Secondary Logo

Journal Logo

Atrioventricular Disruption After Mitral Valve Replacement: The Role of Intraoperative Transesophageal Echocardiography

Chui, Jason MB ChB*; Roscoe, Andrew MB ChB*; Tsang, Wendy MD, SM

doi: 10.1213/ANE.0000000000000306
Cardiovascular Anesthesiology: Echo Rounds

Supplemental Digital Content is available in the text.

From the Departments of *Anesthesia and Pain Management, and Cardiology, University of Toronto, Toronto, Canada.

Accepted for publication March 7, 2014.

Funding: None.

The authors declare no conflicts of interest.

Patient Consent Statement: The patient’s family was contacted and gave permission for publication.

Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s website.

Reprints will not be available from the authors.

Address correspondence to Jason Chui, MB ChB, Department of Anesthesia and Pain Management, Toronto General Hospital, 200 Elizabeth Street, Toronto, ON M5G 2C4, Canada. Address e-mail to

An 81-year-old man with coronary artery disease and severe mitral regurgitation from myxomatous mitral valve disease was referred for surgery due to progressive congestive heart failure. He was scheduled to have mitral valve replacement (MVR) and coronary artery bypass grafting. His intraoperative transesophageal echocardiogram (TEE) revealed that the main mechanism of his mitral regurgitation was anterior leaflet prolapse. He had severe left ventricular (LV) dysfunction (ejection fraction 25%). There was right ventricular and tricuspid annular dilation with mild-to-moderate tricuspid regurgitation, but right ventricular function was preserved. After completion of the MVR (29 mm Hancock II, Medtronic, Mississauga, Canada) and coronary artery bypass grafting (4 grafts), the patient was weaned from cardiopulmonary bypass (CPB) with inotropic support. Immediate post-CPB TEE revealed paravalvular leak, a small hematoma in the interatrial septum (IAS) and global LV hypokinesia (ejection fraction 10%–20%) (Fig. 1; Video 1, Supplemental Digital Content 1, The initial diagnosis was myocardial ischemia due to coronary air. The surgeon decided to decannulate the aorta, and protamine was administered. The patient became more hypotensive, despite increasing inotropic requirements. The TEE showed worsening biventricular function, subtle rocking of the bioprosthetic MVR, and hematoma in the IAS (Fig. 2; Video 2, Supplemental Digital Content 2, The patient remained hypotensive, and CPB was reinstituted for surgical exploration. A large hematoma was found in the posterior atrioventricular groove extending to the inferior vena cava. The surgeon attempted but failed to repair the disrupted atrioventricular groove, and the patient died on the operating table. Retrospective analysis of the intraoperative 3-D TEE imaging using Qlab software (Philips Medical Systems) revealed a crescentic gap around the posterior aspect of the MVR, consistent with atrioventricular separation (Fig. 3; Video 3, Supplemental Digital Content 3, With the use of multiple 2D planes (MPR mode) reconstructed from a 3D data set, the extent of atrioventricular separation was more readily appreciated (Fig. 3).

Figure 1

Figure 1

Figure 2

Figure 2

Figure 3

Figure 3

Atrioventricular disruption is a rare but fatal complication after MVR, with a reported incidence between 0.5% and 14% and mortality between 50% and 75%.1,2 Classification has been divided into 3 types based on the location of the tear. Type I is the most common type and is associated with MVR. Types II and III have been virtually abolished, primarily due to the introduction of surgical preservation techniques and low-profile mitral prostheses (Table 1).

Table 1

Table 1

The main clinical presentation for atrioventricular disruption is either unstable hemodynamics after weaning from CPB or failure to wean from CPB. Frank rupture with massive bleeding from the LV can alert the clinicians to the diagnosis of LV rupture. However, LV failure and ventricular arrhythmias are usually nonspecific presentations in many cases. More specific signs, such as dissecting hematoma in the posterior atrioventricular groove, are often not appreciated by the surgeon from the anterior operating view. Therefore, TEE diagnosis is important to provide an earlier diagnosis, allowing immediate surgical intervention.

Echocardiographic features of late or delayed LV rupture with pseudoaneurysm have been well reported,3,4 but intraoperative TEE features of early LV type 1 rupture after MVR have not yet been described. We found that the initial 2D echocardiographic features were very subtle. The most striking features in our patient were worsening of biventricular failure and rocking of the bioprosthetic MVR. The small IAS hematoma in the initial images after separation from CPB is atypical after MVR and should prompt further evaluation. The differential should always include the possibility of atrioventricular separation.5 There was a subtle rocking motion of the bioprosthetic MVR on immediate post-CPB images, but both the degree of rocking and the size of hematoma progressed with the worsening of the atrioventricular separation, clarifying the diagnosis. The distinction among MVR dehiscence, paravalvular leak, and atrioventricular disruption may be difficult by 2D TEE alone but has both management and prognostic implications. Left atrial (LA) dissection may be associated with LV rupture5,6 and typically appears as a hypoechoic space from the mitral origin extending along the IAS or LA wall. The presence of systolic collapsing and demonstration of communicating flow across the false cavity are the hallmarks of LA dissection.7

Three-dimensional TEE has the advantage of better delineation of the extent of the atrioventricular separation. Our 3D image demonstrated the presence of a large crescentic gap between the prosthetic valve annulus and the LV, which was larger than appreciated on the 2D images. When we used multiplanar analysis, we were able to deconstruct the 3D data set to obtain an en face cut plane image of the entire prosthetic mitral valve annulus where it was not attached to the LV. With this unique plane, we can truly appreciate the extensive nature and size of the disruption. Given the clinical scenario, the almost circumferential gap was not consistent with a residual paravalvular leak but atrioventricular disruption. Beyond the qualitative assessment, this cut plane through the 3D data set also allowed us to quantitate the size. This would have been important if any annular reconstruction techniques are considered in an attempt to correct the situation.

In summary, atrioventricular disruption is an uncommon but often fatal complication of MVR and prompt diagnosis is crucial. Initial echocardiographic features, including LV failure, IAS hematoma, paravalvular leak, and rocking motion of the MVR, can be subtle and nonspecific. Three-dimensional echocardiographic examination may be helpful for a more accurate and earlier diagnosis.

Back to Top | Article Outline

Clinician’s Key Teaching Points

By Kent H. Rehfeldt, MD, Roman M. Sniecinski, MD, and Martin J. London, MD

  • Atrioventricular disruption is a rare but often fatal complication of mitral valve replacement resulting from excessive debridement of annular calcification, placement of annular sutures partially within the posterior left ventricular wall, or lifting the heart in a manner that uses the atrioventricular groove as a fulcrum.
  • Although the initial transesophageal echocardiogram (TEE) findings may be nonspecific, it can help distinguish atrioventricular disruption from other complications of mitral replacement such as paravalvular leak or left atrial dissection. Isolated paravalvular regurgitation typically presents without adjacent hematoma. Left atrial dissection is characterized by an echolucent space within the atrial septum or lateral atrial wall that may demonstrate blood flow on Doppler imaging or systolic collapse of the space.
  • In this case, besides identifying paravalvular regurgitation, TEE demonstrated an enlarging hematoma of the atrial septum and an increasing degree of prosthesis rocking indicative of partial separation of the sewing ring from the annulus. In addition, a progressive decline in biventricular systolic dysfunction was observed.
  • Three-dimensional TEE may offer advantages over 2D imaging in the recognition of atrioventricular disruption. In particular, an en face view of the mitral prosthesis may demonstrate extensive separation of the sewing ring away from the annulus, distinguishing this condition from an isolated paravalvular leak.
Back to Top | Article Outline


Name: Jason Chui, MB ChB.

Contribution: This author helped conduct the study and write the manuscript.

Attestation: Jason Chui approved the final manuscript.

Name: Andrew Roscoe, MB ChB.

Contribution: This author helped conduct the study and write the manuscript.

Attestation: Andrew Roscoe approved the final manuscript.

Name: Wendy Tsang, MD, SM.

Contribution: This author helped conduct the study and write the manuscript.

Attestation: Wendy Tsang approved the final manuscript.

This manuscript was handled by: Martin J. London, MD.

Back to Top | Article Outline


1. Wei J, Wu C, Hong G, Tung DY, Chang CY, Chuang YC. Autotransplantation of heart for repair of left ventricular rupture after mitral valve replacement. Transplant Proc. 2001;33:3553–4
2. Zhang HJ, Ma WG, Xu JP, Hu SS, Zhu XD. Left ventricular rupture after mitral valve replacement: a report of 13 cases. Asian Cardiovasc Thorac Ann. 2006;14:26–9
3. Mihaljevic T, Couper GS, Byrne JG, Paul S, Nathan N, Aranki SF, Cohn LH. Echocardiographic localization of left ventricular free wall rupture after minimally invasive mitral valve replacement. J Cardiothorac Vasc Anesth. 2003;17:733–5
4. Biyikoglu SF, Guray Y, Turkvatan A, Boyaci A, Katircioglu F. A serious complication late after mitral valve replacement: left ventricular rupture with pseudoaneurysm. J Am Soc Echocardiogr. 2008;21:1178.e1–3
5. Milla F, Adams DH, Mittnacht AJ. Contained left ventricular rupture with left atrial dissection after mitral valve repair. J Cardiothorac Vasc Anesth. 2010;24:817–9
6. Tang D, Liu H. Acute left atrial intramural wall dissection after mitral valve replacement. J Cardiothorac Vasc Anesth. 2011;25:498–500
7. Tolpin DA, Collard CD, Thomas Z, Pan W. Left atrial dissection associated with pulmonary vein cannulation. Anesth Analg. 2009;109:1409–12
8. Zacharias A. Repair of spontaneous rupture of the posterior wall of the left ventricle after mitral valve replacement. Thorac Cardiovasc Surg. 2003;8:36–41

    Supplemental Digital Content

    Back to Top | Article Outline
    © 2014 International Anesthesia Research Society