Complicated endocarditis—to spare or not to spare the mitral valve

Infective endocarditis involving the native or prosthetic aortic valve presents itself through a myriad of pathological features ranging from vegetations to perforations to complete cusp destruction with or without annular and/or periannular extension. The clinical presentation of the patient is not only determined by the systemic effects of bacteraemia, septic emboli and sepsis but also the haemodynamic consequences of intracardiac lesions such as valve insufficiency and intracardiac fistulas and in some cases septic cardiomyopathy. Therefore, patients presenting for surgery may either be very sick due to the systemic manifestation of the infection or pose a technical challenge due to the complicated nature of the intracardiac pathology or in the worst-case scenario, both. The type of surgery depends on the extent of involvement of the aortic valve and perivalvular structures. When infection is limited to the aortic cusps, conventional valve repair techniques or more commonly replacement are usually sufficient to adequately address the infection. However, perivalvular extension of the infective process frequently involves the fibrous annulus of the aortic valve, often spreading to the intervalvular fibrous body (IFB) and further on to the anterior mitral leaflet (AML). Such patients require the so-called Commando/UFO operation, first developed by the Toronto group, which involves radical resection of all macroscopically infected tissues including the aortic valve or prosthesis, the IFB, and the mitral valve, particularly if a substantial portion of the AML is involved. This is followed by aortic and mitral valve replacement and reconstruction of the IFB with a 2-tongued bovine pericardial or Dacron patch [1–3]. Sometimes, however, only a small segment of the base of the AML is affected. In such situations, the so-called Hemi-commando operation described by the Cleveland clinic group, which involves replacement of the aortic valve/root and the IFB, while sparing the mitral valve, is performed [4]. It is common knowledge that late survival following mitral valve repair (MVRp) is better than replacement in patients undergoing aortic and mitral valve surgery [5], as is also the case in patients undergoing isolated mitral valve surgery for native mitral valve endocarditis [6]. MVRp is associated with lower incidence of recurrent endocarditis than replacement [7]. However, very little data concerning outcomes in patients undergoing the Hemi-commando operation is available in literature. Navia et al. described the technique involving aortic root replacement with a homograft and reconstruction of the AML and IFB with that of the donor homograft. Preservation of the free edge of the patient’s AML and its subvalvular apparatus is mandatory. Furthermore, the MVRp is supported with a ring annuloplasty [4]. The main advantage of this technique is the resistance to re-infection that is provided by homograft tissue, which is well known with the use of homografts in patients with aortic root abscesses [8]. Nevertheless, homografts are not always available in urgent or emergent situations, may not be accessible to many institutions all over the world and not all surgeons are experienced in their use, particularly in techniques, in which the height of the AML must also be appropriately sized. Use of alternate techniques is vital in such circumstances. Tomsic et al. have described a technique utilizing the Freestyle conduit for aortic root replacement and an annuloplasty ring and separate pericardial patches to reconstruct the AML and IFB in all patients undergoing a Hemi-commando procedure [9]. Following radical resection of all infected tissue, an appropriately sized annuloplasty ring is implanted, followed by suturing of a pericardial patch to the healthy edge of the AML and the anterior part of the ring. Thereafter, the folded margin of a 2-tongued pericardial patch is sutured to the anterior part of the ring and the 2 tongues are used to reconstruct the IFB and roof of the left atrium. An aortic root replacement is then performed with a Freestyle conduit to complete the operation. This procedure should not be used if the free edge of the AML, commissures or the posterior leaflet or annulus are involved in the infectious process. Irrespective of the technique used, all surgeries involving reconstruction of the IFB are fundamentally very complex and require much longer clamp and cardiopulmonary bypass times as compared to a double valve replacement. Adding MVRp and aortic root replacement as a routine in such cases increases the complexity manifold in this very sick group of patients. Being a complex MVRp, the probability of post-repair residual MR is higher than in other pathologies. It is important to note that another attempt at re-repair or replacement would not only increase the already prolonged clamp and cardiopulmonary bypass times, but would also be technically extremely challenging with an aortic bioprosthesis already in position. Furthermore, an aortic root with abscesses/infection involving less than half of its circumference can often be reconstructed by using a wider pericardial patch that is used to reconstruct the IFB. Moreover, if the infection is limited to the IFB and a few millimetres of the base of the AML, it may not even be necessary to open the roof of the left atrium. A deep-seated patch or aortic root replacement in the left ventricular outflow tract would suffice. The authors also prefer the Freestyle stentless root as it conforms better to the newly reconstructed, often irregularly shaped outflow tract, than a rigid stented graft. However, reoperations on Freestyle stentless roots are extremely difficult and complex. A customized root replacement with a valve conduit assembled on the operating table with a Dacron graft and a stented prosthesis implanted within it is a good alternative. The Dacron graft is first sewn to the newly reconstructed left ventricular outflow tract, thereby avoiding a rigid stented valve or conduit. Thereafter, a mechanical or stented bioprosthesis is implanted within the Dacron graft at a higher level than the proximal suture line, taking care not to be higher than the plane of attachment of the coronary buttons. Maintenance of adequate distance between the base of the AML and the sewing cuff of the aortic prosthesis is of utmost importance to prevent excessive traction on the AML, thereby reducing the possibility of residual mitral regurgitation. Therefore, the question that arises is whether MVRp would confer a survival benefit over mitral valve replacement in such patients. It would, of course, be inappropriate to compare outcomes of a Hemi-commando operation to double valve replacement with reconstruction of the IFB, as the choice of procedure is frequently dependent on the burden of disease and surgeon experience in MVRp with the latter procedure being performed when the mitral valve is more extensively involved. Surgeons who are adequately experienced to perform such complex procedures, but do not have expertise in MVRp may hesitate to spare the mitral valve and preferably replace it. The authors reported a 5- and 10-year survival of nearly 55% and 52%, respectively. Reinterventions were required in 7 patients (14.5%) undergoing MVRp and aortic root replacement, 5 of them being for mitral insufficiency. The Leipzig group recently reported a 42% 5-year survival among 127 patients undergoing double valve replacement and reconstruction of the IFB for complicated left-sided endocarditis. The survival, however, was 70% in 90-day survivors. Overall, 7% of patients underwent reoperations, most of them for recurrent infections resulting in an 85% 5-year freedom from reintervention [10]. To summarize, patients with advanced left-sided endocarditis involving the IFB should undergo urgent/emergent surgery in a tertiary or quaternary referral centre by surgeons highly experienced in performing complex reconstructive surgery for such extensive disease. Sparing the mitral valve is always preferable, but the decision should be based on the extent of involvement of the mitral valve in the disease process, the clinical status of the patient and the experience of the surgeon in the mitral valve-sparing approach. Complete eradication of infection should be the primary goal of the operation.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.