[Aortoiliac graft infection as a diagnostic and treatment problem].

UNLABELLED

Aortoiliac graft infection occurs in 2-6% of patients with such prosthesis. This condition is seldom properly diagnosed by conventional radiographic methods, leading to high morbidity and mortality. Clinically, the diagnosis of aortic graft infection is difficult because patients may have a variety of nondescript clinical complaints. The diagnosis of graft infection when associated with minimal or absent clinical signs of low-grade infection is uncertain, but is critically important to avoid frequently catastrophic complications such as sepsis, gastrointestinal hemorrhage, and suture line disruption.

AIM OF THE STUDY

identification of bacterial flora present in aortoiliac graft infection; the presentation of my own experience in the detection of aortoiliac graft infection with special description of isotopic study with WBC labeled 99mTc HM-PAO and estimation of the usefulness of this test in comparison with computed tomography, ultrasonography, fistulography and angiography; evaluation of the usefulness of various treatment methods; establishing principles (algorithm) of diagnostic and therapeutic procedures in the case of the suspicion of aortoiliac graft infection. As many as 1190 patients with implanted aortoiliac graft in 1986-1996 at the General and Vascular Surgery Clinic in Szczecin were studied (Tab. 2). Thirty-one patients in the study had deep aortoiliac graft infection (Tab. 3), while 9 patients had, in addition, prosthetic-enteric fistulae and 1 had arterio-enteric secondary fistulae. The group of 31 patients with deep graft infection, that is 2.6% of all patients (1190), had aortoiliac graft implanted at the mentioned time period (Tab. 4, 5). Test results for detection of graft infection have been analysed (Tab. 17). The results of isotopic investigation (Tab. 16), computed tomography (Tab. 11), ultrasonography (Tab. 13), fistulography (Tab. 14) and angiography (Tab. 15) were compared with intraoperative state or in a case of exclusion of infection, with results of follow up. Results of various paths of treatment were estimated (Tab. 18). Based on performed cultures most common bacterial flora from infected grafts, was identified (Tab. 9, 10). The sensitivity of the isotopic study with labeled white blood cells in detection of graft infection was 88%, specificity was 97%, accuracy 93%, positive predictive value 96%. Other useful diagnostic procedures in detection of aortic graft infection are: computed tomography with an accuracy of 75%, endoscopic investigation useful in detection of arterio-enteric fistulae with an accuracy of 50% and ultrasonography with an accuracy of 35.5% (Tab. 17). The choice of the best treatment is still controversial. In my material total excision of infected graft and extraanatomic revascularization were burdened with 50% mortality rate. Among patients treated less radically the mortality rate was considerably lower. In a group of patients with the excision of the infected graft only, the mortality rate was 9% but the amputation rate was 36.4% and in a group of patients with excision of infected graft and reconstruction in situ, the mortality rate was 25% (Tab. 18). Taking into consideration our results, less aggressive methods of aortic graft infection treatment such as the excision of the infected part of the prosthesis with or without in situ revascularization if only possible should be recommended. Most common in bacterial cultures from infected aortoiliac grafts with prosthetic-enteric fistulae were Escherichia coli found (Tab. 10). In infections without fistula various types of Staphylococcus aureus were identified (Tab. 9).

CONCLUSIONS

1. In cases of aortoiliac graft infection the most common cultured bacteria are found to be Staphylococcus aureus. If there is additional prosthetic-enteric fistula Escherichia coli is the most common cultured bacteria. 2. In a case there was suspicion of aortoiliac graft infection, proper diagnostic procedures are most important for effective ma