A significant proportion of patients treated for acute decompensated heart failure (ADHF) suffer from worsening renal function, which is often associated with medical therapy resistance and poor outcome. In this setting, haemofiltration has been used for more than 30 years, despite inconclusive evidence for its advantages. In the last decade, a major technological advances have made available a new technique, ultrafiltration, which works at lower blood flow rates and requires only a venous access. As in a first proof-of-concept study (EUPHORIA), ultrafiltration proved to be efficacious in fluid removal in ADHF patients; this treatment was further investigated in randomized controlled trials. The RAPID-CHF trial demonstrated that ultrafiltration was more effective than medical therapy in fluid removal, even though it did not provide a greater weight loss. The UNLOAD trial thereafter showed a greater weight loss with ultrafiltration compared with diuretic therapy at 48 h after admission and a lower readmission rate at 90 days. Based on these results, the AHA/ACC and ESC guidelines consider ultrafiltration as a reasonable approach in ADHF patients with unresolved congestion notwithstanding optimal medical therapy and/or hyponatremia. However, the recently published CARRESS-HF trial would appear to challenge these recommendations as it failed to demonstrate an advantage of ultrafiltration compared with medical therapy, based on the finding of subtle clinically irrelevant changes in renal function between treatments. This review focused on the current evidence supporting the use of ultrafiltration and on a critical appraisal of the recently published CARRESS-HF trial.