What is the positive predictive value for preterm birth (<37 weeks of gestation) for a shortened cervix (<2.5 cm) found at the time of the fetal anatomical scan at 20 weeks of gestation?

Positive Predictive Value for Preterm Birth with Shortened Cervix at 20 Weeks

For a cervical length <2.5 cm (≤25 mm) detected at the 20-week anatomical scan, the positive predictive value for preterm birth before 37 weeks of gestation is approximately 16.2%. 1

Specific PPV Data by Cervical Length Threshold

The Society for Maternal-Fetal Medicine provides precise predictive values based on cervical length cutoffs measured at 16-22 weeks of gestation (which includes the typical 20-week anatomical scan): 1

For Cervical Length ≤25 mm at 16-22 weeks:

• PPV for PTB <37 weeks: 16.2% 1
• PPV for PTB <32 weeks: 7.4% 1
• Sensitivity: 8.0% 1
• Specificity: 97.8% 1

For Cervical Length ≤20 mm at 16-22 weeks:

• PPV for PTB <37 weeks: 15.5% 1
• PPV for PTB <32 weeks: 8.6% 1
• Sensitivity: 4.1% 1
• Specificity: 98.8% 1

Clinical Interpretation

Despite the relatively modest positive predictive value of 16.2%, the high specificity (97.8%) and negative predictive value (94.7% based on symptomatic populations) make cervical length screening clinically valuable. 1, 2 This means that while most women with a short cervix will still deliver at term, the finding consistently identifies a high-risk population that benefits from intervention.

Key Clinical Points:

• The finding of a short cervix at 20 weeks increases preterm birth risk approximately 3.5-fold (positive likelihood ratio of 3.67 for ≤25 mm). 1

• The majority (approximately 84%) of women with cervical length ≤25 mm at 20 weeks will still deliver at term, but this represents a significantly elevated risk compared to women with normal cervical length. 1

• For cervical lengths ≤20 mm, the Society for Maternal-Fetal Medicine strongly recommends vaginal progesterone (GRADE 1A) to reduce preterm birth risk, despite the modest PPV, because the intervention is safe and effective. 1, 3

Important Caveats

These predictive values apply specifically to nulliparous singleton gestations without prior spontaneous preterm birth. 1 Women with different risk profiles (prior preterm birth, multiple gestations) will have different predictive values.

Measurement technique is critical: These values are based on standardized transvaginal ultrasound measurements with proper technique and quality assurance. 1, 4 Transabdominal measurements are insufficient for clinical decision-making. 1, 4

The PPV increases if cervical length continues to shorten on subsequent measurements, with cervical length shortening >10 mm from the anatomical scan to a later symptomatic presentation increasing the PPV to 21.2%. 2