Combined extracorporeal membrane oxygenation support and patent ductus arteriosus ligation following surgical correction for congenital diaphragmatic hernia, a case report and literature review

Abstract

Our department recently achieved a successful outcome in a case of right-sided CDH. The patient required emergent surgery for right-sided diaphragmatic hernia repair due to severe hypoxemia, followed by urgent bedside venous-arterial ECMO and simultaneous bedside ECMO-assisted patent ductus arteriosus ligation, necessitated by severe hypoxemia and pulmonary hypertension. During the second day on ECMO, the patient developed intracranial hemorrhage, leading to the performance of a lateral ventriculostomy. Fortunately, the patient was successfully weaned off ECMO. This report presents an analysis of the clinical data from this case and shares insights from our experience.

1. Introduction

The prevalence of congenital diaphragmatic hernia (CDH) in newborns ranges from 1 in 2500 to 5000 births, classifying it as a significant congenital structural anomaly during the neonatal period [1]. CDH arises from a failure in diaphragmatic development during embryogenesis, leading to abdominal organs herniating into the thoracic cavity due to pressure differences between the thorax and abdomen. This disorder is frequently accompanied by underdeveloped alveoli and pulmonary vessels, contributing to pulmonary hypertension (PH), which can result in life-threatening persistent hypoxemia and/or hypercapnia [2]. Despite advancements in prenatal diagnostics, in utero interventions, neonatal intensive care, and surgical techniques, CDH's mortality rate remains high, ranging from 20 % to 30 %, primarily due to complications like pulmonary hypoplasia, cardiac dysfunction, and pulmonary hypertension [3]. However, with improved prenatal screening and the adoption of multidisciplinary treatment strategies, the overall mortality rate for neonatal CDH has shown an encouraging decline [4]. This improvement provides more infants diagnosed prenatally with CDH a chance for effective postnatal intervention. Yet, severe cases, often marked by pulmonary hypoplasia and persistent neonatal pulmonary hypertension, may necessitate extracorporeal membrane oxygenation (ECMO) [5,6]. Our department recently achieved a successful outcome in a case of right-sided CDH. This report presents an analysis of the clinical data from this case and shares insights from our experience.

2. Case presentation

The patient, a male neonate, was admitted presenting with “diaphragmatic hernia detected in utero” and “postnatal dyspnea.” He was the firstborn, delivered at 37 weeks and 1 day via cesarean section due to fetal distress. His birth weight was 2990g, and he had a sail-shaped placenta and a twisted umbilical cord. His Apgar scores were 8 at 1 minute, 9 at 5 minutes, and 9 at 10 minutes. A routine ultrasound at 24 weeks of gestation, performed at our hospital, revealed a right-sided diaphragmatic hernia, including the liver and intestines in the herniated area. The mother reported no significant medication use or exposure to harmful physical or chemical agents during pregnancy. Follow-up ultrasounds consistently showed the right-sided diaphragmatic hernia, with lung-to-head circumference ratio (LHR) measurements at 24, 26, 28, 30, 32, and 34 weeks being 1.24, 1.62, 1.33, 1.34, 1.38, and 1.38, respectively. Chromosomal karyotype analysis of umbilical cord blood at 26 weeks revealed no significant abnormalities, and CNV-seq showed no notable pathogenic variations. However, a small segmental deletion of approximately 211.37kb on chromosome 9 was noted, deemed a variation of unknown clinical significance, and inherited from the mother. Upon admission, a bedside chest X-ray showed liver and intestinal shadows in the right thoracic cavity, consistent with a right-sided diaphragmatic hernia (Fig. 1).

Fig. 1.

Chest X-ray before surgical repair of right-sided diaphragmatic hernia.

The neonate exhibited respiratory distress and cyanosis immediately following birth, necessitating tracheal intubation and mechanical ventilation. He was subsequently transferred to the neonatal intensive care unit (NICU). His treatment included high-frequency oscillatory ventilation (HFOV, RR 8HZ, FiO₂ 70 %, MAP 20 cmH₂O), inhaled nitric oxide (iNO) for pulmonary artery pressure reduction, adrenaline for cardiac support, and sedation and analgesia. Despite intensive medical interventions, the patient's oxygenation remained poor (pre-ductal and post-ductal oxygen saturation were 85 % and 79 % respectively), and he developed concurrent circulatory failure. His oxygenation index (OI) fluctuated between 31.7 and 40. Arterial blood gas readings showed a pH ranging from 7.143 to 7.239, PaO₂ from 40 to 65 mmHg, PaCO₂ from 60 to 75 mmHg, lactate at 4.6 mmol/L, and base excess (BE) between −3.92 and −5.34 mmol/L. Arterial blood pressure was 55–65/40-45 mmHg. His vasoactive-inotropic score was 30. Echocardiography revealed a patent ductus arteriosus (0.25 cm, predominantly right-to-left shunt), a patent foramen ovale with mainly bidirectional shunting, and elevated pulmonary artery systolic pressure (Fig. 2). The patient's pulmonary artery pressure continued to rise, accompanied by an increasing volume of right-to-left shunting, severe right heart failure, and left heart involvement. Despite adjustments to ventilatory settings and medication regimens, there was no significant improvement in his condition.

Fig. 2.

Pre-ECMO echocardiography revealed a right-to-left shunt across the arterial duct.

Twenty-one hours after birth, the arterial duct flow became bidirectional with a predominant left-to-right shunt, indicating a need for surgical intervention. Following discussions with a multidisciplinary team and obtaining family consent, the patient underwent right-sided thoracic diaphragmatic hernia repair surgery under general anesthesia, supported by HFOV, at 22 hours post-birth. Intraoperatively, the diaphragm was found to be extremely thin with a 4cm × 4cm defect in the central tendon area, covered by a hernial sac. The intestines and liver had protruded into the thoracic cavity. Upon opening the hernial sac, the liver was dark red, and the intestines, while well-perfused, were underdeveloped. Post diaphragmatic repair, the thoracic cavity noticeably increased in size. However, the lungs were underdeveloped and dark red, indicative of incomplete lung expansion and right-sided pulmonary hypoplasia. Perioperatively, the patient exhibited severe respiratory and circulatory failure, evidenced by fluctuating SPO₂ levels between 75 and 85 %, PaO₂ of 40–50 mmHg, blood pressure readings of 40–50/20–30 mmHg, and a lactate level of 15 mmol/L. Consequently, ECMO support was initiated 1 h after the surgery (24 hours post-birth).

Venous-arterial ECMO (V-A ECMO) was initiated by inserting cannulas into the right common carotid artery and right internal jugular vein. We used a Medtronic arterial cannula (size 8) and venous cannula (size 10), along with a German MEDOS800 kit equipped with a Maquet centrifugal pump. The initial ECMO flow rate was set at 0.4 L/min with a pump speed of 2500 RPM, aiming to maintain an activated clotting time (ACT) of 180–220 seconds and an activated partial thromboplastin time (APTT) of 60–80 seconds. Following the initiation of ECMO, the patient's hemodynamics and blood gases stabilized. His SPO₂ remained above 98 %, lactate levels decreased progressively, urine output was within normal limits, and liver and kidney functions were monitored and found to be normal. A head ultrasound showed no significant abnormalities. However, a repeat echocardiogram revealed a patent ductus arteriosus (0.55 cm, left-to-right shunt), mild-to-moderate tricuspid regurgitation, and elevated pulmonary artery pressure (72 mmHg). Chest radiographs showed lung perfusion injury with lactic acid rising to a maximum of 18 mmol/L. This indicated a significant left-to-right shunt at the ductal level with a high shunt volume (Fig. 3). To mitigate the risk of pulmonary edema due to the high shunt volume, bedside arterial duct ligation was successfully performed under ECMO support 4 h later (28 hours post-birth) (see Fig. 4).

Fig. 3.

Post-ECMO echocardiography showed a left-to-right shunt across the arterial duct.

Fig. 4.

Cranial ultrasound revealed intraventricular hemorrhage in the right lateral ventricle; R: right; L: left.

During ECMO therapy, heparin anticoagulation was administered, with dynamic monitoring of ACT, APTT, anti-Xa levels, platelet count (PLT), and other relevant parameters (refer to Table 1 for details). At 34 hours post-ECMO initiation (58 hours post-birth), a significant increase in arterial blood pressure was noted. Subsequent head ultrasound imaging revealed severe intracranial hemorrhage, with marked ventriculomegaly and ventricular dilation. Consequently, bilateral lateral ventriculostomy was performed at 46 hours of ECMO (70 hours post-birth). By 60 hours after ECMO initiation (84 hours post-birth), the patient's cardiac function showed improvement, hemodynamics were stabilized, and internal environment parameters were stable, which facilitated the weaning process from ECMO.

Table 1.

Anticoagulation therapy and coagulation monitoring during ECMO.

ECMO	PT (s)	APTT (s)	d-dimer (mg/L)	Fib (g/L)	PLT ( × 109/L)	ACT (s)	Heparin (U/kg.hr)
Before ECMO	13	59.8	2.6	1.63	199	–	–
Day 1	16.6 (12.1–17.6)	78.5 (67.4–115.3)	5.6 (3.6–9.4)	1.71 (1.63–2.43)	139	229 (163–251)	10 (8–12)
Day 2	17.8 (13.5–20.8)	110.9 (59.3–131.4)	15.6 (7.2–25.3)	3.08 (2.69–3.68)	76	205 (175–232)	13 (9–18)
Day 3	11.7 (9.9–15.3)	54 (49.8–86.1)	28.3 (19.5–32.3)	2.56 (2.18–3.01)	85	182 (159–220)	14 (8–20)

Abbreviation: ECMO: extracorporeal membrane oxygenation; APTT: activated partial thromboplastin time; PT: prothrombin time; Fib: fibrinogen; PLT: platelets; ACT: activated coagulation time.

Postoperative monitoring via cranial ultrasound indicated significant improvement in hydrocephalus, with no progression of intracranial hemorrhage. The neurological assessment score of 35 points suggested favorable neurological recovery. Electroencephalogram (EEG) findings demonstrated normal background brain activity, with no evidence of abnormal discharges or epileptic seizures, indicating stable cerebral function. Overall, the infant exhibited a promising neurological prognosis without severe complications.

3. Discussion

Advancements in prenatal screening, diagnosis, and prognostic factor identification have continually improved the treatment approaches for CDH. Prognostic indicators for CDH encompass a range of clinical markers assessing cardiopulmonary development and function. These include the LHR, observed/expected LHR (O/E LHR), fetal lung volume (FLV), observed/expected total fetal lung volume (O/E FLV), liver position, and chromosomal abnormalities. A critical LHR threshold of 1.0 is commonly utilized. At 24–26 weeks of gestation, an LHR >1.4 signals a favorable prognosis with a 100 % survival rate, an LHR between 0.6 and 1.4 indicates a less favorable prognosis with a survival rate of approximately 61 %, and an LHR <0.6 correlates with a 100 % mortality rate [7]. The 2018 Canadian guidelines recommend ultrasound measurements of O/E LHR between 22 and 32 weeks of gestation to predict lung underdevelopment severity in isolated CDH fetuses. These guidelines classify >45 % O/E LHR as mild, 25–45 % as moderate, 15–25 % as severe, and ≤15 % as extremely severe fetal CDH [8]. Research indicates that prenatal diagnosis before 25 weeks of gestation, right-sided defects, liver herniation, and genetic abnormalities are significant prognostic factors for CDH [9,10]. While CDH predominantly occurs on the left side, right-sided CDH (RCHD) is rarer, representing about 10–15 % of cases. RCHD poses greater diagnostic challenges prenatally, often leads to a higher rate of ECMO support postnatally, and is associated with higher mortality. Patients with RCHD generally have worse outcomes than those with left-sided hernias, even when lung volumes are comparable [11]. The primary causes of death in CDH patients are linked to associated pulmonary hypoplasia and persistent pulmonary hypertension [12,13]. The European CDH Consortium's 2015 consensus on CDH suggests planning deliveries after 39 weeks of gestation at a large tertiary center, considering lung development [14]. Over the past two decades, severe neonatal CDH has been an indication for ECMO internationally, with survival rates around 50 %, which is lower compared to other diseases treated with ECMO [5,15,16].

According to the CDH EU Consensus, the Canadian CDH Treatment Guidelines, and the 2018 Chinese Expert Consensus on Neonatal Respiratory Failure ECMO Support, ECMO indications include: (1) an OI greater than 40 for over 4 hours, where the OI is calculated as the mean airway pressure multiplied by the FiO₂ and then multiplied by 100, divided by the post-ductal arterial oxygen tension; (2) an OI exceeding 20 for more than 24 hours or ongoing worsening of respiratory distress; (3) rapid deterioration in clinical condition despite aggressive respiratory support, characterized by severe hypoxemia with PaO₂ less than 40 mmHg; (4) blood pH lower than 7.15, lactate levels of 5 mmol/L or higher, and urine output less than 0.5 mL per kilogram per hour for 12–24 hours; (5) pulmonary hypertension leading to right ventricular dysfunction, requiring continuous high-dose positive inotropes, with a vasoactive inotropic score of 40 or higher, calculated as Adrenaline times 100 plus Isoprenaline times 100 plus Milrinone times 10 plus Amrinone times 1 plus Dopamine times 1 plus Dobutamine times 1 [8,12,17].

In this case, the patient, diagnosed with a right-sided diaphragmatic hernia (involving herniated liver and intestines), was identified at 24 weeks of gestation with a LHR of 1.24 and an O/E LHR of 80 %. Following a multidisciplinary consultation at our hospital at 27 weeks gestation, a plan was made for delivery at 40 weeks. However, due to fetal distress, delivery occurred at 37 weeks and 1 day, categorizing this CDH case within the high-risk prenatal screening category. Postnatally, the patient experienced severe respiratory failure and hypoxemia, necessitating high ventilator settings. The patient had a high OI of 40, severe acidosis, and progressively increasing lactate levels. Echocardiographic findings included continuously rising pulmonary artery pressure, leading to right atrial enlargement, left atrial compression, and bidirectional shunting through the oval foramen. These conditions further exacerbated PH.

It was previously believed that the primary issues in CDH were diaphragmatic defects, abdominal organ herniation, and compression of the lungs, leading to impaired lung development and respiratory function. The key to alleviating respiratory distress was thought to be the prompt relief of lung compression to allow for lung re-expansion. Consequently, urgent surgery was commonly performed in neonates with CDH in the past [18,19]. Research indicates that over 90 % of infants not receiving ECMO undergo repair within the first two weeks of birth, with two-thirds undergoing surgery between the 2nd and 5th days. Moderate to severe CDH patients are generally recommended to have surgery within 24–48 hours after birth, following the stabilization of respiratory and circulatory functions. The European CDH organization advises surgery when mean arterial blood pressure is normal, pre-intubation oxygen saturation is between 85 % and 95 %, FiO₂ is below 0.5, lactate is less than 3 mmol·L-1·h-1, and urine output is over 1 mL·kg-1·h-1 [14]. Experts on CDH surgical and laparoscopic repair guidelines suggest considering early surgery if normal respiratory function remains difficult to maintain despite the use of vasoactive drugs and HFOV [20]. Repair of CDH can be performed earlier when HFOV and iNO are utilized. Employing these techniques may offer significant benefits, including the prompt restoration of normal anatomy and an earlier initiation of enteral feeding, while concurrently minimizing the risk of lung injury [21]. For newborns experiencing severe immediate respiratory and circulatory dysfunction unresponsive to medication, ECMO can stabilize clinical status. Thus, the timing of surgery in relation to ECMO support is a critical consideration. Debate continues over the optimal timing for surgical repair, especially in patients requiring ECMO [22]. Surgery timing categories include pre-ECMO cannulation, early surgery (within 72 hours of ECMO cannulation), late surgery (after 72 hours of ECMO cannulation), and post-ECMO decannulation surgery. Partridge et al. reported survival rates of 67 %, 43.9 %, and 100 % for surgeries conducted before ECMO, during ECMO, and after ECMO decannulation, respectively [23]. Kays et al. found that pre-ECMO cannulation surgery in left-sided hernia patients with liver herniation had a higher survival rate than delayed post-cannulation surgery. However, risks include severe respiratory distress during surgery, increased bleeding risk with early ECMO transition, and potentially heightened ECMO necessity [24]. Surgery during ECMO necessitates systemic anticoagulation therapy, which may elevate complication risks like bleeding, thus potentially destabilizing the patient's clinical state. Some scholars advocate for delayed surgery when the patient is stable, either close to or following ECMO decannulation. However, late surgery, particularly post-decannulation, may prolong ECMO duration compared to early surgery, increasing complication rates and potentially impacting prognosis. Studies show survival rates around 95 % for patients with small defects amenable to direct repair [25]. PDA blood flow patterns can act as reliable indicators of neonatal PH for optimal CDH repair timing, with the best intervention period being when PDA blood flow shifts to a left-to-right shunt [26]. Despite a reduction in pulmonary artery pressure, hypoxemia may persist, likely due to insufficient tidal volume and excessive pulmonary circulation, leading to pulmonary congestion. This imbalance disrupts the ventilation-perfusion ratio, even with high-intensity mechanical ventilation support. Tidal volume significantly improved after the repair of the CDH. This case also illustrates that pre-ECMO cannulation surgery can promptly restore normal anatomy, alleviate thoracic compression, and create space for lung expansion. Postoperative benefits from stable ECMO operation, stable hemodynamics, and lung-protective strategies allowed the lungs to rest adequately, potentially aiding in the improvement of respiratory function and pulmonary hypertension.

CDH's clinical manifestations stem from the pathological effects of herniated organs compressing the lungs during pivotal periods of lung development. As lung compression intensifies, there's a corresponding decrease in bronchial and pulmonary artery branches, disruption of developing lung structures, incomplete alveolar development, reduced terminal bronchioles, and impaired alveolarization, leading to progressive pulmonary hypoplasia [27]. Additionally, reduced pulmonary vascular beds, vascular remodeling, and altered vascular reactivity contribute to pulmonary hypertension in neonates with CDH [28]. Neonates with severe CDH often present with underdeveloped lungs, resulting in significant postnatal pulmonary hypertension and hypoxemia, with right-to-left shunting through the PDA. Since the presence of PDA relieves right ventricular afterload, it should not be prematurely closed to prevent exacerbating right heart failure. Post-surgical CDH treatment alleviates thoracic compression, facilitating gradual lung expansion, while ECMO support improves oxygenation and reduces pulmonary artery pressure. When pulmonary artery pressure drops below aortic pressure, PDA shunting shifts from right-to-left to left-to-right. A large PDA at this stage can lead to substantial left-to-right shunting, increasing pulmonary blood flow while reducing systemic circulation, potentially resulting in increased pulmonary blood volume, reduced systemic blood flow, perfusion lung injury, and cardiac volume overload [29]. Conversely, reduced systemic circulation can cause tissue ischemia and hypoxia, leading to lactate accumulation, necrotizing enterocolitis (NEC), and other complications. For neonates with a large PDA causing significant pulmonary congestion or hemorrhage, left ventricular overload, or reduced systemic circulation with elevated lactate, early PDA closure is often recommended [30]. The literature on PDA closure during ECMO is limited. Wang et al. reported a case of bedside PDA ligation in a neonate with persistent pulmonary hypertension of the newborn during ECMO, where a large PDA was closed after pulmonary artery pressure decreased and the PDA began shunting left-to-right [31]. Persistent fetal circulation caused severe pulmonary insufficiency in a patient who initially demonstrated adequate lung function after CDH repair [32]. In this case, the diameter of the PDA increased to 0.55 cm post-repair, and massive left-to-right shunts led to systemic failure, as indicated by a continued rise in lactate levels to 15 mmol/L. Consequently, we performed a bedside PDA ligation to obstruct the left-to-right shunt. Following the procedure, there was a notable decrease in lactate levels, and lung effusions significantly reduced by the second day. We believe that if the PDA becomes hemodynamically significant, particularly when extensive left-to-right shunting occurs, PDA closure may be necessary to maintain the protective effect on the right ventricle, despite the associated risk of sudden right ventricular failure.

ECMO circuits, along with venous and arterial cannulation, typically utilize various coating techniques to enhance biocompatibility. However, during ECMO, blood contact with non-endothelial surfaces triggers platelet activation, increasing the risk of thrombus formation and blood damage. Consequently, anticoagulation during ECMO is essential to prevent thrombosis and minimize blood destruction. Therefore, managing anticoagulation during ECMO is a critical balancing act between preventing thrombosis and avoiding bleeding. Comprehensive monitoring, analysis, and control of the patient's coagulation function are imperative for effective anticoagulation management. Drawing from both domestic and international data, as well as our experience, we recommend maintaining a platelet count of at least 100 × 10^9/L and hemoglobin levels above 120g/L prior to surgery. The ACT should be controlled between 180 and 220 seconds, and APTT around 60 seconds. Particularly regarding anticoagulation management after intracranial hemorrhage, this study aimed to maintain coagulation parameters within the recommended range by monitoring for thrombotic events and closely observing changes in intracranial hemorrhage. During PDA ligation, mean arterial pressure can be cautiously reduced to 30–35 mmHg using sodium nitroprusside to mitigate the risk of acute bleeding from sudden pressure increases. Post-ligation, the pressure should be gradually returned to 45–50 mmHg [30]. This case demonstrates that arterial duct ligation and lateral ventriculostomy under ECMO, with appropriate anticoagulation, are safe and feasible procedures.

4. Conclusion

Treatment strategies for CDH must be tailored to each patient's unique clinical situation, encompassing aspects such as ventilation tactics, surgical timing, anesthesia choices, and surgical techniques. This article, by detailing the treatment journey of an infant with a right-sided diaphragmatic hernia, affirms the feasibility and safety of undertaking pre-ECMO cannulation diaphragmatic hernia surgery, arterial duct ligation, and lateral ventriculostomy, all under ECMO support. These interventions illustrate the critical importance of customizing CDH management to optimize patient outcomes.

CRediT authorship contribution statement

Kun-Yao Hong: Writing – original draft, Methodology, Conceptualization. Zhi Zheng: Methodology, Investigation, Funding acquisition, Conceptualization. Yi-Rong Zheng: Methodology, Investigation. Hong Liang: Resources, Project administration, Formal analysis. Liang Gao: Methodology, Investigation, Formal analysis. Yu-Cong Lin: Investigation. Jin-Xi Huang: Methodology, Investigation. Qiang Chen: Writing – review & editing, Supervision, Methodology, Investigation. Xin-Zhu Lin: Writing – review & editing, Project administration, Funding acquisition, Conceptualization.

Ethical approval and consent to participate

This study was approved by the ethics committee of Xiamen Women and Children's Hospital and followed the guidelines outlined in the Declaration of Helsinki. Written informed consent was obtained from all the patient's parents.

Clinical trial number

No applicable.

Consent to publish declaration

Written informed consent was obtained from all the patient's parents to publish identifying information or images in the journal.

Data availability

The data supporting the findings of this study are available upon request from the corresponding author on reasonable request.

Funding

This research received no external funding.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Handling Editor: DR AC Amit Chopra