Objective We aimed to assess the optimal frequency for changing single-use enteral delivery sets during postoperative enteral feeding in infants with congenital heart disease (CHD). Methods We enrolled 120 CHD infants who were fed using an enteral nutrition pump directly connected to a milk bottle with a single-use enteral delivery set in a four-arm randomized controlled trial (ChiCTR2000039544). Patients were randomized into four groups based on the replacement frequency of the enteral delivery set (6 h, 12 h, 18 h, and 24 h groups). The primary outcome was the percentage of contaminated enteral delivery sets (overgrowth of microbiota and colonization of pathogenic bacteria). Secondary outcomes included evidence of infection, gastrointestinal tolerance, intestinal microflora dysbiosis, and healthcare costs. Results The percentages of microbial overgrowth detected in the 6 h, 12 h, 18 h, and 24 h groups were 6.7%, 30.0%, 46.7%, and 80%, respectively (P < 0.001). Significant differences were observed between the 6 h and 18 h groups (P < 0.001), the 6 h and 24 h groups (P < 0.001), and the 18 h and 24 h groups (P = 0.007). Meanwhile, pathogenic bacterial colonization was detected in 0, 4, 6, and 11 delivery sets in the 6 h, 12 h, 18 h, and 24 h groups, respectively (P = 0.002). No difference in clinical symptoms was found among the four groups. The total cost per patient in the 12 h group and the 18 h group was 340.2 RMB and 226.8 RMB, respectively. Conclusion Taking into consideration both microbial overgrowth and cost-effectiveness, the results of this study indicate that for children receiving continuous enteral feeding following CHD surgery, the optimal frequency for changing the single-use enteral delivery set when formula reconstituted from powder is used is 18 hours.