Objective To investigate the effects of normal saline flushing and its frequency on extracorporeal circuit lifespan and solute removal in continuous renal replacement therapy (CRRT) without anticoagulation. Methods Patients undergoing continuous veno-venous hemodiafiltration (CVVHDF) without anticoagulation between June and September 2021 were prospectively collected. The patients were randomly divided into three groups by envelope method, namely 30 min-flushing group (flushing every 30 minutes for extracorporeal circulation), 2 h-flushing group (flushing every 2 hours for extracorporeal circulation), and non-flushing group (no flushing for extracorporeal circulation during treatment). All patients were treated with Prismaflex V8.0 CRRT machine and matched AN69ST-ST150 extracorporeal circuit, through either femoral or internal jugular venous double-lumen catheter. CVVHDF was adopted, the blood pump rate was 200 mL/min, and the rates of replacement fluid and dialysate were both 1 000 mL/h. The replacement fluid was pre-post dilution. Extracorporeal circuit lifespan, treatment time delayed by flushing, overall treatment time of CRRT, actual treatment time of CRRT, proportion of actual treatment time achieved, delayed daily treatment time, and small molecule solute removal efficiency before and after treatment were recorded. Results A total of 83 patients were included, including 24 in the 30 min-flushing group, 30 in the 2 h-flushing group, and 29 in the non-flushing group. There were significant differences in the indexes of extracorporeal circuit lifespan and various treatment time indicators among the three groups (P<0.05). Compared with the 2 h-flushing group and the non-flushing group, the 30 min-flushing group significantly shortened the extracorporeal circuit lifespan, delayed more treatment time by flushing, and delayed the longest daily treatment time (P<0.05). The proportion of actual treatment time in the non-flushing group was significantly higher than that in the 30 min-flushing group and the 2 h-flushing group (P<0.05), and in the 2 h-flushing group was also higher than that in the 30 min-flushing group (P<0.05). There was no significant difference in the blood urea nitrogen clearance rate among the three groups (P=0.570), but the serum creatinine clearance rate was significantly different among the three groups (P=0.020). Compared with the 30 min-flushing group, the 2 h-flushing group had a higher serum creatinine clearance rate, and there was statistical significance (P<0.05). Twenty-five patients had hypotension during treatment. The frequency of 30 min-flushing caused a higher risk of coagulation during cardiopulmonary bypass (hazard ratio=2.502, P=0.001). Conclusion For CVVHDF without anticoagulation, longer extracorporeal circuit lifespan and longer effective treatment time can be achieved without using normal saline flush.