A Case Study on Implantation Strategies to Mitigate Coronary Obstruction in a Patient Receiving Transcatheter Aortic Valve Replacement

Abstract

Coronary obstruction is a life threatening complication during and post-transcatheter aortic valve replacement(TAVR). The objective of this preliminary work is to investigate the mechanisms underlying coronary obstruction in an expired patient after TAVR, in whom coronary obstruction was detected as the cause of death in addition to highlighting the importance of pre-procedural planning. Thus, the sensitivity of coronary obstruction with respect to TAV type, size, and positioning needs to be investigated. The aortic root of an 80-year old male patient who expired due to coronary obstruction during TAVR–where a 29mm SAPIEN 3 was deployed-was segmented from Computed Tomography scans and 3D-printed with compliant material. Flow and pressure data were acquired using a pulse duplicator under physiological conditions. Measurements were acquired using a 29mm SAPIEN3, a 26mm SAPIEN3 expanded with a 29mm balloon, and a 31mm Medtronic-CoreValve deployed annularly, supra and sub-annularly. Only the CoreValve in sub-annular axial position and the 29 S3 yielded pressure gradients(PG)lower than 10mmHg(6.76±0.52 and 5.72±0.13mmHg respectively)while the 26S3, CoreValve in normal and supra-annular positions yielded higher PGs(15.5±0.48, 12.2±0.15 and 10.8±0.24mmHg respectively). 29mm SAPIEN3 implantation yielded an FFR value of 45.7±0.6%. However, 31mm CoreValve in any of the 3 different annular positions yielded FFR values going from 89.6±1.1% in supra-annular position to 98.3±1.1% in sub-annular. Implantation with a 26mm SAPIEN3 expanded with a 29mm-balloon also yielded an FFR of 92.1±1.2%. Coronary obstruction in this patient could have been prevented through usage of different valve types and/or through usage of a different combination of valve size-balloon sizes.

Introduction

Transcatheter aortic valve (TAV) replacement (TAVR) emerged as an alternative to surgical valve replacement procedure that is highly invasive for high-risk patients³. Despite being percutaneous and less invasive, TAVR is associated with several adverse effects such as elevated pressure gradients, regurgitation, subclinical leaflet thrombosis and coronary obstruction³. Particularly, coronary obstruction is a life threatening⁶ complication during and post-transcatheter aortic valve replacement, that can occur when the TAV stent pushes the calcified native leaflets to partially or totally occlude the coronary ostium⁹. Prevention of this adverse effect is indispensable. While coronary ostium height and sinus of Valsalva size have emerged as predictive of the complication⁷, there is a need to better investigate the sensitivity of coronary obstruction with respect to TAV type, size, and positioning to help bring TAVR benefits to patients at risk. The objective of this preliminary work is to investigate the mechanisms underlying coronary obstruction in an expired patient after TAVR, in whom coronary obstruction was detected as the cause of death in addition to highlighting the importance of pre-procedural planning. In addition, this study aims to assess whether different deployment heights could have avoided the fatal coronary obstruction seen in this patient. The overarching goal of this study is to introduce pre-procedural patient-specific analysis as a pre-requisite for every patient case at high-risk of coronary obstruction.

Methods

From an institutional review board (IRB) approved database, the aortic root of an 80-year-old male patient who expired due to coronary obstruction during TAVR (Fig. 1 and Video 1) was selected. Contrast in Fig. 1 and Video 1 was injected in the left coronary main to highlight the obstruction. The aortic root was segmented at the 70% phase from pre-operation Computed Tomography (CT) scans using anatomic modeling software (Mimics, Materialise, Belgium) and 3D-printed with compliant material using a Stratasys Connex Printer as shown in Fig. 2. Cusp calcification was generated from the patient-specific CT scans using rigid print material VeroWhite clear (Stratasys, Farmington Hills, Mich) and soft tissue structures were replicated using a rubber-like material (TangoPlus FLX930, Stratasys). Further Methodology details can be found in previous publications⁵. The patient’s mean pressure gradient (PG) was 43mmHg and the aortic valve area was calculated to be 0.56cm². He was treated with a 29mm SAPIEN3 TAV which immediately led to coronary obstruction and subsequent mortality. Pre-TAVR hemodynamics (Cardiac output 4L/min; mean aortic pressure 71mmHg, heart rate 81 beats per minute) were re-created through the 3D-printed aortic root with a pulse duplicator left heart simulator which included a physiological left coronary circuit as previously described⁴. Briefly, the left coronary flow is controlled by a Starling resistor that was collapsed or expanded during specific intervals to match the changes in coronary flow during myocardial contraction or relaxation respectively. Sixty cycles of aortic and left coronary flow waveforms and pressure waveform data were acquired at 100Hz. In order to explore novel strategies to mitigate the risk of coronary obstruction, measurements were repeated following implantation of (a) Edwards 29mm S3; (b) 26mm S3 expanded with a 29mm balloon; and (c) Medtronic CoreValve 31 placed in normal, supra-annular and sub-annular positions. Fractional flow reserve (FFR) was computed directly from measured coronary flow rate. FFR measures the maximum myocardial blood flow in the presence of a vessel obstruction as a percentage of the maximum flow in the case of a completely normal and unobstructed vessel.

Fig.1:

Coronary obstruction at the time of the procedure.

Fig.2:

Representative 3D printed patient-specific aortic root model.

Results

Fig. 3 shows the results of PGs and FFRs of the different valves. Only the CoreValve in sub-annular axial position and the 29 S3 yielded PGs lower than 10mmHg (6.76±0.52 and 5.72±0.13mmHg respectively) while the 26 S6, CoreValve in normal and supra-annular positions yielded higher PGs (15.5±0.48, 12.2±0.15 and 10.8±0.24mmHg respectively).

Fig.3:

Pressure Gradient and FFR results for the different valves.

SAPIEN 3 29mm implantation yielded an FFR value of 45.7±0.6%. However, 31mm CoreValve in any of the 3 different annular positions yielded FFR values going from 89.6±1.1% in supra-annular position to 98.3±1.1% in sub-annular. Implantation with a 26mm S3 expanded with a 29mm balloon also yielded an FFR of 92.1±1.2%.

Discussion

It is expected to obtain the lowest PG with the 29 S3 compared to the 31 CoreValve as the SAPIEN 3 balloon-expandable characteristic allows for more expansion upon implantation thus a higher orifice area.

Clinically, an ideal FFR is supposed to be 100%. However, the FFR range of 75%–80% is considered a gray zone, in which clinical judgment must complement quantitative assessments in forming the final treatment decision, thus the threshold adopted for this study is 75%⁸. Any value below 75% is considered to be indicating a probable inducible ischemia. The experiment recapitulated the patient’s complication following the SAPIEN 3 29mm implantation with an FFR of 45.7±0.6%.

Despite being the appropriate TAV to be selected for this patient, the larger height of the 29mm S3 frame may have contributed to the occlusion. The favorable FFR with 26 S3 with a 29mm balloon implantation can be attributed to the TAV frame height shortening, significantly altering the apposition of the native leaflet thereby avoiding the occlusion. The self-expanding nature combined with the stent gaps characterizing the CoreValve stent, independently from axial position, may explain the improved FFR compared to 29mm S3. CoreValve supra-annular deployment yielded the lowest FFR compared to the normal and sub-annular positions. This may be attributed to several factors. One factor may be because of a higher radial force concentrated on the lower diamonds of the CoreValve stent. Deeper implants spare the native leaflets from stronger radial forces, explaining the decrease in likelihood of obstructing the coronary ostium. Another factor may be the space available for blood to flow towards the coronary ostium with deeper implants. The deeper the valve is implanted with respect to the sinotubular junction, the less likely that it will obstruct blood flow towards the ostium and more likely that the aortic sinus vortex will propagate towards the sinus entraining blood flow towards the coronary ostium⁴. Figure 4 summarizes these different deployment cases and their impact on the flow in the coronary ostium.

Fig.4:

Schematic showing the 5 different TAV cases in relation with the coronary ostium with (a) SAPIEN 3 29, (b) SAPIEN 3 26 expanded with a 29mm balloon, (c) CoreValve in supra-annular position, (d) CoreValve in normal position and (e) CoreValve in sub-annular position. The images of Medtronic CoreValve and Edwards SAPIEN 3 are taken from Refs ¹ and ² respectively.

In this study, a fatal case of coronary obstruction that occurred in a patient was not only re-created in-vitro, but also could have been avoided by pre-procedural planning. The results of this study would generate new interest in pre-procedural biomechanical analysis for patients that may be at high risk of coronary obstruction. Knowing that patient-specific analysis is currently highly required, including it for intra-procedural guidance and planning is not only beneficial for a case like the one described in this study, but also for future clinical therapy. Overall, fatal outcomes such as the case highlighted in this study will be avoided when new knowledge and insights are provided.

Limitations

In this study, the CT scans were segmented in diastole. As the aortic root undergoes changes throughout the cardiac cycle, segmenting it at this particular phase is one of the limitations of this study. In addition, this study is a preliminary work with just one sample that may not be applicable to other patients or geometries.

Conclusion

Coronary obstruction potential is not always easy to be judged with traditional clinical means leading to unexpected risks once the TAV is deployed despite respecting the annulus anatomical dimensions. In this specific in-vitro study case, it was demonstrated that coronary obstruction could have been mitigated with careful selection of another valve type or same valve type with a smaller size combined with an appropriate balloon. Overall, pre-procedural patient-specific planning is important to avoid such complications.

Supplementary Material

1

Video 1: Coronary obstruction fluoroscopy video at the time of TAVR procedure. Contrast is injected from the left coronary main.