Abstract: Aims. Mitral regurgitation (MR) is a complication that may occur during transcatheter aortic valve implantation (TAVI) in a certain percentage of cases and may require different treatments depending on the mechanism. Our purpose was to describe the occurrence rate of this complication during TAVI with the CoreValve prosthesis, as well as to assess the usefulness of transesophageal echocardiogram (TEE) in the detection of the mechanism of MR. Methods and Results. We analyzed a total of 129 cases of severe aortic stenosis treated with CoreValve prosthesis from June 2008 to October 2011. We defined a significant MR after TAVI as grade III MR or higher, considering either the new appearance of MR or the worsening of a preexisting MR, as assessed by both TEE and angiography. In our series, there was a total of 11 cases of significant MR after TAVI (8.5%). Angiography detected 100% of the MR cases, but was unable to determine the mechanism of MR in any case. TEE, on the other hand, determined 100% of the MR cases, and determined that 1 case was caused by mechanical asynchrony due to a new left bundle branch block, 3 cases were due to an aortic prosthesis impingement on the anterior mitral leaflet, 2 cases were due to the appearance of a systolic anterior movement of the anterior mitral leaflet with dynamic obstruction of the left ventricular outflow tract, 1 case was caused by a commissural tearing of the valve, and 4 cases were explained by a “functional” mechanism, probably due to transient damage of the subvalvular mitral apparatus by the delivery system. All cases had an MR grade II or less as evidenced by transthoracic echocardiography at discharge. Surgery was not required in any case. Knowledge of the mechanism of MR made it possible to provide the best treatment option in each case. Conclusion. There is a certain percentage of patients treated with CoreValve prosthesis who develop significant MR during the procedure. TEE, unlike angiography, can define the very diverse mechanisms of MR in 100% of cases, and elucidates the best approach to this complication. Surgery was not required in any case.

J INVASIVE CARDIOL 2014;26(11):603-608

Key words: cardiac imaging, transesophageal echocardiography, TEE

__________________________________

Percutaneous aortic prostheses are a valid option in the treatment of severe symptomatic aortic stenosis for patients at high surgical risk.¹ Without doubt, the role of echocardiography is relevant in the proper selection of patients to be treated using this type of prosthesis,² as well as in the follow-up after valve implantation. The European and American Societies of Echocardiography³ have recently published recommendations regarding the use of ultrasound during transcatheter aortic valve implantation (TAVI), highlighting an important role for this technique in detecting complications during such procedures. 

One of the commonly reported TAVI complications is the appearance of significant mitral regurgitation (MR). However, the actual incidence of this complication and the various mechanisms that produce it are not entirely known. In addition, it has been proven that transesophageal echocardiography (TEE) is superior to contrast ventriculography in detecting MR, especially in evaluating its degree and mechanism.⁴ 

Thus, the purpose of this study was to analyze the frequency with which significant MR occurs during the implant of a CoreValve percutaneous aortic prosthesis, as well as to study the utility of TEE in comparison to angiography for diagnosing its mechanism. 

Methods

Patients. From June 2008 to October 2011, we selected a total of 129 consecutive patients with severe symptomatic aortic stenosis and high surgical risk who underwent implantation of a CoreValve percutaneous aortic prosthesis. The design of the study corresponds to a series of cases. The patients were selected through a previous study, the protocol of which included a clinical evaluation of the patient and their cardiac structures through transthoracic echocardiography (TTE), TEE, angiography, hemodynamics, and a 64-slice computed tomography, as described in previous publications by our group.⁵ 

Transesophageal study. Intraprocedural TEE was performed by an expert echocardiographer. The intraprocedure TEE was not performed routinely in our center for the first 50 study patients, but rather was indicated when a possible complication was suspected (for example, after the appearance of hemodynamic instability of the patient, elevation of the pulmonary capillary pressure with a  V-wave, or when the postimplant angiography indicated significant mitral or aortic regurgitation). For the rest of the study, which included the remaining 79 patients, the procedure was routinely monitored with TEE for all patients. 

Study with contrast ventriculography. Standard ventriculography with contrast was performed in the right anterior oblique projection at 30° in the basal study for all patients in the study using an automatic injector. Likewise, ventriculography was performed during the procedure when a complication (including significant MR) was suspected. MR severity was determined based on the method of the opacification of the left auricle according to the criteria of Grossman.⁶ 

Definition of variables. Significant MR was defined as the appearance of mitral regurgitation greater than grade IIa through TEE following the recommendations of the last guidelines,⁷ which were published in 2012. MR was assessed as grade I (mild), grade IIA (mild-moderate), IIB (moderate-severe), and grade III. Therefore, we could have a significant MR in the following cases: (1) if the patient had previous MR (grade ≤IIa), MR was considered significant if it was at least one grade worse (grade >IIa); or (2) the appearance of “new” MR grade >IIa. 

Transthoracic echocardiography at discharge. At discharge, all surviving patients underwent a complete TTE performed by an expert echocardiographer using the same echocardiographic equipment (Philips iE33), where the degree of MR was evaluated in accordance with appropriate echocardiographic guidelines.⁷ The decision to perform TEE at discharge was made only in selected cases.    

Statistical analysis. Quantitative data are given as average ± standard deviation. Qualitative data are given as frequencies (percentages).    

Results

In our center, during the study period, the implantation of a CoreValve aortic prosthesis was attempted in 131 patients. In this population, 59% were women (76 cases) with an average age of 78 ± 5 years; the basal features of the patients appear in Table 1. The prosthesis was successfully implanted in 129 patients (98.5%), with only 2 deaths, which were due to the perforation of the left ventricle with the guide on which the prosthesis is mounted prior to its implantation. Four patients were excluded for previously receiving metallic mitral prostheses.

Significant MR occurred in 11 cases (8.5% of the entire study). In 100% of the cases in which angiography and TEE were performed, both techniques diagnosed the same number of cases of significant MR; however, 2 cases of MR were evaluated by TEE as grade III and as grade IIB by angiography. According to TEE, 8 cases had grade IIB, and 3 cases had grade III (Table 2). None of these patients developed more than mild periprosthetic aortic regurgitation.

With respect to the mechanism of occurrence of significant MR, angiography was not able to determine the mechanism in any of the cases (Figure 1), while TEE ascertained the mechanism in 100% of cases. One case (9% of the total cases of MR) developed grade IIB significant MR due to mechanical asynchrony because of the appearance of a new left branch block, which was transitory. Three cases (27%) were due to an “impingement” of the prosthesis on the anterior leaflet of the mitral valve because of low implantation of the prosthesis. The problem was resolved by repositioning the device in the direction of the ascending aorta prior to definitive release. In 2 cases (18%), the mechanism was the appearance of an anterior systolic movement of the mitral valve with the dynamic obstruction of the left ventricle outflow tract and moderate-severe MR immediately after the valvuloplasty (Figure 2), which resolved after the administration of intravenous beta-blockers. In 1 case (9%), after balloon valvuloplasty and prior to prosthesis implantation, a laceration of the anterior commissure of the mitral valve resulted in grade III MR (Figure 2).  After implantation of the aortic percutaneous prosthesis and subsequent improvement of the hemodynamic conditions, the degree of MR was reduced at the end of the procedure to grade III. In 4 cases (36%), the mechanism was attributed to a functional cause, ie, the transitory alteration of the left ventricular function during the release of the device, which continued to diminish progressively during the succeeding minutes.  

In our study, TEE at the end of the procedure revealed that MR was lower than grade IIB in all cases, except 2 in which it was grade IIB (the patient with the laceration of the mitral commissure and a patient in which MR was due to mitral systolic anterior motion). TTE at discharge showed that only the patient with laceration of the mitral commissure continued to exhibit grade IIB MR. All other cases of mitral insufficiency showed a grade ≤IIA, and no case required surgical treatment on the mitral valve. 

Discussion

The appearance of significant MR is one of the possible complications of a percutaneous prosthesis implant procedure, with a lower incidence in our study (8.5%) than in other studies of the same type,⁸ which is most likely due to stricter criteria for identifying significant MR.    

Regarding mechanisms for the occurrence of MR, one of the mechanisms most commonly found in our study was the “impingement” of the prosthetic structure on the anterior mitral leaflet due to a low implantation of the device. This problem was the second most frequent cause of MR and could be reduced in all cases by pulling and repositioning the prosthesis before the complete release of the device.⁹ Nevertheless, other patients with the same lower position did not develop significant MR; the repositioning was not performed in these cases. In our laboratory routine, the repositioning of the prosthesis is performed only if it extremely exceeds the limit specified by the manufacturer, or if serious complications such as MR or paravalvular aortic regurgitation occur.

Another of the causes found for MR was the “organic” damage of the mitral valvular/subvalvular apparatus, a complication that occurs more frequently during the implantation of the prosthesis via the transapical approach (Edwards Sapien prosthesis),¹⁰ even though in our study, the damage to the mitral valve was due to a laceration on the free edge of the anterior mitral commissure after the valvuloplasty using a balloon on an intensely fibrosated valve.  This MR mechanism had previously been described by other authors while performing valvuloplasty with balloon on calcified aortic valves as a palliative treatment for a severe aortic stenosis or on a congenital obstruction of the left ventricular outflow tract in a pediatric patient.^11-13  

In our study, 2 cases of MR were due to the anterior systolic movement of the mitral valve leaflets secondary to a dynamic obstruction of the left ventricular outflow tract. This etiology has been described after surgery for severe aortic stenosis;¹⁴ however, to the best of our knowledge, it has not been reported by other authors in important studies of  TAVI, nor has it been described in the recommendations mentioned above for the use of echocardiography in the implantation of this type of prosthesis. In our study, the frequency of this complication has not been negligible, and it seems important to report it as a mechanism of occurrence of MR that should be considered when this type of procedure is performed, as TEE is essential at the point of diagnosis and offers good guidance during the pharmacological management of the patient.   

The mechanism of MR that we have labeled as “functional” is more difficult to explain. This mechanism can be identified when TEE makes it possible to rule out any organic damage to the mitral valvular and/or subvalvular structure and none of the other described mechanisms applies. It is well known that during the release of the device, an important systolic dysfunction occurs that can practically lead to an asystolic situation in certain cases, and clearly produces myocardial damage affecting the mitral subvalvular apparatus. In the 4 cases described in our study, a notably severe ventricular dysfunction was observed with TEE during the implantation, with later recuperation. However, the reason that this significant functional MR does not occur in all patients is not well known; it’s possible that it occurs in a subgroup of patients with intrinsically greater myocardial dysfunction, even though they have a similar ejection fraction. Studies with cardio-resonance have shown a higher degree of fibrosis in the myocardium of certain patients with aortic stenosis in spite of similar degrees of ejection fraction. In our study, in these cases, MR was reduced to non-significant levels at the end of the procedure.¹⁵ 

Finally, another MR mechanism described during or after the implant of the CoreValve prosthesis was the development of an asynchronic contraction of both ventricles due to stimulation by a pacemaker in the right ventricle.³ In our study, no MR due to this mechanism was found, but we did find a case where transitory MR developed due to the presence of a new left bundle block. 

Generally, the development of MR is less common after implantation of the Edwards Sapien prosthesis.¹⁷ Possible reasons could be the shorter length of this prosthesis without impingement on the anterior mitral leaflet and the lower incidence of conduction disturbance.

In treating these patients, management was performed depending on the MR mechanism diagnosed by TEE and varied from treatment with beta-blockers for the mitral [systolic anterior movement] to traction on the not-yet-completely released prosthesis in cases where mechanism was the impingement,¹⁶ and observation and a conservative attitude in cases where it was due to a functional mechanism or the development of a new transitory left branch block. In the case of the laceration of the mitral anterior commissure, the improvement in hemodynamic conditions and the partial reduction of regurgitation suggested a “wait and see” attitude.    

The rapid resolution of MR to a non-significant degree in all but 1 case meant that surgery was not necessary for any of these cases. Only the patient with a grade IIB MR that persisted after the laceration of the mitral anterior leaflet is being followed closely, continuing 1 year of follow-up in functional grade II, and has not required surgery to date.  

As is evident in our study, angiography is not enough to determine the mechanism of occurrence of significant MR during the implantation of a CoreValve prosthesis. TEE is most likely the ideal diagnostic test for this purpose. Therefore, we recommend the monitoring of all TAVI procedures with a CoreValve prosthesis by means of TEE.  

Study limitations. The principal limitation of this study is that TEE was not performed systematically in the first 50 patients; however, the fact that hemodynamic monitoring did not suggest at any point the appearance of significant MR, together with the absence of MR grade >IIa during TTE at discharge in the rest of the study, indicates that underdiagnosing this complication during this period was highly unlikely.    

The diagnosis of “functional” MR was chosen after ruling out any other cause of mitral dysfunction by TEE; however, the presence of greater myocardial damage in these patients made evident by the elevation of biochemical parameters, such as creatine phosphokinase or troponin, was not tested.   

Conclusion 

A non-negligible percentage of patients treated with TAVI develop significant MR during the procedure through highly diverse mechanisms. In contrast to angiography, TEE differentiates the mechanism of MR after implantation of the CoreValve, ie, those due to mitral valve injury vs those due to functional abnormalities of the leaflets, as in the case of anterior systolic movement of the mitral valve or “tenting” due to myocardial dysfunction. 

REFERENCES

1. Kodali SK, Williams MR, Smith CR, et al. Two-year outcomes after transcatheter or surgical aortic-valve replacement. N Engl J Med. 2012;366(18):1686-1695. Epub 2012 Mar 26.
2. Mesa Rubio D, Suárez de Lezo Cruz Conde J, Álvarez-Osorio MP, et al. Measurement of aortic valve annulus using different cardiac imaging techniques in transcatheter aortic valve implantation: agreement with finally implanted prosthesis size. Echocardiography. 2011:28(4):388-396.
3. Zamorano JL, Badano LP, Bruce C, et al. EAE/ASE recommendations for the use of echocardiography in new transcatheter interventions for valvular heart disease. Eur Heart J. 2011;32(17):2189-2214. 
4. Kamp O, Huitink H, van Eenige MJ, Visser CA, Roos JP. Value of pulmonary venous flow characteristics in the assessment of severity of native mitral valve regurgitation: an angiographic correlated study. J Am Soc Echocardiogr. 1992;5(3):239-246.
5. León C, Suárez de Lezo J, Mesa D, et al. Early development of leaks in the CoreValve percutaneous aortic valve prosthesis: echocardiographic assessment. Rev Esp Cardiol. 2011:64(1):67-70. 
6. Grossman W, Dexter L. Profiles in valvular heart disease. In: Grossman W, ed. Cardiac Catheterization and Angiography. 2nd ed. Philadelphia, Pa: Lea & Febiger. 1980:305-324.
7. Lancellotti P, Tribouilloy C, Hagendorff A, et al. Recommendations for the echocardiographic assessment of native valvular regurgitation: an executive summary from the European Association of Cardiovascular Imaging. Eur Heart J Cardiovasc Imaging. 2013;14(7):611-644. 
8. Gotzmann M, Pljakic A, Bojara W, et al. Transcatheter aortic valve implantation in patients with severe symptomatic aortic valve stenosis — predictors of mortality and poor treatment response. Am Heart J. 2011;162(2):238-245.e1. Epub 2011 Jul 18.
9. Tzikas A, Piazza N, van Dalen BM, et al. Changes in mitral regurgitation after transcatheter aortic valve implantation. Catheter Cardiovasc Interv. 2010;75(1):43-49. 
10. Hekimian G, Detaint D, Messika-Zeitoun D, et al. Mitral regurgitation in patients referred for transcatheter aortic valve implantation using the Edwards Sapien prosthesis: mechanisms and early postprocedural changes. J Am Soc Echocardiogr. 2012;25(2):160-165. Epub 2011 Nov 8.
11. De Ubago JL, Vazquez de Prada JA, Moujir F, Olalla JJ, Figueroa A, Colman T. Mitral valve rupture during percutaneous dilatation of aortic valve stenosis. Cathet Cardiovasc Diagn. 1989;16(2):115-118.
12. Yamamoto Y, Shiraishi I, Yamagishi M, Hamaoka K. Mitral valve injury during balloon valvuloplasty for an infant with severe aortic stenosis: spatial evaluation using three-dimensional echocardiography. Pediatr Cardiol, 2003;24(3):300-303. Epub 2003 Apr 16.
13. De Lezo JS, Romero M, Segura J, et al. Long-term outcome of patients with isolated thin discrete subaortic stenosis treated by balloon dilation: a 25-year study. Circulation. 2011;124(13):1461-1468. Epub 2011 Aug 29.
14. Kerut EK, Hanawalt C, Dearstine M, et al. Mitral systolic anterior motion (SAM) with dynamic left ventricular outflow obstruction following aortic valve replacement. Echocardiography. 2007;24(6):658-660. 
15. Rodés-Cabau J, Gutiérrez M, Bagur R, et al. Incidence, predictive factors, and prognostic value of myocardial injury following uncomplicated transcatheter aortic valve implantation. J Am Coll Cardiol. 2011;57(20):1988-1999. 
16. Laborde JC, Brecker JD, Roy D, Jahangiri M. Complications at the time of transcatheter aortic valve implantation. Methodist Debakey Cardiovasc J. 2012;8(2):38-41. 
17. Spethmann S, Dreger H, Baldenhofer G, et al. Long-term Doppler hemodynamics and effective orifice areas of Edwards Sapien and Medtronic CoreValve prostheses after TAVI. 2014;31(3):302-310. Epub 2013 Sep 24. 

_______________________________________

From the Servicio de Cardiología, Hospital Universitario Reina Sofía, Córdoba, Spain.

Disclosure: The authors have completed and returned the ICMJE Form for Disclosure of Potential Conflicts of Interest. The authors report no conflicts of interest regarding the content herein.

Manuscript submitted October 24, 2013, provisional acceptance given November 18, 2013, final version accepted January 28, 2014.

Address for correspondence: Dr José López-Aguilera, Servicio de Cardiología. Hospital Universitario Reina Sofía, Avd. Menéndez Pidal s/n 14005 Córdoba, Spain. Email: mircardjla@gmail.com