Diaphragmatic hernia is a condition in which intra-abdominal contents herniate into the thoracic cavity via a defect in the diaphragm. The defect may be congenital or acquired (eg, as a consequence of blunt or, less often, penetrating trauma). Congenital diaphragmatic hernia (CDH) is considerably more common than acquired diaphragmatic hernia (ADH) and is the primary focus of this article.
CDH is often diagnosed on 20-week screening antenatal ultrasonography (US).[1] Left-side defects account for 80-85% of cases and right-side defects for 10-15%; bilateral defects are rare, accounting for only 1-2% of cases. The majority (90%) of defects are posterolateral and are referred to as Bochdalek hernias. The remaining 10% are anteromedial and are referred to as Morgagni hernias.[2, 3] Because CDH compresses the lung during a critical time of lung development, infants with this condition are at increased risk for the development of pulmonary hypoplasia and pulmonary hypertension.
CDH was first described in 1679 in a postmortem examination of a 24-year-old man.[4] Surgical repair of a CDH was first attempted in 1888 in a 19-year-old who presented with respiratory distress and an acute abdomen; findings from clinical examination prompted a laparotomy that revealed the diaphragmatic defect. The first successful repair was performed in 1905 in a 9-year-old male; after reduction of the herniated content, the diaphragmatic defect was closed through a midline laparotomy incision. Approximately two decades later, the first outcomes associated with CDH were reported, with a 58% mortality among patients who underwent surgical intervention.
It was not until 1940 that Ladd and Gross based their diagnosis of CDH on history, physical examination, and chest radiographic findings.[4] They advocated early surgical intervention (≤ 48 hr after diagnosis). Gross also described a two-stage closure of the abdominal wall in challenging cases, with closure of the skin and subcutaneous fascia in the initial surgery and closure of the abdominal wall 5-6 days later. In 1950, Koop and Johnson suggested the transthoracic approach as a means of closing the defect under more direct vision.
As surgical expertise has improved, innovative strategies have been developed to address large diaphragmatic defects and agenesis of the hemidiaphragm. These techniques have included the use of rotational muscle flaps, perirenal fascia, and synthetic patch repairs.
Over the past few decades, improved understanding of the cardiopulmonary sequelae of CDH has shaped the management of this complex patient population, and this development has resulted in better outcomes. CDH is no longer considered a primarily surgical disease but, rather, a disease associated with pulmonary hypoplasia, pulmonary hypertension, and an increased susceptibility to ventilation-induced lung injury.
Contemporary management of CDH emphasizes the importance of the management of pulmonary hypoplasia and persistent pulmonary hypertension. Various gentle alveolar recruitment strategies are employed, and a nonurgent approach is taken to operative treatment, once therapies for pulmonary hypertension have been initiated. Urgent surgical repair is almost never necessary and may induce a pulmonary hypertension crisis.
The diaphragm is a musculotendinous structure that plays a crucial role in pulmonary mechanics. It consists of a continuous convex muscle surrounding a central tendon, with peripheral attachments that include the xiphoid process, ribs, costal cartilage, and vertebral bodies.
The diaphragm functions to regulate the thoracic volume by promoting pressure changes as it contracts and relaxes and is particularly influential with respect to forced vital capacity (FVC) and maximal inspiratory pressure (MIP). When the diaphragm contracts, it generates negative thoracic pressure, drawing air into the lungs; this ultimately results in gas exchange at the alveolar-capillary beds. Conversely, as it relaxes, pressure increases, promoting expulsion of carbon dioxide. In patients with CDH, the anatomic defect results in disruption of these physiologic processes.
During embryology, the earliest diaphragm precursor is seen at week 4 of gestation, and it is derived from four embryonic structures: septum transversum, pleuroperitoneal membranes, mesoderm of the body wall, and esophageal mesenchyme. The septum transversum is located ventrally and eventually develops into the central tendon. Pleuroperitoneal membranes are dorsolateral, and crura of the esophageal mesentery are dorsal.[5] By week 8 of gestation, pleural and peritoneal cavities are separated. Muscularization begins with migration of phrenic nerve axons and recruitment of myoblasts to form a mature diaphragm.[5, 6]
Bochdalek hernias are thought to be caused by an embryologic failure in the fusion of the pleuroperitoneal folds and the transverse septum with the intercostal muscles. Morgagni hernias, on the other hand, are believed to be caused by a defect in the union of the transverse septum and lateral body wall.[2, 3]
Most authors have postulated that CDH develops from failure of diaphragm muscularization prior to closure of the pleuroperitoneal canals, resulting in diaphragmatic weakness prone to herniation. Others have suggested that abnormal lung development results in a weakened mesenchymal plate with impaired diaphragm fusion. Disruptions or mutations in myofibroblasts derived from pleuroperitoneal folds may also play a role, ultimately causing abnormal diaphragm development.[6, 7, 8] Further research into the pathogenesis is needed to facilitate the potential development of in-utero therapies targeted at failed embyogenesis.
CDH occurs early in fetal development as a consequence of incomplete fusion of diaphragmatic structures leading to herniation of intra-abdominal organs into the thoracic cavity. Although infants born with CDH suffer from sequelae related to multiple organs, the leading causes of morbidity and mortality in these patients are the cardiopulmonary consequences of the condition—specifically, pulmonary hypoplasia and pulmonary hypertension.
The herniation of intra-abdominal organs during critical stages of lung development results in significantly reduced bronchial branching, alveolar surface area, and pulmonary vascularization—hallmark pathologic findings in pulmonary hypoplasia. Whereas normal airway development results in approximately 23-25 branched divisions, individuals affected by CDH often only have 12-14 divisions within the ipsilateral lung and 16-18 within the contralateral lung.[9, 10] This disruption of bronchial branching is associated with impaired alveolar development and pulmonary hypoplasia.[11]
Abnormal vascular smooth-muscle development is another hallmark feature associated with CDH. Specifically, pulmonary arterial smooth-muscle hypertrophy results in increased pulmonary artery resistance and consequently in pulmonary hypertension.[11, 12] Pulmonary hypertension may lead to sustained fetal circulation postnatally, with right-to-left shunting, progressive hypoxemia, hypercarbia, and acidosis. There is an association between the severity of pulmonary hypertension and the morbidity and mortality of CDH, with high mortality reported among those with persistent suprasystemic pulmonary artery pressures beyond 21 days of life.[13, 14]
In ADH, the pathophysiology includes circulatory and respiratory depression secondary to impaired diaphragmatic function, intrathoracic presence of abdominal contents leading to compression of the lungs, shifting of the mediastinum, and cardiac compromise.[15] If the hernia is relatively small, it may not be identified until months or years after the initial insult, when the patient presents with dyspnea, strangulated intra-abdominal organs, or nonspecific gastrointestinal (GI) complaints.
The etiology of CDH is largely unknown but is probably multifactorial. However, there are several reported maternal risk factors for CDH have been reported, including the following[16] :
Multiple studies are currently investigating genetic factors leading to CDH; there is a known association between CDH and certain chromosomal abnormalities. An estimated 60% of CDH cases are isolated (ie, occur in healthy term pregnancies), whereas 40% are nonisolated (ie, occur in infants with an associated chromosomal or other structural anomaly).[17] Disorders commonly associated with CDH include the following[18] :
Inheritance of CDH is also poorly understood, with familial occurrence of CDH estimated to be about 2%.[19]
Abnormalities in the retinoid signaling pathway have been hypothesized to contribute to the etiology.[20, 21] The earliest link was observed when 25-40% of the offspring of rat dams fed a vitamin A–deficient diet developed CDH, and the proportion of affected pups decreased when vitamin A was reintroduced into the diet in midgestation.[22] Subsequently, vitamin A was found to decrease the incidence and severity of CDH caused by exposure to the herbicide nitrofen in utero.[23] Two clinical studies demonstrated that plasma retinol and retinol-binding protein levels in cord blood were significantly lower in newborns with CDH than in control subjects, independent of maternal retinol levels.[24, 23]
To achieve a better understanding of the underlying etiology of CDH, further research will be required.
ADH is most commonly caused by blunt or (less frequently) penetrating trauma. Motor vehicle accidents (MVAs) are the leading cause of blunt diaphragmatic injuries, whereas gunshot or stab wounds are the leading cause of penetrating diaphragmatic injuries. Rarer causes of traumatic diaphragmatic rupture include the following:
These rare cases occur more frequently on the left side than on the right (68.5% vs 24.2%), presumably because of the liver protection and increased strength afforded by the right hemidiaphragm.
CDH has been estimated to occur in 1 in 5000 live births in the United States[1] and approximately 2.3 per 10,000 live births globally. Prevalence varies across geographic regions, and estimates may include both isolated and nonisolated cases.[30, 31] It is likely that prevalence is underestimated, given that stillbirth and pregnancy termination are often excluded. Another challenge to obtaining accurate figures is variation in antenatal detection, which has dramatically improved over time but remains dependent on geographical location. Such challenges are likely to result in underreporting or missing data in epidemiologic studies.[10]
ADH is relatively uncommon. Fewer than 1% of patients who sustain trauma have an associated diaphragmatic injury.[32] Most cases occur in the third decade of life. The male-to-female ratio is 4:1.
Historically, mortality for those with a CDH has been in the range of 25-35%, and even higher figures have been reported when termination of pregnancy and fetal loss in late gestation (~8%) were included. With advances in neonatal care, mortality for CDH has decreased substantially, falling to 15-20% in most centers[33] and to 5-10% in higher-functioning centers. This decrease is probably due to better ventilation practices, with reduced ventilator-induced lung injury, and to better control of postnatal infection, especially central line–associated bloodstream infection (BSI) and ventilator-associated pneumonia.[34]
In all studies, the only reliable predictor of mortality has been the presence of associated anomalies (see Etiology), which increases mortality to the range of 70-85%. Mortality in infants requiring extracorporeal membrane oxygenation (ECMO) remains around 50% and has stagnated at this level for the past two decades.[35]
In 46-97% of cases, congenital diaphragmatic hernia (CDH) is detected in the antenatal period before week 25 of gestation, depending on the use of level II ultrasonography (US) techniques. Assessment of the severity of the defect requires the following:
Through high-resolution US and fetal magnetic resonance imaging (MRI), it is possible to estimate the extent of lung underdevelopment and to check for additional fetal anomalies. A comprehensive evaluation also includes fetal echocardiography to assess potential cardiac anomalies.
Liver herniation into the thoracic cavity is a key predictor of CDH severity.[36, 37, 38] A 2010 meta-analysis showed that liver herniation was associated with a 45% survival rate and an 80% rate of ECMO use, compared with a 74% survival rate and a 25% rate of ECMO use without herniation.[38] Data from a subsequent study indicated survival rates of 45% for "liver-up" CDH and 94% for "liver-down" CDH.[39] Whereas US can detect liver herniation, fetal MRI can quantify it, and patients with greater than 20% herniation have an elevated risk of chronic lung disease.[40]
The lung-to-head ratio (LHR), which is calculated by dividing the contralateral lung area at the four-chamber heart view by the head circumference, is an established measure for estimating lung volume in infants with CDH. Contemporary studies have suggested an LHR cutoff value of 1.0 for poor prognosis.[41, 42] Initially, LHR calculations did not consider gestational age, but studies have since documented significant gestational changes in lung and head growth, leading to the use of the observed-to-expected LHR (O/E LHR) to account for gestational age.[43, 44, 45]
In left-side CDH, O/E LHR classifications range from extreme (< 15%) to mild (36-45%), and survival rates correlate with these classifications.[46] Infants with an O/E LHR lower than 25% are now considered candidates for fetal intervention, such as fetoscopic endoluminal tracheal occlusion (FETO), and additional trials have been exploring the benefits of intervention in moderate CDH cases (O/E LHR 26-45%) as well.[46] In right-side CDH, limited data suggest a high mortality (>90%) for an O/E LHR lower than 45%, warranting consideration of FETO.[47]
Fetal MRI has become a valuable adjunct for assessing the severity of lung hypoplasia in CDH.[39, 48, 49] Key parameters include the following:
PPLV, calculated by subtracting mediastinal volume from total thoracic volume, is predictive of outcomes: A PPLV lower than 15% is correlated with a 100% rate of ECMO requirement, longer hospital stays, and higher mortality.[50]
Late-gestation TLV, measured around 34 weeks, also predicts outcomes: Patients who have a TLV greater than 40 mL show improved survival and reduced ECMO needs as compared with those who have a TLV of less than 20 mL.[51]
O/E TLV, derived from Rypens’ method, predicts survival and ECMO need with an accuracy of 76-77%; when it is combined with percentage of liver herniation (%LH), its predictive accuracy improves to 80-83%.[40] Some studies have suggested that O/E TLV is a better predictor than O/E LHR,[52] ; however, other studies have found PPLV or O/E LHR alone to be sufficient for prognosis.[53, 54]
History and clinical findings often vary according to the severity of disease. In an infant who presents in the neonatal period without an antenatal diagnosis of CDH, variable respiratory distress and cyanosis, feeding intolerance, and tachycardia are often evident. On physical examination, the abdomen may be scaphoid if significant visceral herniation is present (see the image below). Upon auscultation, breath sounds are diminished in the chest, and heart sounds are distant or displaced.
![]() View Image | Photograph of 1-day-old infant with congenital diaphragmatic hernia. Note scaphoid abdomen; this occurs if significant visceral herniation into chest .... |
Patients may present outside the neonatal period with respiratory symptoms, intestinal obstruction, bowel ischemia, and necrosis following volvulus. In rare cases, CDH may be found incidentally.
In cases of acquired diaphragmatic hernia (ADH), typically from blunt or penetrating trauma, clinical findings may include the following:
It should be kept in mind that diagnosis of ADH requires a high level of clinical suspicion. These lesions are often overlooked even on cross-sectional imaging. (See Workup.) Missed diagnoses of ADH are associated with significant morbidity and mortality. Occasionally, the diagnosis of ADH is made at the time of laparotomy or thoracotomy performed to address other traumatic injuries. In cases where an ADH is strongly suspected, diagnostic laparoscopy is the most definitive means for diagnosis.
Although mortality associated with CDH has decreased over the past few decades, morbidity remains a significant problem. The most important adverse is the development of long-term neurodevelopmental handicaps. CDH has been associated with significant impairments in widespread neurodevelopmental domains, including the following:
Both neurocognitive and neuromotor scores are in the low-normal range. Along with these defects, specific learning disability, attention deficit/hyperactivity disorder (ADHD), problems with executive function, cognitive and attentional weaknesses, and social difficulties are more common in patients with CDH.[55]
Many studies have implicated patient-related factors as strong predictors of impaired neurologic function in CDH survivors, including the following:
Although the severity of CDH and the degree of illness are strongly correlated with adverse neurodevelopmental outcomes, there is increasing evidence that central nervous system (CNS) development may be abnormal in children with CDH. Reports have demonstrated a high incidence of cerebral abnormalities found on MRI in newborns with CDH, 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 extracranial collateral vessels.[56]
Newborns with CDH are also exposed to potential 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. Evidence suggests that the fetal and neonatal systematic inflammatory response is a key mechanism of white-matter injury. Exposure of the brain to certain drugs used in pediatric anesthesia (eg, barbiturates, benzodiazepine) or intensive care medicine (eg, muscle relaxants) and increased cytokine release during the systemic inflammatory response may cause widespread apoptotic neurodegeneration, decreased oligodendrocyte myelination, and increased astrocytosis.
Sensorineural hearing loss has been described in a number of case series of CDH survivors and seems to occur in infants regardless of whether they were treated with ECMO (though the incidence is higher with ECMO treatment).[57] The cause remains unknown but may be related to treatments for respiratory failure (such as hyperventilation, ototoxic medications, or neuromuscular blockade).[58] Severe hypoxemia, prolonged ventilation, and ECMO are risk factors. Approximately half of infants with initially normal hearing assessments develop hearing loss later in infancy.[59]
Along with adverse neurodevelopmental outcomes, CDHs can lead to significant respiratory morbidity. Survivors of CDH are at increased risk for viral respiratory infections and repeated hospitalizations in the first year of life. The prevalence of recurrent respiratory infections (>3 episodes/y) has been reported to be as high as 25-50% in the first year of life. For this reason, vaccination against influenza and respiratory syncytial virus (RSV) is strongly recommended for all survivors of CDH.
CDH survivors commonly exhibit failure to thrive, with significantly below-average weight and height. This may be secondary to severe gastroesophageal reflux disease (GERD), inadequate oral intake, or recurrent infections. In one clinical series, more than 50% of infants with CDH weighed below the 25th percentile.[60] Gastrostomy tube placement was performed in 33% of infants in this series. Another report demonstrated that more than 40% of CDH survivors weighed below the fifth percentile at age 2 years.[61]
Chest-wall deformities and scoliosis occur more frequently in infants with CDH, especially after patch or muscle flap repair, and patients should therefore be screened for such disorders.
Recurrent hernias have been reported in 8-20% of patients with CDH. The single most important predictor of hernia recurrence is the presence of a large defect that requires a patch or muscle flap to repair. Recurrences may present from months to years after the initial hospitalization, or the patient may remain asymptomatic. In some instances, recurrences may be discovered incidentally on chest radiographs performed for surveillance or other reasons. The lifetime risk of recurrence for a patient with a patch or muscle flap repair has not been established.
Antenatal studies to be considered include the following:
Postnatal studies to be considered include the following:
The Score for Neonatal Acute Physiology (SNAP)-II has been suggested as a useful means of assessing the risk of mortality and the need for extracorporeal membrane oxygenation (ECMO) therapy in neonates with CDH.[63] This score is determined on the basis of arterial blood pressure, pH, ratio of arterial oxygen tension (PaO2) to fraction of inspired oxygen (FiO2), body temperature, diuresis, and seizure activity.
The oxygenation index (OI) is another useful means of assessing the need for ECMO. It is calculated by using the FiO2 (reported as a decimal), the PaO2 (reported in cm H2O), and the mean airway pressure (MAP; reported in mm Hg), as follows:
The OI is used to categorize the predicted degree of oxygen impairment, as follows:
Many centers use an OI greater than 40 as an indication for ECMO in neonates with CDH.
An early chest radiograph is obtained to confirm the diagnosis of CDH. Findings include loops of bowel in the chest, a mediastinal shift, a paucity of bowel gas in the abdomen, and the presence of the tip of a nasogastric tube in the thoracic stomach (see the image below). Repeated chest radiography may reveal a change in the intrathoracic gas pattern. Right-side lesions are difficult to differentiate from diaphragmatic eventration and lobar consolidation.
![]() View Image | Radiograph of infant with congenital diaphragmatic hernia. Note shift of mediastinum to right, air-filled bowel in left chest, and position of orogast.... |
Chest radiography is a standard component of the Advanced Trauma Life Support (ATLS) protocol for a trauma workup. In 23-73% of cases of traumatic diaphragmatic rupture, the injury is identified on the initial radiograph; in an additional 25%, it is found on subsequent films.[64] Findings indicative of traumatic diaphragmatic rupture on chest radiography include the following:
![]() View Image | Acquired diaphragmatic hernia. Preoperative chest radiograph in 53-year-old woman who was restrained passenger in automobile accident. Note bowel cont.... |
![]() View Image | Acquired diaphragmatic hernia. Postoperative chest radiograph in 53-year-old woman who was restrained passenger in automobile accident. |
Level III ultrasonography (US) and echocardiography should accompany a diagnosis of CDH. Antenatal echocardiography may identify cardiac anomalies (more commonly, ventricular hypoplasia, atrial septal defects, and ventricular septal defects).[65] Decreased left ventricular mass, poor ventricular contractility, pulmonary and tricuspid valve regurgitation, and right-to-left shunting may be seen. Repeated echocardiography is recommended to measure changes in pulmonary artery pressure, left-to-right shunt, and flow across the ductus arteriosus.
In a study involving 18 neonates with CDH, Tanaka et al found that using M-mode imaging to measure diastolic wall strain was a useful method for evaluating the diastolic function of CDH patients.[66]
US—in particular, focused assessment with sonography for trauma (FAST)—has been reported as a means of detecting an acquired diaphragmatic hernia (ADH).[67] During visualization of each upper quadrant, the movement of the diaphragm can be seen to be decreased in patients with ADHs. Application of this technique is limited in patients who are on mechanical ventilation because of the positive pressure of the thoracic cavity.[67]
Helical (spiral) computed tomography (CT) has a sensitivity of approximately 71-100% for traumatic diaphragmatic injury; sensitivity is greater on the left side than on the right.[64] Findings indicative of an acquired diaphragmatic injury on CT include the following:
The timing of congenital diaphragmatic hernia (CDH) repair remains controversial, with no consensus on the optimal approach. Guidelines recommend balancing the need for stability prior to repair against the risks posed by prolonged central venous access and enteral starvation. Some centers advocate waiting until pulmonary hypertension resolves before providing surgical treatment,[68] but this can take months, increasing the risks of infection and sepsis. Thus, an ideal approach involves stabilizing the infant while minimizing these associated risks. (See Timing of Surgical Repair, below.)
After repair, pneumothorax will typically be present, but drainage and chest-tube placement are not required unless tension physiology develops. Because of the lung hypoplasia and the associated mediastinal shift, it may prove difficult to determine whether the pneumothorax is under tension. The mediastinum should shift back to the midline once the repair is completed, and within 24 hours after repair, the air-filled space often fills with fluid. Failure to fill with fluid should raise concern for pneumothorax and the possible need for chest-tube placement.
If the situation is uncertain, a butterfly needle may be advanced into the chest, with the back end placed to water seal in a sterile container. If continuous bubbling is seen, an air leak is probably present, and thoracostomy tube placement is indicated. The tube should be placed to water seal, not suction, because excessive suction could pull the entire mediastinum over, creating a tension effect and hemodynamic instability due to the potential space created by the hypoplastic lung on the side of the defect. In repairs done with extracorporeal membrane oxygenation (ECMO), chest-tube placement is recommended to monitor for bleeding and allow intervention before clot formation in the thoracic cavity.
Several considerations must be taken into account in determining the treatment approach, as listed below.
Historically, pulmonary hypoplasia and hypertension have been implicated as contributors to the abnormal transition from intrauterine to extrauterine life. Subsequently, increased attention has been given to the contribution of cardiac dysfunction to cardiopulmonary failure during the neonatal transition—specifically, failure of the left ventricle. Whereas pulmonary hypoplasia and hypertension contribute to hypoxemia and abnormalities in gas exchange after birth,[69] failure of the left and right ventricles to compensate for the significant afterload following birth results in cardiac dysfunction, with biventricular dysfunction the biggest contributor to the need for ECMO.[70]
Autopsy studies from infants with CDH highlight the importance of lung-protective ventilation, with all patients demonstrating significant lung injury post mortem.[71] How best to protect the lungs from injury remains controversial.
The VICI trial compared high-frequency oscillatory ventilation (HFOV) with conventional ventilation during the initial stabilization period.[72] Although no differences were seen in mortality or the development of bronchopulmonary dysplasia, patients who initially underwent conventional mechanical ventilation were ventilated for fewer days, required vasoactive agents for shorter periods, had fewer treatment failures, and less frequently needed ECMO, inhaled nitric oxide (iNO), or intravenous (IV) sildenafil. In this report, conventional ventilation was performed by emplying a pressure mode, with lower positive end-expiratory pressure (PEEP) and ventilator rates of 40-60 breaths/min.
Because the optimal tidal volume to target in the initial stabilization is unknown, volume-targeted ventilation was less commonly adopted early on. As a result, no randomized trials have yet been performed to determine the optimal tidal volume targets in infants with pulmonary hypoplasia secondary to CDH.
Hemodynamic support during the neonatal transition is complicated by the presence of left ventricular, right ventricular, or biventricular dysfunction. Although in a number of other neonatal disease processes, dopamine is considered the first vasopressor of choice, there continues to be concern about the use of this agent in CDH—specifically, because of the detrimental effects on the pulmonary vasculature. In pathologic states, the increase in pulmonary vascular resistance (PVR) with dopamine use exceeds that in systemic vascular resistance (SVR), raising concern regarding the use of dopamine in the setting of CDH and pulmonary hypertension.[73]
Because of the predominantly alpha effects, norepinephrine may be preferred to epinephrine for hypotension in the setting of normal cardiac function.[74] Agents such as vasopressin have also been explored in patients with CDH. Small case series have demonstrated efficacy, as well as a vasodilatory effect on the pulmonary vasculature.[75, 76]
Historically, sedation with paralysis was standard in the management of CDH, with some centers administering paralytics intramuscularly (IM) during delivery. This was partially to reduce air entrainment into the intestines but also to assist with the transition and facilitate gas exchange. A study examining the effect of paralysis on lung compliance found that paralytic use in the delivery room reduced lung compliance, tidal volume, and minute ventilation.[77]
Because spontaneous minute ventilation is integral to initial stabilization, use of paralytics is no longer recommended. Patient agitation can increase PVR and exacerbate pulmonary hypertension. Newer agents (eg, dexmedetomidine) that have minimal effect on spontaneous respiratory drive may be preferred to narcotics for sedation.
As noted, the optimal timing of CDH repair has not yet been definitively established. Ideally, the defect should be repaired when the infant is stable. However, this approach also means prolonged exposure to central venous access and extended periods of enteral starvation, which may have their own risks and complications. The optimal approach is to find a balance between repairing the defect when the infant is stable and minimizing prolonged exposure to central venous access and enteral starvation.
Retrospective studies, though limited by selection bias, have suggested that for infants who do not require ECMO, early repair (≤ 48 hr) may lead to fewer ventilator days, reduced supplemental oxygen needs, and shorter hospital stays.[78, 79, 80, 81] For neonates who do require ECMO, early repair (≤ 48 hr post cannulation) has also been associated with better outcomes.[82, 83, 84]
Thus, if an infant can be stabilized without ECMO, and hemodynamics and gas exchange are stable, then repair is likely to be safe even if pulmonary hypertension has not resolved. If the patient requires ECMO therapy, early repair is recommended and is associated with improved outcomes.[85]
Although repair of the diaphragm defect is unlikely to improve pulmonary mechanics and pulmonary hypertension, CDH repair allows for initiation of enteral feeding and removal of central venous catheters and respiratory support.
Delivery room management begins with intubation as soon as possible after the first breath to prevent air from entering the stomach and bowel during the infant's respiratory efforts. In CDH patients, the carina is often positioned higher than it is in healthy neonates, necessitating careful monitoring of the endotracheal (ET) tube placement to avoid right mainstem intubation.[86]
In cases of mild CDH, a spontaneous breathing approach (SBA) has shown promise; a study of eight patients indicated that those who underwent SBA had shorter hospital stays, less mechanical ventilation, and reduced oxygen needs in comparison with matched control subjects, with earlier tolerance of enteral feeding.[87] Larger studies are ongoing to evaluate safety and replicate these findings while determining optimal respiratory support.
Once the airway is secured, an orogastric (OG) tube should be placed slightly deeper than the standard measurement to optimize drainage. Positive-pressure ventilation (PPV) should be initiated, preferably using a T-piece resuscitator, with close attention paid to peak inspiratory pressure (PIP) so as to mitigate the risk of barotrauma due to hypoplastic lungs.
There are no established guidelines specifying the fraction of inspired oxygen (FiO2) in the resuscitation of these infants; however, animal studies have suggested that a high FiO2 may blunt later responses to pulmonary vasodilators in cases of persistent pulmonary hypertension of the newborn (PPHN).[88] Acute hypoxia can cause vasoconstriction in an already underdeveloped pulmonary vascular bed, warranting close monitoring of preductal saturations and careful titration of inspired oxygen.
The literature has suggested a PIP of 25 cm H2O or less for resuscitating infants, with strong recommendations to avoid exceeding 30 cm H2O except in an emergency (eg, bradycardia unresponsive to resuscitation). A high respiratory rate of 60-80 breaths/min is recommended in view of the common issues with ventilation and hypercapnic respiratory failure.[69] Goal saturations in the delivery room range from 75% to 85% and are usually rapidly achieved, whereas hypercarbia takes longer to resolve.
Once goal saturations are met, time should be allowed for improvement in hypercarbia before respiratory support is increased. It is crucial to aim for a pH greater than 7.25 and an arterial carbon dioxide tension (PaCO2) lower than 65 mm Hg, given that acidosis exacerbates pulmonary vasoconstriction. PEEP should be maintained in the range of 2-5 cm H2O, a range that is associated with improved respiratory compliance, tidal volume, and pulmonary blood flow.[89] Rapid onset of metabolic acidosis complicates the transition, potentially worsening acidemia and cardiac dysfunction.[69] Sodium bicarbonate can be administered to buffer acidosis while gas exchange improves.
Hypotension is common in infants with CDH; it is typically related to cardiac dysfunction, though multiple factors may contribute. Only limited guidance is available for managing hemodynamics in CDH, with initial stabilization strategies varying according to whether dysfunction involves the left ventricle, the right, or both. In cases of left ventricular dysfunction, agents such as milrinone and low-dose epinephrine may be beneficial. For right ventricular dysfunction, maintaining ductus arteriosus patency, potentially with the use of iNO, may be helpful. In instances of biventricular dysfunction, ECMO may be the optimal therapeutic approach.[90]
The use of iNO in CDH remains controversial and has not been approved by the US Food and Drug Administration (FDA), because of inconclusive randomized trial results. Nevertheless, its application has increased from 30% to 80% over the past decade, and this increase is correlated with improved survival rates.[91] Selective use of iNO may benefit a subset of patients, particularly if instituted early in the course (48-72 hr post birth) after recovery of left ventricular function.[92, 93]
Vasopressor agents are often required to enhance systemic blood pressure and tissue perfusion, especially when iNO fails to reduce PVR. When SVR exceeds PVR, pulmonary blood flow and oxygenation improve. However, excessive afterload from left ventricular systolic and diastolic dysfunction can reduce tissue perfusion and worsen left ventricular function, making ECMO crucial for left ventricular dysfunction in CDH. The use of dopamine raises concerns due to its adverse effects on pulmonary vasculature, where the increase in PVR can exceed that of SVR.[73]
In cases of catecholamine-resistant shock, vasopressin may be more effective, inducing vasoconstriction through V1, V2, and V3 receptor activation. Case reports have found that vasopressin improves tissue perfusion, pulmonary blood flow, and outcomes in CDH. There is, however, a risk of water retention with vasopressin use, and monitoring serum sodium is therefore essential.
Adrenal insufficiency is another potential cause of hemodynamic compromise, often presenting without hypoglycemia, hyponatremia, or hyperkalemia. Diagnosis involves measuring random cortisol levels and administering adrenocorticotropic hormone (ACTH) to assess cortisol response. Hydrocortisone may improve hemodynamics, though the indications for treatment remain unclear.
Patient agitation can increase PVR, leading to pulmonary hypertensive crises. Accordingly, sedation is crucial for initial stabilization. Minimal stimulation techniques (eg, maintaining the infant in an isolette and using earmuffs to minimize noise) can effectively reduce stress. In some cases, medications may be required. Narcotics, often combined with paralytics, have traditionally been used for sedation; however, because spontaneous minute ventilation is critical for stabilization and paralytics can reduce lung compliance, their use is no longer recommended. Newer sedatives, such as dexmedetomidine, may be preferable by virtue of their minimal effects on spontaneous respiratory drive.
Postdelivery care is complex and requires close monitoring of respiratory status, gas exchange, pulmonary hypertension, and hemodynamics (including cardiac performance). Continuous monitoring should include a cardiorespiratory monitor, pre- and postductal saturation measurements, and frequent noninvasive blood pressure readings until invasive arterial access is established. Umbilical arterial and venous access should be secured promptly, with continuous blood pressure transduction once arterial access is obtained. Umbilical venous access may be challenging if the liver is mispositioned; a low-lying umbilical venous line can be a temporary solution until central venous access is achieved.
Monitoring respiratory status and gas exchange in infants with CDH is made particularly challenging by lung hypoplasia, immaturity, and pulmonary hypertension, all of which influence gas exchange. Achieving optimal lung inflation (functional residual capacity [FRC]) can be difficult, and both under- and overinflation can worsen pulmonary hypertension. Eight-rib inflation is thought to be ideal in CDH, but mediastinal shifts may obscure visualization of the contralateral lung, complicating inflation assessment. Often, only a small triangle at the costophrenic angle of the contralateral lung is visible. Overdistention can shift the mediastinum back to midline and flatten the contralateral diaphragm.
The optimal tidal volume for ventilation is unknown. Initially, pressure ventilation with a PIP lower than 25 cm H2O is recommended. Monitoring exhaled tidal volumes is essential, and volumes exceeding 5 mL/kg should prompt a decrease in PIP. Lower levels of PEEP (2-5 cm H2O) are favored over higher levels, in that they correlate with improved respiratory compliance, tidal volume, and pulmonary blood flow.[89] High respiratory rates of 60-80 breaths/min may improve outcomes.
One study found that using high-rate PPV (HPPV) before HFOV was associated with higher rates of survival to discharge (80% vs 50%) and surgical intervention (95.6% vs. 68.8%), with surgery occurring an average of 2 days earlier.[94] This study employed pressure-controlled ventilation with a PIP of 20 cm H2O, 100 breaths/min, an inspiratory time of 0.3 seconds, and low PEEP (1-1.5 cm H2O). Low PEEP is crucial for preventing auto-PEEP, which can elevate measured PEEP beyond set values. The role of HFOV in initial stabilization is debated, with conflicting results reported on mortality, ECMO necessity, and chronic pulmonary morbidity.[95, 96, 97, 98, 99, 100]
The VICI trial, designed to determine the optimal initial ventilation mode in CDH, compared conventional mechanical ventilation (PEEP 2-5 cm H2O; 40-60 breaths/min) with HFOV in 171 patients with antenatally diagnosed CDH.[72] Although mortality and bronchopulmonary dysplasia (BPD) rates did not differ, conventional ventilation resulted in fewer ventilation days, decreased need for ECMO and iNO, and shorter durations of vasoactive drug administration.
Another study demonstrated that utilizing HFOV with a lower mean airway pressure (MAP) of 10-12 cm H2O as compared with 13-15 cm H2O led to decreased ECMO use (from 37% to 13%), increased survival without ECMO (from 53% to 79%), and improved overall survival (from 74% to 89%).
After the initial stabilization period, these infants often need continued intensive respiratory support and careful monitoring for pulmonary hypertension and hemodynamic stability. Transition to ECMO might be required for infants with severe pulmonary hypertension or those who do not stabilize on conventional respiratory support. ECMO provides temporary respiratory and circulatory support, giving the lungs time to mature before the transition back to ventilator support. (See the image below.) The decision to use ECMO should involve a multidisciplinary team and is often influenced by factors like the oxygenation index (OI), pH, and hemodynamic stability.
![]() View Image | Newborn with congenital diaphragmatic hernia on venoarterial extracorporeal membrane oxygenation (ECMO). Note arterial and venous cannulas connected t.... |
In addition to respiratory management, other supportive care should include thermoregulation, correction of fluid and electrolyte imbalances, and nutritional support tailored to minimize metabolic demand. Monitoring for sepsis is essential, in that these infants are particularly vulnerable to infection.
For infants on ECMO, a tailored approach to pharmacologic management, including pulmonary vasodilators and inotropes, is vital in optimizing pulmonary and cardiac function. These infants may also benefit from the use of structured sedation protocols to maintain stability without excessive suppression of spontaneous breathing efforts; this can aid in weaning from ECMO.
Finally, the timing for surgical repair of CDH varies according to the infant’s stabilization status. In general, the consensus is to delay surgery until the infant is stabilized both hemodynamically and in terms of respiratory support, which can sometimes take several days or even weeks after delivery.
Over the long term, these infants often require ongoing respiratory and developmental follow-up with particular attention to potential chronic lung disease, neurodevelopmental impacts, and nutritional challenges. Coordinated follow-up with pediatric pulmonologists, nutritionists, and developmental specialists can be instrumental in optimizing outcomes for CDH survivors.
Surgical repair can often be safely delayed in stable patients, and the operation can be scheduled on a semielective basis. Urgent surgical repair is almost never necessary and may induce a pulmonary hypertension crisis.
The priorities in preoperative care are to provide appropriate ventilatory management of the newborn and to determine whether the patient has any other associated congenital anomalies, particularly cardiac abnormalities. Echocardiography should always be performed before surgical repair, and pulmonary hypertension should be well controlled before the patient is subjected to the stress of surgery.
CDH is typically repaired either through an open abdominal incision or thoracoscopically. In open repairs, a subcostal or supraumbilical transverse incision is made. The abdominal viscera are examined, and the herniated content is reduced by gentle traction. A hernia sac is sought and, if found, excised.
With right-side defects, it is occasionally necessary to exteriorize the right lobe of the liver to permit visualization of the native diaphragm. Care must be taken to avoid disruption of the hepatic veins, which can result in hepatic congestion and reduced preload. With both left-side and right-side defects, after careful dissection of the posterior leaf of the diaphragm, primary repair can be accomplished with nonabsorbable sutures. If the diaphragmatic defect is large enough to preclude primary closure, a prosthetic patch or a transversus abdominis muscle flap is used. Open transthoracic repair of a left-side and right-side diaphragmatic hernia has been reported; however, this approach is not commonly used.
In open cases, if abdominal closure is not feasible because of interference with the chest wall, potential effect on diaphragmatic compliance, or concern that it will lead to abdominal compartment syndrome, then a temporary silo with delayed primary closure of the fascia or skin can be safely accomplished.
The use of chest tubes is controversial. The authors prefer to use a chest tube when the repair is performed with the patient on ECMO so as to minimize the pleural effusions that can easily accumulate while the lung is expanding. For repairs performed without ECMO, chest tubes are not routinely used.
Thoracoscopic repair was initially described for late presenters and neonates requiring minimal ventilator support. Today, there is a tremendous amount of institutional variability, with some centers offering thoracoscopic repair on HFOV.[101] Thoracoscopic repair on ECMO has been described, but this approach is not considered routine practice. Exclusion criteria for thoracoscopic repair are not well defined; however, an intrathoracic liver or stomach, inability to tolerate a period of manual ventilation, and large defects have all been cited as reasons for choosing initial open repair or conversion to open repair.
Thoracoscopic repair of CDH offers several advantages (eg, improved visibility, reduced postoperative opioid use, and shorter ventilation duration), though outcomes may vary, depending on patient selection. Studies by Gander et al[101] and Okazaki et al[102] reported recurrence rates as high as 23% in infants undergoing thoracoscopic repair in the neonatal period.
A systematic review and meta-analysis by Terui et al suggested that although minimally invasive surgery for CDH may lower mortality, it may also have a higher recurrence rate[103] ; the authors therefore recommended that it be used selectively rather than routinely in neonates. In contrast, a single-center study by Tyson et al found thoracoscopic repair to be as safe as open repair, with no hernia recurrences observed at a median follow-up of 27 months.[104]
Surgical repair with the patient on ECMO was initially associated with increases in mortality, surgical-site hemorrhage, and intracranial hemorrhage.[105] As noted previously, early repair on ECMO has been associated with improved outcomes.[85] Although repair of the diaphragmatic hernia after decannulation results in better outcomes than early or late repair on ECMO, the authors typically use antenatal lung volumes (which can be estimated via the lung-to-head ratio [LHR]) to predict whether the neonate can successfully be decannulated before repair. If the CDH is considered severe (observed-to-expected LHR [O/E LHR] < 25%), the authors favor early repair on ECMO.
Survival rates vary widely among centers, with some centers citing figures higher than 90% for severe cases and others citing figures in the area of 50% for such cases. Among survivors, long-term morbidities—including neurodevelopmental, nutritional, hearing, and pulmonary function deficits—are often attributed to the severity of lung hypoplasia and pulmonary hypertension associated with CDH.[106, 107, 108] Given the severity of these outcomes, researchers have sought ways to address CDH, or the resulting lung hypoplasia, before birth so as to provide better support of normal postnatal pulmonary function.
Early attempts at fetal CDH repair involved open fetal surgery in midgestation. In these cases, returning the liver to the abdomen could occlude blood flow through the umbilical vein and inferior vena cava, leading to fetal death.[109, 110] In cases where liver herniation was absent, however, fetuses often had sufficient lung volume to render fetal surgery unnecessary.
The evolution of fetal intervention for CDH has been shaped by two developments: the advent of minimally invasive surgical techniques[111] and insights from the study of congenital high airway obstruction syndrome (CHAOS), in which airway occlusion led to lung overexpansion. In the 1990s, advances in fetoscopic technology allowed the performance of minimally invasive procedures, while preterm labor risks highlighted the need to minimize uterine manipulation. This led to the concept of fetoscopic tracheal occlusion (FETO) to promote lung growth by occluding the trachea, enhancing lung fluid retention and expansion.[112, 113, 114]
Studies using a fetal lamb model of CDH confirmed the feasibility of FETO.[115] (See the image below.) In the initial human studies, surgical clips were used for tracheal occlusion, but adverse effects (eg, tracheal stenosis) caused the technique to be refined to include fetal bronchoscopy with a detachable balloon for occlusion.[116, 117, 118, 119, 120] Subsequently, the National Institutes of Health (NIH) supported a randomized controlled trial (RCT) comparing FETO with standard postnatal care for severe CDH.[121] FETO was not found to confer any overall survival advantage, but infants undergoing FETO were born earlier, and some, despite prematurity, had marginally improved lung function.[122]
![]() View Image | Diagram illustrating sheep model of PLUG (trachea used for fetal management of congenital diaphragmatic hernia). Image from Michael Harrison, MD. |
These results have fueled worldwide interest in FETO as an investigational therapy, especially given that overall survival rates for CDH have plateaued over the past few decades and that survivors of severe CDH face significant comorbidities.[107] However, the long-term benefits of FETO have not yet been established in controlled studies.[123, 124, 125, 126, 127, 128]
In 2021, the findings of the Tracheal Occlusion to Accelerate Lung Growth (TOTAL) trial (N = 80) were published.[129] This RCT examined the management of severe left CDH with either FETO (n = 40) or expectant care (n = 40) in 10 European centers. The survival-to-discharge rate was found to be 40% in the FETO group and 15% in the no-FETO group. One criticism of this trial is the low use of ECMO—shown to improve survival in neonates with CDH—in both groups (17% in the TOTAL trial vs ~30% in the United States).[130, 131] Another is that the survival rates for both groups in the TOTAL trial are lower than the survival rate for severe CDH in the United States (~50%).[132]
Selection of CDH cases that may benefit from FETO involves multiple assessments. Once CDH is confirmed by an experienced radiologist via ultrasonography (US) or magnetic resonance imaging (MRI), any major fetal anomalies (eg, cardiac defects or chromosomal abnormalities) and maternal health risks are carefully ruled out. If no contraindications are found, the severity of isolated CDH is assessed on the basis of antenatal factors, including the following:
An LHR of 1.0 or less is generally indicative of severe lung hypoplasia, and newer MRI-based measurements (eg, PPLV and TLV) provide more detailed insights into potential respiratory outcomes.
In the TOTAL trial, criteria for FETO included the following[129] :
Fetal positioning is critical for successful balloon deployment. External fetal manipulation may be required to line up the fetoscope for entry into the mouth of the fetus. Before initiation of FETO, an intramuscular fetal cocktail containing a paralytic, a narcotic, and atropine is administered percutaneously. Under US guidance, a 3.3-mm cannula with an inner pyramidal trocar is introduced into the amniotic cavity. A 1.3-mm fiberoptic endoscope is introduced, as well as a delivery catheter to inflate and deploy an inflatable 0.8-mL balloon. The balloon should be deployed about 1 cm above the carina.
There are multiple ways of deflating the balloon in utero. The authors prefer a percutaneous approach using US guidance through the fetal back. Another option is repeat fetoscopy with balloon puncture. As a last resort, an ex-utero intrapartum treatment (EXIT) procedure may be required, in which the fetus is maintained on placental circulation after the head is delivered via an open hysterotomy. The balloon is then deflated with the help of a rigid bronchoscope, and the fetus is delivered.
Once an acquired diaphragmatic hernia (ADH) is diagnosed, it should always be repaired, whether the patient is presenting immediately after sustaining trauma or after some delay; there are no known contraindications for ADH repair. In the acute trauma setting, the patient must be adequately resuscitated before being transported to the operating room. The high incidence of concomitant intra-abdominal injuries dictates the need for emergency abdominal exploration after initial resuscitation is accomplished. Patients who have a delayed presentation also require repair, in that the hernia content may become strangulated and ischemic.
Many small diaphragmatic injuries are discovered during exploratory laparotomy for the repair of other intra-abdominal injuries. A distinguishing feature of ADHs as compared with CDHs is that they arise as a consequence of trauma or other external factors rather than congenital maldevelopment, which generally means there is adequate diaphragm tissue available for primary repair. However, in rare cases of ADH where the diaphragmatic defect is extensive or the quality of the diaphragm tissue is compromised, a synthetic patch may be necessary to reinforce or replace part of the diaphragm.
Complications observed in the early postoperative period include recurrent pulmonary hypertension and deterioration in respiratory mechanics and gaseous exchange. Less commonly observed complications include recurrence of the CDH, which is more common with patch repair,[133] leakage of peritoneal fluid and blood into the thorax; and development of an ipsilateral hydrothorax. Small-bowel obstruction may occur secondary to adhesions or volvulus.
Continued care post discharge is recommended for survivors of CDH, to be provided by a multidisciplinary team consisting of a social worker, a nutritionist, a physiotherapist, a pediatrician/neonatologist, a neurologist, a child development specialist, a pulmonologist, a cardiologist, and a pediatric surgeon.
The following screening tests may be performed before discharge:
In 2008, the American Academy of Pediatrics (AAP) published postdischarge follow-up recommendations for infants with CDH,[134] which included scheduling of the following: