Background/Aims: Renal replacement therapy (RRT) is a rescue therapy for patients with type 1 cardiorenal syndrome (CRS) with poor prognoses. However, the optimal timing for initiation and cessation of RRT remains controversial. The purpose of this study was to determine the optimal timing of initiation and cessation of RRT for patients with type 1 CRS. Methods: In this retrospective analysis, patients with refractory type 1 CRS receiving RRT were divided into 3 groups according to weaning from RRT and death within 90 days. Baseline characteristics, underlying heart disease, comorbidities, drug use before RRT, indicators of RRT initiation, and prognosis were compared between the 3 groups. Results: Fifty-two patients were enrolled, which included 27 males and 25 females with a mean age of 70.7 ± 16.1 years and a 90-day mortality rate of 65.4%. The mean urine output before RRT initiation was 800 mL/ 24 h in the RRT-independent group, 650 mL/24 h in the RRT-dependent group, and 345 mL/ 24 h in the death group (p = 0.021). Additionally, there were obvious differences in fluid balance between the 3 groups (167, 250, and 1,270 mL, respectively, p = 0.016). Patients could be successfully weaned from RRT when urine output was >880 mL and fluid balance volume was <150 mL. Conclusion: The mean fluid balance of survivors was remarkably less than that of the death group at RRT initiation. RRT termination can be considered when urine output is >880 mL/24 h and volume balance is <150 mL/24 h.