Infected aneurysm of the thoracic aorta

Setting 

National Taiwan University Hospital (Taipei, Taiwan) is a 2200-bed tertiary care hospital. It serves an urban population of 2 million persons as both first-line and tertiary facilities. It serves also as a referral center for other hospitals in the country with a population of 22 million persons.

Patients 

The study patient cohort was obtained from a retrospective analysis of the clinical data for patients with infected aneurysms of the thoracic aorta from September 1995 to June 2007. Patients with infection of ascending aorta, thoracoabdominal aorta, and abdominal aorta were excluded.

Diagnosis 

The diagnosis of infected aneurysm was made by a combination of clinical evidence of infection, imaging evidence of infected aorta, and pathologic evidence of inflammation. We excluded patients with thoracoabdominal, iatrogenic, or traumatic aneurysms. Clinically, infected aortic aneurysm was usually preceded by an infected aorta or aortitis. An infected aorta was diagnosed with clinical evidence of infection (fever and leukocytosis) and periaortic soft tissue infiltration demonstrated by a computed tomography or magnetic resonance imaging study.20 Blood and tissue culture was repeated to find the responsible microorganism. The diagnosis of infected aneurysm was further confirmed by the presence of frank pus during the operation and the presence of acute suppurative inflammation or bacterial clumps, or both, on pathologic examination.21

Medical treatment 

As described previously,16, 17 an intravenous antibiotic was given once the diagnosed was confirmed. Patients with Salmonella spp infections received intravenous ceftriaxone (1000-2000 g every 12 hours). For patients with non-Salmonella infections, the antibiotic use was determined by culture result and sensitivity test. In patients with good response to antibiotic treatment (no fever, no localized pain and declining white cell count), surgical intervention was considered after the infection was controlled. The imaging study was repeated if patients had a new symptom or sign suggesting aortic pseudoaneurysm formation (shock, a new localized pain, or a new palpable mass) or after complete antibiotic treatment.

Early surgical intervention, which was defined as an operation before the infection was controlled, was performed only in the situations of uncontrolled infection (persistent fever or septic shock) or if there was evidence of impending aortic rupture demonstrated by persistent pain, shock, or enlarging pseudoaneurysm formation on the repeated imaging study.

Surgical treatment 

Surgical management consisted of wide débridement of infected tissue, copious saline irrigation, and in situ repair with a Dacron graft (DuPont, Wilmington, Del). Neither homografts nor antibiotic bonded grafts were available.

The selected operative approach depended on the surgeon’s preference. Median sternotomy was selected in patients with proximal aortic arch or distal aortic arch aneurysms. Cardiopulmonary bypass was instituted through arterial cannulation of ascending aorta or right subclavian artery and venous cannulation of right atrium.

The reconstruction of the aortic arch and arch vessels was performed under deep hypothermic circulatory arrest. Methods of brain protection during circulatory arrest, including deep hypothermia alone, retrograde cerebral perfusion, and selective antegrade cerebral perfusion, also depended on the surgeon’s preference.

Left posterolateral thoracotomy was selected for patients with distal arch or descending thoracic aneurysms. Cardiopulmonary bypass was instituted through arterial cannulation of femoral artery and venous cannulation of femoral vein.

For those patients with distal arch aneurysms, a proximal aortic anastomosis was performed under deep hypothermic circulatory arrest and no adjunctive cerebral perfusion. For those patients with descending thoracic aneurysms, a proximal and distal aortic anastomosis was performed under aortic clamping.

Antibiotics were administered intravenously in the hospital for at least 6 to 8 weeks and until the clinical and laboratory variables of fever, white cell count, and C-reactive protein were normalized. The antibiotic was continued orally after discharge, with the duration of at least 6 months or according to the surgeon’s preference.

Data collection 

Data on age, sex, medical comorbidities, operation status, location of infected aneurysms, surgical procedure, and clinical outcome were collected retrospectively from medical records. For those patients not followed up at our outpatient clinic, follow-up information and survival data were obtained by telephone contact or from the National Death Index. Aneurysm-related mortality was defined as a death caused by persistent infection or postoperative complications directly related to the operation. A death that occurred ≤1year after operation and was caused by a diagnosis not related to the infected aneurysm was considered not aneurysm-related. Patients who survived >1 year were considered disease-free.

Patient characteristics 

From 1995 to 2007, 32 patients (24 men and 8 women) with infected aneurysms of the thoracic aorta were treated in our hospital. The patient demographics are summarized in Table I, Table II, Table III. Their median age was 74 years (range, 50-88 years). The medical comorbidities included hypertension in 9 patients, diabetes mellitus in 11, chronic obstructive pulmonary disease in 3, coronary artery disease in 7, pulmonary tuberculosis in 5, end-stage renal disease with regular hemodialysis in 3, and liver cirrhosis in 2.

Table I. Data of 13 patients with infected aneurysms of the aortic arch

	No.	Sex	Age, y	Pathogen	Clinical presentation	Medical diseases	Imaging finding	Operation, extent of arch replacement	Hospital outcome	Post-op complication	Late outcome
	1	M	67	S typhimurium	Fever	None	PSA 6 cm	In situ grafting; partial distal arch	Alive	Respiratory failure, hypoxic encephalopathy	Died 14 mon; heart disease
	2	M	57	S choleraesuis	Fever, chest pain	HT	PSA 4.8 cm	In situ grafting; partial distal arch	Alive	Wound infection	Died 4 mon; late prosthetic infection
	3	M	58	K pneumoniae	Fever, LoC	CAD, bacterial meningitis	PSA 5.5 cm	In situ grafting; partial distal arch	Alive	None	Alive 47 mon
	4	M	79	S enteritidis	Fever	Pulmonary TB, DM, HT	PSA 5 cm	In situ grafting; total arch	Alive	Hypoxic encephalopathy	Died 3 mon; multiple organ failure
	5	F	66	S choleraesuis	Fever	DM, HT, CAD	PSA 3 cm	In situ grafting; partial distal arch	Alive	Respiratory failure, nosocomial bacteremia	Alive 44 mon
	6	M	79	E coli	Fever, LoC, chest pain	COPD, heart failure	PSA 6 cm	In situ grafting; partial distal arch	Alive	Hypoxic encephalopathy	Died 14 mon; multiple organ failure
	7	M	78	Ss aureus	Fever, ischemic stroke	Toxicoderma	PSA 5 cm	In situ grafting; total arch	Died early prosthetic infection	Early prosthetic infection
	8	M	68	S aureus	Fever	DM, HT, CAD, uremia	PSA 5.5 cm	In situ grafting; total arch	Alive	Hypoxic encephalopathy, reopen for bleeding, sternal wound infection	Died 9 mon; multiple organ failure
	9	M	50	S enteritidis	Fever, hoarseness	None	PSA 4 cm	In situ grafting; partial distal arch	Alive	None	Alive 18 mon
	10	F	70	S pneumoniae	Hemoptysis, Drowsiness	DM, central pontine myelinosis	PSA 6.5 cm	In situ grafting; partial distal arch	Alive	None	Alive 12 mon
	11	M	70	S choleraesuis	Fever	Uremia, bladder carcinoma, CAD	Aortitis	None	Alive		Died 22 mon; bladder carcinoma
	12	M	49	S choleraesuis	Fever	DM	PSA 5 cm	None	Alive		Alive 78 mon
	13	M	82	Unknown	Fever	DM, HT, CAD	PSA 2.3 cm	None	Died

PSA, Pseudoaneurysm; DM, diabetes mellitus; CAD, coronary artery disease; TB, tuberculosis; LoC, loss of consciousness; HT, hypertension; COPD, chronic obstructive pulmonary disease.

Table II. Data of 10 patients with infected aneurysms of the proximal descending thoracic aorta

	No.	Sex	Age, y	Pathogen	Clinical presentation	Medical diseases	Imaging finding	Operation	Hospital outcome	Post-op complication	Late outcome
	1	M	82	M tuberculosis	Fever, chest pain hemoptysis	Pulmonary TB	PSA, 3 cm	In situ grafting	Alive	None	Died 101 mon, unknown cause
	2	F	69	S typhimurium	Fever, urinary tract infection	DM, HT	PSA, 4 cm	In situ grafting	Alive	None	Alive 91 mon
	3	F	61	Unknown	Respiratory failure	COPD	PSA, 9 cm	In situ grafting	Alive	Respiratory failure	Died 3 mon, respiratory failure
	4	F	84	M tuberculosis	Back pain, hemoptysis	Pulmonary TB	PSA, 6 cm	In situ grafting	Alive	MRSA bacteremia	Died 41 mon, unknown cause
	5	F	77	S enteritidis	Fever, chest pain	DM, HT, lung carcinoma	PSA, 5 cm	In situ grafting	Alive	None	Alive 27 mon
	6	M	68	S aureus	Chest pain, hemoptysis,	Uremia, DM, HT, old stroke	PSA, 5.1 cm	In situ grafting	Alive	None	Alive 2 mon
	7	M	88	S typhimurium	Fever, chest pain	DM	PSA, 4.5 cm	None	Died aneurysm rupture
	8	M	80	M tuberculosis	Hemoptysis	Pulmonary TB	PSA, 6 cm	None	Alive		Died 3 mon, sudden death
	9	M	83	S aureus	Fever	HT, COPD, liver cirrhosis	PSA, 5 cm	None	Died aneurysm rupture
	10	M	75	Unknown	Chest pain	Liver cirrhosis, hepatocellular carcinoma	PSA, 4.4 cm	None	Died aneurysm rupture

TB, Tuberculosis; PSA, pseudoaneurysm; BDM, diabetes mellitus; HT, hypertension; COPD, chronic obstructive pulmonary disease; CAD, coronary artery disease; MRSA, methicillin-resistant Staphylococcus aureus.

Table III. Data of nine patients with infected aneurysms of the distal descending thoracic aorta

	No.	Sex	Age, y	Pathogen	Clinical presentation	Medical diseases	Imaging finding	Operation	Hospital outcome	Postoperative complication	Late outcome
	1	M	74	S typhimurium	Fever	Steroid use	PSA, 6 cm	In situ grafting	Alive	None	Died 133 mon, prostate cancer
	2	F	74	E coli	Fever, chest pain	CAD	PSA, 10 cm	In situ grafting	Alive	Seizure	Alive 134 mon
	3	M	51	S enteritidis	Fever, chest pain	HT, DM, old stroke	PSA, 7 cm	In situ grafting	Alive	None	Alive 112 mon
	4	M	78	Unknown	Fever, dyspnea	Anemia	PSA, 6.5 cm	In situ grafting	Died sepsis	Re-op for bleeding, early graft infection
	5	F	81	S enteritidis	Fever, back pain	DM	PSA, 5 cm	In situ grafting	Alive	Respiratory failure, wound infection, Veillonella bacteremia	Died 11 mon, renal failure
	6	M	84	S typhimurium	Shock,	CAD, pulmonary TB, old stroke	PSA, 4 cm	In situ grafting	Died sepsis	Esophageal perforation
	7	M	74	S enteritidis	Fever, back pain	None	PSA, 5 cm	In situ grafting	Alive	Early graft infection	Died 24 mon, sepsis
	8	M	56	S choleraesuis	Chest pain	None	PSA, 4.5 cm	In situ grafting	Alive	None	Alive 8 mon
	9	M	56	S pneumoniae	Fever, back pain	None	PSA, 7 cm	In situ grafting	Alive	None	Alive 35 mon

TB, Tuberculosis; PSA, pseudoaneurysm; DM, diabetes mellitus; HT, hypertension; CAD, coronary artery disease.

At presentation, 24 patients were febrile and 16 patients had localized chest or back pain. One patient was in shock status at the time of presentation. Four patients presented with consciousness disturbance, five with hemoptysis, and two with hemothorax. The four patients presenting with disturbance of consciousness had an infected aortic arch aneurysm, and one patient had acute left middle cerebral artery ischemic stroke because of external compression of left common carotid artery by an infected distal arch pseudoaneurysm (Table I).

The initial imaging findings were aortitis in one patient and aortic pseudoaneurysm or saccular aneurysm in 31 patients. The pseudoaneurysms were 2.3 to 10 cm in maximal diameter. The site of infection was the aortic arch in 13 patients, proximal descending thoracic aorta in 10, and the distal descending thoracic aorta in nine.

Microbiology 

All patients except four had a blood or tissue culture result positive for a microorganism. Of the 28 isolates, the most common responsible microorganism was nontyphoid Salmonella in 16 patients (57%), followed by Staphylococcus aureus in 4 (14%), Mycobacterium tuberculosis in 3 (11%), Escherichia coli in 2 (7%), Streptococcus pneumoniae in 2 (7%), and Klebsiella pneumoniae in 1. The serotypes of isolated nontyphoid Salmonella spp were S choleraesuis in 5, S enteritidis in 6, and S typhimurium in 5.

Treatment 

Operation was refused in seven patients, because of old age in four, severe medical comorbidities in two, and one patient was at high surgical risk. In situ graft replacement was done in 25 patients, none of which underwent extra-anatomic bypass. The median duration of preoperative antibiotic use was 8 days (range, 0-45 days). The aneurysm location, treatment, and outcome are summarized in the Fig.

View Large Image Download to PowerPoint

Fig. Aneurysm location, treatment, and outcome in patients with infected aneurysms of the thoracic aorta.

Aortic arch 

Three of the 13 patients with infected aortic arch aneurysm received medical treatment only, and 10 patients underwent in situ graft replacement (Fig). One of the three patients with medical treatment alone died of aneurysm rupture 1 day after admission.

The operation was performed through median sternotomy in four patients and thoracotomy in six. One of the 10 patients undergoing in situ graft replacement died, for a hospital mortality rate of 10%. The only hospital death occurred in a 78-year-old patient with a distal aortic arch infected aneurysm caused by methicillin-resistant S aureus. He underwent urgent operation because of progressive left cerebral hemispheric ischemic stroke caused by localized rupture of infected aneurysm and compression of left common carotid artery. He died of early prosthetic graft infection and overwhelming sepsis 4 weeks after the operation.

Seven of 10 patients (70%) had major postoperative complications (Table I). Postoperative hypoxic encephalopathy with persistent neurologic deficits occurred in four patients, respiratory failure with prolonged ventilator support occurred in two, and reoperation for bleeding, sternal wound infection, nosocomial bacteremia, surgical wound infection, and early prosthetic graft infection each occurred in one patient. Three of the four patients with preoperative consciousness disturbance recovered, and one patient had postoperative hypoxic encephalopathy. Five late deaths occurred because of multiple organ failure in three patients, late prosthetic graft infection in one patient, and heart disease in one patient (Table I). In summary, the aneurysm-related mortality rate of in-situ grafting in patients with infected aortic arch aneurysm was 40%, and the disease-free survival at 1 year was 50% (Fig).

Proximal descending thoracic aorta 

Four of the 10 patients with an infected proximal descending thoracic aortic aneurysm received medical treatment alone, and six underwent in situ graft replacement through posterolateral thoracotomy (Fig). Of the four patients with medical treatment alone, three died of aneurysm rupture at 5, 11, and 16 days after admission, and one patient died suddenly at home about 3 months after treatment. There was no hospital mortality in 6 patients undergoing in situ graft replacement. Major postoperative complications occurred in two of six patients (33%) and included respiratory failure with prolonged ventilator support and bacteremia caused by methicillin-resistant S aureus. In summary, the aneurysm-related mortality rate of in situ grafting in patients with proximal descending thoracic aortic aneurysm was 0% and the disease-free survival at 1 year was 83% (Fig).

Distal descending thoracic aorta 

None of the nine patients with infected distal descending thoracic aortic aneurysm received medical treatment alone, and nine patients underwent in situ graft replacement through thoracotomy (Figure). The hospital mortality rate in the nine patients who had in situ graft replacement was 22% (2 of 9). The causes of hospital death were early prosthetic graft infection and esophageal perforation (Table III). Major postoperative complications occurred in five of nine (56%) patients. Early prosthetic graft infection occurred in two patients, and respiratory failure with prolonged ventilator support, reoperation for bleeding, wound infection, nosocomial bacteremia, and esophageal perforation each occurred in one patient. Of the two patients with early graft infection, one patient received conservative treatment but died, and one recovered after flap reconstruction (Table III). One patient died 11 months after operation because of renal failure. In summary, the aneurysm-related mortality rate of in situ grafting in patients with distal descending thoracic aortic aneurysm was 33% and the disease-free survival at 1 year was 67% (Fig).

Overall, the hospital mortality rate of medical treatment alone was 57% and the hospital mortality rate of in-situ graft replacement was 12% (Fig).

Late outcome 

The median follow-up duration for the 22 operated-on survivors was 16 months (range, 2-134 months). Follow-up information was complete in all patients. There were 11 late deaths, of which four were aneurysm-related, and seven were not.

One of the aneurysm-related mortalities occurred in 57-year-old man with late prosthetic graft infection in an aortic arch aneurysm caused by S choleraesuis. He underwent a successful operation after the preoperative use of ceftriaxone for 8 days. However, he died of aortoenteric fistula and massive bleeding 4 months after the operation (Table I). One 79-year-old man with an infected aortic arch aneurysm caused by S enteritidis underwent an emergency operation that was complicated with hypoxic encephalopathy. He was discharged with long-term tracheostomy care but died of multiple organ failure 3 months after the operation (Table I). A 68-year-old man with an infected aortic arch aneurysm caused by S aureus underwent an emergency operation that was complicated with hypoxic encephalopathy. He was discharged with long-term tracheostomy care but died of multiple organ failure 9 months after operation (Table I). One 81-year-old woman with an infected distal thoracic aortic aneurysm caused by S enteritidis underwent an emergency operation that was complicated with respiratory failure. She was discharged with long-term tracheostomy care but died of renal failure 11 months after operation (Table III).

Of the seven deaths that were not aneurysm-related, acute respiratory failure occurred in a 61-year-old woman with an infected proximal descending thoracic aortic aneurysm who had a medical history of chronic obstructive pulmonary disease. She underwent a successful operation, but died of acute respiratory failure 3 months after the operation. The repeat imaging study showed no evidence of recurrent graft infection (Table II). Six patients died >1 year after operation because of heart disease in 1 (Table I), multiple organ failure in 1 (Table I), prostate carcinoma in 1 (Table III), sepsis in 1 (Table III), and an unknown cause in 2 (Table II).

In summary, the aneurysm-related mortality rate of in situ graft replacement in patients with infected thoracic aortic aneurysm was 28% (7 of 25; Fig). Of the 23 patients undergoing in situ graft replacement before July 2006, 15 patients (65%) were alive and infection-free at 1 year after operation.

Infected thoracic aortic aneurysm 

Infected aneurysms of the thoracic aorta are extremely rare.11, 12, 13 S aureus and Salmonella spp are the predominant pathogens. Presenting symptoms are often nonspecific. Because of insidious onset, the clinical presentation is often due to the complication of aneurysm rupture.13

Surgery remains the definitive treatment, but in the early years, the hospital mortality rate and the aneurysm-related mortality rate was as high as 38% and 40% in patients with infected suprarenal or thoracic aortic aneurysm.1, 8 However, the reported clinical experience is limited to small series and case reports.1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12

The largest single-center experience of 11 patients with infected thoracic aortic aneurysm was reported in 2003. There were four ascending aortic aneurysms, four aortic arch aneurysms, and three descending thoracic aortic aneurysms. Nine patients underwent surgery, and one died, for a hospital mortality rate of 11%. None of the eight operative survivors had evidence of recurrent infection with a follow-up duration of 19 months.13

In this current series, 32 patients had an infected thoracic aortic aneurysm. None of our patients had ascending aortic infection, 13 had an aortic arch aneurysm, and 19 had a descending thoracic aortic aneurysm. The initial imaging study showed that >96% of our patients had aneurysm rupture or pseudoaneurysm formation.

Of the 25 operated on cases, the hospital mortality rate of 12% and aneurysm-related mortality rate of 28% were much lower than the rates in previous reports.1, 8, 11, 12, 13 The clinical results of current treatment method using in situ graft replacement are improving. The low mortality rate was probably attributed to clinical alertness and the extensive surgical experience of treating infected aortic aneurysms in Taiwan.16, 17 Nevertheless, major postoperative complications, including prosthetic graft infection and hypoxic encephalopathy in infected aortic arch aneurysms, were still problems. In this current series, the incidence of prosthetic graft infection was 16% (3 early and 1 late).

The possible ways to improve the outcomes include aggressive surgical débridement, possibly placement of pedicled vascular tissue such as omentum or intercostals muscle to separate the graft from an infected bed of tissue, and the use of antibiotic-bonded grafts, autogenous grafts, or cryopreserved allografts. Previous studies have shown that the use of pedicled flap, antibiotic-bonded graft, autogenous grafts, and cryopreserved allograft is an effective strategy for treating aortic infection.22, 23, 24, 25 However, the clinical experience is limited in only small case series. Certain antibiotic-bonded grafts, such as rifampin-soaked gelatin-sealed polyester grafts, are not effective against infection caused by Salmonella spp or methicillin-resistant S aureus.26 Further studies are needed to prove these points.

In addition, although preoperative consciousness disturbance was present in four of 10 patients with an infected aortic arch aneurysm, postoperative hypoxic encephalopathy occurred in four of nine operated-on survivors (Table I). Proper use of adjunctive cerebral perfusion during deep hypothermic circulatory arrest probably will improve the outcomes.

Open surgical repair for infected thoracic aortic aneurysms carries significant mortality and morbidity. Endovascular stent grafts combined with antibiotic therapy may be an alternative to conventional thoracotomy in managing infected aneurysms of the thoracic aorta. The preliminary results are encouraging27, 28 However, the use of endovascular stent grafts in the treatment of infected thoracic aneurysm is contentious because the lack of surgical débridement confers the risk of continued infection and the long-term durability of stent grafts is still unclear.29

Salmonella infection 

In Western countries, the most common responsible organism of infected aortic aneurysm was S aureus and Streptococcus spp.1, 2, 3, 4, 5, 6, 7, 8, 9, 10 In Taiwan, nontyphoid Salmonella was the most common pathogen.14, 15, 16, 17, 18, 19 Salmonella-infected aneurysms of the thoracic aorta are rare and the prognosis is poor.30, 31 In a collected series of 14 previously reported patients with Salmonella-infected aneurysms of the thoracic aorta, the overall outcome was abysmal, with 10 patients undergoing operation and six patients (60%) dying ≤1 month after the diagnosis.11 In this current series, however, 16 patients had infected thoracic aortic aneurysms caused by nontyphoid Salmonella (Table I, Table II, Table III). Three patients had medical treatment alone, and one patient died of aneurysm rupture (33%). Thirteen patients underwent in situ graft replacement, with a hospital mortality rate of 8% and an aneurysm-related mortality rate of 31%. The prognosis of infected thoracic aortic aneurysm caused by nontyphoid Salmonella was not dismal.

Study limitation 

The major limitations of this study are that it was retrospective and lacked long-term follow-up information. Because it was retrospective, there was no consistent approach to these patients and the duration of preoperative antibiotic use was also variable. However, to our knowledge, this study is the largest case series ever reported in the literature and provides useful information of treating infected aneurysms of the thoracic aorta.