Parameter
Mild
Moderate
Severe
Valve structure and motion
Mechanical or bioprosthetic
Usually normal
Usually abnormal
Usually abnormal
Structural parameters
Left ventricular size
Normal
Normal/mildly dilated
Dilated
Doppler parameters (qualitative or semi-quantitative)
Jet width in central jets (% LVO diameter): colora
Narrow (≤25 %)
Intermediate (26–64 %)
Large (≥65 %)
Jet density: CW Doppler
Incomplete or faint
Dense
Dense
Jet deceleration rate (PHT, ms): CW Dopplerb
Slow (>500)
Variable (200–500)
Steep (<200)
LV outflow vs. pulmonary flow: PW Doppler
Slightly increased
Intermediate
Greatly increased
Diastolic flow reversal in the descending aorta:
PW Doppler
Absent or brief early diastolic
Intermediate
Prominent, holodiastolic
Circumferential extent of paraprosthetic AR (%)c
<10
10–20
>20
Doppler parameters (quantitative)
Regurgitant volume (mL/beat)
<30
30–59
>60
Regurgitant fraction (%)
<30
30–50
>50
Table 14.2
VARC-2 definitions for PVR
Prosthetic aortic valve regurgitation | |||
---|---|---|---|
Mild | Moderate | Severe | |
Semi-quantitative parameters | |||
Diastolic flow reversal in the descending aorta – PW | Absent or brief early diastolic | Intermediate | Prominent, holodiastolic |
Circumferential extent of prosthetic valve paravalvular regurgitation (%) | <10 % | 10–29 % | ≥30 % |
Quantitative parameters | |||
Regurgitant volume (ml/beat) | <30 ml | 30–59 ml | ≥60 ml |
Regurgitant fraction (%) | <30 % | 30–49 % | ≥50 % |
EROA (cm2) | 0.10 cm2 | 0.10–0.29 cm2 | ≥0.30 cm2 |
The strength of the VARC definition may be that it’s generally inclusive and acknowledges the fact that not one single parameter is sufficient for determining the severity of PVR. Important new parameters, such as the extent of PVR across the circumference of the annulus, expressed as a percentage, to the use of RVOT/LVOT regurgitant fraction, are incorporated to traditional Doppler colour flow as width of jet, and extend of jet into the ventricle.
Given the lack of standardization in previous reports, it remains to be defined the natural history of PVR in TAVR. There is likely inter-observer variability in non-core lab reported PVR. It’s unclear whether PVR generally worsens over time [93], what medication may retard its progress [89], or whether it tends to improve as the previously hypertrophied left ventricle remodels after the aortic stenosis is relieved.
The temporal trend for PVR is important to guide whether TAVR PVR needs to be percutaneously treated or whether medical therapy may suffice. There have been various reports of percutaneous delivery of off label vascular plug into severe PVR to seal of defects between the THV and aortic annulus [94–97]. The observational data remains limited but generally demonstrated an improvement in PVR severity.
Case Presentations and Imaging Considerations
Case 1: Aortic Stenosis Echocardiography
Figure 14.1a, b show TTE with good image quality can provide important and accurate information of aortic valvar apparatus including LVOT diameter, sinus height, sinotubular junction dimension and degree of calcification. Figure 14.1c shows critical aortic stenosis with a transaortic Doppler velocity of 7.5 m/s. In Fig. 14.1d, e, accurate calculation of aortic valve area relies on accurate determination of LVOT diameter; calcification in the RCC as reflected on the parasternal short axis as well. Whilst TTE with good image quality can provide important information on calcium burden, MSCT remains the gold standard.
Fig. 14.1
Aortic stenosis echocardiography. (a) Sinotubular height, (b) annular and sinotubular dimensions, (c) CW of the aortic valve, (d) LVOT measurement, and (e) short axis of the aortic valve
Case 2: Dobutamine Stress Echocardiography
In aortic stenosis with impaired left ventricular systolic function, the transaortic gradient may be underestimated due to poor ventricular function or that the aortic stenosis is indeed less than severe. In this case, the use of dobutamine demonstrated contractile reserve (23 % increase in SV) as well as moderate AS only with the increased ventricular function pushing open the aortic leaflets better (Fig. 14.2).
Fig. 14.2
Dobutamine stress echocardiography. From top to bottom; at baseline, 5 mcg, 10 mcg, and 20 mcg of dobutamine. Note how the stroke volume, gradient, and aortic valve area are increasing. Note how the aortic valve on 2D echocardiography appears to be opening further with increased flow
Case 3: TTE vs. MSCT
Figure 14.3 shows the spatial resolution of MSCT is far superior to TTE. This is particular obvious in calcium quantification. Additionally, the ovoid nature of the aortic annulus is well appreciated in MSCT (Fig. 14.3e vs. Fig. 14.3c). Note suggestion of in the aortomitral continuity well demonstrated on MSCT (Fig. 14.3a vs. Fig. 14.3d). Calcium is likely an important area of research for the TAVR technology.
Fig. 14.3
(a–e) TTE vs. MSCT of the aortic valve apparatus. (a) and (c) TTE, and (b), (d), (e) MSCT
Case 4: 3mensio Workup
Commercially available 3D reconstruction computer program, such as this 3mensio example (3mensio®, Pie Medical Imaging BV, Netherlands) shown in Fig. 14.4, vastly improves annular plane determination, thus allowing coaxial assessment of the aortic valvular structure. Unlike traditional method without 3D reconstruction, which required many steps of adjustment focusing on each aortic sinus, 3D reconstruction allows the determination of a centerline (yellow line in Fig. 14.4a, b). Once this line is achieved, in Fig. 14.4c, the tangential plane, the cross hair can be spun around the annulus, ensuring that this particular plane transects the bottom of each sinus (Fig. 14.4a, b), the so-called “spinning the line” technique. Once this is achieved, in Fig. 14.4d, a favorable angle for the operators can be selected for implantation angle, to ensure coaxial placement of the aortic root on fluoroscopy.
Fig. 14.4
(a–d) 3mensio workup for identifying and measuring the aortic annulus
Case 5: THV Sizing Dilemma
In Fig. 14.5a, MSCT reconstruction yielded an aortic annulus area of 4.13 mm2, with moderate eccentricity but a focal calcium, extending into a severe aorto-mitral continuity calcification. TTE yielded an LVOT diameter of 22 mm. Borderline for a 23 mm Edwards SAPIEN THV but given the calcification this THV was chosen. In Fig. 14.5b, 23 mm THV placed with severe focal paravalvular regurgitation, likely due to calcification. A second THV with further postdilatation did not significant mitigate PVR. A 10 × 5 mm vascular plug (arrow) was placed reducing PVR to one fourth (Fig. 14.5d) from four fourths (Fig. 14.5c).
Fig. 14.5
(a–d) THV sizing dilemma with PVR and need for a vascular plug
Case 6: MSCT Pitfalls
Figure 14.6 shows examples of poorly performed MSCT. Figure 14.6a shows contrast poorly captured in left ventricular outflow tract and most contrast density in descending aorta (Fig. 14.6c). To obtain an accurate aortic annulus measurement sufficient contrast is needed in the left ventricle. In Fig. 14.6d the double shadow in the outline of the aorta and aortic annulus, likely to be patient movement artifact. This significantly compromises the accuracy of annulus measurement.
Fig. 14.6
(a–d) MSCT pitfalls with poor annular contrast opacification on MSCT
Case 7: Bicuspid Aortic Valve
A 68 year old with previous coronary bypass grafts adherent to his sternum has now developed severe symptomatic bicuspid aortic stenosis. Figure 14.7a shows bicuspid aortic valve with moderate leaflet tip calcification – the presence of calcification is important to ensure satisfactory anchoring of THV. Figure 14.7b shows MSCT annular determination (area 5.41 cm2; perimeter 84.1 mm). Figure 14.7c–e show difficult to determine annular plane given bicuspid nature, MSCT demonstrating no prohibitive sinotubular junction or aortic root dilatation. Figure 14.7f shows successful transapical implant of a 29 mm Edwards Sapien XT ® THV. Figure 14.7g shows D4 post THV TTE showing trivial paravalvular regurgitation (blue jet).
Fig. 14.7
(a–g) Bicuspid aortic valve (a–e) 3mensio, (f) TAVR implantation, and (g) echocardiography
Case 8: Vasculature Reconstruction
Figure 14.8a shows conventional angiography of ilio-femoral axis, with a marker pigtail with 10 mm interval markers. Figure 14.8b shows MSCT 3D rendered image of vascularture providing information of tortuosity and calcification not attainable from angiography. Figure 14.8c shows proprietary program such as 3mension providing phantom “straightened” artery and caliber of a 18Fr sheath (see yellow straight line).
Fig. 14.8
(a–c) Vasculature reconstruction with 3mensio
Case 9: Edwards and CoreValve Fluoroscopy
Edwards (upper panel) and CoreValve (lower panel) fluoroscopy revealing excellent, acceptable, and poor implantation position (left, middle, and right, respectively) (Fig. 14.9).
Fig. 14.9
Edwards (top panels) and CoreValve (bottom panels) fluoroscopy revealing excellent (left panels), adequate (middle panels), and poor (right panels)
Future Perspective
Magnetic resonance imaging – thus far there is little data on the use of MRI in TAVR planning. There is some MRI data on the aortic root dynamics on systole compared to diastole. The most promising data may be the volumetric assessment of the regurgitant flow possible with MRI [92].
Paravalvular regurgitation – whilst second generation THVs such as Lotus (SADRA) or Edwards 3 seem to have significant reduced the incidence of moderate to severe PVR, much research is still needed in this area. The temporal pattern to PVR, the individual patient factors contributing to an adverse outcome, and the role of medications in treating PVR are but some of the areas to be defined. A most pressing area is standardization and validation of PVR. Quantification of PVR remains challenging, and despite the expert consensus of VARC2 definition, the later still needs to be formally validated. Mitigation of PVR is an important prerequisite for the extending the TAVR to a lower risk group of patients.
MSCT – better understanding of calcification – how best to decide THV choice based on calcium
Conclusion
TAVR has transformed the management of aortic stenosis. Two large RCTs comparing two different THV platform to SAVR and many ongoing registries have demonstrated the importance of this technology and in the foreseeable future, superiority to SAVR, particularly in intermediate to high risk subset. Without doubt with improvement in each iteration of the technology it is likely TAVR will one day become the dominant treatment option for aortic stenosis. Concurrent with this TAVR development has been the enormous improvement in understanding of the complex three dimensional aortic valve apparatus through advancement in imaging modalties particularly through the use of three dimensional imaging such as MSCT. The same lessons learnt from TAVR are now extending to other structural heart disease interventions such as left atrial appendage closure devices. It would not be an exaggeration to claim that TAVR is now equivalent if not superior to SAVR in high risk patients mostly due to mitigation of issues related to imaging techniques and THV design and size algorithm over the years. Imaging, particularly MSCT, will likely play a pivotal role in enhancing THV design and probably will contribute to individualized THV choice in the future based on specific aortic annulus size, calcification, aorto-ventricular angulation and other as yet to be defined important parameters. It therefore behoves the interventional cardiologist and cardiac surgeon to enhance their imaging knowledge through close collaboration with the imaging specialist in the heart team. The issues and techniques highlighted in this chapter will surely evolve, be improved and hopefully in doing so this will propel this disruptive technology to be the forefront of aortic stenosis treatment.
References
1.
2.
3.
4.
5.
6.
Turina J, et al. Spontaneous course of aortic valve disease. Eur Heart J. 1987;8(5):471–83.PubMed
7.
Joint Task Force on the Management of Valvular Heart Disease of the European Society of Cardiology, et al. Guidelines on the management of valvular heart disease (version 2012). Eur Heart J. 2012;33(19):2451–96.CrossRef
8.
Nishimura RA, et al. 2014 AHA/ACC guideline for the management of patients with valvular heart disease: a report of the American College of Cardiology/American Heart Association Task Force on practice guidelines. Circulation. 2014;129(23):e521–643.PubMedCrossRef