Introduction
Preterm birth affected 11.4% of all births in the United States in 2013; it is the leading cause of neonatal and child mortality worldwide. Accumulating evidence suggests that preterm birth is a syndrome caused by multiple pathologic processes. A short cervix, traditionally defined as a transvaginal sonographic cervical length (CL) ≤25 mm in the midtrimester of pregnancy, is a risk factor for preterm delivery. Untimely cervical ripening is one of the mechanisms of disease implicated in the etiology of the preterm parturition syndrome.
A prospective multicenter cohort study reported that women with a singleton pregnancy and a CL ≤25 mm at 22-24 weeks of gestation were more than 6 times more likely than women with a CL >40 mm to deliver spontaneously before 35 weeks of gestation. Today, a short cervix is a well-accepted risk factor for preterm birth and one of the strongest predictors of preterm birth in both singleton and twin gestations. In the absence of any intervention, an estimated 17% of infants born to mothers with a CL ≤25 mm before 25 weeks of gestation will die during the neonatal period or develop major neonatal complications such as respiratory distress syndrome, intraventricular hemorrhage, necrotizing enterocolitis, or neonatal sepsis.
Efficacy of vaginal progesterone in pregnant women with a short cervix
Progesterone, a key hormone for pregnancy maintenance, plays a role in the control of cervical ripening through several mechanisms, including the regulation of extracellular matrix metabolism and antiinflammatory effects. Progesterone also has other effects on the common pathway of parturition, which may contribute to the prevention of preterm birth in women with a short cervix.
The most powerful evidence of the efficacy of vaginal progesterone in the prevention of preterm birth derives from randomized clinical trials and individual patient data metaanalyses. Fonseca et al randomly assigned 250 women with a CL ≤15 mm at 20-25 weeks of gestation to receive vaginal progesterone capsules (200 mg/d) or a placebo at 24-34 weeks of gestation, and reported a significant reduction of 44% of the risk for spontaneous preterm birth before 34 weeks of gestation [relative risk (RR), 0.56; 95% confidence interval (CI), 0.36–0.86] and a nonsignificant decrease of 41% of the risk for major neonatal morbidity (RR, 0.59; 95% CI, 0.26–1.25). Hassan et al reported a trial in which 458 women with a CL of 10-20 mm at 19-23 weeks of gestation were randomly selected to receive vaginal progesterone gel (90 mg/d) or a placebo at 20-23 to 36 weeks of gestation. The administration of vaginal progesterone was associated with a significant reduction of the risk for preterm birth before 33 weeks of gestation (RR, 0.55; 95% CI, 0.33–0.92), respiratory distress syndrome (RR, 0.39; 95% CI, 0.17–0.92), and composite neonatal morbidity and mortality (RR, 0.57; 95% CI, 0.33–0.99).
In 2012, an individual patient data metaanalysis assessed the efficacy of vaginal progesterone to prevent preterm birth in asymptomatic women with a transvaginal sonographic CL of ≤25 mm in the midtrimester. A total of 775 women (723 with singleton gestations and 52 with twin gestations) and 827 infants (723 from singleton gestations and 104 from twin gestations) from 5 randomized controlled trials were included in the study. In addition to the 2 trials specifically designed to evaluate vaginal progesterone in women with a sonographic short cervix, the individual patient data metaanalysis included information from 3 trials that assessed the use of vaginal progesterone in women at high risk for preterm birth, which provided data for women with a CL of ≤25 mm before randomization. Treatment with vaginal progesterone was associated with a statistically significant reduction of the risk for preterm birth from <28 to <35 weeks of gestation, respiratory distress syndrome, composite neonatal morbidity and mortality, birth weight <1500 g, admission to the neonatal intensive care unit, and requirement for mechanical ventilation. Importantly, the reduction of preterm birth before 33 weeks of gestation and of composite neonatal morbidity and mortality was demonstrated by women with no previous spontaneous preterm birth as well as by those with at least 1 previous spontaneous preterm birth.
Another important prespecified subgroup analysis showed that the beneficial effects of vaginal progesterone did not differ significantly among women as a function of CL, which stratified patients into 3 groups: <10 mm, 10-20 mm, or 21-25 mm. The P values for the interaction effect of progesterone based on CL were nonsignificant (0.32 for preterm birth before 33 weeks of gestation and 0.93 for composite neonatal morbidity and mortality). The interpretation of these nonsignificant interaction P values is that the beneficial effects of vaginal progesterone for patients with a CL between 21-25 mm and <10 mm are not significantly different from those for patients with a CL of 10-20 mm.
The beneficial effects of vaginal progesterone in women with a singleton gestation and a CL ≤25 mm are shown in Table 1 . Among women with a twin gestation and a CL ≤25 mm, the use of vaginal progesterone was associated with a significant 44% reduction of the risk for composite neonatal morbidity and mortality (adjusted RR, 0.56; 95% CI, 0.30–0.97) and a 30% nonsignificant reduction of the risk for preterm birth before 33 weeks of gestation (RR, 0.70; 95% CI, 0.34–1.44). Similar results were recently reported in an individual patient data metaanalysis by Schuit et al, which has been subject to editorial comment, and in a randomized clinical trial among women with a twin gestation and a CL between 20-25 mm at 20-24 weeks of gestation (N = 224).
| Outcome | RR (95% CI) | NNT (95% CI) | NNS (95% CI) a | |
|---|---|---|---|---|
| Preterm birth <35 weeks | 0.67 (0.51–0.87) | 11 (7–28) | 138 (88–350) | |
| Preterm birth <34 weeks | 0.60 (0.44–0.82) | 10 (7–23) | 125 (88–288) | |
| Preterm birth <33 weeks | 0.56 (0.40–0.80) | 11 (8–25) | 138 (100–313) | |
| Preterm birth <32 weeks | 0.56 (0.38–0.82) | 13 (9–32) | 163 (113–400) | |
| Preterm birth <30 weeks | 0.59 (0.37–0.92) | 19 (12–97) | 238 (150–1213) | |
| Preterm birth <28 weeks | 0.51 (0.31–0.85) | 19 (13–61) | 238 (163–763) | |
| Respiratory distress syndrome | 0.47 (0.27–0.81) | 18 (13–51) | 225 (163–638) | |
| Birth weight <1500 g | 0.52 (0.34–0.81) | 14 (10–35) | 175 (125–438) | |
| Admission to NICU | 0.67 (0.50–0.91) | 12 (8–46) | 150 (100–575) | |
| Major neonatal morbidity/neonatal mortality b | 0.59 (0.38–0.91) | 18 (12–81) | 225 (150–1013) | |
a Assuming a prevalence of 8% for cervical length ≤25 mm
b Occurrence of any of the following events: neonatal death, respiratory distress syndrome, intraventricular hemorrhage, necrotizing enterocolitis, or neonatal sepsis.
Efficacy of vaginal progesterone in pregnant women with a short cervix
Progesterone, a key hormone for pregnancy maintenance, plays a role in the control of cervical ripening through several mechanisms, including the regulation of extracellular matrix metabolism and antiinflammatory effects. Progesterone also has other effects on the common pathway of parturition, which may contribute to the prevention of preterm birth in women with a short cervix.
The most powerful evidence of the efficacy of vaginal progesterone in the prevention of preterm birth derives from randomized clinical trials and individual patient data metaanalyses. Fonseca et al randomly assigned 250 women with a CL ≤15 mm at 20-25 weeks of gestation to receive vaginal progesterone capsules (200 mg/d) or a placebo at 24-34 weeks of gestation, and reported a significant reduction of 44% of the risk for spontaneous preterm birth before 34 weeks of gestation [relative risk (RR), 0.56; 95% confidence interval (CI), 0.36–0.86] and a nonsignificant decrease of 41% of the risk for major neonatal morbidity (RR, 0.59; 95% CI, 0.26–1.25). Hassan et al reported a trial in which 458 women with a CL of 10-20 mm at 19-23 weeks of gestation were randomly selected to receive vaginal progesterone gel (90 mg/d) or a placebo at 20-23 to 36 weeks of gestation. The administration of vaginal progesterone was associated with a significant reduction of the risk for preterm birth before 33 weeks of gestation (RR, 0.55; 95% CI, 0.33–0.92), respiratory distress syndrome (RR, 0.39; 95% CI, 0.17–0.92), and composite neonatal morbidity and mortality (RR, 0.57; 95% CI, 0.33–0.99).
In 2012, an individual patient data metaanalysis assessed the efficacy of vaginal progesterone to prevent preterm birth in asymptomatic women with a transvaginal sonographic CL of ≤25 mm in the midtrimester. A total of 775 women (723 with singleton gestations and 52 with twin gestations) and 827 infants (723 from singleton gestations and 104 from twin gestations) from 5 randomized controlled trials were included in the study. In addition to the 2 trials specifically designed to evaluate vaginal progesterone in women with a sonographic short cervix, the individual patient data metaanalysis included information from 3 trials that assessed the use of vaginal progesterone in women at high risk for preterm birth, which provided data for women with a CL of ≤25 mm before randomization. Treatment with vaginal progesterone was associated with a statistically significant reduction of the risk for preterm birth from <28 to <35 weeks of gestation, respiratory distress syndrome, composite neonatal morbidity and mortality, birth weight <1500 g, admission to the neonatal intensive care unit, and requirement for mechanical ventilation. Importantly, the reduction of preterm birth before 33 weeks of gestation and of composite neonatal morbidity and mortality was demonstrated by women with no previous spontaneous preterm birth as well as by those with at least 1 previous spontaneous preterm birth.
Another important prespecified subgroup analysis showed that the beneficial effects of vaginal progesterone did not differ significantly among women as a function of CL, which stratified patients into 3 groups: <10 mm, 10-20 mm, or 21-25 mm. The P values for the interaction effect of progesterone based on CL were nonsignificant (0.32 for preterm birth before 33 weeks of gestation and 0.93 for composite neonatal morbidity and mortality). The interpretation of these nonsignificant interaction P values is that the beneficial effects of vaginal progesterone for patients with a CL between 21-25 mm and <10 mm are not significantly different from those for patients with a CL of 10-20 mm.
The beneficial effects of vaginal progesterone in women with a singleton gestation and a CL ≤25 mm are shown in Table 1 . Among women with a twin gestation and a CL ≤25 mm, the use of vaginal progesterone was associated with a significant 44% reduction of the risk for composite neonatal morbidity and mortality (adjusted RR, 0.56; 95% CI, 0.30–0.97) and a 30% nonsignificant reduction of the risk for preterm birth before 33 weeks of gestation (RR, 0.70; 95% CI, 0.34–1.44). Similar results were recently reported in an individual patient data metaanalysis by Schuit et al, which has been subject to editorial comment, and in a randomized clinical trial among women with a twin gestation and a CL between 20-25 mm at 20-24 weeks of gestation (N = 224).
| Outcome | RR (95% CI) | NNT (95% CI) | NNS (95% CI) a | |
|---|---|---|---|---|
| Preterm birth <35 weeks | 0.67 (0.51–0.87) | 11 (7–28) | 138 (88–350) | |
| Preterm birth <34 weeks | 0.60 (0.44–0.82) | 10 (7–23) | 125 (88–288) | |
| Preterm birth <33 weeks | 0.56 (0.40–0.80) | 11 (8–25) | 138 (100–313) | |
| Preterm birth <32 weeks | 0.56 (0.38–0.82) | 13 (9–32) | 163 (113–400) | |
| Preterm birth <30 weeks | 0.59 (0.37–0.92) | 19 (12–97) | 238 (150–1213) | |
| Preterm birth <28 weeks | 0.51 (0.31–0.85) | 19 (13–61) | 238 (163–763) | |
| Respiratory distress syndrome | 0.47 (0.27–0.81) | 18 (13–51) | 225 (163–638) | |
| Birth weight <1500 g | 0.52 (0.34–0.81) | 14 (10–35) | 175 (125–438) | |
| Admission to NICU | 0.67 (0.50–0.91) | 12 (8–46) | 150 (100–575) | |
| Major neonatal morbidity/neonatal mortality b | 0.59 (0.38–0.91) | 18 (12–81) | 225 (150–1013) | |
a Assuming a prevalence of 8% for cervical length ≤25 mm
b Occurrence of any of the following events: neonatal death, respiratory distress syndrome, intraventricular hemorrhage, necrotizing enterocolitis, or neonatal sepsis.
Public health implications
The number of patients needed to treat with vaginal progesterone to prevent 1 case of preterm birth before 33 weeks or adverse neonatal outcomes varies from 10-19 women ( Table 1 ). This compares favorably to the numbers needed to treat by other obstetrical and perinatal interventions such as magnesium sulfate to prevent eclampsia or cerebral palsy, aspirin to prevent preeclampsia, active management of the third stage of labor to prevent postpartum hemorrhage, and antenatal corticosteroids to prevent respiratory distress syndrome and neonatal death ( Table 2 ).
| Intervention | Target population | Adverse outcome to prevent | RR (95% CI) | NNT (95% CI) |
|---|---|---|---|---|
| Magnesium sulfate | Women with preeclampsia | Eclampsia | 0.41 (0.29–0.58) | 91 (75–127) |
| Women at risk of preterm birth before 34 weeks of gestation | Cerebral palsy | 0.69 (0.55–0.88) | 52 (31–154) | |
| Aspirin | All women | Preeclampsia | 0.90 (0.84–0.97) | 167 (104–556) |
| Active management of the third stage of labor | All women | Postpartum hemorrhage >1000 mL | 0.34 (0.14–0.87) | 62 (48–315) |
| Blood transfusion | 0.35 (0.22–0.55) | 53 (44–76) | ||
| Antenatal corticosteroids | Women at risk of preterm birth | Respiratory distress syndrome | 0.66 (0.59–0.73) | 11 (9–14) |
| Neonatal death | 0.69 (0.58–0.81) | 22 (16–36) | ||
| Vaginal progesterone | Women with a singleton gestation and a short cervix | Preterm birth <33 wk | 0.56 (0.40–0.80) | 11 (8–25) |
| Major neonatal morbidity/neonatal mortality | 0.59 (0.38–0.91) | 18 (12–81) |
During the late 1990s, Iams et al conducted a multicenter study in the United States and found that the prevalence of a CL ≤25 mm during the midtrimester in women with a singleton gestation was 9%; Heath et al reported an 8% prevalence in a large cohort from the United Kingdom. At our institution, during the period of 2006 through 2015, the prevalence of a CL ≤25 mm was 8.9% (Edgar Hernandez Andrade, MD, PhD, unpublished observations, June 28, 2015). Based on a prevalence of 8%, we calculated that the number of patients with a singleton gestation who need to be screened for a transvaginal sonographic CL to prevent 1 case of preterm birth before 34 weeks of gestation is 125 women, assuming that all patients with a CL ≤25 mm would receive vaginal progesterone ( Table 1 ). The number of patients who need to be screened to prevent 1 case of major neonatal morbidity or neonatal mortality is 225 women.
Currently, there is strong evidence indicating that universal transvaginal sonographic CL screening and vaginal progesterone administration for women with a short cervix is a cost-effective intervention to prevent preterm birth and associated morbidity and mortality, regardless of the cutoff used to define a short cervix in the decision and economic analyses (≤15 mm, ≤20 mm, ≤25 mm, 10-20 mm, or 10-30 mm ). Moreover, the analysis of Combs provided compelling evidence that this strategy satisfies the 10 criteria that the World Health Organization considers to be a good screening test.
Using current national vital statistics, we estimated that 303,613 women with a singleton gestation and a CL ≤25 mm would have been identified in the United States in 2013 if universal transvaginal sonographic CL screening had been implemented. Based on the results of the individual patient data metaanalysis, if all the women with a CL ≤25 mm had received vaginal progesterone, then 30,545 preterm births occurring before 34 weeks of gestation (95% CI, 13,434–41,793) and 17,433 cases of major neonatal morbidity or neonatal mortality (95% CI, 3740–25,765) could have been prevented. Since the rate of preterm birth occurring before 34 weeks of gestation among singleton gestations was 2.72% (n = 103,228), the implementation of universal transvaginal sonographic CL screening and treatment with vaginal progesterone could contribute to a decrease of 30% (30,545/103,228) in the rate of preterm birth occurring before 34 weeks of gestation in the United States. Universal transvaginal sonographic screening for a short cervix and treatment with vaginal progesterone appears to be a relevant public health strategy when compared to universal screening for maternal group B streptococcal colonization and intrapartum antibiotic prophylaxis ( Table 3 ).
