No definite adverse effects on mother or infant were observed in any of the studies, though concerns remained—and, indeed, still exist—about an increased risk of infection in women with PROM. Of note, no studies published after the original Liggins and Howie article demonstrated an increased risk for the infants of hypertensive mothers.
It became common practice at many centers to repeat weekly steroid courses, so the fetus was never more than 7 days from treatment.
Very little data exist on the long-term development of children treated with these drugs. One study followed children who received glucocorticoids and those who were given placebo up to the age of 12 years.9 Of 102 survivors in the study, 90 were available for follow up. The result: Antenatal corticosteroid therapy appeared to have no adverse effects on growth or development in these children, though there was a trend toward delayed pubertal development in treated males.
TABLE 1
Effective doses of antepartum corticoids for fetal maturation
| BETAMETHASONE | 12 mg IM every 24 hours for 2 doses* |
| DEXAMETHASONE | 6 mg IM every 12 hours for 4 doses* |
| *Optimal timing is to start treatment between 2 and 7 days before delivery. | |
Effects of the NIH report
Following the 1994 report, antepartum corticoid use greatly increased. In many areas, up to 60% of infants delivered prematurely were receiving treatment. But a crucial question remained unanswered by the consensus panel: How soon after the first injection would the infant have to be delivered in order for the glucocorticoids to maintain a protective effect?
In their original article, Liggins and Howie noted they could not demonstrate a statistically significant reduction in mortality or in the incidence or severity of RDS for neonates born more than 7 days after treatment.3 Unfortunately, many readers mistakenly assumed no statistically significant benefit meant there was no benefit at all. This error was eventually developed into the concept of the “steroid window”: a period extending from 48 hours to 7 days after the first steroid injection, during which the fetus would derive benefit from treatment.
What some clinicians failed to consider was that by 1 week after admission to the study, many women in both the control group and the treated group had delivered. As a result, the analysis for infants born more than 7 days after treatment were based on less data than the statistics for those delivered earlier in the study. Also, because these fetuses experienced an extra week of gestational development, they were not at as high a risk for morbidity and mortality as those born closer to the time of treatment initiation. Therefore, the statistical improvements would not be as dramatic. Nevertheless, once this “window of benefit” concept was accepted, it became common practice at many centers to repeat weekly steroid courses, so the fetus was never more than 7 days from steroid treatment. These weekly courses were continued until the fetus reached the gestational age of presumed pulmonary maturity.
Many institutions also expanded the indications for glucocorticoid treatment. At some hospitals, patients with a history of preterm delivery were started on weekly courses of steroids as early as 24 weeks into the pregnancy, to “keep them in the window” in case of premature delivery. Patients admitted to a hospital with a placenta previa were put on similar treatment regimens. Although this liberal use of therapy benefited many neonates, it also exposed numerous patients to repeated courses for which there was no scientific evidence. This led the NIH to reconvene the Consensus Development Conference in 2000 to consider the issue of repeat courses.
Liberal use of therapy exposed numerous patients to repeated courses for which there was no scientific evidence.
Updated conclusions, protocols
The reconvened consensus panel reviewed the available published human and animal studies, as well as information presented at both consensus conferences.10
At least one animal study suggested that repeat courses resulted in improved lung mechanics and gas exchange compared to single courses,11 though others demonstrated that repeat courses had deleterious effects on somatic and brain growth, lung growth and organization, and retinal development.12-14
No randomized controlled human studies addressing this question were available, and data from nonrandomized human studies, all of which were retrospective and inadequately controlled, were of limited quality. These studies suggested that repeat courses resulted in a reduction in the severity of respiratory disease,15 but did not show convincing evidence of reduced mortality or lower long-term morbidity. Of note, none of the humans studies included infants who had multiple courses of antepartum steroids and then delivered near term. All told, conclusions were inconsistent and often conflicting, with some suggesting adverse effects while others found none.