Objective
The objective of the study was to estimate the predictive value of cervical length (CL) and fetal fibronectin (fFN) in patients being treated with 17-alpha-hydroxyprogesterone caproate (17P).
Methods
This was a retrospective cohort of 176 patients with a prior spontaneous preterm birth being treated with weekly injections of 17P who underwent serial CL and fFN screening.
Results
A short CL (≤25 mm) was significantly associated with an earlier gestational age at delivery and with recurrent preterm birth at less than 37, less than 35, less than 34, and less than 32 weeks. A positive fFN was not significantly associated with recurrent preterm birth. As a screening test for recurrent preterm birth, the positive and negative likelihood ratios for CL were 2.04 and 0.35, respectively, whereas for fFN they were 1.22 and 0.98, respectively, indicating that fFN did not offer any additional predictive value.
Conclusion
In patients being treated with 17P, cervical length at 22-32 weeks is predictive of recurrent preterm birth, but fetal fibronectin is not.
Patients with a history of preterm birth have an increased risk of preterm birth in subsequent pregnancies. In the Preterm Prediction Study, Iams et el reported that in patients with a prior spontaneous preterm birth, the recurrence risk of preterm birth varies widely according to both fetal fibronectin (fFN) and cervical length (CL) at 22-24 weeks. Although the American College of Obstetricians and Gynecologists (ACOG) does not endorse routine CL or fFN screening in low-risk singleton pregnancies because of the low positive predictive value in these patients at low risk for preterm birth, based on the data from the Preterm Prediction Study, ACOG considers screening high-risk women, such as those with a history of preterm birth, with CL or fFN to be a reasonable strategy to predict recurrent preterm birth.
Subsequent to the Preterm Prediction Study, one large randomized trial reported that the administration of weekly injections of 17-alpha-hydroxyprogesterone caproate (17P) beginning at 16-20 weeks significantly reduced the risk of recurrent preterm birth in patients with a prior spontaneous preterm birth of a singleton pregnancy. Based on these data, ACOG supports the use of progesterone supplementation for the prevention of recurrent preterm birth in women with a singleton pregnancy and a prior spontaneous preterm birth. Because the indications for 17P administration and CL/fFN testing are identical (prior spontaneous preterm birth of a singleton pregnancy), most patients who currently undergo screening for recurrent preterm birth with CL and fFN will also be taking 17P injections.
Considering that the original data supporting the association between CL and fFN and recurrent preterm birth preceded the widespread use of 17P, it is unknown whether progesterone use modifies the predictive value of CL and fFN in patients with a prior preterm birth. It is possible that for patients receiving 17P, the predictive value of either CL or fFN, or both, in predicting recurrent preterm birth could be significantly enhanced or reduced.
The objective of this study was to estimate the predictive value of CL and fFN in patients with a prior spontaneous preterm birth of a singleton pregnancy who are being treated with 17P.
Materials and Methods
After Biomedical Research Alliance of New York Institutional Review Board approval was obtained, we searched the patient database of one maternal-fetal medicine practice, which includes all patients delivered from July 2005 (when the database was created) to July 2011. We searched the database for all patients we delivered who had a history of a preterm birth in a prior pregnancy. All patients in the database have a listed parity (full-term births, preterm birth >20 weeks, pregnancy losses at <20 weeks, live births), and the search was done by identifying all patients with a number greater than zero for prior preterm birth.
The complete obstetrical histories for all of these patients were reviewed to select patients with a history of at least 1 prior spontaneous preterm birth (20 0/7 to 36 6/7 weeks) of a singleton pregnancy at any time in their past who were then managed by us in a subsequent singleton pregnancy. We then reviewed the complete charts for these patients.
In our practice, for patients with a prior spontaneous preterm birth of a singleton pregnancy, we recommend weekly 17P in a protocol similar to the study by Meis et al. We also perform serial (every 2-3 weeks) sonographic CL measurements from 16 to 32 weeks. Beginning at 22 weeks, we perform serial fFN testing just prior to the CL measurement. Certain patients with a prior spontaneous preterm birth of a singleton pregnancy do not receive 17P or CL/fFN screening, including patients who decline, some patients with mixed obstetrical histories including some term and some preterm births, and certain patients who have a history of a late preterm birth (34 to 36 6/7 weeks) who, after counseling, decide against 17P and CL/fFN screening.
For this study, we excluded all the patients who did not have a combined CL and fFN screening from 22 to 32 weeks and all the patients who were not receiving weekly 17P. We also excluded patients who underwent cerclage placement at any time in pregnancy.
All CL measurements and fFN testing were done in an outpatient setting on asymptomatic patients. All tests done on labor and delivery were excluded because they were done on symptomatic patients as part of a preterm labor evaluation. Patients and obstetricians were not blinded to the CL measurements or fFN results. Gestational age was determined by the last menstrual period and confirmed by ultrasound in all patients. The pregnancy was redated if there was a more than 5 day discrepancy up to 14 weeks or a greater than 7 day discrepancy after 14 weeks. If the pregnancy was the result of in vitro fertilization (IVF), gestational age was determined from IVF dating.
All CL measurements were measured by 4-8 MHz transvaginal probes (LOGIQ a200 and Voluson 530 and 730 Expert; GE Healthcare, Milwaukee, WI) with an empty bladder with the optimal image defined according to the criteria reported by Iams et al. The shortest functional CL was used as this has been found to be the most reproducible measurement. A short CL was defined as a CL of 25 mm or less.
Fetal fibronectin testing was performed using a Dacron swab without the use of a speculum according to an established protocol that has been validated previously by both our group and others. Fetal fibronectin testing was performed more than 24 hours from the last reported intercourse or endovaginal ultrasound and was not performed in the setting of vaginal bleeding. Vaginal swabs were sent for quantitative determination of fetal fibronectin concentration using the rapid TLi system qualitative method (fetal fibronectin immunoassay; Hologic Inc, Bedford, MA). A fetal fibronectin concentration of 50 ng/mL or greater was considered to be positive.
In our practice, we do not hospitalize patients with a short CL or positive fFN aside from those actually in preterm labor. We typically administer corticosteroids to asymptomatic patients with both a short CL and positive fFN. Patients in preterm labor are given a course of corticosteroids and tocolytics, but they are typically not advised bed rest.
Because we perform CL and fFN testing every 2-3 weeks, most of our patients have multiple test results. For the purpose of this study, a patient with a CL of 25 mm or less at any time between 22 0/7 and 31 6/7 weeks was considered as having a short CL, and a patient with a positive fFN at any time between 22 0/7 and 31 6/7 weeks was considered as having a positive fFN. We compared patients with and without a short CL, with and without a positive fFN, and the combination of these tests (neither positive, either positive, both positive).
Our primary outcome was recurrent preterm birth less than 34 weeks. We also looked at recurrent preterm births of less than 32 weeks, less than 35 weeks, and less than 37 weeks, and a gestational age at delivery. To have 80% power to find a 4-fold increase in the rate of preterm birth in patients with a positive screening test (as seem by Iams et al ) with an alpha error of 5%, we needed approximately 175 patients.
A Fisher exact test, a χ 2 test, and a Student’s t test were used when appropriate (SPSS for Windows 16.0, 2007; SPSS, Inc., Chicago, IL). A regression analysis was performed to control for significant baseline characteristics. A P value of .05 or less was considered significant.
Results
Our initial search revealed 778 patients over the study period with a history of any preterm birth. After reviewing the complete obstetrical history for these patients and excluding 415 patients with multiple pregnancies, a history of indicated preterm births, and a history of preterm births of a multiple pregnancy, there were 363 patients with a singleton pregnancy with a history of a spontaneous preterm birth of a singleton pregnancy.
Of these patients, 195 patients received weekly 17P. Reasons for not receiving 17P were mixed obstetrical histories (term and preterm births), late preterm births, patient refusal, or a combination of the above. A further 10 patients with a cerclage were excluded, as were 9 patients who did not undergo combined fFN and CL screening from 22 to 32 weeks. This left 176 patients for analysis.
Descriptive characteristics for these patients are shown in Table 1 . Overall, this population was at high risk for preterm birth. A total of 63.6% had a history of 1 prior spontaneous preterm birth, and 36.4% had more than 1 prior spontaneous preterm birth. The mean gestational age at delivery was 37.5 ± 2.4 weeks. The recurrence risk for preterm birth at less than 37 weeks, less than 35 weeks, less than 34 weeks, and less than 32 weeks was 31.8%, 7.4%, 5.1%, and 4.0%, respectively.
| Characteristics | Values |
|---|---|
| Number of patients | 176 |
| Number of prior spontaneous singleton preterm births <37 wks, n (%) | |
| 1 | 112 (63.6) |
| 2 | 37 (21.0) |
| 3 | 18 (10.2) |
| >3 | 9 (5.1) |
| Gestational age of earliest prior preterm birth, wks | 29.5 ± 5.2 |
| Gestational age of most recent birth, wks | 32.8 ± 5.8 |
| Most recent birth was preterm | 132 (75%) |
| Any prior term births | 91 (51.7%) |
| Prior LEEP or cervical conization | 5 (2.8%) |
| Age, y | 31.8 ± 5.8 |
| IVF | 11 (6.2%) |
| Prepregnancy body mass index, kg/m 2 | 23.5 ± 8.8 |
| White race | 169 (96%) |
| Positive fFN | 16 (9.1%) |
| Gestational age at positive fFN | 28.2 ± 2.0 |
| Shortened CL, ≤25 mm | 67 (38.1%) |
| Gestational age at diagnosis of shortened CL, wks | 27.0 ± 3.2 |
| Mean CL at diagnosis of shortened CL, mm | 21.0 ± 4.8 |
| Both positive fFN and shortened CL, ≤25 mm | 10 (5.7%) |
| Gestational age at delivery, wks | 37.5 ± 2.4 |
| Preterm birth <37 wks | 56 (31.8%) |
| Preterm birth <35 wks | 13 (7.4%) |
| Preterm birth <34 wks | 9 (5.1%) |
| Preterm birth <32 wks | 7 (4.0%) |
Sixty-seven (38.1%) patients had a short CL of 25 mm or less from 22 0/7 to 31 6/7 weeks. A short CL was significantly associated with an earlier gestational age at delivery and recurrent preterm birth at less than 37 weeks, less than 35 weeks, less than 34 weeks, and less than 32 weeks ( Table 2 ).
| Variable | CL >25 mm | CL ≤25 mm | P value |
|---|---|---|---|
| (n = 109) | (n = 67) | ||
| Mean GA at delivery, wks | 37.94 ± 1.87 | 36.84 ± 2.94 | .003 |
| <32 | 1 (0.9%) | 6 (9.0%) | .008 |
| <34 | 2 (1.8%) | 7 (10.4%) | .012 |
| <35 | 3 (2.8%) | 10 (14.9%) | .003 |
| <37 | 28 (25.7%) | 28 (41.8%) | .026 |
Sixteen patients (9.1%) had a positive fFN from 22 0/7 to 31 6/7 weeks. A positive fFN was not associated with gestational age at delivery or recurrent preterm birth at any gestational age ( Table 3 ). The change in CL from the first to last measurement in patients who delivered at 37 weeks or longer, 34 to 36 6/7 weeks, and less than 34 weeks is shown in the Figure .
