Abstract
Single-stage repair of infected aortic grafts through a clamshell incision (bilateral thoracotomy with transverse sternotomy) is rarely undertaken, even when this approach could offer better exposure. We describe the use and considerations of a clamshell approach in the context of a rare prosthetic graft infection involving ascending, arch, and descending aortic segments.
A 59-year-old man underwent aortic valve plus ascending aorta and total arch replacement for acute type A dissection. Eleven months later, he presented with a contained rupture at the distal anastomosis of his aortic arch graft. A thoracic endovascular aortic repair was performed extending into the descending aorta. However, within a few weeks, the patient developed fever and persistent distal anastomotic pseudoaneurysms, suggesting graft infection (without an obvious culprit organism). Replacement of the entire aortic arch, ascending, and descending aorta, preserving the aortic valve, was ultimately performed via a clamshell incision, requiring total circulatory arrest (eight minutes) and antegrade cerebral perfusion (28 minutes). Postoperatively, the patient was ventilated for 2.7 days and was discharged home on postoperative day 26 with empirical antibiotic therapy. The culprit organism was later identified as Parvimonas micra (P. micra) following extensive microbiological analysis (16S ribosomal deoxyribonucleic acid sequencing). The patient presented with hemoptysis and recurrent signs of infection on aortic imaging at approximately three months following his final operation. After discussion with the patient and family, nonoperative management was determined as the most appropriate path forward.
This case illustrates the technical feasibility of this approach to address graft material present in the ascending, arch, and descending aorta. While the outcome was not ideal for this patient, our approach made this very extensive removal of graft material possible; such a procedure would not have been feasible with a midline sternotomy and would have required a multi-stage repair. Further, P. micra was an unexpected culprit, difficult to detect and detected late; this underscores the potentially growing importance of considering unexpected culprits and using advanced sequencing in culture-negative graft infections.
Keywords: bilateral thoracosternotomy, clamshell incision, parvimonas micra, prosthetic aortic graft infection, reoperation
Introduction
Effective management of aortic graft infections represents one of the most formidable challenges in cardiovascular surgery. These infections occur in approximately 1% of aortic grafts and, while rare, are associated with high morbidity and mortality due to their insidious onset, diagnostic complexity, and the technical demands of surgical management [1,2]. The presence of prosthetic material in a contaminated field, coupled with the proximity to vital mediastinal structures, further complicates eradication and reconstruction efforts [3].
Among existing surgical approaches, the clamshell incision (bilateral anterior thoracotomy with transverse sternotomy) offers unparalleled exposure to the aortic root and the ascending, arch, and descending segments. The clamshell approach for addressing thoracic aortic pathology has been described with excellent outcomes in reoperative settings; however, because of its invasiveness and unfamiliarity, many cardiac surgeons appear to avoid it in the context of graft infections. It is well established that approaching multiple aortic segments, including the aortic arch, necessitates periods of circulatory arrest with some cerebral perfusion strategy [4]. Taken together, the perceived technical complexity of this procedure has likely contributed to surgeons' persistent reluctance to use it.
The clamshell approach has been described primarily in the context of aneurysmal disease and has not, to our knowledge, been employed for infected graft material; in these situations, staged interventions have largely been favored [5,6]. Therefore, this report describes an extensive aortic graft infection in a patient who had previously undergone a total arch replacement and subsequent thoracic endovascular aortic repair for acute type A aortic dissection. Our case illustrates the importance of the clamshell approach for simultaneously addressing ascending, arch, and descending aortic pathology and supports a more robust consideration of this technique in similar contexts.
Case presentation
Preface
A 59-year-old man with an unremarkable past medical history presented with an acute type A aortic dissection. The patient underwent emergent ascending and total aortic arch replacement. At that time, the supra-aortic vessels were reconstructed as an island due to adequate tissue quality. A surgical aortic valve replacement (25 mm Resilia™, 11500A, Inspiris, CA, USA) was performed concomitantly. CT at eight months postoperatively demonstrated an excellent result (Figure 1A).
Figure 1. Initial infection was repaired endovascularly. Sagittal CT demonstrates (A) a patent synthetic arch graft (red arrows) and anastomotic site (blue arrow) at nine months after open repair of a type A dissection. Follow-up CT at 11 months (B) reveals probable mediastinal hematoma (red arrow) of the native graft alongside false aneurysm of the distal arch (blue arrow). This was repaired endovascularly without residual hematoma, as confirmed by (C) a postoperative arch aortogram.
CT: computed tomography
At 11 months after surgery, the patient presented with a small blister at the distal end of his sternal incision, which produced sanguineous discharge after being lanced. A sample was sent for culture. The sternum and incision were otherwise healed and stable. Associated symptoms included a dry cough; at the time, the patient was afebrile and had no elevation in white blood cell (WBC) count. A CT angiogram suggested a contained rupture with a false aneurysm and mediastinal hematoma. A contained rupture at the distal arch was confirmed angiographically. This was corrected endovascularly using a covered stent graft (15 cm × 3.4 cm, Zenith Alpha™, ZTA-P-34-161-W, Cook Medical, United States), with a zone-two proximal landing zone and a zone-four distal landing zone. A bare-metal dissection stent (12 cm × 4.6 cm, Zenith Alpha™, ZDES-46-120-US) was then deployed, extending to the distal abdominal aorta using a target landing zone of 2 cm on the existing 26 mm arch graft (Figure 1C). Given the blister on the anterior sternum, additional incisional tissue from the sternum was sent for microbiological analysis. The initial postoperative course was unremarkable, with resolution of the dry cough. The sternal wound healed well.
Four weeks after discharge from the hospital, the patient became persistently febrile without clear positive bacterial cultures and received intravenous piperacillin/tazobactam (3.375 mg QID) followed by vancomycin (1250 mg UID). A tagged WBC scan was positive, demonstrating uptake at the junction of the arch graft and endovascular graft (Figure 2A-2B). A subsequent CT scan demonstrated multiple areas of contrast extravasation and gas surrounding the graft, suspicious for graft infection (Figure 2C). Despite ongoing antibiotic therapy, signs and symptoms of infection persisted, with evidence of compromised graft integrity demonstrated by contrast leakage. This represented a clear indication for surgical removal of all existing graft material. Given the extensive graft material involving the ascending, arch, and descending aorta, we did not believe that a median sternotomy would provide sufficient exposure to remove all infected material. To maximize operative exposure and enable a single-stage repair, a clamshell incision was chosen to facilitate the removal and replacement of the ascending, arch, and descending aortic grafts.
Figure 2. Persistent infections necessitated further intervention. (A) Axial and (B) coronal leukocyte uptake scans demonstrated distinct immunologic presence at the graft site one month after endovascular intervention. An axial CT of the arch to descending aorta (C) revealed persistent sanguinous leakage at the arch accounted for by ruptured graft material (blue arrows) with air bubbles (red arrows) surrounding the descending stent graft.
CT: computed tomography
Operative technique
The patient was anesthetized, and arterial monitoring lines were placed in the brachial and femoral arteries. Cardiopulmonary bypass was established via femoral artery-femoral vein cannulation, and the patient was cooled to 20°C. A bilateral thoracotomy and transverse sternotomy were performed through the fourth intercostal space, and the ascending, arch, and descending thoracic aorta were mobilized. Once the target temperature was reached, the ascending aorta was cross-clamped, the heart was arrested with antegrade cardioplegia, and the descending aorta was clamped at the level of the diaphragm. Lower body perfusion was maintained. The subclavian and innominate arteries were snared, and the left carotid artery was cannulated with a soft balloon-tip catheter. Cerebral perfusion was maintained at a flow rate of 2.5 L/min. Once cerebral perfusion was established, all existing grafts were removed, followed by reconstruction of the head vessels using a multibranched plexus graft (28 mm × 40 cm × 4 cm, Gelweave™ Plexus, 73281088/8, Terumo, United Kingdom), which was anastomosed to the innominate, subclavian, and left carotid arteries. This allowed restoration of antegrade perfusion to the entire upper body through the antegrade limb of the graft.
Lower body perfusion was then discontinued, and the cross-clamp at the diaphragm was removed to facilitate extraction of all remaining dissection stents from the descending aorta. The distal end-to-end anastomosis of the plexus graft to the descending aorta at the level of the diaphragm was then completed. Lower body perfusion was subsequently re-established via the antegrade limb of the plexus graft. Finally, the proximal anastomosis to the sinotubular junction was completed, and myocardial reperfusion was initiated. The total cardiopulmonary bypass time was 300 minutes, the aortic cross-clamp time was 137 minutes, and the circulatory arrest time was eight minutes. Antegrade cerebral perfusion was maintained for 28 minutes. Inspection of the aortic valve demonstrated endothelialized cuff surfaces without evidence of structural abnormalities. The aortic valve was therefore preserved.
Postoperative course
The patient was extubated after 2.7 days and was subsequently weaned from supplemental oxygen and vasoactive support. Transient seizure activity on postoperative day (POD) 2 resolved with levetiracetam (500 mg BID), and no other neurological deficits were observed. No tissue cultures yielded bacterial growth. However, 16S ribosomal deoxyribonucleic acid polymerase chain reaction analysis of a tissue sample from the false aortic lumen identified the anaerobe Parvimonas micra (P. micra). Once the causative organism was identified, postoperative antimicrobial therapy was tailored to include vancomycin (1250 mg UID), daptomycin (500 mg BID), and ertapenem (1000 mg BID). The patient recovered well, was discharged from the intensive care unit on POD 9, and was discharged home on POD 26 with ongoing antibiotic therapy.
Unfortunately, despite extensive efforts to eradicate the infection, persistent signs of infection recurred approximately three months after the clamshell procedure, with interval CT changes and intermittent fever. The patient remains on long-term antibiotic therapy; however, management has since focused on quality of life and nonoperative treatment. At this stage, no reasonable surgical options remain.
Discussion
This patient developed a refractory graft infection after total arch replacement for an acute type A dissection. The causative organism was later identified as P. micra, a commensal of the oral cavity that is capable of colonizing prosthetic material following hematogenous dissemination. P. micra bacteremia is exceedingly rare in patients without classic risk factors (e.g., periodontitis); therefore, its presence was unexpected in our patient. Our experience adds to a growing body of clinical observations highlighting increasing antibiotic resistance and the emergence of unusual bacteremia in postsurgical patients, including cardiac surgical patients [7,8].
Because the infection persisted despite antimicrobial therapy, aggressive surgical intervention was ultimately required. We believe that the clamshell approach was the optimal strategy for the simultaneous resection of all infected graft material. This goal would not have been achievable with conventional surgical approaches. From a technical standpoint, the reoperation was more straightforward than anticipated, likely because the clamshell incision provided access through tissue planes that remained relatively undisturbed compared with those encountered during the original median sternotomy for type A dissection repair. Although the immediate surgical outcome was favorable, the persistent infection at five months likely reflected P. micra's propensity to adhere to prosthetic material and form biofilms. This was probably compounded by the delayed identification of the organism, which ultimately required external microbiological analysis for diagnosis [9]. This case therefore underscores the importance of early recognition and targeted management of P. micra infections.
This report illustrates a unique clinical scenario in which the clamshell approach enabled complete removal of extensive infected aortic graft material. Such complex procedures are often approached with caution by both patients and surgeons because of their technical demands. Nevertheless, our experience demonstrates that, with meticulous planning, careful preparation, and multidisciplinary collaboration, these challenging cases can be managed successfully.
Although complete eradication of the infection was not achieved, the operative objectives were accomplished and would not have been feasible through more conventional approaches such as median sternotomy. In carefully selected patients, surgical success should be judged not only by long-term infectious control but also by the ability to achieve definitive anatomic correction and meaningful functional recovery in the setting of otherwise life-threatening disease.
From the patient's perspective, this operation represented the only realistic opportunity to avoid imminent death from progressive graft destruction and overwhelming infection. Despite its technical complexity, the procedure enabled meaningful recovery, discharge home, and a return to an acceptable quality of life following a prolonged critical illness. Regaining functional independence and spending additional time at home with preserved autonomy were outcomes that the patient and family considered highly valuable.
The delayed identification of P. micra likely contributed to persistent infection despite aggressive surgical and antimicrobial therapy. Early consideration of atypical pathogens and broader implementation of molecular sequencing techniques may therefore represent important adjuncts for optimizing outcomes in similar cases.
Conclusions
While the clamshell approach remains underutilized due to its perceived complexity, our experience supports considering it a viable strategy for managing extensive infected thoracic aortic pathology. With meticulous preoperative planning, multidisciplinary collaboration, and careful cerebral protection strategies, this approach can expand the surgical options available to patients who might otherwise have only limited or palliative treatment.
Disclosures
Human subjects: Informed consent for treatment and open access publication was obtained or waived by all participants in this study.
Conflicts of interest: In compliance with the ICMJE uniform disclosure form, all authors declare the following:
Payment/services info: All authors have declared that no financial support was received from any organization for the submitted work.
Financial relationships: All authors have declared that they have no financial relationships at present or within the previous three years with any organizations that might have an interest in the submitted work.
Other relationships: All authors have declared that there are no other relationships or activities that could appear to have influenced the submitted work.
Author Contributions
Acquisition, analysis, or interpretation of data: Malcolm J. MacKinnon
Drafting of the manuscript: Malcolm J. MacKinnon, Christopher W. White, Jean-François Légaré
Critical review of the manuscript for important intellectual content: Malcolm J. MacKinnon, Darren B. Ferguson, Christopher W. White, Jennifer Cloutier, Jean-François Légaré
Concept and design: Darren B. Ferguson, Christopher W. White, Jennifer Cloutier, Jean-François Légaré
Supervision: Jean-François Légaré
References
- 1.Characteristics and outcome of vascular graft infections: a risk factor and survival analysis. Stockschläder L, Margaryan D, Omran S, Schomaker M, Greiner A, Trampuz A. Open Forum Infect Dis. 2024;11:271. doi: 10.1093/ofid/ofae271. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Incidence, management, and outcomes of aortic graft infection. Shiraev T, Barrett S, Heywood S, et al. Ann Vasc Surg. 2019;59:73–83. doi: 10.1016/j.avsg.2019.01.027. [DOI] [PubMed] [Google Scholar]
- 3.Treatment of infections associated with surgical implants. Darouiche RO. N Engl J Med. 2004;350:1422–1429. doi: 10.1056/NEJMra035415. [DOI] [PubMed] [Google Scholar]
- 4.Single-stage extensive replacement of the thoracic aorta: the arch-first technique. Rokkas CK, Kouchoukos NT. J Thorac Cardiovasc Surg. 1999;117:99–105. doi: 10.1016/s0022-5223(99)70473-3. [DOI] [PubMed] [Google Scholar]
- 5.Clamshell thoracosternotomy for single-stage repair of select aortic arch pathologies. Prabhu NK, Vekstein AM, Jensen CW, Williams AR, Gaca JG, Hughes GC. JTCVS Open. 2025;23:44–50. doi: 10.1016/j.xjon.2024.10.032. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Staged repair for infected aortic arch graft and branched endograft post type A dissection repair. Tavolacci SC, Spielvogel D, Ohira S. Multimed Man Cardiothorac Surg. 2025;2025 doi: 10.1510/mmcts.2025.009. [DOI] [PubMed] [Google Scholar]
- 7.Pacemaker infection and endocarditis due to Parvimonas micra: a case report and systematic review. Suzuki T, Ishikawa K, Matsuo T, et al. Anaerobe. 2021;72:102459. doi: 10.1016/j.anaerobe.2021.102459. [DOI] [PubMed] [Google Scholar]
- 8.Parvimonas micra bacteremia: a rare complication after esophagogastroduodenoscopy for upper gastrointestinal bleeding. Raza HS, Love JS, Mikolajczyk AE. ACG Case Rep J. 2024;11:1378. doi: 10.14309/crj.0000000000001378. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Polymicrobial late-onset knee prosthetic joint infection involving Parvimonas micra: a case report and genomic characterization. Salles MJ, Pereira DL, Oliveira ÍS, et al. Microorganisms. 2026;14:413. doi: 10.3390/microorganisms14020413. [DOI] [PMC free article] [PubMed] [Google Scholar]


