Abdominal aortic endograft infection with Clostridium difficile

Introduction

Aortic graft infection is a devastating complication of endovascular aortic repair. Luckily, this complication is rare with a reported incidence as low as 0.2% and up to 5% in small series.^1–4 The most common organisms associated with endograft infections are Gram-positive organisms, namely, Streptococcus and Staphylococcus species, followed by polymicrobial infections, Gram-negative infections, and fungal infections.^3–5

Other organisms, the atypicals, make up about 2% of the cultured micoorganisms. ⁵ Though rare, as endovascular aortic aneurysm repair has increased, so too have reports of endograft infections, including atypical infections. Cases of atypical infections with clostridia, Mycobacterium avium complex, Listeria monocytogenes, and Cutibacterium acnes (formerly Propionibacterium acnes) have been reported. We report a Clostridium difficile (C. difficile) culture–positive aortic endograft infection. To our knowledge this is the second report of an abdominal aortic endograft infection with C. difficile. ⁶

Case report

Informed consent for the publication of this case report was obtained from the patient and documented accordingly per institutional guidelines. The patient is an 82-year-old male with a history of a 6.4-cm infrarenal abdominal aortic aneurysm. He previously underwent endovascular aortic repair with a Gore Excluder endograft. He presented as a transfer to our facility 3 years later after routine follow-up imaging demonstrated peri-aortic inflammation (Figure 1). The patient endorsed intermittent fevers, chills, and right flank and abdominal pain over the past 3 weeks. He also noted around the time his symptoms started, he had a 3-day period of diarrhea, abdominal cramping, and nausea. He was not tested for C. difficile infection at the time and denied a history of C. difficile infection in the past. He also denied any recent antibiotic treatment. He was hemodynamically stable and afebrile. Labs were notable for a leukocytosis on admission. Blood cultures were drawn and were negative. He was admitted to the ICU and started on broad spectrum antibiotics. Gastroenterology was consulted due to concerns for an aortoenteric fistula as plans for operative repair were developed. An esophagogastroduodenoscopy was performed and was negative for an aortoenteric fistula, ulcer, or inflammation.

OPEN IN VIEWER

Figure 1.

Abdominal aortic endograft within rim-enhanced, inflamed aorta.

He was taken to the operating room and underwent an open, complete explant of his endograft and in-situ repair using a rifampin-soaked Dacron graft that was wrapped with an omental pedicle flap. This was performed via a transperitoneal approach. The fourth portion of the duodenum was thickened, inflamed, and adherent to the aneurysm sac. Once carefully dissected, no fistulous tracts nor enterotomies of the duodenum were noted, and therefore, no bowel resection was performed. The remainder of the small and large bowel appeared normal without evidence of inflammation or infection. The aneurysm sac was filled with thick, purulent material and thrombus. Tissues cultures and Gram stains of the aortic contents were sent. The tissue culture of aortic contents grew C. difficile. Two other samples of the aortic contents and the explanted endograft did not identify C. difficile infection nor any other organisms. The specific strain of C. difficile was not tested for. Infectious disease was consulted. Post-operatively, a C. difficile stool polymerase chain reaction was sent and was positive for C. difficile toxin. A toxin enzyme immunoassay test was performed and was negative. He remained on intravenous vancomycin and piperacillin-tazobactam for presumed infection with polymicrobial enteric pathogens not detected in operative culture. A peripherally inserted central line was placed, and he was discharged on 6 weeks of intravenous antibiotic therapy from the date of surgery.

The patient did well post-operatively and was discharged home with home healthcare on post-operative day eight. He followed up in the clinic 1 month later. His recovery remained uneventful, and he was compliant with his intravenous antibiotic therapy. He completed 6 weeks of intravenous antibiotics and was transitioned to oral sulfamethoxazole-trimethoprim for lifelong suppression. He continues to do well.

Discussion

Clostridial infections have been reported in both native aortic infections and aortic endograft infections. Among the Clostridium species, there are more reports of Clostridium septicum (C. septicum) aortic infections than C. difficile aortic infections.^6,7 C. septicum is an anaerobic, spore-forming, toxin-producing, Gram-positive Bacillus. It is associated with malignancy, most commonly hematologic and colonic malignancies. ⁸ C. septicum virulence is attributed to its toxin, alpha toxin, which causes gas gangrene. A recent seven-patient case series on C. septicum–infected aortic aneurysms and aortic endografts reported 100% mortality in the two patients who did not undergo surgical intervention and average 1 year mortality of 75% in the remaining five patients who did undergo surgery. ⁷

C. difficile is also an anaerobic, spore-forming, toxin-producing, Gram-positive Bacillus. C. difficile is a prevalent strain of clostridia. Up to 5% of adults are colonized with C. difficile, and of those 5%, about 25–30% will develop symptoms.^9,10 The spectrum of disease manifestation ranges from asymptomatic carriers to fulminant colitis. Fulminant colitis has morbidity and mortality rates up to 34% in patients requiring ICU-level care. C. difficile produces two toxins, toxin A and toxin B, also known as enterotoxin A and cytotoxin B, respectively. ⁹ These toxins lead to the inflammatory response and tissue destruction associated with C. difficile infection. Risk factors for the development of C. difficile infection include antibiotic exposure, hospitalization, institutionalization in nursing homes, and age > 65 years. ⁹ Nearly all antibiotics have been associated with the development of C. difficile infection, and therefore, any antibiotic exposure is relevant to its diagnosis. There have also been significant increases in community-acquired cases defined as C. difficile infection in patients without an overnight hospital stay within 12 weeks of the infection. ¹⁰

Our patient’s presentation is most consistent with a mild C. difficile infection in the weeks prior to him presenting with endograft infection. His infection likely developed from being a carrier or from a community-acquired infection. He also had presumed polymicrobial infection not detected on intra-operative cultures due to pre-operative intravenous antibiotic therapy. In contrast to the more severe presentation of the previously reported C. difficile aortic endograft infection with bacteremia, our patient underscores the spectrum of disease patients can present with. ⁶

Most patients with infected aortic endografts have at least one risk factor for developing infection including, but not limited to, endovascular reinterventions, specifically reinterventions for endoleaks, aortoduodenal fistulae, urosepsis, pneumonia, bowel surgery, groin infections, spine infections, and recent tooth extraction. ³ This complex problem is further complicated by rare infections. Unless the patient is at prohibitive risk, the management of infected aortic endograft is surgical due to the high mortality rate with medical management. Both medical and surgical treatments are tailored to the microorganism. Further challenges emerge when pre-operative cultures cannot be obtained or rare infections are identified. As a result, decision making in these scenarios is educated and experiential, not evidenced based.

There are four options for the treatment of an infected aortic endograft: three in-situ reconstructions and one ex-situ, extra-anatomic reconstruction. The three in-situ reconstruction options are cryopreserved allograft reconstruction; neoaortoiliac system; and antibiotic-soaked, prosthetic repair. The antibiotic soaking improves survival of patients repaired with prosthetic graft repair. ⁸ Of the tested antibiotics, rifampin has a better affinity to bind to Dacron and is used for antibiotic soaking. It covers Staphylococcus aureus, Streptococcus species, as well as Gram-negative bacteria. The ex-situ, extranatomic reconstruction consists of an axillofemoral bypass followed by explant and over sewing of aortic and iliac stumps. Each repair offers advantages and disadvantages and should be individualized to each patient based on their risks factors and the risks each option poses to the patient. Additionally, there are institutional preferences that influence the type of repair. At our institution, we prefer in-situ repair when appropriate due to the increased risks of limb complications and aortic stump issues with ex-situ, extra-anatomic reconstruction. We also prefer either neoaortoiliac reconstruction or antibiotic-soaked prosthetic repair because of the potential for aneurysmal degeneration of cryopreserved allografts. Even though for his age this patient was of good functional status, an in-situ reconstruction with rifampin-soaked Dacron was selected over a neo-aortoiliac reconstruction to decrease operative morbidity and time in this critically ill, elderly patient.

There is no consensus on the long-term, post-operative antibiotic therapy for infected aortic endografts. Most patients are treated with 4 to 6 weeks of intravenous antibiotic therapy with further long-term antibiotic therapy determined after reassessment of patient-specific risk factors. Atypical infections again present yet another treatment challenge as efficacy and long-term outcomes are not known. At the Mayo Clinic, the institutional criteria for long-term suppressive antibiotics include virulent organisms, abscess, frank pus and partial graft excision. ³ Our institutional preference is to continue long-term suppressive antibiotic therapy if it is well tolerated by the patient. Therefore, the plan for this patient is to continue on lifelong suppressive therapy with oral sulfamethoxazole-trimethoprim.

This case highlights a few important points when treating atypical infections of the aortic endograft. First, the importance of a thorough history to include any recent symptoms and risk factors for aortic endograft infection such as interval infections, reinterventions, and antibiotic use. Second, treating with broad spectrum antibiotic coverage and then tailor treatment based on pre-operative and intra-operative culture data. There should be involvement of infectious disease specialists in a multi-disciplinary fashion. Finally, considerations should be made to cover less common pathogens if the history is suggestive as more atypical infections are being identified.

Conclusion

We report a C. difficile aortic endograft infection in a patient with mild C. difficile infection in the weeks prior to presentation. This case illustrates successful surgical treatment of an infected aortic endograft and underscores the importance of broad spectrum antibiotic coverage until all culture data are finalized and multi-disciplinary care for this complex patient population.