Abstract
Background and Aims
Acute kidney injury (AKI) is a common and serious problem associated with poor prognosis. The aim of the study was to reveal the prevalence and predictors of community-acquired AKI in patients with acute cardiac diseases.
Method
566 patients (278 with acute decompensated heart failure (ADHF), 288 with non-ST-elevation acute coronary syndrome (NSTE-ACS), 46% male, 71±11 years (M±SD), smokers 26%, arterial hypertension 91%, previous myocardial infarction (MI) 45%, diabetes mellitus (DM) 28%, atrial fibrillation 35%, chronic kidney disease (CKD) 46%, previous hospitalization with ADHF 36%, ejection fraction (EF) <35% 15%, blood pressure (BP) 142±30/83±16 mmHg) were examined. AKI was diagnosed according 2012 KDIGO Guidelines. Community-acquired AKI was identified in patients with elevated serum creatinine levels on admission, which decreased during hospitalization.
Results:
Incidence of AKI in all patients, patients with ADHF and NSTE-ACS was 40, 43.5 and 37.2%. In-hospital mortality in patients with AKI was higher than in those with stable kidney function (14.9 vs 3.6%, p<0.001). Community-acquired AKI was present in 18% of patients (20.5 and 15.6% in ADHF and NSTE-ACS respectively), in-hospital mortality was 16.7% (10.5 and 24.4% respectively). The risk assessment scale for community-acquired AKI was developed based on independent predictors of AKI, using binary logistic regression and ROC analysis (AUC 0.860, 95% CI 0.821-0.898).
Independent variables included in the model, and the corresponding points (pts) are listed below: clinical and demographic characteristics (male gender - 6 pts, alcohol abuse - 7 pts, DM - 1 pt), present on admission (MI - 5 pts, AHF/ADHF - 9 pts, systolic BP <120 - 10 pts, <110 - 15 pts, <90 mmHg - 27 pts; state of kidney function on admission: serum creatinine >98 and >128 mkmol/L - 14 and 22 pts, GFRCKD-EPI <45 and <15 ml/min/1.73 m2 - 7 and 14 pts; glucose level >7 mmol/L - 4 pts), outpatient intake of ACE inhibitors - 4 pts, absence of spironolactone in outpatient therapy - 1 pt. Diagnostically significant risk score for predicting AKI was >30 pts, the risk prediction model showed sensitivity 89%, specificity 66%.
Conclusion
Community-acquired AKI is common in patients in acute cardiovascular events, is associated with high mortality, and often is underdiagnosed. Usage of risk assessment scale in clinical practice may help to detect patients with high-risk of AKI on admission. Baseline kidney function and blood pressure level are main predictors of AKI in patients admitted with acute cardiac diseases.