https://orcid.org/0009-0006-0339-5791
Abstract
The aim of this study was to evaluate the outcomes of transposition of the omentum into the mediastinum to support the replacement of infected aortic grafts or to cover infected aortic grafts that are not amenable for surgical replacement.
All patients with thoracic aortic graft infections who underwent mediastinal transposition of the omentum at our institution between 2005 and 2023 were included in this study. Mediastinal transposition of the omentum was performed either after replacement of the infected graft (‘curative concept’) or solely as bailout procedure by wrapping the infected graft (‘palliative concept’). The diagnosis, including computed tomography scans during follow-up, was made according to the criteria of the Management of Aortic Graft Infection Collaboration.
The patient cohort consisted of 31 patients. Both in-hospital and 1-year mortality were 0% (n = 0) for the curative concept (n = 9) compared to 23% (n = 5) and 41% (n = 9) for the palliative concept (n = 22), respectively. There was no graft infection-associated death or recurrence of infection after 3 years in the curative group. Survival was 52% at 3 years in the palliative group, with freedom of infection in 59% of the patients (n = 13).
Transposition of the omentum and wrapping of the infected aortic prosthetic graft is a useful bailout strategy for patients who are ineligible for replacement of an infected aortic graft. However, mortality stays high. For radical treatment of aortic graft infections, it may prove an effective supportive therapy and represents an important tool in the armamentarium of cardiac surgeons.
INTRODUCTION
Aortic graft infection (AGI) is a rare but serious complication after aortic surgery. According to current literature, the incidence of AGI lies between 0.6% and 3%, though the number of unreported cases may be higher than expected [1]. Reported mortality rates vary widely and are influenced by a multitude of factors [2]. These findings highlight that conservative treatment is associated with poor survival, amounting to mortality rates of 75% for conservative management compared to 40% in case of a surgical approach [1]. Thoracic AGI of the ascending aorta and/or the aortic arch requires in-situ reconstruction, which typically involves complete resection of all infected prosthetic material and surgical reconstruction, preferably using pericardial tubes or homografts (if available) [3]. According to this, the treatment of AGI represents a historically lethal problem, and outcome of patients is still poor. Various techniques to treat AGI of the ascending aorta and the aortic arch have been described in the past, among them mediastinal transposition of the greater omentum to wrap the aortic graft [2, 4–6]. However, this procedure is not commonly performed, and contemporary outcome reports are lacking [7]. In this retrospective single-centre study, we report our experience in 31 patients undergoing mediastinal transposition of the greater omentum for the treatment of AGI of the ascending aorta and the aortic arch. The procedure was performed after replacement of the infected graft as well as for local infection control in cases not amenable for radical graft replacement.
PATIENTS AND METHODS
Ethics approval
The local institutional review board (No. EA1/249/22) officially approved this study on 12 December 2022. The study complies with the Declaration of Helsinki.
Patient population
All patients with AGI of the ascending aorta and/or the aortic arch who underwent mediastinal transposition of the greater omentum at our centre between January 2005 and June 2023 were included in this study. Previous aortic surgery included open replacement of the aortic root and/or the ascending aorta and/or the aortic arch, mostly due to acute type A aortic dissection or aortic aneurysm. Infected hybrid prostheses such as the Ascyrus Medical Dissection Stent® (AMDS®; Artivion, Hechingen, Germany) or total arch replacement with a frozen elephant trunk did not constitute exclusion criteria; patients with AGI of complete endovascular grafts, on the other hand, were not included [8].
Diagnosis of aortic graft infection
The diagnosis of AGI was made according to the criteria of the Management of Aortic Graft Infection Collaboration [9]. Major and minor criteria for AGI are shown in Supplementary Material, Table S1. According to these criteria, AGI was suspected in the presence of 1 major or 2 minor criteria and diagnosed when there was at least 1 single major criterion and any other criterion.
Decision for treatment concept
Patients were either treated with a curative intention, including radical complete replacement of the infected aortic graft and consecutive omental wrapping (‘curative concept’), or with a palliative intention, preserving the infected graft and performing omental wrapping alone (‘palliative concept’). The decision for treatment was made on an individual basis by a multidisciplinary aortic team considering the patients’ condition and the surgical feasibility of extensive redo aortic surgery. The decision-making process is illustrated in Fig. 1. The feasibility of surgery primarily hinges on the surgeon’s expertise, with considerations including an assessment of the patient’s overall condition (such as sepsis, catecholamine levels, and cardiac function) and a thorough evaluation of local technical operability. These factors collectively inform the decision-making process regarding surgical operability. Euroscore II was calculated additionally to estimate the operative risk, but a specific cut-off value for each treatment concept was not defined for decision-making. Solely omental wrapping was performed as a bailout procedure if the patients’ condition was considered unstable and/or the surgical feasibility for complete radical replacement of the aortic graft was not given (‘palliative concept’). In both groups, meticulous surgical debridement and mediastinal lavage were performed additionally. Furthermore, all patients received specific intravenous perioperative antibiotic treatment for at least 6 weeks considering the identified pathogen, supported by the recommendations of the Department of Infectious Diseases. If recommended, subsequent antibiotic oral suppression was added. In this regard, the duration of postoperative antibiotic treatment differed depending on the identified pathogen and individual recommendations.
Patient selection process for treatment concept.
Surgical procedure
Median sternotomy was the access for all patients in this study. In patients deemed eligible for a curative treatment concept, complete replacement of the aortic graft was performed during cardiopulmonary bypass and cardiac arrest, preferably using pericardial roles or biological homografts (if commercially available). Surgical debridement and mediastinal lavage were performed complementarily in all patients. In addition, microbiological swabs of pus and suspicious areas were taken and resected parts of the infected graft and/or surrounding tissue were collected and sent for further microbiological analyses. Mediastinal transposition of the greater omentum was performed as the last step.
Abdominal access was achieved from the caudal end of the median sternotomy taking care not to extend the skin incision or cut the diaphragm. The latter may only be necessary in case of adhesions and hard ‘block-like’ tissue to not to compress the feeding arteries of the omentum. Only in a few cases, e.g., in the presence of severe abdominal adhesions after previous visceral surgery, median laparotomy was necessary to gain access to the abdominal cavity and prepare the omentum. Then, the greater omentum was carefully dissected from its attachments to the transverse colon and the greater curvature of the stomach while preserving the right gastroepiploic artery. This was performed step-by-step using multiple single ligations and bipolar cautery until achieving sufficient mobility and length for mediastinal transposition. Finally, the pedicled omentum was lifted gently into the mediastinum and carefully wrapped around the aortic graft using additional absorbable sutures to ensure complete prosthesis coverage. This procedure is shown in Video 1. Sternal closure was performed in standard fashion using steel wires in the most cases. In case of sternum osteomyelitis, meticulous debridement of the sternum using a sharp curette combined with excessive lavage was performed. The modified Robicsek technique was used in case of multiple transversely fragmented sternal locations in a few patients [10]. Vacuum-assisted closure was used additionally in a few cases to promote tissue regeneration.
Definition of outcome and follow-up
Follow-up was performed by postoperative appointments in our aortic department including computed tomography (CT) scans. The median follow-up was 700 days (234–2260 days). Three-year survival and freedom of infection was 100% complete except for patients who could not complete the follow-up due to a more recent date of surgery. Data collection for the follow-up was closed in October 2023. In-hospital mortality, 1- and 3-year survival and freedom of infection at discharge including during follow-up were the primary study end-points. Freedom of infection at discharge and during follow-up was defined as the absence of clinical, radiological, and laboratory signs of AGI analogous to the criteria defined by the Management of Aortic Graft Infection Collaboration group [9]. Postoperative CT scans during follow-up in our aortic department were also screened for omental transposition-associated pathologies, and patients were screened for gastrointestinal symptoms. Causes of death were divided into cardiac or non-cardiac causes or septic multiple organ failure, most likely due to AGI.
Statistical analysis
Continuous variables were tested for a normal distribution using the Shapiro-Wilk test and by visualization using histograms. None of the continuous variables exhibited a normal distribution. Therefore, they were presented as median with corresponding interquartile ranges (25th–75th percentile). Categorical data were represented as absolute numbers with corresponding percentages. Reverse Kaplan–Meier curves were prepared to illustrate the probability of death according to the 2 treatment groups. Statistical analysis was performed using R (The R Foundation for Statistical Computing) version 4.1.1.
RESULTS
Preoperative variables
Preoperative variables are shown in Table 1. One-third (n = 12) of the previous aortic surgeries were elective procedures due to a thoracic aortic aneurysm (ascending and/or arch with or without the need for root replacement), and two-thirds (n = 18) were emergency procedures due to acute type A aortic dissection. There was 1 case of previous fulminant endocarditis involving replacement of the aortic root and ascending aorta. Previous aortic procedures were restricted mainly to ascending/hemiarch replacements in 90% (n = 28). Aortic root replacement was performed in 39% (n = 12) of the cases. Additional AMDS implantation in combination with ascending/hemiarch replacement was performed in 13% (n = 4). Concomitant coronary artery bypass grafting due to coronary artery disease was performed in 19% (n = 6).
Preoperative variables
| Baseline characteristics, n = 31 | n (%)/median (IQR) |
|---|---|
| Gender (female) | 9 (29) |
| Age (years) | 64 (55–71) |
| BMI (kg/m2) | 29 (25–33) |
| Arterial hypertension | 28 (90) |
| Diabetes mellitus | 4 (13) |
| COPD | 5 (16) |
| Renal insufficiency | 9 (29) |
| Coronary artery disease | 7 (23) |
| LVEF | 55 (52–59) |
| Indication for previous aortic surgery | |
| 12 (39) |
| 18 (58) |
| 1 (3) |
| Type of previous aortic/cardiac surgery | |
| 28 (90) |
| 12 (39) |
| 4 (13) |
| 3 (10) |
| 6 (19) |
| Time until omentum transposition (days) | 54 (25–177) |
| Preoperative ICU treatment | 14 (45) |
| Preoperative inotropic medication | 10 (32) |
| Preoperative mechanical ventilation | 10 (32) |
| Preoperative VAC | 10 (32) |
| Euroscore II | 29.09 (16.59–55.34) |
| 15.76 (13.82–25.98) |
| 44.52 (22.41–58.38) |
| Baseline characteristics, n = 31 | n (%)/median (IQR) |
|---|---|
| Gender (female) | 9 (29) |
| Age (years) | 64 (55–71) |
| BMI (kg/m2) | 29 (25–33) |
| Arterial hypertension | 28 (90) |
| Diabetes mellitus | 4 (13) |
| COPD | 5 (16) |
| Renal insufficiency | 9 (29) |
| Coronary artery disease | 7 (23) |
| LVEF | 55 (52–59) |
| Indication for previous aortic surgery | |
| 12 (39) |
| 18 (58) |
| 1 (3) |
| Type of previous aortic/cardiac surgery | |
| 28 (90) |
| 12 (39) |
| 4 (13) |
| 3 (10) |
| 6 (19) |
| Time until omentum transposition (days) | 54 (25–177) |
| Preoperative ICU treatment | 14 (45) |
| Preoperative inotropic medication | 10 (32) |
| Preoperative mechanical ventilation | 10 (32) |
| Preoperative VAC | 10 (32) |
| Euroscore II | 29.09 (16.59–55.34) |
| 15.76 (13.82–25.98) |
| 44.52 (22.41–58.38) |
AMDS®: Ascyrus Medical Dissection Stent; BMI: body mass index; CABG: coronary artery bypass graft; COPD: chronic obstructive pulmonary disease; FET: frozen elephant trunk; ICU: intensive care unit; IQR: interquartile range; LVEF: left ventricular ejection fraction; VAC: vacuum-assisted closure.
Preoperative variables
| Baseline characteristics, n = 31 | n (%)/median (IQR) |
|---|---|
| Gender (female) | 9 (29) |
| Age (years) | 64 (55–71) |
| BMI (kg/m2) | 29 (25–33) |
| Arterial hypertension | 28 (90) |
| Diabetes mellitus | 4 (13) |
| COPD | 5 (16) |
| Renal insufficiency | 9 (29) |
| Coronary artery disease | 7 (23) |
| LVEF | 55 (52–59) |
| Indication for previous aortic surgery | |
| 12 (39) |
| 18 (58) |
| 1 (3) |
| Type of previous aortic/cardiac surgery | |
| 28 (90) |
| 12 (39) |
| 4 (13) |
| 3 (10) |
| 6 (19) |
| Time until omentum transposition (days) | 54 (25–177) |
| Preoperative ICU treatment | 14 (45) |
| Preoperative inotropic medication | 10 (32) |
| Preoperative mechanical ventilation | 10 (32) |
| Preoperative VAC | 10 (32) |
| Euroscore II | 29.09 (16.59–55.34) |
| 15.76 (13.82–25.98) |
| 44.52 (22.41–58.38) |
| Baseline characteristics, n = 31 | n (%)/median (IQR) |
|---|---|
| Gender (female) | 9 (29) |
| Age (years) | 64 (55–71) |
| BMI (kg/m2) | 29 (25–33) |
| Arterial hypertension | 28 (90) |
| Diabetes mellitus | 4 (13) |
| COPD | 5 (16) |
| Renal insufficiency | 9 (29) |
| Coronary artery disease | 7 (23) |
| LVEF | 55 (52–59) |
| Indication for previous aortic surgery | |
| 12 (39) |
| 18 (58) |
| 1 (3) |
| Type of previous aortic/cardiac surgery | |
| 28 (90) |
| 12 (39) |
| 4 (13) |
| 3 (10) |
| 6 (19) |
| Time until omentum transposition (days) | 54 (25–177) |
| Preoperative ICU treatment | 14 (45) |
| Preoperative inotropic medication | 10 (32) |
| Preoperative mechanical ventilation | 10 (32) |
| Preoperative VAC | 10 (32) |
| Euroscore II | 29.09 (16.59–55.34) |
| 15.76 (13.82–25.98) |
| 44.52 (22.41–58.38) |
AMDS®: Ascyrus Medical Dissection Stent; BMI: body mass index; CABG: coronary artery bypass graft; COPD: chronic obstructive pulmonary disease; FET: frozen elephant trunk; ICU: intensive care unit; IQR: interquartile range; LVEF: left ventricular ejection fraction; VAC: vacuum-assisted closure.
The median time between the index aortic surgery and mediastinal omentum transposition was 54 days (25–177). One-third of the patients (n = 10) were in a critical preoperative state, defined by the need for inotropic medication and invasive mechanical ventilation. Preoperative vacuum-assisted closure therapy was used in 10 (32%) to reduce bacterial load and promote tissue regeneration before adding omentum transposition. The median Euroscore II was 29.09 (16.59–55.34) for the cohort as a whole, whereas the median Euroscore II was more than twice as high (44.52 vs 15.76) in patients who were deemed ineligible for reoperation including complete graft replacement and therefore underwent a palliative treatment concept with solely omentum wrapping. Two of 3 patients who received previous total arch replacement underwent a palliative treatment concept and 1 underwent a curative concept including redo total arch replacement.
Postoperative variables
No case of intraoperative death occurred in the study cohort. Postoperative variables are shown in Table 2. Out of 9 patients undergoing a curative concept including complete aortic graft replacement, a biological homograft was used in 55% of patients (n = 5), a custom-made bovine pericardium prosthesis in 33% of patients (n = 3), and a mechanical valved Dacron conduit in 1 patient. Postoperative vacuum-assisted closure due to concomitant sternal wound infection was performed in 19% of patients (n = 6). In all other cases, primary wound closure was achieved. The pathogens identified in cultures of tissue and graft material are shown in Supplementary Material, Table S2. Here, Staphylococcus species were the most common pathogens and were identified in 42% of patients (n = 13, 2 mixed infections). Candida species were identified in 13% of patients (n = 4), and mixed infections with >1 pathogen were also confirmed in 13% of patients (n = 4). In 19% of patients (n = 6), microbiological analysis showed no pathogen. No trend could be observed between patient’s age and the identified pathogens. Gastrointestinal complications occurred in 13% of patients (n = 4); however, in all cases, these were gastrointestinal bleedings that occurred during the stay on the intensive care unit and were not associated with the mediastinal omentum transposition procedure. In-hospital mortality was 0% (n = 0) for patients who underwent reoperation with a curative treatment concept including aortic graft replacement. In these patients, no case of recurrent AGI was present at the time of discharge. In-hospital mortality among patients receiving solely omentum transposition in a palliative treatment concept was 23% (n = 5). Of these patients, 80% (n = 4) died due to septic multiple organ failure in the further course of AGI. The other death was related to sudden haemorrhaging after removing a mediastinal drainage; the patient died on the way to the operating room during cardiopulmonary resuscitation. At the time of discharge, no signs of recurrent or persistent AGI were observed in 59% of patients (n = 13) undergoing solely omentum transposition in a palliative treatment concept. One-year mortality was 5% (n = 1) in patients with no signs of AGI at discharge and 89% (n = 8) for patients with persistent AGI. The results are shown in Table 3. In patients with no signs of AGI at discharge, no case of recurrence of infection, pseudoaneurysm or anastomotic leakage was observed during the follow-up. The probability of death according to the treatment concept is illustrated in Fig. 2 as an inverse Kaplan–Meier curve. No mortality was observed in the curative treatment group during the follow-up. No symptomatic omentum transposition-related complications were reported in the study cohort during follow-up. One case of an intrathoracic upside-down stomach was identified by CT in the survivors during follow-up. However, the patient was asymptomatic and did not suffer from any gastrointestinal symptoms.
Probability of death according to treatment concept.
Mediastinal omentum transposition and wrapping of the aortic graft.
Postoperative variables
| Short-term outcome, n = 31 | n (%)/median (IQR) |
|---|---|
| ICU treatment time (days) | 23 (8–33) |
| Ventilation time (days) | 16 (4–30) |
| Tracheotomy | 14 (45) |
| Postoperative VAC | 6 (19) |
| Gastrointestinal complications | 4 (13) |
| Postoperative stroke | 1 (3) |
| Septic multiple organ failure | 4 (13) |
| Outcome curative concept | 9 (29%) |
| 0 (0%) |
| 9 (100%) |
| Outcome palliative concept | 22 (71%) |
| 5 (23%) |
| 13 (59%) |
| Short-term outcome, n = 31 | n (%)/median (IQR) |
|---|---|
| ICU treatment time (days) | 23 (8–33) |
| Ventilation time (days) | 16 (4–30) |
| Tracheotomy | 14 (45) |
| Postoperative VAC | 6 (19) |
| Gastrointestinal complications | 4 (13) |
| Postoperative stroke | 1 (3) |
| Septic multiple organ failure | 4 (13) |
| Outcome curative concept | 9 (29%) |
| 0 (0%) |
| 9 (100%) |
| Outcome palliative concept | 22 (71%) |
| 5 (23%) |
| 13 (59%) |
ICU: intensive care unit; IQR: interquartile range; VAC: vacuum-assisted closure.
Postoperative variables
| Short-term outcome, n = 31 | n (%)/median (IQR) |
|---|---|
| ICU treatment time (days) | 23 (8–33) |
| Ventilation time (days) | 16 (4–30) |
| Tracheotomy | 14 (45) |
| Postoperative VAC | 6 (19) |
| Gastrointestinal complications | 4 (13) |
| Postoperative stroke | 1 (3) |
| Septic multiple organ failure | 4 (13) |
| Outcome curative concept | 9 (29%) |
| 0 (0%) |
| 9 (100%) |
| Outcome palliative concept | 22 (71%) |
| 5 (23%) |
| 13 (59%) |
| Short-term outcome, n = 31 | n (%)/median (IQR) |
|---|---|
| ICU treatment time (days) | 23 (8–33) |
| Ventilation time (days) | 16 (4–30) |
| Tracheotomy | 14 (45) |
| Postoperative VAC | 6 (19) |
| Gastrointestinal complications | 4 (13) |
| Postoperative stroke | 1 (3) |
| Septic multiple organ failure | 4 (13) |
| Outcome curative concept | 9 (29%) |
| 0 (0%) |
| 9 (100%) |
| Outcome palliative concept | 22 (71%) |
| 5 (23%) |
| 13 (59%) |
ICU: intensive care unit; IQR: interquartile range; VAC: vacuum-assisted closure.
One-year survival according to the status of aortic graft infection
| Outcome | Freedom of AGI at discharge, n = 22, n (%) | Persistence of AGI at discharge, n = 9, n (%) |
|---|---|---|
| Treatment concept | ||
| 9 (41) | 0 (0) |
| 13 (59) | 9 (100) |
| One-year mortality | 1 (5) | 8 (89) |
| Cause of death | ||
| 1 (100) | 0 (0) |
| 0 (0) | 3 (33) |
| 0 (0) | 5 (56) |
| Outcome | Freedom of AGI at discharge, n = 22, n (%) | Persistence of AGI at discharge, n = 9, n (%) |
|---|---|---|
| Treatment concept | ||
| 9 (41) | 0 (0) |
| 13 (59) | 9 (100) |
| One-year mortality | 1 (5) | 8 (89) |
| Cause of death | ||
| 1 (100) | 0 (0) |
| 0 (0) | 3 (33) |
| 0 (0) | 5 (56) |
AGI: aortic graft infection.
One-year survival according to the status of aortic graft infection
| Outcome | Freedom of AGI at discharge, n = 22, n (%) | Persistence of AGI at discharge, n = 9, n (%) |
|---|---|---|
| Treatment concept | ||
| 9 (41) | 0 (0) |
| 13 (59) | 9 (100) |
| One-year mortality | 1 (5) | 8 (89) |
| Cause of death | ||
| 1 (100) | 0 (0) |
| 0 (0) | 3 (33) |
| 0 (0) | 5 (56) |
| Outcome | Freedom of AGI at discharge, n = 22, n (%) | Persistence of AGI at discharge, n = 9, n (%) |
|---|---|---|
| Treatment concept | ||
| 9 (41) | 0 (0) |
| 13 (59) | 9 (100) |
| One-year mortality | 1 (5) | 8 (89) |
| Cause of death | ||
| 1 (100) | 0 (0) |
| 0 (0) | 3 (33) |
| 0 (0) | 5 (56) |
AGI: aortic graft infection.
DISCUSSION
In this retrospective study, we share our experience including the largest currently available clinical series according to current literature on mediastinal omentum transposition for the treatment of AGI of the ascending aorta and/or the aortic arch. Bearing in mind that this study has several major limitations, we believe that transposition of the greater omentum into the mediastinum is an effective, sustainable, and low-risk procedure which can be performed to control infection in the context of thoracic AGI.
Originally, omentum transposition was applied to prevent and treat infections and to lower the risk of bleeding and anastomotic insufficiency in abdominal surgery [11]. It was also used as a ‘filler’ for residual cavities to avoid infection after resection. The immunological properties of the greater omentum are unique and explain its natural ability to contain infectious tissue and support its use for protecting vulnerable surfaces of organs and vessels. The omentum is an immunologically active tissue that embeds lymphoid aggregates which contribute essentially to a variety of immune responses by detecting antigens, particulates, and pathogens in their surroundings [12]. They contain a cluster of T-cells, B-cells, macrophages, and other important components for immune response and are formed around a glomerulus-like knot of blood vessels to provide an additional connection to the lymphatic system [13]. These mechanisms create a unique biological barrier which supports the containment of infections and potentially contributes to the eradication of native and graft infections.
Based on these findings, mediastinal transposition of the omentum has been used for the treatment of AGI with promising results. Shah et al. reported their experience with a cohort of 11 patients, 4 of whom were treated with a curative concept including complete replacement of the infected graft, and 7 of whom with a palliative concept by solely omental wrapping [5]. There was only 1 perioperative death following septic multiple organ failure. A mean follow-up of 3 years revealed a survival rate in excess of 90%. However, information about the preoperative risk profile, a major determinant for the outcome, is lacking. A similar study reported about omentum transposition for the treatment of thoracic AGI in 20 patients, including graft replacement in 11 patients and graft preservation in 9 patients [14]. In-hospital mortality within the preserved graft subgroup was 44%, compared to 18% for graft replacement. Delayed omental flap coverage was associated with a greater mortality than immediate coverage as a 1-step procedure (P = 0.043).
A recent multicentre review analysed the treatment outcomes of thoracic AGI, demonstrating a 1-year survival of 58.6% for all patients [15]. The in-hospital mortality rate for patients who received additional pedicled omental flaps was 25.9% and the risk of in-hospital death increased in the absence of pedicled flaps (muscle or omentum) to cover the graft (P = 0.001). These findings are comparable to the results of our palliative cohort, emphasizing the potential of omental wrapping in patients in whom graft replacement is not feasible.
The dismal results of non-operative management of thoracic AGI should encourage an aggressive surgical treatment [16]. The presence of concomitant sternal wound infection, the type of pathogen, the presence of a pseudoaneurysm, previous CABG, aortic root or total arch replacement may be considered additionally during the individual decision process but should not represent definite criteria for or against a curative or palliative approach in general. Especially in case of candida infection, very early and aggressive radical surgical treatment may offer the only chance for patients. Delayed reconstruction in case of unstable conditions due to septicaemia may be a feasible option in selected cases but should be assessed on an individual basis. Only after a multidisciplinary aortic team has discussed the possibility of a curative concept including radical resection and replacement of the infected aortic graft in terms of surgical feasibility and the patient’s condition, isolated omental wrapping should be performed as a bailout procedure. Four cases with AGI including AMDS were observed, which might be concerning. In case of AGI involving an AMDS prosthesis, curative treatment should include removal of the dacron graft and the cuff that are more prone for infection. The felt cuff of the AMDS may represent a vulnerable spot for AGI due to its surface for circulating pathogens. However, this remains speculation and no conclusions can be drawn at this point of time.
According to our findings, early-onset AGI were more likely to be the result of fulminant mediastinitis and sternal wound infection during the postoperative course, whereas late-onset AGI which developed slowly seemed to be caused by bacteraemia and led to coexisting mediastinitis. However, this is just an observation, and no causalities can be drawn from this. Primary goal is closure of the sternum after omentum transposition. In several cases we did not perform it to prevent compression in case of haemodynamic instability. To prevent postoperative compression, it is of upmost importance to open both pleurae with deep incisions, though it is not always possible because of severe adhesions. According to the drainage management, we recommend using pleural drainages on both sides after opening of the pleurae instead of mediastinal drainages. If mediastinal drainages are urgently needed, we recommend using BLAKE® drainages to lower the risk for obstruction and damage in case of removal. Additionally, we recommend using an ‘easy-flow’ drainage system for the abdomen after omentum transposition.
Since we did not observe any omentum transposition-related gastrointestinal complications, and given the potential benefits, transposition of the greater omentum should be encouraged additionally for the treatment of AGI. If the expertise for dissecting the omentum is not readily available within the thoracic surgical team, an interdisciplinary approach with the assistance of plastic and/or abdominal surgeons is recommended.
Limitations
This study is limited by its retrospective and unicentric nature as well as by the size of the study cohort. The data should be interpreted with caution due to the limited sample size. The reported cohort lacks a comparator with thoracic AGI and either radical replacement without omentum transposition or conservative medical treatment alone, which leads to careful interpretation of the study results. The fact that the decision for the treatment concept was made on an individual basis considering the patients’ condition and the technical and anatomical feasibility results in difficult comparability, but also highlights the patient-tailored treatment approach.
CONCLUSIONS
The poor results of non-operative management of thoracic AGI should encourage an aggressive surgical treatment approach, preferably with resection of all infected tissue and graft material. Transposition of the greater omentum is an effective supportive measure for radical treatment after complete replacement of the infected aortic graft to prevent re-infection and stabilize the suture lines. When combined with surgical debridement and adequate systemic anti-infective treatment, it also may serve as a bailout strategy for patients in very poor status and/or in the case of difficult local surgical conditions. However, mortality stays high. The individual decision for the treatment concept should be guided by a multidisciplinary aortic team in centres experienced in AGI.
SUPPLEMENTARY MATERIAL
Supplementary material is available at EJCTS online.
FUNDING
No funding was provided.
Conflict of interest: none declared.
DATA AVAILABILITY
The data underlying this article are available in the article and in its online supplementary material.
Author contributions
Leonard Pitts: Conceptualization; Data curation; Formal analysis; Investigation; Methodology; Project administration; Validation; Visualization; Writing—original draft; Writing—review & editing. Miralem Pasic: Conceptualization; Data curation; Investigation; Project administration; Supervision; Validation; Writing—review & editing. Leonhard Wert: Supervision; Writing—review & editing. Gaik Nersesian: Supervision; Writing—review & editing. Julius Kaemmel: Supervision; Writing—review & editing. Semih Buz: Conceptualization; Data curation; Supervision; Validation; Writing—review & editing. Christoph Knosalla: Conceptualization; Data curation; Supervision; Writing—review & editing. Volker Düsterhöft: Conceptualization; Data curation; Supervision; Writing—review & editing. Christoph Starck: Supervision; Writing—review & editing. Jörg Kempfert: Supervision; Validation; Writing—review & editing. Stephan Jacobs: Supervision; Writing—review & editing. Volkmar Falk: Project administration; Supervision; Validation; Writing—review & editing.
Reviewer information
European Journal of Cardio-Thoracic Surgery thanks Jürg Schmidli, Efrem Civilini, Hitoshi Ogino and the other anonymous reviewers for their contribution to the peer review process of this article.
REFERENCES
ABBREVIATIONS
- AGI
Aortic graft infection
- AMDS®
Ascyrus Medical Dissection Stent®
- CT
Computed tomography
