Abstract:
Rationale The relationship between kidney function and venous thromboembolism (VTE) in critically ill patients is not well studied. The main objective of this study was to evaluate this relationship in patients admitted to a medical-surgical intensive care unit (ICU). Methods This was a retrospective study of 798 patients admitted to a tertiary-care ICU and prospectively followed for the development of clinically suspected and radiologically diagnosed deep venous thrombosis or pulmonary embolism. Patients were divided based on admission creatinine and dialysis history into five groups: normal kidney function, RIFLE classes R, I and F (combined = acute kidney injury [AKI]) and endstage renal disease (ESRD). We compared VTE prophylaxis practices and VTE incidence in these groups and evaluated renal failure as a VTE risk factor using multivariate Cox regression analysis. Results Of the 798 patients, 27.2percent had AKI and 10.1percent had ESRD. Unfractionated heparin use was similar in the five groups but enoxaparin use was less frequent in AKI (13.4percent) and ESRD (3.8percent) patients compared with patients with normal kidney function (39.0percent). VTE occurred in 7.6percent of patients with normal renal function, 7.8percent AKI patients and 2.5percent ESRD patients (p = 0.22). The adjusted hazard ratios for VTE compared to patients with normal kidney function were 0.35 (95percent confidence interval [CI], 0.08-1.47) for RIFLE class R, 1.19 (95percent CI, 0.83-1.70) for RIFLE class I, 0.82 (95percent CI, 0.59-1.14) for RIFLE class F and 0.71 (95percent CI, 0.49-1.02, p = 0.06) for ESRD. Conclusions Neither AKI nor ESRD was an independent risk factors for critically ill patients. © 2013 The Authors.