4.4 Strengths and limitations
The main strength of our study lies in the following points. The first one is that only healthy uncomplicated twin pregnancies were included. Secondly, our reference charts were based on longitudinal data as well adequate sample size, which made it available to develop growth centiles to identify patterns and differences in fetal growth at different gestational ages14. Thirdly, taking both chorionicity and mode of conception into account promised our growth reference chart a more reliable and sensitive tool in identifying growth restriction in twins. Forth, the GAMLSS model has unique advantages in fitting the changing curve over time, which has been applied by WHO to establish child growth standards10. Moreover, making comparison with data from singleton provided evidence to confirm that twin has intrinsic growth pattern, emphasizing the importance of this study.
However, several weaknesses also warrant acknowledgement. First, this study was a single-center design, there might be a selection bias. However, International Peace Maternity & Child Health Hospital is the top quality maternal and child health hospital in Shanghai, which attracts pregnant women all around Shanghai. The hospital has opened twin pregnancy clinic since 2015, which is one of three specialist consultation settings for twin pregnancy in Shanghai. Therefore, the twin pregnant women recruited through four years should have a certain representative in Shanghai area. Secondly, very minority (n=23) was MCDA twins conceived by ART, therefore, they were not included in this study. Finally, although we adopted strict inclusion criteria to acquire qualified target population, other high-risk individual with unclear adverse perinatal outcomes was possible.