IMPORTANCE Clinicians aim to extubate preterm infants as early as possible, to minimize the risks of mechanical ventilation. Extubation is often unsuccessful owing to lung disease or inadequate respiratory drive. OBJECTIVE To conduct a systematic review and meta-analysis of interventions to improve rates of successful extubation in preterm infants. DATA SOURCES Searches were undertaken in PubMed and The Cochrane Library. STUDY SELECTION The review was conducted using the methods of the Cochrane Collaboration and Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. Studies were included if they were randomized clinical trials published in English, enrolled intubated preterm infants (born <37 weeks' gestation), and reported 1 or both of the primary outcomes. DATA EXTRACTION AND SYNTHESIS One thousand three hundred seventy-nine titleswere screened independently by 2 investigators to assess need for full-text review. Disagreements were resolved via consensus of all authors. Where no Cochrane Review existed for an intervention, or not all identified studies were included, a new pooled analysis was performed. MAIN OUTCOMES AND MEASURES Primary outcomeswere treatment failure or reintubation within 7 days of extubation. RESULTS Fifty studies were eligible for inclusion. Continuous positive airway pressure reduced extubation failure in comparison with head-box oxygen (risk ratio [RR], 0.59; 95%CI, 0.48-0.72; number needed to treat [NNT], 6; 95%CI, 3-9). Nasal intermittent positive pressure ventilation was superior to continuous positive airway pressure in preventing extubation failure (RR, 0.70; 95%CI, 0.60-0.81; NNT, 8; 95%CI, 5-13). High-flow nasal cannula therapy and continuous positive airway pressure had similar efficacy (RR, 1.11; 95%CI, 0.84-1.47). Methylxanthines reduced extubation failure (RR, 0.48; 95%CI, 0.32-0.71; NNT, 4; 95%CI, 2-7) compared with placebo or no treatment. Corticosteroids (RR, 0.18; 95%CI, 0.04-0.97; NNT, 12; 95%CI, 6-100) and chest physiotherapy (RR, 0.32; 95%CI, 0.13-0.82; NNT, 15; 95%CI, 7-50) both reduced extubation failure rates but were associated with significant adverse effects. Doxapram did not aid successful extubation (RR, 0.80; 95%CI, 0.22-2.97). CONCLUSIONS AND RELEVANCE Preterm infants should be extubated to noninvasive respiratory support. Caffeine should be used routinely, while corticosteroids should be used judiciously, weighing up the competing risks of bronchopulmonary dysplasia and neurodevelopmental harm.