Twins with short cervix: hope ahead.

Abstract

As I often said, human beings were made to carry one baby at a time. Twin gestations, although seemingly exiting for both mothers and providers, are indeed associated with much higher complications for both mothers and babies. The most frequent and deadly of these complications is preterm birth (PTB). So the quest to decrease PTB in twins is an extremely important one. The best preventive measure for now is surely reducing the incidence of twins in the first place, through the better use of assisted reproductive techniques. This is because once a woman is carrying twins, nothing seems to work. The contribution from Jarde et al. to this issue of BJOG confirms this truth. Though a systematic review and meta-analysis of the literature, they showed that neither pessary, nor cerclage, nor vaginal progesterone prevented PTB in twins with a short transvaginal ultrasound (TVU) cervical length (CL). A few things should be highlighted. First, most of the 23 randomised controlled trials (RCTs) in the meta-analysis included mostly ‘unselected’ twins. It appears to be true that unselected twin gestations do not benefit from vaginal progesterone, pessary, or cerclage. Any possible benefit came in subgroups with short TVU CL. Just ‘throwing’ an intervention at all twins with the hope of preventing PTB does not seem to work, and such RCTs should probably not be prioritised anymore. Second, twins with prior PTB have not been studied separately and sufficiently for these three interventions of progesterone, cerclage, and pessary. Third, the addition of short TVU CL to the risk factor of twins seemed to be helpful in identifying a subgroup of twins who could benefit from intervention. Unfortunately, TVU CL in twins only has ~40% sensitivity for PTB (Conde-Agudelo et al. Am J Obstet Gynecol 2010; 203:128.e1–12.). That is, 60% of twins have PTB without showing a short CL in the second-trimester screening. In contrast, TVU CL has a 70% sensitivity for PTB in singletons with both a prior PTB and a short TVU CL, and it's not surprising that in this population cerclage has been shown to be associated with a reduction in PTB (Berghella et al. Obstet Gynecol. 2011;117:663–71). Fourth, in twins with a short TVU CL, vaginal progesterone seems to be for now the most promising intervention. Although Jarde et al. emphasise the possible benefit in secondary outcomes of this intervention in all twins, on a close reading of each related RCT the benefit of vaginal progesterone seems to come just in twins with a short TVU CL. These data confirm two prior meta-analyses on vaginal progesterone alone in twins with short CL (Schuit et al. BJOG 2015;122:27–37, Romero et al. Am J Obstet Gynecol 2012;206:124.e1–19). Most of the women included had a TVU CL of ≤25 mm. Fifth, in twins with short TVU CL, a pessary has been reported to be beneficial in two RCTs (Liem et al Lancet 2013;382:1341–9 and Goya et al Am J Obstet Gynecol. 2016 Feb;214(2):145–52) and not beneficial in two other RCTs (Nicolaides et al Am J Obstet Gynecol. 2016;214:3.e1–9 and Berghella et al. Ultrasound Obstet Gynecol, 2017; doi: 10.1002/uog.17430); therefore, it remains unclear whether a pessary is beneficial in this population. Finally, only 49 sets of twins with short CL have been included in RCTs of cerclage versus no cerclage (Saccone Acta Obstet Gynecol Scand 2015;94:352–8). Clearly more RCTs are needed to answer this question. For now, we at our institution are continuing to perform RCTs in twins with a short TVU CL: we just finished a study on pessary [Berghella et al. AJOG 2017 (in press)], and are starting another on cerclage. Surely the completion and publication of the continuing RCTs on interventions for twins with a short TVU CL (Table S1) will soon revolutionise the management of twins. None declared. Completed disclosure of interests form available to view online as supporting information. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.