Risk factors and outcomes for nosocomial infection after prosthetic vascular grafts.

OBJECTIVE

The objective of this study was to determine risk factors for nosocomial infections (NIs) and predictors of mortality in patients with prosthetic vascular grafts (PVGs).

METHODS

This was a prospective cohort study of all consecutive patients who underwent PVG of the abdominal aorta with or without iliac-femoral involvement and peripheral PVG from April 2008 to August 2009 at a university hospital. Patients younger than 15 years and those with severe immunodeficiency were excluded. The follow-up period was until 3 years after surgery or until death.

RESULTS

There were 261 patients included; 230 (88.12%) were male, and the mean age was 67.57 (standard deviation, 10.82) years. The reason for operation was aortic aneurysm in 49 (18.77%) patients or lower limb arteriopathy in 212 (81.23%) patients. NIs occurred in 71 (27.20%) patients. Of these, 42 were surgical site infections (SSIs), of which 61.9% occurred in the lower extremities (14 superficial, 10 deep, and 2 PVG infections) and 38.1% in the abdomen (7 superficial, 7 deep, and 2 PVG infections); 15 were respiratory tract infections; and 15 were urinary tract infections. Active lower extremity skin and soft tissue infection (SSTI) at the time of surgery was a significant predictor of NI for both types of PVG (abdominal aortic PVG: adjusted odds ratio [OR], 12.6; 95% confidence interval [CI], 1.15-138.19; peripheral PVG: adjusted OR, 2.43; 95% CI, 1.08-5.47). Other independent predictors of NI were mechanical ventilation (adjusted OR, 55.96; 95% CI, 3.9-802.39) for abdominal aortic PVG and low hemoglobin levels on admission (adjusted OR, 0.84; 95% CI, 0.71-0.99) and emergent surgery (adjusted OR, 4.39; 95% CI, 1.51-12.74) for peripheral PVG. The in-hospital mortality rate was 1.92%. The probability of surviving the first month was 0.96, and significant predictors of mortality were active lower extremity SSTI (adjusted risk ratio [RR], 12.07; 95% CI, 1.04-154.75), high postsurgical glucose levels (adjusted RR, 1.02; 95% CI, 1.00-1.04), and noninfectious surgical complications (adjusted RR, 19.38; 95% CI, 2.25-167.29). The long-term mortality rate was 11.88%. The probability of surviving at 12, 24, and 36 months was 0.94, 0.92, and 0.87, respectively. Variables significantly associated with long-term death were older age (adjusted RR, 1.08; 95% CI, 1.01-1.15), high values of creatinine on discharge (adjusted RR, 1.91; 95% CI, 1.08-3.38), and an SSI with the highest adjusted RR (6.35; 95% CI, 1.87-21.53).

CONCLUSIONS

SSI was the primary NI. The risk of NI depended primarily on the presence of a lower extremity SSTI at the time of surgery, whereas mortality was determined by age, surgical complications during the operation, and SSI. These findings suggest that in those cases in which surgery is reasonably delayed, surgery should be deferred until the lower extremity SSTIs are resolved.