Editor—Diaphragmatic hernia occurs with an incidence of 1/4000-1/5000 newborns.1 2 About half of the cases occur as isolated malformations; anomalies of the central nervous system are well known associated defects, as are midline malformations such as cleft lip and palate and heart defects.1 2 Among syndromic causes of diaphragmatic hernia are numerous chromosomal imbalances (4p deletion, trisomy 18, i(12p), and some single gene defects such as the autosomal recessive Fryns syndrome3and an X linked form, the thoracicoabdominal syndrome.4 5Fryns syndrome is probably the most frequent non-chromosomal malformation syndrome associated with diaphragmatic hernia,3 6 but this disorder presents with typical dysmorphic features, and both heart and CNS malformations are infrequent in Fryns syndrome.

The family reported here, along with several other published reports, combining CNS, heart, and diaphragmatic defects, may help to delineate a distinct syndromic form of diaphragmatic hernia.

Healthy, non-consanguineous parents, of French origin and with unremarkable family histories, had six unsuccessful pregnancies. Two resulted in first trimester spontaneous abortion. The other four led to pregnancy terminations between 15 and 23 weeks' gestation because of malformations detected on ultrasound (table 1). Left diaphragmatic hernia was the earliest observed and most constant feature. Cardiac malformations, present in all cases, were of varying types: VSD/ASD was seen twice and hypoplastic left heart and a complex heart malformation once each. Hydrocephalus was also constant on morphological examination of the fetus, although not yet visible on ultrasound in the pregnancy terminated at 15 weeks. In cases 3 and 4, examined by the same neuropathologist, this was a tetraventricular, non-obstructive hydrocephalus. The only inconstant malformation was imperforate anus, described in one male fetus. Muscle histology was normal in the two cases investigated in detail.

We examined the last two fetuses personally. No major dysmorphic features were present (fig 1). Specifically, the craniofacial features and nail/distal digit hypoplasia typical of Fryns syndrome were lacking.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Fetuses 3 (A) and 4 (B) after pregnancy termination (18 and 16 weeks' gestation). Note lack of distinct dysmorphic features.

Genetic analyses included karyotyping of fetuses 2-4 on amniocentesis (one) or fibroblasts post-termination (one), or both (one). Cytogenetic studies using RHG, RTBG, and GTBG banding were normal at the 550-600 band resolution level. Standard karyotypes of both parents were also normal, 46,XX and 46,XY. A 22q11 microdeletion was also excluded in the parents and one fetus by fluorescence in situ hybridisation using a critical region probe (Oncor). The parents were investigated with a probe set covering 41 subtelomeric regions (Cytocell Telomeric Multiprobe), which confirmed the presence of all telomeres. Analysis of fetus number 4 with the MultiFish Spectra Vision set (Vysis) was also normal.

We present here a couple whose six pregnancies resulted in two spontaneous abortions and four pregnancy terminations for multiple malformations. The malformed fetuses, two male and two female, presented with a remarkably similar malformation syndrome comprising left diaphragmatic hernia, heart malformation, and non-obstructive hydrocephalus of early onset. The diaphragmatic hernia was suspected as early as 12 weeks on ultrasound and confirmed by 15-16 weeks; the hydrocephalus developed somewhat later but was clearly present on echographic examination at 19-20 weeks' gestation.

A chromosomal anomaly could have explained the combination of first trimester miscarriage and malformed fetuses in the same family, but cytogenetic and molecular cytogenetic studies, including the use of telomeric probes for detection of a cryptic translocation, were repeatedly normal. We considered a mitochondrial disorder, given that all six pregnancies were abnormal. However, we know of no precedent for such an aetiology in malformation syndromes and fetal muscle histology was normal.

An autosomal recessive aetiology seems the most likely, given four affected fetuses of both sexes born to normal parents. The constancy of the phenotype is anecdotal evidence for a recessive disorder, as is the resemblance to Fryns syndrome, which follows this hereditary transmission. Among syndromic causes of diaphragmatic hernia, Fryns syndrome is the closest clinically. However, the typical dysmorphic features of facies and limbs were lacking in the fetuses and CNS pathology is not frequent in this disorder. Heart malformations are known but are rarely severe and imperforate anus is rarely, if ever, described.3 6 7

The thoracicoabdominal syndrome4 5 combines diaphragmatic defects with CNS anomalies, including hydrocephalus in some cases, but there are other midline defects (omphalocele, clefts of lip and palate) and inheritance is X linked recessive.5 The family reported here had similarly affected fetuses of both sexes. We also considered the CHARGE association and the hydrolethalus syndrome7 8 in the differential diagnosis, but excluded them given inadequate clinical fit (table 2).

Several other families, some of which are described within the Fryns syndrome entry (MIM 229850) in the catalogue ofMendelian inheritance in man,9 have been reported with a malformation complex similar to that of our patients. Fitch et al 10 reported a child with absent left hemidiaphragm, arhinencephaly, and cardiac malformations born to consanguineous parents; as distal digital hypoplasia was described, it is possible that their case represents Fryns syndrome, which was not delineated nominally until the following year. However, other publications also report the association of diaphragmatic hernia and CNS anomalies, sometimes in association with heart defects.11 Bieber et al 2 presented a family in which two sibs, of different sexes, had a right diaphragmatic hernia, pulmonary hypoplasia/agenesis, and hydrocephalus. Their family differs from the one we report in that the hernias were right sided (the rarer form), there was no cardiac defect, and the hydrocephalus was obstructive.

Therefore, although we are not aware of another reported family with exactly the same combination of anomalies, there is some precedent for the combination of heart, CNS, and diaphragmatic defects outside Fryns syndrome. We strongly suggest, based on the family reported here, the existence of a monogenic, probably autosomal recessive aetiology for this combination. Whether this disorder will prove to be the result of a distinct mutated gene, or might perhaps be allelic with that for Fryns syndrome, will await identification of the gene or genes responsible. In the meantime, descriptions of additional families would greatly aid genetic counselling and prenatal diagnosis in severe cases such as the family presented here.