Holoprosencephaly is the most common structural malformation of the forebrain, with a prevalence of 1:8,000 in the second trimester.1 It is characterized by the failure of the forebrain to bifurcate into two hemispheres, a process that is normally complete by the fifth week of gestation. The condition is classified according to the degree of division. In its most severe form, termed alobar holoprosencephaly, there is no interhemispheric fissure; a single brain ventricle is present, and there may be cyclopia or a proboscis-like nasal structure. The interhemispheric fissure is present only posteriorly in semilobar holoprosencephaly, and in the lobar form, most of the cerebral hemispheres and lateral ventricles are separated.2
Although many fetuses with holoprosencephaly miscarry spontaneously, there is a common misconception that children with holoprosencephaly do not survive beyond infancy. Early mortality is common in severe forms of holoprosencephaly but those with mild to moderate forms can survive into childhood and beyond.3 Affected children experience many medical and neurologic problems, including mental retardation, epilepsy, spasticity, dystonia, and endocrine disorders, such as diabetes insipidus and growth hormone deficiency.4,5
Prenatal diagnosis of holoprosencephaly can be made by ultrasonography, from the first trimester, particularly when the more severe types of holoprosencephaly, such as alobar and lobar, are present. The milder forms however, such as semilobar, may be more difficult to diagnose early in pregnancy. Facial malformations are commonly associated and include cyclopia, hypotelorism, and cleft lip and palate. Other abnormalities include polydactyly, exomphalos, renal dysplasia, and hydrops. In up to 40% of cases the fetus is aneuploid, commonly trisomy 13. Fetal karyotyping is offered prenatally to women whose fetus has holoprosencephaly. The recurrence risk for trisomy 13 is low, but can be higher if an unbalanced rearrangement is detected secondary to a balanced familial translocation. In euploid fetuses, however, mutations in holoprosencephaly genes, such as sonic hedgehog (SHH), are known to occur. The recurrence risk has been estimated to be 6% in studies of families.6 In this study we sought to determine the underlying cause and, hence, the recurrence risk of holoprosencephaly for women referred to a regional tertiary referral fetal medicine unit to inform good practice regarding their management.
MATERIALS AND METHODS
This is a retrospective cohort study of holoprosencephaly cases referred to the Fetal Medicine Unit at University College London Hospital from June 1991 to June 2006 for a detailed ultrasound scan. The Fetal Medicine Unit database, the North East London Regional Cytogenetics Laboratory, and the Department of Clinical Genetics database at Great Ormond Street Hospital were searched using the terms holoprosencephaly, trisomy 13, trisomy 18, and facial cleft, and case notes were examined to confirm the diagnosis of holoprosencephaly. We examined maternal, neonatal, genetic, and pathology records for prenatal and postnatal management, outcome, and genetic follow-up. Postmortem findings and outcomes were ascertained from the perinatal pathology department, referring hospitals, regional genetics centers, and congenital malformation registers. This study was approved by the Institutional Review Board of the University College London Hospital. Exact binomial confidence intervals were calculated using Stata 9.2 (StataCorp LP, College Station, TX).
During the 15-year study period, 43 women were seen on the Fetal Medicine Unit whose fetus had an antenatal diagnosis of holoprosencephaly (Fig. 1). In one woman, ventriculomegaly was diagnosed at 20 weeks of gestation, but holoprosencephaly was only suspected at 33 weeks; postnatally the diagnosis was amended to an absent septum pellucidum and a small corpus callosum in the infant after magnetic resonance imaging examination. One woman moved away from the area before delivery and was lost to follow-up. We excluded these two patients.
There were 41 women with complete follow-up. The mean gestational age at diagnosis was 19 weeks of gestation (median 20 weeks, range 12–28 weeks, Table 1); seven cases were diagnosed in the first trimester, of which one was a recurrence in a woman who had five previously affected pregnancies (see below). The mean maternal age was 32 years (median 33 years, range 16–42 years, Table 1). There was one multiple pregnancy (twins) with a normal co-twin. The type of holoprosencephaly was predicted to be alobar in 24 cases (58.5%), semilobar in nine cases (22%), and lobar in eight cases (19.5%). Postmortem examination was performed in 16 cases and confirmed the type of holoprosencephaly suggested by ultrasound examination in 13 cases; a postnatal scan confirmed the diagnosis in two further cases, showing that the prenatal ultrasound was correct in identifying the type of holoprosencephaly in 83% of cases examined.
Prenatal karyotyping was offered to all women, and accepted by 36 (88%). Of those karyotyped, 21 (58%) were abnormal, including 18 with trisomy 13 (50%), two with trisomy 18 (6%), and one with a ring chromosome 13 (3%). Of the five women who declined karyotyping, two fetuses had additional structural abnormalities consistent with an abnormal karyotype such as a cardiac abnormality (atrioventricular septal defect and large ventricular septal defect) and echogenic kidneys; in the remaining three fetuses the holoprosencephaly was isolated.
In 15 of the 36 women opting for invasive testing (42%), the karyotype of the fetus was normal. Three women with affected fetuses and a normal karyotype had a recurrence of holoprosencephaly (three of 41 cases, 7.3%). Therefore, 20% (95% confidence interval [CI] 4–48%) of women with a confirmed euploid fetus had a recurrence. If the three cases with isolated holoprosencephaly that were not karyotyped had been euploid, then three of 18 women (17%, 95% CI 4%–41%) would have had a recurrence. None of the women with a recurrence were in a consanguineous relationship, and they did not have any risk factors known to be associated with holoprosencephaly, such as type 1 diabetes, smoking, alcohol, or salicylate medication. All women with a euploid fetus were reviewed postnatally with their partners in the genetics clinic, and parental karyotyping was performed. Of the 12 women with a single euploid case, five are known to have had at least one further pregnancy with a normal outcome; no data are available for the other seven women. The 15 women with an affected euploid fetus attempted 45 pregnancies in total, of which 12 were normal live births, 11 miscarried in the first trimester, and 22 were affected cases of holoprosencephaly.
In one woman with recurrent holoprosencephaly, there were five previously affected pregnancies; all had a euploid karyotype. The first case was diagnosed at 20 weeks of gestation, whereas the next five recurrences were diagnosed by 14 weeks of gestation on ultrasound scan. The family history revealed a maternal sibling who had died 20 years earlier with microcephaly and a bilateral cleft lip and palate. Detailed examinations, including magnetic resonance imaging scanning of this mother, her mother, and surviving siblings were unremarkable; karyotypes were also normal. Some years after the first affected pregnancy, molecular analysis of the affected fetuses showed a deletion in the sonic hedgehog gene (c.415–416delCT, p.Leu139ArgfsX13, exon 2), which was subsequently detected in the mother, grandmother, uncle, and first cousin. In two recurrence cases there was one previously affected pregnancy that was euploid; all affected pregnancies were diagnosed at birth (n=1) or in the second trimester (n=3). In two women with recurrences diagnosed at 22 and 24 weeks, there was one previously affected pregnancy that was euploid, diagnosed at 27 weeks and birth, respectively. A missense mutation was found in the VAX2 gene in one case; parental clinical examination was normal. The other woman with a recurrence declined follow-up.
The majority of women (35 of 41, 85%) chose to terminate the pregnancy. Of those who decided to continue, two had stillbirths and four had a live birth, of whom three children died within the first 2 years of life.
Our findings are in agreement with other studies that show holoprosencephaly is associated with chromosomal abnormalities, especially trisomy 13.7 Importantly, in our series we found that, in 20% (95% CI 4–48%) of women with a confirmed euploid fetus, there was a recurrence of holoprosencephaly. This is a relatively small study, and the recurrence risk is higher than has previously been published. However the recurrence risks from other published studies fall within the 95% CI calculated for our observed recurrence risk, which supports our findings. One study of 30 families gave an empiric recurrence risk of 6%.6 A survey in the west of Scotland identified 28 fetuses of 50 cases of holoprosencephaly with a normal karyotype. Of these, there were three families where a recurrence or related cerebral malformation or mental handicap was found, giving an overall recurrence risk for serious neurologic disability of 12%.8 A genetic model using segregation analysis performed on 79 nuclear families with nonsyndromic and nonchromosomal holoprosencephaly predicted a recurrence risk of up to 14%.9 Although our study is not population based, our findings support the high recurrence risk that was calculated in this genetic model.
Referral to a geneticist is important in a woman whose fetus is euploid because of the risk of recurrence of holoprosencephaly. Prenatal detection of holoprosencephaly has improved in the last 20 years.1,10 Holoprosencephaly can be detected in the first trimester, as we have shown, although the sensitivity of ultrasound diagnosis before 24 weeks of gestation was 69% and up to 77% overall in one recent study.10
Clinical examination of the parents or other family members is useful to elicit subtle dysmorphic signs such as microcephaly, a single maxillary central incisor, flattening of the nose with an abnormal columella, bifid uvula, absent labial frenula, or hypotelorism, which would suggest a familial holoprosencephaly.11 A proportion of these families have mutations identified in the SHH, TGIF, ZIC2, and SIX3 genes12 as was found in one of our recurrence cases. The SHH gene plays a critical role in early forebrain and central nervous system development. Mutations in the SHH gene seem to be the main causal gene implicated in holoprosencephaly to date, being observed in 8.5% of patients in one cohort.12 Submicroscopic chromosome deletions, only detectable by fluorescent in-situ hybridization, may also account for a significant proportion. The VAX2 gene that was mutated in another recurrence case is involved in the control of eye development.13
The survival rate of children with holoprosencephaly is much lower for those with cytogenetic abnormalities or those with malformations in other parts of the body than for those with isolated holoprosencephaly. A recent review from a United Kingdom regional congenital malformation register showed that seven of 16 liveborn children with holoprosencephaly survived infancy.10 Children with “less severe” forms such as lobar holoprosencephaly can survive for longer and some may live well into adulthood.7 However, these children still have major difficulties such as profound mental retardation and brain stem dysfunction causing feeding and breathing difficulties and recurrent infections.
A recommended appropriate work-up for a pregnancy complicated by fetal holoprosencephaly would begin with fetal karyotyping to detect aneuploidy. If the fetal karyotype is normal, the parents should be referred for genetic counseling, including a detailed family history, clinical examination for subtle dysmorphic features that might suggest minor degrees of holoprosencephaly, and parental mutation analysis. A fetal postmortem examination should be performed. In future pregnancies, women should be offered a detailed ultrasound scan in the first trimester to diagnose a severe recurrence early and a further ultrasound scan in the second trimester to look for more subtle abnormalities.
In conclusion, genetic review is important in those women who have a euploid fetus with holoprosencephaly to adequately counsel and estimate the risk of recurrence in a subsequent pregnancy, as well as initiating appropriate molecular studies. Identification of mutations responsible for holoprosencephaly can confer significant recurrence risks in a clinically normal family. In our series, the risk of recurrence was found to be 20% (95% CI 4–48%) for women whose fetus was karyotypically normal, and such women should be scanned early in their next pregnancy to detect recurrence of holoprosencephaly.
1. Bullen PJ, Rankin JM, Robson SC. Investigation of the epidemiology and prenatal diagnosis of holoprosencephaly in the North of England. Am J Obstet Gynecol 2001;184:1256–62.
2. Nanni L, Schelper RL, Muenke MT. Molecular genetics of holoprosencephaly. Front Biosci 2000;5:D334–42.
3. Stashinko EE, Clegg NJ, Kammann HA, Sweet VT, Delgado MR, Hahn JS, et al. A retrospective survey of perinatal risk factors of 104 living children with holoprosencephaly. Am J Med Genet A 2004;128:114–9.
4. Plawner LL, Delgado MR, Miller VS, Levey EB, Kinsman SL, Barkovich AJ, et al. Neuroanatomy of holoprosencephaly as predictor of function: beyond the face predicting the brain. Neurology 2002;59:1058–66.
5. Barr M Jr, Cohen MM Jr. Holoprosencephaly survival and performance. Am J Med Genet 1999;89:116–20.
6. Roach E, Demyer W, Conneally PM, Palmer C, Merritt AD. Holoprosencephaly: birth data, genetic and demographic analyses of 30 families. Birth Defects Orig Artic Ser 1975;11:294–313.
7. Croen LA, Shaw GM, Lammer EJ. Risk factors for cytogenetically normal holoprosencephaly in California: a population-based case-control study. Am J Med Genet 2000;90:320–5.
8. Whiteford ML, Tolmie JL. Holoprosencephaly in the west of Scotland 1975–1994. J Med Genet 1996;33:578–84.
9. Odent S, Le Marec B, Munnich A, Le Merrer M, Bonaïti-Pellié C. Segregation analysis in nonsyndromic holoprosencephaly. Am J Med Genet 1998;77:139–43.
10. Ong S, Tonks A, Woodward ER, Wyldes MP, Kilby MD. An epidemiological study of holoprosencephaly from a regional congenital anomaly register: 1995–2004. Prenat Diagn 2007;27:340–7.
11. Lazaro L, Dubourg C, Pasquier L, Le Duff F, Blayau M, Durou MR, et al. Phenotypic and molecular variability of the holoprosencephalic spectrum. Am J Med Genet A 2004;129:21–4.
12. Dubourg C, Lazaro L, Pasquier L, Bendavid C, Blayau M, Le Duff F, et al. Molecular screening of SHH, ZIC2, SIX3, and TGIF genes in patients with features of holoprosencephaly spectrum: mutation review and genotype-phenotype correlations. Hum Mutat 2004;24:43–51.
13. Kim JW, Lemke G. Hedgehog-regulated localization of Vax2 controls eye development. Genes Dev 2006;20:2833–47.