The aim of this study was to determine whether elective use of nasal continuous positive airways pressure (CPAP) following extubation of preterm infants was well tolerated and improved short- and long-term outcomes. A randomized comparison of nasal CPAP to headbox oxygen was undertaken and a meta-analysis performed including similar randomized trials involving premature infants less than 28 days of age. A total of 150 infants (median gestational age 30 weeks, range 24-34 weeks) were randomized in two centres. Fifteen nasal CPAP infants and 25 headbox infants required increased respiratory support post-extubation and 15 nasal CPAP infants and nine headbox infants required reintubation (non significant). Eight infants became intolerant of CPAP and were changed to headbox oxygen within 48 h of extubation; 19 headbox infants developed apnoeas and respiratory acidosis requiring rescue nasal CPAP, 3 ultimately were re-intubated. Seven other trials were identified, giving a total number of 569 infants. Overall, nasal CPAP significantly reduced the need for increased respiratory support (relative risk, 0.57, 95% CI 0.43-0.73), but not for re-intubation (relative risk 0.89, 95% CI 0.68-1.17). Nasal CPAP neither influenced significantly the intraventricular haemorrhage rate reported in four studies (relative risk 1.0, 95% CI 0.55, 1.82) nor that of oxygen dependency at 28 days reported in six studies (relative risk 1.0, 95% CI 0.8, 1.25). In two studies nasal CPAP had to be discontinued in 10% of infants either because of intolerance or hyperoxia. CONCLUSION: Elective use of nasal continuous positive airways pressure post-extubation is not universally tolerated, but does reduce the need for additional support.