• congenital anomalies
• prenatal diagnosis
• ultrasound screening in pregnancy

SUMMARY
The fetal anatomy evaluation and the subsequent timely detection of congenital anomalies is usually included in the aims of second trimester ultrasound screening. However, some structural anomalies can be detected before or after this time period. A growing body of literature is dealing with first trimester diagnosis, often in the setting of the nuchal translucency screening programme for trisomy 21. On the other hand, the term ‘’ late- onset’’ anomalies had been used since almost 20 years to refer to anomalies only detectable in the third trimester.
The aim of the present study was to evaluate the gestational age when the diagnosis of fetal structural anomalies is performed in the unselected obstetric population.
The pregnant women with a suspicion of fetal structural anomaly at routine prenatal sonography and those with an increased risk for fetal anomalies are referred to the tertiary unit of University of Pisa based on geographic criteria. Among those examined between January 2011 and December 2015, 203 cases of fetal structural anomalies were eligible for the study, on the basis of their confirmation after completion of the pregnancy.
In our series, the majority of congenital anomalies were diagnosed between 15 and 22 weeks’ gestation (69.5 %), with a significant lower proportion detected before 14 (11.3 %) and after 23 weeks’ gestation (19.2 %) (p< 0.0001). The majority (52.7 %) of fetuses were diagnosed with Central Nervous System or Urinary System anomaly, whereas CHDs were the third more represented anomalies, probably because our unit is not the referral center for diagnosis and treatment of the CHDs.
Also among the cases diagnosed before 14 weeks, the majority were CNS or Urinary system anomalies, but the specific diagnoses (i.e. anencephaly or megacystis) were often different from those detected in the subsequent periods. On the other hand, 61.5 % of the anomalies diagnosed after 23 weeks’ gestation were late onset anomalies (i.e. achondroplasia or autosomal dominant polycystic kidney disease) or anomalies with variable onset according to the specific cause (e.g. hydrocephalus, effusion, etc.). A relevant cause of late diagnosis was a disadvantaged social condition of the mother (20.5 %), whereas medical errors or gaps in the organization system were observed in a minority of cases (respectively, 10.3% and 7.7 %).
In the present study, at the end of diagnostic procedure, 40 anomalies were regarded as lethal (8 of them with a lethal chromosomal anomaly), whereas 75 were severe and 88 were moderate; the gestational age at diagnosis was significantly lower according to the increasing severity of the anomaly (p< 0.0001).
Also the pregnancy outcomes were significantly different according to the severity of the anomaly (p< 0.0001), as termination of the pregnancy was requested by 72.5 % of patients with a lethal fetal anomaly, 52.0 % of those with a severe anomaly and 8.0 % of those with a moderate anomaly. Among pregnancies intended to continue, the rate of intrauterine fetal death was higher for lethal anomalies (45.5 %) than for severe (17.1 %) and moderate (1.3 %) anomalies.
In conclusion, the accuracy of sonography in the diagnosis of fetal structural anomalies at different gestational ages may depend on a variety of factors: besides the resolution of the equipment used and the ability of the sonographic operators, the natural history of the specific anomaly and the organization of the sonographic network play an important and often unrecognized role.