Thank you for the interest in our metaanalysis. We congratulate Dr Martinez de Tejada et al on their recent publication on vaginal progesterone for prevention of preterm birth (4P trial).

One of the most important aspects of a metaanalysis is the inclusion criteria. Strict inclusion criteria are needed to reduce both clinical and statistical heterogeneity. Moreover, the protocol of every metaanalysis should be decided a priori before the data extraction and should not be modified. These are key elements that are needed to evaluate the reliability of a metaanalysis. In 2011, the first international prospective register of systematic reviews (PROSPERO) was launched by the Centre for Reviews and Dissemination, University of York, UK. All journals should encourage prospective registration of all planned systematic review protocols because it helps to promote transparency and safeguards against publication bias and duplication of reviews. Recently, preferred reporting items for systematic review and meta-analysis guidelines for protocols have also been published. In our protocol, which is registered with PROSPERO (CRD42014013706), we a priori decided to include all published randomized controlled trials (RCTs) of singleton gestations that had arrested preterm labor (PTL) and were randomized to maintenance tocolysis treatment with either vaginal progesterone or control.

As Martinez de Tejada et al knows well, we tried our best to include their trial in our metaanalysis, as we can confirm by the several emails that we exchanged directly with her early in 2015. In their RCT, vaginal progesterone was given within 48 hours of the start of acute tocolysis and was used as an additional agent with primary tocolysis, not as maintenance tocolysis, which was also pointed out by the Commentary to their study. Indeed, in their RCT, vaginal progesterone appears to be used for women both who had (perhaps) arrested PTL and those who did not. Maintenance tocolysis means that preterm contractions have resolved, at least 48 hours have elapsed from presentation, and steroids have been given; now the patient is being considered for discharge. This is not at all what happened in the 4P RCT. The 4P authors did not mention whether the study subjects were assessed for arrested PTL before randomization and allocation of vaginal progesterone vs placebo. For these reasons, including their RCT would have been methodologically incorrect, which would have compromised the reliability of our metaanalysis on maintenance tocolysis. We had already explained this well in several emails to Martinez de Tejada, who was aware. There are indeed other RCTs that use progesterone as an additional agent to primary tocolysis. The RCT by Martinez de Tejada et al can, if desired, possibly be combined with this RCT and other future similar trials.

After carefully reviewing the 4P trial, we agree with Dr Norwitz’s commentary to this trial and then his reply again to Martinez de Tejada et al that the 4P study is underpowered to conclude that there is no benefit of vaginal progesterone as a tocolytic, given, among other reasons, the fact that they did not reach their own precalculated sample size. Unlike these authors, we did not make any absolute and definitive clinical recommendations in our study. In our metaanalysis, we concluded that “Maintenance tocolysis with vaginal progesterone is associated with prevention of PTB, significant prolongation of pregnancy, and lower neonatal sepsis. However, given the frequent lack of blinding and the generally poor quality of the trials, we do not currently suggest a change in clinical care of women with arrested PTL. We suggest instead well-designed placebo-controlled randomized trials to confirm the findings of our meta-analysis.”

Once again, congratulations to Martinez de Tejada et al for their 4P trial. We look forward to more excellent work and publications from their group.