Prenatal Medical Genetics: Screening, Diagnosis and Reproductive Counselling

Literature and regulatory information updated: 6 October 2026.

Medical Genetics Lecture — Prenatal Diagnosis and Reproductive Genetics

---

0. Initial Reference Table

Additional table notes: a dating ultrasound between 8 and 12 weeks, when crown–rump length (CRL) correlates most closely with gestational age, is recommended before any invasive procedure and before serial studies of growth or limb length [1]. Estimates of procedure-attributable pregnancy loss vary by source and study design: the meta-analysis by Akolekar and colleagues (2015) estimated weighted procedure-attributable risks of 0.11% for amniocentesis and 0.22% for CVS, neither statistically significant compared with background losses, whereas the values used for consent in several centres are approximately 0.5% (amniocentesis) and approximately 1% (CVS) [2]; the cardiogenetics chapter by Nannenberg and Arens reports 0.3% and 0.5%, respectively [3]. This variation is discussed in section 7.1.

---

1. Introduction

Prenatal genetics as practised today arose from the convergence of two lines of research that began independently in the early 1970s and, decades later, merged into a continuum of care extending from biochemical screening to sequencing of the complete fetal genome. The first began with the finding, published by Brock and Sutcliffe in the Lancet in 1972, that alpha-fetoprotein (AFP) was markedly elevated in the amniotic fluid of pregnancies complicated by anencephaly and spina bifida, establishing the basis for maternal serum biochemical screening for neural tube defects (NTDs) during that decade [4]. This foundational article first established the possibility of identifying a severe fetal malformation from a maternal sample, a principle that, extended decades later to cell-free fetal DNA, became the conceptual basis of contemporary non-invasive prenatal screening.

The second, later line of research was marked by Lo and colleagues' demonstration, also in the Lancet, in 1997, that DNA fragments of fetal origin (subsequently attributed to the placental trophoblast) circulate freely in maternal plasma and serum, detectable through Y chromosome-specific sequences in pregnancies with male fetuses [5]. This discovery, made using a relatively simple PCR technique on samples of only 10 microlitres of plasma, provided the foundation for the development over the following decade of the non-invasive prenatal testing (NIPT) and non-invasive prenatal diagnosis (NIPD) technologies now used routinely. Large-scale clinical validation of these approaches was consolidated through meta-analyses of diagnostic performance, including Gil and colleagues' synthesis of cfDNA screening performance for the main trisomies in tens of thousands of pregnancies [6], and prospective validation studies of the conventional combined test, such as Santorum and colleagues' study published in 2017 [7].

Over half a century, prenatal genetics therefore evolved from a single biochemical measurement in amniotic fluid to a range of methods including high-resolution ultrasound, combined first- and second-trimester biochemical tests, massively parallel sequencing of circulating cell-free fetal DNA, invasive cytogenetic techniques (amniocentesis, chorionic villus sampling and cordocentesis), array comparative genomic hybridisation (array-CGH) and, more recently, rapid exome and whole-genome sequencing for fetal structural anomalies without an established cytogenetic diagnosis [8]. In parallel, assisted reproductive medicine—in vitro fertilisation, intracytoplasmic sperm injection and preimplantation genetic testing—moved part of reproductive genetic counselling into the preconception period, allowing embryo selection before pregnancy is established.

In most circumstances, prenatal diagnosis is offered to give parents options for managing their reproductive risk; fetal treatment is possible for only a small number of conditions [9]. Next-generation sequencing-based NIPT for common aneuploidies has been commercially available since autumn 2011 and, by the reference edition of the Oxford Desk Reference, had already been performed in more than 500,000 pregnant women in more than 15 countries [10].

The clinical continuum proceeds from the biological and technological foundations of each testing modality through clinical indications, diagnostic algorithms and interpretation of complex results to reproductive options, population and ethical considerations, and the long-term prognosis of families receiving care from these services.

2. Molecular Genetics and Pathophysiology

2.0 Biological and Technological Foundations of Screening and Diagnostic Modalities

Prenatal genetics, unlike a single monogenic disorder, has no single reference gene or locus; its molecular basis is heterogeneous and depends on the testing modality used. Understanding the biological mechanism underlying each technique is essential to interpret its results and limitations correctly.

Maternal serum biochemical markers. The first-trimester combined test measures two placental glycoproteins—pregnancy-associated plasma protein A (PAPP-A) and the free beta fraction of human chorionic gonadotropin (free β-hCG)—whose concentrations deviate from the normal median (expressed as multiples of the median, MoM) in pregnancies affected by trisomies. In trisomy 21, PAPP-A tends to be reduced and free β-hCG elevated; distinct patterns of deviation characterise trisomies 18 and 13, reflecting differences in placental function associated with each aneuploidy [7]. The second-trimester quadruple test adds AFP and unconjugated oestriol (uE3) to total hCG and inhibin-A and is the screening test of choice when the first-trimester window has been missed.

Nuchal translucency and structural ultrasound. Nuchal translucency (NT) represents the transient physiological accumulation of fluid in the subcutaneous tissue of the fetal neck between 11 and 14 weeks; increased thickness correlates with lymphatic and cardiac dysfunction associated with aneuploidies and isolated congenital heart defects. The risk of a chromosomal abnormality and an adverse outcome increases non-linearly with thickness; in euploid fetuses, the risk of an adverse perinatal outcome becomes statistically significantly increased only from 3.5 mm (above the 99th percentile) and subsequently rises exponentially, but if the fetus survives to mid-gestation and the targeted ultrasound at 20–22 weeks is normal, this risk is no longer significantly increased [11]. Section 3.5 presents the distribution of outcomes by NT interval in Bilardo and colleagues' series. The second-trimester fetal anomaly scan (18–22 weeks), in contrast, depends on direct anatomical resolution rather than biochemical markers, identifying major structural defects (cardiac defects, neural tube defects, omphalocele and skeletal dysplasias) and soft markers whose probabilistic significance is discussed in section 6.

Circulating cell-free fetal DNA (cffDNA) and NIPT/NIPD. Cell-free fetal DNA—more precisely, DNA of placental/trophoblastic origin—enters the maternal circulation from approximately 6–7 weeks of gestation and is rapidly cleared after delivery, ensuring that the test is specific to the current pregnancy. This cffDNA coexists in maternal plasma with a predominant fraction of cell-free maternal DNA, typically accounting for 80% to 90% of the total; the minimum fetal fraction required for a reliable result is generally above 4% and is reduced in pregnant women with a high body mass index, a recognised technical limitation [12]. NIPT technology uses massively parallel (“shotgun”) sequencing of millions of cfDNA fragments, mapped to the reference genome and quantified by chromosome: a statistically significant excess of reads mapping to chromosome 21, for example, indicates trisomy 21. In Gil and colleagues' updated meta-analysis (35 validation or implementation studies published between 2011 and 2016, singleton pregnancies), weighted detection rates were 99.7% (95% CI 99.1–99.9%) for trisomy 21 (1963 affected cases; 223,932 unaffected), 97.9% (94.9–99.1%) for trisomy 18, 99.0% (65.8–100%) for trisomy 13 and 95.8% (70.3–99.5%) for monosomy X, with false-positive rates of 0.04% for each of the three trisomies and 0.14% for monosomy X; for other sex chromosome aneuploidies and twin pregnancies, case numbers were too small for precise assessment [6]. These values describe performance in studies with follow-up of more than 85% of pregnancies, rather than the positive predictive value of an individual result, which depends on the condition's prevalence in the tested population. NIPT remains a probability-based screening test rather than a confirmatory diagnostic test, a distinction reinforced in the latest American College of Medical Genetics and Genomics (ACMG) [13] and American College of Obstetricians and Gynecologists (ACOG) guidelines [14]. NIPD (non-invasive prenatal diagnosis) uses the same biological principle in a targeted manner: rather than quantifying overall chromosomal representation, it tests for the presence or absence of a specific paternal pathogenic variant (in recessive disorders) or fetal Y chromosome/RHD gene sequences, allowing, for example, early fetal sex determination from the sixth to seventh week in families at risk of an X-linked recessive disorder.

Biological properties of cffDNA relevant to interpretation. cffDNA derives mainly from apoptotic trophoblastic fragments and therefore displays placental methylation signatures; cell-free fetal RNA (cffRNA), described in 2000, appears to be packaged in microvesicles [10]. cffDNA has been detected in maternal plasma from the fourth week of gestation, and its concentration increases with gestational age; the maternal-to-fetal cell-free DNA ratio is approximately 60:1, and fetal DNA accounts for 10% to 20% of circulating cell-free DNA [10]. Clearance after delivery is biphasic, with a rapid phase having a mean half-life of approximately 1 hour and a slow phase having a mean half-life of approximately 13 hours; circulating fetal DNA disappears approximately 1 to 2 days after delivery, and cffDNA levels may be elevated in placental disorders such as pre-eclampsia and fetal growth restriction [10]. Fetal fragments are shorter than maternal fragments (a dominant peak at approximately 162 bp and a smaller peak at approximately 340 bp), with two implications: the fetal fraction can be enriched through size fractionation, and diagnosis of expansions, duplications or insertions larger than approximately 200 bp by this method may be unreliable [10]. Although cffDNA is detectable from approximately 4 weeks, aneuploidy testing is more reliable from 10 weeks, when fetal DNA quantity is consistently higher [10]; the 6–7-week value above refers to the period from which fetal sex determination is practised, while reliable sex determination using ffDNA is described from 8 weeks [1].

Methodological development of NIPT. Early approaches to non-invasive aneuploidy detection used allelic ratios of differentially methylated sequences (trisomy 18, Tong and colleagues, 2006), allelic ratios of single-nucleotide polymorphisms (SNPs) in placental PLAC4 mRNA, transcribed from chromosome 21 (Lo and colleagues, 2007), and SNP analysis to distinguish fetal from maternal DNA (Dhallan and colleagues, 2007) [10]. Counting single molecules, initially through digital PCR and, from 2008, through next-generation sequencing (Chiu and colleagues), enabled detection of overrepresentation of chromosome 21 molecules: the percentage of unique sequences mapping to each chromosome in the sample is compared with the mean in a euploid reference population and expressed as a z-score (z-score = [% chrN sample − mean % chrN reference] / standard deviation % chrN reference) [10]. All methods assessed in large-scale validation studies showed approximately 1 in 200 discordance between NIPT results and fetal results obtained through invasive testing [10]. Early reports of cffDNA sequencing documented sensitivities above 95% with very low false-positive rates, and the technique has potential applicability to other copy-number changes [9].

NIPD for monogenic disorders. NIPD is technically simpler when the variant is absent from the mother (paternally inherited or de novo) and is useful for molecular confirmation of conditions such as achondroplasia or thanatophoric dysplasia when ultrasound findings are suggestive; in autosomal recessive disorders, the presence of the maternal allele makes analysis more difficult, but the variant on the paternal allele can be detected. PCR, digital PCR and next-generation sequencing have been assessed for this purpose (for example, beta-thalassaemia and haemophilia) [10]. NIPD also applies to fetal RhD genotyping to guide immunoprophylaxis [10].

Performance of combined and integrated screening. Using maternal serum free β-hCG and PAPP-A with NT thickness at 12 weeks, the trisomy 21 detection rate was 88% at a screen-positive rate of 2% (Kagan and colleagues, 2008) [15]; combining first- and second-trimester results (integrated screening) can achieve a Down syndrome detection rate of 90% with fewer than 1% false positives [9]. In pregnancies screened using serum markers and ultrasound, serum screening combined with detailed ultrasound should detect more than 90% of pregnancies affected by open neural tube defects, the only multifactorial condition with specific population screening [9].

Conventional cytogenetics, QF-PCR and array-CGH. Invasive techniques (amniocentesis, CVS and cordocentesis) provide cells or tissue of genuine fetal origin, allowing direct chromosomal analysis. Quantitative fluorescent polymerase chain reaction (QF-PCR) analyses short tandem repeat markers specific to chromosomes 13, 18, 21, X and Y, producing a rapid result (24 to 72 hours) on their copy number without prior cell culture. Conventional karyotyping, which requires 10 to 14 days of cell culture, remains necessary to characterise balanced structural rearrangements (translocations and inversions) that QF-PCR does not detect. Array comparative genomic hybridisation (array-CGH) has become the standard method of chromosomal analysis when invasive testing is indicated, as it identifies submicroscopic microdeletions and microduplications undetectable by conventional karyotyping, although it does not detect balanced rearrangements or low-level mosaicism.

Exome and whole-genome sequencing. When ultrasound identifies a complex fetal structural anomaly and the karyotype and array-CGH are normal, whole exome sequencing (WES) can investigate pathogenic variants in genes associated with Mendelian malformation syndromes. The PAGE cohort study, published in the Lancet in 2019, demonstrated a relevant additional diagnostic yield of WES in fetuses with structural anomalies and normal conventional chromosomal tests, supporting its growing role, although limited by turnaround time and the complexity of interpreting variants of uncertain significance in the prenatal setting [8,16]. Detailed gene-by-gene characterisation of the hundreds of Mendelian conditions amenable to prenatal diagnosis cited throughout this lecture is available through the continuously curated GeneReviews and Orphanet databases, updated independently of professional society guidelines [17,18].

3. Clinical Spectrum and Manifestations by System

Spectrum of Clinical Applications by Modality

3.1 Ultrasound throughout pregnancy

Obstetric ultrasound serves progressively expanding functions as gestational age increases and fetal anatomy develops. Up to 14 weeks, the examination focuses on viability and basic body-plan parameters: number of embryos, embryofetal vitality, presence of the cephalic pole, assessment of the spine and presence of limbs; NT is measured during this period. Between 17 and 20 weeks, assessment extends to the face, abdominal wall, diaphragm, stomach, bladder, external genitalia and more detailed examination of the limbs. Between 20 and 24 weeks—the period of the conventional fetal anomaly scan—the cerebral structures, chest and heart, abdomen and digestive tract, and kidneys are assessed, with further refinement of limb examination. Between 28 and 32 weeks, the focus shifts to late skeletal anomalies, investigation of skeletal dysplasias (dwarfism) and identification of fetal ovarian cysts.

Structural findings particularly relevant to suspected chromosomal abnormality include fetal hydrops unrelated to Rhesus isoimmunisation, diaphragmatic hernia, anencephaly, gastroschisis, encephalocele and meningocele, cardiac hypoplasia (especially of the left ventricle), obstructive uropathy with the associated Potter sequence, and femoral agenesis with limb shortening. Certain combinations are particularly suggestive of a specific trisomy: hydrocephalus with an intraventricular communication, the “double bubble” sign of duodenal atresia and polyhydramnios associated with intrauterine growth restriction suggest trisomy 21; predominantly left-sided cardiac hypoplasia, omphalocele, Dandy–Walker malformation and a choroid plexus cyst associated with enlargement of the cisterna magna are more characteristic of trisomy 18, whose rocker-bottom foot is another classic, although non-specific, sign. The following table summarises the correspondence between ultrasound findings and the chromosomal abnormalities most likely to be associated with them:

When indicated, fetal magnetic resonance imaging, available from 24 weeks, complements three-dimensional ultrasound in the detailed characterisation of central nervous system anomalies, involving multidisciplinary fetal medicine and neuroradiology teams to establish prognosis.

The limitations of ultrasound should be discussed with the couple: it detects many, but not all, major structural anomalies—isolated cleft palate without cleft lip and alveolus is very difficult to diagnose, an atrial septal defect is often difficult to identify, and ultrasound signs of some anomalies appear late (hydrocephalus, microcephaly, achondroplasia and some renal anomalies, such as autosomal dominant polycystic kidney disease); ultrasound provides no information on cognition, vision or hearing except in the presence of gross structural central nervous system defects [1]. In the Eurofetus study, which assessed ultrasound detection of fetal malformations in an unselected population, the overall detection rate was approximately 60%, meaning that 40% of defects were not identified [9]. When there is a previous history of specific structural anomalies, ultrasound should be scheduled with a fetal medicine specialist, because the appropriate gestational age varies with the condition, as shown in the following table (adapted from Pathak and Lees, 2009) [1]:

Routine obstetric ultrasound includes at least the four-chamber and outflow tract views; suspected congenital heart disease or increased risk (family history or increased NT) warrants specialist fetal echocardiography, including ventricular outflow tract views, Doppler assessment of valvular, aortic and pulmonary arterial flow, and visualisation of the pulmonary veins [1]. Up to 11–12 weeks, the external genitalia appear similar in both sexes (with a difference in phallic orientation), allowing sex determination with 90% accuracy at 13–14 weeks, although dependent on image quality and fetal position [1]. Three-dimensional ultrasound may help assess facial features, the brain, and hand and foot anomalies [1]. Fetal MRI requires ultrafast technology to minimise movement artefacts and is used mainly to delineate intracranial anomalies, particularly neuronal migration defects, and in pregnancies at high risk of central nervous system malformations; the prospective multicentre MERIDIAN study showed that in utero MRI can improve diagnostic accuracy and confidence for fetal brain anomalies, although data on the limits of normality in specific pregnancies remain limited [1].

3.2 Amniocentesis

Amniocentesis involves transabdominal aspiration, under ultrasound guidance, of 10 to 20 mL of amniotic fluid from the fifteenth week of gestation. After centrifugation, the cell pellet—derived from the amnion, fetal skin and urinary tract epithelium—is cultured to obtain analysable metaphases, a process taking approximately 14 days; use of the supernatant, formerly used for AFP measurement, has largely been replaced by ultrasound in neural tube defect investigation. QF-PCR provides a preliminary result for common aneuploidies in approximately three days, with full karyotyping and array-CGH reserved for confirmation and detection of structural rearrangements. Beyond cytogenetics, amniocentesis allows a range of biochemical and molecular analyses, summarised in the following table.

The procedure-attributable miscarriage risk ranges from 0.5% to 1%; early amniocentesis (12–14 weeks) is technically possible but seldom practised because of the low amniotic fluid volume at that stage and a pregnancy loss risk similar to the conventional procedure. Classical indications include advanced maternal age (35–37 years or older), a family history or previous child with a chromosomal abnormality, a family or fetal history of a neural tube defect, a previous child with multiple congenital anomalies, and a fetal structural abnormality identified by ultrasound.

In the technique described in the Oxford Desk Reference, a 22-gauge spinal needle is introduced transabdominally under continuous ultrasound guidance, usually aspirating approximately 15 mL of amniotic fluid; continuous ultrasound guidance, operator experience and gestational age are the strongest determinants of post-procedure loss risk [2]. Post-procedure loss in Tabor and colleagues' 11-year Danish national registry was 1.4%, whereas the meta-analysis by Akolekar and colleagues (2015) estimated an attributable risk of 0.11%; 0.5% was the value used for counselling at the centre where that study was conducted [2]. A retrospective case–control study of 1296 children born after amniocentesis showed no increase in registrable disability (hearing impairment, learning difficulties, visual problems and limb anomalies) over 7 to 18 years of follow-up [2]. In a retrospective analysis of more than 10,000 amniotic fluid samples (O'Brien and colleagues, 2016), karyotyping failed in 2.3% (0.3% at 16 weeks to 43.4% at 36–40 weeks), QF-PCR in 2.3% (from 1.9% at 16 weeks to 7.1% at 36 weeks) and array-CGH in 3.4% [2]. The Canadian CEMAT trial showed that early amniocentesis (before 13 weeks) is associated with increased fetal loss and talipes equinovarus (1.3% versus 0.1%) compared with second-trimester amniocentesis, leading to a recommendation to discontinue the early procedure [2]. Second-trimester amniocentesis is usually performed between 15 and 17 weeks, although the Oxford Desk Reference gives 16 weeks as the lower limit [2]; Smith's Recognizable Patterns of Human Malformation places it between 15 and 18 weeks, with a miscarriage risk of approximately 1 in 400 [9].

3.3 Chorionic villus sampling

Chorionic villus sampling (CVS) brings diagnosis forward to the first trimester (11 to 13 weeks and 6 days), aspirating trophoblastic tissue—of genuine fetal origin—most commonly through a transabdominal route, or transcervically. After removal of contaminating maternal decidua, part of the sample is cultured and DNA is extracted directly from the remainder for mutation or risk-haplotype analysis. Pregnancy loss risk is approximately 1%, decreasing in the hands of experienced operators. The procedure is particularly indicated when there is a known familial risk of a specific metabolic or molecular disorder, allowing direct diagnosis through enzyme, biochemical or DNA analysis rather than relying solely on probabilistic screening. Metabolic disorders most frequently investigated through CVS include organic acidurias (argininosuccinic, methylmalonic and propionic), leukodystrophies (adrenoleukodystrophy, Krabbe disease and metachromatic leukodystrophy), sphingolipidoses (Tay–Sachs, Sandhoff, all three Gaucher subtypes, Fabry and Niemann–Pick A-B), mucopolysaccharidoses (Hurler, Hunter and Morquio) and mucolipidoses, among many other rarer lysosomal and peroxisomal conditions, such as Zellweger syndrome and Menkes disease. Direct molecular applications include cystic fibrosis, alpha- and beta-thalassaemia, Duchenne and Becker muscular dystrophies, haemophilia A and B, 21-hydroxylase deficiency, OTC deficiency, phenylketonuria, fragile X syndrome and osteogenesis imperfecta; CVS also allows PCR testing for congenital infections such as toxoplasmosis, rubella, cytomegalovirus, parvovirus B19, herpes simplex, adenovirus and Coxsackie virus. This list, drawn from classical indication protocols, is not exhaustive and should always be adapted to the specific family context.

CVS was introduced into UK clinical practice in the 1980s; in the second and third trimesters, the same technique is termed placental biopsy [2]. The trophoblast and embryo derive from the same fertilised egg and therefore share the same genetic constitution, supporting CVS as an indirect means of determining fetal genetic constitution; as a highly cellular tissue, it is the sample of choice for many molecular genetic investigations (mean weight 15.2 ± 6.0 mg in one series; larger samples may be needed for biochemical enzyme diagnosis in uncultured material) [2]. In the early 1990s, two clusters of newborns with limb defects after CVS were reported, raising the possibility of a causal association between early CVS and transverse limb deficiency; risk extends throughout limb morphogenesis, falling from 10 to 20 times background levels at 9 weeks or earlier to near-background levels (or only a few times higher) from 11 weeks. A review by Stoler and colleagues (1999) also found increased vascular disruption defects (gastroschisis, intestinal atresias and clubfoot) in children exposed to CVS [2]. The following table compares CVS and amniocentesis according to the Oxford Desk Reference (turnaround times are indicative and should be confirmed with the local laboratory) [2]:

Comparing the two techniques, the 2000 Cochrane review concluded that excess miscarriages after CVS appeared procedure-related and that second-trimester amniocentesis appeared safer; in Akolekar and colleagues' meta-analysis (324 losses in 42,716 amniocenteses and 207 losses in 8899 CVS procedures), miscarriage risk before 24 weeks was 0.81% after amniocentesis and 2.18% after CVS, with background rates in controls of 0.67% and 1.79%, respectively [2]. Uncertain results (for example, mosaicism) are more frequent with CVS than with second-trimester amniocentesis; the benefits of earlier diagnosis should therefore be weighed against this risk and the small additional pregnancy loss risk [2]. Nannenberg and Arens' cardiogenetics chapter describes CVS between 10 and 13 weeks through an abdominal or cervical route, and amniocentesis from 15 weeks [3].

Multiple pregnancies. First-trimester ultrasound is necessary to establish chorionicity (the presence or absence of the lambda sign allowed correct assignment in all cases in a series of more than 200 multiple pregnancies) and thereby estimate zygosity: in a monozygotic pregnancy, the risk of both twins being affected approaches the risk in a singleton pregnancy, whereas in a dizygotic pregnancy each twin is assessed independently, as siblings [2]. When an invasive procedure is indicated in twins, CVS allows selective termination, if required, in the first trimester, with a lower pregnancy loss risk than in the second trimester (when selective termination carries a 5% to 10% risk of loss or very preterm delivery), but produces ambiguous results in up to 2% of cases; a review by Agarwal and Alfirevic (2012) estimated an excess risk of 1% for both CVS and amniocentesis in twin pregnancies [2].

3.4 Cordocentesis, fetoscopy and targeted fetal biopsies

Cordocentesis—puncture of an umbilical cord vessel, technically feasible from approximately 20 weeks—has become rare as a purely diagnostic tool because QF-PCR on CVS or amniotic fluid samples provides rapid cytogenetic results through less invasive approaches. It nevertheless retains a relevant role in obtaining a rapid karyotype at advanced gestational ages, resolving ambiguous mosaicism identified in other samples and, above all, providing a therapeutic route for intrauterine red cell or platelet transfusions in Rhesus isoimmunisation or non-immune fetal hydrops, direct administration of fetal medication (for example, digoxin for fetal arrhythmias) and monitoring of transplacental maternal treatment. The associated fetal loss risk is approximately 1% to 2%.

Fetoscopy has largely been replaced by high-resolution ultrasound as a diagnostic tool but retains therapeutic indications in specific circumstances—insertion of vesicoamniotic shunts for posterior urethral valves, laser division of amniotic bands and laser ablation of placental vascular anastomoses in twin-to-twin transfusion syndrome—always in specialist centres because of the substantial risk of fetal loss or preterm delivery inherent in direct intrauterine instrumentation. Targeted skin biopsies (for investigation of epidermolysis bullosa or anhidrotic ectodermal dysplasia), liver biopsies (for X-linked enzyme deficiencies, glycogen storage disorders, urea cycle defects and phenylketonuria), and aspiration of serous or cystic collections (ascites, pleural effusion, hydrops, pulmonary or renal cysts) complete the range of targeted invasive procedures, now reserved for highly selected situations.

3.5 Specific ultrasound findings with genetic implications

Increased nuchal translucency, cystic hygroma and hydrops. The mechanism underlying nuchal oedema, cystic hygromas or hydrops in chromosomally abnormal fetuses is not fully understood; cystic hygroma appears to result from lymphatic dysplasia and is particularly associated with 45,X monosomy (Turner syndrome) [15]. Increased NT may resolve spontaneously even in pregnancies with an abnormal karyotype, and fetal tissue distension may leave residual dysmorphic features (webbed neck, nuchal folds, hypertelorism, epicanthic folds, low-set ears and widely spaced nipples) [15]. A value of 3 mm corresponds to the 95th percentile, and many fetuses with mildly increased NT are normal at birth [1]. Fetal echocardiography is recommended in all cases with NT of 3.5 mm or greater: in a review of six studies, the mean prevalence of major cardiac defects in chromosomally normal fetuses with NT ≥3.5 mm was 6.27% (3.1–12.2%) [15]. Some cardiac defects, such as aortic coarctation and hypoplastic left heart, may be sequelae of fetal lymphoedema [15]. Outcomes in euploid fetuses with increased NT (Bilardo and colleagues, 2010; 451 cases) are summarised in the following table; if the 20-week fetal anomaly scan is entirely normal and NT has resolved completely, risk returns to baseline [15].

*For the 5.5–6.4 mm row, the table reproduced in the Oxford Desk Reference gives only two values (42% and 58%); the distribution of the 42% between fetal death and structural/genetic anomalies is not specified in the source consulted.

In fetal cystic hygroma (septated or non-septated, both associated with aneuploidy), a series of 72 cases (Graesslin, 2007) found a chromosomal abnormality in 52.7% (37% with trisomy 21), and only 22% of the total had no detectable problems; in another series of 42 cases (Ganapathy, 2004), 47.6% had aneuploidy (55% Turner syndrome, 20% trisomy 21), and 29.4% of euploid fetuses had a major cardiac defect, with hygroma resolution having no diagnostic value. In Boyd's series (1996; 92 cases), 89% of continued pregnancies with normal chromosomes and no additional risk factors were normal at 1 to 5 years of follow-up [15]. Fetal hydrops is defined ultrasonographically as skin oedema greater than 5 mm with other serous effusions (ascites, pericardial effusion, hydrothorax) and/or polyhydramnios and/or placental thickening greater than 6 cm; non-immune hydrops accounts for 90% of cases in the Western world and is a non-specific terminal state of multiple disorders, with a high pregnancy loss rate [15]. The most common causes in a systematic review of 5437 cases (Bellini and colleagues, 2009) were [15]:

Several metabolic disorders may present in utero with hydrops—congenital disorders of glycosylation, Farber disease, galactosialidosis, Gaucher disease, glycogen storage disease IV, GM1 gangliosidosis, mucolipidosis II (I-cell disease), mucopolysaccharidoses I, IV and VII (with amniotic fluid glycosaminoglycan measurement), multiple sulfatase deficiency, Niemann–Pick A and C, Salla disease and sialic acid storage disease, sialidosis, Wolman disease and, among mitochondrial disorders, Pearson syndrome—and some specialist centres offer a lysosomal hydrops screen in amniotic fluid and cultured amniocytes [15].

In non-immune fetal hydrops without an identified cause after routine assessment, prenatal exome sequencing has a demonstrated diagnostic yield: in a series of 127 consecutive cases of unexplained non-immune hydrops (defined by ascites, pleural or pericardial effusions, skin oedema, cystic hygroma, increased NT or a combination of these findings), Sparks and colleagues (2020) identified diagnostic variants (pathogenic or likely pathogenic under ACMG criteria) in 37 cases (29%); RASopathies (RAS-MAPK signalling pathway disorders) accounted for 30% of genetic diagnoses, inborn errors of metabolism and musculoskeletal disorders for 11% each, and lymphatic, neurodevelopmental, cardiovascular and haematological disorders for 8% each. Of the diagnoses, 68% (25/37) were autosomal dominant (88% de novo and 12% inherited) and 27% (10/37) autosomal recessive (95% inherited), with 1 inherited X-linked recessive case and potentially diagnostic variants in another 12 cases [19]. Among monogenic haematological causes, germline GATA1 variants (GATA1-related cytopenia, with X-linked inheritance) cause anaemia of variable severity, whose severe end comprises fetal hydrops requiring intrauterine transfusion [20].

Echogenic bowel. Bowel is classified as echogenic when its echogenicity equals that of the iliac crests; the finding occurs in approximately 0.8% of pregnancies in the second or third trimester and may result from intra-amniotic bleeding with swallowing of blood pigments, hypoperistalsis (for example, trisomy 21) or reduced meconium fluid content (cystic fibrosis, growth restriction) [21]. In a French multicentre study of 682 fetuses (Simon-Bouy, 2003), 65.5% had a normal outcome, 3.5% a significant chromosomal abnormality, 3% cystic fibrosis and 2.8% a viral infection (15 CMV and 4 parvovirus), with prematurity in 6.2%, growth restriction in 4.1% and intrauterine death in 1.9%; in an Austrian study of 97 isolated cases, growth restriction occurred in 9.9% (versus 1.3% in controls) and intrauterine death in 8.9% (versus 0.5%) [21]. Approximate frequencies of the main causes are: chromosomal abnormality, predominantly trisomy 21, 3–5% (lower after low-risk trisomy 21 screening); cystic fibrosis approximately 3%; CMV approximately 2–3%; gastrointestinal anomaly approximately 3%; and adverse pregnancy outcome (growth restriction, intrauterine death) approximately 7–10% [21].

Short limbs. Femoral length is routinely measured at the 18–20-week fetal anomaly scan; a short femur may reflect a dating error, a normal small fetus, intrauterine growth restriction (which would need to be very severe to be detected at 20 weeks), a chromosomal disorder, a skeletal dysplasia or a malformation syndrome with growth restriction or limb reduction [22]. Suspected skeletal dysplasia requires assessment of long bone length and shortening pattern (proximal, distal or both), epiphyseal stippling, bowing, mineralisation and fractures, skull shape and mineralisation (cloverleaf skull in thanatophoric dysplasia), vertebrae, thorax (cardiothoracic ratio, rib length and fractures, thoracic/abdominal circumference ratio), hands and feet (trident hand, detectable from 25 weeks; clubfoot; polydactyly; hitchhiker thumb in diastrophic dysplasia), face and external genitalia [22]. Predictors of lethality (Parilla, 2003) include early severe long bone shortening, a femoral length/abdominal circumference ratio below 0.16, a small thorax with short ribs and a high cardiothoracic ratio, and marked bowing or fractures; in that series of 31 prenatally detected cases, a final diagnosis was possible in 80% [22]. Most fetuses with achondroplasia show no significant shortening at 20 weeks (limb lengths usually at or above the fifth percentile until 24 weeks), and prenatal confirmation of FGFR3-related dysplasias can be obtained through NIPD [22]. The ACMG guidelines for prenatal diagnosis of skeletal dysplasias (Krakow, Lachman and Rimoin, 2009) add that skeletal dysplasia should be strongly suspected in fetuses with long bone measurements more than 3 standard deviations below the mean, particularly if head circumference exceeds the 75th percentile; that the femur/foot ratio approaches 1.0 throughout gestation and is below 1 in dysplasias with rhizomelia; that a thoracic/abdominal circumference ratio below 0.6, like a femoral length/abdominal circumference ratio below 0.16, strongly suggests a perinatally lethal dysplasia (particularly because of pulmonary hypoplasia), with exceptions; that the accuracy of prenatal diagnosis of these dysplasias through routine ultrasound approaches 40%, so all prenatally diagnosed cases should receive a final diagnosis through specialist clinical and radiological assessment, with autopsy and growth plate cartilage histology when available and storage of tissue or DNA; and that molecular testing should be offered in pregnancies at risk of homozygosity or compound heterozygosity, ideally with both parents' variants identified before pregnancy [23].

4. Clinical Recognition and Warning Signs

4.0 Warning Signs for Referral

Referral for prenatal or preconception genetic counselling is prompted by a recognisable set of warning signs, which can be organised to support effective screening in obstetric practice.

Advanced maternal age remains the most commonly cited criterion, reflecting the exponential rather than linear increase in trisomy 21 risk with age: from approximately 1 in 1500 births at age 20 years, risk rises to approximately 1 in 900 at 30, 1 in 400 at 35, 1 in 100 at 40 and approximately 1 in 30 at 45, according to the classical estimates of Cuckle, Wald and Thompson [24]. Couples should be reminded that trisomy 21 incidence at CVS (11–12 weeks) and amniocentesis (16 weeks) is consistently higher than at birth, because some of these pregnancies end spontaneously in fetal loss during the second and third trimesters—a relevant consideration in reproductive prognosis counselling [25].

The values above, from Cuckle, Wald and Thompson, are close to the maternal age-specific probabilities of a live birth with Down syndrome reviewed by Morris and colleagues (2002), whose rounded probabilities for counselling are: 1:1450 at age 20 years, 1:1350 at 25, 1:950 at 30, 1:350 at 35, 1:260 at 36, 1:200 at 37, 1:150 at 38, 1:110 at 39, 1:85 at 40, 1:55 at 42 and 1:35 at 45 [26]. Approximately 43% of affected pregnancies are lost spontaneously between CVS and term, and approximately 23% between amniocentesis and term [26]. The overall risk of any aneuploidy (including age-dependent +21, +18, +13, 47,XXX and 47,XXY, and 45,X, 47,XYY and triploidy, which are unaffected by maternal age) is approximately twice the risk of Down syndrome [26]. Meiosis I nondisjunction is the main age-dependent factor; advanced maternal age also increases the risk of subfertility, miscarriage (30–40% early pregnancy loss at age 40 years), dizygotic twinning and, to a lesser extent than paternal age, de novo dominant variants. In a study of more than 1000 cleavage-stage embryos, aneuploidy increased from 3.1% (ages 20–34 years) to 17% (40 years or older) [26].

Advanced paternal age. The association between advanced paternal age and monogenic conditions such as achondroplasia and Apert syndrome has been recognised since 1957; the mechanism involves gain-of-function variants in FGFR2 (Apert, Crouzon, Pfeiffer), FGFR3 (thanatophoric dysplasia, achondroplasia), PTPN11 (Noonan), HRAS (Costello) and RET (MEN2A, MEN2B) that confer a selective advantage and clonal expansion on spermatogonia [27]. Genome sequencing studies show that approximately 80% of de novo variants are paternal in origin, with each year of paternal age adding an average of 1 to 2 variants; there are associations with autism, schizophrenia, bipolar disorder and developmental delay, but little evidence of increased aneuploidy [27]. Without a family history, no specific additional testing related to paternal age is recommended (Toriello and Meck, ACMG 2008), and the germline mosaicism recurrence risk for age-related variants is substantially below 1% [27].

A family or personal history of a chromosomal abnormality is another central warning sign. A couple with a previous child with Down syndrome, without a structural rearrangement, has a modestly increased empirical recurrence risk, typically approximately 1%, above the baseline risk for maternal age; when one parent carries a balanced chromosomal rearrangement (Robertsonian or reciprocal translocation), the risk of recurrent chromosomal imbalance in offspring may range from 1–2% to 15–20%, depending on the specific rearrangement and chromosomal segments involved.

According to Smith's Recognizable Patterns of Human Malformation, amniocentesis or CVS should be offered in the following circumstances: (1) abnormal combined ultrasound and biochemical screening for autosomal trisomies; (2) a previous child with trisomy 21 or another trisomy; (3) a parental balanced translocation; (4) a parent affected by a microdeletion or microduplication syndrome; (5) any de novo abnormality in a previous child where the chromosomally normal parents wish to exclude recurrence through gonadal mosaicism; (6) isolated or multiple ultrasound defects or markers requiring more precise definition of fetal prognosis; (7) parental preference to test for as many genomic conditions as possible [9].

A family history of a known monogenic disorder, particularly one that is severe or limits life expectancy and quality of life, warrants preconception counselling, ideally before pregnancy, to allow timely identification of the familial pathogenic variant and discussion of all available options, including preimplantation diagnosis. Communities with organised carrier screening, such as certain Orthodox Jewish communities for Tay–Sachs disease, illustrate the value of targeted preconception screening; however, most at-risk couples seek assessment only after pregnancy has begun, sometimes limiting available clinical options.

Abnormal ultrasound findings, whether major structural anomalies, persistent soft markers or NT of 3.5 mm or greater, directly indicate invasive investigation (QF-PCR and array-CGH), irrespective of previous biochemical screening results.

Other high-risk factors include parental consanguinity, which increases the risk of a shared rare recessive disorder; carrier screening for recessive disorders common in the couple's ethnic group or family history; adverse obstetric history, particularly three or more unexplained miscarriages, warranting cytogenetic analysis of products of conception and, if necessary, the parents; and poorly controlled chronic maternal disease, such as pregestational diabetes, associated with an increased risk of fetal structural anomalies and warranting detailed targeted ultrasound.

Regarding recurrent miscarriage, 25% to 30% of all pregnancies end in early pregnancy loss, and at least 50% of miscarriages have a chromosomal abnormality (60–70% with microarray analysis of chorionic tissue); the most common abnormalities are autosomal trisomies (50% of chromosomal abnormalities, with trisomy 16 the most frequent, accounting for 30% of trisomies), triploidy (15%), 45,X monosomy (10%), tetraploidy (5%) and unbalanced structural abnormalities (4%, half de novo) [28]. In a large Quebec study of more than 22,000 couples with two or more miscarriages, 4.7% had one member carrying a chromosomal rearrangement (versus 0.5% at baseline); Barber and colleagues identified 1.9% balanced rearrangements in 20,432 parents and questioned the cost-effectiveness of systematic karyotyping, and the Royal College of Obstetricians and Gynaecologists (2011) recommended parental karyotyping only after detection of chromosomal imbalance in products of conception [28]. Recurrent miscarriage with normal parental chromosomes does not, by itself, indicate a genetics consultation; if the pregnancy loss is aneuploid, the probability of success in the next pregnancy is higher than after a euploid loss (68% versus 41% subsequent live births) [28].

5. Diagnosis: Criteria, Tests and Algorithms

The diagnostic pathway in prenatal genetics follows a sequence of screening followed by confirmation, each stage reducing uncertainty without generally eliminating it completely before the next.

Stage 1 — First- or second-trimester screening. In pregnancies with a low prior risk, the most widely adopted initial strategy internationally is first-trimester combined screening (maternal age, PAPP-A, free β-hCG and NT), with prospectively validated accuracy demonstrating high overall sensitivities for trisomies 21, 18 and 13 and a controlled false-positive rate [7]. Recent ACOG (2020, reaffirmed and updated in subsequent reviews) and ACMG (2023) guidelines have recommended offering cfDNA screening to all pregnant women, irrespective of maternal age or previous risk factors, because of its superior sensitivity and specificity compared with conventional biochemical methods [13,14]. In the United States, this shift from risk-based eligibility to universal offering was formalised in ACOG Practice Bulletin No. 226 of 2020 [14]. Nevertheless, combined serum marker and NT screening remains clinically relevant, particularly because it can indicate increased risk of pre-eclampsia and other placental complications, and because cfDNA cost coverage varies with the healthcare system and the couple's insurance; the peer-reviewed sources consulted do not quantify this coverage, and circulating estimates derive mainly from commercial laboratory announcements.

Guideline updates (2023–2026). The 2023 ACMG guideline, based on the ACMG's own systematic review and developed using the GRADE Evidence to Decision framework, strongly recommends cfDNA screening rather than traditional methods for trisomies 21, 18 and 13 in all pregnant women with singleton or twin pregnancies, and strongly recommends offering sex chromosome aneuploidy screening; detection of rare autosomal trisomies and microdeletions is described as an emerging field [13]. In November 2025, the Society for Maternal-Fetal Medicine (SMFM) published Consult Series No. 74, replacing No. 36: it recommends that cfDNA screening for trisomies 21, 18 and 13 be routinely available to all pregnant women (GRADE 1B) and considers it the most sensitive and specific screening test in any population, reaffirming the right to accept or decline screening and diagnosis; sex chromosome aneuploidy screening should be an opt-in choice following counselling (1C); it does not recommend routine population microdeletion screening, including 22q11.2 outside an informed choice, or routine genome-wide screening for large copy-number variants (1C), and recommends diagnostic testing for those seeking CNV information. It recommends cfDNA as first-line screening for trisomy 21 in twin pregnancies (1B) and considers it appropriate for trisomies 18 and 13 in that setting, but not for sex chromosome aneuploidies in twins or higher-order pregnancies (1C) [29]. In 2025, the ACMG also published a laboratory technical standard on biochemical and ultrasound screening for trisomy 21 (first- and second-trimester, integrated, contingent and reflex), confirming that these methods remain in use and require their own quality control [30]. A Brazilian integrative review of 2026, comparing nine international guidelines published through December 2025, concluded that most societies recommend offering aneuploidy screening to all pregnant women, describe NIPT as the most sensitive screening method but not diagnostic, retain amniocentesis and CVS as the reference for confirmation, generally recommend microarray for structural anomalies and restrict exome sequencing to selected settings, with continuing differences in implementation [31].

Performance in a routine screening population and national implementation. In the prospective multicentre NEXT study (35 international centres; 15,841 analysable pregnant women, mean age 30.7 years, tested at 10–14 weeks), cfDNA detected 38 of 38 trisomy 21 cases (100%; 95% CI 90.7–100), compared with 30 of 38 (78.9%) using combined screening, with false-positive rates of 0.06% and 5.4% and positive predictive values of 80.9% and 3.4%, respectively; the study was funded by the test manufacturer [32]. In the Netherlands, the national TRIDENT-2 study offered genome-wide NIPT as first-line screening to all pregnant women from April 2017: in the first year, 73,239 pregnancies were tested (42% of all pregnancies), 4% chose the combined test and 54% did not participate; positive predictive values were 96% for trisomy 21, 98% for trisomy 18 and 53% for trisomy 13. Among the 78% of women requesting additional findings, 207 findings occurred (0.36%): other trisomies (0.18%; positive predictive value 6%, many probably confined placental mosaicism), structural aberrations (0.16%; positive predictive value 32%) and complex profiles suggestive of maternal malignancy (0.02%; positive predictive value 64%) [33]. These data illustrate that positive predictive value varies considerably by condition and that expanding the panel increases discordant results.

Stage 2 — Interpretation of screening results. A “high-risk” combined test or cfDNA result is not a diagnosis; it makes the pregnant woman eligible for genetic counselling and an offer of confirmatory diagnostic testing. Couples must understand that, despite its high sensitivity, cfDNA retains a residual rate of false-positive and inconclusive results (particularly with a low fetal fraction, maternal obesity, multiple pregnancies or confined placental mosaicism), so definitive clinical decisions, including whether to continue the pregnancy, should never rest solely on a screening result [13,14].

Stage 3 — Invasive diagnostic confirmation. After high-risk screening, an ultrasound structural anomaly or a direct indication (advanced maternal age where screening is unavailable, relevant family history), amniocentesis or CVS is performed according to gestational age [34], with QF-PCR for a rapid result for common aneuploidies and array-CGH as the standard method of chromosomal characterisation; full karyotyping is reserved for confirmation of suspected structural rearrangements [35]. The International Society of Ultrasound in Obstetrics and Gynecology (ISUOG) practice guidelines on invasive procedures, published in 2016, organise the technical criteria, periprocedural care and risk communication to couples at this stage [35].

Stage 4 — Molecular investigation in selected circumstances. When a complex fetal structural anomaly remains unexplained after normal karyotyping and array-CGH, rapid prenatal exome sequencing (results in weeks rather than months) may be considered, particularly for multiple anomalies or phenotypes suggestive of a specific Mendelian syndrome, with a relevant additional diagnostic yield according to the PAGE cohort study [8]. This stage requires prior specialist genetic counselling because of the risk of incidental findings and variants of uncertain significance inherent in the technology.

Laboratory prerequisites for molecular prenatal diagnosis. A maternal DNA sample must be available for all PCR-based molecular analyses to exclude maternal cell contamination; in autosomal recessive disorders, each parent's carrier status must be established before prenatal diagnosis; in X-linked recessive disorders, the mother's carrier status must be established (accounting for germline mosaicism risk); and PCR diagnosis of large deletions (for example, spinal muscular atrophy and congenital adrenal hyperplasia) should combine linkage analysis with variant detection [2]. Maternal blood group must be known before any invasive procedure, and anti-D prophylaxis should be offered to all Rh-negative women; counselling should explain that a normal karyotype does not guarantee a normal newborn and that unexpected or difficult-to-interpret results may arise (sex chromosome aneuploidies, mosaicism) [2].

Strategies for prenatal diagnosis of monogenic disorders. The approach depends on knowledge of the condition's molecular basis; testing may target the gene (DNA), message (RNA), product (biochemical analysis) or phenotype (morphology), and the literature should be reviewed when testing is requested because of rapid developments in this field [9]: (1) when the familial variant is known, direct analysis in cells obtained through amniocentesis or CVS is the usual approach, straightforward when a common variant explains most cases (achondroplasia) and possible only after identifying the familial variant in conditions with allelic heterogeneity (Marfan syndrome); (2) when only the gene's location is known, linkage analysis requires samples from several relatives, may reveal non-paternity and carries a recombination-related error risk, requiring multiple intragenic or nearby markers; (3) biochemical studies of the gene product may be performed directly in amniotic fluid or require cultured fetal cells; (4) in X-linked conditions without available direct or linkage analysis, fetal sex determination may be an option, although 50% of male fetuses of carriers are normal; (5) in conditions diagnosed by phenotype, diagnosis depends on ultrasound visualisation of specific features (for example, severe limb shortening in skeletal dysplasias); (6) in conditions with variable expressivity or incomplete penetrance, the variant may not predict phenotype—in EEC syndrome due to TP63 variants, ultrasound identification of cleft lip and palate or ectrodactyly may be more useful than molecular testing alone; (7) for ultrasound anomalies, next-generation sequencing panels targeted to groups of defects (for example, a skeletal dysplasia panel or a Noonan syndrome panel for increased NT) [9]. In multifactorial conditions without a chromosomal or genetic marker, prenatal diagnosis depends entirely on ultrasound visualisation: holoprosencephaly is readily visualised, whereas isolated cleft palate is not always detected [9]. When familial risk concerns a specific monogenic disorder with a known variant, the joint 2026 ESHG and ISPD recommendations indicate gene-targeted testing rather than broader genomic analysis [36].

Indications and limitations of genomic analysis for ultrasound anomalies. Some anomalies, when isolated on detailed ultrasound, do not warrant consideration of karyotyping—gastroschisis, unilateral multicystic dysplastic kidney and clubfoot—whereas most others confer increased chromosomal abnormality risk. The need for genetic investigation should be assessed individually, considering other anomalies, the detail of ultrasound assessment and additional risk factors (maternal age, NT, serum screening) [1]. Several studies established a higher detection rate for prenatal array analysis than for karyotyping, particularly with ultrasound anomalies, also documenting significant detection of pathogenic copy-number variants (CNVs) in low-risk pregnancies; CNVs of uncertain significance pose an even greater counselling challenge prenatally, especially without ultrasound findings [9]. In products of conception, a meta-analysis (Dhillon, 2014) showed agreement between array and karyotyping in 86% of cases, additional detection of 13% abnormalities by array and 3% by karyotyping (for example, balanced translocations), with 2% variants of uncertain significance; triploidy is not detected by array-CGH alone [28].

Prenatal exome and genome sequencing: yield and indications. The two major prospective studies of 2019 defined the expected yield. In the UK PAGE study, after exclusion of aneuploidies and CNVs, 610 fetuses with structural anomalies underwent exome analysis (596 trios and 14 duos); a diagnostic variant was identified in 52 (8.5%; 95% CI 6.4–11.0) and a potentially useful variant of uncertain significance in another 24 (3.9%). Yield was 15.4% in multisystem anomalies (22/143), 11.1% in cardiac anomalies (9/81), 15.4% in skeletal anomalies (10/65) and only 3.2% in isolated NT ≥4.0 mm (3/93) [8]. In the Columbia study (New York), only 234 trios (45%) of 517 eligible couples were analysed, and exome sequencing identified diagnostic variants in 24 (10%), with suggestive but insufficient signatures for diagnosis in another 46 (20%) [37]. Yield therefore depends on the fetal phenotype, with isolated increased NT having the lowest yield. In its updated 2022 position statement, ISPD considered prenatal sequencing beneficial in fetuses with an isolated major anomaly or multisystem anomalies without a diagnosis after microarray, when review by a geneticist considers the phenotype suggestive of a genetic cause or when the pattern strongly suggests a monogenic disorder. Because exome sequencing had not been validated to detect all CNVs, microarray should be performed before or in parallel; ISPD also allowed recurrence of similar anomalies after a previous undiagnosed fetus and stated that there was no evidence for routine sequencing, even at parental request, in invasive samples without a fetal anomaly [38]. In 2020, the ACMG published a points-to-consider document on fetal exome use in prenatal diagnosis [39].

The global recommendations of the European Society of Human Genetics (ESHG) and ISPD, approved in June 2026 and published in September 2026, update this framework [36]. Indications remain a single major or multisystem anomaly suggestive of a genetic aetiology (beyond the usual microarray indications), recurrent anomalies after a previous undiagnosed fetus or child, patterns of minor anomalies strongly suggestive of a monogenic disorder, and recurrent stillbirths of unknown cause with a recurrent pattern in the current pregnancy; isolated findings such as increased NT, polyhydramnios or oligohydramnios and fetal growth restriction currently lack sufficient evidence. They advise against sequencing as screening for fetuses without ultrasound anomalies: in a systematic review of four studies involving 1916 phenotypically normal fetuses with normal karyotype or microarray, 1.6% had a pathogenic or likely pathogenic variant, and the probability of association with severe disease was 0.5%; nevertheless, this predictive sequencing is performed in some countries. They recommend trio analysis whenever possible (de novo variants explained 47% of diagnoses in one prenatal study), multidisciplinary discussion (fetal medicine, laboratory genetics and clinical genetics or genetic counselling), pretest counselling covering timelines and the local legal framework, and recognise the transition from capture methods to genome sequencing, with improved structural variant detection; parallel microarray may become unnecessary if CNVs are analysed in sequencing data. The most frequently reported turnaround time was 3 weeks, and rapid QF-PCR or FISH should precede sequencing.

Genetic test limitations and report content. No prenatal technique detects all classes of variation, and a negative result must be interpreted in relation to the method's coverage and the ultrasound phenotype. The following table summarises what each approach detects and what tends to escape detection [2,10,36]:

The report should unambiguously identify the pregnancy and sample type (villi, uncultured or cultured amniocytes), state maternal cell contamination controls, testing and analysis strategy (virtual panel, inheritance- or phenotype-based filtering), limitations and low-coverage regions, classification guidelines used (ACMG/AMP and subsequent specifications), and, for each variant, the gene, versioned reference transcript, normalised HGVS nomenclature, zygosity, parental origin and classification; parental results should be reported separately [36]. Examples illustrate normalisation: the Dutch PLN founder variant, published as c.40_42del (p.Arg14del), is recorded in ClinVar as NM_002667.5(PLN):c.37AGA[1] (p.Arg14del), the MANE Select transcript, with both notations representing the same AGA codon deletion in a repeat sequence; the classical achondroplasia variants “G1138A” and “G1138C” correspond to NM_000142.5(FGFR3):c.1138G>A and c.1138G>C, both p.(Gly380Arg), and the thanatophoric dysplasia type II variant to NM_000142.5(FGFR3):c.1948A>G p.(Lys650Glu); in the Pompe disease case in section 10.4, the familial variant is NM_000152.5(GAA):c.525_526del p.(Asn177fs). In chorionic villi, a low-allele-fraction variant may reflect true mosaicism, confined placental mosaicism, maternal contamination or an artefact; confirmation requires a suitable orthogonal method (Sanger sequencing has limited sensitivity at fractions near 10%, so mosaicism requires digital PCR or amplicon sequencing) and, if the variant is found only in the placenta, confirmation in amniocytes [36]. For an apparently de novo variant, parental sequencing data should be reviewed for low-level somatic mosaicism, present in approximately 5% (4.4% and 5.7% in two studies) of de novo variants, where recurrence risk may reach 50% [36].

There are no prenatal-specific classification criteria; as a rule, only pathogenic or likely pathogenic variants related to the phenotype and compatible with the inheritance pattern are reported, although multidisciplinary discussion may support reporting a strongly supported variant of uncertain significance in a gene closely matching the phenotype (a “hot VUS”) [36]. A VUS may be reclassified as the phenotype evolves or after confirmatory non-genetic testing (for example, demonstration of enzyme deficiency in amniocytes for biallelic DHCR7 variants), but should not be treated as pathogenic on the basis of partial segregation or any response to an intervention; in a study of neurodevelopmental disorders diagnosed through postnatal exome sequencing, 24% of non-truncating causal variants would have been classified as VUS if sequenced during pregnancy because of the absence of a specific phenotype [36]. According to ISPD, fetal incidental findings should be restricted to highly penetrant genes causing moderate to severe childhood neurodevelopmental, intellectual disability or metabolic disorders; ESHG does not recommend actively searching for secondary findings as standard care, whereas ACMG permits reporting with an opt-out option, a divergence that counselling should explain [36,38]. Postnatal reanalysis, informed by careful phenotyping, may establish diagnoses not achieved during pregnancy, and all families should have a postnatal review of results before using them for subsequent prenatal or preimplantation diagnosis [36].

6. Differential Diagnosis

6.0 Interpretation of Complex Results and Differential Diagnosis

A substantial part of prenatal genetics practice concerns intermediate situations rather than clearly normal or abnormal results, requiring careful differential reasoning.

Sampling or culture failure. Occurs in fewer than 1% of procedures and generally requires repeat sampling.

Chromosomal mosaicism. Approximately 1% of CVS samples show evidence of two or more cell lines with different chromosomal constitutions. Possible causes include maternal cell contamination, culture artefact, confined placental mosaicism (CPM)—resulting from a mitotic error restricted to the trophoblastic lineage and not necessarily representing the true fetal karyotype—or, more rarely, genuine fetal mosaicism. In amniocentesis, mosaicism is classified by level (level 1 pseudomosaicism, confirmed level 2 or 3 mosaicism), a distinction guiding the need for resampling. Definitive resolution may require additional amniocentesis or, exceptionally, fetal blood sampling, although tissue obtained at delivery is often the most reliable means of retrospective clarification.

According to Wang and colleagues' series (1993), confined placental mosaicism occurs in 1.9% of CVS procedures and true fetal mosaicism in 0.19%; in amniocentesis, mosaicism occurs in 0.3%, with more than 50% of these cases representing true fetal mosaicism [2].

Unexpected numerical chromosomal result. Beyond the major autosomal trisomies, invasive procedures may reveal sex chromosome aneuploidies (Turner syndrome, Klinefelter syndrome), for which counselling is particularly challenging because of wide phenotypic variability and the difficulty of anticipating the full spectrum of manifestations for the couple.

Prenatal diagnosis of a sex chromosome aneuploidy is almost always incidental, arising, for example, in a karyotype requested for trisomy 21 risk; in consecutive live-birth series, these aneuploidies occur in approximately 1 in 400 births but are rarely recognised at birth or in childhood, and frequency in prenatal diagnosis is higher still, estimated at 1 in 250 to 1 in 300 (European collaborative study of 52,965 amniocenteses) [40]. When NIPT suggests a fetal sex chromosome aneuploidy, maternal sex chromosome aneuploidy or mosaicism should be considered as a false-positive cause, depending on the technology used [40]. Counselling should emphasise that the information obtained was not the test's intended target, describe the specific condition (47,XXX, 47,XXY, 47,XYY or 45,X) and expectations if pregnancy continues (usually normal appearance at birth, except in 45,X with a large cystic hygroma, and mainstream schooling for most), explain that 47,XXY, 47,XYY and 47,XXX are substantially underdiagnosed in the population, discourage hasty decisions and plan follow-up (for example, referral of a 47,XXY boy to paediatric endocrinology). Recurrence risks for 45,X, 47,XXX, 47,XXY and 47,XYY are very low and probably do not exceed maternal age-specific risks, so the indication for prenatal diagnosis in a subsequent pregnancy is weak relative to procedural risk; disclosure of the karyotype to the child and others should follow the “need to know” principle [40].

De novo balanced structural rearrangement. Identification of a balanced translocation or inversion absent from the parents carries an approximately 5% to 10% risk of functional imbalance (through gene disruption at a breakpoint), potentially causing physical anomalies or developmental delay; array-CGH helps exclude loss or gain of genetic material at breakpoints, and parental chromosome and extended family investigation is indicated whenever an inherited rearrangement is confirmed.

These estimates derive from Warburton's US study (1991), predating molecular methods: a de novo balanced translocation occurred in approximately 1 in 2000 pregnancies, and, with normal ultrasound, 6.1% of pregnancies had anomalies at birth or autopsy, compared with approximately 3% at baseline, a value excluding late-onset problems. A risk of 2 to 3 times baseline for developmental problems (5–10%) was therefore used when detailed ultrasound, head circumference and growth were normal [41,42]. Balanced rearrangements are detected by neither microarray nor NIPT; an apparently balanced rearrangement warrants urgent fetal microarray and parental karyotyping, considering cryptic imbalance, unmasking of a recessive allele on the normal homologue, breakpoint gene disruption and involvement of imprinted regions (including uniparental disomy in familial rearrangements). Microarray added 8.2% to karyotyping's diagnostic yield in fetuses with abnormal ultrasound, particularly benefiting apparently balanced translocations and marker chromosomes [41]. If the rearrangement is familial and balanced in a parent, the fetus is unlikely to have a phenotypic effect, but the child's clinical record should document the finding and the family should understand that the child's own reproductive counselling should be sought in adolescence; the UK support group ARC (Antenatal Results and Choices) is recommended for couples in this situation [41].

Supernumerary marker chromosome. An additional chromosome whose origin cannot be identified through conventional cytogenetics confers an abnormal phenotype risk of up to 15% when de novo; fluorescence in situ hybridisation (FISH) and array-CGH are essential to establish its chromosomal origin and refine prognosis. The most frequently identified marker derives from chromosome 15.

Incidental findings. Prenatal array-CGH may reveal information unrelated to the original test indication—for example, a dystrophin gene deletion in a female fetus, establishing Duchenne muscular dystrophy carrier status with relevant implications for maternal cardiac screening and future pregnancies. Incompletely penetrant neurosusceptibility loci are generally not routinely reported prenatally, except when particularly relevant to the fetus or family, following the UK Joint Committee on Genomics in Medicine recommendations [43].

Ultrasound soft markers. Choroid plexus cysts, formerly associated almost deterministically with trisomy 18, are now recognised as frequent findings in chromosomally normal fetuses, becoming relevant only when large, persistent or associated with other findings. Similarly, increased bowel echogenicity, initially associated with a cystic fibrosis risk estimated as high as 10%, is now recognised as conferring a residual risk closer to 1% to 2%, warranting a cautious approach with serial ultrasound rather than immediate invasive intervention based on a single isolated finding.

The Oxford Desk Reference data on echogenic bowel refine this estimate: cystic fibrosis frequency in French series was approximately 3%; when only one parent carries a common variant, Muller and colleagues (2002) found 11% cystic fibrosis in fetuses with echogenic bowel and one variant, and application of Bayes' theorem (Caucasian population) gives a risk of approximately 16.7% when one variant has been identified in a fetal sample and one parent is a carrier, and approximately 9.5% without a fetal sample [21]. If both parents are carriers, fetal risk is very high and confirmation can be performed through amniocentesis or placental biopsy, or in cord blood at term if pregnancy continues. With negative cystic fibrosis and infection screening and low-risk trisomy screening, the risk of early growth restriction and an adverse outcome remains, warranting growth scans, particularly if maternal serum PAPP-A is low or Doppler findings are abnormal; amniocentesis is no longer offered after low-risk trisomy 21 screening [21].

Specific limitations of NIPT/NIPD. Failure to detect a specific sequence (for example, a Y sequence in a male pregnancy) does not prove its absence and may result from extraction or amplification failure or a very low fetal fraction; other error sources include the minute amount of DNA (approximately 25 genome equivalents per mL of plasma), placental mosaicism (present in approximately 1–2% of CVS procedures), twin pregnancies, resorption of a second gestational sac in the first trimester, and a maternal chromosomal abnormality or tumour. Careful ultrasound before NIPD/T is therefore mandatory to estimate gestational age and exclude multiple pregnancy and an empty sac from a non-viable pregnancy [10].

Differential diagnosis of a short femur and syndromes with limb shortening. A short femur is an ultrasound marker of trisomy 21; fetuses with trisomies 13 or 18 are small but have other distinctive ultrasound signs; triploid fetuses have very severe growth restriction (the digynic type more often survives into the second trimester, with relative macrocephaly); bilateral radial ray defects occur in trisomy 18 and Fanconi anaemia (whose chromosomal breaks are not detected by routine cytogenetics) [22]. With a normal karyotype, considerations include Cornelia de Lange syndrome (growth restriction and asymmetric radial defects), Silver–Russell syndrome (growth restriction with near-normal head circumference; approximately 10% with maternal uniparental disomy of chromosome 7), Seckel syndrome and microcephalic primordial dwarfism, and Smith–Lemli–Opitz syndrome (low oestriol and the maternal urinary steroid profile as clues) [22]. Among skeletal dysplasias, thanatophoric dysplasia (FGFR3, de novo, approximately 1/20,000) has extremely short limbs and a very narrow thorax, with curved “telephone receiver” femurs in type I and a cloverleaf skull in type II (p.Lys650Glu in all cases); achondroplasia is the most common viable dysplasia (80% de novo, G1138A and G1138C variants, paternal age effect); severe osteogenesis imperfecta, infantile hypophosphatasia (ALPL), diastrophic dysplasia (SLC26A2), short-rib polydactyly syndromes (ciliary genes such as IFT80, DYNC2H1, WDR60, NEK1) and achondrogenesis complete the main differential [22].

The overall differential diagnosis after positive screening must therefore distinguish true aneuploidy, confined placental mosaicism without fetal effects, a benign structural variant without pathological significance, and a technical limitation of the screening method itself—a distinction that only confirmatory invasive diagnostic testing can resolve with clinical confidence.

7. Treatments and Risk Management Strategies

Management, Interventions and Risk Reduction Strategies

7.1 Periprocedural counselling and management of pregnancy loss risk

Counselling before any invasive procedure should explicitly quantify the associated pregnancy loss risk—approximately 0.5% to 1% for amniocentesis and close to 1% for CVS, decreasing with operator experience—and discuss the implications of an abnormal result in advance, including the possible option of mid-trimester termination, which in some legal settings may discourage undertaking the test itself. The 2016 ISUOG practice guidelines organise the recommended technical precautions—access route, number of attempts, local anaesthesia and post-procedure follow-up—to minimise procedural risk [35].

The heterogeneity of published risk estimates should be explained to the couple: the values of 0.5–1% (amniocentesis) and approximately 1% (CVS) used here are traditional consent values; Akolekar and colleagues' meta-analysis (2015) estimated procedure-attributable risks of 0.11% and 0.22%, respectively, whereas other sources give 0.3% and 0.5% [2,3]. Salomon and colleagues' updated meta-analysis (2019), including large controlled studies and losses before 24 weeks, estimated an attributable risk of 0.30% (95% CI 0.11–0.49%) for amniocentesis (580 losses in 63,723 procedures, 0.91%, versus 0.58% in controls) and 0.20% (95% CI −0.13 to 0.52%) for CVS (1.39% versus 1.23%); restricting analysis to studies with controls of similar risk profiles gave 0.12% and −0.11%, neither statistically significant, leading the authors to conclude that traditionally communicated risks are overestimated [44]. Between-study heterogeneity was considerable (I² of 70.1% for amniocentesis), so these values should be presented as ranges alongside the centre's experience. Given miscarriage risk, some authors prefer invasive testing only when the couple would consider termination following a positive result, with postnatal testing as an alternative; emotional, religious, ethical and moral considerations should be discussed in depth [3]. Maternal blood group should be known for every invasive procedure, anti-D prophylaxis offered to Rh-negative women, and the arrangements for communicating results agreed in advance [2].

7.2 Targeted fetal treatment

Historically limited to therapeutic cordocentesis—intrauterine red cell or platelet transfusion for Rhesus isoimmunisation, direct administration of medicines such as digoxin or curare, and monitoring of transplacental maternal treatment (corticosteroids, immunoglobulin, antibiotics, oxygen)—fetal treatment has expanded in recent years to molecular interventions. A recent illustrative milestone is the first documented in utero treatment of spinal muscular atrophy (SMA) with risdiplam (Evrysdi®), reported by the St. Jude Children's Research Hospital team in a letter to the New England Journal of Medicine published online on 19 February 2025 [45]. The fetus, a sibling of a child who died with genetically confirmed SMA type 1, was tested through amniocentesis: zero SMN1 copies (diagnostic of SMA) and two SMN2 copies, predictive of SMA type 1. Under a single-patient investigational plan approved by the FDA and local ethics committee, the mother received oral risdiplam 5 mg daily from 32 weeks and 5 days until delivery at 38 weeks and 6 days, with weekly obstetric monitoring; at birth, drug concentrations in amniotic fluid and cord blood were 33% and 69% of the maternal plasma concentration, respectively. The newborn started oral risdiplam on day 8 and continued daily until 30 months, when there was no hypotonia, weakness, areflexia or fasciculation, with normal six-monthly motor, muscle ultrasound and electrophysiological studies and markers (SMN protein, neurofilaments) consistent with target engagement [45]. The child nevertheless had a ventricular septal defect that resolved, bilateral optic nerve hypoplasia with mildly reduced visual acuity and transient nystagmus, mild right hemiparesis associated with left midbrain hypoplasia, and global developmental delay without regression; microarray and long-read genome sequencing identified no other genetic disorder, and the investigators considered these anomalies to predate exposure, with no identified cause [45]. This is a single case, published as correspondence, whose results the authors themselves state are not generalisable; in the evidence hierarchy, it is proof of concept rather than a trial. In September 2026, St. Jude registered an observational registry of prenatally diagnosed SMA cases treated in utero with disease-modifying therapy (NCT07805278; planned start October 2026, 30 participants), with maternal–fetal surveillance, pregnancy and delivery outcomes [46].

In cell therapy, prenatal stem cell transplantation has mainly been studied in osteogenesis imperfecta using allogeneic fetal mesenchymal stem cells (the European phase I/II BOOSTB4 trial, NCT03706482, sponsored by Karolinska Institutet, compares postnatal or prenatal plus postnatal administration with untreated historical and prospective controls, has safety as its primary endpoint and was active, not recruiting, with primary completion expected in 2030 in its latest September 2025 update), and in severe combined immunodeficiency (SCID) using haematopoietic stem cells, exploiting physiological fetal immune tolerance to foreign antigens introduced in utero, which reduces rejection risk compared with postnatal transplantation [47,48]. In utero enzyme replacement therapy has progressed from animal experimentation to clinical use: the phase 1 PEARL trial (NCT04532047, University of California, San Francisco) assesses the safety and feasibility of umbilical vein infusions in fetuses from 18 weeks to 34 weeks and 6 days with mucopolysaccharidoses I, II, IVA, VI and VII, Gaucher disease types 2 and 3, infantile-onset Pompe disease and Wolman disease, and is recruiting (10 planned participants; primary completion expected in 2031) [49]; the first case treated under this protocol is described in section 10.4 [50]. Prenatal gene therapy, still experimental, seeks to exploit the same immune tolerance window and the smaller fetal body volume (allowing proportionally smaller vector doses) to prevent irreversible damage to organs such as the central nervous system in early-onset neurodegenerative disorders, with advanced preclinical studies and an initial proposed human clinical trial targeting GM1 gangliosidosis [47,51].

Other therapeutic applications guided by prenatal diagnosis include congenital adrenal hyperplasia: fetal sex determination by NIPD allows maternal dexamethasone to be discontinued in male pregnancies, and NIPT may permit non-invasive diagnosis before the ninth week so that treatment, whose indication is controversial and should follow current guidelines, is directed only to mothers carrying affected female fetuses [10]. Prenatally diagnosed achondroplasia warrants delivery planning in a specialist unit because of cephalopelvic disproportion risk from relative macrocephaly; in thanatophoric dysplasia, predelivery diagnostic confirmation, possible through NIPD, allows neonatal management planning [22].

Investigation after fetal loss with an anomaly. If the fetus or newborn dies, clinical photography, skeletal radiography (some skeletal dysplasias may present with hydrops, and oedema hampers limb length assessment), tissue storage (for example, a fibroblast cell line) and DNA storage should be undertaken, and autopsy including the placenta offered; if myopathy with fetal akinesia is suspected, muscle biopsy should be obtained as soon as possible after delivery. In recurrent undiagnosed hydrops, a fibroblast line should be established and amniotic fluid frozen; in severe or lethal hydrops without an identified cause, particularly in consanguineous families, exome, genome or panel sequencing should be considered [15]. In skeletal dysplasia, DNA may be stored from cord blood, fetal blood, placental tissue or skin biopsy, complemented by full skeletal radiography, photography and clinical measurements; even if autopsy is declined, most necessary information can be obtained, and, if performed, it should include bone histology [22].

7.3 Termination of pregnancy: clinical and ethical considerations

A severe fetal anomaly provides legally recognised grounds for termination in some legal systems; however, this option is not universally accepted, and the decision involves ethical, moral and religious considerations varying with culture, the legal system and the personal convictions of each couple and professional. Specialist genetic counselling should remain non-directive in every setting, accurately presenting available prognostic information and supporting informed family decision-making, whatever the chosen option—further investigation, preparation for the birth of a child with special needs, or termination where legally available. Section 9 addresses this issue through different bioethical frameworks.

7.4 Regulatory framework for tests and fetal therapies

As a thematic chapter, there is no medicine specific to “prenatal diagnosis”; regulatory rules apply to two distinct subjects: screening and diagnostic tests (in vitro diagnostic medical devices or laboratory-developed tests), and medicines used with the intention of treating the fetus, none of which has a fetal indication in its label in the sources consulted. The following table summarises the status verified on 6 October 2026.

The distinction between evidence extrapolation and off-label use is essential: giving a medicine approved for postnatal SMA to a pregnant woman to treat the fetus is off-label use (population, exposure route and life stage outside the label), and its scientific rationale is a biological extrapolation (transplacental passage, animal data) supported by one published case. Medicine authorisation also does not guarantee reimbursement or access in each country; screening test reimbursement depends on the national programme or insurance.

Pivotal trials and studies in prenatal genetics. The following table brings together the studies supporting the main recommendations, including negative trials:

These studies should not be compared directly. cfDNA detection rates derive mainly from validation studies with enriched populations and manufacturer-funded studies, and positive predictive value falls in low-prevalence populations and rare conditions (trisomy 13, additional findings). Procedural risks differ according to whether controls have comparable aneuploidy risk; a small relative difference must always be interpreted alongside the absolute baseline rate. Exome yields depend on phenotype selection, trio analysis and prior exclusion of CNVs. The two randomised PGT-A trials were negative or demonstrated non-inferiority for the primary outcome, and STAR's benefit in women aged 35–40 years derives from a post hoc analysis. Fetal treatment cases are single reports without controls, whose outcomes do not establish efficacy.

8. Reproductive Options and Family Counselling

The main international references, consensus statements and guidelines on human reproduction and prenatal and preconception counselling are centralised through the European Society of Human Reproduction and Embryology (ESHRE) and the American Society for Reproductive Medicine (ASRM), whose regularly updated clinical practice documents form the basis of these recommendations [59,60].

8.1 Preimplantation genetic testing (PGT)

Preimplantation genetic diagnosis, historically termed PGD (for monogenic disorders or structural rearrangements) and PGS (for aneuploidy screening), was reclassified in 2017–2018 under the ESHRE/ASRM international infertility terminology as the unified term PGT (preimplantation genetic testing), subdivided into PGT-A (aneuploidies), PGT-M (monogenic disorders) and PGT-SR (structural rearrangements) [61]. The procedure begins with controlled ovarian hyperstimulation and transcervical oocyte retrieval under sedation and ultrasound guidance; after in vitro fertilisation, embryos are usually biopsied at the blastocyst stage and cellular material is amplified (multiple displacement amplification) and analysed through haplotype markers or targeted sequencing, allowing selective transfer of unaffected embryos. Historically, success per treatment cycle is approximately 30%, with technical alternatives including polar body removal instead of direct embryo biopsy. Exceptionally, PGT is also used to select an embryo with a human leucocyte antigen (HLA) match to a sibling with a severe haematological disorder, allowing related-donor bone marrow transplantation.

Historical and outcome data for PGT. Preimplantation diagnosis was introduced in 1990 for embryo sex determination in X-linked disorders; the first birth after preimplantation diagnosis of a monogenic disorder (cystic fibrosis) was reported in 1992, and the first after diagnosis of a dystrophin gene deletion in 1995 [62]. The ESHRE consortium's thirteenth data collection (62 centres) reported 17,721 cycles reaching oocyte retrieval (6968 for chromosomal abnormalities, 2814 for X-linked disorders, including 1484 for sex determination alone, 7294 for autosomal monogenic disorders and 643 for HLA typing), with 12,785 embryo transfer cycles and clinical pregnancy rates of 21% per retrieval and 29% per transfer; most couples using PGT have had at least one previous pregnancy with an affected fetus or child, and most have no unaffected living children [62]. The development of preimplantation genetic haplotyping, which avoids a variant-specific assay by identifying informative linked markers on each side of the gene, and more recently SNP-based platforms, enabled PGT to be offered to any couple carrying a monogenic disorder with a known gene and sufficient familial samples to establish phase [62]. Most couples request PGT to avoid the possibility of terminating a pregnancy after prenatal diagnosis; others require assisted reproduction because of a fertility problem that may have a genetic cause (for example, congenital bilateral absence of the vas deferens) [62]. ICSI is used in PCR-based PGT protocols to avoid contamination by sperm retained in the zona pellucida [62].

Adverse effects and diagnostic error. Ovarian hyperstimulation syndrome occurs in a mild form in 8% to 23% of cases and a severe form in 0.1% to 2%, particularly relevant to those caring for a child with a disability at home, although changes in stimulation protocols have reduced this risk [62]. Twelve diagnostic errors were reported to ESHRE; causes include mosaicism (less frequent with blastocyst biopsy), allele drop-out (ADO), which can remove the affected allele in a dominant disorder, and contamination during the multiple PCR cycles required; multiplex PCR protocols and sufficient informative markers reduce this risk [62]. Confirmatory prenatal diagnosis may therefore be offered after PGT—in Nannenberg and Arens' case, because of a 2% diagnostic error risk [3]. Neonatal outcomes after PGT are considered similar to those after ICSI without embryo biopsy, and most biopsies are now performed at the blastocyst stage (days 5 to 7), with frozen embryo transfer [62].

PGT with HLA typing and aneuploidy screening. When selecting an unaffected embryo that is HLA-compatible with an affected sibling, each embryo has only a 3/16 probability of being both unaffected by a recessive disorder (3/4) and HLA-compatible (1/4); in a study of 242 couples, the probability of successful transplantation was 9.5% [62]. For aneuploidy screening (PGS, now PGT-A), approved in principle by the HFEA in 2001 for couples with repeated IVF/ICSI failure, a meta-analysis of randomised trials (Mastenbroek, 2011) showed no improvement in live-birth rates and a significant reduction in women of advanced maternal age; debate continues over which patients, if any, should be offered it [62]. Randomised trials using next-generation sequencing in blastocysts have not changed this conclusion: in STAR (661 women aged 25–40 years), ongoing pregnancy rates did not differ between PGT-A and morphology selection, and in Yan and colleagues' Chinese trial (1212 women with at least three good-quality blastocysts), conventional IVF was non-inferior to PGT-A for cumulative live births (81.8% versus 77.2%), although clinical pregnancy loss was lower with PGT-A (8.7% versus 12.6%) [57,58]. In its 2023 good practice recommendations on add-ons, ESHRE concluded that current data indicate limited improvement in live-birth rates, that purported reductions in miscarriage or time to pregnancy in subgroups (for example, advanced maternal age) rely on post hoc analyses, and that PGT-A—like non-invasive PGT, still investigational—is not recommended for routine clinical use [63]. PGT-M and PGT-SR, indicated by known monogenic disorder or structural rearrangement risk, follow the ESHRE PGT consortium's 2020 good practice recommendations on centre organisation, biopsy and detection methods [64].

PGT and prenatal diagnosis in inherited cardiac diseases. Inherited cardiomyopathies and arrhythmia syndromes are predominantly autosomal dominant, with incomplete penetrance and variable expressivity, and have preventive and therapeutic options; reports of prenatal diagnosis and PGT for these disorders are sparse [3]. In the Netherlands between 2001 and 2015, prenatal diagnosis was performed 8 times for inherited cardiac disease (3 couples for hypertrophic cardiomyopathy, 2 for dilated cardiomyopathy and 1 for long QT syndrome), approximately 0.3–0.7% of annual prenatal diagnoses; in Europe between 2006 and 2010, 24 PGT cycles were undertaken for inherited cardiac diseases (1 dilated cardiomyopathy, 7 hypertrophic, 2 arrhythmogenic right ventricular, 1 congenital, 4 long QT, 7 Brugada syndrome, 2 catecholaminergic polymorphic ventricular tachycardia) [3]. Of 3280 couples referred for PGT in the Netherlands between 1995 and 2017, 45 (approximately 1%) had an inherited cardiac disease (76% cardiomyopathies); after counselling, 11 proceeded, and 4 couples achieved a pregnancy with a child without the variant [3]. Dutch PGT eligibility criteria are as follows [3]:

Developing a PGT test for each specific indication requires considerable time (9 months in the reported case), given the small amount of DNA from one or a few embryonic cells, and often requires participation by affected and unaffected relatives for linked markers [3]. Couples using PGT are generally fertile and undergo IVF solely for embryo selection; they should be informed that children born after IVF may have increased risks of prematurity and low and very low birth weight [3].

The following table gives a non-exhaustive selection of conditions for which PGT-M or PGT-SR is available in licensed centres, grouped by inheritance pattern:

Regulation and licensing of PGT centres vary internationally: in the United Kingdom, the Human Fertilisation and Embryology Authority (HFEA) maintains a regularly updated public list of conditions approved for PGT [65], within a substantially more centralised regulatory framework than in the United States, where no specific federal legislation regulates assisted conception apart from IVF and ICSI outcome reporting requirements.

8.2 In vitro fertilisation and intracytoplasmic sperm injection

In vitro fertilisation (IVF), available since 1978, now accounts for 1% to 3% of births in several Western countries, with subfertility—affecting approximately one in seven couples—as its main indication. Accumulated evidence shows a relative increase of 30% to 40% in congenital defect risk in IVF-conceived children compared with natural conception, corresponding to absolute rates of approximately 7% to 8% versus 6% in the general population; a small excess of epigenetic conditions associated with defective genomic imprinting, particularly Beckwith–Wiedemann and Angelman syndromes, has also been described, with the precise mechanism incompletely understood and hypothetically related to prolonged in vitro embryo culture.

In Davies and colleagues' study (2012) in South Australia, 7.2% of IVF-conceived children had a congenital defect diagnosed before age five, versus 5.8% of those conceived without treatment (adjusted OR 1.07; 95% CI 0.90–1.26), and 9.9% of ICSI-conceived children (adjusted OR 1.57; 95% CI 1.30–1.90) [62]. Assisted reproduction increases low birth weight risk in term singleton pregnancies (OR 1.4); Beckwith–Wiedemann syndrome risk after IVF was estimated at 1 in 4000 (Halliday, 2004), although a large Danish registry found no excess cases [62]. In IVF, controlled ovarian stimulation with exogenous gonadotropins is followed by a hormonal trigger for oocyte maturation and ultrasound-guided transvaginal retrieval 34 to 38 hours later, with embryo transfer 2 to 7 days after fertilisation [62].

Intracytoplasmic sperm injection (ICSI), used predominantly for male subfertility—low sperm count, abnormal motility or morphology, or vas deferens obstruction—raises specific genetic considerations: approximately 5% of men eligible for ICSI have chromosomal rearrangements, increasing to 10–12% in azoospermia or severe oligospermia, with characteristic examples including the 13:14 Robertsonian translocation and Y chromosome microdeletions. Men with congenital bilateral absence of the vas deferens (CBAVD) frequently carry CFTR variants associated with cystic fibrosis, warranting preconception carrier screening of the couple; ICSI is also the preferred approach for men with Klinefelter syndrome, using testicular sperm aspiration. Chromosomal abnormality risk in ICSI-conceived offspring is slightly increased (approximately 1.6%), and approximately 10% of children conceived through this method may have congenital defects identified before age five, compared with up to 6% after natural conception—differences partly reflecting the underlying aetiology of subfertility rather than the technique alone.

8.3 Donor insemination and gamete donation

Donor insemination (DI), used to treat male infertility or avoid transmission of paternal genetic disease, underwent increased genetic screening of sperm donors after documented cases of children conceived with paternally derived chromosomal disorders or cystic fibrosis; from 2000, European andrology societies formally recommended screening donors for cystic fibrosis variants and chromosomal rearrangements. Transmission of neurodegenerative disorders to multiple offspring of a single donor reinforced the need to limit the number of recipient families per donor; older men also have a small increase in the risk of new de novo genetic variants in sperm. The right of a DI-conceived child to know the genetic donor's identity is a legal and ethical issue varying substantially between jurisdictions, applying equally to oocyte donation.

8.4 Mitochondrial donation

In severe maternally transmitted mitochondrial disorders, where recurrence risk may approach 100%, nuclear replacement (transfer of maternal nuclear genetic material—in pronuclear transfer, the pronuclei of the zygote of the patient carrying the pathogenic mitochondrial variant—into an enucleated donor oocyte or zygote with healthy mitochondria) can markedly reduce the burden of mutant mitochondrial DNA in offspring. This technique, popularly described as producing “three-parent babies”, prompted substantial ethical controversy despite the donor's genetic contribution being limited to a minute fraction of the mitochondrial genome. The United Kingdom pioneered its regulation, with Newcastle Fertility Centre receiving the first clinical licence in 2017 and the first confirmed birth in 2023; data published in the New England Journal of Medicine in 2025 document eight children born through pronuclear transfer, all without clinical evidence of mitochondrial disease, although three had low residual mutant mitochondrial DNA levels at birth (5%, 12% and 16% in blood and 9%, 13% and 20% in urine, respectively; thus between 5% and 20% depending on the tissue), undetectable by pyrosequencing in the other five. In the child with 5% in blood at birth, the level was undetectable at 18 months [66–68]. To date, the Human Fertilisation and Embryology Authority restricts mitochondrial donation to severe mitochondrial disease without a viable therapeutic alternative, including PGT-M; it is not licensed for general infertility treatment.

Data published in July 2025 refine these results: in the Newcastle programme, which offered PGT to women with heteroplasmy and pronuclear transfer to women with homoplasmy or high heteroplasmy, clinical pregnancy was confirmed in 8 of 22 patients (36%) undergoing ICSI for pronuclear transfer and 16 of 39 (41%) for PGT; pronuclear transfer resulted in 8 live births and 1 ongoing pregnancy, and PGT in 18 live births [66]. Blood heteroplasmy levels in the 8 children born after pronuclear transfer ranged from undetectable to 16%, with a 95–100% reduction (6 newborns) and 77–88% reduction (2 newborns) in maternal variant level compared with the enucleated zygotes; in PGT-born children with available data (10 of 18), heteroplasmy ranged from undetectable to 7% [66]. During clinical follow-up, all 8 children were healthy at birth; one developed hyperlipidaemia and cardiac arrhythmia (the mother had hyperlipidaemia during pregnancy), both responding to treatment, and another infantile myoclonic epilepsy with spontaneous remission. At reporting, all were making normal developmental progress [68].

8.5 Recurrence counselling and family pedigree

Across these reproductive pathways, a detailed family pedigree remains the fundamental tool for empirical risk assessment, allowing estimation of whether recurrence risk for a structural anomaly without an established molecular diagnosis is consistent with a recessive pattern (usually 25%), a dominant pattern or a de novo variant with very low recurrence risk. When no definitive molecular diagnosis can be achieved despite extensive investigation, detailed fetal ultrasound in subsequent pregnancies from the sixteenth week remains the main surveillance tool available to the couple.

9. Special Considerations: Population Diversity and Epidemiology

Population Diversity, Founder Effects and Ethical Framework

9.1 Carrier screening in founder populations

Modern biochemical and molecular screening already allows, in private healthcare, carrier screening for more than 2000 common pathogenic variants associated with approximately 250 disorders, mostly autosomal recessive or X-linked recessive, including “pan-ethnic” panels. Historically, however, targeted screening arose in populations with well-characterised founder effects: the Ashkenazi Jewish community has substantially elevated carrier frequencies for Tay–Sachs disease, cystic fibrosis, Gaucher disease and other recessive conditions, prompting UK charity-funded preconception screening programmes for this community, such as Jnetics. Similar founder effects are described in other genetically isolated populations worldwide, although availability of published population epidemiological data varies considerably by region.

Carrier screening was long available only for autosomal recessive disorders highly prevalent in particular populations (Tay–Sachs disease, sickle cell anaemia, cystic fibrosis), but next-generation sequencing enabled rapid, accessible testing with large panels for disorders for which the individual has no known increased risk; the number of screenable conditions is expected to increase as sequencing costs fall [9]. Cystic fibrosis carrier frequency varies with ethnicity, so Bayesian risk calculations, such as those for echogenic bowel, should use the frequency in the couple's population of origin (commonly published values refer to the Caucasian population) [21]. PGT regulation varies by country and is generally assigned to the bodies regulating assisted reproduction; in the United Kingdom, the HFEA was established in 1990 under the Human Fertilisation and Embryology Act [62].

The American College of Medical Genetics and Genomics (ACMG) first recommended pan-ethnic carrier screening for cystic fibrosis and, in 2008, for spinal muscular atrophy (SMA), whose carrier frequency was then estimated at 1/40 to 1/60, with a prevalence of approximately 1 in 10,000 live births; SMN1 dosage testing detects more than 90% of carriers but does not identify carriers with two SMN1 copies on one chromosome and none on the other (2+0 configuration), or carriers of point variants, which account for approximately 5% of patients (compound heterozygotes for a deletion and a point variant) [69]. In 2021, the ACMG replaced the non-specific term “expanded screening” with a tiered system: tier 1 corresponds to pan-ethnic cystic fibrosis and SMA screening; tier 2, proposed by ACOG, includes conditions with a moderate or severe phenotype and carrier frequency ≥1/100; tier 3 includes conditions with carrier frequency ≥1/200 in any subpopulation represented in the United States (97 autosomal recessive genes and 16 X-linked genes); and tier 4 includes less frequent genes without a lower frequency limit [70]. The ACMG recommends offering tier 3 screening to all pregnant women and those planning pregnancy (the male partner may be screened concurrently for autosomal recessive conditions), considering tier 4 in pregnancies of consanguineous couples (second cousins or closer) or when justified by personal or family history, and not offering tiers 1 and 2 because they do not provide equitable assessment across population groups. If the partner is unavailable, declines testing or would not receive results in time for reproductive decision-making, a diagnostic procedure should be offered with counselling about the limitations of fetal gene analysis in these circumstances [70].

9.2 Regional disparities in access and regulation

Access to cfDNA prenatal screening and its coverage by healthcare systems or private insurance are markedly unequal even among high-income countries: in the United States, private insurance coverage of NIPT in general-risk pregnancies varies and is not quantified in the peer-reviewed sources consulted; in the Netherlands, first-line NIPT was offered to all pregnant women through a single national counselling, testing and follow-up programme, and in the first year 54% of pregnant women did not participate in any screening, illustrating that universal availability does not mean universal uptake [33]. Assisted conception and preimplantation genetic testing also follow different regulatory models: the United Kingdom operates under centralised HFEA licensing through the Human Fertilisation and Embryology Act of 1990 (updated in 2008), with mandatory registration of all treatment cycles; the United States has no specific federal legislation regulating these practices beyond individual outcome reporting requirements. The literature consulted does not allow precise quantification of coverage, relative cost or access barriers for prenatal screening and preimplantation diagnosis in Latin America, sub-Saharan Africa, the Middle East or much of Asia; this regional data gap should be explicitly acknowledged rather than extrapolated from high-income settings, and also reflects the limitations of international literature predominantly published by Western European and North American centres, a disparity identified in Allyse and colleagues' review of international NIPT implementation challenges [71].

Among the sources consulted, Portuguese-speaking participation appears mainly in consensus documents and reviews rather than population data: the 2026 ESHG and ISPD global recommendations included authors from centres in Porto, Coimbra and Lisbon [36], and an integrative review by a Campinas group (Brazil) compared international prenatal screening and diagnostic guidelines in 2026 [31]; neither source provides national rates of screening uptake, invasive procedures or PGT in Portugal or Brazil, which does not mean that such data are absent from other publications. The 2026 recommendations also emphasise that local genomic resources help reduce uncertainty in variant interpretation for populations under-represented in reference databases [36].

European networks and patient organisations. No European Reference Network (ERN) is dedicated to prenatal diagnosis; couples with an affected fetus are directed to the ERN for the relevant condition—for example, ERN ITHACA (congenital anomalies and malformation syndromes, intellectual disability), ERN BOND (skeletal dysplasias), ERNICA (congenital digestive anomalies) or EURO-NMD (neuromuscular disorders such as SMA) [72]. Support resources include EURORDIS, representing European rare disease patient organisations, and, in the United Kingdom, ARC (Antenatal Results and Choices), which specifically supports couples facing abnormal prenatal results and pregnancy decisions [41,73].

9.3 Bioethical perspectives on prenatal diagnosis, termination and assisted reproduction

Ethical assessment of prenatal diagnosis, particularly its possible use as the basis for a termination decision, is not universally agreed and reflects distinct philosophical, religious and legal frameworks that medical genetics professionals should recognise without imposing them on the couples they counsel.

Among religious frameworks historically influential in Western clinical practice, the Catechism of the Catholic Church maintains that human life must be absolutely respected from conception, considering direct abortion—sought as an end or a means—a grave moral offence, also subject to canonical sanction. Within this doctrinal perspective, prenatal diagnosis is considered morally permissible only when directed towards protecting or treating the embryo or fetus itself, and morally unacceptable when intended as a route to possible termination depending on the result. The same doctrinal framework extends ethical reservations to assisted reproductive techniques involving third-party gamete donors or surrogacy because they separate biological parenthood from the marital relationship, and regards any conception separating the marital act from procreation as morally problematic, including artificial insemination and fertilisation even when undertaken exclusively between the couple's members.

This position differs from secular bioethical frameworks widely used in other genetic counselling settings, such as the four-principle model (autonomy, beneficence, non-maleficence and justice), in which the couple's informed reproductive decision following non-directive counselling is generally the central basis of ethical practice, without a prenatal test result itself being morally equated with a judgement on the value of the fetus's life. A third relevant framework, arising from the disability rights movement, critically questions—irrespective of the position on termination itself—whether widespread prenatal screening for conditions such as Down syndrome implicitly reinforces diminished social valuation of the lives of people with intellectual disabilities, an argument spanning religious and secular debates. Responsible prenatal genetic counselling, irrespective of the professional's ethical framework, follows non-directivity: providing complete, accurate information without pressing the family towards a specific decision, explicitly recognising the plurality of values involved.

10. Illustrative Clinical Case

Illustrative Clinical Cases

10.1 Clinical Case 1 — High-risk combined screening due to isolated low PAPP-A

Presentation. A 36-year-old nulliparous woman receives a high-risk combined screening result at 12 weeks of gestation, determined by maternal age combined with markedly reduced PAPP-A (below 0.4 multiples of the median). NT and the remaining ultrasound assessment show no additional abnormality.

Discussion and management options. The first counselling step is to explain that a “high-risk” combined test result represents a revised statistical probability rather than a diagnosis: at age 36 years, baseline risk is already higher than in younger women, and such low PAPP-A shifts combined risk to a level warranting further investigation, yet still consistent with a euploid pregnancy in most cases. Three management pathways are clinically reasonable and should be presented non-directively: (1) proceed directly to CVS, still technically feasible at 12 weeks, with QF-PCR providing a common aneuploidy result in approximately three days; (2) request cfDNA as secondary screening to refine probability before deciding on an invasive procedure, although this may delay definitive diagnosis by another one to two weeks; or (3), if the couple wishes to avoid any first-trimester intervention, wait until 15–16 weeks for amniocentesis. CVS and amniocentesis have equivalent cytogenetic sensitivity and specificity, so choice depends mainly on gestational age and the couple's preference about result timing.

A frequently underestimated teaching point is that isolated very low PAPP-A, even with a subsequently normal cytogenetic result, independently marks increased risk of late placental complications—fetal growth restriction, pre-eclampsia and, in extreme cases, fetal death—reflecting impaired trophoblastic invasion. Even with a normal fetal karyotype, the pregnancy should therefore be flagged for serial ultrasound surveillance of fetal growth and uterine artery Doppler in the second and third trimesters; low-dose acetylsalicylic acid prophylaxis should be considered within routine obstetric follow-up, although this therapeutic decision extends beyond the strict scope of genetic counselling.

If cytogenetic testing (QF-PCR/array-CGH) confirms a trisomy, the couple should receive further specialist genetic counselling, discussing the prognosis of the identified trisomy, options for continuing pregnancy with multidisciplinary support and, where legally available, termination, always non-directively.

10.2 Clinical Case 2 — Couple with a family history of cystic fibrosis, 6-week pregnancy

Presentation. A couple is referred to genetics because of a family history of cystic fibrosis (CF), an autosomal recessive disorder caused by pathogenic CFTR variants. Pregnancy is ongoing at 6 weeks.

Discussion and management options. Fetal risk depends entirely on each parent's carrier status, so the first and most urgent step, given the tight time window, is rapid determination of whether both carry a pathogenic CFTR variant, ideally identifying the specific familial variant through the previously diagnosed relative, substantially accelerating subsequent laboratory analysis compared with expanded screening. If only one parent is a carrier, the risk of an affected fetus is practically negligible (limited to the rare possibility of a new variant in the other parent or both alleles coming from the same carrier, which does not apply to classical autosomal recessive CF), and pregnancy can continue without CF-targeted invasive testing. If both are confirmed carriers, each pregnancy has a 25% risk of an affected fetus, 50% of an asymptomatic carrier and 25% of inheriting neither variant.

Given the time pressure—the pregnancy is only at the sixth week and the optimal CVS window (11 to 13 weeks and 6 days) leaves a tight deadline for both parents' carrier testing—blood collection should be arranged promptly for molecular analysis targeted to the known familial variant (or an expanded panel if no familial variant is identified), allowing results in time to guide the invasive testing decision within the first-trimester window. If both are carriers, CVS between 11 and 13+6 weeks allows direct fetal DNA analysis for the specific parental variants, typically yielding results within a few days; amniocentesis from 15 weeks is an alternative if the CVS window is missed. Both involve direct mutation analysis without prolonged cell culture and therefore have a rapid turnaround.

Beyond prenatal diagnosis in the current pregnancy, future reproductive counselling should cover PGT-M in a subsequent IVF cycle, selecting unaffected embryos before transfer, and, in principle, NIPD targeting absence of the specific paternal variant in future pregnancies, an approach already clinically validated for CF in specialist centres but with availability varying by country and laboratory. If the current fetus is identified as affected, counselling should present the CF severity spectrum in a balanced way—now substantially modified by highly effective postnatal CFTR modulators—and the couple's available options, with specialist psychological and genetic support.

10.3 Clinical Case 3 — PGT-M for familial dilated cardiomyopathy due to a PLN founder variant

Presentation. A couple requested PGT for familial dilated cardiomyopathy with a history of sudden death (a brother died at age 18 years after cardiac arrest). A nuclear panel of 46 cardiomyopathy genes identified the Dutch PLN founder variant c.40_42del (p.Arg14del); the male partner was heterozygous, without cardiomyopathy signs on cardiac assessment [3].

Discussion and course. After counselling by a clinical geneticist and psychologist, the PGT working group accepted the request; developing the specific test took 9 months, during which the couple was accepted for IVF, and treatment was scheduled 3 months later [3]. The first attempt produced no suitable embryos; the second transferred an unaffected embryo and achieved pregnancy [3]. Confirmatory prenatal diagnosis was offered because of the 2% diagnostic error risk associated with the small DNA quantity available in PGT, but the couple declined; a healthy child was born, with negative variant status available for confirmation by postnatal DNA analysis [3]. This case illustrates the full PGT-M pathway in an incompletely penetrant disorder—from variant identification to birth—and the timelines and decisions the couple should anticipate.

10.4 Clinical Case 4 (published) — CVS diagnosis and in utero enzyme replacement therapy for infantile Pompe disease

Presentation. A 37-year-old woman (gravida 12, para 4) from a consanguineous family of Pakistani ancestry, with three previous pregnancies affected by CRIM-negative infantile-onset Pompe disease, presented in the first trimester; fetal diagnosis was established through CVS, showing homozygosity for the familial variant NM_000152.5(GAA):c.525_526del p.(Asn177fs) [50]. The first affected child was diagnosed at 5.5 months, received immune tolerance induction and enzyme replacement from 6.6 months and died at 29 months with progressive disease; the second affected pregnancy was terminated; the third child, diagnosed through amniocentesis and showing severe left ventricular hypertrophy on fetal echocardiography at 34 weeks (z-score 7.0), received palliative care and died at 8 months of cardiorespiratory failure [50]. The pedigree and siblings' clinical courses appear in Figure 1A of the original article (https://pmc.ncbi.nlm.nih.gov/articles/PMC10794051/).

Intervention and course. After non-directive counselling about all options, the family chose fetal treatment; because the pandemic prevented travel to the trial centre, treatment was performed locally in Ottawa with clinical ethics approval and according to the trial protocol [50]. Alglucosidase alfa was administered through the umbilical vein under ultrasound guidance from 24 weeks and 5 days, every 2 weeks until 34 weeks and 5 days (six infusions), without maternal or fetal adverse reactions; vaginal delivery occurred at 37 weeks and 4 days, and the child underwent immune tolerance induction on day 1 and postnatal enzyme therapy from day 4 [50]. Fetal echocardiography at 34 weeks showed no hypertrophy (z-score 0.6), and the left ventricular mass index at birth was normal, unlike the siblings (Figure 1C–D); the placenta lacked the glycogen deposits seen in untreated placentas (Figure 2); antidrug antibodies appeared after the third infusion, peaking at 1:3200 in utero and 1:6400 after birth, always below the threshold considered clinically significant (Figure 3). At 13 months, the child had normal cardiac and motor function, independent walking at 11.5 months, and normal creatine kinase and biomarkers [50].

Discussion. The case integrates almost every stage of the clinical pathway: family history and consanguinity, early invasive diagnosis targeted to the familial variant, normalised HGVS nomenclature, non-directive counselling and experimental fetal treatment. It also illustrates the evidence hierarchy: one patient compared with siblings and historical cohorts, rather than a controlled trial; long-term effects are unknown, and treatment should take place only in a trial (such as PEARL) or with specific ethical approval [49,50].

11. Prognosis and Long-Term Follow-up

Prognosis in prenatal genetics comprises several overlapping dimensions over time: the condition-specific fetal prognosis, the obstetric prognosis of the current pregnancy, and the couple's reproductive prognosis for future pregnancies.

Follow-up after confirmed cytogenetic diagnosis. Once a trisomy or other major chromosomal abnormality is confirmed, multidisciplinary follow-up—maternal–fetal medicine, paediatric cardiology, neonatology and clinical genetics—allows anticipation of care needs at birth (for example, systematic cardiac screening in trisomy 21 because of the high associated congenital heart disease prevalence). For couples continuing pregnancy, care includes fetal growth monitoring and serial fetal cardiac surveillance; for those choosing termination where legally available, structured subsequent psychological support is integral to care, and cytogenetic analysis of obtained fetal tissue should always be completed to provide accurate recurrence counselling for future pregnancies.

Follow-up after ambiguous results or mosaicism. Obtaining fetal tissue at delivery—even when pregnancy continues without further intervention—often remains the only route to definitive diagnostic resolution and should be planned with the genetics team in advance and communicated to the neonatal team.

Follow-up after increased NT with a normal karyotype. One study found postnatal anomalies in 11% of fetuses with NT of 3.5 mm or greater and unremarkable ultrasound (Senat, 2007); if the 20-week anomaly scan is entirely normal and NT has resolved completely, risk returns to baseline, although further studies are needed to confirm this [15].

Reproductive follow-up after PGT and IVF/ICSI. Long-term fertility centre and regulatory registries (such as the UK HFEA) continue following cohorts of IVF-, ICSI- and PGT-conceived children to monitor long-term developmental outcomes, given the slight excess of congenital anomaly and epigenetic condition risks described in section 8; this continuing population surveillance is particularly relevant to newer techniques such as mitochondrial donation, whose first UK cohort of eight pronuclear-transfer-born children is under structured paediatric follow-up, with periodic monitoring of residual mutant mitochondrial DNA levels in accessible tissues (blood and urine) beyond the immediate neonatal period [66,68].

The need for international collaboration in long-term follow-up of children born after PGT has been recognised, but systematic organisation exists in only some centres [62]. Demand for prenatal diagnosis and PGT in inherited cardiac diseases remains low, attributed to incomplete penetrance and variable expressivity (hindering prediction of sudden death risk in a carrier child), available preventive and therapeutic options, perceived disease burden and lack of information on reproductive options; most couples seeking PGT have a severe phenotype or family history of sudden cardiac death, while those not proceeding choose natural conception without testing, donor gametes or adoption [3].

Follow-up after experimental fetal treatment. Pioneering interventions such as prenatal risdiplam for SMA, by their nature as single-patient proof of concept, require prolonged systematic clinical follow-up—more than two years at publication in the reported case—with neurological, electrophysiological and biomarker assessment before considering extension to a larger cohort in a controlled clinical trial [45]; the observational registry NCT07805278, registered in September 2026, was designed specifically to collect maternal, pregnancy, delivery and postnatal outcomes systematically [46]. Published follow-up of the child with Pompe disease treated in utero extends to 13 months, with cardiac, motor, biomarker and antidrug antibody surveillance [50]. For risdiplam, the US label provides a pregnancy exposure registry applicable to exposed pregnant women [55]. Similarly, any child exposed to experimental prenatal gene or cell therapy should, as a standard regulatory requirement in this field, enter long-term follow-up programmes typically extending fifteen years or more after exposure, because of theoretical risks of late insertional mutagenesis or delayed immune responses to the viral vectors used.

Recurrence counselling for future pregnancies. In all scenarios, closure of the care episode should include a clear documented recurrence risk estimate for future pregnancies—whether based on a defined Mendelian pattern, an empirical population risk or a de novo variant with very low recurrence—and explicit identification of screening or diagnostic options in a subsequent pregnancy, including preconception PGT planning whenever clinically applicable. Comprehensive reproductive risk assessment and prenatal screening and diagnosis references, such as those by Evans, Odibo and Krantz, and Pereira and Soria, remain useful resources for more detailed clinical team consultation during this counselling stage [74–76].

12. Summary and Final Considerations

Contemporary prenatal genetics forms a continuum ideally beginning before conception—with carrier screening and, when indicated, preimplantation genetic testing—and extending through pregnancy through successive layers of non-invasive screening (biochemical, ultrasound and increasingly cell-free fetal DNA) and, when necessary, invasive diagnostic confirmation. NIPT development since the foundational cffDNA discovery in 1997 irreversibly changed the balance between screening and diagnosis, substantially reducing invasive procedure numbers without eliminating their necessity: NIPT remains high-performance screening rather than a replacement for confirmatory cytogenetic diagnosis, a distinction explicitly reinforced by the latest ACOG and ACMG guidelines even as they recommend universal offering. Exome and whole-genome sequencing are increasingly elucidating fetal structural anomalies without a conventional cytogenetic diagnosis, while fetal treatment is progressing from predominantly haematological and transfusion interventions towards pharmacological and gene-based interventions targeting the underlying molecular cause, illustrated by the pioneering prenatal SMA treatment case.

This technological progress reinforces the central role of non-directive genetic counselling: each new layer of information—a positive screen, an ultrasound soft marker, ambiguous mosaicism or an incidental array-CGH finding—adds probabilistic uncertainty requiring clear communication, rigorous statistical context and respect for couples' diverse ethical, religious and cultural values. The increasing popularity of preimplantation genetic testing as an alternative to traditional prenatal diagnosis in several international centres illustrates the movement of reproductive decisions to a point before pregnancy is established, without eliminating conventional prenatal screening and diagnosis in naturally conceived pregnancies or even after fertility treatment, as emphasised by the latest professional guidelines.

Developments in the past twelve months include the SMFM cfDNA update (November 2025), making common trisomy screening routinely available to all pregnant women, retaining sex chromosome aneuploidy screening as an opt-in choice and advising against population screening for microdeletions and genome-wide CNVs [29]; the ESHG and ISPD global prenatal genomic sequencing recommendations (September 2026), restricting sequencing to fetuses with anomalies suggestive of a genetic aetiology, advising against its use to screen fetuses without anomalies, and standardising report content and VUS and incidental finding policies [36]; reversal of the US rule that would have subjected laboratory tests to device regulation (September 2025) [53]; and St. Jude's registration of a registry of in utero-treated SMA cases (September 2026) [46]. None changes the central principle: screening identifies probabilities, diagnosis requires a fetal sample and a method suited to the clinical question, and the decision belongs to the adequately informed couple.

13. References

1. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. Imaging in prenatal diagnosis; p. 742–743.
2. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. Invasive techniques and genetic tests in prenatal diagnosis; p. 744–747.
3. Nannenberg EA, Arens Y. Prenatal diagnosis and preimplantation genetic testing for inherited cardiac diseases. In: Baars HF, van der Smagt JJ, Doevendans PAFM, van Tintelen JP, editors. Clinical Cardiogenetics. 3rd ed. Cham: Springer; 2020. p. 457–461. doi:10.1007/978-3-030-45457-9_27
4. Brock DJH, Sutcliffe RG. Alpha-fetoprotein in the antenatal diagnosis of anencephaly and spina bifida. Lancet. 1972;2(7770):197-199. doi:10.1016/S0140-6736(72)91634-0
5. Lo YMD, Corbetta N, Chamberlain PF, Rai V, Sargent IL, Redman CWG, Wainscoat JS. Presence of fetal DNA in maternal plasma and serum. Lancet. 1997;350(9076):485-487. doi:10.1016/S0140-6736(97)02174-0
6. Gil MM, Accurti V, Santacruz B, Plana MN, Nicolaides KH. Analysis of cell-free DNA in maternal blood in screening for aneuploidies: updated meta-analysis. Ultrasound Obstet Gynecol. 2017;50(3):302-314. doi:10.1002/uog.17484
7. Santorum M, Wright D, Syngelaki A, Karagioti N, Nicolaides KH. Accuracy of first-trimester combined test in screening for trisomies 21, 18 and 13. Ultrasound Obstet Gynecol. 2017;49(6):714-720. doi:10.1002/uog.17283
8. Lord J, McMullan DJ, Eberhardt RY, Rinck G, Hamilton SJ, Quinlan-Jones E, et al. Prenatal exome sequencing analysis in fetal structural anomalies detected by ultrasonography (PAGE): a cohort study. Lancet. 2019;393(10173):747-757. doi:10.1016/S0140-6736(18)31940-8
9. Jones KL, Jones MC, del Campo M. Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. Prenatal diagnosis; p. 950–952.
10. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. Non-invasive prenatal diagnosis/testing (NIPD/T); p. 764–767.
11. Souka AP, Von Kaisenberg CS, Hyett JA, Sonek JD, Nicolaides KH. Increased nuchal translucency with normal karyotype. Am J Obstet Gynecol. 2005;192(4):1005-1021. doi:10.1016/j.ajog.2004.12.093. PMID: 15846173
12. Muzzey D, Goldberg JD, Haverty C. Noninvasive prenatal screening for patients with high body mass index: evaluating the impact of a customized whole genome sequencing workflow on sensitivity and residual risk. Prenat Diagn. 2020;40(3):333-341. doi:10.1002/pd.5603
13. Dungan JS, Klugman S, Darilek S, Malinowski J, Akkari YMN, Monaghan KG, et al. Noninvasive prenatal screening (NIPS) for fetal chromosome abnormalities in a general-risk population: an evidence-based clinical guideline of the American College of Medical Genetics and Genomics (ACMG). Genet Med. 2023;25(2):100336. doi:10.1016/j.gim.2022.11.004
14. American College of Obstetricians and Gynecologists' Committee on Practice Bulletins—Obstetrics, Committee on Genetics, and the Society for Maternal-Fetal Medicine. Screening for fetal chromosomal abnormalities: ACOG Practice Bulletin No. 226. Obstet Gynecol. 2020;136(4):e48-e69.
15. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. Oedema — increased nuchal translucency, cystic hygroma, and hydrops; p. 768–769.
16. Best S, Wou K, Vora N, Van der Veyver IB, Wapner R, Chitty LS. Promises, pitfalls and practicalities of prenatal whole exome sequencing. Prenat Diagn. 2018;38(1):10-19. doi:10.1002/pd.5102
17. Adam MP, Feldman J, Mirzaa GM, Pagon RA, Wallace SE, Bean LJH, Gripp KW, Amemiya A, editors. GeneReviews® [Internet]. Seattle: University of Washington, Seattle; 1993-2026. Available at: https://www.ncbi.nlm.nih.gov/books/NBK1116/
18. Orphanet: the portal for rare diseases and orphan drugs. Paris: INSERM. Available at: https://www.orpha.net
19. Sparks TN, Lianoglou BR, Adami RR, Pluym ID, Holliman K, Duffy J, et al. Exome sequencing for prenatal diagnosis in nonimmune hydrops fetalis. N Engl J Med. 2020;383(18):1746-1756. doi:10.1056/NEJMoa2023643
20. Takasaki K, Kacena MA, Raskind WH, Weiss MJ, Chou ST. GATA1-related cytopenia. In: Adam MP, Feldman J, Mirzaa GM, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993-2026 [updated on 16 Feb 2023]. PMID: 20301538. Available at: https://www.ncbi.nlm.nih.gov/books/NBK1364/
21. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. Hyperechogenic bowel; p. 738–739.
22. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. Short limbs; p. 784–786.
23. Krakow D, Lachman RS, Rimoin DL. Guidelines for the prenatal diagnosis of fetal skeletal dysplasias. Genet Med. 2009;11(2):127-133. doi:10.1097/GIM.0b013e3181971ccb
24. Cuckle HS, Wald NJ, Thompson SG. Estimating a woman's risk of having a pregnancy associated with Down's syndrome, using her age and serum alpha-fetoprotein level. Br J Obstet Gynaecol. 1987;94(5):387-402. doi:10.1111/j.1471-0528.1987.tb03114.x
25. Hook EB, Cross PK, Jackson L, Pergament E, Brambati B. Maternal age-specific rates of 47,+21 and other cytogenetic abnormalities diagnosed in the first trimester of pregnancy in chorionic villus biopsy specimens. Am J Hum Genet. 1988;42(6):797-807.
26. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. Maternal age; p. 754–755.
27. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. Paternal age; p. 772–773.
28. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. Miscarriage and recurrent miscarriage; p. 760–761.
29. Society for Maternal-Fetal Medicine (SMFM); Rink BD, Dugoff L, Kuller JA; SMFM Publications Committee. Society for Maternal-Fetal Medicine Consult Series #74: Cell-free DNA screening for aneuploidies: updated guidance. Pregnancy (Hoboken). 2025;1(6):e70139. doi:10.1002/pmf2.70139. PMID: 42597105
30. Palomaki GE, Smith WE, Bajaj K, Rowsey R, Hurst ACE, Messerlian G, Best RG; ACMG Laboratory Quality Assurance Committee. Prenatal screening for trisomy 21 (Down syndrome) using first- and second-trimester biochemistry and nuchal translucency: a technical standard of the American College of Medical Genetics and Genomics (ACMG). Genet Med. 2025;27(9):101505. doi:10.1016/j.gim.2025.101505. PMID: 40778613
31. Veras GS, Bennini JR. Prenatal screening and diagnostic strategies for fetal genetic abnormalities: comparison of international clinical guidelines. Int J Gynaecol Obstet. 2026 [advance online publication, 8 Aug 2026]. doi:10.1002/ijgo.71259. PMID: 42568303
32. Norton ME, Jacobsson B, Swamy GK, Laurent LC, Ranzini AC, Brar H, et al. Cell-free DNA analysis for noninvasive examination of trisomy. N Engl J Med. 2015;372(17):1589-1597. doi:10.1056/NEJMoa1407349. PMID: 25830321
33. van der Meij KRM, Sistermans EA, Macville MVE, Stevens SJC, Bax CJ, Bekker MN, et al.; Dutch NIPT Consortium. TRIDENT-2: national implementation of genome-wide non-invasive prenatal testing as a first-tier screening test in the Netherlands. Am J Hum Genet. 2019;105(6):1091-1101. doi:10.1016/j.ajhg.2019.10.005. PMID: 31708118
34. American College of Obstetricians and Gynecologists. Prenatal diagnostic testing for genetic disorders: ACOG Practice Bulletin No. 162. Obstet Gynecol. 2016;127(5):e108-e122.
35. Ghi T, Sotiriadis A, Calda P, Da Silva Costa F, Raine-Fenning N, Alfirevic Z, McGillivray G. ISUOG Practice Guidelines: invasive procedures for prenatal diagnosis. Ultrasound Obstet Gynecol. 2016;48(2):256-268. doi:10.1002/uog.15945
36. Deans ZC, Basel Salmon L, Choy KW, de Wert G, Dong Z, Downie L, et al. Global recommendations for the use of diagnostic genomic sequencing in the prenatal setting on behalf of the ESHG and ISPD. Eur J Hum Genet. 2026 [advance online publication, 16 Sep 2026]. doi:10.1038/s41431-026-02217-4. PMID: 42744877. Simultaneous publication: Prenat Diagn. 2026. doi:10.1002/pd.70248
37. Petrovski S, Aggarwal V, Giordano JL, Stosic M, Wou K, Bier L, et al. Whole-exome sequencing in the evaluation of fetal structural anomalies: a prospective cohort study. Lancet. 2019;393(10173):758-767. doi:10.1016/S0140-6736(18)32042-7. PMID: 30712878
38. Van den Veyver IB, Chandler N, Wilkins-Haug LE, Wapner RJ, Chitty LS; ISPD Board of Directors. International Society for Prenatal Diagnosis updated position statement on the use of genome-wide sequencing for prenatal diagnosis. Prenat Diagn. 2022;42(6):796-803. doi:10.1002/pd.6157. PMID: 35583085
39. Monaghan KG, Leach NT, Pekarek D, Prasad P, Rose NC; ACMG Professional Practice and Guidelines Committee. The use of fetal exome sequencing in prenatal diagnosis: a points to consider document of the American College of Medical Genetics and Genomics (ACMG). Genet Med. 2020;22(4):675-680. doi:10.1038/s41436-019-0731-7. PMID: 31911674
40. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. Prenatal diagnosis of sex chromosome aneuploidy; p. 680–681.
41. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. Autosomal reciprocal translocations — prenatal; p. 644–645.
42. Warburton D. De novo balanced chromosome rearrangements and extra marker chromosomes identified at prenatal diagnosis: clinical significance and distribution of breakpoints. Am J Hum Genet. 1991;49(5):995-1013. PMID: 1928105
43. Joint Committee on Genomics in Medicine. Recommendations for the use of chromosome microarray in pregnancy. London: The Royal College of Pathologists; 2015. Available at: https://www.rcpath.org/uploads/assets/06664c28-0f90-4230-86158c91fea14be6/Recommendations-for-the-use-of-chromosome-microarray-in-pregnancy.pdf
44. Salomon LJ, Sotiriadis A, Wulff CB, Odibo A, Akolekar R. Risk of miscarriage following amniocentesis or chorionic villus sampling: systematic review of literature and updated meta-analysis. Ultrasound Obstet Gynecol. 2019;54(4):442-451. doi:10.1002/uog.20353. PMID: 31124209
45. Finkel RS, Hughes SH, Parker J, Civitello M, Lavado A, Mefford HC, Mueller L, Kletzl H; Prenatal SMA Risdiplam Study Group. Risdiplam for prenatal therapy of spinal muscular atrophy. N Engl J Med. 2025;392(11):1138-1140. doi:10.1056/NEJMc2300802
46. St. Jude Children's Research Hospital. A registry of cases of spinal muscular atrophy diagnosed in the prenatal period that received disease-modifying therapy. ClinicalTrials.gov: NCT07805278. Updated on 30 Sep 2026. Available at: https://clinicaltrials.gov/study/NCT07805278
47. Sagar R, David AL. Fetal therapies — stem cell transplantation, enzyme replacement therapy, in utero genetic therapies. Best Pract Res Clin Obstet Gynaecol. 2024;97:102542. doi:10.1016/j.bpobgyn.2024.102542
48. Karolinska Institutet. Boost Brittle Bones Before Birth (BOOSTB4). ClinicalTrials.gov: NCT03706482. Updated on 12 Sep 2025. Available at: https://clinicaltrials.gov/study/NCT03706482
49. University of California, San Francisco. PEARL (PrEnAtal Enzyme Replacement Therapy for Lysosomal Storage Disorders). ClinicalTrials.gov: NCT04532047. Updated on 17 Mar 2026. Available at: https://clinicaltrials.gov/study/NCT04532047
50. Cohen JL, Chakraborty P, Fung-Kee-Fung K, Schwab ME, Bali D, Young SP, et al. In utero enzyme-replacement therapy for infantile-onset Pompe's disease. N Engl J Med. 2022;387(23):2150-2158. doi:10.1056/NEJMoa2200587. PMID: 36351280. PMCID: PMC10794051
51. Waddington SN, Peranteau WH, Rahim AA, David AL, Perocheau DP, Suff N, Attilakos G. Fetal gene therapy. J Inherit Metab Dis. 2024;47(1):192-210. doi:10.1002/jimd.12659
52. U.S. Food and Drug Administration. Genetic non-invasive prenatal screening tests may have false results: FDA safety communication. 19 Apr 2022. Available at: https://www.fda.gov/medical-devices/safety-communications/genetic-non-invasive-prenatal-screening-tests-may-have-false-results-fda-safety-communication
53. Food and Drug Administration. Medical devices; laboratory developed tests; implementation of vacatur (RIN 0910-AJ05; Docket FDA-2025-N-1730; 21 CFR Part 809). Final rule. Fed Regist. 2025 Sep 19;90:45134. Doc. 2025-18239 [accessed 2026 Oct 6]. Available at: https://www.federalregister.gov/documents/2025/09/19/2025-18239/
54. Regulamento (UE) 2017/746 do Parlamento Europeu e do Conselho, de 5 de abril de 2017, relativo aos dispositivos médicos para diagnóstico in vitro. Jornal Oficial da União Europeia. 2017;L 117:176-332. Available at: https://eur-lex.europa.eu/eli/reg/2017/746/oj
55. Genentech. EVRYSDI (risdiplam) for oral solution; EVRYSDI (risdiplam) tablets. Prescribing information (label accessed through openFDA, version of 19 Feb 2026); Drugs@FDA NDA 213535 (initial approval 7 Aug 2020) and NDA 219285 (11 Feb 2025). Available at: https://www.accessdata.fda.gov/scripts/cder/daf/
56. European Medicines Agency. Evrysdi (risdiplam): EPAR — medicine overview and product information. Authorised on 26 Mar 2021; latest European Commission decision of 20 Apr 2026. Available at: https://www.ema.europa.eu/en/medicines/human/EPAR/evrysdi
57. Munné S, Kaplan B, Frattarelli JL, Child T, Nakhuda G, Shamma FN, et al.; STAR Study Group. Preimplantation genetic testing for aneuploidy versus morphology as selection criteria for single frozen-thawed embryo transfer in good-prognosis patients: a multicenter randomized clinical trial. Fertil Steril. 2019;112(6):1071-1079.e7. doi:10.1016/j.fertnstert.2019.07.1346. PMID: 31551155
58. Yan J, Qin Y, Zhao H, Sun Y, Gong F, Li R, et al. Live birth with or without preimplantation genetic testing for aneuploidy. N Engl J Med. 2021;385(22):2047-2058. doi:10.1056/NEJMoa2103613. PMID: 34818479
59. European Society of Human Reproduction and Embryology (ESHRE). Guidelines and legal documents on reproductive medicine. Available at: https://www.eshre.eu/Guidelines-and-Legal
60. American Society for Reproductive Medicine (ASRM). Practice guidance. Available at: https://www.asrm.org/practice-guidance
61. Zegers-Hochschild F, Adamson GD, Dyer S, Racowsky C, de Mouzon J, Sokol R, et al. The International Glossary on Infertility and Fertility Care, 2017. Fertil Steril. 2017;108(3):393-406. doi:10.1016/j.fertnstert.2017.06.005
62. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. Assisted reproductive technology: IVF, ICSI and PGD; p. 706–708.
63. ESHRE Add-ons working group; Lundin K, Bentzen JG, Bozdag G, Ebner T, Harper J, Le Clef N, et al. Good practice recommendations on add-ons in reproductive medicine. Hum Reprod. 2023;38(11):2062-2104. doi:10.1093/humrep/dead184. PMID: 37747409
64. ESHRE PGT Consortium Steering Committee; Carvalho F, Coonen E, Goossens V, Kokkali G, Rubio C, et al. ESHRE PGT Consortium good practice recommendations for the organisation of PGT. Hum Reprod Open. 2020;2020(3):hoaa021. doi:10.1093/hropen/hoaa021. PMID: 32524036
65. Human Fertilisation and Embryology Authority. Approved conditions for preimplantation genetic testing (PGT). Available at: https://www.hfea.gov.uk/pgd-conditions/
66. Hyslop LA, Blakely EL, Aushev M, Marley J, Takeda Y, Pyle A, et al. Mitochondrial donation and preimplantation genetic testing for mtDNA disease. N Engl J Med. 2025;393(5):438-449. doi:10.1056/NEJMoa2415539
67. Human Fertilisation and Embryology Authority. Mitochondrial donation treatment [press release]. 10 May 2023. Available at: https://www.hfea.gov.uk/about-us/news-and-press-releases/2023-news-and-press-releases/mitochondrial-donation-treatment/
68. McFarland R, Hyslop LA, Feeney C, Pillai RN, Blakely EL, Moody E, et al. Mitochondrial donation in a reproductive care pathway for mtDNA disease. N Engl J Med. 2025;393(5):461-468. doi:10.1056/NEJMoa2503658
69. Prior TW; Professional Practice and Guidelines Committee. Carrier screening for spinal muscular atrophy. Genet Med. 2008;10(11):840-842. doi:10.1097/GIM.0b013e318188d069
70. Gregg AR, Aarabi M, Klugman S, Leach NT, Bashford MT, Goldwaser T, et al.; ACMG Professional Practice and Guidelines Committee. Screening for autosomal recessive and X-linked conditions during pregnancy and preconception: a practice resource of the American College of Medical Genetics and Genomics (ACMG). Genet Med. 2021;23(10):1793-1806. doi:10.1038/s41436-021-01203-z
71. Allyse M, Minear MA, Berson E, Sridhar S, Rote M, Hung A, Chandrasekharan S. Non-invasive prenatal testing: a review of international implementation and challenges. Int J Womens Health. 2015;7:113-126. doi:10.2147/IJWH.S67124
72. European Commission. European Reference Networks. Available at: https://health.ec.europa.eu/rare-diseases-and-european-reference-networks/european-reference-networks_en
73. EURORDIS — Rare Diseases Europe. Available at: https://www.eurordis.org
74. Evans MI, editor. Prenatal Diagnosis. New York: McGraw-Hill Professional; 2006.
75. Odibo AO, Krantz D, editors. Prenatal Screening and Diagnosis, An Issue of Clinics in Laboratory Medicine. Philadelphia: Saunders/Elsevier; 2010.
76. Pereira E, Soria J, editors. Handbook of Prenatal Diagnosis Methods, Issues and Health Impact. New York: Nova Science Publishers; 2010.