Abstract Background and Aims Severe acute kidney injury (AKI) requiring renal replacement therapy (RRT) is a serious clinical disorder in the intensive care unit (ICU), occurring in a substantial proportion of critically ill patients. The aim of our single centre retrospective observational study was to analyse the outcomes of patients admitted to a non-surgical ICU and treated with RRT and with/without continuous hemoperfusion with Cytosorb. Method One hundred critically ill patients (mean age, 64.3 years; 69 men) admitted to the ICU and requiring RRT for AKI were analysed. Patient demographics, concomitant diseases, type of RRT, and survival were obtained from the medical record. APACHE II and Sofa Scores on admission were calculated. 30-day mortality was assessed using Kaplan-Meyer or Cox proportional hazards models. Results Reasons for ICU admission were acute respiratory failure (39%), cardiopulmonary resuscitation (14%), shock (13%), acute coronary syndrome (9%), sepsis (3%), acute pancreatitis (3%), and other (19%). Prior comorbidities were hypertension (70%), diabetes (43%), heart failure (32%), chronic kidney disease (CKD) (30%), coronary artery disease (CAD) (27%), chronic obstructive pulmonary disease (COPD) (15%), malignancies (12%). Eighty-six patients were treated with continuous RRT (CRRT) and 14 with intermittent hemodialysis (IHD). Twenty-four (24%) of patients treated with CRRT were also treated with hemoperfusion with Cytosorb. Using an independent-samples T test, we compared the two groups of patients with respect to the use of Cytosorb (Table 1). 30-day mortality was 82% in all patients and 87.5% in patients treated with CRRT and Cytosorb. Among concomitant diseases, only patients with previous heart failure had worse survival (p = 0.032), previous CKD, CAD, COPD, malignancy had no statistically significant impact. We found no statistically significant differences in 30-day mortality between patients treated with CRRT+Cytosorb and patients treated with CRRT or IHD alone. Multivariate Cox proportional hazard regression showed that of all the variables in the statistical model (age, sex, body mass index, previous diabetes, C-reactive protein, lactate, procalcitonin, serum creatinine, mean arterial pressure, APACHE II, ultrafiltration between CRRT), only lactate levels on admission (p = 0.002; 95%CI 1.08-1.38) were significant predictor of survival. Conclusion The use of hemoperfusion with Cytosorb in ICU patients with AKI did not reduce 30-day mortality. Patients with prior heart failure had a worse outcome. Serum lactate levels at ICU admission were an independent highly prognostic factor for death within 30 days of admission.