Materials and Methods
This is a secondary analysis of data from the Preterm Prediction Study, a multicenter prospective observational cohort study conducted by the National Institute of Child Health and Human Development Maternal-Fetal Medicine Units Network between 1992 and 1994 to study the predictors of preterm birth.
Women with a singleton gestation from the general obstetric population with well-dated pregnancies less than 24 weeks’ gestation were included. Women with multiple gestation, placenta previa, cerclage , and fetal anomalies were excluded. Women who had sonographic CL ultrasounds performed at a gestation at 24 weeks (range, 21–28 weeks) and at 28 weeks (range, 25–33 weeks) and had maternal blood samples collected at these visits were included in the secondary analysis.
The technique for sonographic measurement of CL has been described in detail in prior reports of this cohort. Sonographic CL measurements were subjected to a quality assurance protocol. Each ultrasound visit included 3 transvaginal CL measurements, and the shortest measurement was recorded. Fundal pressure was not used to assess the shortest CL. When a cervical funnel was present, the CL below the funnel was recorded. Sonogram results were not reported to managing physicians unless fetal death, advanced cervical dilation, prolapsed membranes, oligo- or polyhydramnios, or regular uterine contractions were detected.
As part of the original study protocol for the Preterm Prediction Study, maternal serum samples for CRH, CRP, and thrombin-antithrombin III (TAT) were collected at the first study visit at 24 weeks and at a follow-up visit at 28 weeks. Serum samples were stored at –70°C until analysis. All serum assays were performed after 1 freeze-thaw cycle.
CRH concentrations were determined using an enzyme-linked immunosorbent assay from Peninsula Laboratories (San Carlos, CA). Free CRH, the biologically active form of CRH, was measured using a Sep Pak C18 column (Waters, Milford, MA). CRH levels reported in this secondary analysis refer to free CRH. CRP levels were obtained using an enzyme-linked immunosorbent assay from Kamiya Biomedical Co (Seattle, WA). An immunoassay for TAT was performed using a kit from American Diagnostica (Greenwich, CT), and absorbances were measured using a microtiter plate reader and software from Molecular Devices (Menlo Park, CA). Assays had been performed prior to this dataset, having been made available for the present secondary analysis.
The differences between maternal serum concentrations of CRH, CRP, and TAT at 24 weeks and 28 weeks were our primary outcome variables of interest. The bivariate relationships between the change in CL and plasma levels of these proteins were determined using simple linear regression. Covariate analyses used multiple linear regression and included age, race, unmarried status, tobacco use, and a history of prior preterm birth. A value of P < .05 was considered statistically significant. Statistical analyses were performed using Stata 10 (StataCorp, College Station, TX).
Results
Three hundred thirty-four women completed both study visits. Demographic information is listed in the Table . One hundred nine subjects (33%) were nulliparous and 225 (67%) were multiparous. Eighty-four women (26%) had a history of prior SPTB. Visit 1 occurred at approximately 24 weeks’ gestation (range, 22.4–24.9 weeks). Fifty-two women (16%) had sonographic CL less than 25 mm at visit 1. The median time between visits was 28 days (range, 18–55 days).
| Demographic | Number of subjects | % |
|---|---|---|
| Total | 334 | 100 |
| Age, y | ||
| Median (range) | 26 (18–44) | |
| BMI, kg/m 2 | ||
| Median (range) | 23.2 (15.4–47.5) | |
| Race | ||
| White | 242 | 73.8 |
| African American | 84 | 25.6 |
| Other | 2 | 0.6 |
| Parity | ||
| Nulliparous | 109 | 33.2 |
| Multiparous | 225 | 66.8 |
| Marital status | ||
| Married | 161 | 49.1 |
| Unmarried | 173 | 50.9 |
| Tobacco | ||
| Yes | 90 | 27.4 |
| No | 244 | 72.6 |
| Prior PTB | ||
| Yes | 84 | 25.6 |
| No | 250 | 74.4 |
| Gestational age, wks | ||
| Visit 1 median (range) | 23.9 (22.4–24.9) | |
| Visit 2 median (range) | 28 (26.2–31.7) | |
| Time between visits, wks | ||
| Median (range) | 4 (2.6–7.9) | |
| CL, visit 1, mm | ||
| <25 | 52 | 16 |
| ≥25 | 282 | 84 |
Among women with a sonographic short CL at visit 1, there was a significant association between cervical shortening and a rise of CRH (r 2 = 0.34, P = .014) ( Figure 1 ), and a rise of CRP (r 2 = 0.44, P = .001) ( Figure 2 ). For every 10 mm decrease in CL between visits, the concentration of maternal serum CRH increased 0.0523 ng/mL, and the concentration of maternal serum CRP increased 0.62 mg/dL. Therefore, the slope of the linear relationship between CRH and CL was greater than that of CRP and CL.
For the cohort overall, change in CL between visits was not associated with concentrations of CRP or CRH. Change in CL was not associated with maternal serum TAT concentration, either for women with a CL less than 25 mm (r 2 = 0.005) or for the cohort overall. Adjustment for age, race, unmarried status, tobacco use, and a history of prior preterm birth did not alter the magnitude or significance of these associations.
Results
Three hundred thirty-four women completed both study visits. Demographic information is listed in the Table . One hundred nine subjects (33%) were nulliparous and 225 (67%) were multiparous. Eighty-four women (26%) had a history of prior SPTB. Visit 1 occurred at approximately 24 weeks’ gestation (range, 22.4–24.9 weeks). Fifty-two women (16%) had sonographic CL less than 25 mm at visit 1. The median time between visits was 28 days (range, 18–55 days).
