Third Trimester Screening of Fetal Growth Restriction

The Aim of this trial is to assess the efficacy and efficiency of two screening protocols for fetal growth disorders: one providing an early fetal growth assessment at 28-32 weeks' (single scan) and the second one with the addition of a late third trimester screening at 35-37 weeks (longitudinal scan).

This study is designed to generate level 1 evidence of diagnostic efficacy.

MAIN OUTCOME: The main outcome of the two protocols in screening true late SGA fetuses is measured based on the antenatal detection of small for gestational age newborn basing on the IG21th newborn weight charts centiles \<the 10th according to gestational age and gender at delivery.

SECONDARY OUTCOME

• Antenatal detection of severe SGA \<3rd centile
• Mode of delivery
• Caesarean section rate according to Robson classes for singleton
• Composite mild adverse neonatal outcome (Apgar score at 5 minutes \<7, pH \< 7.10, or BD >8mmol/L, admission at NICU)
• Composite severe adverse perinatal (stillbirth or term live birth associated with neonatal death, hypoxic ischaemic encephalopathy, use of inotropes, need for mechanical ventilation, or severe metabolic acidosis (defined as a cord blood pH \<7•0 and base deficit >12 mmol/L)

TERTIARY OUTCOME

• Number of ultrasound scans performed in outpatient clinics by medical indication (SSR-prescription) beyond Early-TT and late-TT scheduled exams, and number of ultrasound scans performed on indication in outpatient clinics minus the first one indicated.
• Number of ultrasound scans by patient choice
• Econometrics of the two protocols, and of the estimated sanitary costs of outcomes

4.2 EXPERIMENTAL PLAN

Women who agree to participate are randomized to a single scan Protocol vs longitudinal Scan protocol.

5. STUDY POPULATION

Sample size is based on sensitivity in detecting SGA below 10th centile weight at birth. For the 28-32 group, sensitivity is based on results by Roma et al. (22.5%). The results published by Sovio et al. are instead taken for the "longitudinal" group (57% sensitivity). The sample size calculation was carried out for the comparison between the 28-32 only group vs. the "longitudinal". The calculation considered a confidence level of 95% (1-alpha) and an 80% power (1-beta), and a one to one ratio between the 28-32 only groups and the "longitudinal". According to Fleiss (Fleiss JL. Statistical methodsf or rates and proportions, II Ed. John Wiley \& Sons, New York, 1981. page45, formulas 3.18 \& 3.19), for the 28-32 only group vs. the "longitudinal",each arm should be 32 women, which implies 64 for two arms. SGA represent, as by definition, 10% of the total population: the sample size needed considering both AGA and SGA, is then 64x10=640. Considering 10% of drop-out and incomplete data the final sample size should be increased of such percentage, reaching 704 subjects. The sample should be recruited according to the number of deliveries per year of each institute involved in the multicenter study.

5.2 PATIENT SELECTION 5.2.1 INCLUSION CRITERIA Eligible cases are nulliparous pregnant women, with first trimester ultrasound assessment of gestational age, who conceived singleton fetuses. If the crown-rump-length (CRL) differs of more than ± 3-5 days from the last menstrual period, gestational age is calculated on the CRL.

5.2.2 EXCLUSION CRITERIA

• major medical disease
• high risk for preeclampsia on history or detected in those centres that perform pre-eclampsia screening at first trimester
• women older than 40 years on ASA low dose prophylaxis
• known immune disorders or clinical thrombophilic conditions;
• twin pregnancies;
• IVF
• suspected fetal anomalies
• Papp-A at Combined-Test\<0.2

5.3 WITHDRAWAL PROCEDURES If a patient, during the study protocol, wanted to withdrawal, this will not modify the subsequent monitoring of her pregnancy following local protocols.

6. SUBJECT TREATMENT

• RANDOMIZATION : Patients who agree to participate and sign an informed consent, Women who agree to participate are randomized to the single scan Protocol vs Longitudinal-TT protocols.