Incidence and mortality of acute kidney injury in coronary care unit: a retrospective study from a single center

Abstract

Objective To evaluate the incidence and mortality of acute kidney injury (AKI) in coronary care unit (CCU), and to identify the risk factors of the incidence of AKI and the mortality of CCU patients. Methods A total of 414 patients in CCU from January 1, 2014 to June 1, 2015 at Zhongnan Hospital of Wuhan University were enrolled. Based on the KDIGO-AKI criteria, these patients were classified into two groups: NAKI group (patients without AKI) and AKI group. Clinical characteristics and laboratory data of two groups were compared. The risk factors of the incidence of AKI and the mortality of CCU patients was analyzed by logistic regression, and then the receiver operating characteristic (ROC) curve was drawn to evaluate the predictive value of these risk factors. Results (1) Among 414 patients, 136(32.9%) patients fulfilled the criteria for AKI, and 14.0% patients in AKI stage 1, 10.9% in AKI stage 2 and 8.0% in AKI stage 3. (2) The total CCU mortality was 15.0%. Mortality of AKI patients in the CCU was 33.3%, higher than 6.1% in patients without AKI (OR=7.735, 95%CI 4.215-14.196, P<0.001). The mortality worsened with increasing severity of AKI (22.4% for AKI stage 1 group, 37.8% for AKI stage 2 group, 45.4% for AKI stage 3 group). (3) Anemia (OR=8.274, 95% CI 4.363-15.689), history of chronic illness (OR=2.582, 95% CI 1.400-4.760), APACHEⅡ scores (OR=1.813, 95%CI 1.739-1.895), male (OR=3.666, 95%CI 1.860-7.226) were the independent risk factors for AKI, while the normal mean arterial pressure (MAP) (OR=0.292, 95%CI 0.153-0.556) and normal estimated glomerular filtration rate (eGFR) (OR=0.166, 95%CI 0.090-0.306) are the protective factors for AKI (all P<0.05). (4) AKI was the most powerful independent factor associated with the mortality of CCU patients (OR=7.050, 95% CI 2.970-16.735, P<0.001). Other independent risk factors for CCU mortality included history of chronic illness, ejection fraction and APACHE Ⅱ ≥15 scores (all P<0.05), while the normal MAP and normal eGFR were the protective factors (all P<0.05). (5) For predicting AKI, eGFR displayed an excellent areas under the ROC curve (AUC=0.815, P<0.001), and for CCU mortality, APACHEⅡ scores had the highest overall correctness of prediction (AUC=0.757 P<0.001). Conclusions CCU patients have high morbidity of AKI, which is the most powerful independent factor associated with the increased CCU mortality. The eGFR is the best predictor for AKI, and then through the evaluation of eGFR for CCU patients, we can evaluate high-risk groups, make early interventions and then improve the prognosis of CCU patients. Key words: Coronary care units; Renal insufficiency, acute; Incidence; Risk factors