Infection in ICU patients with acute kidney injury treated with renal replacement therapy

Introduction

Patients with chronic kidney disease have decreased immunity and increased risk for infection. We evaluated the occurrence rate of infection in ICU patients with acute kidney injury, and its impact on hospital mortality.

Methods

Retrospective evaluation of all ICU patients with acute kidney injury treated with renal replacement therapy (RRT) during a 5-year period (2000–2004) in the 22-bed SICU, eight-bed CSICU, and six-bed Burn Unit of a Belgian tertiary care hospital. Infection was scored based on chart review, and the pharmacy database on antibiotic prescription.

Results

During the study period 406 patients were included. The median (IQR) age was 64 (53–73) years, 68% of patients were male, and the APACHE II score was 24 (17–30). At the start of renal replacement therapy, 69% of patients were treated with vaso-active therapy, and 73% were ventilated. Infection occurred in 87% of patients; 38% had more than one infection. The start of infection was during the treatment period with RRT in 41.4% of all infections, and in 58.6% before or after the treatment period with RRT. The incidence rate of infection was 5.92 per 100 patient-days. Patients who developed infection had a longer length of stay in the ICU (7 [15–31] days vs 5 [3–7] days; P < 0.001). There was no difference in mortality between patients with and without infection (62.5% respectively 54.9%, P = 0.295). Infection was not associated with inhospital mortality, even after adjustment for age, type of ICU admitted to, APACHE II score, and severity of illness at start of RRT (i.e. treatment with vasoactive therapy) or mechanical ventilation (odds ratio for inhospital death: 0.52 [0.21–1.28], P = 0.152) (logistic regression analysis).

Conclusion

In this cohort of ICU patients with acute kidney injury and treated with RRT, infection had a high occurrence rate of almost 90%. Also, more than one-third of these patients had more than one infection episode. Infection was not associated with increased inhospital mortality.