Objective
The purpose of the study was to evaluate whether 17-alpha-hydroxyprogesterone caproate (17-OHPC) exposure is associated with the rate of cervical shortening.
Study Design
Women with a history of spontaneous preterm delivery (PTD) at <37 weeks’ gestation who had serial cervical length measurements (2009-2012) were identified. 17-OHPC administration and outcome data were collected. We excluded patients with multiple gestations, indicated PTDs, major fetal anomalies, cerclage, and vaginal progesterone use. The rate of cervical shortening was modeled in relation to 17-OHPC status with the use of methods for longitudinal data analysis.
Results
Two hundred thirty-seven patients with 1171 cervical length measurements were included, of whom 184 patients (77.6%) were exposed to 17-OHPC. Gestational age, number of previous PTDs, gestational age at initiation, and interval between cervical length examinations were similar between the 2 groups, although women who were not exposed to 17-OHPC were more likely to have delivered multiples in their previous PTD (24.5% vs 4.4%; P < .01). In the entire cohort, the rate of cervical shortening was identical, regardless of 17-OHPC exposure (0.85 mm per week). Among term deliveries, the rates of cervical shortening per week, on average, were 0.9 and 0.8 mm per week among women with and without 17-OHPC, respectively ( P = .76). Among preterm deliveries, the corresponding rates were 0.8 and 1.2 mm, respectively, among women with and without 17-OHPC ( P = .67).
Conclusion
Cervical shortening among women with previous preterm delivery occurs at a similar rate, regardless of exposure to 17-OHPC.
Cervical length measurement in pregnancy has proved to be a useful tool in the identification of women who are at risk of spontaneous preterm delivery. Serial assessments of the cervix and weekly intramuscular injections of 17-alpha-hydroxyprogesterone caproate (17-OHPC) are 2 strategies that are aimed at reducing the rates of recurrent preterm delivery. It is well established that cervical length decreases with advancing gestational age and that the cervix begins to shorten physiologically after 28 weeks’ gestation, even in women who are destined to deliver at term. However, several studies have shown that the rate at which cervical shortening occurs may identify those women who are at increased risk of spontaneous preterm delivery.
Although the mechanism by which 17-OHPC reduces the risk of recurrent preterm delivery is not well established, pathways that are involved in myometrial quiescence, the modulation of the immune response, and alterations in cervical remodeling and ripening have been proposed. If progesterone inhibits the cervical ripening process, then it might be expected that cervical shortening would occur more gradually among women who are exposed to 17-OHPC compared with women with similar histories who are not exposed to 17-OHPC.
A previous study reported no difference in the rate of cervical shortening among women with a history of spontaneous preterm delivery according to 17-OHPC exposure. However, whether changes in cervical length in relation to 17-OHPC exposure differed between those who delivered preterm vs term remains unknown. Previous work has suggested inherent differences between women who experience isolated vs recurrent spontaneous preterm deliveries.
Because aberrations in cervical remodeling may predispose women to recurrent preterm deliveries and progesterone may function by targeting these aberrations, we hypothesized that women who experience recurrent preterm delivery and were exposed to progesterone may exhibit a slower rate of cervical shortening when compared with women who were not exposed to 17-OHPC. Thus, the objective of this study was to evaluate the association between 17-OHPC exposure and the rate of cervical shortening, specifically among women who have a history of spontaneous preterm delivery and go on to deliver preterm again in a subsequent pregnancy.
Materials and Methods
With Institutional Review Board approval, we performed a retrospective study among women with a history of spontaneous preterm delivery (defined as ≥1 spontaneous births between 16 weeks and 36 weeks 6 days’ gestation) who underwent serial cervical length assessments to monitor for cervical shortening in our institution between 2009 and 2012. Women with a singleton pregnancy who underwent at least 2 cervical length measurements during their pregnancy were included. The exposure of interest was treatment with 17-OHPC, and the primary outcome was the rate of cervical shortening. Pregnancies with major fetal anomalies, vaginal progesterone use at any time in the index pregnancy, <2 cervical length measurements, medically indicated preterm delivery, or the presence or placement of an abdominal or vaginal cerclage were excluded. The latter exclusion applied to both history and ultrasound-indicated vaginal cerclage, because the presence of either potentially could interfere with cervical remodeling and may confound the association between cervical shortening and progesterone exposure.
Electronic medical records were reviewed to abstract pertinent obstetric history, including the number of previous preterm deliveries, gestational age of each preterm delivery, and whether that pregnancy involved multiple gestations. Variables that were collected included date and gestational age at each cervical length measurement along with the measurement itself, maternal age, race/ethnicity, prepregnancy and weight at delivery, type of provider (private maternal-fetal medicine specialist, private generalist obstetrician and gynecologist (OB/GYN), or government-insured OB/GYN low-risk clinic or high-risk maternal-fetal medicine clinic), use of in vitro fertilization, smoking status, and illicit drug use during pregnancy. Previous surgical manipulations of the cervix were recorded, including loop electrosurgical excision procedure, cold knife cone biopsy, dilation, curettage, and evacuation. Details regarding timing and indications for delivery in the current pregnancy, when applicable, were also collected.
Our cohort was comprised of both private and publically insured patients. At our center, women with a history of spontaneous delivery at <37 weeks’ gestation are offered weekly treatment with 17-OHPC starting at 16 gestational weeks. The women included in our study who were not exposed to 17-OHPC may have declined the medication or did not receive therapy because of noncompliance. Review of the medical record was used to abstract details regarding noncompliance. Once 17-OHPC therapy was initiated, injections were given on a weekly basis until 36 weeks’ gestation or delivery.
Review of our electronic ultrasound database was used to record cervical length measurements (in millimeters) from 16-32 weeks’ gestation. During the study period, a variety of General Electric (Fairfield, CT), Philips (Andover, MA), Medison (Seoul, South Korea), and Acuson (Mountain View, CA) ultrasound machines that had been approved for clinical use through the institution and department were used with the appropriate transvaginal probes (5-9 MHz). Based on our institution’s protocol, cervical length measurements take place every 2 weeks, unless indicated more frequently by symptoms or significant sonographic findings. Each transvaginal cervical length assessment was obtained with the use of the technique published by Iams et al in 1996. Sonographer training at our site had been performed in conjunction with a Maternal-Fetal Medicine Units (MFMU) Network study by Grobman et al and involved a series of didactics on the methods of Iams et al followed by the submission of 3 images from each of 5 separate patients. Each sonographer specified the best image for each patient, the shortest cervical length, and whether a funnel or debris was present. Central review was performed by an expert in the field, and study personnel were certified only when at least 4 of 5 images were deemed acceptable. Once this MFMU study was completed, the same certification process continued for incoming sonographers in preparation for an ongoing National Institute of Child Health and Human Development study (Nulliparous Pregnancy Outcomes Study; clinicaltrials.gov ; ID NCT01322529 ). The obstetrics ultrasound guidelines at Columbia University Medical Center require 3 measurements of the cervix that include at least 1 assessment while the patient performs the Valsalva maneuver. The shortest of the 3 cervical length values is reported clinically; this measurement was recorded into our database for each visit.
Short cervix is defined at our institution as <25 mm, which represents the 10th percentile at 24 weeks’ gestation. Once this is identified in our ultrasound unit, patient treatment varies according to practitioner and individual patient details and may involve cerclage placement, the initiation of vaginal progesterone, or expectant treatment and observation either as an inpatient or outpatient. Previous work has confirmed the predictive value of cervical length at <29 weeks’ gestation, although our institution’s protocol includes continued assessments until 32 weeks’ gestation. Although the risk of preterm delivery with a short cervix at this advanced gestational age has not been confirmed, the goal of our protocol was to identify all women who may benefit from time-sensitive interventions such as antenatal steroid and magnesium administration.
We examined the distribution of maternal sociodemographic characteristics and obstetric history in relation to 17-OHPC status. Covariates that were categoric were compared based on exposure to 17-OHPC with the use of either the χ 2 test or the Fisher exact probability test; for continuous covariates, we applied the t test for normally distributed data and the 2-sample median test for nonnormal data.
The shortest cervical length was plotted against gestational age for every patient starting at 16 weeks, and the trajectory of women who were treated with 17-OHPC was compared with women who did not receive this medication. Data for these women were then analyzed separately according to whether the current pregnancy delivered at term or preterm. In an analysis that was stratified based on whether women delivered at term or preterm gestations, the rate of cervical change across gestational age initially was modeled with restricted cubic spline transformations (for gestational age at the time of cervical length assessment); because these data exhibited a linear change in cervical lengths across gestational age, we modeled gestational age with a single linear term. All the aforementioned analyses were adjusted for the confounding effects of maternal age, history of recurrent preterm deliveries, smoking and illicit drug use, and race/ethnicity. Confounders were chosen by a comparison of odds ratios from the unadjusted and adjusted analyses, such that the confounder was retained in the models for adjustment if the odds ratio differed by at least 10%. Because women contributed multiple cervical length assessments during pregnancy (thereby violating the statistical assumption of independence of observations ), we estimated the parameters on the basis of the methods of generalized estimating equations in these linear regression models.
All statistical analysis was performed using SAS software (version 9.3; SAS Institute, Cary, NC).
Results
Of 17,400 deliveries from 2009-2012, 376 women with a history of spontaneous preterm delivery met inclusion criteria. We excluded 139 women: major fetal anomalies, 5 women; the presence of placement of a cerclage (2 abdominal and 78 vaginal), 80 women; exposure to vaginal progesterone, 32 women; indicated preterm delivery, 22 women. This resulted in 237 women for analysis. Of the included patients, 184 (77.6%) were exposed to 17-OHPC in the current pregnancy.
The distribution of maternal characteristics at baseline in relation to 17-OHPC status is described in Table 1 . Women who were exposed to 17-OHPC were, on average, 2.2 years younger than those who were not exposed to 17-OHPC. None of the other characteristics, including previous uterine or cervical surgery, differed between the groups. No subject in our cohort had a history of a cold knife cone biopsy.
| Baseline characteristic | 17-alpha-hydroxyprogesterone caproate | P value | |
|---|---|---|---|
| Exposed (n = 184) | Not exposed (n = 53) | ||
| Maternal age at delivery, y a | 30.4 ± 6.0 | 32.6 ± 5.6 | .02 |
| Weight at delivery, kg a | 79.6 ± 19.5 | 81.6 ± 16.6 | .59 |
| Weight gain, kg a | 9.8 ± 7.2 | 8.5 ± 11.0 | .54 |
| In vitro fertilization pregnancy, n (%) | 6 (3.3) | 5 (9.4) | .07 |
| Race/ethnicity, n (%) | .67 | ||
| White | 29 (17.6) | 8 (17.4) | |
| African American | 15 (9.1) | 6 (13) | |
| Hispanic | 37 (22.4) | 7 (15.2) | |
| Other | 84 (50.9) | 25 (54.35) | |
| Smoking, n (%) | 3 (1.6) | 0 | .99 |
| Illicit drug use, n (%) | 1 (0.5) | 2 (3.7) | .13 |
| Previous gynecologic surgeries, n (%) | |||
| Loop electrosurgical excision procedure | 6 (3.3) | 2 (3.8) | .99 |
| Curettage | 41 (22.3) | 9 (17) | .67 |
| Evacuation | 12 (6.5) | 4 (7.6) | .73 |
| Prenatal care, n (%) | |||
| Private general obstetrician/gynecologist | 10 (5.4) | 8 (15.1) | .03 |
| Private maternal-fetal medicine specialist | 44 (23.9) | 15 (28.3) | .59 |
| Government-insured clinic | 123 (66.9) | 27 (50.9) | .04 |
| Government-insured clinic with maternal-fetal medicine consultation | 125 (67.9) | 19 (35.9) | < .01 |
We assessed history-based obstetric risk in relation to 17-OHPC exposure ( Table 2 ). Subjects who were not exposed to 17-OHPC were more likely to have a previous preterm delivery that involved a multiple gestation. One patient in our study population had previously delivered both preterm singleton and preterm multiple gestations. Otherwise, obstetric history of recurrent preterm delivery, gestational age at earliest preterm delivery, and history of at least 1 term delivery was similar between those who were exposed to 17-OHPC and those not.
| Historic variable | 17-alpha-hydroxyprogesterone caproate | P value | |
|---|---|---|---|
| Exposed | Not exposed | ||
| History of ≥1 term delivery, n (%) | 110 (59.8) | 25 (47.2) | .12 |
| Previous preterm deliveries, n (%) | .76 | ||
| 1 | 89 (48.4) | 24 (45.3) | |
| ≥2 | 95 (51.6) | 29 (54.7) | |
| Gestational age of earliest birth, wk a | 28.6 ± 5.5 | 29.2 ± 6.4 | .49 |
| Previous preterm deliveries, n (%) | < .01 | ||
| Singleton gestation | 175 (96.2) | 42 (79.3) | |
| Multiple gestation | 8 (4.4) | 13 (24.5) | |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
