic dissection. The inability of valve leaflets to remain coapted during diastole leads to blood flow back into the left ventricle. This leads to increase in end-diastolic volume and elevated wall stress, eventually leading to compensatory eccentric hypertrophy from volume overload. In contrast to AS (where both pressure and volume overload occurs), this eccentric LV change predominantly affects the circumferentially arranged fibers leading to more severe impairment in GCS as compared to GLS.

Current recommendations

ACC/AHA 2014 guidelines define severe AR as a Doppler jet width ≥ 65% of LVOT, vena contracta > 0.6 cm, regurgitant volume ≥ 60 mL/beat, regurgitant fraction of ≥ 50%, and effective orifice area of ≥ 0.3 cm2[3]. Guidelines from the ACC/AHA (2014)[3], European Society of Cardiology (ESC; 2012)[4], and Canadian Cardiovascular Society (2004)[5] on the management of valvular heart disease recommend AVR as a class I indication in symptomatic patients with severe AR, asymptomatic patients with severe AR and LVEF of < 50%, and patients with severe AR who are undergoing other cardiac surgery. AVR is also recommended as a Class IIa indication for patients with severe AR, normal LVEF ≥ 50% but severe LV dilation [LV end-systolic diameter (LVESD) of > 50 mm], or indexed LVESD of > 25 mm/m2. It is also reasonable to pursue AVR in patients with moderate AR undergoing cardiac surgery. However, severe dilation and decreased LVEF represent the late stage of the disease, and other parameters to identify subtle LV dysfunction early in the course of disease are desired.

Speckle tracking echocardiography in AR

STE has also been used in the assessment of AR. Stefani et al[70] examined 60 patients including young athletes with bicuspid aortic valves and mild AR with matched controls and found a reduced longitudinal peak systolic strain in LV basal segments. A single center retrospective study of 314 patients with chronic moderate to severe AR noted that global longitudinal strain was independently predictive of mortality (at a threshold of -12.5%)[71]. Di Salvo et al[72] showed that patients with moderate to severe AR with a progressive pattern and symptom development had significantly reduced longitudinal strain compared to those with stable disease (-17.8% ± 3.9% vs -22.7% ± 2.7%, P = 0.001) despite having similar LVEFs. In a longitudinal study of 64 patients with moderate to severe AR, Olsen et al[73] noted that patients with reduced GLS, strain rate, and early diastolic strain were more likely to have progressive disease (symptom development or worsening LVEF) in patients managed conservatively and also poorer outcomes following surgery. In a study of 90 patients, Iida et al[74] demonstrated that the LVEF may continue to be normal despite a significant lowering of longitudinal strain due to compensatory increases in wall thickening in the subepicardium.

Gabriel et al[75] reported a correlation between reduced longitudinal strain rate and subclinical LV dysfunction on exercise echocardiography in patients with moderate to severe AR. Similarly, Marciniak et al[76] also found that longitudinal and radial strain rates were a sensitive indicator of subclinical dysfunction in severe AR. Interestingly, Onishi et al[77] noted that LV radial systolic strain rate was predictive of LVEF post surgery. In moderate to severe AR with preserved EF, Ewe et al[78] noted impairment in all three strains (longitudinal, circumferential, and radial) in symptomatic compared to asymptomatic patients. Also, impaired baseline GLS (per 1% decrease, HR = 1.21, P = 0.04) or GCS (per 1% decrease, HR= 1.22, P = 0.04) was predictive of the need for surgery in asymptomatic patients. Li et al[44] noted that circumferential strain is a more sensitive marker for AR/volume overload compared to longitudinal strain for AS/pressure overload. Similarly, Broch et al[79] showed that in asymptomatic patients with moderate to severe AR and preserved LVEF, GCS was higher and thought to contribute to reduced GLS.

Decreased LV apical rotation and torsion have been detected in moderate to severe AR with preserved EF compared to normal healthy subjects[80], while others[81] have noted lower LV torsion in the severe AR group compared to the moderate AR group but no difference in the apical rotation compared to moderate AR and control patients. Finally, in a prospective study by Chen et al[82] for severe AR and reduced EF, there were significant improvements in GLS and GCS at three months following AVR when measured using 3D-STE.

3D echocardiography

Although many different 2D echocardiography parameters can be used to quantify AR, it is still challenging. Not only are there subtle differences in the scan plane with altered size of the vena contracta jet, but also its shape is not always circular but sometimes irregular or ellipsoid. 3D-echocardiography can be used to directly measure the vena contracta area (3D-VCA). Sato et al[83] showed that the 3D-VCA was 32 mm2 in severe AR (sensitivity: 89%; specificity: 98%). 2D-derived proximal isovelocity surface area (PISA) and regurgitant volume (Rvol) suffer from geometric assumptions, angle correction, and problems with assessing multiple jets. Choi et al[84] reported 3D full volume color Doppler echocardiography to be more accurate then 2D-PISA for assessment of all AR grades, especially with eccentric or multiple jets. Perez de Isla et al[85] reported that 3D color Doppler echocardiography was both accurate and reproducible for AR evaluation with a high correlation with gold standard cMR. Real-time 3D-TEE was also able to reveal specific anatomical differences between type I (annular dilation) and type II (prolapsed) AR[86].

CLINICAL APPLICATION OF EARLY DIAGNOSTICS

Reduced strain is now established for early diagnosis, prognosis, and risk stratification, predicting post-op recovery and showing associations with mortality. Decreased GLS is a robust parameter to diagnose subclinical LV dysfunction before the LVEF deteriorates and the patient develops symptoms. GLS also correlates with disease severity and helps to identify patients with excess risk of cardiovascular events and death who are likely to benefit from earlier surgical intervention. The high accuracy and reproducibly of 3D echocardiography has made the precise assessment of volume and AR possible and, therefore, the early recognition of its severity.

CONCLUSION

Newer echocardiographic techniques have significantly enhanced our understanding of the pathophysiology of aortic valve diseases and are being increasingly employed to assess disease severity and to risk-stratify patients. Early detection of LV dysfunction is paramount in these patients to allow timely intervention before irreversible impairment occurs. Novel non-invasive imaging techniques such as speckle tracking echocardiography and 3D imaging have shown promise in the quantification of subclinical myocardial damage in vivo. With development of higher processing power echocardiograms, newer generation probes, and with increasing familiarity and experience with these techniques, these are likely to become the new norm.