Aortic Homografts in Surgical Management of Prosthetic Valve Endocarditis: A Case Series from Greece

Introduction

Infective endocarditis (IE) is a severe, life-threatening, and relatively common complication in patients that had valve replacement operations, with an incidence rate varying between 1.8% and 5.8% with an in-hospital mortality rate of up to 20% [1,2]. Common microorganisms are (by decreasing incidence) Streptococci, Staphylococcus aureus, Enterococci, bacteria of the HACEK group, and fungi. The pathogenesis involves adhesion of bacteria to the damaged endothelial or artificial surface, followed by formation of vegetations consisting of platelets, fibrin, and microorganisms. Lytic enzymes released by the bacteria simultaneously with the inflammatory response lead to further tissue damage or even perforation. Clinically, IE presents with a range of signs and symptoms, including fever, heart murmurs, sepsis, and various embolic phenomena, while complications are severe and may include heart failure, valvular dysfunction, systemic embolism, abscess, and mycotic aneurysm formation. Diagnosis is primarily based on the modified Duke criteria, involving clinical, microbiological, and echocardiographic findings [3]. Treatment of IE is complex, typically involving prolonged courses of antibiotics. The presence of prosthetic valve endocarditis combined with embolic events is an indication for surgical intervention. The conventional surgical approach involves replacement or repair of the affected valve, and surgical debridement of surrounding infected tissue. Especially in cases of aortic prosthetic valve endocarditis, root abscess formation with involvement of the aorto-mitral skeleton is not uncommon and complex surgical intervention is required, including structural reconstruction of cardiac skeleton with closure of fistulas or defects with pericardial patches if needed. However, the use of artificial aortic grafts and valves has an innate higher risk of re-infection, as well as surgical reimplantation difficulties related to friability of the surrounding tissue on which the new stiff artificial valve annulus must be sewed.

One of the notable advancements in surgical management of these cases is the use of homografts for valve repair or replacement [4], which are particularly advantageous in IE compared to synthetic graft and valve replacement, as the use of live tissue offers inherent resistance to re-infection, excellent hemodynamic performance, enhanced recovery, and no need for anticoagulation therapy, unlike mechanical valves.

Case Reports

We report the first case series of 8 patients successfully operated for prosthetic valve endocarditis with extensive aortic root abscess formation in Greece at Onassis Cardiac Surgery Center with the use of homografts. Patient characteristics are shown in Table 1. All homografts were implanted as aortic root grafts with direct suturing on the left ventricular outflow tract (LVOT) (Figures 1, 2). All cases were redo surgeries, with 2 of them being third-time sternotomies. Interestingly, one of the patients (Patient #6) had extensive aortic root involvement with abscess formation extending to the aorto-mitral fibrous skeleton, which led to aortic–left atrium shunt (TOE – Figure 3) requiring aortic root replacement with homografts, aorto-mitral skeleton reconstruction with bovine pericardium, and mitral valve replacement with a mechanical prosthesis (commando procedure) as previously described by David et al [5].

Another patient (Patient #7) had an extensive aortic paravalvular leak (80% of the perimeter). Intraoperative evaluation of this patient showed complete detachment of the mechanical aortic prosthesis (Figure 4).

Two other patients (Patients #5 and #8) required concomitant coronary bypass grafting of the right coronary artery with reversed saphenous vein grafts. In the first case (Patient #8), we used this approach to reconstruct the pre-existing graft to the right coronary artery (RCA) due to ligation of the RCA during the previous operation. In the second case (Patient #5), the RCA saphenous vein graft was performed as a rescue approach due to inability to re-implant the right coronary ostium to the homograft, and was ligated. This patient (Patient #5) had also extensive ascending aorta replacement with valved synthetic woven polyester graft in the first surgery. To minimize the risk of re-infection, all synthetic material was removed, and a neo-aorta was created by folding a bovine pericardial patch and suturing it to the distal end of the homograft and proximal aortic arch (Figure 5).

Only 1 of the patients (Patient #7) was in shock due to heart failure attributed both to infection and valvular disfunction due to dehiscence and was operated on as soon as the homograft was available. Seven patients were treated as elective cases after appropriate antibiotic therapy duration.

Fungus was identified in a single patient (Patient #8) as the culprit organism of endocarditis. Initial blood cultures were negative for all types of microorganisms and empirical antibiotic treatment was prescribed for 2 weeks. However, culture of the removed aortic valve specimen showed extensive growth of a fungicide-sensitive strain of Candida parapsilosis and appropriate parenteral treatment was administered for 4 weeks postoperatively. Further per os antibiotic regimen was prescribed to all patients upon discharge from the hospital to complete the required minimum 6-week duration of treatment for bacterial endocarditis and 6 months for the patient with fungal infection.

Six patients (Patients #3–8) had prolonged in-hospital stays due to difficulty in weaning from ventilator and inotropic drugs. Three of them (Patients #5, 6, and 8) needed reoperation within the first 72 hours due to bleeding and pericardial clot accumulation, with a single patient presenting tamponade.

Two patients (Patient #5 and 8) developed sustained conduction abnormalities and a permanent pacemaker was implanted 2 weeks postoperatively after total remission of infection (negative blood cultures, negative transesophageal echo, and normal values of inflammatory biomarkers).

Despite the variability in the length of stay of these patients in the hospital due to the severity of their condition, all patients had good outcomes and were discharged from the hospital with full functional status. No relapses or long-term surgical complications were reported during their follow-up visits (1–2 years postoperatively).

Discussion

Surgical management of endocarditis is always a challenge of escalating complexity depending on the extent of infected tissue. Simpler cases involve a single affected valve with the presence of vegetations. Several centers, including ours, manage these IE cases by replacing the affected valve with a new artificial one (bioprosthetic or mechanical). However, annular extension of the infection is not uncommon and in these cases several centers prefer replacing the valve with an artificial one and reconstruction of the annulus with pericardium. The use of homografts for aortic root reconstruction is the preferred method as they are resistant to recurrent infection because they are made of biologic human material. Furthermore, from a surgical technique perspective, it is the recommended approach, as the tissue is more pliable and permits hemostatic suturing of the homograft to the left ventricular out-flow tract (LVOT) because the aortic root is destroyed due to abscess formation. However, the timing of the surgery should be individualized based on the clinical condition of the patient (presence of hemodynamic instability, embolic phenomena, sepsis, multiorgan failure) and classified as emergency (within 24 hours), urgent (3–5 days), and non-urgent/elective (1–2 weeks) [6]. This may affect surgical management options due to limited and time-sensitive availability of homografts. Specifically, in Greece, the homografts have to be shipped from cadaver banks in Western Europe and their cost is not yet fully covered by public or private health insurance. To overcome the lack of immediate availability, some centers use a full aortic root, stentless porcine bioprosthesis (Medtronic Freestyle), which shares the anatomic features and long-term benefits of a homograft [7] but still has the innate weakness of introducing artificial tissue susceptible to re-infection. For this reason, this approach is not currently used in our center.

All the above-mentioned features make homografts a preferable choice in young patients, women of childbearing age, and those with contraindications to anticoagulation. A cost-effectiveness study [8] showed that homografts were superior, even for the initial valve replacement surgery. Although homografts seem to have multiple benefits, long-term disadvantages like calcification necessitating reoperation are still present [4,9]. Existing evidence is still ambiguous regarding their superiority. A recent meta-analysis indicated no significantly different mortality or re-infection rates in patients treated for endocarditis with homografts versus valved conduits [10]. However, the high heterogeneity of studies and patients included may preclude firm conclusions, and further randomized studies are needed.

Conclusions

In conclusion, aortic root replacement with aortic homograft is the preferred choice for prosthetic valve endocarditis with aortic root abscess formation. Despite the technical complexity of implantation, this option offers a second chance for survival in patients with this critical condition.