Increased stillbirth in uncomplicated monochorionic twin pregnancies: A systematic review and meta-analysis

摘要

We read the review by Danon et al with interest. Estimation of risk of stillbirth in twin pregnancies provides invaluable information for clinical practice. We would like to raise the following issues in their systematic review. Firstly, we question the appropriateness of estimation of survival curves using the Kaplan Meier method (see Figure 3 in their article).This method assumes that censors are noninformative and happen merely by chance. Under this random noninformative censoring assumption, censored cases are thought to behave in the same way as women continuing with their twin pregnancies. This is accounted for in the calculations of risks because it is expected that some events will occur in censored cases.