Abstract Background and Aims Acute kidney injury (AKI) is a clinical syndrome with heterogeneous pathophysiological mechanisms, associated with higher morbidity and mortality risk. The existing evidence shows that an episode of AKI increases the risk of chronic kidney disease and end-stage kidney disease, with considerable effects on long term outcomes. Current risk prediction scores were mostly calculated in the setting of contrast induced nephropathy and cardiac surgery reporting the patient's risk of AKI immediately before and after the invasive procedure. To date, no scoring models are available to assess the risk of progression and necessity of renal replacement therapy (RRT) for patients in the course of AKI. Method A prospective observational cohort study was conducted enrolling patients with AKI treated within the Nephrology, Dialysis and Kidney Transplant Unit of IRCCS Azienda Ospedaliero-Universitaria di Bologna from January 2022 to July 2023. The inclusion criteria were: 1) age >18 years, 2) stage 3 AKI according to the current KDIGO guidelines or presence of urgent RRT indication (severe hyperkalemia, metabolic acidosis or fluid overload refractory to medical treatment). The indication for RRT was given based on clinical judgement. Type of RRT, dialyzer and ultrafiltration rate were applied according to the clinical features of the patient. The primary endpoint was the association between AKI features and RRT. Secondary endpoints were evaluation of patients treated with RRT for factors associated with renal recovery. Results We studied 56 patients. The mean age ± SD was 69.8 ± 15.3 years (range, 18-94) and 37 (66.1%) were males. 18 patients (32.1%) were treated with conservative therapy and 38 (67.9%) with RRT. The factors correlated with the necessity of RRT were previous cardiovascular disease (p = 0.002), diabetes mellitus (p = 0.017), Charlson score >5 (p = 0.035), volume overload (p = 0.002), urine output level (p = 0.033) and C reactive protein (CRP; p = 0.020). Multivariate analysis found significant association of cardiovascular disease (OR, 5.94; 95% CI, 1.45 to 24.34; p = 0.013) and elevated CRP (OR, 1.08; 95% CI, 1.00 to 1.16; p = 0.049) during AKI episode with the RRT need in the study cohort. Higher ultrafiltration rate during RRT was correlated with non recovery of kidney function (OR, 1.86; 95% CI, 1.09 to 3.14; p = 0.021). Conclusion Cardiovascular disease and higher CRP were associated with necessity of RRT. More intensive fluid removal during RRT resulted as negative predictor of kidney function recovery. Further studies with larger study samples and randomized design are necessary to confirm these findings.