Multicenter Randomized Trial of Cerclage for Preterm Birth Prevention in High-Risk Women With Shortened Midtrimester Cervical Length (Owen 2009)
Source: View publication →
In high-risk women with a history of preterm birth and a midtrimester short cervix, ultrasound-indicated cerclage did not significantly reduce preterm birth before 35 weeks overall, but it significantly decreased previable birth and perinatal mortality, especially in those with profound cervical shortening.
Key Findings
Study Design
Study Limitations
Clinical Significance
This trial solidified the clinical paradigm that obstetric history and midtrimester transvaginal cervical length screening should be used in tandem to guide intervention. It demonstrated that ultrasound-indicated cerclage is a highly effective, life-saving intervention for preventing previable births and improving perinatal survival in women with prior preterm births and a midtrimester cervix < 25 mm.
Historical Context
Prior to this trial, the efficacy of cerclage for women found to have a short cervix was heavily debated. Previous large trials evaluated cerclage in a general, predominantly low-risk screening population with a short cervix and found no significant benefit. The 2009 Owen trial specifically targeted a "high-risk" population—those with a prior spontaneous preterm birth—combining obstetric history with ultrasound findings. This landmark trial provided the crucial evidence base that shaped current ACOG and SMFM guidelines, which now specifically recommend ultrasound-indicated cerclage for women with both a prior spontaneous preterm birth and a shortened midtrimester cervical length.
Guided Discussion
High-yield insights from every perspective
Anatomically and pathophysiologically, how does cervical insufficiency lead to midtrimester pregnancy loss, and what specific mechanical or structural role does a McDonald or Shirodkar cerclage play in preventing it?
Key Response
Understanding the cervix structural matrix (collagen and elastin) is foundational. Cervical insufficiency involves premature remodeling and dilation. A cerclage provides mechanical support against gravitational and uterine forces and helps maintain the mucosal plug barrier, preventing ascending infections which are a major trigger of preterm labor.
Based on the Owen 2009 trial and current clinical practice, how do you differentiate the criteria for an ultrasound-indicated cerclage from a history-indicated or physical exam-indicated cerclage, and which specific patient population benefits from serial cervical length screening?
Key Response
Residents must categorize cerclage indications. History-indicated is placed at 12-14 weeks purely based on obstetric history. Exam-indicated is placed emergently for painless dilation in the second trimester. Ultrasound-indicated (the Owen population) requires a history of prior spontaneous preterm birth and a shortened cervix (<25 mm) detected on serial transvaginal ultrasound screening between 16 and 24 weeks.
The Owen 2009 trial did not reach statistical significance for its primary outcome of preterm birth <35 weeks (P=0.09), yet it profoundly influenced MFM practice. How do you integrate the borderline primary outcome with the significant secondary outcomes (previable birth, perinatal mortality) and the <15mm cervical length subgroup findings when counseling a high-risk patient?
Key Response
Fellows must grapple with technically negative trials that drive clinical care. The trial was likely underpowered due to a lower-than-expected event rate, but the drastic, statistically significant reduction in catastrophic outcomes (previable births and perinatal mortality), combined with profound benefit in the <15mm subgroup, justifies the intervention clinically despite the primary outcome missing the p<0.05 threshold.
Given that the Owen trial was conducted before the widespread prophylactic use of vaginal progesterone for a short cervix, how should modern obstetricians balance or sequence the use of vaginal progesterone versus cerclage in a patient with a prior spontaneous preterm birth who presents with a midtrimester cervical length of 20 mm?
Key Response
Attendings must navigate overlapping therapies. Current data suggests cerclage is the definitive preferred intervention for a patient with a prior spontaneous preterm birth and a current short cervix (<25 mm). While progesterone is the standard for a short cervix without a prior preterm birth, the additive benefit of using both therapies in the Owen population remains a complex, highly debated area requiring individualized shared decision-making.
Scholarly Review
Critical appraisal through the lens of expert reviewers and guideline development
The Owen trial sample size was calculated based on an expected 50% rate of preterm birth in the control group, but the actual rate was 42%, leading to an underpowered P-value of 0.09 for the primary outcome. How does overestimating the baseline event rate impact study power, and what adaptive trial designs could have been employed to mitigate this type II error?
Key Response
Overestimating the control event rate reduces the statistical power to detect a true difference, often leading to a Type II error. Methodologists should consider adaptive sample size re-estimation designs, which allow researchers to conduct blinded interim analyses of the overall event rate and adjust the final target enrollment to preserve the study original power without inflating the Type I error rate.
As a journal editor reviewing this manuscript, how do you evaluate and ethically publish a paper where the primary endpoint fails to achieve statistical significance, but the authors emphasize statistically significant secondary outcomes and subgroup analyses to suggest a clinical benefit?
Key Response
Editors must be highly vigilant against outcome reporting bias and spin. While the Owen trials secondary findings were biologically plausible and pre-planned, reviewers must ensure the abstract and conclusion accurately reflect the primary outcome failure first. Emphasizing secondary outcomes is acceptable only as hypothesis-generating or contextualized cautiously, leaving it to systematic reviews to definitively prove efficacy.
ACOG and SMFM currently recommend ultrasound-indicated cerclage for women with a prior singleton spontaneous preterm birth and a cervical length <25 mm before 24 weeks. How did the meta-analysis combining the Owen 2009 data with other trials overcome the individual trial lack of primary significance to establish this strong Level A recommendation?
Key Response
Guideline committees rely heavily on pooled data. Berghella patient-level meta-analysis, which included the Owen 2009 data, provided the necessary statistical power to prove a significant, definitive reduction in preterm birth <35 weeks. This demonstrates how clinical guidelines are often shaped not by a single perfectly powered trial, but by the systematic aggregation of rigorous, clinically similar, but individually underpowered RCTs.
Clinical Landscape
Noteworthy Related Trials
MRC/RCOG Cervical Cerclage Trial
Tested
Cervical cerclage
Population
Women at high risk of preterm delivery based on obstetric history
Comparator
Expectant management
Endpoint
Delivery before 33 weeks of gestation
To Trial
Tested
Cervical cerclage
Population
Asymptomatic women with a short cervix (<15 mm) identified on routine mid-trimester transvaginal ultrasound
Comparator
Expectant management
Endpoint
Preterm delivery before 33 weeks
PREGNANT Trial
Tested
Vaginal progesterone gel (90 mg daily)
Population
Women with a short cervix (10-20 mm) at 19-24 weeks of gestation
Comparator
Placebo
Endpoint
Preterm birth before 33 weeks
Tailored to your role
Want this tailored to you?
Add your specialty or training stage to get role-specific takeaways and more questions.
Personalize this analysis