3.2 | Risk of bias assessment for all included studies 
All included studies were generally considered to be low or moderate risk according to the Cochrane assessment standards for RCT. Among them, 11 trials 14,16,18-26 reported the method of random sequence generation, and 11 trials 14,16,19-26described the specific allocation concealment methods. Due to the particularity of the intervention, blinding was not possible in all trials, and blinding of researchers and subjects was assessed as high risk. However, almost all studies described the criteria for surfactant delivery 14-26 and rescue intubation14,16-26, so as to minimize such bias. Twelve articles14,16-26 reported complete outcome data. For reporting bias, five studies 15,17,18,22,23,26 showed unclear bias because they were not registered in a public trial registry, and three studies 20,24,25 showed high risk (Table 2).