Purpose: To compare diagnostic accuracy of sonographic cervical length (CL) measurement and uterine electric activity assessed by electromyography (EMG) in second trimester regarding prediction of preterm delivery (PTD).
Methods: Prospective study of 308 low-risk women. Shortened CL was defined as ≤25 mm (≤5th centile), while raised EMG activity was defined as the presence of ≥20 action potentials in 20 min of assessment (≥95th centile). Outcome measures were diagnostic accuracy of both tests alone or in combination for prediction of PTD and early PTD (≤34 weeks).
Results: The incidence of PTD was 23/308 (7.4%) while the incidence of early PTD was 9/308 (2.9%). Shortened CL and raised EMG activity were significantly related to PTD [prevalence-weighted likelihood ratio (pw-LR) 1.9, 95% CI 1.0-3.5 vs. 9.5, 95% CI 2.5-35.7], but not to early PTD (pw-LR 0.4, 95% CI 0.2-0.8 vs. 0.6, 95% CI 0.3-1.7). Significant predictive value for early PTD was found only if both tests were combined (pw-LR 4, 95% CI 1.3-14.3).
Conclusion: Shortened CL and raised EMG activity in second trimester have significant diagnostic accuracy regarding prediction of PTD in a low-risk population. However, in order to be useful as a predictor for early PTD both tests must be positive.