Online Exclusive

Successful Percutaneous Management of Partial Avulsion of the Native Aortic Valve Complex Complicating Transcatheter Aortic Valve Replacement

Jeffrey Zimmet, MD, PhD1;  Erhard Kaiser, MD3;  Elaine Tseng, MD2;  Kendrick Shunk, MD, PhD

Jeffrey Zimmet, MD, PhD1;  Erhard Kaiser, MD3;  Elaine Tseng, MD2;  Kendrick Shunk, MD, PhD

Abstract: This is the first reported case of partial aortic valve and annulus rupture during transcatheter aortic valve replacement, and its successful management through percutaneous means. It stresses the fact that even severe procedural complications can often be treated by a heart team endovascularly, without requiring sternotomy. 

J INVASIVE CARDIOL 2014;26(10):E137-E140

Key words: balloon aortic valvuloplasty, aortic valve stenosis


An 83-year-old male military veteran with a history of lymphoma presented with severe symptomatic aortic valve stenosis. Despite an STS score of 3%, he was deemed a non-candidate for surgical aortic valve replacement by two cardiac surgeons due primarily to mantle cell lymphoma under ongoing treatment. For the lymphoma, he had initially received 6 cycles of cyclophosphamide, vincristine, doxorubicine, and dexamethasone (CVAD) chemotherapy, followed by a relapse and then an additional 4 cycles of bendamustine/rituximab and was on maintenance rituximab at the time of evaluation; his 5-year survival associated with this lymphoma was estimated by his oncologists at ≤50%. Transcatheter aortic valve replacement (TAVR) was recommended by the heart team.

Preprocedure transthoracic (TTE) and transesophageal echocardiography (TEE) were performed and showed normal  left ventricular systolic function, with an ejection fraction of 65%. The aortic valve was trileaflet, with moderate calcification and severe restriction of leaflet opening. The mean gradient across the valve was 42 mm Hg and aortic insufficiency was trivial. The annulus diameter was measured by TEE at 25 mm and the annular area by computed tomography (CT) scan at 522 mm2 (Figure 1). The aortic and iliofemoral anatomy was suitable for a transfemoral approach with minimal calcification and sufficient lumen. For procedural planning, the aortic root anatomy and valve anatomy were segmented using a preprocedural three-dimensional (3D) CT dataset. A 3D Dyna-CT dataset (Siemens) obtained intraprocedurally with rapid pacing at 180 bpm was used to further analyze the anatomy and select the optimal deployment angle. 

The left groin was accessed surgically by traditional cut-down, and a 24 Fr Edwards Retroflex3 sheath was inserted over a Lunderquist wire without difficulty following serial dilatation. A 6 Fr pigtail catheter was inserted from the right femoral artery, and placed in the right coronary cusp. From the right femoral vein, a balloon-tipped 5 Fr temporary pacing catheter was placed in the right ventricular apex and capture was confirmed at low threshold.

The aortic valve was crossed without difficulty using a 6 Fr AL1 diagnostic catheter (Boston Scientific) and a straight 0.035˝ guidewire (Cook Medical). This wire was exchanged for a 260 cm Amplatz Extra-Stiff guidewire with a custom-shaped tip per standard TAVR practice. TEE and fluoroscopy showed the wire to be biased toward the posterior aspect of the annulus in the region of the commissure between right coronary and non-coronary cusps, as expected.

A 23 mm x 4 cm valvuloplasty balloon (Edwards Lifesciences) was advanced to the aortic valve. The balloon crossed easily on the second approach after mild resistance was encountered on the initial approach as the tip of the balloon catheter encountered the valve. Rapid pacing was initiated at 180 bpm, with appropriate drop in blood pressure and pulse pressure, and valvuloplasty was performed in typical fashion. Upon discontinuation of rapid pacing, immediate hypotension with a diastolic pressure in the 20s was present, requiring support with multiple vasopressors. TEE demonstrated “wide-open” aortic regurgitation, and a hypermobile linear echodense peduncular structure was observed with focal attachment near the sinotubular junction (Figure 2). 

This pedunculated structure extended into the ascending aorta during systole, and prolapsed into the ventricle during diastole (Figure 3).

TEE evaluation showed no pericardial effusion, and no evidence of dissection, hematoma, or false lumen in the ascending aorta, arch, or descending thoracic aorta. A 26 mm Edwards Sapien valve was rapidly crimped, and successfully deployed in appropriate position under rapid pacing. TEE showed adequate positioning of the prosthesis, but with continued severe aortic regurgitation, primarily central, and of variable appearance as the pedunculated structure prolapsed into the prosthetic valve with each diastole and prevented complete closure of the prosthetic leaflets. Preparation was made for conversion to open surgery. However, the patient had regained a modicum of hemodynamic stability with inotropic support and volume administration, so sternotomy was deferred while additional data were obtained and alternatives were considered. TEE evaluation once again showed no evidence for dissection beyond the sinotubular junction. A brief pacing test under TEE visualization demonstrated that the peduncular structure ceased prolapsing into the ventricle and remained in the antegrade position in the ascending aorta during rapid pacing. Based on this information, a 49 mm Palmaz XL stent (Cordis Corporation) was crimped onto a 30 mm Z-Med balloon (NuMed, Inc) and prepared for deployment in the ascending aorta, with the intent of trapping the pedunculated structure against the aortic wall to prevent it from interfering with prosthetic valve function. The Palmaz stent was deployed during rapid pacing, but only temporarily trapped the mobile tissue structure due to incomplete stent apposition (Figure 4).

The Z-Med balloon was removed, and a 32 mm Coda balloon (Cook Medical) was used to fully appose the Palmaz stent to the ascending aortic wall proximal to the brachiocephalic artery.

The echodensity continued to be highly mobile, prolapsing through the newly-implanted Sapien valve and causing marked valve dysfunction. 

At this point, with the ascending aorta stented, the risk of creating an ascending aortic dissection that might propagate over the arch was serendipitously no longer present, so we determined to attempt snare removal of the pedunculated mass.  An Atrieve 27-45 mm vascular snare (Angiotech) was advanced through a 7 Fr AL1 guide catheter via the right femoral artery and used to capture and restrain the mass.  With gentle traction on it, prolapse through the valve and central AI were no longer present by TEE. By applying moderate further traction on the snare, the structure was removed from its point of attachment near the sinotubular junction and withdrawn from the body via cut-down at the right femoral artery due to its size (Figure 5).  

Gross examination by the cardiac surgeon (ET) suggested two contiguous leaflets with calcific nodules. Subsequent gross and microscopic pathologic examination with H&E staining confirmed leaflet tissue with fibrocalcific degeneration. Final TEE showed no dissection or pericardial effusion, but persistent substantial perivalvular leak. Postdilatation was performed, with a modest reduction in paravalvular leak and no further central aortic insufficiency. The patient was transferred to the intensive care unit with acceptable hemodynamics, weaned from his vasopressors, and extubated.   

CT of the chest the following day showed a well-positioned Sapien valve.  The Palmaz stent was well-apposed in the ascending aorta, without evidence for dissection. Paravalvular leak was mild at follow-up echocardiogram. At 2 months of follow-up, he was asymptomatic and reported doing 80 push-ups each morning.


Severe AI has been reported to occur following balloon aortic valvuloplasty (BAV) in approximately 1% of cases.1,2 Prior to the development of TAVR, the development of severe AI after valvuloplasty had catastrophic consequences in most cases.3,4 Rapid deployment of a transcatheter aortic valve is generally effective at resolving acute BAV-induced AI.5-7 In the current case, the mechanism for severe AI was partial avulsion of the native aortic valve and partial annulus rupture, resulting in a large mobile tissue mass, consisting of annular and valve leaflet tissue, that interfered with proper functioning of the implanted Sapien valve during diastole.  Percutaneous management in this case involved first an attempt to jail the hypermobile structure between stent and aortic root wall. This attempt failed due to undersizing on our part (30 mm delivery balloon) of the stent relative to the actual diameter of the aorta in that location, but after deploying the stent adequately at 32 mm further up in the ascending aorta, the stent serendipitously provided a level of insurance against the possibility of subsequent endovascular manipulations producing an ascending aortic dissection capable of propagating into the arch. Subsequently, the partially avulsed tissue was able to be safely snared and removed, restoring normal function of the Sapien valve. Paravalvular leak was greater than anticipated, likely related to the removal of some annular tissue connecting the two retrieved leaflets prior to valve implantation, but improved after postdilatation and became only mild by the following day as assessed by TEE. The relatively large annulus size relative to the implant may explain some of the persistent paravalvular leak. It has to be stressed that intraprocedural TEE was invaluable in localizing the hypermobile structure, positioning the Palmaz stent just above the coronaries, and finally snaring the avulsed annulus and valve tissue.

The mechanism of partial valvular rupture in this case is unknown.8 Perforation of a valve leaflet by the initial wire crossing is possible, but relatively unlikely given the use of a soft-tipped non-hydrophilic wire, proper preprocedural planning, and uncomplicated valve crossing. Valvuloplasty was performed with adequate rapid pacing and a lack of balloon pistoning during inflation. There was only minor subvalvular and left ventricular outflow tract calcification and the implanted Sapien prosthetic valve was only slightly oversized. Whether the history of lymphoma and subsequent chemotherapy played any role in weakening the tissues and predisposing to tear along an unusual plane rather than along the commissures is unknown.

Trapping the hypermobile structure with a Palmaz stent was unsuccessful in this situation because we slightly undersized the Palmaz stent relative to the size of the aortic root at this location.  Trapping it by using a valve-in-valve strategy was considered, but rejected as unlikely to be effective due to the considerable length of the pedunculated mass.9

This is the first reported case of partial aortic valve and annulus rupture during TAVR, and its successful management through percutaneous means. This case stresses the fact that even severe procedural complications can often be treated by a heart team endovascularly without requiring sternotomy.10,11


  1. Percutaneous balloon aortic valvuloplasty. acute and 30-day follow-up results in 674 patients from the NHLBI balloon valvuloplasty registry. Circulation. 1991;84(6):2383-2397. 
  2. Ben-Dor I, Pichard AD, Satler LF, et al. Complications and outcome of balloon aortic valvuloplasty in high-risk or inoperable patients. JACC Cardiovasc Interv. 2010;3(11):1150-1156. 
  3. Safian RD, Berman AD, Diver DJ, et al. Balloon aortic valvuloplasty in 170 consecutive patients. N Engl J Med. 1988;319(3):125-130. 
  4. Ussia GP, Sarkar K, Tamburino C. Aortic valve perforation during aortic valvuloplasty: Identification and strategies for prevention. Catheter Cardiovasc Interv. 2011;77(6):876-880. 
  5. Genereux P, Head SJ, Van Mieghem NM, et al. Clinical outcomes after transcatheter aortic valve replacement using valve academic research consortium definitions: a weighted meta-analysis of 3,519 patients from 16 studies. J Am Coll Cardiol. 2012;59(25):2317-2326. 
  6. Leon MB, Smith CR, Mack M, et al. Transcatheter aortic-valve implantation for aortic stenosis in patients who cannot undergo surgery. N Engl J Med. 2010;363(17):1597-1607. 
  7. Smith CR, Leon MB, Mack MJ, et al. Transcatheter versus surgical aortic-valve replacement in high-risk patients. N Engl J Med. 2011;364(23):2187-2198. 
  8. Russ C, Hopf R, Hirsch S, et al. Simulation of transcatheter aortic valve implantation under consideration of leaflet calcification. Conf Proc IEEE Eng Med Biol Soc. 2013;2013:711-714.
  9. Yu Y, Vallely M, Ng MK. Valve-in-valve implantation for aortic annular rupture complicating transcatheter aortic valve replacement (TAVR). J Invasive Cardiol. 2013;25(8):409-410.
  10. Eggebrecht H, Mehta RH, Kahlert P, et al. Emergent cardiac surgery during transcatheter aortic valve implantation (TAVI): insights from the Edwards SAPIEN Aortic Bioprosthesis European Outcome (SOURCE) registry. EuroIntervention. 2013 Nov 12 (Epub ahead of print).
  11. Barbanti M, Yang TH, Rodès Cabau J, et al. Anatomical and procedural features associated with aortic root rupture during balloon-expandable transcatheter aortic valve replacement. Circulation. 2013;128(3):244-253.


From 1the Department of Medicine and 2Department of Surgery, University of California San Francisco and San Francisco Veterans Affairs Medical Center, San Francisco, California; and 3Kardiologische Privatarztpraxis, Frankfurt am Main, Germany.

Disclosure: The authors have completed and returned the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Zimmet reports no disclosures. Dr Kaiser reports personal fees from Edwards Lifesciences. Dr Tseng reports grants from University of California and the Coulter Foundation. Dr Shunk reports a grant from Siemens Medical Systems and consulting fees from TransAortic Medical Inc. 

Manuscript submitted March 6, 2014, provisional acceptance given March 28, 2014, final version accepted April 14, 2014.

Address for correspondence: Jeffrey Zimmet, MD, PhD, San Francisco VA Medical Center, 4150 Clement Street, Cardiology 111C, San Francisco, CA 94121. Email: