The rate of recurrent preterm birth for women in the placebo arm of the National Institute of Child Health and Human Development Maternal-Fetal Medicine Units (MFMU) Network randomized trial of 17 alpha-hydroxyprogesterone caproate (17-OHPC) was 54.9%. This rate has been called “strikingly high,” thus “calling into question the apparent risk reductions associated with progesterone therapy.” Because the recurrence rate in the treatment arm (36.9%) was similar to recurrence rates in another MFMU Network observational study, the high rate of recurrence has been suggested to be the result of uterotonic effects of castor oil, the vehicle used in both arms of the Meis trial. Despite explanations from the study’s lead author and sponsor that the population enrolled was at particularly high risk, the Meis trial has been haunted by these concerns since the day of publication.
See related article, page 453
The study reported by O’Sullivan et al in this issue of the Journal is driven by this concern, and may contribute further to a perception that the rate in the placebo group of the Meis trial was in fact too high. As the American Journal of Obstetrics and Gynecology editor who accepted the manuscript by O’Sullivan et al, and a coauthor of the Meis study, I’d like to provide an alternate view. I think a recurrence rate of 55% is about what one would expect from the demography and obstetric history of women enrolled into the placebo arm of the Meis trial. There are 3 historical factors associated with the likelihood of recurrent preterm birth: maternal race (black vs non-black), gestational age of the index preterm birth, and the number of previous preterm births. Each confers about a 1.5- to 2-fold increase in the risk of preterm birth beyond the 1.5- to 2-fold risk associated with a prior preterm birth not further specified. Recalling that women who enrolled may have been strongly motivated to participate in that they agreed to receive 5 months of weekly injections, let’s look at the study population in the MFMU Network trial of 17-OHPC.
- •
First, their demography: almost 60% of women enrolled were African American, a population whose preterm birth rate exceeds 17%. The risk of recurrent preterm birth is also increased in African Americans compared with women from other backgrounds. See Table from Kistka et al.
TABLE 3
Risk of subsequent preterm (20-34 6/7 weeks’ gestation) birth in Missouri to a mother with either an initial preterm or full term birth, according to race, 1989-1997
Rate of preterm birth (per 100 live births)
Initial preterm
Initial full term
OR
CI (95%)
All births
12.34
3.59
3.78
3.66-3.91
Black
21.50
8.51
3.68
3.53-3.85
White
9.18
2.49
2.53
2.40-2.66
Iams. Preterm birth rate in the MeisMFMUNetwork trial of 17-OHPC. Am J Obstet Gynecol 2010.
Reproduced, with permission, from Kistka et al.
- •
Second, their obstetric history: the risk of recurrent preterm birth rises with the number of prior preterm births and as gestational age of the qualifying preterm birth declines (see Figures 1 and 2 from McManemy et al ). In the Meis study, the mean gestational age of the qualifying preterm birth was 31 weeks, and 32% of enrollees had >1 prior preterm birth. Notably, there were significantly more women in the placebo group who had >1 prior preterm birth (27.7% in the treatment group vs 41.2% in the placebo group). In the dataset of McManemy et al, among women of all races who had ≥2 preterm births in which 1 was <32 weeks, the recurrence rates exceeded 50% (see Figure 2 from McManemy et al ).
