This review focuses particularly on recent evidence to determine which respiratory support techniques may be most advantageous to prematurely born infants. Meta-analyses of the results of randomised trials have demonstrated that, of invasive ventilation modes, only prophylactic high frequency oscillation is associated with a significant, but modest reduction in bronchopulmonary dysplasia, but both patient-triggered ventilation and volume-targeted ventilation (VTV) are associated with shorter durations of ventilation. Small numbers of infants, however, were included in the VTV trials and the level of volume targeting may be crucial to its success. Published randomised trials have failed to confirm the advantages of CPAP seen in non-randomised studies. Results of nasal non-invasive ventilation are encouraging, particularly with regard to reducing extubation failure, but larger trials are required to determine whether there are important side-effects.