|
|
Background In singleton gestations, 17 alpha-hydroxyprogesterone caproate (17P) has been shown to reduce the rate of recurrent preterm birth. This study was undertaken to evaluate whether 17P would reduce the rate of preterm birth in twin gestations.
Methods We performed a randomized, double-blind, placebo-controlled trial in 14 centers. Healthy women with twin gestations were assigned to weekly intramuscular injections of 250 mg of 17P or matching placebo, starting at 16 to 20 weeks of gestation and ending at 35 weeks. The primary study outcome was delivery or fetal death before 35 weeks of gestation.
Results Six hundred sixty-one women were randomly assigned to treatment. Baseline demographic data were similar in the two study groups. Six women were lost to follow-up; data from 655 were analyzed (325 in the 17P group and 330 in the placebo group). Delivery or fetal death before 35 weeks occurred in 41.5% of pregnancies in the 17P group and 37.3% of those in the placebo group (relative risk, 1.1; 95% confidence interval [CI], 0.9 to 1.3). The rate of the prespecified composite outcome of serious adverse fetal or neonatal events was 20.2% in the 17P group and 18.0% in the placebo group (relative risk, 1.1; 95% CI, 0.9 to 1.5). Side effects of the injections were frequent in both groups, occurring in 65.9% and 64.4% of subjects, respectively (P=0.69), but were generally mild and limited to the injection site.
Conclusions Treatment with 17 alpha-hydroxyprogesterone caproate did not reduce the rate of preterm birth in women with twin gestations. (ClinicalTrials.gov number, NCT00099164
[ClinicalTrials.gov]
.)
It is unknown whether 17P can reduce the rate of preterm birth in women with other risk factors. One logical candidate risk factor is twin gestation. Twin gestations are increasingly common, and more than half result in premature births. Between 1980 and 2004, the rate of twin births rose dramatically in the United States (from 18.9 to 32.2 per 1000 live births), as did the absolute number (from 68,339 to 132,219).1 Consequently, the morbidity and mortality burden of twins is increasingly disproportionate and substantial: almost one in four very-low-birth-weight infants (below 1500 g) born in the United States are twins, as are one in six infants who die in the first month of life.1,4 Therefore, we evaluated the efficacy of 17P in preventing preterm birth in women with twin gestations.
Methods
Recruitment
Recruitment to this placebo-controlled, double-blind, randomized clinical trial of 17P for the prevention of preterm birth in twin gestations was undertaken at 14 sites from April 2004 through February 2006. Women carrying twins with a gestational age of at least 16 weeks and no more than 20 weeks 3 days were eligible. The exclusion criteria were serious fetal anomalies, spontaneous death of a fetus after 12 weeks, presumed monoamnionic placenta, suspected twin-to-twin transfusion syndrome, marked ultrasonographic growth discordance (a difference of at least 3 weeks of estimated gestational age between fetuses), planned nonstudy progesterone therapy after 16 weeks, in-place or planned cerclage, major uterine anomaly (e.g., bicornuate uterus), treatment with 10,000 or more units of unfractionated heparin per day, treatment with low-molecular-weight heparin at any dose, and major chronic medical diseases (e.g., insulin-requiring diabetes mellitus or pharmacologically treated hypertension). Twin gestations that were the result of intentional fetal reduction were also excluded.
An ultrasonographic examination was required between 12 weeks and 20 weeks 6 days of gestation to confirm the duration of gestation and to screen for major fetal anomalies. For women who conceived spontaneously, the duration of gestation at the time of randomization was determined according to a previously described algorithm on the basis of the last menstrual period and the results of ultrasonography of the larger fetus.5 For women who conceived by in vitro fertilization, the duration of gestation was calculated on the basis of the date of embryo transfer and the age of the embryos when transferred.
The study was approved by the institutional review boards at each clinical site and at the data-coordinating center. All women gave written informed consent before enrollment in the study.
Protocol
Eligible, consenting women were given a trial intramuscular injection of the placebo and were scheduled for a randomization visit not later than 20 weeks 6 days of gestation. At that visit, women who still met none of the exclusion criteria were assigned to receive identical-appearing injections of active agent (250 mg of 17P) or placebo (castor oil) prepared by a research pharmacy. The simple urn method of randomization6 with stratification according to clinical center was used by the George Washington University Biostatistical Coordinating Center to create a randomization sequence for each center, and the boxes of 17P and placebo were packaged for each center according to the randomization sequences. The participating women, their caregivers, and the research personnel were unaware of the women's study-group assignments. After entering the study, the women returned for weekly injections through the end of the 34th week of gestation or until delivery, whichever occurred first. At each visit, they underwent systematic assessment for side effects. Otherwise, the women received usual clinical care.
After delivery, study personnel reviewed delivery, newborn, and postpartum records and documented the date of delivery, the birth weight of the infants, and the neonatal course, as well as the occurrence of complications of pregnancy and obstetrical interventions. The infants were followed until discharge from the hospital of birth, or, if they were transferred, until discharge from the transfer hospital.
Study Outcomes
The primary study outcome was a composite of delivery or fetal death before 35 completed weeks of gestation (245 days). Fetal death includes miscarriage, termination of pregnancy, and stillbirth. Prespecified secondary outcomes included the time from randomization to fetal death or delivery, a composite of serious adverse fetal or neonatal outcomes, and selected individual maternal and neonatal outcomes. In analyses that were not prespecified, we assessed the proportion of preterm births in each group at different gestational-age thresholds.
The data were analyzed according to the intention-to-treat principle. The unit of analysis was the pregnancy, and if the outcome occurred in either fetus or neonate, the pregnancy was considered to have met the outcome. For example, a pregnancy in which one fetus died at 22 weeks and the other was born alive at 37 weeks would be counted as meeting the primary outcome. However, for fetal or neonatal death, a proportional-odds model, which permits the distinction between the death of one fetus or neonate and the death of both, was also used to compare the two groups.
Statistical Analysis
The time to delivery or death of the first fetus was compared between the two groups by means of a proportional-hazards model with left truncation (i.e., with adjustment for gestational age at entry). Similarly, survival curves were plotted with the use of a modified product-limit estimator.7 Continuous variables were compared with the Wilcoxon rank-sum test, and categorical variables with the chi-square test or Fisher's exact test, as appropriate.
On the basis of data from two studies performed by the National Institute of Child Health and Human Development Maternal–Fetal Medicine Units Network, we estimated conservatively that in the placebo group, 35% of twins would be delivered before 35 completed weeks of gestation.8,9 Thus, a total sample size of 600 was deemed sufficient to detect a 33% reduction in the rate of delivery or death before 35 weeks, under the assumptions of a type I error (two-sided) of 5% and a power of greater than 80%. This sample size also yielded 70% power if it was assumed that there would be no reduction in indicated preterm delivery and that the one-third reduction would apply only to spontaneous preterm deliveries (on the assumption of a 4:1 ratio of spontaneous to indicated deliveries at less than 35 weeks of gestation).
We estimated that a sample size of 600 would provide at least 70% power to detect a 33% reduction in the rate of the composite of the following serious adverse fetal or neonatal outcomes: fetal or neonatal death, respiratory distress syndrome, grade 3 or 4 intraventricular hemorrhage, stage 2 or 3 necrotizing enterocolitis, periventricular leukomalacia, bronchopulmonary dysplasia, severe retinopathy of prematurity, or early-onset, culture-proven sepsis. On the basis of our previous data, we expected that at least one component of this outcome would occur in one or both fetuses or neonates in 25% of pregnancies.9
An independent data and safety monitoring committee monitored the trial and reviewed the interim results. Before the study started, the group sequential method of Lan and DeMets with the modified O'Brien–Fleming spending function was chosen for adjustment of the significance level in interim analyses.10 Two interim analyses were performed, and in the final analysis of the primary outcome, two-tailed P values of less than 0.048 were chosen to indicate statistical significance. However, since the adjustment is minimal, 95% confidence intervals are reported. For all other outcomes, a nominal P value of less than 0.05 was considered to indicate significance, and no adjustments were made for multiple comparisons.
Results
We identified 1526 eligible women, of whom 699 (45.8%) gave consent and 661 (43.3%) were randomly assigned to treatment. Outcome data were available for 655 of these 661 women and for 1310 of 1322 fetuses or infants (Figure 1). The baseline characteristics of the two study groups were similar (Table 1).
|
|
The rate of the primary outcome did not differ significantly between groups. Fetal death or delivery before 35 weeks of gestation occurred in 41.5% of pregnancies in the 17P group (135 of 325) and 37.3% of pregnancies in the placebo group (123 of 330) (relative risk, 1.1; 95% confidence interval [CI], 0.9 to 1.3). Of those deliveries occurring before 35 weeks of gestation, 72.8% were spontaneous and 27.2% were medically indicated; the proportions of spontaneous and medically indicated deliveries were similar between the groups (Table 2). The mean gestational age at delivery did not differ significantly between groups, nor did the proportion of deliveries occurring before 37 weeks (the standard threshold for preterm birth), 32 weeks, or 28 weeks (Table 2). From the time of randomization, the estimated distributions of time to first miscarriage, fetal death, or delivery for the two groups were very similar (Figure 2). The rates of selected obstetrical interventions were similar between the groups (Table 2).
|
|
|
Side effects of the injections were frequent in both the 17P and the placebo groups, occurring in 65.9% and 64.4% of subjects, respectively (P=0.69), but were generally mild and most often limited to the injection site (Table 2). Three women (two in the 17P group and one in the placebo group) discontinued injections because of side effects; the two women receiving 17P had intense injection-site reactions. In addition, one woman in the 17P group had subjective heart palpitations and presyncope immediately after the first injection. She did not return for any study follow-up and did not provide any outcome information.
Discussion
In this randomized trial conducted among women carrying twins, weekly 250-mg injections of 17P failed to lower the rate of preterm birth, to prolong gestation, or to improve fetal or neonatal outcome.
Our results are generalizable to most women in the United States who are pregnant with twins. The study subjects were drawn from a broad geographic area and were racially and ethnically diverse. Roughly half were nulliparous, and two thirds conceived spontaneously. Although most of the subjects were recruited from academic medical centers, they were not at unusually high risk for preterm birth. On average, the women in this trial delivered at 34.8 weeks, as compared with a national average of 35.2 weeks for women carrying twins.1 Our results may not be applicable to women carrying twins as a result of intentional fetal reduction from a higher-order multiple gestation, since these women were excluded from the trial.
In previous trials demonstrating a benefit of 17P, the study participants were carrying singletons, and most were at risk for spontaneous preterm birth because of preterm birth in a previous pregnancy.11 One previous trial conducted in women with twin gestations failed to find a benefit of 17P; however, this trial was statistically underpowered, with only 77 participants, and 17P was initiated much later in gestation than in our study, at a mean of 29 weeks.12 In a trial reported elsewhere in this issue of the Journal,13 progesterone treatment reduced the rate of preterm birth among women who were at high risk for preterm birth because of a short cervix. In that trial, which involved mostly women with singleton gestations, both the formulation (micronized progesterone) and the route of administration (vaginal) were different from those in our trial. Concurrently with our twins trial, we enrolled triplets in a companion trial, the results of which are currently being analyzed.
Potential limitations of our study should be noted. Because less than 10% of women in our trial had previously given birth prematurely, it is uncertain whether 17P might be of benefit in twin gestations in which the mother has a history of spontaneous preterm birth. Our choice of using the same dose of 17P (250 mg per week) that was used in our previous trial2 of 17P in women with singleton gestations might be questioned, because plasma volume is known to be approximately 20% greater in twin than in singleton gestations.14 Thus, it is possible that a larger dose of 17P might have been efficacious. However, unless there is a threshold effect for 17P, which has not previously been suggested, a lower-than-optimal dosage would have been expected to have an attenuated effect on preterm birth, rather than no effect.
We based our primary outcome on delivery or fetal death before 35 completed weeks of gestation, rather than on rates of adverse fetal or neonatal outcomes. Short- and long-term complications of preterm birth are a direct function of gestational age, and delivery before 35 weeks is a more stringent cutoff point for preterm birth than is the standard definition of delivery before 37 weeks. Moreover, in well-dated pregnancies (as in this trial), the assessment of gestational age at delivery is highly reliable and objective.
In summary, the results of our trial do not support the use of 17P to reduce the risk of preterm birth in twin gestations. Why 17P has been effective in women with singleton gestations and a history of spontaneous preterm birth2 but was not effective in the present trial in women carrying twins is a question that will be answered only when the mechanisms underlying preterm birth and the actions of 17P are better understood. Further investigation is warranted to assess whether 17P is effective in other conditions in which the risk of preterm birth is increased.
Supported by grants (HD27869, HD21410, HD40512, HD34136, HD34208, HD40485, HD27915, HD40544, HD40560, HD27917, HD40500, HD34116, HD40545, HD27860, and HD36801) from the National Institute of Child Health and Human Development.
No potential conflict of interest relevant to this article was reported.
We thank Valerija Momirova for protocol and data management and statistical analysis; Allison Northen and Margaret Cotroneo for protocol development and coordination between clinical research centers; Jeffrey S.A. Stringer, M.D., of the University of Alabama at Birmingham for providing thoughtful manuscript review; and Joyce A. Martin, M.P.H., of the National Center for Health Statistics for graciously supplying data on U.S. natality and infant mortality.
Source Information
From the Departments of Obstetrics and Gynecology, Center for Women's Reproductive Health, University of Alabama at Birmingham, Birmingham (D.J.R.); the University of Pittsburgh, Pittsburgh (S.N.C.); Northwestern University, Chicago (A.M.P.); Drexel University, Philadelphia (A.S.); George Washington University Biostatistics Center, Washington, DC (E.A.T.); the National Institute of Child Health and Human Development, Bethesda, MD (C.Y.S.); the University of Utah, Salt Lake City (M.V.); Columbia University, New York (F.M.); Ohio State University, Columbus (J.D.I.); Case Western Reserve University, Cleveland (B.M.M.); the University of North Carolina, Chapel Hill (J.T.); Wayne State University, Detroit (Y.S.); Brown University, Providence, RI (M.C.); the University of Texas Southwestern Medical Center, Dallas (J.L.); the University of Texas at Houston, Houston (S.R.); Wake Forest University, Winston-Salem, NC (M.H.); and the University of Texas Medical Branch, Galveston (G.A.).
Address reprint requests to Dr. Rouse at the Department of Obstetrics and Gynecology, University of Alabama at Birmingham, 619 19th St. S., OHB 457, Birmingham, AL 35249-7333, or at drouse{at}uab.edu.
References
In addition to the authors, the members of the National Institute of Child Health and Human Development Maternal–Fetal Medicine Units Network are as follows: University of Alabama at Birmingham — W.W. Andrews, A. Northen, J. Sheppard; Brown University — J. Tillinghast, D. Allard; Case Western Reserve University — C. Milluzzi, C. Heggie, H. Ehrenberg, B. Stetzer, A. Merlino; Columbia University — R. Berkowitz, S. Bousleiman, S. South, L. Paley, V. Carmona, R. Wapner; Drexel University — M. Hoffman, M. Talucci, S. Wilson, C. Tocci, M. Lake; University of North Carolina — K. Boggess, K. Dorman, S. Timlin; Northwestern University — W. Grobman, G. Mallett, M. Dinsmoor, P. Simon, M. Huntley, M. Ramos; Ohio State University — F. Johnson, M. Landon, C. Latimer; University of Pittsburgh — H. Simhan, M. Cotroneo, E. Daugherty; University of Texas at Houston — D. Soebbing-Cross, J. Martinez, B. Glenn-Cole, L. Gilstrap; University of Texas Southwestern Medical Center — K. Leveno, L. Moseley; University of Utah — K. Anderson, F. Porter, C. Jolley, S. Quinn, A. Guzman; Wake Forest University — P. Meis, M. Swain, K. Johnson, K. Lanier, C. Leftwich; Wayne State University — G. Norman, C. Sudz, S. Blackwell; George Washington University Biostatistics Center — V. Momirova, A. Braga, E. Cardenas, L. Leuchtenburg; National Institute of Child Health and Human Development — S. Pagliaro.
| |||||||||||||||||||||||||||||||||||||||||||||||||||||
Related Letters:
Progesterone and Preterm Birth
O'Brien J. M., Chandiramani M., Tribe R., Shennan A., Nicolaides K. H., Celik E., Fonseca E. B., Rouse D. J., Thom E. A., Spong C. Y., the NICHD Maternal Fetal Medicine Units Network
Extract |
Full Text |
PDF
N Engl J Med 2007;
357:2306-2307, Nov 29, 2007.
Correspondence
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2008 Massachusetts Medical Society. All rights reserved. |
Previous
