[Is universal screening for cervical length among singleton pregnancies with no history of preterm birth justified?]

Abstract

The ultrasonographic measurement of cervical length with a cutoff of 15mm is currently the best method to identify a group of asymptomatic women in the general population at risk of spontaneous preterm birth, especially among asymptomatic patients with a singleton pregnancy with no history of preterm birth. Cerclage and 17 alpha-hydroxyprogesterone caproate (17OHP-C) are ineffective to reduce the risk of preterm birth among asymptomatic patients with a short cervix in midtrimester. However, vaginal progesterone (200-mg capsules of micronized progesterone or gel containing 90mg progesterone) has been demonstrated effective in 2 large randomized trials to reduce the risk of preterm birth and possibly the composite morbidity and perinatal mortality associated among asymptomatic women with a short cervix in the general population screened by ultrasound of the cervix in midtrimester. Three cost-effectiveness analyses are converging to show that universal screening for cervical length with vaginal progesterone treatment seems to be cost-effective compared with no screening. However, it is too early to definitively conclude that universal screening is justified for several reasons: many women must be screened to prevent a relatively small number of preterm births. Moreover, the epidemiology of preterm delivery is such that the use of progesterone in asymptomatic women with a short cervix screened by ultrasound in midtrimester in the general population will not significantly reduce the prevalence of preterm births; there are no data comparing the effectiveness of universal ultrasound screening followed by vaginal progesterone treatment in case of short cervix versus no universal screening associated to a progesterone treatment in case of incidentally observed short cervix; the universal ultrasound screening may not produce the same results in practice than those observed in published randomized trials, due to population differences, "indication creep", or "stretching of the cutoff" defining the short cervix. Moreover, the implementation of unevaluated or not recommended treatments, such as bed rest, tocolytics, 17OHP-C or cerclage, can potentially cause unintended deleterious consequences and reduce the cost-effectiveness; the cost-effectiveness analyses evaluating universal screening for cervical length present uncertainties on critical variables, notably the short cervix prevalence and the progesterone efficacy. In conclusion, although the implementation of such a screening strategy can be considered by individual practitioners, this screening cannot be universally mandated.

Keywords: Accouchement prématuré; Cervical length; Dépistage universel; Longueur du col; Preterm birth; Progesterone; Progestérone; Ultrasound; Universal screening; Échographie.