Twins with omphalocele in Denmark (1970–1989)^†

INTRODUCTION

Omphalocele is an abdominal wall defect limited to an open umbilical ring, with organs herniated into the open base of the umbilical cord covered by membranes. Omphalocele must be distinguished from gastroschisis, in which the abdominal wall defect is located laterally to a normally closed umbilical ring with herniation of uncovered organs [Moore and Stokes, 1953].

The cause of omphalocele is unknown and very little is known about the role of genetic factors.

Study of the occurrence of the malformation in twins is a method to evaluate the role of genetic factors. To our knowledge, no study of omphalocele in a representative sample of twins has been reported. Twins with omphalocele have been reported, but they are rather few and most are sporadic case reports.

A Danish registry of 469 live- and stillborn children with abdominal wall defects is a research registry based on a nation-wide epidemiological study of 20 Danish birth cohorts. The registry contains 258 cases of omphalocele and 166 of gastroschisis, classified according to the above-mentioned definition [Moore and Stokes, 1953]. A tentative estimate of the completeness of the register is 95% of live born and 90% of stillborn infants [Bugge and Holm, 2002]. The registry is based on all the material available in Danish healthcare system and contains clinical information retrieved from all sources.

We report on seven pairs of twins in which one or both twins were born with omphalocele ascertained in the Danish registry of infants with abdominal wall defects from 1970 to 1989.

MATERIALS AND METHODS

RESULTS

Seven twin pairs included eight omphalocele children. Two pairs were monozygotic, one of them was a pair of conjoined twins. Two pairs were of different sex and thus probably dizygotic. In three sex-like pairs, zygosity could not be determined as the children had died. Discordance for omphalocele was found in all twin pairs except the conjoined twins. Three twins had omphalocele as isolated malformation (case 1970; 1978, a; and case 1983, a) and five had omphalocele and associated malformations (case 1978, b; 1978, c; 1983, b: twin A and B; and case 1989) (Table I). Autopsy was done on the stillborn and the livebirths who died.

In the 20-year-period 1,244,138 children were born in Denmark and 25,670 (12,835 pairs) were twins which is 2% [Sundhedsstyrelsen, 1991]. In the 20-year-period 258 children with omphalocele were identified [Bugge and Holm, 2002]. The eight twins with omphalocele represent 3.1% of all. This is not significantly different from the expected (χ² = 1.305, df: 1, 0.50 > P > 0.10, NS).

The MZ twins (case 1978a) and the conjoined twins had normal karyotypes.

No case of twins with gastroschisis occurred in the same period. One monozygotic discordant pair was born in 1992 [Bugge et al., 1994].

DISCUSSION

The cause of omphalocele and the role of heredity are unknown. Omphalocele occur as isolated malformation in about 50% of cases and as part of syndromes or associated with other malformations in the remaining 50% (syndromic) (Danish registry of abdominal wall defects, unpublished data). A number of the syndromic cases are due to defects of blastogenesis [Opitz et al., 2002], but the cause of these defects seems to be unknown and both environmental and/or genetics factors are suggested to be involved. Omphalocele is part of several aneuploid and polyploidy syndromes as trisomies 13, 18, and 21 as well as triploidy. Omphalocele is also part of Beckwith–Widemann syndrome which is caused by paternal uniparental disomy of chromosome region 11p15. Omphalocele is not a part of the phenotype in all cases of these genetic syndromes and to our knowledge it is still unknown what causes omphalocele in the phenotype in some of these infants and not in others [Labbe et al., 1981; Hauge et al., 1983; Loevy et al., 1985; Beattie et al., 1993].

Study of the occurrence of the malformation in twins is a method to evaluate the role of genetic factors. To our knowledge no study of a representative sample of twins with omphalocele has been reported.

Omphalocele has been subject to many epidemiological studies during the last 2–3 decades, but few described the occurrence of twins. Only one study [Hwang and Koussef, 2004] compared the results with background data. The authors described 127 infants with omphalocele retrieved from a genetic center covering an area in Southern Florida and found nine pairs of twins, eight dizygotics and one monozygotic, all discordant. The nine pairs represented 7% of the total number of children with omphalocele, which is significantly higher than the occurrence of twins in the general population. To our knowledge, this is the highest occurrence of twins with omphalocele described.

One study showed that omphalocele/gastroschisis were more frequent in twins than singletons, but the authors did not separate the two malformations [Layde et al., 1980]. From the birth registries in two areas in Australia in 1980–1990 five pairs of twins with omphalocele including one pair of conjoined twins were identified among 104 cases [Byron-Scott et al., 1998]; and 3.8% of 448 omphalocele cases were twins reported from England and Wales in the period 1987–1993 [Tan et al., 1996], but the reports did not contain further information. Rankin et al. 1999 reported on 6 pairs of discordant twins in 296 cases of abdominal wall defects from a population-based register in northern England from 1986 to 1996, but it was not indicated whether they had omphalocele or another abdominal wall defect. Salihu et al. 2002 mentioned briefly that one discordant pair with omphalocele occurred out of 29 cases seen in 8 years in an obstetric unit in Germany, but there was no information on sex or zygosity of the twins. Brantberg et al. 2005 found six pairs of discordant twin pairs with omphalocele in a Norwegian obstetric unit from 1985 to 2004. Neither sex nor zygosity were indicated. None of the above-mentioned reports indicated any comparison between the frequency of twins with omphalocele and the occurrence of twins in the population.

Omphalocele in twins has been described in a number of case reports (Table II). Seven pairs of discordant verified monozygotic twins [Adeyokunnu, 1981; Lowry and Baird, 1982; Loevy et al., 1985; Beattie et al., 1993; McCoy et al., 1994; Opitz et al., 2002; Bhat et al., 2006] and five pairs, four discordant, and one concordant, of dizygotic twins [Adeyokunnu, 1981; Lindfors, 1883; Steele and Nevin, 1985; Noack et al., 2005]. In nine pairs, eight of them of same sex, zygosity was not proven, six discordant and three concordant [Herbert, 1928; Yuzpe and Johnson, 1968; Petersen et al., 1972; Labbe et al., 1981; Lowry and Baird, 1982; Warburton, 1991; Heydanus et al., 1993] (Table II). In these case reports atypical cases with rare associated malformations are likely to be overrepresented.

The Danish research registry of abdominal wall defects over 20 years is based on a nationwide epidemiological study with high ascertainment. The occurrence of omphalocele in twins in the register did not differ from the occurrence of twins in the Danish population in the same period. Discordance for omphalocele was found in all twin pairs except the conjoined twins. In only one pair (case 1978a), both twins were alive and available for zygosity test. They were monozygotic. Discordance for a malformation in MZ twin pairs indicates that the malformation cannot be determined by simple monogenic inheritance. In a number of cases it has been demonstrated that MZ twins are not always genotypically identical and the difference can explain the phenotypical difference. Postzygotic somatic mutation, genomic imprinting and skewed X-inactivation in only one of the twins has been one explanation of the discordance [Bugge et al., 1994; Machin, 1996, 2009]. To our knowledge no investigation of imprinting as a cause of omphalocele has been reported except for omphalocele occurring in some cases of Beckwith–Wiedemann syndrome. None of the twins had this syndrome. In our discordant MZ twin pair we have not investigated X-inactivation. X-linked inheritance has been suggested in one report [Havalad et al., 1979]: in one family four males with omphalocele as isolated malformation were described. Our data from the registry showed that 115 infants had omphalocele as isolated malformation 63 males and 52 females. The M/F ratio is thus 1.21 while the M/F ratio in the Danish population was 1.06 in the period. The M/F ratio for infants with omphalocele as isolated does not seem to be significantly different from the ratio in the Danish population, and we do not suspect omphalocele as isolated malformation to be of X-linked inheritance.

Comparison of concordance rates in MZ and DZ twins with omphalocele is a method to evaluate the role of genetics and environmental factors in the cause of the malformation, but the method demands large series of unselected cases for reliable results. The method was used in a Danish study of twins with nonsyndromic (without associated malformations) cleft lip with or without cleft palate [CL(P)]. The authors concluded, that the frequency of twins with [CL(P)] was not different from the occurrence of twins in the population, and 47 twins (39 twin pairs) were included. The study was performed in the same period as our study [Christensen and Fogh-Anderson, 1993]. The difference in prevalence between the two malformations is significant as [CL(P)] occur in 1.5–2 per 1,000 newborn [Christensen and Fogh-Anderson, 1993] and omphalocele in 2 per 10,000 [Bugge and Holm, 2002]. Therefore although we found that the occurrence of omphalocele in twins is the same as the occurrence of twins in the population, we found only seven pairs of twins due to the low prevalence of omphalocele. Furthermore, the cause of omphalocele as isolated malformation and omphalocele associated with other malformation or part of a syndrome might not be the same, and the two groups should be analyzed separately.

Our data allow the conclusion, that the occurrence of twins with omphalocele does not differ from the occurrence of twins in the Danish population, but does not allow any conclusion as to the role of genetic factors in the cause of omphalocele.