2 Year Neurodevelopmental and Intermediate Perinatal Outcomes in Infants With Very Preterm Fetal Growth Restriction (TRUFFLE): A Randomised Trial

There is no consensus on the best methods to monitor fetal growth restriction or to trigger delivery. Previous studies have suggested that the abnormal ductus venosus (DV) pulsatility index is the best discriminating variable for neonatal outcome. This study hypothesized that changes in the fetal DV Doppler waveform, rather than cardiotocograph short-term variation, should be used as indications for delivery. The study (TRUFFLE [TRrial of Umbilical and Fetal Flow in Europe]) was conducted in 20 tertiary-care centers in 5 European countries from January 1, 2005, to October 1, 2010. Women were assigned randomly to 3 different groups. One group decided on the timing of delivery with criteria for reduced short-term variation (CTG STV). The others assessed timing of delivery based on abnormalities of the DV waveform. In the second group, delivery timing was based on early DV changes, and in the third group, delivery timing was based on late DV changes. Early DV changes were defined as pulsatility index greater than the 95th percentile (DV p95), and late DV changes were defined as no or reversed flow in the DV A wave (DV no A). Survival without neurodevelopmental impairment at 2 years of age, corrected for prematurity, was the primary outcome. Data were obtained for 503 women included in the study. Of those, 461 infants survived until 2 years of age, and 402 completed follow up at 2 years. Of the 503 infants, 354 (69%) survived without severe neonatal morbidity. One hundred forty-four infants were assigned to the CTG STV group, of whom 111 survived without impairment (77%). Of the 142 assigned to the DV-p95 group, 118 (84%) survived without impairment, and 133 (85%) of 157 survived without impairment in the DV-no-A group (P-trend = 0.09). In the CTG STV group, there were 13 deaths in 166 infants (8%); in the DV p85 group, 11 (7%) deaths of 167, and in the DV-no-A group, 17 (10%) of 170 (P-trend = 0.35). Impairment in survivors was found in 15% of the CTG STV group (20 of 131), 9% in the DV-p95 group (12 or 131), and 5% of the DV-no-A group (7 of 140) (P-trend = 0.004). No differences in short-term outcomes or baseline variables were seen between the 3 study groups. Improved outcome in survivors was shown in the DV-no-A group compared with the CTG STV group (P-trend = 0.005). At 2 years, neuroimpairment was less frequent in the DV-no-A group than in the CTG STV group, and these findings support waiting for late DV changes, contrasting with previous study suggesting a worse outcome is associated with this method.