Determining when to deliver a fetus is the only strategy for managing fetal growth restriction (FGR). FGR is a common condition in pregnancy that is associated with various perinatal outcomes [1] . It is a key cause of perinatal and neonatal mortality as well as suboptimal neurodevelopment [2] [3] [4] , with adverse effects in adolescence and adulthood [5] .

Due to limited preventive and treatment measures, the management of FGR is based on optimal fetal surveillance and timely delivery, both of which aim to reduce perinatal mortality and morbidity. The goal of perinatal management of FGR is to deliver the neonate at a higher gestational age with fewer neurological sequelae while avoiding fetal death whenever possible. In most cases, the timing of delivery is decided on an individual basis using antenatal fetal surveillance, including cardiotocography (CTG), biophysical profile score (BPS), estimated weight, amniotic fluid volume, and umbilical artery Doppler velocimetry [1] .

A previous meta-analysis reported that the odds ratios for the risk of fetal death in fetuses with FGR with absent or reversed umbilical artery velocities prior to 34 weeks of gestation were 3.59 and 7.27, respectively [6] . The Society for Maternal-Fetal Medicine’s (SMFM) Consult series recommends delivery at 33 - 34 weeks of gestation for fetuses with FGR and absent end-diastolic velocity (AEDV) in the umbilical artery and at 30 - 32 weeks for fetuses with reversed end-diastolic velocity (REDV) [7] . However, there are no recommendations for fetuses with FGR with AEDV or REDV at <30 weeks [7] . Additionally, there is no clear criterion for delivery in Japan for abnormal umbilical artery Doppler velocity at any point during gestation [8] . Improvements in neonatal care have enabled neonates with severe FGR to be managed, and the impact on child development is now the focus of attention. Since there is no established treatment for FGR, it is important to identify and share the management strategies and prognosis of neonates in each facility. Therefore, this study aimed to identify perinatal morbidity, mortality, and neurodevelopmental outcomes in the management of FGR at a single university hospital.

In this study, we retrospectively reviewed all cases managed in our department in 2013. The inclusion criteria were all cases of FGR. The exclusion criteria were cases where intrauterine fetal death (IUFD) occurred at the time of the initial visit and cases of fetal congenital anomalies.

In Japan, FGR is defined as an estimated fetal weight of < −1.5 standard deviations (SDs) of the mean weight for gestational age [8] . The current management policy of our department is as follows: when FGR is diagnosed, the mother is hospitalized. CTG is performed twice daily, and abdominal ultrasonography is performed 2 - 3 times per week to assess the following parameters: estimated fetal birth weight (EFW), head circumference, BPS, amniotic fluid volume (amniotic fluid index/amniotic fluid pocket), and Doppler velocity of the umbilical, middle cerebral, and ductal venous arteries. Indications for delivery are made comprehensively based on the following abnormal findings according to Japanese guidelines [8] : CTG abnormalities (moderate or severe variant pattern), abnormal umbilical artery Doppler velocity (AEDV/REDV), low BPS (<6), and no significant change in the estimated weight for 2 weeks. FGR is classified as early or late onset based on the gestational age at the time of prenatal ultrasound. Early- and late-onset FGR is diagnosed prior to and after 32 weeks of gestation, respectively [9] . The term “small for gestational age” (SGA) has been used to describe a newborn whose birthweight is less than the 10th percentile for gestational age, and “severe SGA” describes a newborn whose birthweight is less than the 5th percentile for gestational age. Neonates are followed up until 1 year of age and discharged if there are no problems. However, follow-up is continued if developmental abnormalities are suspected. Neurodevelopmental disorders include intellectual disabilities, communication disorders, autism spectrum disorders, attention deficit/hyperactivity, localized learning difficulties, and motor disorders caused by abnormal neurodevelopment in neonates, children, and adolescents.

All procedures were performed according to the ethical standards of the institutional and/or national research committee and in accordance with the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards. This retrospective study was approved by the Institutional Review Board of Ryukyu University, Okinawa, Japan (approval no. 1805).

All statistical analyses were performed using JMP software (SAS Institute, Inc., Cary, NC, US). Proportions were compared using Fisher’s exact test or chi-square test. Continuous variables were analyzed and compared using the student t-test or Mann-Whitney U test. A p-value of <0.05 was considered statistically significant. Factors independently associated with neurodevelopmental disorders were determined by logistic regression analysis for gestational age at birth, birth weight, and umbilical artery resistance index value.

Among the 2465 cases managed in our department from January 2013 to December 2019, 113 (4.6%) had FGR. Three twins and one case of IUFD at the time of initial visit were excluded, and the remaining 109 cases were reviewed for the causes of FGR, indications for delivery, perinatal death, overall neonatal outcomes, and long-term prognosis.

Seventeen cases of fetal congenital anomalies that were included in the analysis of the causes of FGR were excluded from the analyses of other parameters.