Use of the SAPIEN 3 Transcatheter Heart Valve in High-Risk Scenarios
The transcatheter aortic valve replacement procedure is used in patients with aortic stenosis. Transcatheter aortic valve replacement devices are quite versatile; thus, they are increasingly being used for nonaortic applications, such as tricuspid valve-in-valve implantation. This case series describes a transcatheter aortic valve replacement procedure in 4 patients with anatomic challenges (eg, aortic tortuosity, high valvular calcium burden, highly calcified bicuspid valve, low coronary artery takeoff, left main coronary artery occlusion, and large aortic annulus) and a fifth patient who had a failed tricuspid bioprosthesis and underwent a tricuspid valve-in-valve implantation with the Edwards SAPIEN 3 transcatheter heart valve (Edwards Lifesciences). All procedures required adjustments to the standard protocol, and each procedure was successful. The critical, technical adjustments in the deployment technique and preprocedural planning of the procedures are detailed to provide a road map for other cardiologists who encounter similar challenges.
Surgical valve replacement or repair has been the standard of care for the treatment of valvular stenosis in patients with the presence of favorable anatomies. Transcatheter aortic valve replacement (TAVR) has emerged as an alternative to surgical valve replacement in patients with variable risk profiles.
In this case report, we describe 4 patients with severe aortic valvular stenosis who underwent TAVR using the Edwards SAPIEN 3 transcatheter heart valve (S3-THV) (Edwards Lifesciences), which is used because of its low profile and refined deployment technique. In addition, we report on a challenging case of a degenerated tricuspid bioprosthesis that was successfully treated with a transcatheter tricuspid valve-in-valve (TViV) procedure, which demonstrates the growing utility of the S3-THV. Patients presented with anatomic challenges in addition to elevated surgical risks (eg, aortic tortuosity, high valvular calcium burden, highly calcified bicuspid valve, low coronary artery takeoff, left main coronary artery [LMCA] occlusion, large aortic annulus, tricuspid intervention, and right ventricular [RV] dysfunction). This case series describes the growing versatility of valve bioprosthesis with specific detail on how to manage treatment in patients presenting with certain challenging anatomies and risk profiles. Deidentified data are available upon request from the corresponding author.
Case Series
All patients underwent transthoracic echocardiography (TTE) or transesophageal echocardiography and a computed tomography (CT) scan per institutional protocol as part of the preprocedural planning for a TAVR or TViV procedure.
Case 1: Tortuous Aorta
Patient 1 is a 78-year-old woman with severe symptomatic aortic stenosis, decompensated chronic heart failure, paroxysmal atrial fibrillation (AF), and hypothyroidism. She was deemed ineligible for surgery by the cardiovascular surgeons because of her overall frailty (Table I). Therefore, she was offered transcatheter implantation on a compassionate use basis.
Baseline TTE revealed severe aortic stenosis (Table I). Mild aortic regurgitation was noted in the presence of a preserved left ventricular ejection fraction (LVEF). The CT scan of the chest revealed the presence of a significant degree of valvular calcification (82.6 mm3, Table II).
The calcium volume score was measured in accordance with the method described by Bettinger and colleagues.1 The calcium volume threshold was set at luminal attenuation + 100 Hounsfield units (HU). Of note, the patient had an extremely tortuous thoracic aorta (Fig. 1A). Based on preprocedural planning, a 26-mm S3-THV was selected to match the annular size, with a 17% annular oversizing.
Citation: Texas Heart Institute Journal 49, 5; 10.14503/THIJ-21-7579
Main vascular access was obtained through the right common femoral artery, and a 14F Edwards Lifesciences sheath was inserted without complications over an extra-stiff Amplatz wire (Boston Scientific) and positioned in the abdominal aorta. Next, the aortic valve was crossed with the standard technique.2 A Safari wire (Boston Scientific) was exchanged, and a 26-mm S3-THV was advanced over a Commander delivery system (Edwards Lifesciences); effective pushability was maintained to easily to cross the native valve despite an S-shaped tortuous aorta. The valve was deployed (Fig. 1B) under rapid pacing.
The postdeployment echocardiogram revealed the presence of a well-seated bioprosthetic aortic valve appearing to be functioning normally with no paravalvular leak. The patient was discharged a few days later without any evidence of aortic or access-related complications. At the 30-day follow-up visit, she showed significant cardiac improvement.
The S3-THV delivery system offers adequate support, and its “pushability” granted an access diameter of greater than 5 mm2.
Case 2: Bicuspid Valve With High Calcification Burden
This case involves an 81-year-old man with a long medical history of coronary artery disease, arterial hypertension, hyperlipidemia, stage 3 chronic kidney disease, paroxysmal AF, and hypothyroidism. During a prior right carotid endarterectomy, the patient experienced a cerebrovascular accident. He received a coronary artery bypass graft in 1997. Specifically, the left internal mammary artery was grafted to the left anterior descending (LAD) coronary artery, a saphenous vein graft (SVG) to the obtuse marginal 1 artery was completed with a jump graft to the first diagonal artery, and another SVG to the posterior descending branch artery was performed. One month before presenting to our center, he was treated for heart failure exacerbation in an outside hospital and received a diagnosis of severe low-flow, low-gradient aortic stenosis with moderate aortic regurgitation. On presentation, his symptoms (worsening dyspnea on exertion, 2-pillow orthopnea, and lower extremity edema) classified his heart failure as New York Heart Association (NYHA) class III. He was considered for TAVR because of his poor functional status and very poor postoperative rehabilitation potential (Table I).
Baseline echocardiography confirmed the presence of severe aortic stenosis (Table I). A CT scan of the chest revealed the presence of a Sievers type 1 bicuspid valve with a left-to-right fusion. In addition, heavily calcified, diffused raphes were noted. Of particular interest, the total calcification of the aortic annulus was measured to 778.4 mm3 (Fig. 2; Table II). The calcium distribution within the valve had a spiral distribution, which placed the patient at high risk for annular rupture. The minimum intercommissural distance on CT was 32.8 mm.
Citation: Texas Heart Institute Journal 49, 5; 10.14503/THIJ-21-7579
A 29-mm S3-THV was selected to match the annular size, with a 9% valvular undersizing; however, because of the large annular area (716 mm2) and intercommissural distance (32.8 mm), an additional 2 mL of contrast were added to the nominal volume of the delivery system balloon. The initial deployment was performed according to the standard femoral approach; however, the postdeployment aortic angiogram demonstrated moderate (2+) aortic insufficiency. Postdilation with an additional 1 mL of contrast was conducted. The following aortic angiogram demonstrated resolution of the paravalvular regurgitation, but complete heart block developed and required the implant of a pacemaker that was later converted to a permanent pacemaker (PPM) after 24 hours of persistent atrioventricular (AV) block. The patient was discharged a few days later and was neurologically intact. A TTE taken 3 months after the TAVR showed normal function of a prosthetic aortic valve without a paravalvular leak.
The bicuspid valve presents multiple challenges. Pre-dilation with a 20-mm balloon and additional postdilation should be anticipated.
Case 3: Large Aortic Annulus
A 76-year-old man with a medical history of chronic AF, hypertension, hyperlipidemia, nonischemic cardiomyopathy (LVEF, 35%), type 2 diabetes mellitus, hypothyroidism, obstructive sleep apnea, morbid obesity, and severe chronic obstructive pulmonary disease presented to our clinic with complaints of worsening dyspnea on exertion, fatigue, and dizziness. His baseline TTE revealed severe aortic stenosis. The Society of Thoracic Surgeons (STS) score of 4.8% placed him at intermediate risk for surgical aortic valve replacement; however, his frailty scores deemed him a patient at high risk (Table I). The CT scan revealed an annular area of 730.8 mm2 and a moderate degree of calcification (562.7 mm3, Table II).
Given this extreme anatomy, imaging analysis, and prior experience, a 29-mm S3-THV was selected, and an additional 3 mL of contrast was added to the nominal volume in the deployment system. The valve was undersized by 11%. The S3-THV implant was performed through a standard femoral route approach. The postdeployment aortic angiogram demonstrated excellent placement of the prosthetic valve and no paravalvular regurgitation, without the need for any postdilation. The patient had no postprocedural complications and was discharged home the following day. At the routine 30-day post-TAVR follow-up, TTE showed a well-seated bioprosthetic aortic valve with normal function.
The S3-THV device performs well even when oversized up to 790 mm2 (+20% increase of the nominal area of 649 mm2).
Case 4: TAVR With a Partial LMCA Occlusion
Patient 4 is a 71-year-old man with a medical history of hypertension, hypothyroidism, and mild chronic obstructive pulmonary disease. Transthoracic echocardiography revealed severe aortic stenosis (Table I). He was considered to be at intermediate risk for surgical aortic valve replacement, but because of his poor functional status, he was considered for TAVR (Table I).
The right and left coronary heights (18.0 mm and 12.8 mm, respectively) did not raise any particular concern for acute LMCA occlusion. However, the left coronary cusp (LCC) leaflet tip was heavily calcified (906.4 mm3). The sinus of Valsalva (SoV) diameters were 35.0 mm (L), 32.4 mm (N), and 31.3 mm (R). Based on the anatomy as outlined above, a 26-mm S3-THV was selected, with 14.7% oversizing.
The valve was deployed according to the standard technique, but postdeployment hemodynamics revealed elevated left ventricular end-diastolic pressure in excess of 40 mm Hg and a new-onset left bundle branch block (Fig. 3). Aortic root angiography confirmed the presence of a high-grade stenosis of the LMCA ostium from the proximity of the native LCC now displaced by the S3-THV device. A 6F JL4 Vista Brite Guiding Catheter (Cordis) was used, and we were able to navigate a 0.014-inch ChoICE Floppy Guide Wire (Boston Scientific) across the LMCA and into the distal LAD artery. The proximal LMCA lesion was predilated with a 3.0- × 12.0-mm Trek compliant balloon (Abbott Vascular); the LMCA stenting was done with a 3.5- × 12.0-mm SYNERGY drug-eluting stent (Boston Scientific) (Fig. 3C). The stent was postdilated with a 3.75 × 12.0-mm noncompliant EMERGE coronary balloon (Boston Scientific) because of significant stent underexpansion seen on intravascular ultrasound (IVUS; OptiCross, Boston Scientific). The patient's hemodynamics and ST-segment depression normalized.
Citation: Texas Heart Institute Journal 49, 5; 10.14503/THIJ-21-7579
The patient was discharged a few days later following an uneventful postoperative stay. At the 30-day follow-up visit, TTE showed a well-seated prosthetic valve in aortic position with normal function and without a valvular or paravalvular leak, and the patient reported significant functional improvement.
The degree of leaflet calcification, as well as the coronary height and SoV diameters, should always be carefully evaluated. In the presence of unfavorable features of these 3 factors (alone or in combination), preemptive coronary wiring should be used.
Case 5: TViV Procedure With Intentional Annular Fracture
The final case involves a 61-year-old woman with rheumatic heart disease. The patient received a mechanical mitral valve replacement in 2013 and underwent bioprosthetic tricuspid valve (29 mm) replacements in 2004 and 2013. Her permanent dual-chamber pacemaker with leads in the right atrium and right ventricle underwent lead extraction and the right side was reimplanted in 2017. The patient previously received radiation therapy for breast cancer, which was in remission on admission. At presentation, she complained of worsening dyspnea on minimal activity, episodic presyncope, and orthopnea. On exam, a faint mid-diastolic rumble on the right side and mechanical click of the mitral valve could be heard. Mild lymphedema of the left chest with mild bilateral ankle edema was noted.
Her last echocardiogram showed an LVEF of 60% to 65% with mean gradient across the tricuspid valve of 10 mm Hg at a heart rate of 62 beats per minute (bpm) (Fig. 4A). Taken together, clinical findings were suggestive of significant tricuspid stenosis. Her STS score (3.6%), previous sternotomies, and prior left-sided radiation therapy for breast cancer also qualified this case as high risk. Thus, a percutaneous TViV procedure was chosen to avoid an additional sternotomy. The patients' anticoagulation therapy could not be halted because of the mechanical mitral valve. Further, there was particular concern that the pacemaker's RV lead could be compromised when the new valve was deployed; therefore, the intervention plan included interrogation of the lead and the decision to jail the RV lead between the previous bioprosthesis and the new S3-THV. A lead extraction/reimplant plan was ready in the case of significant lead failure or fracture.
Citation: Texas Heart Institute Journal 49, 5; 10.14503/THIJ-21-7579
Given the lack of documentation at prior treatment sites and lack of registration of the valve, our team had to rely on CT imaging for identification of the valve type and size. The internal diameter measured 25 mm; thus, we selected a 29-mm S3-THV. After the patient was sedated, the pacemaker was interrogated and the baseline capture thresholds and impedance were measured. Using the standard femoral approach, a Swan-Ganz catheter (Edwards Lifesciences) was positioned to allow a J-wire into the right pulmonary artery. The wire was exchanged for a 0.035-inch STORQ wire to advance a 29-mm S3-THV across the tricuspid valve prosthesis. Given the lack of support, we exchanged the STORQ wire for a 0.035-inch Safari wire with its tip positioned in the RV (Fig. 4B). Under rapid ventricular pacing at 120 bpm, the device was deployed.
Postdeployment visualization showed that the valve was well positioned across the annulus and there were no significant paravalvular leaks; however, a significant waist was present (Fig. 4C). An echocardiogram confirmed a transvalvular gradient of 5 mm Hg at 57 bpm, suggestive of a significant residual transtricuspid gradient. As expected, the RV lead became impinged by the S3-THV struts; however, interrogation of the RV lead showed no increase in impedance and a limited rise in capture threshold. To address the gradient, we fractured the sewing ring of the surgical valve by inflating a 26- × 45-mm True balloon (BD) to 12 atm, and the mean gradient decreased to 1.9 mm Hg (Fig. 4D). Subsequent interrogation of the RV lead confirmed no change in impedance or threshold. The patient was then transferred to the recovery area and discharged within 2 days. The patient's NYHA functional class improved from III/IV to I within 1 week. Normal pacemaker activity was noted alongside the absence of dyspnea on exertion.
Aggressive postdilation and surgical bioprosthetic cracking is an effective strategy to eliminate a residual transtricuspid gradient.
Discussion
With the expanding indications for percutaneous valve replacement, clinicians are confronted with more challenging cases and opportunities. Each case presented herein represents the versatility that the S3-THV now has. Compared with first- and second-generation devices, the S3-THV has improved in terms of a lower sheath/delivery system profile and overall stability of the valve at the time of deployment. Despite the extreme anatomic conditions and calcifications noted in this case series, the S3-THV conformed without any significant paravalvular leak after 1 postdilation step.
Aortic tortuosity and calcified aorta may present additional risk of serious and potentially lethal complications and have been associated with reduced 30-day survival.3 With the advancement of TAVR techniques and valve equipment, the number of aortic complications has decreased. During the deployment step, the optimal stability of the valve under rapid pacing is a result of the differential inflation profile of the S3-THV valve delivery balloon.4,5 This critical step of valve placement at the annulus is now very reproducible.
The bicuspid valve poses a significant challenge that is exacerbated by calcium burden and distribution. Undersizing of the valve can result in paravalvular leaks, while oversizing of the valve risks AV block (via AV-node compression by displaced calcium) or even annular rupture.6 There is still a limited consensus on the definition of high calcium burden and its implications.7,8 The current implant strategy is to limit valve oversizing to 20%,9 and a recent multicenter study reported a cumulative annulus rupture rate of 0.9%10 when less than 20% oversizing was used. In case 2, the combination of high calcium, especially in proximity to the LCC, and extra nominal volume might have contributed to the AV-node compression and consequent AV block requiring a PPM implant. The presenting electrocardiogram in case 2 diagnosed the presence of a preexisting first-degree AV block (PR interval, 216 ms) and right bundle branch block (QRS interval duration, 146 ms); therefore, the AV block may have been related to the underlying condition of the patient's conduction system rather than the excess radial strength applied by the device on the AV node.
There is growing evidence that S3-THV oversizing yields satisfactory results.11 The third case presents a large annular/device mismatch with a resulting 11% valve undersizing, and it demonstrates the feasibility of TAVR with a 730-mm2 annular size. Similarly, Mathur et al published a small series of 3 patients with annuli ranging from 748 to 793 mm2 who were successfully treated by increasing the loading volume of the delivery system (up to 4 mL of contrast in addition to the 33 mL nominal volume of a 29-mm S3-THV).12 Based on our experience and that of others, 12 the S3-THV device performs well at up to 793 mm2, which is 18% oversized. The degree of calcification can alter this boundary by providing additional volume. We avoid oversizing by any valve greater than 20% as a general principle.
Our predeployment imaging of the fourth case identified severe calcification of the LCC leaflet (906.4 mm3), an LCC height of 12.8 mm, and a left SoV of 35 mm. Thus, we oversized the valve (14.7%). The oversizing of more than 10% and the anatomic characteristic may have resulted in significant postdeployment stenosis of the LMCA ostium and required prompt recognition and treatment with IVUS-guided LMCA stenting.
Acute coronary artery occlusion is a relatively rare complication in TAVR and has potentially catastrophic clinical consequences.13 Specific anatomic features have been shown to predispose patients to this eventuality (right coronary ostial height <10 mm and left coronary ostial height <10 mm).7 Based on our institutional experience, occlusion can occur in the absence of these specific features because of the dynamic interplay among coronary height, degree of calcification of the LCC (>200 mm3 HU), and size of SoV (<4 mm of the valve size). In the presence of high-risk features, the preemptive LMCA protection technique should be considered. The sole presence of the wire and delivery system in distal LAD may protect the ostium of the LMCA from being partially covered by the LCC once the valve is deployed. Also, it obviates the necessity of wiring an occluded LMCA in an unstable patient and allows LMCA stenting by pulling back the stent from the LAD to the ostium of the LMCA.13
Although the S3-THV has been used in nonaortic applications, the use of the off-label TViV approach to treat bioprosthetic valve regurgitation is not yet fully validated.14 Based on well-established practice, valve fracturing is performed after transcatheter heart valve deployment to further decrease gradients. Use of the TViV approach and valve-in-ring implantation for a failed tricuspid bioprosthesis is particularly appealing because of the S3-THV's safety profile, success rate,15,16 and reported (although limited) favorable outcomes.17,18 Compared with the surgical risk associated with a repeated tricuspid surgery, the TViV procedure offers an advantage because of the RV dysfunction that is frequently associated with a failed tricuspid prosthesis.19–21
We elected to use an S3-THV 29-mm valve in case 5 because of the internal valve diameter (25 mm), prior surgical reports, and radiologic appearance and features. We also selected the 29-mm valve (as opposed to the 26-mm valve) because we anticipated a significant residual gradient that would require postdilation. Postdilation and intentional fracture of the bioprosthesis were necessary to minimize the transvalvular gradient. After the deployment of a 29-mm S3-THV showed underexpansion, Hensey and colleagues fractured the bioprosthesis valve in a separate procedure to improve valve expansion and documented a reduction in the residual gradient from 4 to 1 mm Hg.22 Thus, we deployed and fractured the valve in a concomitant procedure, which resulted in a final transtricuspid gradient of less than 2 mm Hg. Given the patient's history of previous RV lead extraction and the need for anticoagulation therapy with her mechanical mitral valve, the decision to jail the RV lead between the surgical bioprosthesis and the S3-THV was made. Even though the lead was impinged, its function remained normal.
In conclusion, device size selection and careful radiologic, angiographic, and anatomic planning remain the mainstay for procedural success. In addition, the S3-THV device is particularly suitable in challenging anatomy because the low profile (14F), ease in deployment, and skirt design, which ensure successful and effective deployment. The expanding utility of the device to treat nonaortic valves opens new therapeutic opportunities but requires even more stringent planning and careful consideration.
A) Computed tomography 3-dimensional reconstruction for case 1. The abdominal and thoracic aorta with complete loop in the abdominal and the thoracic portion of the vessel in addition to a steep vertical annular aortic plane are shown. B) Fluoroscopy image of postdeployment of heart valve in case 1. Extreme aortic tortuosity is evident from the loops described by the supporting 0.035-inch wire.
Computed tomography reconstruction and calcium analysis of an aortic valve using Hounsfield units revealed an extreme calcium volume of 778.4 mm3 in case 2.
A) Pre– and B) post–left ventricular and aortic pressure with relative transaortic gradient in case 4. Note that the resolution of the left ventricle–aorta gradient in panel B is associated with a reduced systolic pressure and marked increase in left ventricular end-diastolic pressure (from 20 mm Hg in panel A to 40 mm Hg in panel B) as a result of deployment of the Edwards SAPIEN 3 transcatheter heart valve and secondary high-degree left main coronary artery stenosis. C) Coronary angiogram of the left coronary after deployment of valve in case 4. Note the close proximity of the heart valve's struts with the ostium of the left
A) A Doppler image of the significant bioprosthetic tricuspid valve dysfunction in case 5. B) The fluoroscopic image of the Edwards SAPIEN 3 transcatheter heart valve predeployment position in relationship to the bioprosthetic annulus and right ventricle lead in case 5. C) A fluoroscopic image of the deployed heart valve in case 5. Note the circumferential constraint around the new valve. A residual 5-mm Hg mean gradient was measured, which required postdilation to intentionally fracture the bioprosthetic valve. D) Doppler imaging of the valve's final gradient after postdilation shows a marked reduction in the transtricuspid gradient (1 mm Hg) with significant bioprosthetic tricuspid valve dysfunction in case 5.
Contributor Notes
