Objective: The objective was to review the evidence regarding multiple versus single courses of antenatal corticosteroids (ACS), devise dosing regimens based on this evidence, and then, by using decision analysis, compare 5 ACS regimens: (1) single course, (2) weekly courses, (3) every-other-week dosing, (4) every-other-week dosing with no more than 2 courses, and (5) an every-other-week dosing regimen with no third courses after 30 weeks and no second course after 32 weeks. Study Design: The literature was examined for evidence regarding a dose response of the benefits and detriments of ACS. A decision analytic model was designed by using odds ratios, risks of morbidity and mortality, and birth rates from the literature. Mathematic modeling, based on data in the literature, was used to estimate the number of patients who would present at risk for preterm delivery and at what gestational age they would subsequently be delivered. A sensitivity analysis was performed to account for uncertainties in the data. Results: Of approximately 4 million annual births in the United States, a conservative estimate of 138,000 patients will present between 24 and 34 weeks of gestation at risk for preterm delivery, with 91,915 deliveries. If the 5 different dosing regimens are applied to these patients, strategy 1, the single course of ACS, would result in 30,232 cases of respiratory distress syndrome (RDS) and 4032 neonatal deaths. Three of the other dosing regimens (2, 3, and 5) would all decrease the number of cases of RDS but would concomitantly increase neonatal deaths. Only dosing regimen 4 would decrease the cases of RDS by 2187 without increasing the number of neonatal deaths. Conclusion: There is evidence that there may be a decrease in the incidence of RDS secondary to the use of multiple doses of ACS; however, a concomitant increase in neonatal deaths may also occur. On the basis of our results, we recommend the following: (1) all fetuses between 24 and 34 weeks' gestation at risk for preterm delivery should be given the first course, (2) if there is a persisting risk of preterm delivery subsequent to this, the next course should be given 2 weeks later, and (3) no more than 2 courses should be given. These recommendations need to be examined in a randomized, controlled trial. (Am J Obstet Gynecol 2002;186:1221-9.)
Read full abstract