Pediatrics

Diaphragmatic hernia

OVERVIEW: What every practitioner needs to know

Are you sure your patient has congenital diaphragmatic hernia? What are the typical findings for this disease?

Diagnosis

Most cases of Congenital Diaphragmatic Hernia (CDH) are diagnosed prenatally (greater than 50%). Diagnosis is made on either routine prenatal ultrasound or as part of the work up for polyhydramnios. CDH can be left sided (80-85%) or right sided (10-15%) and rarely bilateral (5%).

Prenatal

Left sided CDH is characterized by the presence of a heterogeneous mass in the chest that often results in right mediastinal shift. The fluid filled stomach may be identified in the chest cavity next to or just behind the heart, and noted to be absent in the abdomen. Fluid in the small bowel also helps to distinguish bowel from lung or other thoracic neoplasm. The liver may be herniated as well and appear as a homogeneous mass in the chest at the level of the heart and continuous with the intraabdominal liver. The gallbladder and hepatic or umbilical veins may be abnormally located within the abdomen, which can be scaphoid.

Right sided CDH is characterized by the presence of a homogeneous mass (liver) in the right chest that often results in left mediastinal shift. Pleural fluid is often present, and bowel may herniate, as well. The left shift of the heart is a key finding since sonographically the liver is similar in appearance to fetal lung. Identifying the gallbladder in the chest is also diagnostic of right sided CDH.

Color Doppler ultrasound can be used to document the location of the liver by demonstrating the course of the intrahepatic vessels.

Polyhydramnios may be present due to esophageal compression. Hydrops fetalis can occur from mediastinal shift and compression of the great vessels.

Postnatal

When undiagnosed prenatally the infant will present with respiratory distress immediately after birth with retractions and tachypnea, tachycardia and cyanosis. The classic finding of CDH in the delivery room are asymmetry of the chest with an abdomen that appears concave due to absence of abdominal organs and are only occasionally seen.

What other disease/condition shares some of these symptoms?

CDH may be confused with other thoracic lesions such as congenital cystic adenomatoid malformation (CCAM), bronchopulmonary sequestration (BPS), bronchogenic cysts, bronchial atresia and teratomas. Infants with these congenital lung malformations may have similar ultrasound findings prenatally and a similar clinical presentation after birth.

The definitive prenatal sonographic diagnosis of fetal CDH relies on the visualization of abdominal organs in the fetal chest. The sonographic hallmark of a left CDH is a fluid-filled stomach just behind the left atrium and ventricle in the lower thorax with the absence of the stomach below the diaphragm, mediastinal shift to the right and a small abdominal circumference. Right CDH is more frequently missed or misdiagnosed because the herniated viscera consists predominantly of the right lobe of the liver, which may have similar echogenicity to the lung, or be confused with a solid mass in the chest such as a CCAM or BPS. Cystic lesions in the chest are more likely to be CCAM. The presence of a systemic feeding vessel to a cystic or solid mass is consistent with a CCAM or BPS.

Postnatally ultrasound of the chest can help differentiate CDH from other thoracic lesions. If ultrasound fails to differentiate between CDH and other lesions CT of the chest is indicated.

CDH should also be distinguished from diaphragmatic eventration (displacement of abdominal contents into the thorax as a result of a congenitally thin, hypoplastic but intact diaphragm). It is rare and usually has less severe effects on lung development and function than CDH.

What caused this disease to develop at this time?

The etiology of CDH has not been established definitively. The leading theories are that it is due to failure of normal closure of the pleuroperitoneal folds during the 4th to 10th weeks postfertilization resulting in a diaphragm defect. During this same time the abdominal viscera which are external to the fetus, return to the abdominal cavity allowing for closure of the anterior abdominal wall. Failure of diaphragm closure during this time of organ migration results in herniation of abdominal contents into the thoracic cavity.

CDH may occur as an isolated diaphragm defect (50-60% cases) or as part of a genetic syndrome known as syndromic or non-isolated CDH. Chromosomal anomalies are identified in 10 to 20% of prenatally identified cases (trisomies 18, 13, and 21). Other karyotype abnormalities, such as monosomy X, tetrasomy 12 p (isochromosome 12p), partial trisomy 5, partial trisomy 20, and polyploidies, have also been reported.

An underlying syndrome is present in approximately 10% of CDH cases occurring with associated anomalies. CDH is a prominent finding in the Fryns phenotype, Apert, Killian/Teschler-Nicola (Pallister-Killian), CHARGE, Coffin-Siris, Goltz, Perlman, Swyer, Brachmann-Cornelia De Lange, Goldenhar sequence, Beckwith Wiedemann, Simpson-Golabi-Behmel, Donnai-Barrow, Mathew-Wood, Jarcho-Levin, and others.

In stillborn infants with CDH or infants with bilateral CDH associated anomalies are more common (95%) including neural tube defects (anencephaly, myelomeningocele, hydrocephalus, and encephaloceles) and cardiac defects (ventriculoseptal defects, vascular rings, and coarctation of the aorta). Other midline developmental anomalies have also been reported, and include esophageal atresia, omphalocele, and cleft palate.

The vast majority of CDH occurs sporadically, with no identifiable familial link. While CDH has occurred in monozygotic twins, concordance for CDH is rare: in a large CDH registry, there were no concordant cases among the five monozygotic twin pairs. Familial cases involving autosomal recessive, autosomal dominant, and X-linked inheritance patterns have been reported. CDH inherited in an autosomal recessive pattern usually results in agenesis of the diaphragm and is usually fatal.

Many different chromosomal anomalies (e.g., deletions, duplications, translocations) and gene mutations have been identified among sporadic cases. Occurrence of CDH in several patients with deletions 15q26, 8p23, 8q22, 4p16, 1q42, and 3q22 allows to propose that these segments harbor the genes which, when deleted (or truncated) may be responsible for CDH. Segments 22q11, 4q28.3q32, 1q25q31.2 and 2p23p25 are good candidates for the location of genes which cause CDH in trisomic condition. These cases may represent de novo mutational events in genes for normal diaphragmatic development or reflect polygenic or multifactorial inheritance, or both.

The possibility of an environmental trigger is supported by cases of CDH as a manifestation of vitamin A deficiency and after exposure to thalidomide, anticonvulsants, and quinine. The etiology of CDH remains unknown at this time.

What laboratory studies should you request to help confirm the diagnosis? How should you interpret the results?

There are no laboratory studies currently available that help to confirm the diagnosis of CDH. Prenatally amniocentesis should be performed on all infants with suspected CDH especially in the presence of other anomalies to rule out chromosomal anomalies. Diagnosis of CDH is made radiographically with prenatal ultrasound/MRI and postnatal CXR.

Would imaging studies be helpful? If so, which ones?

Prenatally ultrasound is used to make the diagnosis of CDH with CDH diagnosed on either routine ultrasound or as part of the work up for polyhydramnios. Further radiologic evaluation is only useful to help determine outcome in infants with CDH.

Measurement of lung head ratio (LHR) prenatally by ultrasound have traditionally been used to determine outcome with CDH. This is most accurate when performed at 24-26 weeks gestation. LHR increases throughout gestation so most accurate assessments are determined at this time point. It is an estimate of contralateral lung size and mediastinal shift at the level of the atria on transverse scan of the fetal thorax. In left CDH, the LHR is calculated using a two-dimensional perpendicular linear measurement of right lung area (in square millimeters) divided by the head circumference (in millimeters) to minimize lung size differences owing to gestational age.

Initial outcome studies using LHR as a predictor of outcome, used a LHR of 1.4 as a cut-off for good versus poor prognosis with fetuses with a LHR greater than 1.4 surviving 100% of the time. More recently 1.0 has been used as a cut-off for poor prognosis with less than 50% survival if LHR is less than 1.0. With advances in neonatal care this measurement has become increasingly more unreliable and even, when measured at less than 1.0 seems to predict morbidity rather than mortality. For this reason other prenatal studies have been used to try and predict outcome.

Prenatal magnetic resonance imaging (MRI) has become increasingly used to help define the anatomy as well as help prognosticate outcome in CDH. This imaging modality while expensive can help better define the anatomy especially in terms of liver position and may be useful in predicting outcome. As LHR has become less predictive of outcome in CDH, liver position has become the only predictor of mortality in CDH with liver in the chest predicting mortality in 45-50% of cases.

MRI is able to more accurately determine liver position as well as measure volume of liver in the chest as it relates to thoracic volume. When liver volume to thoracic volume exceeds 20% mortality is around 85%. MRI has also been useful in predicting lung volumes. An assessment of lung to thoracic volume can be obtained, with values less than 15% indicating a poor prognosis and values in excess of 30% associated with 100% survival.

Another measurement is the modified McGoon index, which can be obtained by both MRI as well as fetal cardiac echo. This is a ratio between the diameters of the 2 pulmonary arteries and the aorta at the level of the diaphragm and calculated using the following equation - modified McGoon index (MGI) [=(diameter of the right pulmonary artery + LPA(d))/diameter of aorta]. A value of less than 0.88 is predictive of developing suprasystemic pulmonary artery hypertension after birth with a 90% sensitivity and specificity.

Along with these measurements fetal MRI is also useful in terms of ruling out other fetal anomalies that may be present as high as 50% of the time in infants with CDH. A detailed fetal examination with MRI can help to rule out other associated anomalies.

Fetal echo may also have value in terms of helping to prognosticate outcome. Fetal echo helps to confirm normal fetal cardiac anatomy as associated cardiac malformations may be present in up to 20% of CDH cases. This modality can also be used to calculate pulmonary artery (PA) size which correlates with degree of lung hypoplasia.

PA size in mm should roughly equal gestational age. Using this measurement one can estimate degree of lung hypoplasia on the ipsilateral and contralateral side of the defects and predict mortality. A main PA measurement of less than 1, contralateral PA size less than 0.82 and ipsilateral PA size less than 0.67 observed/expected PA size correlates with mortality in CDH. Along with this measurement one can calculate the MGI to see if this correlates with MRI measurements of this index.

Maternal hyperoxia has also been used as a predictor of lethal pulmonary hypoplasia. Administering 60% oxygen by facemask to the mother and measuring fetal pulmonary blood flow by echo has been shown to have predictive value with respect to mortality from pulmonary hypoplasia. A reactive test (greater than 20% increase in pulmonary blood flow) predicted survival for 92% of infants; a nonreactive test (less than 20% increase in pulmonary blood flow) predicted fetal death from pulmonary hypoplasia for 79% of infants.

Despite these imaging studies prenatal ascertainment of CDH outcomes and survival remains a challenge with poor predictive value. Better predictors of outcomes are desperately needed to help with decisions around prenatal termination of pregnancy, prenatal intervention for CDH, postnatal management and ECMO.

Post delivery plain radiographs of the chest confirm the diagnosis with bowel or stomach in the chest on CXR. The presence of an OG tube coiled on the stomach on CXR with the stomach in the chest is diagnostic. Ultrasound with Doppler looking at the hepatic venous flow pattern can confirm the diagnosis of liver in the chest postnatally. Echocardiogram post delivery to evaluate pulmonary artery pressure and left and right ventricular function is useful.

If you are able to confirm that the patient has congenital diaphragmatic hernia, what treatment should be initiated?

The approach to postnatal management of CDH is complex and involves a multidisciplinary care team approach including MFM, neonatology, pediatric surgery and ECMO as well as feeding, developmental and rehabilitation specialists. Despite advances in neonatal care mortality for CDH still ranges between 25 and 50% with significant short and long term morbidity.

There are many controversies surrounding postnatal care from place of delivery to timing of delivery and subsequent postnatal management. There are to date no studies that support unequivocally delivery of all CDH patients in a center that can provide Extracorporeal Membrane Oxygenation Therapy (ECMO). Despite this it is still highly recommended that infants who postnatal ECMO therapy will be offered deliver in a center able to provide ECMO. Transfer of these infants post delivery to an ECMO center has significant risks and is not recommended electively.

Timing of delivery for infants with CDH is not clearly defined in the literature as well. There are two studies in the literature looking at timing of delivery related to gestational age both with conflicting results. The first study compared 38-40 week gestation delivery to 40-42 week gestation delivery and found greater survival (63% vs 52%), less time on ECMO and shorter hospital stay with late delivery. The second study compared 37-39 weeks and 39-41 weeks gestation delivery and found improved survival (75% versus 65%), and less use of ECMO provided infants were greater than 3.1kg in the early delivery group.

There are currently no guidelines for when to deliver infants with CDH, but close adherence to ACOG recommendations and avoiding elective deliveries prior to 39 weeks is advisable. There is no benefit of caesarian section delivery over vaginal birth.

Delivery room management

Delivery room management consists of intubation as soon after the first breath as possible to avoid air entrainment into the stomach and bowel with the infants respiratory effort. In CDH patients the position of the carina is often higher than in normal neonates so close attention to ET tube position should be taken to prevent right mainstem intubation.

There are currently no published guidelines or recommendations for fraction of inspired oxygen to resuscitate these infants with, but animal studies would suggest that resuscitation with high inspired oxygen blunts the later response to pulmonary vasodilators in animals with PPHN. Acute hypoxia results in vasoconstriction in a pulmonary vascular bed that is already underdeveloped and abnormal so if low inspired oxygen is used for resuscitation close monitoring of preductal saturations and titration of inspired oxygen is recommended.

As soon as the airway is secured, an orogastric tube (OG) should be placed and advanced a few cm further than the standard measurement to optimize drainage. Once placed the OG tube should be placed to low intermittent suction. Positive pressure ventilation (PPV) should be initiated with preferably a T piece resuscitator with close attention to peak inspiratory pressure (PIP) as lungs are hypoplastic and at high risk for barotrauma.

There is variance in the literature about what PIP to resuscitate infants with, with a PIP of 25 or less recommended most frequently. Avoiding PIPs greater than 30 is strongly advised unless needed in an emergent situation (baby bradycardic and not responding to resuscitation). A high rate 40-60 is recommended as these infants very commonly have issues with ventilation and hypercapneic respiratory failure. Goal saturations preductally in the delivery room range from 75%-85%. Once goal saturations are reached dialing back the PIP is recommended to avoid lung injury. In the infant not responding to routine NRP resuscitation one should have a low threshold for ruling out a pneumothorax on the contralateral side of the defect. Once stable infant should be transferred to the NICU.

Post delivery room care

Post delivery room care is complex and involves close monitoring of respiratory status, gas exchange, pulmonary hypertension and hemodynamics including cardiac performance. Monitoring should include a CR monitor, continuous pre and postductal saturation measurements and frequent non invasive blood pressure measurements until invasive arterial access can be obtained. Umbilical arterial and venous access should be obtained as soon as possible and blood pressure transduced continuously once arterial access is obtained. Umbilical venous access is often difficult when the liver is malpositioned and a low lying umbilical venous line is acceptable until central venous access can be obtained.

Close monitoring of respiratory status and gas exchange postnatally is extremely difficult in infants with CDH as degree of lung hypoplasia, relative lung immaturity and pulmonary hypertension all determine gas exchange. Optimal lung inflation (Functional Residual Capacity (FRC)) can be difficult to achieve and determine, and pulmonary hypertension can be exacerbated by under and overinflation of the lung. A 8 rib inflation is thought to be optimal in the setting of CDH, but often due to mediastinal shift, the cardiothymic silhouette obscures visualization of the lung fields on the contralateral side and inflation cannot be assessed by counting the number of ribs.

In most cases the only lung that is seen is a small triangle at the costophrenic angle on the contralateral side. Overdistention of the contralateral lung will result in shift of the mediastinum back to the midline and flattening of the diaphragm on the contralateral side. A PIP of less than 25 should be used if ventilating in a pressure control mode. If ventilating in a volume control mode optimal tidal volume should be determined by degree of lung hypoplasia (range 4-6mls/kg) with 5mls/kg as a starting point. If a PIP greater than 25 and rate greater than 40 is needed to optimally inflate the lung and ventilate the baby one should consider the use of high frequency oscillatory ventilation (HFOV).

Echocardiography is key in terms of optimal management of hemodynamics. Initial assessments include measurements of right and left ventricular systolic function, right ventricular systolic pressure (RVSP) as measured by TR jet, presence or absence of patent ductus arteriosus (PDA) and direction of flow across the PDA and patent foramen ovale (PFO).

Systemic or suprasystemic RVSP is a common finding in the setting of CDH. Whether this is contributing to hypoxemia depends on the direction of blood flow across fetal conduits (PFO, PDA). A common echocardiographic finding in the setting of CDH is a PDA with right to left flow indicating a suprasystemic RVSP and PFO with left to right flow. Left to right flow across the PFO in the setting of a suprasystemic RVSP raises the concern for left ventricular (LV) diastolic dysfunction and a left atrial (LA) pressure that exceeds right atrial (RA) pressure. These patients have a significant differential in pre and postductal saturations and placement of a preductal arterial line (right radial or brachial artery) is extremely beneficial.

In the presence of a left to right PFO an arterial blood gas (ABG) drawn from a preductal site, provides a direct measure of the ability of the lungs to oxygenate and ventilate. With a large right to left shunt across the PDA a postductal ABG may be equivalent to a venous blood gas (VBG) and markedly different from a preductal ABG. Along with preductal blood gas sampling use of an agent to reduce LV afterload such as milrinone may be more beneficial than inhaled nitric oxide (iNO), allowing for LV relaxation during diastole and better LV filling.

Use of iNO in this setting is likely to increase LA pressure further by increasing the amount of blood flow to the lungs and returning to the LV with resultant pulmonary hemorrhage. Initial trials with iNO in infants with CDH showed increased need for ECMO in infants treated with iNO raising the concern for the use of iNO in infants with LV diastolic dysfunction. Infants with CDH often have profound LV systolic dysfunction and it is usually in this setting that infants require ECMO therapy.

LV systolic dysfunction results in hemodynamic compromise and inadequate tissue perfusion with shock. There have been published case reports of LV dysfunction in CDH that is so severe that in the presence of a wide open PDA and suprasystemic RVSP, retrograde flow of ductal blood up the aortic arch occurs with preductal mixing of oxygenated and deoxygenated blood and resultant severe preductal and postductal desaturation. For this subset of infants ECMO therapy can be lifesaving.

Due to the risk of increasing preload to an already compromised LV, use of iNO in infants with severe LV dysfunction is not recommended. When right to left shunt is seen across the PFO, severe hypoxemia ensues and use of iNO is indicated and usually beneficial. iNO reduces RV afterload and promotes forward flow to the lungs, with less regurgitation of blood from the RV to the RA, decreasing RA pressure with less intracardiac mixing of deoxygenated blood with oxygenated blood. This usually results in a marked improvement in oxygenation.

In infants who fail to respond to iNO RV failure often occurs, due to a suprasystemic RVSP. Use of prostaglandin E1 is often beneficial to recruit the PDA and offload the RV. In the presence of a wide open ductus the suprasystemic RVSP offloads into the systemic circulation, with improvement in RV function and often improvement in oxygenation, as there is less regurgitation of deoxygenated blood from the RV into the RA and less intracardiac mixing.

The cardiopulmonary interactions that exist in the setting of CDH are quite complex and the decision whether or not to dilate the pulmonary circulation with iNO is often confusing. Understanding the contribution of LV diastolic and systolic dysfunction to the pathogenesis of pulmonary hypertension and cardiopulmonary failure in CDH, allows for selective iNO and ECMO use for the subset of infants most likely to benefit from these therapies.

iNO is not FDA approved for use in CDH due to failure to show benefit in randomized trials. Despite this use of iNO in infants with CDH has increased over the last decade from 30% to 80% (data from CDH study group). Along with increased iNO use in CDH there has been increased survival as well from 46% to 60% over the same time period. These data suggest that selective use of iNO, in patients with physiology that is likely to benefit from selective pulmonary vasodilation, can improve survival in CDH.

Hemodynamic support with vasopressors is often needed in infants with CDH. Cardiopulmonary collapse that accompanies severe LV failure in CDH is best treated with ECMO, making ECMO a key component in the management of infants with CDH. In the presence of adequate LV systolic function hypotension often occurs. The etiology is often multifactorial. Inadequate LV and RV preload is often a contributing factor and aggressive fluid resuscitation often needed for treatment of hypotension. Using a pressure transducer attached to a central venous catheter (CVP) can assist with whether or not a fluid bolus is indicated. A CVP of less than 6 should alert the clinician to the need for volume resuscitation.

Often vasopressor agents such as dopamine and epinephrine are needed to increase systemic blood pressure and improve tissue perfusion. In cases where iNO fails to reduce pulmonary vascular resistance (PVR) these agents can also be used to artificially increase systemic vascular resistance (SVR) above PVR. When SVR increases above PVR pulmonary blood flow improves along with oxygenation. Excessive afterload in the setting of LV systolic and diastolic dysfunction can decrease tissue perfusion and adversely affect LV function, making ECMO an important therapy for LV dysfunction in CDH.

Catecholamine resistant shock is a frequent occurrence in CDH, and it is in this setting that agents such as vasopressin are often more effective. Vasopressin causes catecholamine independent vasoconstriction through activation of V1, V2 and V3 receptors. Recently case reports of vasopressin use in CDH have demonstrated its efficacy in improving tissue perfusion, pulmonary blood flow and outcomes in infants with CDH. Water retention is a common side effect of vasopressin and close monitoring of serum sodium indicated in all infants needing vasopressin therapy.

An often overlooked cause of hemodynamic compromise in infants with CDH is adrenal insufficiency. Adrenal insufficiency results in profound shock, often without hypoglycemia. hyponatremia or hyperkalemia. The diagnosis is made by measuring a random cortisol and if low administering ACTH to see if a cortisol response with an increase in serum cortisol occurs. Supplementation with hydrocortisone in stress doses often results in significant hemodynamic stabilization.

Timing of repair in CDH is controversial and optimal timing of repair is yet to be elucidated. The best time to repair the CDH is when the infant is stable, however the corollary to that premise is the longer one waits to repair the defect the longer the duration of exposure to central venous access and the longer the duration of enteral starvation. There are many centers that advocate for resolution of pulmonary hypertension prior to repair. This often takes months with increased risk of central line associated infection and sepsis.

Finding a balance between repairing the defect when the infant in stable, and minimizing exposure to central venous access and prolonged enteral starvation is optimal. If an infant can be stabilized without ECMO and hemodynamics and gas exchange are stable, repair is probably safe even if pulmonary hypertension has not resolved. While repair of the diaphragm defect is unlikely to improve pulmonary mechanics and pulmonary hypertension, CDH repair allows for initiation of enteral feeds and removal of central venous catheters and respiratory support.

Post repair of CDH a pneumothorax will be present within the left chest, that does not require drainage or chest tube placement unless under tension. Determining whether or not the pneumothorax is under tension may be difficult due to the lung hypolasia and mediastinal shift associated with that. Post repair the mediastinum should shift back to the midline and within 24 hours of repair the air filled space should be replaced with fluid.

Failure of the space to fill with fluid should raise concern for a pneumothorax and possible need for chest tube placement. If still unsure one could place a butterfly needle into the chest with the back end of the butterfly to water seal in a sterile water container. If continuous bubbling is seen, an airleak is most likely present and thoracostomy tube placement indicated. The thoracaostomy tube should be placed to water seal and not suction as excessive suction could pull the entire mediastinum over creating a tension effect and hemodynamic instability due to the potential space created by the hypolastic lung on the ipsilateral side of the defect.

Persistent pulmonary hypertension of the newborn complicates CDH in the long term and has significant morbidity and mortality. CDH can be divided into 3 groups with respect to the pulmonary vasculature. The 1st group has a no pulmonary hypertension and has a survival of 100%, the 2nd group has pulmonary hypertension that is present at birth and may be quite severe, but resolves stepwise by 1 month to 6 weeks of age, survival for this group is 75%. The 3rd group has pulmonary hypertension that remains systemic beyond 6 weeks of age, reported mortality for this group is as high as 100%.

For this reason patients with systemic pulmonary hypertension at 4-6 weeks of age are best cared for in a center that has expertise in management of severe pulmonary hypertension. Management of pulmonary hypertension involves optimization of lung recruitment and reliable iNO delivery with tracheostomy and chronic ventilation, aggressive management of gastro-esophageal reflux and nutrition with Nissan Funduplication and gastrostomy tube placement and cardiac catheterization once lung disease is optimized to quantify degree of pulmonary hypertension and rule out structural abnormalities in the pulmonary arteries and veins.

Because LV hypoplasia commonly complicates CDH, measurements of LV end diastolic pressure (LVEDP) and LA pressure are essential to determine if pulmonary venous hypertension is complicating the course and contributing to the severity of the pulmonary hypertension. In this setting LV afterload reduction with milrinone or ACE inhibitors is often beneficial. In the absence of structural abnormalities of the pulmonary arteries and veins and elevated LVEDP and LA pressure, the addition of alternative pulmonary vasodilators such as sildenafil (phosphodiesterase 5 inhibitor) and bosentan (non selective endothelin-1 receptor blocker) may be of benefit. Referral to a center with experience with chronic ventilation and use of these agents is strongly advised.

The most devastating long term complication with CDH is hospital acquired infection, central line associated blood stream infection and ventilator associated pneumonia. These conditions can dramatically alter the course in CDH and usually result in significant morbidity and mortality and often increased hospital days. Repair of the diaphragm defect as soon as the infant is medically stable, early initiation of enteral feeds with removal of central venous catheters and extubation as soon as the patient is medically stable are all key in terms of infection prevention.

When infection is suspected early evaluation and initiation of antibiotic therapy is recommended. Removal of foreign bodies as soon as possible is the only effective way of preventing infection.

What are the adverse effects associated with each treatment option?

While the mortality for CDH has improved over the last decade or so, morbidity associated with this disease remains an significant problem. The most significant adverse outcome and morbidity associated with CDH is long term neurodevelopmental handicap. CDH has been associated with significant impairments in widespread neurodevelopmental domains: fine and gross motor skills, visuospatial skills, and cognition, attention and behavioral skills, and speech and language skills. Both neurocognitive and neuromotor scores are in the low normal range.

Along with these defects specific learning disability, attention deficit hyperactivity disorder and problems with executive function, cognitive and attentional weaknesses, and social difficulties are more common in CDH patients. Sensorineural hearing loss is a common complication of CDH both in infants treated with ECMO and without, with a higher incidence in ECMO treated infants.

Many studies have implicated patient related factors such as intrathoracic liver position, need for extracorporeal membrane oxygenation (ECMO), and patch requirement, supplemental oxygen requirement beyond 30 days of life, hypotonicity, socio-economic status, and even surgical intervention itself as strong predictors of impaired neurological function in CDH survivors.

In all studies liver position has most strongly been associated with adverse outcome. Perioperative hypocapnia is often needed to manage the pulmonary hypertension and allow for adequate oxygenation. Perioperative hypocapnea was especially linked to executive dysfunction, behavioral problems, lowered intelligence, and poor achievement in mathematics.

While severity of CDH and degree of illness strongly correlate with adverse neurodevelopmental outcome, there is increasing evidence that the central nervous system development might be abnormal in children with CDH. Recent reports have demonstrated a high incidence of cerebral abnormalities found by magnetic resonance imaging in CDH newborns including gray and white matter injury, subpendymal hemorrhage, and impairment of the posterior limb of the internal capsule. Others have demonstrated that cerebral blood flow is significantly altered as evidenced by high grade stenosis of intracranial vessels and the development of intra- and extra-cranial collateral vessels.

Newborns with CDH are also exposed to the potential risk of hypoxia/ischemia, emboli, reactive oxygen species, acidosis, hypotension, and inflammatory microvasculopathy before and after surgery, all of which may increase the risk of neurodevelopmental delay. There is increasing evidence that systematic inflammatory response of the fetus and neonate is an important mechanism of white matter injury. Exposure of the brain to neurotrophic drugs (e.g., barbiturates, benzodiazepine) commonly used in pediatric anesthesia, intensive care medicine (e.g., muscle relaxants) and increased cytokine release during the systemic inflammatory response may cause widespread apoptotic neurodegeneration, decrease of oligodendrocytes myelination, and increase astrocytosis.

Along with adverse neurodevelopmental outcomes in CDH, significant respiratory morbidity co-travels with CDH. Survivors of CDH are at increased risk for viral respiratory infections and repeated hospitalizations in the 1st year of life. The prevalence of recurrent respiratory infections (>3 episodes per year) has been reported up to 25% to 50% of affected children in the first year of life. For this reason vaccination against influenza and RSV immune globulin are strongly recommended for all survivors of CDH.

Failure to thrive with significantly impaired weight and height are common in CDH survivors. This may be secondary to severe gastroesophageal reflux disease, inadequate oral intake or recurrent infections. After 1 year of age there is slow and progressive improvement in growth.

Chest wall deformities and scoliosis occur much more frequently in infants with CDH especially after patch repair and should be screened for at provider visits.

What are the possible outcomes of congenital diaphragmatic hernia?

Despite advances in neonatal care, mortality for CDH ranges from 25-35% with an even higher mortality when one includes prenatal termination of pregnancy. In some reports mortality for CDH is as high as 50%. Prenatal predictors of outcome are unreliable with no single prenatal factor predictive of mortality with a high enough sensitivity and specificity to adequately counsel families.

Due to the severity of CDH as a birth defect, many centers are recommending termination of pregnancy for CDH. Many factors have been used prenatally to predict outcome, LHR, lung volume by ultrasound or MRI, pulmonary artery size by fetal echo or MRI, liver position and volume of liver in the chest and the presence of other anomalies. In all studies the only reliable predictor of mortality is the presence of other anomalies, and either a genetic or chromosomal abnormality, which increases mortality to 70-85%.

For patients with isolated CDH, position of the liver is the most reliable predictor of outcome and probably predicts morbidity rather than mortality. Intrathoracic liver position is associated with increased mortality with survival in this group ranging from 40-50% with much higher long term morbidities.

With this in mind counseling of patients remains a challenge. Informing families that their fetus has a severe birth defect that is likely to alter their long term pregnancy outcome and quality of life is important so that families understand the significance of carrying a pregnancy with an CDH fetus to term. While the diagnosis of CDH is likely to alter pregnancy and long term outcomes, due to the unreliability of prenatal factors in CDH, predicting mortality prenatally is very difficult. An understanding for which infants are likely to have severe disease as well as informing families that until the baby is born it is almost impossible to predict outcome and the postnatal course, is probably the best approach to prenatal counseling in CDH.

Educating families about CDH, the short and long term treatments and complications as mentioned above is an important part of counseling about outcomes. ECMO should be included in all discussions around CDH, both the risks and benefits of this therapy as well as the importance of ECMO as a treatment for LV failure and pulmonary hypertension in CDH.

What causes this disease and how frequent is it?

CDH occurs in 1:1000-1:200 live births and probably more commonly in utero as many infants born stillborn have associated CDH. The etiology of CDH has not been established definitively. The leading theories are that it is due to failure of normal closure of the pleuroperitoneal folds during the 4th to 10th weeks postfertilization resulting in a diaphragm defect. During this same time the abdominal viscera which are external to the fetus, return to the abdominal cavity allowing for closure of the anterior abdominal wall. Failure of diaphragm closure during this time of organ migration results in herniation of abdominal contents into the thoracic cavity.

CDH may occur as an isolated diaphragm defect (50-60% cases) or as part of a genetic syndrome known as syndromic or non-isolated CDH. Chromosomal anomalies are identified in 10 to 20% of prenatally identified cases (trisomies 18, 13, and 21). Other karyotype abnormalities, such as monosomy X, tetrasomy 12 p (isochromosome 12p), partial trisomy 5, partial trisomy 20, and polyploidies, have also been reported.

An underlying syndrome is present in approximately 10% of CDH cases occurring with associated anomalies. CDH is a prominent finding in the Fryns phenotype, Apert, Killian/Teschler-Nicola (Pallister-Killian), CHARGE, Coffin-Siris, Goltz, Perlman, Swyer, Brachmann-Cornelia De Lange, Goldenhar sequence, Beckwith Wiedemann, Simpson-Golabi-Behmel, Donnai-Barrow, Mathew-Wood, Jarcho-Levin, and others.

In stillborn infants with CDH or infants with bilateral CDH associated anomalies are more common (95%) including neural tube defects (anencephaly, myelomeningocele, hydrocephalus, and encephaloceles) and cardiac defects (ventriculoseptal defects, vascular rings, and coarctation of the aorta). Other midline developmental anomalies have also been reported, and include esophageal atresia, omphalocele, and cleft palate.

The vast majority of CDH occurs sporadically, with no identifiable familial link. While CDH has occurred in monozygotic twins, concordance for CDH is rare: in a large CDH registry, there were no concordant cases among the five monozygotic twin pairs. Familial cases involving autosomal recessive, autosomal dominant, and X-linked inheritance patterns have been reported. CDH inherited in an autosomal recessive pattern usually results in agenesis of the diaphragm and is usually fatal.

Many different chromosomal anomalies (e.g., deletions, duplications, translocations) and gene mutations have been identified among sporadic cases. Occurrence of CDH in several patients with deletions 15q26, 8p23, 8q22, 4p16, 1q42, and 3q22 allows to propose that these segments harbor the genes which, when deleted (or truncated) may be responsible for CDH. Segments 22q11, 4q28.3q32, 1q25q31.2 and 2p23p25 are good candidates for the location of genes which cause CDH in trisomic condition. These cases may represent de novo mutational events in genes for normal diaphragmatic development or reflect polygenic or multifactorial inheritance, or both.

The possibility of an environmental trigger is supported by cases of CDH as a manifestation of vitamin A deficiency and after exposure to thalidomide, anticonvulsants, and quinine.The etiology of CDH remains unknown at this time.

How do these pathogens/genes/exposures cause the disease?

Unknown, it is an epidemiologic association that connects CDH to vitamin A deficiency and after exposure to thalidomide, anticonvulsants, and quinine.

What is the evidence?

Slavotinek, AM. "The genetics of congenital diaphragmatic hernia". Semin Perinatol. vol. 29. 2005. pp. 77.

(While the etiology of CDH is multifactorial, some candidate genes contributing to CDH have been described. This paper provides insights into those genes.)

Kling, DE, Schnitzer, JJ. "Vitamin A deficiency (VAD), teratogenic, and surgical models of congenital diaphragmatic hernia (CDH)". Am J Med Genet C Semin Med Genet. vol. 145C. 2007. pp. 139.

(This paper describes other etiologies for CDH.)

Sweed, Y, Puri, P. "Congenital diaphragmatic hernia: influence of associated malformations on survival". Arch Dis Child. vol. 69. 1993. pp. 68.

(This paper describes factors affecting outcome in CDH.)

Worley, KC, Dashe, JS, Barber, RG. "Fetal magnetic resonance imaging in isolated diaphragmatic hernia: volume of herniated liver and neonatal outcome". Am J Obstet Gynecol. vol. 200. 2009. pp. 318.e1.

(This paper describes the utility of prenatal MRI in assisting with the diagnosis and prognosis in CDH.)

Hubbard, AM, Crombleholme, TM, Adzick, NS. "Prenatal MRI evaluation of congenital diaphragmatic hernia". Am J Perinatol. vol. 16. 1999. pp. 407.

(This paper describes the utility of prenatal MRI in assisting with the diagnosis and prognosis in CDH.)

Hubbard, AM, Adzick, NS, Crombleholme, TM, Haselgrove, JC. "Left-sided congenital diaphragmatic hernia: value of prenatal MR imaging in preparation for fetal surgery". Radiology. vol. 203. 1997. pp. 636.

(This is the first description of prenatal MRI for the evaluation of CDH.)

Ruano, Rodrigo, Aubry, Marie-Cécile, Barthe, Bruno, Mitanchez, Delphine, Dumez, Yves, Benachi, Alexandra. "Predicting perinatal outcome in isolated congenital diaphragmatic hernia using fetal pulmonary artery diameters". Journal of Pediatric Surgery. vol. 43. 2008. pp. 606-611.

Sokol, Jenni, Shimizu, Naoki, Bohn, Desmond, Doherty, Dorota, Ryan, Greg, Hornberger, Lisa K. "Fetal pulmonary artery diameter measurements as a predictor of morbidity in antenatally diagnosed congenital diaphragmatic hernia: A prospective study". American Journal of Obstetrics and Gynecology. vol. 195. 2006. pp. 470-7.

(This paper describes the use of prenatal echo measurements of pulmonary artery size as a marker of outcome in CDH.)

Broth, Richard E, Wood, Dennis C, Rasanen, Juha, Sabogal, Juan Carlos, Komwilaisak, Ratana, Weiner, Stuart, Berghella, Vincenzo. "Prenatal prediction of lethal pulmonary hypoplasia: The hyperoxygenation test for pulmonary artery reactivity". Am J Obstet Gynecol. vol. 187. 2002. pp. 940-5.

Vuletin, JF, Lim, FY, Cnota, J, Kline-Fath, B, Salisbury, S, Haberman, B. "Prenatal pulmonary hypertension index: novel prenatal predictor of severe postnatal pulmonary artery hypertension in antenatally diagnosed congenital diaphragmatic hernia". J Pediatr Surg. vol. 45. 2010. pp. 703-8.

Seetharamaiah, R, Younger, JG, Bartlett, RH, Hirschl, RB. "Congenital Diaphragmatic Hernia Study Group Factors associated with survival in infants with congenital diaphragmatic hernia requiring extracorporeal membrane oxygenation: a report from the Congenital Diaphragmatic Hernia Study Group". J Pediatr Surg. vol. 44. 2009. pp. 1315-21.

(This report from the CDH study group describes factors determining survival in infants with CDH needing ECMO.)

Frisk, V, Jakobson, LS, Unger, S, Trachsel, D, O'Brien, K. "Long-term neurodevelopmental outcomes of congenital diaphragmatic hernia survivors not treated with extracorporeal membrane oxygenation". J Pediatr Surg. vol. 46. 2011. pp. 1309-18.

(This paper addresses long term neurodevelopmental outcomes in CDH infants.)

Frenckner, BP, Lally, PA, Hintz, SR, Lally, KP. "Prenatal diagnosis of congenital diaphragmatic hernia: how should the babies be delivered?". J Pediatr Surg. vol. 42. 2007. pp. 1533-8.

(This paper addresses the issue of caesarian section versus vaginal birth for infants with CDH.)

Moya, Fernando R, Lally, Kevin P. "Evidence-Based Management of Infants with Congenital Diaphragmatic Hernia". Semin Perinatol. vol. 29. 2005. pp. 112-117.

(This review addresses the evidence for various treatment strategies in the management of infants with CDH.)

Logan, J. Wells, Cotten, C. Michael, Goldberg, Ronald N, Clark, Reese H. "Mechanical ventilation strategies in the management of congenital diaphragmatic hernia". Seminars in Pediatric Surgery. vol. 16. 2007. pp. 115-125.

(This paper describes different ventilation strategies for infants with CDH.)

"Inhaled nitric oxide and hypoxic respiratory failure in infants with congenital diaphragmatic hernia. The Neonatal Inhaled Nitric Oxide Study Group (NINOS)". Pediatrics. vol. 99. 1997. pp. 838-45.

(This paper describes a randomized trial of inhaled NO in infants with CDH, demonstrating worse outcomes in infants treated with inhaled NO.)

Ledbetter, DJ, Sawin, RS. "Adrenal insufficiency in newborns with congenital diaphragmatic hernia". J Pediatr. vol. 157. 2010. pp. 696.

(Describes the prevalence of adrenal insufficiency in infants with CDH and the utility of supplementation with hydrocortisone.)

Kamath, BD, Fashaw, L, Kinsella, JP. "Adrenal insufficiency in newborns with congenital diaphragmatic hernia". J Pediatr. vol. 156. 2010. pp. 495-497.e1.

(Describes the prevalence of adrenal insufficiency in infants with CDH and the utility of supplementation with hydrocortisone.)

Dillon, PW, Cilley, RE, Mauger, D, Zachary, C, Meier, A. "The relationship of pulmonary artery pressure and survival in congenital diaphragmatic hernia". J Pediatr Surg. vol. 39. 2004. pp. 307-12.

(This paper correlates survival with pulmonary artery pressure at various post delivery ages.)

Tracy, S, Estroff, J, Valim, C, Friedman, S, Chen, C. "Abnormal neuroimaging and neurodevelopmental findings in a cohort of antenatally diagnosed congenital diaphragmatic hernia survivors". J Pediatr Surg. vol. 45. 2010. pp. 958-65.

Trachsel, D, Selvadurai, H, Bohn, D, Langer, JC, Coates, AL. "Long-term pulmonary morbidity in survivors of congenital diaphragmatic hernia". Pediatr Pulmonol. vol. 39. 2005. pp. 433-9.

(This paper describes the long term pulmonary morbidity for infants with CDH.)

Harrison, MR, Mychaliska, GB, Albanese, CT, Jennings, RW, Farrell, JA, Hawgood, S. "Correction of congenital diaphragmatic hernia in utero IX: fetuses with poor prognosis (liver herniation and low lung-to-head ratio) can be saved by fetoscopic temporary tracheal occlusion". J Pediatr Surg. vol. 33. 1998. pp. 1017-22.

(This paper describes a fetoscopic approach to treating CDH in utero with baloon occlusion of the trachea.)

Mychaliska, GB, Bullard, KM, Harrison, MR. "In utero management of congenital diaphragmatic hernia". Clin Perinatol. vol. 23. 1996. pp. 823-41.

(This paper describes a fetoscopic approach to treating CDH in utero with balloon occlusion of the trachea.)

Harrison, MR, Keller, RL, Hawgood, SB, Kitterman, JA, Sandberg, PL, Farmer, DL. "A randomized trial of fetal endoscopic tracheal occlusion for severe fetal congenital diaphragmatic hernia". N Engl J Med. vol. 349. 2003. pp. 1916.

(This paper describes a fetoscopic approach to treating CDH in utero with balloon occlusion of the trachea and the results of a randomized trial in human infants with CDH.)

Harrison, MR, Sydorak, RM, Farrell, JA, Kitterman, JA, Filly, RA, Albanese, CT. "Fetoscopic temporary tracheal occlusion for congenital diaphragmatic hernia: prelude to a randomized, controlled trial". J Pediatr Surg. vol. 38. 2003. pp. 1012-20.

(This paper describes the rationale for undertaking a randomized trial of tracheal occlusion in utero for infants with severe CDH.)

Ongoing controversies regarding etiology, diagnosis and treatment

The biggest controversy in CDH is around prenatal intervention for CDH. CDH with a low LHR and presence of liver in the chest is associated with increased morbidity and mortality and the question that remains unanswered is whether there is a prenatal intervention that can alter these outcomes for infants with CDH. The rationale for prenatal therapy in severe CDH is to prevent or reverse pulmonary hypoplasia, and restore adequate lung growth for neonatal survival. This hypothesis was supported by work in sheep, which showed that if the space occupying abdominal viscera were reduced from the chest cavity, then the lungs would grow and develop normally.

Open repair of the diaphragm defect in utero while successful in treating CDH was abandoned early on due to fetal demise in infants with intrathoracic liver position secondary to kinking of the umbilical vein and in patients where the liver was intraabdominal, in utero repair of the diaphragm defect was limited by preterm birth and the consequences of prematurity.

Fetal tracheal occlusion (TO) is an alternative prenatal therapy which avoids kinking of the umbilical vein, and, more importantly, reverses lung hypoplasia. The rationale for this approach is that the dynamics of fetal lung fluid can dramatically affect lung growth.

Under normal circumstances, the lungs are net producers of amniotic fluid with lung liquid volume and intratracheal pressure maintained at constant values by fetal laryngeal mechanisms. Increasing the egress of lung fluid by fetal tracheostomy, induced oligohydramnios, or recurrent laryngeal nerve or cord transection results in pulmonary hypoplasia. Conversely, prenatal TO obstructs the normal egress of lung fluid during pulmonary development, increasing transpulmonic pressure and resulting in large fluid-filled lungs. This hypothesis was supported by multiple studies in animal models that showed prenatal TO accelerated parenchymal lung growth in normal and hypoplastic lungs.

In the first human CDH/TO study, exposure of the fetal head and neck via a hysterotomy was performed to occlude the fetal trachea; this procedure was associated with a 13% survival rate. Using a similar open-fetal surgical approach, a subsequent study reported a 33% postnatal survival rate. In both studies, predicted mortality approached 100%; therefore, surgery improved the predicted outcome. However, lung growth was variable, postnatal respiratory function was marginal, and infants often required long-term, aggressive ventilatory support. This appears to be due to a diminished number of alveolar epithelial type-II cells after prolonged periods of TO, leading to decreased levels of pulmonary surfactant gene and protein expression.

Techniques to achieve minimally invasive fetoscopic reversible fetal TO have been developed to decrease the risks of preterm labor and restore surfactant deficiency. A NIH sponsored trial of fetoscopic TO for fetuses with liver herniation and LHR <1.4 was halted after enrollment of 24 patients demonstrated no survival advantage and an increased rate of prematurity (mean gestational age at delivery 31 weeks versus 37 weeks for standard postnatal care).

Important factors in offering prenatal therapy continue to be, first and foremost, determining which fetuses have a poor prognosis. Due to the unpredictable nature of prenatal factors in CDH which patients will benefit from a prenatal intervention is unclear. Along with this the optimal timing, duration, and release of occlusion in humans is not known.

Fetal TO in sheep has been shown by multiple investigators to be of benefit, but the same has not been demonstrated in humans. The Eurofetus group has had early success with fetoscopic reversible TO in fetuses with liver herniation and LHR less than 1.0, and these studies are ongoing. In their report of TO in 24 poor prognosis fetuses (LHR <1.0 and liver herniation) in whom occlusion was performed at 26 to 28 weeks and the balloon was retrieved at 34 weeks, mean gestational age at delivery was 33.5 weeks and survival to discharge was 12 of 24 (50%).

A randomized trial of standard postnatal therapy and fetoscopic TO in the severest of CDH (liver herniation, LHR <1) is needed to examine the utility of TO in CDH.

EXIT to ECMO (Ex Utero Intrapartum Treatment to Extracorporeal Membrane Oxygenation) is another available treatment for CDH that is controversial. It involves delivery of the head and shoulders, placement of a endotracheal tube and cannulation to ECMO prior to separation from the placental circulation. This allows for a smooth transition from the womb to extrauterine life. There is clearly a subset of infants who would benefit from this therapy, but defining which patients these are is very difficult.

Prenatal predictors of outcome in CDH continue to be unreliable and until better prenatal indicators of outcome are available determining which patients will benefit from an EXIT to ECMO procedure is unclear.

You must be a registered member of Infectious Disease Advisor to post a comment.

Sign Up for Free e-newsletters