PAEDIATRIC RESPIRATORY REVIEWS (2004) 5(Suppl A), S277–S282
Congenital diaphragmatic hernia;
many questions, few answers
M. Hösgor and D. Tibboel
Department of Pediatric Surgery, Sophia Children’s Hospital, Erasmus Medical Center,
Rotterdam, The Netherlands
INTRODUCTION 22q11). In many instances however clinicians remain
empty handed when trying to explain to the parents
Congenital diaphragmatic hernia remains one of the
the cause of their offspring’s anomaly.
major challenges in the neonatal and paediatric
They then tend to fall back on the term
surgical ICU for various reasons: the wide variability
“multifactorial”, which is synonymous with “I don’t
in clinical presentation, the severe respiratory
know”, because it is very hard to identify specific
insufficiency, and the high pulmonary hypertension,
environmental factors including dietary components,
either continuously or intermittently.1,2
medication or environmental pollutants as the
It is the combination of pulmonary hypoplasia
causes of major congenital anomalies.8 CDH is
together with the unpredictable response to a variety
not an exception to this rule. Considering the wide
of treatment modalities applied to decrease the pul-
variability in phenotypic appearance of children
monary vascular resistance that mainly determines
with CDH, the real incidence including cases with
the outcome in the individual newborn.3 Overall
intra-uterine fetal death as well as terminations
mortality is reported to range from 20% to 60%,
of pregnancies following prenatal chromosomal
but these rates are hardly comparable because of
evaluation (Trisomy 13,18,21), and syndromic cases
differences in case selection.4−6 Likewise, it is hard
(Fryn’s syndrome) without a clear underlying genetic
to compare treatment results, especially because
background, CDH should be considered a disease
there are no standards for documenting patients
rather than a single congenital anomaly. In this
with congenital diaphragmatic hernia. Although the
way it is hard to identify “the patient” with CDH.
condition has derived its name from the primary
Both the genetic background and the effect of
anomaly, the hole in the diaphragm, in the last two
environmental factors may vary considerably in
decades clinicians have come to realise that the
the individual patient, leading to abnormal early
diaphragmatic defect does not determine survival.
lung development and closure of the diaphragm.
Animal studies and evaluation of serial sections
No candidate genes have been identified so far,
of early human embryos have convincingly shown
although chromosome-15 anomalies and a variety
that disturbed pulmonary development during the
of other chromosomal abnormalities have been
embryonic phase should be considered the primary
suggested in the literature. CDH is well known
defect.7
to occur as part of major numerical chromosomal
anomalies such as the trisomies 18, 13 and
ETIOLOGICAL ASPECTS 21.4 Routine chromosomal analysis with sensitive
methods to identify even point mutations or small
It is now generally accepted that major structural
deletions, together with availability of DNA data
abnormalities result from specific genetic abnormal-
bases of patients and parents, might provide
ities in selected cases such as Smith–Lemli–Opitz
important information to elucidate the etiology in
syndrome and DiGeorge syndrome (chromosome
individual cases.
* Correspondence to: D. Tibboel. Tel.: +31-(10)-4636567;
Fax: +31-(10)-4636288; E-mail: DEVELOPMENTAL ASPECTS
Correspondence address: Dr. Molewaterplein 60, 3015 GJ
Rotterdam, Netherlands Paediatric surgeons and obstetricians such as
1526-0542/$ – see front matter © 2004 Elsevier Science Ltd. All rights reserved.
, S278 M. HÖSGOR AND D. TIBBOEL
Michael Harrison, Scott Adzick and Jay Wilson, PRENATAL MODULATION OF LUNG
and Jan Deprest and Bruno Piedboeuf performed GROWTH
a wonderful series of experiments especially in the
Prenatal modulation of lung growth has been
sheep and rabbit, to induce congenital diaphragmatic
investigated, through both surgical correction and
hernia at different developmental stages. Much of
hormone treatment, in animal models as well as in
our knowledge is based on these animal models selected cases of human CDH.
in which the primary development of the lungs The disappointing effects of postnatal treatment
and diaphragm is normal, and both insult to the together with the unpredictable responses to
developing lung and the hole of the diaphragm are new treatment modalities such as high-frequency
created surgically.9−13 oscillation (HFO), extra-corporeal membrane oxy-
These models have provided valuable data with genation (ECMO) and finally liquid ventilation, have
regard to type-II cell differentiation, pulmonary vas- stimulated studies aimed at correcting the defect
cular developmental abnormalities, and ventilatory intra-uterine and subsequent catch up growth, or at
strategies in newborn lambs with CDH, including the modulating differentiation of the gas-exchange area
effects of surfactant therapy and liquid ventilation. by hormones like corticosteroids. The San Francisco
Other investigators have used this approach in group headed by Michael Harrison as well as
rodent animal models (rats and mice), which the research groups in Boston (Jay Wilson) and
provided information about the closure process of Philadelphia (Scott Adzick) have shown that intra-
the diaphragm and the early stages of (ab)normal uterine surgery, either open or closed, is a
lung development. In these models the defect is relatively safe event even in the human fetus.20
induced at the stage when the foregut has just An NIH-sponsored trial failed to show the benefit
separated into the esophagus and the trachea, of intra-uterine surgical correction in unselected
which in rodents is usually at 9 to 10 days of patients following prenatal diagnoses compared with
gestation. In rats the diaphragm does not close postnatal treatment. This is why identification of
until day 17 (term day 21).14−18 These models even the individual patient suitable for prenatal lung-
permit one to isolate the lung buds and to culture growth modulating procedures, such as temporary
them with or without the heart, so that branching obstruction of the trachea, has become one of
morphogenesis in detail and the effect of mutational the major goals of the recently finalised NIH-
deletions can be investigated.19 sponsored trial. Only patients with a predicted high
A different model is the Nitrofen-induced rodent mortality as revealed by a low lung-to-head ratio
model. This, however, has the disadvantage that in case of so-called “liver-up patients” are eligible
the toxic substance applied does not only affect for inclusion in studies using a so-called “closed
the developing diaphragm and lungs but also has procedure” with a detachable balloon to obstruct
a teratogenic effect on other organ systems, such the trachea. Experimental studies evaluating the
effect of tracheal obstruction of the fetal lung have
as the central nervous system, skeleton and heart.
provided data evidencing that lung growth really
What in the near future will turn out to be the ideal
takes place, i.e. increased DNA\protein levels and
model for unravelling the etiology and pathogenesis
cell counts. Even the vascular abnormalities are
of CDH is still unknown. The classical hypothesis
reported to diminish following tracheal obstruction.
in the rodent model, declaring the diaphragm the
The effects on the type-II cell differentiation remain
primary defect and the pulmonary hypoplasia a
to be documented in more detail, especially when
secondary phenomenon caused by compression combined therapies such as tracheal obstruction
of the developing lung by the bowel loop, has with prenatal corticosteroids are considered. Intra-
become more and more disputed. Recent studies uterine fetal surgical manipulation of pulmonary
have shown that the pulmonary hypoplasia occurs growth performed in selected centres should yield
independently of the presence of the defect in the an adequate number of patients for randomised
diaphragm and is already present for several days controlled trials. Otherwise we will only have
before the normal closure process of the diaphragm anecdotal reports and thus remain in uncertainty
takes place during ontogeny.7 about patient selection.
In the rodent model competition for space between In view of the well-documented positive effects
the developing lungs and the liver has been proven of prenatal corticosteroids in threatened premature
by longitudinal underwater dissection. Although all labour and those of additional corticosteroids follow-
these studies provide some information about the ing prenatal induction of CDH in the Nitrofen rodent
pathogenesis of CDH, it is still unclear what the model, corticosteroids have been used sporadically,
etiology is in the majority of cases at least in with varying results. The use of corticosteroids
humans. in prenatally diagnosed CDH in humans is partly
Congenital diaphragmatic hernia;
many questions, few answers
M. Hösgor and D. Tibboel
Department of Pediatric Surgery, Sophia Children’s Hospital, Erasmus Medical Center,
Rotterdam, The Netherlands
INTRODUCTION 22q11). In many instances however clinicians remain
empty handed when trying to explain to the parents
Congenital diaphragmatic hernia remains one of the
the cause of their offspring’s anomaly.
major challenges in the neonatal and paediatric
They then tend to fall back on the term
surgical ICU for various reasons: the wide variability
“multifactorial”, which is synonymous with “I don’t
in clinical presentation, the severe respiratory
know”, because it is very hard to identify specific
insufficiency, and the high pulmonary hypertension,
environmental factors including dietary components,
either continuously or intermittently.1,2
medication or environmental pollutants as the
It is the combination of pulmonary hypoplasia
causes of major congenital anomalies.8 CDH is
together with the unpredictable response to a variety
not an exception to this rule. Considering the wide
of treatment modalities applied to decrease the pul-
variability in phenotypic appearance of children
monary vascular resistance that mainly determines
with CDH, the real incidence including cases with
the outcome in the individual newborn.3 Overall
intra-uterine fetal death as well as terminations
mortality is reported to range from 20% to 60%,
of pregnancies following prenatal chromosomal
but these rates are hardly comparable because of
evaluation (Trisomy 13,18,21), and syndromic cases
differences in case selection.4−6 Likewise, it is hard
(Fryn’s syndrome) without a clear underlying genetic
to compare treatment results, especially because
background, CDH should be considered a disease
there are no standards for documenting patients
rather than a single congenital anomaly. In this
with congenital diaphragmatic hernia. Although the
way it is hard to identify “the patient” with CDH.
condition has derived its name from the primary
Both the genetic background and the effect of
anomaly, the hole in the diaphragm, in the last two
environmental factors may vary considerably in
decades clinicians have come to realise that the
the individual patient, leading to abnormal early
diaphragmatic defect does not determine survival.
lung development and closure of the diaphragm.
Animal studies and evaluation of serial sections
No candidate genes have been identified so far,
of early human embryos have convincingly shown
although chromosome-15 anomalies and a variety
that disturbed pulmonary development during the
of other chromosomal abnormalities have been
embryonic phase should be considered the primary
suggested in the literature. CDH is well known
defect.7
to occur as part of major numerical chromosomal
anomalies such as the trisomies 18, 13 and
ETIOLOGICAL ASPECTS 21.4 Routine chromosomal analysis with sensitive
methods to identify even point mutations or small
It is now generally accepted that major structural
deletions, together with availability of DNA data
abnormalities result from specific genetic abnormal-
bases of patients and parents, might provide
ities in selected cases such as Smith–Lemli–Opitz
important information to elucidate the etiology in
syndrome and DiGeorge syndrome (chromosome
individual cases.
* Correspondence to: D. Tibboel. Tel.: +31-(10)-4636567;
Fax: +31-(10)-4636288; E-mail: DEVELOPMENTAL ASPECTS
Correspondence address: Dr. Molewaterplein 60, 3015 GJ
Rotterdam, Netherlands Paediatric surgeons and obstetricians such as
1526-0542/$ – see front matter © 2004 Elsevier Science Ltd. All rights reserved.
, S278 M. HÖSGOR AND D. TIBBOEL
Michael Harrison, Scott Adzick and Jay Wilson, PRENATAL MODULATION OF LUNG
and Jan Deprest and Bruno Piedboeuf performed GROWTH
a wonderful series of experiments especially in the
Prenatal modulation of lung growth has been
sheep and rabbit, to induce congenital diaphragmatic
investigated, through both surgical correction and
hernia at different developmental stages. Much of
hormone treatment, in animal models as well as in
our knowledge is based on these animal models selected cases of human CDH.
in which the primary development of the lungs The disappointing effects of postnatal treatment
and diaphragm is normal, and both insult to the together with the unpredictable responses to
developing lung and the hole of the diaphragm are new treatment modalities such as high-frequency
created surgically.9−13 oscillation (HFO), extra-corporeal membrane oxy-
These models have provided valuable data with genation (ECMO) and finally liquid ventilation, have
regard to type-II cell differentiation, pulmonary vas- stimulated studies aimed at correcting the defect
cular developmental abnormalities, and ventilatory intra-uterine and subsequent catch up growth, or at
strategies in newborn lambs with CDH, including the modulating differentiation of the gas-exchange area
effects of surfactant therapy and liquid ventilation. by hormones like corticosteroids. The San Francisco
Other investigators have used this approach in group headed by Michael Harrison as well as
rodent animal models (rats and mice), which the research groups in Boston (Jay Wilson) and
provided information about the closure process of Philadelphia (Scott Adzick) have shown that intra-
the diaphragm and the early stages of (ab)normal uterine surgery, either open or closed, is a
lung development. In these models the defect is relatively safe event even in the human fetus.20
induced at the stage when the foregut has just An NIH-sponsored trial failed to show the benefit
separated into the esophagus and the trachea, of intra-uterine surgical correction in unselected
which in rodents is usually at 9 to 10 days of patients following prenatal diagnoses compared with
gestation. In rats the diaphragm does not close postnatal treatment. This is why identification of
until day 17 (term day 21).14−18 These models even the individual patient suitable for prenatal lung-
permit one to isolate the lung buds and to culture growth modulating procedures, such as temporary
them with or without the heart, so that branching obstruction of the trachea, has become one of
morphogenesis in detail and the effect of mutational the major goals of the recently finalised NIH-
deletions can be investigated.19 sponsored trial. Only patients with a predicted high
A different model is the Nitrofen-induced rodent mortality as revealed by a low lung-to-head ratio
model. This, however, has the disadvantage that in case of so-called “liver-up patients” are eligible
the toxic substance applied does not only affect for inclusion in studies using a so-called “closed
the developing diaphragm and lungs but also has procedure” with a detachable balloon to obstruct
a teratogenic effect on other organ systems, such the trachea. Experimental studies evaluating the
effect of tracheal obstruction of the fetal lung have
as the central nervous system, skeleton and heart.
provided data evidencing that lung growth really
What in the near future will turn out to be the ideal
takes place, i.e. increased DNA\protein levels and
model for unravelling the etiology and pathogenesis
cell counts. Even the vascular abnormalities are
of CDH is still unknown. The classical hypothesis
reported to diminish following tracheal obstruction.
in the rodent model, declaring the diaphragm the
The effects on the type-II cell differentiation remain
primary defect and the pulmonary hypoplasia a
to be documented in more detail, especially when
secondary phenomenon caused by compression combined therapies such as tracheal obstruction
of the developing lung by the bowel loop, has with prenatal corticosteroids are considered. Intra-
become more and more disputed. Recent studies uterine fetal surgical manipulation of pulmonary
have shown that the pulmonary hypoplasia occurs growth performed in selected centres should yield
independently of the presence of the defect in the an adequate number of patients for randomised
diaphragm and is already present for several days controlled trials. Otherwise we will only have
before the normal closure process of the diaphragm anecdotal reports and thus remain in uncertainty
takes place during ontogeny.7 about patient selection.
In the rodent model competition for space between In view of the well-documented positive effects
the developing lungs and the liver has been proven of prenatal corticosteroids in threatened premature
by longitudinal underwater dissection. Although all labour and those of additional corticosteroids follow-
these studies provide some information about the ing prenatal induction of CDH in the Nitrofen rodent
pathogenesis of CDH, it is still unclear what the model, corticosteroids have been used sporadically,
etiology is in the majority of cases at least in with varying results. The use of corticosteroids
humans. in prenatally diagnosed CDH in humans is partly
