Aortic Stent Graft Infection in a Patient with Marfan Syndrome.
Fu Wei-guo,Shi De-bing,Wang Yu-qi,Guo Da-qiao,Chen Bin,Shi Zhen-yu
DOI: https://doi.org/10.1097/00029330-200705020-00016
IF: 6.133
2007-01-01
Chinese Medical Journal
Abstract:Endovascular repair of ruptured abdominal aortic aneurysm (AAA) has been adopted clinically.1,2 It is especially suitable for the patients who are at high surgical risk or fulfil the technical requirements for endovascular therapy. The use of endoprostheses, however, may raise the risk of infection, as occurred in our AAA patient with Marfan syndrome who had stent-graft infection as a result of Salmonella choleraesuis. CASE REPORT A 57-year-old man with Marfan syndrome was admitted to our department because of severe epigastric pain for 2 months. He had a history of essential hypertension. Emergency CT scan indicated a ruptured infrarenal aortic aneurysm with a maximum transverse diameter of 4.2 cm and the proximal neck of 5.4 cm (Fig. 1A). A 26 × 80-mm tube stent-graft (Shenzhen Lifetech Scientific Inc., China) was then implanted to repair the aneurysm with the polytetrafluoroethylene (PTFE) just below the renal artery and the distal graft terminating in the aortic bifurcation (Fig. 1B). The procedure was performed in a catheterization laboratory equipped for surgical conversion. Cefuroxime at a dose of 1.5 g was administered for perioperative antibiotic prophylaxis.Fig. 1. A:: First preoperative arteriography showing an eccentric ruptured infrarenal aortic aneurysm; B: First postoperative immediate arteriography showing a 26×80-mm straight tube graft positioned satisfactorily; C: Second preoperative arteriography showing a 5-mm migration of the endograft from the aortic bifurcation and a significant dilation of the distal aorta; D: Second postoperative immediate arteriography showing an aortomonoiliac endograft implanted.The postoperative course was uneventful. Spiral CT angiography at one month of follow-up showed neither leakage and expansion of the aneurysm sac, nor inflammatory reaction. Ten weeks later, however, the patient was readmitted for hypogastric and low back pain radiating to the left inguinal region. Enhanced CT scan and arteriography demonstrated persistent distal type I endoleak with a 5-mm migration of the endograft from the aortic bifurcation and a significant dilation of the distal aorta (Fig. 1C). The second procedure was performed at the same catheterization laboratory. An aortomonoiliac endograft (Microport Medical (Shanghai) Co. Ltd., China) was deployed with its distal end terminating in the right external iliac artery (EIA), followed by an occluder in the left common iliac artery (CIA) and a femorofemoral crossover. A second extension piece (Microport Medical (Shanghai) Co. Ltd., China) was required to seal a proximal type I endoleak. Arteriography confirmed that the exclusion of the aneurysm was satisfactory with no endoleak (Fig. 1D). The whole operative procedure lasted for 4.5 hours. Flucloxacillin at a dose of 1 g was given intravenously twice during the procedure, and postoperative antibiotics were given until six days later when the patient presented a fever peaking at 39.8 °C with serious chills, accompanied by malaise and decreased appetite. Blood analysis showed white blood cell count 20.4 × 109/L, neutrophils 90.5%, haemoglobin 79 g/L, hematocrit 24.6%, platelet count 164 × 109/L, blood urea nitrogen 6.6 mmol/L, and serum creatinine 66 μmol/L. The erythrocyte sedimentation rate (ESR) was 90 mm/h (normal <10–15 mm/h) on the 8th day after the second procedure. Meanwhile, blood culture identified the growth of Salmonella choleraesuis, but cultures of urine and sputum showed nothing abnormal. The antibiotics regimen was changed with imipenem (1 g, three times daily) and vancomycin hydrochloride (400 mg, four times daily) in stead of flucloxacillin (1.5 g, twice daily). However, no significant response was observed in the patient. Echocardiography ruled out endocarditis-related infected aneurysm but revealed no evidence of endocardial vegetation. Enhanced CT showed perigraft collection of air and fluid (Fig. 2), indicating an infection of the stent-graft. Laparotomy and other surgical treatments were applied.Fig. 2.: CTA showing perigraft air and fluid collection (arrow).On the 14th day after the second procedure, surgical exploration revealed a pronounced inflammatory reaction and scarring around the stent-graft. Abundant reddish-brown fluid and aortic thrombus in the aneurysm sac appeared to be gelatinous, semi-liquefied, and purulent (Fig. 3A). There was no fresh blood in the sac and the stent-graft struts were intact witout damage to the fabric (Fig. 3B). The purulent fluid was drained and the infected tissue around the stent-graft was carefully debrided. Then two irrigation tubes were inserted into the aneurismal sac and along the endograft. Finally the aneurysm sac was closed and wrapped with the retroperitoneal and omental tissue. The graft and aneurysm sac were irrigated with solutions of ciprofloxacin and metronidazole through the drainage tube twice a day and the patient received concomitant intravenous antibiotic therapy of ceftazidime (2 g, twice daily), and vancomycin hydrochloride (400 mg, four times daily) for six weeks. Cultures of purulent material and samples of the aneurysm sac isolated Salmonella choleraesuis, which confirmed the aortic stent-graft infection, and antibiotic susceptibilities test supported the antibiotics above. The patient recovered slowly with temperature decreasing to normal and ESR, C-reactive protein (CRP) lowered markedly for the first time. After a 4-week course of drainage and irrigation, three consecutive cultures yielded negative results, and the drainage tubes were removed. Discharged from the hospital the patient was given a regimen of oral antibiotics (cefaclor, 250 mg three times daily and levofloxacin, 100 mg twice daily) according to the susceptibilities of antibiotics, which continued till now. At discharge, the CRP level was 10 mg/L. At 6- and 12 months of follow-up, CT scan showed no evidence of recurrent infection. The patient remained to be asymptomatic, with normal leucocyte count and CRP level less than 6 mg/L.Fig. 3. A:: While opening the aneurysm sac, abundant reddish-brown fluid and an aortic thrombus, which was gelatinous and semiliquefied (arrow) were seen; B: No fresh blood in the aneurysmal sac and no endoleak was seen, and the stent-graft struts were intact without damage to the fabric (arrow).DISCUSSION Marfan syndrome is an inherited autosomal-dominant disorder of the connective tissues involving the cardiovascular system. The genetic defect leads to synthesis of mutant fibrillin with inability to bind calcium and eventually to weakening of the elastic layers of the connective tissue.3,4 Ince et al5 suggested that stent-grafting should be considered a bridge rather than a curative procedure for aortic disease with Marfan syndrome because of the high possibility of conversion to surgery in their series. From our case, we learned that not only prudent implantation of stent-graft but also close surveillance are mandatory during the endovascular repair for aortic disease in patients with Marfan syndrome so as to avoid or detect silent complications eventually requiring open surgery. Aortic stent-graft infection is an uncommon but life-threatening complication in aortic endovascular therapy. Ducasse et al6 reported that the incidence of aorto-iliac stent-graft infection averaged 0.43% in their retrospective study. Several factors could influence the risk of stent graft infection, one of which was the procedure setting. About 62.5% of stent graft infections occurred after the endovascular procedure was carried out in the catheterization laboratory, in contrast to 37.5% after those in conventional operating theaters.6 The endovascular technique always leaves the aneurismal content intact compared with open surgery. The reported rates of bacterial colonization of aortic thrombus range from 8% to 37%.7 In certain cases, the bacteria may truly involve in the pathogenic process and inflammatory reaction. Furthermore, the aortic endothelium may be damaged by the procedure-related instrumentation, which makes contamination possible in cases of infected aortic aneurysm. The most common organism related to the aortic stent-graft infection is Staphylococcus aureus,6 or Salmonella.8 In our case, cultures of blood and aneurismal contents both identified the growth of Salmonella choleraesuis. Other possible causes for the infection of our case included the long duration of the second procedure in the catheterization laboratory, the poor condition of the patient who unfortunately suffered from rupture of AAA twice as well as damage to the mucosal barrier of the intestine. Sometimes, huge retroperitoneal hematoma may disturb the perfusion of the bowel, and the bacteria inhabiting in the intestine could pass across the mucosal barrier into the bloodstream and disseminate hematogenously. Classic management of aortic graft infection involves complete excision of the infected graft with reconstruction of the distal circulation via clean and unaffected tissues.9 In practice, application of this classic approach is poorly tolerated by many patients who usually have significant medical comorbidities, resulting in a high mortality. In selected patients, in situ replacement of synthetic graft has been advocated, but its main drawback lies in the risk of severe complication owing to recurrent infection.10,11 Each of these approaches has merit and utility when applied in the appropriate patient. There remains, however, a distinct group of patients who are very debilitated, clinically unstable or have markedly hostile abdomens so that these approaches would be precluded. Our patient had undergone two procedures for ruptured AAA, with poor condition and high surgical risk precluding the procedure of total excision of the infected aortic stent-graft. Ultimately he received conservative treatment consisting of surgical debridement, drainage, systematic antimicrobial therapy, local antibiotic irrigation, with complete preservation of stent graft. Up till now, the conservative treatment has been very successful in this patient, but it awaits further elucidation as the initial approach for the management of infection of aortic stent graft.