Management of early graft infections in the ascending aorta and aortic arch: a comparison between graft replacement and graft preservation techniques

OBJECTIVES: Treatment of infected thoracic aortic grafts is associated with considerable morbidity and mortality. The replacement of an infected graft is an effective strategy, yet a complex surgical endeavour, especially if the aortic root or aortic arch is involved. In situ graft-sparing surgical therapy with continuous mediastinal irrigation after surgical debridement might offer an alternative in the management of early graft infections in the thoracic aorta. METHODS: Between 1996 and August 2015, 25 patients were treated in our institution for early graft infection after thoracic aortic surgery via sternotomy. In 11 patients, the infected prosthesis was replaced by a cryopreserved homograft or a biological valved pericardial xenograft. In 14 patients, an attempt to salvage the graft was made by resternotomy, aggressive debridement and subsequent continuous mediastinal antibiotic irrigation over a course of 2 weeks, accompanied by systemic antibiotic therapy. RESULTS: In-hospital mortality was comparable (replacement group: 2/11 = 18%, graft-sparing group: 2/14 = 14%, P = ns). The time interval from the initial surgery was significantly shorter in the graft-sparing group (replacement group: 165 days [range 95-300 days] and graft-sparing group: 24 days [range 15-93 days], P = 0.004]. Two patients (14%), who were treated with the graft-sparing approach >100 days after the initial surgery, were retreated for infection, and 1 due to an intra-aortic infection of an aortic arch hybrid stent graft was not amenable to external irrigation (median follow-up: 1.5 years [range 1.1-2.1 years]). One patient in the replacement group (9%) was reoperated on due to homograft degeneration (median follow-up: 6.0 years [3.0-8.9 years]). CONCLUSIONS: In situ graft-sparing surgical therapy is safe and effective if diagnosis and treatment of aortic graft infection is initiated promptly and aggressively (ideally <1 month post-surgery). Our method produces good midterm results (3 years). For aortic graft infections that become clinically apparent >3-6 months after surgery, replacement of grafts with biological conduits (homografts or pericardial xenografts) most likely remains the best treatment option.