Birth weight reference for Japanese twins and risk factors for infant mortality: A population-based study.

There is no standard birth weight curve for twins in Japan other than a prototype curve based on 1988-1991. Twins have a high perinatal mortality rate than singletons; therefore, we developed a new standard curve for twin birth weight using data from the 1995-2016 Vital Statistics and compared it with previous reports. We used 469,064 cases for analysis, excluding stillbirths and cases with missing values, and created a standard curve using LMS (statistical methods to vary the distribution by using skewness, median, and coefficient of variation) method. In comparison with previous reports, the mean birth weight decreased by 100-200 g. The groups with the lowest neonatal death rates (NDRs) and infant death rates (IDRs) were those with a birth weight of 1,500-2,499 g (NDR: 0.3%, IDR: 0.6%) and those born at 34-36 weeks (NDR: 0.2%, IDR: 0.4%). Compared to these, the IDR was significantly higher in the 2,500-3,999 g group and the 37-39 weeks group (incidence rate ratio (IRR): 1.1 in the 2,500-3,999 g group, IRR: 1.3 in the 37w0d-39w6d group). In particular, the risks of neonatal mortality and infant mortality were higher in infants born at a birth weight above 3,500 g. Infants born at a birth weight above 3,500 g may include recipients of twin-to-twin transfusion syndrome. The most common causes of infant mortality are accidental death and sudden infant death syndrome (SIDS). We considered the possibility that infants treated as healthy newborns and whose mothers were discharged from the hospital without adequate twin care guidance may be more likely to experience unintentional accidents and SIDS at home. The present study suggested that creating a new twin birth weight standard curve and guidance on managing twins at home for full-term and normal birth weight infants may lead to a reduction in infant deaths.