Perinatal outcome of monochorionic twin pregnancy complicated by selective fetal growth restriction according to management: systematic review and meta-analysis.

Abstract

To explore the impact of severity and management (expectant, laser treatment or selective reduction) on perinatal outcome of monochorionic twin pregnancies complicated by selective fetal growth restriction (sFGR). MEDLINE, EMBASE, CINAHL, ClinicalTrials.gov and The Cochrane Library databases were searched for studies on outcome following expectant management, laser treatment or selective reduction in monochorionic twin pregnancies complicated by sFGR. Only pregnancies affected by sFGR and categorized according to the Gratacós classification (Type I, II or III) were included. The primary outcome was mortality, including single and double intrauterine (IUD), neonatal (NND) and perinatal deaths. Secondary outcomes were neonatal morbidity, abnormal postnatal brain imaging, intraventricular hemorrhage, periventricular leukomalacia, respiratory distress syndrome, admission to neonatal intensive care unit and survival free from neurological complications (intact survival). Meta-analyses of proportions were used to analyze the extracted data according to management, severity of sFGR and fetal size (smaller vs larger twin). Sixteen observational studies (786 monochorionic twin pregnancies) were included. In pregnancies complicated by Type-I sFGR managed expectantly, IUD occurred in 3.1% (95% CI, 1.1-5.9%) of fetuses and 97.9% (95% CI, 93.6-99.9%) of twins had intact survival. In pregnancies complicated by Type-I sFGR treated using laser therapy, IUD occurred in 16.7% (95% CI, 0.4-64.1%) of fetuses and, in those treated using selective reduction, IUD occurred in 0% (95% CI, 0-34.9%) of cotwins, with no evidence of neurological complications in the survivors. In pregnancies complicated by Type-II sFGR managed expectantly, IUD occurred in 16.6% (95% CI, 6.9-29.5%) and NND in 6.4% (95% CI, 0.2-28.2%) of fetuses, and 89.3% (95% CI, 71.8-97.7%) of twins survived without neurological compromise. In Type-II sFGR pregnancies treated using laser therapy, IUD occurred in 44.3% (95% CI, 22.2-67.7%) of fetuses, while none of the affected cases experienced morbidity and survivors were free of neurological complications. Of pregnancies undergoing selective reduction, IUD of the cotwin occurred in 5.0% (95% CI, 0.03-20.5%) and NND in 3.7% (95% CI, 0.2-11.1%), and 90.6% (95% CI, 42.3-94.3%) of surviving cotwins were free from neurological complications. In pregnancies complicated by Type-III sFGR managed expectantly, IUD occurred in 13.2% (95% CI, 7.2-20.5%) and NND in 6.8% (95% CI, 0.7-18.6%) of fetuses, and 61.9% (95% CI, 38.4-81.9%) of twins had intact survival. In pregnancies complicated by Type-III sFGR treated with laser therapy, IUD occurred in 32.9% (95% CI, 20.9-46.2%) of fetuses and all surviving twins were without neurological complications. Finally, in pregnancies with Type-III sFGR treated with selective reduction, NND occurred in 5.2% (95% CI, 0.8-12.8%) of cotwins and 98.8% (95% CI, 93.9-99.9%) had intact survival. Type-I sFGR is characterized by good perinatal outcome when managed expectantly, which represents the most reasonable management strategy for the large majority of affected cases. Pregnancies complicated by Type-II or -III sFGR treated with fetoscopic laser ablation have a higher rate of mortality but lower rate of morbidity compared with those managed expectantly, supporting the use of fetal therapy at gestations remote from neonatal viability. Data on outcome following selective reduction are scarce. In view of the lack of evidence from randomized controlled trials, prenatal management of sFGR should be individualized according to gestational age at diagnosis, severity of growth discordance and magnitude of Doppler anomalies. Copyright © 2018 ISUOG. Published by John Wiley & Sons Ltd.