Search

Background A number of studies have looked at the outcome of vascular graft infections, but these often include small numbers of patients collected over a prolonged period of changing practice. Many of these studies do not contain cases of methicillin-resistant Staphylococcus aureus (MRSA). Methods Between February 1998 and January 1999, a prospective, multicentre audit was performed in order to examine the outcome of complex wound and graft infections in vascular surgery, with particular reference to MRSA. Results Some 66 grade II or III Szilagyi wound infections occurred. Overall, 23 per cent of type II infections progressed to graft infection. Of those with type III infections, 83 per cent had an adverse outcome (death, major amputation or ongoing infection). No vein graft became infected, but 12 of 38 prosthetic grafts were affected. MRSA accounted for 42 per cent of infections and was present in 67 per cent of patients with an adverse outcome. Adverse events occurred in 32 per cent of MRSA cases and in 10 per cent of non-MRSA cases (P = 0·04). Wound infection proceeded to graft infection in 58 per cent of MRSA cases and in 32 per cent of non-MRSA cases (P = 0·08). Seven of eight patients with MRSA before operation developed postoperative MRSA infection. Median length of stay was 31 days for MRSA versus 18 days for non-MRSA infections (P = 0·001). Forty-three graft infections occurred; 12 were associated with a complex wound infection. Median time to presentation was 39 (range 2–2450) days. There were 16 deaths, nine major limb amputations and three ongoing infections among patients with a graft infection. The median length of stay was 36 days. MRSA was present in 36 per cent of all graft infections. Only one of these infections involved a vein graft (P = 0·001). MRSA infection occurred more frequently where the primary procedure had been urgent rather than elective (P = 0·03). MRSA graft infection accounted for seven of the nine major limb amputations. The median length of stay in patients with MRSA graft infection was 25 days longer (P = 0·001). Conclusion Complex wound and graft infections continue to be associated with postoperative deaths and amputations. MRSA infection is now the most common organism involved in complex infections following vascular procedures. It is associated with an increase in the number of adverse outcomes and a significant increase in length of stay.

Background Serious infection complicates a small proportion of prosthetic vascular grafts and usually follows implantation of bacteria at the time of graft insertion. The aim of this study was to determine whether the use of a rifampicin-bonded, gelatin-coated Dacron graft (Gelsoft/Gelseal; Sulzer Vascutek) could reduce the rate of infective complications after extra-anatomic bypass. Methods In a multicentre study involving 14 vascular units, 257 patients underwent extra-anatomic bypass. There were 178 men and 79 women, of median age 69 (range 43–92) years. Some 132 (51 per cent) had no risk factors for wound or graft sepsis, 82 (32 per cent) had one and 43 (17 per cent) had two or more risk factors. Patients were randomized to rifampicin bonding (rifampicin 1 mg ml−1 soak for 15 min before insertion) (n = 123) or control (n = 134). The authors have previously reported small, but not statistically significant, reductions in total perioperative wound complications, antibiotic requirements and delayed hospital discharge in favour of patients with bonded grafts. Results Of 257 patients originally randomized, follow-up to 2 years or death is available for 197 patients; 60 have died. A total of ten patients have had serious wound or graft infections (Szilagyi grade III), four with unbonded and six with bonded grafts (P not significant). All occurred within 6 months of operation. The graft fabric was involved in eight patients. Treatment and outcome are listed in the Table below. Conclusion It has not been possible to show any advantage for the routine use of antibiotic bonding for extra-anatomic reconstruction.

Twelve patients with a clinical diagnosis of possible vascular graft infection have been studied over the last 3 years. All patients had their leucocytes labelled with indium-111 and γ-camera imaging after 24 and 48h. Subsequent management was according to established surgical techniques. Eight patients proved to have vascular graft infection and indium uptake was seen along the length of the graft in six. Two patients with open wound infections and synthetic grafts had localized uptake only on leucocyte scanning. Four patients, all of whom had negative scans were not thought to have infected grafts after further investigation using digital subtraction angiography and computed tomography scanning. These patients have been followed up for a median period of 19 months (range 5–25 months) and have remained symptom free. Initial experience with 111In-labelled leucocyte scanning has been encouraging, both in diagnosis and in planning the management of patients with graft infections.