Background: Invasive respiratory support is associated with risk and complications including mortality and neurological impairments. Consequently, extubation of a ventilated infant should be as early as possible. Objective: This study aimed to assess the efficacy of spontaneous breathing trial as indicator for the success of extubation in mechanically ventilated preterm infants. Patients and methods: A prospective cohort study included 62 preterm born infants who were maintained on mechanical ventilation. They were divided into: (32 infants) group for whom a spontaneous breathing trial was carried out for 5 minutes. Second (30 infants) group for whom a spontaneous breathing test was carried out for 3 minutes. Then, rapid Shallow Breathing Index (RSBI) was calculated for each patient. At the end of the test, the newborns were extubated and placed on continued positive airway pressure (CPAP) or just oxygen, as needed, according to the unit's routine protocol. Results: On multivariate logistic regression of factors associated with failure of weaning, APGAR at 5 minutes was > 6, absence of maternal PIH, birth weight > 400 gm. Pre-extubation MAP < 5.5 and preextubation PO₂ > 28 were protective against failure of weaning. On the other hand, lower preextubation PCO₂ was a predictor of weaning failure (increase risk of failure by about 63 times). Failure of weaning forecated in RSBI trial can predict actual failure of weaning with sensitivity of 97.9%, specificity of 73.3%, positive predictive value of 92%, negative predictive value of 91.7% and accuracy of 91.9%. Conclusion: Failure of weaning associated with lower birth weight, PO2, PCO2 and higher pre-extubation MAP. 3 minutes and 5 minutes spontaneous breathing trial (SBT) can predict actual failure of weaning with sensitivity 97.9%, specificity 73.3%.