Prenatal Profile of Cornelia de Lange Syndrome (CdLS): A Review of 53 Pregnancies

Abstract

INTRODUCTION

Cornelia de Lange Syndrome (CdLS) (OMIM# 122470, 300590, 610759) is a genetically heterogeneous dominant multi-system developmental disorder characterized by growth and cognitive retardation; abnormalities of the upper limbs; gastroesophageal dysfunction; diaphragmatic, cardiac, ocular, palatal, and genitourinary anomalies; hirsutism; and characteristic craniofacial features. Facial findings include synophrys with high arched eyebrows, thin lips with a characteristic protrusion of the maxilla, long and smooth philtrum, small nose with anteverted nostrils, long eyelashes, low-set and posteriorly angulated ears, and micrognathia. There is a wide spectrum of limb abnormalities starting with small hands with subtle changes in the phalanges and metacarpal bones, increasing in severity to include varying degrees of oligodactyly, syndactyly, and limb contractures and as severe as almost complete absence of the upper extremities. Complex congenital heart defects and diaphragmatic hernia are common, and have been reported in prenatal diagnosis of CdLS. Intrauterine growth retardation (IUGR) is almost universal in classic CdLS, but often absent in milder cases. Neurologic function is abnormal, including variable degrees of intellectual disability, behavioral difficulties, and delayed or absent speech.

CdLS is genetically heterogeneous. Three genes in the cohesin pathway have been implicated, with mutations in NIPBL causing approximately 60% of cases and mutations in SMC1A and SMC3 being responsible for an additional 5% [Krantz et al., 2004; Musio et al., 2006; Deardorff et al., 2007]. Prevalence is estimated to be as high as 1 in 10,000 [Opitz, 1985], with most cases being sporadic resulting from de novo dominant mutations. Recurrence in siblings has been reported, usually resulting from germline mosaicism. Recurrence risk is estimated at 1% when neither parents has discernible findings of CdLS [Gillis et al., 2004]. While no proven biochemical marker exists to screen for an increased risk of CdLS prenatally, isolated reports of growth delay, characteristic ultrasound findings, and abnormal biochemical markers suggest that there may exist a particular profile of findings that could lead to a prenatal diagnosis of CdLS, especially in families known to be at risk because of a prior affected child.

Twenty-nine cases with prenatal observations have been reported in the literature. Only nine of these were given a prenatal clinical diagnosis of CdLS, based mainly on a pattern of ultrasound findings that included IUGR, limb anomalies, characteristic facial findings, congenital diaphragmatic hernias (CDH) or congenital heart defects (CHD) [Goolsby et al., 1995; Manouvrier et al., 1996; Ackerman and Gilbert-Barness, 1997; Ranzini et al., 1997; Boog et al., 1999; Urban and Hartung, 2001; Lee et al., 2002; Le Vaillant et al., 2004; Price et al., 2005]. Twenty additional cases describe pregnancies with abnormal findings that resulted in a child diagnosed with CdLS at birth [Lacourt et al., 1977; Bruner and Hsia 1990; Drolshagen et al., 1992; Cunniff et al., 1993; Jelsema et al., 1993; Manouvrier et al., 1996; Sekimoto et al., 2000; Huang and Porto, 2002; Marino et al., 2002; Applewhite et al., 2003; Arbuzova et al., 2003; Hulinsky et al., 2005; Niu et al., 2006; Kennelly and Moran, 2007]. While IUGR is almost universal in classic CdLS [Kliewer et al., 1993], it may not occur until late gestation, or be absent. Upper limb reduction defects are seen in nearly a third of patients, and are often associated with a more severe presentation of CdLS [Jackson et al., 1993; Kennelly and Moran, 2007]. Mid-trimester pregnancy associated plasma protein A (PAPP-A) has been shown to be significantly reduced in pregnancies affected by CdLS [Westergaard et al.,1983; Aiken et al., 1999].

In this report, we review the prenatal findings in 29 literature and 24 previously unreported CdLS cases in order to develop a prenatal profile predictive of CdLS.

MATERIALS AND METHODS

Under an IRB-approved protocol of informed consent, information on pregnancies which resulted in the birth or termination of an infant with CdLS was collected from several centers. For cases diagnosed outside of our institution, clinical details, photographs, and in some cases, blood samples, were obtained to confirm the diagnosis of CdLS. Twenty-four previously unreported cases from the USA, Canada, and Australia were identified, in which CdLS was diagnosed clinically and prenatal ultrasound images, reports, and/or biochemical screening markers were available. Findings identified through ultrasound imaging were reviewed. When possible, nuchal translucency (NT), growth parameters for gestational age, structural anomalies, and facial profiles were evaluated from available ultrasonographic images. PAPP-A level for the patient described in CDL-141-AF was obtained at the Foundation for Blood Research, Scarborough, Maine. Fetal biometric measurements were assigned percentiles for gestational age (GA) based on the data in Diagnostic Images of Fetal Anomalies [Nyberg et al.,2003].

Review of the literature resulted in 29 prenatal cases reported between 1977–2009.

CASE REPORTS

In addition to the 29 literature cases, we summarized the clinical findings in the 24 newly identified patients, including six diagnosed prenatally with CdLS, 15 pregnancies with abnormal findings that resulted in a diagnosis of CdLS at or after birth and three unremarkable pregnancies resulting in an infant born with CdLS. (see Supplemental Case Reports in Supporting Information online).

RESULTS

Data from the 53 patients are summarized in Table I, and clinical synopses of the 24 newly reported cases are documented in the Supporting Information online in Supplemental Case Reports. Prenatals suggestion of CdLS was based on a constellation of abnormalities on second trimester ultrasound, including IUGR, limb anomalies, characteristic facial findings, and major organ anomalies. In eight cases (15%), an abnormally elevated NT and/or low PAPP-A contributed to the diagnosis.

Intra-uterine growth retardation was the most common ultrasound finding, affecting 43/53 pregnancies. Although IUGR was detected on average at 25 weeks (late in the second trimester) and was identified as early as 16 weeks [Bruner and Hsia, 1990], it was minimal in its early appearance (Fig 1). Not until later gestation did it become significant enough in itself to warrant investigation. IUGR was generally symmetric, affecting all measurements, and progressive throughout the late second and third trimester. Table II documents the growth parameters and centiles/standard deviations when detailed sonographic measurements were available. Although IUGR was often not formally diagnosed until growth parameters were below the 5^th–10^th centiles, growth parameters were below the 50^th centile for the majority of scans performed earlier in pregnancy.

Open in a separate window

Figure 1

Plot of fetal growth measurements of four patients with CdLS in whom multiple measurements were available. UA, ultrasound age; GA, gestational age.

Limb anomalies were the finding most suggestive for CdLS, and the most helpful finding leading to a prenatal diagnosis. In total, 35/53 patients had limb abnormalities, ranging in severity from small hands or flexed forearms to complete absence of the upper limbs, evident as early as fourteen weeks, and generally reported on the first ultrasound during the pregnancy. Of the 15 prenatally diagnosed patient, fourteen had limb abnormalities (93%) evident on ultrasound between 13 and 34 weeks (Case 1, Case 2, Case 3, Case 4, Case 5, Case 6 and [Manouvrier et al., 1996; Ackerman and Gilbert-Barness, 1997; Ranzini et al., 1997; Boog et al., 1999; Urban and Hartung, 2001; Lee et al., 2002; Le Vaillant et al., 2004; Price et al., 2005]). In contrast, of the 38 cases without a prenatal diagnosis, only 20 (53%) had significant upper extremity abnormalities noted on ultrasound (Case 10, Case 14, Case 15. Case 16, Case 19, Case 21 and [Drolshagen et al., 1992; Manouvrier et al., 1996; Sekimoto et al., 2000; Huang and Porto 2002; Marino et al., 2002; Applewhite et al., 2003; Arbuzova et al., 2003; Le Vaillant et al., 2004; Hulinsky et al., 2005; Lemire 2006; Lalatta et al., 2007].

Characteristic facial features were reported in 26 of 53 (49%) individuals. This profile was evident by 19–20 weeks. Reports generally described micrognathia with prominent maxilla, prominent skin of the forehead, prominent eyelashes, depression of the nasal root or small nose with anteverted nares, and a long smooth philtrum with prominent convexity. When described, the facial profile became a critical finding in narrowing the differential diagnosis to CdLS. Characteristic ultrasound findings are summarized in Figure 2.

Open in a separate window

Figure 2

Ultrasound findings from select cases with postnatal correlation. a-f) Facial profiles in CdLS pregnancies: a) Facial features at 21 weeks and subsequent 3-D ultrasound (“Case 20” in supplementary digital content); b) Facial features at 28 weeks and postnatal facial features at 28 2/7 weeks (“Case 21” in supplementary digital content); c) Facial profile, ultrasound at 19 weeks, postnatal profile at 1 year (“Case 8” in supplementary digital content); d) Prenatal ultrasound profile at 23 weeks and immediate postnatal profile (“Case 2” in supplementary digital content); e) Prenatal ultrasound of full face at 19 weeks and postnatal full face at 1 year (“Case 8” in supplementary digital content); f) Prenatal ultrasound profile at 18 weeks and postnatal profile at 16 months (“Case 13” in supplementary digital content); g-h) Limb findings in CdLS pregnancies: g) Prenatal ultrasound at 19 weeks showing oligodactyly of upper limbs and 20 week autopsy images (“Case 19” in supplementary digital content); h) Prenatal ultrasound at 23 weeks of right limb showing severe oligodactyly with postnatal conformation (“Case 2” in supplementary digital content).

Surprisingly, life-threatening anatomic defects, such as CDH and CHD, led to prenatal suggestion of CdLS in only 23/53 pregnancies. In two, CDH on prenatal ultrasound led to a clinical diagnosis of Fryns syndrome, which was amended after characteristic CdLS features were identified at birth. Indeed, the prenatal finding of CDH often led to a wide differential diagnosis of genetic disorders. Of the 23 cases with congenital abnormalities, 16 described CDH and 10 CHD (3 had both). Cardiac abnormalities included VSD (4), ToF (3), ASD (1), tortuous aortic arch (1), and one complex, previously undescribed cardiac abnormality (Case 5).

Increased NT/NF (nuchal fold) was described in 6 of 13 cases with measurement (46%). Two additional patients had cystic hygroma, which in one resulted in redundant nuchal skin. Elevated NT/NF was seen in the first and second trimester and ranged from 3.3 mm to 8 mm, with an average of 6.2 mm.

PAPP-A was measured in six pregnancies. Four had a normal value of PAPP-A (Case1, Case 18, Case 22, and Case 24). One had a low PAPP-A value at 0.20 MoM at 11 weeks [Arbuzova et al., 2003] and another had a PAPP-A of 0.16 MoM at 14 weeks (Case 6).

Fifteen pregnancies provided data on maternal serum screening. Seven had completely normal results, with low risk for aneuploidy and neural tube defect. Of the remaining, two had a neural tube defect risk of 1 in 227 (Case 20) and 1 in 260 [Applewhite et al., 2003]; one had a AFP level at 0.4 MoM [Bruner and Hsia, 1990]; one had a trisomy risk greater than 1:10 with AFP at 2.02 MoM, uE3 at 0.47 MoM, and hCG at 0.22 MoM [Marino et al., 2002]; one had a trisomy 18 risk of 1:40 with AFP at 1.02 MoM, uE3 at 0.60MoM, hCG at 0.23MoM, and amniotic AFP 1.58MoM (Case 13); Case 6 had an increased risk for Down syndrome; and Case 7 had an increased risk for Trisomy 18 of 1 in 19 with an AFP at 1.20 MoM, uE3 at 0.73 MoM, hCG at 0.48 MoM, and inhibin A at 1.53 MoM.

Karyotype, generally via amniocentesis, CVS, or skin fibroblast sampling, was reported in 36 individuals. Three were found to have chromosomal abnormalities. One case reported a translocation between chromosomes 3 and 5 (46,XX,t(3;5)(q21;p13)), which was thought to lead to a severe CdLS phenotype [Price et al., 2005]. A second case detected a large deletion on chromosome 5 involving the entire NIPBL region (46,XY,del(5)(p13.1p14.2)) [Hulinsky et al., 2005]. A third, unreported case, detected a balanced reciprocal translocation between chromosomes 16 and 20 (46,XX,t(16;20)(p13.1;p11.2)). This patient also had a postnatally detected NIPBL truncating mutation, and though the translocation is de novo, it is likely not responsible for the CdLS phenotype (Case 13).

In addition to the findings reported above, these cases described a number of other ultrasound abnormalities. Four patients had neurologic findings on sonogram, including cerebellar vermis hypoplasia, agenesis of the cerebellar vermis, ventriculomegaly, dangling choroid plexus, and agenesis of the corpus callosum. In addition, several had umbilical artery abnormalities, including a two-vessel cord in five, and abnormal uterine artery Doppler with middle cerebral artery sparing in one, and absent to reverse diastolic flow in another. Cardiac anomalies included heart displacement, cystic structure behind the heart, tortuous aortic arch, pericardial effusion, left ventricular hypertrophy, and persistent left superior vena cava. Other findings included absent nasal bone, fisting/skin webbing, pleural effusion, fetal edema, renal pyelectasis, echogenic kidneys, premature ossification of the sternum, dilated loops of bowel, and omphalocele. Eight cases reported oligohydramnios and six cases reported polyhydramnios during the course of the pregnancy.

DISCUSSION

Conclusion

The vast majority of infants born with CdLS are sporadic and recurrence in families with unaffected parents is rare. Recurrence due to germ line mosaicism in such families has been estimated to be 1% [Gillis et al., 2004]. For families with documented germ line mosaicism (recurrence of CdLS in a subsequent pregnancy to unaffected parents who do not carry the familial mutation in their studied DNA samples), the risk for subsequent recurrence is as high as 50%. For these families, and especially for those in whom a causative mutation has not been identified, a diagnostic prenatal profile would be useful and reassuring in subsequent pregnancies. Even for families with a single child or prior pregnancy affected by CdLS, the value of reassurance should not be underestimated. Every effort should be made to document a mutation in the affected child in order to allow for testing of future pregnancies.

Anxiety may be high during subsequent pregnancies in families in which a causative mutation cannot be identified. Our study suggests that IUGR and facial differences may be evident earlier than clinically reported. A retrospective review of the initial ultrasound in Case 14 showed micrognathia and upper limb deformities that could have prompted an earlier diagnosis, and IUGR was evident from 18 weeks. Case 11 is instructive for early prenatal diagnosis of CdLS. Ultrasonographic anatomic evaluation should begin early and be repeated frequently. NT can be measured at 11 weeks, and PAPP-A can be assessed from 9 weeks to early mid trimester. As these two measures are increasingly performed routinely for aneuploidy screening in the first trimester, reporting of these values from CdLS pregnancies would allow construction of CdLS risk figures for future pregnancies based on this screening. Early ultrasounds should assess upper limb abnormalities, and later ultrasounds should evaluate facial profile for the characteristics described above. As ultrasound technology advances, three-dimensional sonography could become a useful diagnostic tool for early detection of CdLS. Le Vaillant et al. [2004] described a case detected at 28 weeks via traditional ultrasound with IUGR, VSD, very mild limb abnormalities, and facial findings. Three-dimensional sonography allowed a more detailed look at the facial dysmorphia, and revealed the bulging forehead, depressed nasal root, large philtrum, severe micrognathia, and hirsutism characteristic of CdLS. In addition, they utilized three-dimensional computed tomography of the fetal skeleton to further characterize the upper limb defect as hypoplasia of the radial head and shortened first metacarpal. While in this case the diagnosis likely could have been made based on the 2-D findings, use of 3-D ultrasound and CT could be useful in narrowing a differential diagnosis or estimating the disease severity. Figure 3 provides a flow chart suggesting an algorithm for the prenatal diagnosis of CdLS in families with and without a prior CdLS pregnancy.

Open in a separate window

Figure 3

Flow chart of recommended approach to prenatal diagnosis of CdLS for families with or without a prior CdLS pregnancy.

In a recent publication by Pajkrt et al. [2010], seven CdLS cases were retrospectively reviewed for prenatal findings. In four CdLS was considered prenatally and confirmed shortly after birth. All seven had asymmetrical upper limb defects, five had IUGR, and five had facial features suggestive of CdLS identified by ultrasound. An additional 28 previously reported cases were reviewed as well. Their analysis of the 35 cases concluded that IUGR was identified in 80%, and upper limb anomalies in nearly half. Characteristic facial features, internal organ defects and increased nuchal translucency were commonly seen and aided in diagnosis confirmation.

Many families at average risk undergo prenatal screening tests. Prenatal diagnosis must be made during the mid-trimester to allow for possible termination. Diagnoses made later in pregnancy can be helpful for anticipatory guidance and newborn management. Early signs of fetal difficulty, such as IUGR and increased NT, should suggest a potential CdLS diagnosis after more common causes have been excluded, and the proposed algorithm may be followed.

Supplementary Material

ACKNOWLEDGMENTS

REFERENCES