Materials and Methods
We performed a retrospective cohort study from January 2006 to August 2017. We identified all triplet pregnancies which underwent first trimester ultrasonography to measure the fetal crown-rump length and to determine amnionicity and chronicity. Only trichorionic triamniotic triplet pregnancies were included, and were divided into two groups according to their decisions after counseling on the MFPR; expectant management (EM) group consisting of women who declined MFPR and embryo reduction (ER) group consisting of women who chose MFPR from triplet to twin pregnancy. Women who decided to reduce to singleton pregnancy from triplet pregnancy were excluded. The institutional review boards of Seoul National University Hospital Clinical Research Institute (IRB No.; H-1311-045-533) and CHA Bundang Medical Center approved this study (IRB No.; C 2016-10-007).
MFPR were performed with transvaginal or transabdominal intrathoracic potassium chloride injection before 14 weeks of gestation. We reviewed the clinical records of the mothers and babies for information concerning maternal demographics, clinical presentation, laboratory examinations, ultrasound assessments, and birth outcome. The two groups were compared for the rates of non-viable pregnancy loss before 23 weeks, the rates of preterm birth, and the number of surviving fetuses. Maternal obstetric complications, such as preeclampsia and gestational diabetes, were also evaluated.
To evaluate the short-term birth outcome, we investigated neonatal survival and composite morbidity, which was defined as the occurrence of at least one of the followings: neonatal sepsis, intracranial hemorrhage (ICH), retinopathy of prematurity (ROP), patent ductus arteriosus (PDA), pulmonary hypertension, respiratory distress syndrome (RDS), bronchopulmonary dysplasia (BPD), and necrotizing enterocolitis (NEC). We evaluated SGA using the standard described in the previous report.8 We investigated the long-term neurodevelopment outcomes after one year of corrected age. Developmental delay and cerebral palsy were evaluated by the Korean Ages and Stages Questionnaire (K-ASQ), or Gross Motor Function Measure (GMFM) or chart review. Because the K-ASQ is useful only for young children (4-60 months of age), older children (preschool or elementary school age) were instead evaluated by pediatric charts regarding their ability to perform daily activities or learning ability in school. We considered development to be normal in children described with no difficulties in fulfilling their normal academic obligations; those who attended special schools for mentally retarded children were categorized as cases of developmental delay.
Statistical analysis was performed using SPSS version 23.0 (SPSS Institute, Chicago, IL, USA) and R, version 3.3.1 software (R Foundation for Statistical Computing, Vienna, Austria; http://www.r-project.org). We analyzed discrete data using Fisher’s exact test and comparisons of continuous variables were performed with Mann-Whitney U test, as appropriate. A generalized estimating equation (GEE) was used to assess which variables were associated with neonatal outcomes, accounting for the familial correlation in the model, because the study population consisted of twin or triplet pairs within a single mother.9 We also performed a multivariable analysis including maternal age, BMI and method of conception as covariates, which were selected with p < 0.2 in univariate analysis when comparing cases with neonatal survival or those with neonatal death. A P value <0.05 was considered statistically significant.