Pulmonary inflammation plays an important role in the pathogenesis of Chronic Lung Disease (CLD) in preterm infants. Glucocorticoids have been used to treat or prevent CLD for over a decade ; however because of the systemic side effects, glucocorticoids are not recommended for routine use. Nevertheless, the local beneficial effects of glucocorticoids on the respiratory tract are well documented. One alternative to systemic administration is delivery of glucocorticoids by inhalation or by direct intratracheal instillation. However, inhalation therapy in premature infants is often technically difficult and their effect has been shown limited. Our previous double- blind study indicated that early intratracheal instillation of budesonide using surfactant as vehicle can effectively deliver steroid and improve the combined outcome of mortality or CLD morbidity in preterm infants with severe RDS. However, the crucial question is : does this therapeutic method has long term adverse effects? To answer this question, a follow- up study is therefore proposed. The purpose of the study is to investigate the outcome at about 2-3 years corrected age of infants who were enrolled in a double- blind controlled trial of early intratracheal instillation(< 4 hours) of budesonide for prevention of CLD. A total of 88 children (50 in the budesonide treatment group and 38 in the control group) who survived the initial study period will be followed. All these infants had birth weight < 1500gm and had severe RDS requiring mechaincal ventilation and FIO2 >0.6 shortly after birth. The following variables will be evaluated: Interim medical history, Socioeconomic background, physical growth, neurological examination, mental and psychomotor developmental index (MDI and PDI) of Bailey Scale (), electroencephalogram and auditory Ⅱand visual evoked potential. Since most of the Budesonide treatment group received only one or two doses of budesonide and since budesonide is quickly metabolized in the liver, as shown in our pharmacokinetic data, the long term effects could be minimal and neglible. Therefore we speculate that the outcomes between the Budesonide treated and control group would be comparable. This result will substentiate our recommendation that our therapeutic method can be routinely administered to all infants with severe RDS. This would be a break through in the treatment or prevention of CLD in preterm infant.