A 75 year old independent Caucasian male presented to hospital with three weeks of pyrexia and lethargy, alongside progressive bilateral tender palpable masses to his lower limbs. He was normally well and was physically active prior to this illness. His past medical history was significant for an elective AAA repair 10 months previously, which has been complicated by an upper gastrointestinal bleed post-operatively. He was investigated for this with upper and lower endoscopy which showed gastritis and haemorrhoids only. His regular medication included a statin, clopidogrel and omeprazole.
Upon review, he was pyrexial and tachycardic - HR 124, temperature 38.1oC, BP 138/80, RR 18, SpO2 99% OA. Examination revealed slow gait due to leg pain and hot tender lumps in the lower limbs. There were no peripheral stigmata of infective endocarditis. Chest and abdominal examination was unremarkable. GCS 15/15. Clerking bloods showed high inflammatory markers and three sets of blood cultures were taken prior to antimicrobial therapy. Empiric antimicrobial treatment was intravenous Piperacillin/Tazobactam and intravenous vancomycin.
Table 1. The MAGIC criteria 2020 suggest vascular graft infection diagnosis with >/= 1 major criterion and any other criterion from another category [16].
|
MAJOR
|
MINOR
|
Clinical/surgical
|
• Pus (on microscopy) around graft or in aneurysm sac at surgery
• Open wound with exposed graft of communicating sinus
• Fistula development e.g. aorto-enteric
|
• Localised clinical features of graft infection e.g. erythema, swelling, discharge, pain
• Fever > 38oC with graft infection as most likely cause
|
Radiological
|
• Perigraft fluid on CT scan > 3 months after insertion
• Perigraft gas on CT scan > 7 weeks after insertion
• Increase in perigraft gas volume demonstrated on serial imaging
|
• Other findings e.g. suspicious perigraft gas/fuid; soft tissue inflammation; aneurysm expansion; pseudoaneurysm formation; focal bowel wall thickening; discitis/osteomyelitis; suspicious metabolic activity on PET CT or radiolabelled leukocyte scan
|
Microbiological
|
• Organisms recovered from explanted graft
• Organisms recovered from an intraoperative specimen
• Organisms recovered from a percutaneous, radiologically guided aspirate of perigraft fluid
|
• Blood culture positive, no apparent source except graft infection
|
CT angiogram of the aorta confirmed aortic graft infection with distal emboli: locules of gas and multiple small collections surrounding the aortic graft were noted, as well as soft tissue collections in left gluteus minimus, right inferior gluteus maximus, right vastus lateralis, right tibialis anterior, right posterior calf, and left soleus (Fig. 1).
Of the three sets of blood cultures drawn on admission, the anaerobic bottle of one set flagged positive at 37 hours. Gram stain revealed a combination of Gram positive cocci in chains and small gram negative cocci.
Subsequent MALDI-TOF identification revealed D. pneumosintes (score 1.97). Fluid aspirated from leg collections isolated oral flora - Streptococcus anginosus, Fusobacterium nucleatum, and Parvimonas micra. Given these findings, a maxillofacial review was requested along with orthopantomography. Opinion was of poor dentition but not acute dental infection requiring intervention. Once antibiotic sensitivities were confirmed on all aspirates, antimicrobial spectrum was narrowed with transition from Piperacillin/Tazobactam and Vancomycin to 2.4g IV benzylpenicillin every four hours and 400 mg IV metronidazole every eight hours. Endocarditis dosing regimen of gentamicin (1mg/kg twice daily) was added to cover for possible infective endocarditis despite repeatedly negative transthoracic echocardiograms.
Definitive aortic graft explantation was undertaken three weeks after admission. The planned operation involved graft explantation and replacement with biological graft. However, surgery was complicated by intraoperative findings of a small aortoenteric fistula between the redundant aneurysm sac and the duodenum. Therefore the duodenum was repaired with an omental patch and a new axillo-bifemoral graft away from any local sources of infections. The fistula also prompted the addition of empiric fluconazole (800 mg loading dose, subsequently 400 mg IV once daily). The intraoperative aortic graft and sac samples isolated Citrobacter koseri and Candida albicans (MIC 1.5 fluconazole), in keeping with aorto-enteric fistula. D. penumosintes, S. anginosus, F. nucleatum or P. micra were not isolated from theatre samples, but this is not surprising as our patient received three weeks of effective anti-anaerobic antimicrobial therapy prior to culture.
Antimicrobial therapy was rationalised post operatively to IV co-amoxiclav 1.2g TDS and caspofungin 70 mg loading dose and then 50 mg thereafter OD. Transition to oral co-amoxiclav 625 mg TDS and 600 mg fluconazole OD was recommended when clinically appropriate. A total of six weeks of co-amoxiclav was suggested post explantation, with six months of fluconazole on discharge.