Discussion
Endocarditis after composite valve graft root replacement (Bentall procedure) is not uncommon (16)(17) and is often associated with periannular abscess, disruption of aortic-ventricular junction and aortic false aneurysm (18-20). Redo root surgery is warranted in these cases and may poses difficulties in chest re-entry, in providing an adequate myocardial protection and handling of frail and infected tissue (21,22). These patients can further present severe medical issues with ongoing heart failure and a poor controlled septic status, which account for a high perioperative mortality rate (9). Despite these technical and medical challenges, previous experiences reported good outcomes after reoperation for Bentall infection (9,10,11,21,22) and our results, early mortality of 10% and acceptable survival at 7-year follow-up, are in keeping with these findings.
Two large multi-institutional registries have recently shown that failure to undertake an operation in patients with infective endocarditis and presenting an indication for surgical treatment is a risk factor for early mortality (4,23). On the contrary, a conservative approach seemed providing an acceptable survival especially in patients who did not have local complications, such as abscess, fistula or false aneurysm. These findings were retrieved from mixed populations, including native and prosthetic valve endocarditis, and few evidences exist regarding patients who had a previous root replacement with a valve graft conduit.
We found that a conservative treatment can be associated with a dismal early and mid-term survival also in patients who did not present at the admission with local infective complications and severe tissue disruption. In our series, the most common reason for a conservative treatment was the presence of stable haemodynamic, controlled sepsis and the absence of local infective complications. Among these six patients who had no urgent indication for a surgical procedure (14,24), four did generally well and had a satisfactory survival, while two patients sustained a relapse of infective endocarditis with uncontrolled sepsis and heart failure.
Machelar et al. reported their experience in 8 patients with aortic valve graft conduit infection who were ultimately treated conservatively (25). Among them, one died due to cerebral haemorrhage after 6 weeks and one due to polymicrobial infection after 18 months. Another patient sustained a recurrent infection, and one was still on treatment. Four patients had complete remission during a follow-up time ranging from 30 to 68 months and, noteworthy, none of them had local infective complications at the admission. Similar results were found in our population, as we showed that patients with a stable clinical picture in absence of periannular abscess or aorto-ventricular disruption could be successfully managed with medical treatment in the acute setting. Alongside a less severe anatomical disruption, Staphylococcus aureus was not isolated, both in Machelar’s and our experiences, in any of the patients who sustained a complete remission after lone antibiotic treatment.