1Congenital Diaphragmatic Hernia: Information
This chapter is an overview of fetal congenital diaphragmatic hernia including prenatal ultrasound, measurement, outcomes, and treatment.
A congenital diaphragmatic hernia (CDH) results from defects in the diaphragm through which intra-abdominal organs become present within the fetal chest. A variety of hernias have been described. A Bochdalek hernia consists of a posterior lateral defect in the fetal diaphragm .
The Morgagni hernia (5% or greater of CDH) is an anterior defect in the diaphragm, rarely centrally or bilaterally located. It ranges from a specific defect, to diaphragmatic thinning, to a membranous sac. The size of the defect may range from small to complete agenesis. Size may correlate with clinical severity.
Because of the mass effect from the intra-abdominal contents on fetal lung capacity, neonatal morbidity includes pulmonary hypoplasia as well as persistent pulmonary hypertension.
Hypoplastic lungs in CDH demonstrate a decreased number of airways and fewer alveoli leading to impaired gas exchange. The pulmonary hypertension in CDH relates to a decreased number of vascular branches with a resultant decrease in the cross-sectional area of the pulmonary vascular bed. In addition, there is increased muscularization of the preacinar pulmonary arteries, which is reactive to multiple stimuli and can give rise to severe pulmonary hypertension, even when lung size appears to be an adequate lung size for survival.2
Prevalence of Congenital Diaphragmatic Hernia
CDH occurs in approximately 1 in 2,000 to 3,000 births. In a large epidemiological study, the overall prevalence of CDH was 2.49 per 10,000, and 1.45 per 10,000 for isolated cases (CDH alone).3 In non-isolated CDH (CDH plus malformations), lower risks are noted among blacks, multiparous women, and women not of advanced maternal age. CDH is more common than reported in neonatal surgical practices.4 Overall mortality is underreported due to fetal loss, under-recognition, and death prior to admission to reporting centers.5 Overall mortality in CDH registry of patients who live long enough to be admitted to a participating NICU is 30% for isolated CDH and 33% overall.
Site of Occurrence
Distribution by location of CDH is 75% left-sided, 17% right-sided, and 8% bilateral.4 Distribution of cases in the European registry is somewhat higher for left-sided lesions.
Outcome of CDH is determined by associated malformations, abnormal karyotype/microarray, and pulmonary and/or cardiovascular complications.6 In CDH, at least one malformation is noted in 40% of cases and major malformations in 17% of cases.7 However, the number of associated malformations reported in registries may be lower than this. Associated malformations are heterogeneous, while cardiovascular anomalies are the most common defects reported.
Other studies support the presence of malformations associated with CDH, including cardiovascular, chest, craniofacial, central nervous system, and karyotype abnormalities.8
Associated anomalies are prime determinants of outcome, but do not facilitate CDH detection,9 and lung malformations in the presence of CDH complicate accurate diagnosis of involved lesions.10
Fetuses with CDH have a disproportionately smaller left heart size,11 and the ipsilateral pulmonary artery and branches of the pulmonary artery are normal or smaller than those of normal fetuses.12
Aneuploidy and Deletions
The pathogenesis of CDH is not completely understood. An abnormal pleuroperitoneal fold in the diaphragm at 8 to 10 weeks of gestation underlies the development of certain types of CDH.13 While the majority of cases are idiopathic, approximately 4% are related to chromosomal abnormalities.14 In CDH, 15q2 6.1 deletion is reported. The 15q26 deletion phenotype is associated with left-sided diaphragmatic hernia, cardiac anomalies, and characteristic facial features.15 De novo deletions are reported as 8p23.1 or 15q2 6.1 in fetuses with congenital heart defects.16
Single Gene Disorders
Wolf-Hirschhorn syndrome (WHS), a disorder caused by the deletion of the short arm of chromosome 4, presents as intrauterine growth restriction, severe oligohydramnios, left-sided congenital diaphragmatic hernia, and cystic hygroma.17 In addition, more than 70 syndromes are associated with CDH, some with known gene abnormalities such as Fryns syndrome, Simpson-Golabi-Behmel syndrome, Denys-Drash syndrome, spondylocostal dysostosis, craniofrontonasal syndrome, Cornelia de Lange syndrome, and Marfan syndrome.
Retinoid Signaling Pathway
The disruption of the retinoid signaling pathway may be related to CDH pathogenesis and metabolic and molecular factors of this pathway are under study,18 while candidate genes have been identified.19 Congenital diaphragmatic hernia correlates with low retinol and retinol-binding proteins in the newborn regardless of maternal status of these substances.20 In animals, prenatal treatment with retinoic acid accelerates alveolar development.18
Other Models of Congenital Diaphragmatic Hernia
CDH can be induced in animals with nitrofen (diphenyl ester, a former herbicide) and vitamin A, and anti-inflammatory agents may also play a role in its development.21 Genomic imbalances in CDH may be identified by targeted array comparative genomic hybridization (array CGH).22,23 In utero gene therapy in animal experimentation improves pulmonary hypoplasia associated with congenital diaphragmatic hernia.24
Fetal Surgery for Congenital Diaphragmatic Hernia
Open fetal surgery (hysterotomy) to repair CDH defects and/or perform tracheal occlusion does not have a survival benefit.25 Percutaneous fetal endoscopic tracheal occlusion (FETO) is feasible and safe according to tertiary care units. A subset of fetuses with CDH are likely to die without intervention due to the severity of predicted lung hypoplasia. The most frequent procedural complication is premature rupture of the membranes (PROM). CDH patients undergoing FETO seem more likely to survive than those who do not, findings which will be tested by formal clinical trials.26
In 2004, preliminary results were published, and the initial inclusion criteria were a lung-to-head ratio of < 1.0 and the presence of fetal liver within the chest.27 Successful placement of the endotracheal balloon was possible in all initial cases. PROM occurred in 52.4% of patients and survival was 30% in the first group of 10 fetuses and 63.6% in the second group of 11 fetuses. In a follow-up study, membrane rupture occurred in 35% of patients within 32 weeks and a decrease in the occurrence of PROM was noted over time. Survival to discharge was 50% and long term survival quality was encouraging.28 Retrospective analysis found the position of the liver and LHR correlates to survival. The combination of liver up in the chest and an LHR < 1.0 resulted in a survival rate of only 9%, and when the LHR was less than 0.6, there were no survivors irrespective of fetal liver position. Survival rates increased for right-sided diaphragmatic hernias from 0% to 35.3%.29 Overall, survival for patients undergoing FETO has improved from 15% to approximately 50%. [Deprest J, Devlieger R, Rayyan M, Vanhole C, El handouni N, Claus F, Dymarkowski S, et al. Treatment of congenital diaphragmatic hernia. Fetal MRI. Medical Radiology. 2011;343-360].30
Complimentary support for the effects of FETO is suggested by intrapulmonary arterial Doppler, which predicts increased survival among treated patients,31 and MRI demonstration of increased lung volume in FETO patients compared to controls.32
Changes in the trachea such as tracheal widening are potential complications, which will require long term follow-up.3334
Controversies continue to cloud prenatal surgery, and questions remain as to whether FETO should be offered in any or all cases.35 Until the problems of preterm delivery are resolved, it will be difficult to match the survival rates in some large ECMO centers, recently reported as 85% overall. Some argue that prospective randomized trials are needed prior to the widespread introduction of fetal surgical interventions.36 The use of systemic medications to expand lung volumes and improvements in neonatal respiratory care as well as the excellent outcomes for neonatal surgical management of CHD are reasons to be cautious about widespread prenatal surgical intervention. [Bohn D. Contemporary neonatal intensive care management in congenital diaphragmatic hernia: does this obviate the need for fetal therapy? Fetal and Maternal Medicine Review. 2009;20(4):319-339].
Management Options for Congenital Diaphragmatic Hernia
The morbidity and mortality in CDH is due to fetal lung hypoplasia and pulmonary hypertension. Since the volume and type of intra-abdominal viscera, especially in the fetal liver, is variable, so is their impact on lung growth and pulmonary hypertension, making management difficult. Management options include fetal surgical intervention with FETO at centers competent in it, delivery at tertiary care centers, elective neonatal surgery after stabilization, and the addition of neonatal strategies such as permissive hypercapnia (“gentle” ventilation).37 Tertiary care centers should provide genetic counseling, genetic screening, multiple disciplines, and individual prognoses based upon the presence of associated malformations, lung size assessment, and the presence of liver herniation.38 Neonatal surgical repair of the diaphragmatic defect is no longer emergent and stabilization may be needed with extracorporeal membrane oxygenation (ECMO).39 ECMO may be required in as many as 30% to 40% of patients, often within the first 12 hours of life, and because of these factors, most experts recommend delivery in an ECMO center. Survival has increased to 80% in some tertiary care centers, and mortality is declining in some countries from 67% to 23%.40
Despite these achievements, standardized protocols are urged to define research needs and improve care.41 The availability of outcome data can guide protocol development. From registry data for CDH, the overall survival in 907 patients is 67%. By 2013, among 6,000 patients, the overall survival was 69%. ECMO-treated infants survived better with a gestational age of 38 weeks (+/-2), a birth weight of 3200 grams (+/- 500), and when on ECMO for a shorter period of time.42 Gestational ages of 39 to 40 weeks are likely to have a still higher survival rate.
Route of Delivery
No randomized trials address route of delivery for CDH patients. Two studies suggest no overall survival difference for CDH patients between Cesarean delivery and vaginal delivery.43,44
Neither the results of a randomized trial with 34 patients nor the CDH registry demonstrate a survival benefit for the late prenatal administration of steroids.45 Since the reported series of patients are small, further data may be needed to resolve this.
A discussion of outcomes necessitates a focus upon the anatomic and physiologic changes which define this condition and its outcome. These include pulmonary hypoplasia and cardiovascular changes, and most importantly, the pulmonary vascular changes leading to pulmonary hypertension. Outcome can be viewed from the perspectives o associated anomalies, position of the liver within the fetal chest, site of the defect, and the degree of lung hypoplasia as measured by 2-D ultrasound, 3-D ultrasound, and fetal MRI. Outcome can also depend on antenatal detection, case selection, timing of delivery, delivery in a tertiary center, and the presence of treatment protocols. Predictions of outcome can be based upon lung volume and/or vascular indices, particularly of the pulmonary arteries. Finally, long term outcome is important to discuss.
However, none of the prenatal measurements alone are absolute predictors of outcome, with the exception of extremely small residual lung volume. Pulmonary hypertension is difficult to predict and can be life-threatening, even in neonates with adequate lung volume. Lung size and volume measurements are subject to error and they are not always reproducible. Combinations of predictors may be useful such as the presence of all or some of the liver in the chest.
Overall Outcome in CDH
Ascertainment of true outcome in CDH is difficult due to spontaneous loss and termination of pregnancies as well as variations in case selection and antenatal and neonatal care.46 In a population-based study of 116 cases of CDH, 61% were live born and 52% of those survived beyond 1 year of age, while 32% of CDH infants overall survived beyond one year of age. 47 Termination of pregnancies occurred in 33% and 3% suffered spontaneous abortion while 3% were stillborn. In another study of 88 cases of CDH, 83% survived to discharge. Survival was greater at high volume care centers compared to low volume care centers.48
Another major anomaly occurring in the presence of CDH is reported in 41% of 116 cases. A significant number (18%) of additional anomalies are considered lethal, while any anomaly in the presence of CDH raises the death rate to 78%. Depending upon the population, the reported concomitant anomaly rate may be less than this. Cardiac anomalies are most commonly associated with CDH, followed by lung and craniofacial anomalies. Bronchopulmonary sequestration is reported in association with CHD in 30% to 40% of cases.49
In large autopsy series, approximately 16% of fetal lung hypoplasia is associated with diaphragmatic defects.50 Certain defects carry a poor prognosis such as the association of spinal defects with CDH.51 Finally, the presence of fetal plural effusions does not constitute an adverse prognostic factor for CDH,52 while hydrops fetalis is a poor prognostic factor irrespective of origin.
Liver Position and Outcome
The presence of the fetal liver within the chest is an important prognostic factor for CDH. Survival rates are reported as 45.4% when a portion of the liver is herniated into the chest cavity, and 73.9% when the fetal liver remains within the abdomen.53
In addition, the presence of the fetal liver within the chest is predictive of the need for prosthetic repair in CDH.54 Liver position followed by the position of the fetal stomach are the strongest predictors for an intact neonatal discharge from the hospital.55
In addition, left heart hypoplasia is more likely when the liver is herniated into the fetal chest which causes a distortion in the normal anatomic position and flow from the ductus venosus.56
Right-sided Diaphragmatic Hernia and Outcome
Right-sided diaphragmatic hernia is less common than left-sided CDH and the diagnosis is more difficult compared to left-sided CDH.57
During 18 years, 29 patients were identified with right-sided diaphragmatic hernia and only 27.5% were identified prenatally.58 In that study, associated malformations were reported in 27.5% of the cases. Right-sided diaphragmatic hernias account for 11 to 15% of CDH.59,60 In recent series, prenatal diagnosis is reported in 50% of right-sided CDH, but 75% of left-sided CDH. Higher morbidity and mortality is consistently observed in the patients with right-sided CDH. Increased diagnostic accuracy can be expected when MRI is used as an imaging modality.
The pulmonary pathology in CDH is related to the pulmonary compression from the herniated intra-abdominal viscera. The compression impairs growth of the alveoli and the pulmonary vessels. These changes worsen as the pregnancy progresses, particularly beyond 30 weeks.61
A number of ultrasound and MRI measurements assess the degree of lung hypoplasia and fetal outcome. The LHR is the ratio of the lung area on the contralateral side of the CDH to the head circumference. A value of less than 1.0 is a predictor for mortality, while that of greater than 1.4 is associated with survival.62 The observed-to-expected fetal lung area to head circumference ratio (O/E LHR)and liver position predicts the need for prosthetic patch repair and neonatal morbidity.63 LHR is also a predictive factor for the use of extracorporeal membrane oxygenation (ECMO) and survival in CDH.64 The median lung area to head circumference ratio prior to the performance of fetal endoscopic luminal tracheal occlusion is predictive of postnatal survival.65 Timing of lung size assessment may be important in the prediction of survival for fetuses with CDH. In isolated cases of CDH, accurate measurements of the O/E LHR measurements may be taken at either 22 to 23 weeks or 32 to 33 weeks.66 Prediction models may be less reliable when measurements are taken at 20 to 24 weeks compared to 23 to 34 weeks gestation.67 Other measurements to predict outcome include lung to thorax transverse area ratio at or near term and sonographic fetal lung volume to body weight ratio.68,69
LHR may not always reliably predict fetal outcome. When a LHR of < 0 .6 was used as a cutoff for poor outcome and a LHR of greater than 1.4 was used as a cutoff for survival, no correlation was observed with liver herniation and LHR measurements and fetal outcome.70 In another report, neither a LHR of 1.0 or less nor herniation of the liver identified significant risk factors to warrant fetal intervention.71 Others recommend liver position as the best predictor of outcome in isolated left CDH cases and that LHR alone should not be used for counseling.72 MRI appears to be more accurate than ultrasound in the assessment of LHR.73
3-D ultrasound is a potential assessment methodology for FLV and initial studies suggest this method as a potential predictor for pulmonary hypoplasia and outcome.74 In isolated CDH, 3-D ultrasound assessment is possible for observed-to-expected (O/E) total, contralateral and ipsilateral lung volume ratios, and good correlations were observed between lung ratios and predictors of perinatal outcome including pulmonary hypertension.75
MRI Volume Measurements
A number of studies report the role of MRI in the assessment and prediction of adverse outcome in CDH. The O/E FLV by MRI predicts survival.76 In addition, it predicts pulmonary hypoplasia and neonatal outcome.77 The MRI measurement of FLV correlates with neonatal mortality due to pulmonary hypoplasia,78 and the O/E FLV as well as the LHR are predictors for survival.79,80 A number of studies correlate the need for extracorporeal membrane oxygenation (ECMO) with MRI-derived measures for FLV.81–83 MRI is accurate in detecting the fetal liver within the thoracic cavity and neonatal deaths are significantly increased when greater than 20% of the fetal thorax is occupied by the fetal liver.84 The degree of thoracic liver herniation can be measured by volumetric MRI and this measurement predicts postnatal survival independently from total FLV.85 Finally, after 30 weeks gestation, logistic regression analysis based upon MRI FLV measurements is useful to estimate neonatal survival and the need for ECMO.86
MRI Lung Signal Intensity
The fetal right lung-to-liver signal intensity ratio (LLS IR) may be valid in predicting postnatal outcome.87 However, others suggest FLV/estimated fetal body weight as derived by MRI may be a more useful predictor for fetuses with pulmonary hypoplasia.88 The lung signal intensity to spinal fluid signal intensity (L/SF) may also be calculated using MRI methodology and appears as a predictor for survival and pulmonary hypoplasia in CDH.89
Diffusion-weighted MRI methodology can determine the apparent diffusion coefficient (ADC) values for fetal lung over various gestational ages. There is a significant relationship between ADC and gestational age, making ADC a potential predictor independent of lung volume.90
Antenatal Detection and Outcome
Antenatal detection of CDH leads to transfer to tertiary care centers. Lack of prenatal detection occurs in 45% of cases noted by reporting centers.90 Most undetected cases (73%) are isolated CDH, but experience a better survival rate than those born in tertiary care centers. However, over time, prenatal detection rates are increasing for CDH and left-sided CDH while survival rates for isolated CDH are also improving.91 Many undetected CDH patients die before or during transfer to a reporting center and as such are never counted.
Case Selection and Time of Delivery
Case selection bias probably affects reported survival in CDH. Some reports suggest no change in mortality over time but an increase in the number of pregnancy terminations which correlates inversely with live-born survival.92 Night births do not adversely affect outcome.93 The timing of delivery may be independently related to outcome. Infants free of major malformations and born at 37 weeks with birth weights greater than the group mean (3100 grams) had the highest survival rate (80%).94 However, other reports suggest that neonatal and infant mortality was greater in infants with CDH born at 37 weeks than those born at 40 weeks.95
The implementation of standardized treatment protocols and altering neonatal respiratory support therapy may improve survival rates.96,97 Outcomes from multiple pregnancies affected by CDH appear no different from those from singleton pregnancies.98
Persistent pulmonary hypertension is a major morbidity in newborns with CDH.99 A number of vascular predictors for its severity are reported. MRI-derived diameters of the right pulmonary artery, left pulmonary artery, aorta, and length of the cerebellar vermis are used to calculate the prenatal pulmonary hypertension index (PPHI), which helps predict the severity of neonatal pulmonary hypertension.99 Other parameters include fetal pulmonary artery diameter measurements, which potentially correlate with neonatal respiratory morbidity. The pulmonary artery index may be a measure of mortality in CDH.100 Others support the utility of the pulmonary artery diameters to predict perinatal death, but not pulmonary hypertension since pulmonary artery diameters are likely related to the severity of pulmonary hypoplasia.101 The fetal preload index of the inferior vena cava may also be a predictor for outcome in CHD patients. [Miura H, Ogawa M, Sato A, Fukuda J,Tanaka T. Fetal preload index of the inferior vena cava and neonatal outcome of congenital diaphragmatic hernia. Journal of Medical Ultrasonics. 2009;36(2):77-81]
In left-sided CDH, the left ventricle is compressed and may be smaller than normal. Among fetuses with CDH, hypoplasia and reduced flow in the left heart are common, and this process reverses after birth and repair.102
In summary, there are multiple predictors of neonatal outcome in CDH. The important factors include side of defect with the right side having a poorer prognosis, intrathoracic liver, and MRI derived ratio of fetal lung area to head circumference and FLV.103 Among these, no single predictor appears superior to others.
Adverse neurodevelopment is common among infants with CDH. In one study, approximately half of the infants with CDH show neurodevelopmental abnormalities, including neuromuscular hypotonia and psychomotor dysfunction.104 Neurological impairment is thought to be due to hypoxemia, occurring during the first few days of life, particularly among infants on ECMO.105 Other complications include long term pulmonary dysfunction, cardiovascular complications including pulmonary hypertension, poor feeding and GER, growth restriction, and scoliosis.
Long-term respiratory morbidity is common, which includes growth impairment and impedance of exercise.106 Among 45 patients with CDH, the following was noted: 87% left sided, 22% with cardiac anomalies, 18% congenital anomalies or genetic syndromes, 98% with need for ventilator (median 17 days), 24% ECMO, 84% medical issues at discharge, 68% medical issues at age 1, 77% medical issues at age 3, 32% pulmonary problems at age 1, 47% pulmonary problems at age 3, 46% motor problems at age 1, 71% motor problems at age 3. These findings are from a relatively small cohort and overall outcomes may be more optimistic at some centers.107
Given the overall survival in the 85% range, questions remain as to how best to improve outcomes. Data regarding outcomes with fetal endoscopic tracheal occlusion is still being gathered.108 Outcome may be improved by defining accurate prenatal severity predictors, defining optimal treatment strategies, using multidisciplinary care, and maximizing cooperation among treatment centers.109 Voluntary international registries may centralize data collection and evaluation.110