We read the article by Mendez-Figueroa et al with great interest. They proposed the use of egg white precipitate as a sealant for cases of iatrogenic preterm premature rupture of the membranes. In this study, the authors reported that egg white precipitate sealant significantly decreased the amniotic fluid leakage over a 24-hour period. They already mentioned the immunological concerns of egg white precipitate. We would like to comment on several other issues about the design and conclusions of the study.

A statistically significant result was reported with a 5-mL difference between both groups over 24 hours (55.1 vs 61.6 mL). We thought that the difference was very small considering the time interval. Even though there is a statistical significance, a 5-mL difference over a 24-hour period would not be clinically significant. Moreover, technically, the measurement of a difference of 5 mL using a 3-cm beaker might not be possible. Using the testing equipment as seen in the figure from the original article with thick level line and foamy amniotic fluid without a blinded observer, it is easy to introduce bias while adding fluid to the beakers. Indeed, half of the differences among test tubes were a few milliliters.

We believe that the most important factor that leads to clinical failure of these agents is related with the natural healing of the membranes. Previous studies showed that the membranes do not have the capacity of true histopathologic healing. Therefore, closure of puncture sites in membranes are thought to be accomplished by free membrane edges. In accordance with this theory, the most important finding preceding the membrane complications is amniochorionic separation that is reported in 47% of the cases. Using agents with simple plugging action might fail as with previous attempts with similar agents. Another reason for the failure of plugging might be the difference between the testing environment and real-life situations. In real life, motility of mother, fetus, and the membranes were further complicated with periodically increasing intraamniotic pressure with spontaneous Braxton Hicks contractions. Additionally, the use of antibiotics and autoclaving instruments used in this experiment might also mask potential of bacterial overgrowth with the use of egg white. Instead of agents that plug the holes, true adhesives that stick the free edges together might be a better approach for the prevention of membrane complications.

As a conclusion, success of intrauterine fetal surgical procedures still depends on the effective and lasting closure of the membranes. However, we believe that true adhesive potential should be tested instead of plugging potential. Similarly, we should focus on the prevention of amniochorionic separation with dynamic experimental systems instead of leakage in static testing environment.