IDEAL-ICU examined whether renal-replacement therapy should be started early in patients with septic shock and severe acute kidney injury who did not yet have a life-threatening indication for dialysis. Patients were randomised to an early strategy or a delayed strategy in which renal replacement was deferred for 48 hours if kidney function did not recover. The trial was stopped early because of slow recruitment and did not demonstrate a survival advantage for the early strategy. A meaningful proportion of patients assigned to the delayed group recovered renal function without dialysis. These findings reinforced the principle that laboratory severity alone should not automatically trigger immediate renal-replacement therapy. For the intensivist, the decision should integrate urine output, acid-base status, potassium, fluid balance, uraemic complications and the trajectory of kidney recovery. IDEAL-ICU is especially relevant in septic shock, where patients often have rapidly changing haemodynamics and may recover kidney function once perfusion and infection are controlled. Starting renal replacement too early can expose patients to catheter complications, anticoagulation and treatment-related haemodynamic instability without improving outcomes. The trial therefore supports selective timing rather than a fixed “early” threshold, while reminding clinicians that patients with clear emergency indications should not be delayed simply because they are in a protocol designed to test timing.