178 Evaluation of the microbiological characteristics and risk factors for Gram-negative bloodstream infection in kidney transplant

Friday, March 19, 2010
Grand Hall (Hyatt Regency Atlanta)
Thiago Zinsly S. Camargo, MD , Federal University of São Paulo (UNIFESP-EPM), São Paulo, Brazil
Alexandre R. Marra, MD, PhD , Federal University of São Paulo (UNIFESP-EPM), São Paulo, Brazil
Moacyr Silva Jr, MD, PhD , Federal University of São Paulo (UNIFESP-EPM), São Paulo, Brazil
José Osmar M. Pestana, MD, PhD , Federal University of São Paulo (UNIFESP-EPM), São Paulo, Brazil
Luis Fernando A. Camargo, MD, PhD , Federal University of São Paulo (UNIFESP-EPM), São Paulo, Brazil
Background: Kidney transplantation improves quality and increases the survival of dialysis patients. However, control of rejection and infection are crucial for its success.

Objective: To describe the microbiological characteristics and risk factors for bloodstream infection by Gram-Negative Bacilli (GNB) in renal transplant recipients.

Methods: We conducted a case-control study of GNB bloodstream infection from kidney transplant patients at two large tertiary hospitals in São Paulo, Brazil, between 2000 and 2006. Univariate analysis was performed to evaluate the association of the following variables with GNB bloodstream infection: gender, number of transplants, acute rejection, comorbidities, induction therapy, corticosteroid use, urinary fistula, presence of double-J, cadaver donor, pre-transplant diabetes, previous dialysis, pre-bacteremia gancyclovir use and viral hepatitis. Variables significant in univariate analysis were entered into a logistic regression model.

Results: There were 129 cases of bloodstream infection by GNB during the study period. Of these 58.9% (n = 76) were male, the mean and median hospital stay were 36.1 and 18 days, respectively. Fifty-one percent (n = 66) were from deceased donors and in 20.2% (n = 26) was performed induction therapy. The 30-day mortality was 20.2% (n = 26). Of the isolates, 43.4% (n = 56) were Escherichia coli. In univariate analysis, acute rejection (p = 0.021 - OR 2.46 [1.18-5.12]), induction therapy (p = 0.05 - OR 2.07 [1.03-4.18), corticosteroid use (p = 0.03 - OR 3.26 [1.15-9.26 ) cadaver donor (p = 0.001 - OR 4.58 [2.61-8.04]) and presence of double-J (p = 0.005 - OR 2.45 [1.33-4.48]) were considered risk factors for GNB bloodstream infection. In the multivariate analysis, cadaver donor proved to be a risk factor for the event studied (p = 0.01 - OR 3.98 [2.15-7.37]).

Conclusions: There have been many advances in surgical techniques and immunosuppression for kidney transplant recipients in recent years. However, the infection remains a major problem following these patients. In this sense, knowledge of the microbiological profile and risk factors for bloodstream infection may help to improve care of these patients.