Survival of left ventricular rupture post mitral valve replacement

Abstract

A fatal and uncommon complication after mitral valve replacement is left ventricular (LV) rupture. We describe a case of a woman in her 40s with rheumatic heart disease and mitral regurgitation who underwent mitral valve replacement on cardiopulmonary bypass and experienced LV rupture but survived this catastrophe.

Background

Left ventricular (LV) rupture is the deadliest complication post mitral valve replacement (MVR). Incidence is about 0.5%–2.0%.¹ The first case of LV rupture was reported in 1967 by Robert et al. The mortality rate is 50%–95%. Because of the high mortality rate, it is essential to know further about the prevention, timely identification and management of LV rupture. Here, we share our experience with a patient who underwent external polytetrafluoroethylene (PTFE) felt repair for LV rupture following an MVR and survived the catastrophe by an age-old technique of proper compression of oozing surface.

Case report

A woman in her 40s presented to the cardiology department with complaints of chest pain on exertion for 8 years and shortness of breath for 8 years, initially New York Heart Association grade II (NYHA) and progressed to NYHA III. The patient has no documented history of rheumatic fever. The patient is moderately built with pansystolic murmur in the mitral area and radiating towards the left axilla with mild hepatomegaly. Cardiomegaly is seen on chest X-ray (figure 1A). Two-dimensional echocardiography showed anterior mitral leaflet thickening, diastolic doming and prolapsing with posterior mitral leaflet thickened and restricted, causing severe eccentric mitral regurgitation with peak gradient of 17 and mean gradient of 5 mm Hg with mitral valve area (MVA) of 2.1 cm², with an ejection fraction of 60% (figure 2A). All blood investigations are within normal limits. Blood and urine cultures were sterile. The patient underwent MVR on cardiopulmonary bypass (CPB). The anterior mitral leaflet was discovered to be badly calcified and was removed under the vision and replaced with a 31 # St Jude mechanical bileaflet valve. CPB is weaned off, and aortic decannulation is carried out. We observed that the pericardial space is inundated with blood following aortic decannulation. When haemostasis is examined at each cardiotomy location, we found the posterolateral wall of the heart had LV rupture of type III. Transoesophageal echocardiography (TEE) showed air in the left ventricle and left atrium (figure 2B). The LV rupture is immediately repaired using external PTFE felts stitched with pericardial pledgetted sutures after the patient is heparinised and CPB is restarted. Since adequate haemostasis is not accomplished at the rent site despite attempted repair and local biological glue application twice, the heart is packed with sterile, radiopaque mops, and mediastinal and pleural drains are inserted (figure 3). The standard heparin reversal by dose to dose by intravenous protamine was done. The existence of the mechanical valve precluded the administration of other systemic thrombostatic drugs. The patient was sent to the intensive care unit with an open sternum covered by a sterile dressing. Careful monitoring of vital signs and drain output continued. The patient’s haematocrit did not drop nor experience any haemodynamic instability. So the decision was made for mop removal followed by proper sternal closure. The patient was re-explored on the second postoperative day; all the mops were removed with a thorough saline wash. The repair site was discovered intact without oozing; hence, regular sternal closure was performed. A proper functioning mitral prosthetic valve and satisfactory LV function were seen during intraoperative TEE. The patient made an excellent postoperative recovery with no significant incidents.

Figure 1.

(A) Preoperative chest X-ray showing cardiomegaly and increased bronchovascular marking and (B) postoperative chest X-ray showing metallic mitral valve in position.

Figure 2.

(A) Preoperative transoesophageal echocardiography showing mitral regurgitation jet to the left atrium with dilated left atrium. (B) Intraoperative transoesophageal echocardiography showing bileaflet prosthetic mitral valve with air in the left atrium and ventricle.

Figure 3.

Intraoperative photograph showing LV repair using external felts superimposed with an autologous pericardium. LV, left ventricular.

Investigations

Investigations are already described.

Treatment

Treatment is already described.

Outcome and follow-up

On follow-up, the patient had fully recovered, and a chest X-ray revealed the prosthetic mitral valve in place (figure 1B). Follow-up two-dimensional echocardiogram also revealed that the prosthetic valve was functioning normally, with a mean gradient of 2 mm Hg and a peak gradient of 4 mm Hg with MVA of 2.2 cm², no paravalvular leak and normal biventricular function.

Discussion

LV rupture is a fatal and uncommon post-MVR event. The most frequent causes of LV rupture are excessive traction during excision, ripping of the annulus due to cardiac manipulation following the implantation of a mitral prosthesis and penetration of deep sutures into the left atrioventricular groove posteriorly. Roberts and Morrows initially documented LV rupture following MVR in 1967 during the autopsy of two individuals.^{1 2} LV rupture is divided into three categories (figure 4A). Type I Atrio-ventricular (AV) groove ruptures at the level of the AV groove, type II at the papillary muscle level and type III between types I and II in the posterior LV wall. Type I rupture has a 45% intraoperative mortality rate, while type II rupture has a 35% intraoperative mortality rate. The ‘untethered loop’ idea, by Cobbs et al in 1980, was deemed the most plausible theory.³ By this hypothesis, chordal preservation can prevent LV rupture to a greater extent. Incomplete adhesiolysis during revision surgery, papillary muscle division at the base, excessive decalcification, use of an oversized prosthesis, deep myocardial stitches, excessive tension on sutures, removal of the valve without providing enough exposure, blind usage of vents and hard suckers into the LV, frequent handling of the heart and ventricle being compressed firmly during deairing manoeuvre are additional factors.^{4 5} By careful handling of these steps, LV injury can be avoided. However, the cause of LV rupture cannot be defined in most cases.⁶ In our case, we assume forceful manual massage against the prosthesis as the probable cause. Type I AV rupture is best repaired by the trans-left atrium (LA) approach after explanting the prosthesis. Type II AV rupture is repaired through an epicardial approach. Karlson et al suggested type III can also be repaired externally if the limited defect is within the left ventricle. The epicardial approach can be patch repair or bio-glues.^{7 8} We repaired using epicardial PTFE felts (figure 4B, C). It is essential to take precautions to prevent damage to the left circumflex artery. Type I AV rupture repair has a high chance of sacrificing the left circumflex artery and may require bypass grafting. One drawback of the epicardial technique is that sutures used from outside the heart are not put in healthy myocardium, and bleeding outside the heart may not correspond to a genuine internal tear. The standard pledgetted approach, endocardial and transaortic patch technique, posterior leaflet or left atrial appendage flap technique, atrialisation of the tear, and autotransplantation are other endoventricular repair methods.⁹ Internal repair options include patch closure, direct closure with horizontal mattress sutures buttressed with felt and patch closure, which involves sewing a patch across the annulus to the intact endocardium of the left atrium and ventricle. Once LV rupture has occurred, the CPB has to be reinstituted immediately, even if the patient is already weaned off from CPB and then attempts are made for repair. Without CPB, only 7% of patients with LV rupture survive, compared with 50% with CPB. Survival with the external technique is 67%, whereas internal approach survival is 27%.^{10 11} In this kind of surgical bleeding scenario, systemic reversal of heparin should be done, but other thrombostatic agents have no clear therapeutic role. In our experience, significant oozing persisted after epicardial repair with felts and even using biological sealing glue over the repair as described by Schuetz et al.¹² It left us wondering whether to opt for explanting the valve and do the more exhaustive repair. We decided to pack the heart with mops and wait with careful monitoring for bleeding control due to the increased danger of endoventricular repair. Afterwards, there was no ooze from the repair site, and the patient was discharged without a prolonged hospital stay or any complications.

Figure 4.

(A)Types of LV rupture. Mechanical mitral valve shown as red circular structure. (B)The first layer of LV repair with external PTFE felts sutured in interrupted mattress manner and (C)the second layer of autologous pericardium sutured in a continuous manner over PTFE felt. Illustrations created by ARS. LV, left ventricular; PTFE, polytetrafluoroethylene.

Conclusion

Prevention is always preferable to treatment. Careful manipulation and excision are necessary to prevent damage. Timely identification and intervention can salvage the patient. Attempts to repair a beating heart should never be entertained. Once LV rupture is suspected, CPB should be reinstituted and repair should be done. The age-old technique of proper compression can stop bleeding even from LV rents and oozing from adjoining once.

Patient’s perspective (husband’s version translated).

My wife has had this illness for a long time, so we decided to have heart surgery. After the surgery, knowing that she still had ongoing bleeding and repair of the major part of the heart was done made us too much worried. Also, she was in an intensive care unit with an open chest. Information about repeated blood transfusions made us more uneasy, and the entire family was concerned about whether she would survive. As soon as the patient underwent a second surgery, the chest was closed, and she regained consciousness; we felt more confident that everything would turn out okay. She was moving around well at the time of discharge, which made us very delighted. She was able to manage herself within a week, and she is presently doing great without any problems.

Learning points.

• This fatal complication can be avoided by refraining from high traction and forceful valve excision.

• In high-risk patients, maintaining low threshold suspicion of a left ventricular (LV) rupture can aid in early detection and, consequently, in early management.

• Repair of this kind of tear in a beating heart should never be attempted.

• The reinstitution of cardiopulmonary bypass should be done immediately after diagnosing LV rupture.

• The age-old technique of compression still helps.

Ethics statements

Patient consent for publication

Consent obtained directly from patient(s)