1| INTRODUCTION
Complications of preterm birth (PTB) are the primary cause of death among children in the first 5 years of life, accounting for approximately 35% of deaths among new-borns and 18% of all paediatric deaths.1 Twin gestations are increasing in number and currently account for 3% of all live births and approximately 15-20% of all PTBs.2 The incidence of PTB is 6-8 times higher in twin pregnancy than in singleton pregnancy.3Approximately 70% of twins are born preterm due to preterm premature rupture of membranes or spontaneous labour, while others result from an iatrogenic delivery indicated by monochorionicity, preeclampsia or other maternal or foetal disorders.4,5
Regarding preterm babies, longer gestational age at birth is associated with better prognosis. The EPIPAGE-2 cohort study demonstrated that the survival rates at 22-26, 27-31, and 32-34 weeks of gestation are 51.7%, 93.1%, and 98.6%, respectively.6 Compared with singletons, twins born preterm (before 32 weeks of gestation) are at twice the risk of high-grade intraventricular hemorrhage and periventricular leucomalacia.7 Therefore, better prediction, prevention, and management of PTB is necessary to improve the quality of maternal and neonate care.
To date, strategies for the prevention of PTB in twin pregnancy, such as the use of vaginal progesterone, a cervical pessary and a cervical cerclage, remain controversial or are considered to have limited effects.5, 8-14 To address the growing desire for better guidance for clinical practice, it is a prerequisite to distinguish the patients who are at an increased risk of extreme and very-PTB among the whole twin-pregnancy population, as follow-up treatment may be associated with greater benefit than workload in this high-risk population.
As a demonstrably available and reproducible method, cervical assessment with transvaginal ultrasonography has been increasingly used.15-17 Cervical shortening and cervical funneling are associated with a higher risk of preterm delivery.18-24 However, the predictive accuracy of cervical length for PTB in twin pregnancies are conflicting. Its isolated use as a screening tool has limited value due to low sensitivity.16, 25, 26 In addition, previous research has demonstrated that the risk of PTB is also affected by maternal demographic features, such as ethnic origin, age, primiparity, chorionicity, prepregnancy body mass index (BMI) and history of previous preterm delivery or late-term abortion.27-32In a sense, twin gestation itself is one of the strongest risk factors for PTB.4,33 Assessment of the individual maternal prognosis requires clinicians to consider an array of maternal demographic factors and clinical variables that may be clinically challenging to synthesize. However, the relative contribution of certain characteristics to a given individual’s likelihood of preterm birth and whether or how these features may interact remain poorly understood.
Thus, instead of complicating the clinicians’ lives with close monitoring and administration resulting from an undefined or inherently subjective risk assessment, it would be useful to invest our research efforts in developing a simple and practical algorithm to calculate a risk assessment of PTB for twin pregnancies, similar to the first trimester genetic disease screening tools or the Framingham heart disease score.34
The purpose of this study is to improve our collective understanding of valuable predictive factors related to preterm birth of twin gestation. We then used these factors to develop and validate the prediction model of spontaneous PTB (SPTB) at <32 weeks to provide a comprehensive risk estimation as a clinical assessment tool.