OBJECTIVE The effectiveness of the management of preterm birth remains an important health care issue, especially when considering that more than two thirds of singleton neonatal death occurs in preterm labor. The purpose of this study was to compare oral celecoxib with intravenous magnesium sulfate as tocolytic. METHODS This was a randomized study of patients who were between 24 and 34 weeks of gestation with preterm labor. One hundred and four pregnant women with preterm labor were randomly assigned to receive celecoxib 100 mg b.i.d. for 48 h or intravenous magnesium sulfate (MgSO4) for maximum of 48 h. Outcome variables included delay of delivery for 48 h and the incidence of side-effects. Data was analyzed using the Student t-test and the chi(2) test. RESULTS There was no difference between the groups over the course of the study in demographic characteristics, cervical examination and amniotic fluid index. Labor was arrested for 48 h was in 42 (81%) and 45 (87%) of the patients in the celecoxib and magnesium sulfate groups, respectively (p-0.298). There were no severe maternal or neonatal complications in either group. CONCLUSIONS Celecoxib is as effective as magnesium sulfate for primary tocolysis.