Analysis of Prognostic Factors in Congenital Diaphragmatic Hernia in Neonates.

AIM: The aim is to study the demographic characteristics of neonates with congenital diaphragmatic hernia (CDH) and to analyze the prognostic factors determining mortality. SETTINGS AND
DESIGN: A retrospective cohort of CDH admitted at a tertiary level hospital during January 2005-December 2017.
MATERIALS AND METHODS: All cases of CDH admitted to our tertiary care neonatal intensive care unit before undergoing surgery were included in the study. Babies admitted after surgery and those with eventration of the diaphragm were excluded from the study.
RESULTS: Thirty cases (66.66% males, 53.33% inborn, and 63.3% >37 weeks) formed the study cohort. Mean birth weight was 2762 ± 579.67 g and mean gestational age was 37.12 ± 1.76 weeks. About 56.66% of the cases were detected antenatally. The survival rate was 60%. Predictors of poor survival included herniation of the liver or stomach (P < 0.05), low Apgar score at 5 min (<5), presence of moderate-to-severe persistent pulmonary hypertension of the newborn (PPHN) (P < 0.001), presence of shock (P < 0.003), low partial pressure of oxygen, high alveolar-arterial oxygen gradient, and high oxygenation index during first 24 h.
CONCLUSIONS: Majority of the neonates with CDH at our center were >37 weeks and survival was 60%. The predictors of adverse outcome were low Apgar score, presence of moderate-to-severe PPHN, need for higher ventilatory settings, and shock. Antenatal detection of diaphragmatic hernia did not impact survival rates.

INTRODUCTION

Congenital diaphragmatic hernia (CDH) occurs with a frequency of 1/2000–4000 live births.[1] CDH is characterized by a defect in the diaphragm leading to protrusion of abdominal contents in the thoracic cavity, leading to poor lung development. Despite advances in management of babies with CDH, the mortality rate remains high. Survival rate ranges between 40% and 80% across various centers.[234]

Various poor prognostic factors described included early age at antenatal diagnosis, herniation of the stomach or liver, low Apgar scores, severe respiratory distress, and severe hypoxemia.[5]

The present study was conducted as a retrospective audit to study the demographic profile of neonates with CDH at our center and to analyze factors associated with mortality.

MATERIALS AND METHODS

This study included all cases with CDH admitted from January 2005 to December 2017 at neonatal intensive care services of our center.

Exclusion criteria

The exclusion criteria were as follows:

Those admitted postsurgery

Neonates with eventration of the diaphragm.

Unit protocol

The management of babies with antenatally diagnosed or suspected CDH is standardized in our unit. The protocol followed over the study period is described in detail below.

All resuscitations are attended by a neonatology fellow. Those with antenatally diagnosed CDH are electively intubated in the delivery room and ventilated using T-piece resuscitator and transported to neonatal intensive care unit in transport incubator for further care.

Babies are ventilated on synchronized conventional mode (Acutronic Ventilator) as the initial mode of ventilation. No muscle relaxants are used in our unit. High-frequency ventilation is used as a rescue if requirement of mean airway pressure (MAP) is >14 cm of water. Inhaled nitric oxide (iNO) was used in patients with oxygenation index (OI) >20.

Bedside chest X-ray is obtained at admission and bedside two-dimensional echocardiography (2D ECHO) was performed by pediatric cardiologist within 24 h of admission. Umbilical artery catheter is secured for continuous blood-gas monitoring and continuous invasive blood pressure (BP) monitoring. Fentanyl is used for analgesia at a dose of 2–4 μg/kg/h for all patients; however, muscle relaxants are not used for any neonate. Evidence of shock is managed with appropriate fluid bolus at 10 ml/kg and/or with inotropes dopamine, dobutamine, or norepinephrine.

Surgery is delayed till hemodynamic stabilization is attained with decrease in pulmonary hypertension (preductal saturation >90%, no lability in oxygen saturation, require of fraction of inspired oxygen [FiO2] of <50% on conventional ventilator, review 2D ECHO showing pulmonary pressures at least <2/3rd of the systolic BP). Both open surgery and minimally invasive surgery are being performed at out center.

Data collected

All the demographic characteristics including gestation at antenatal detection, place of delivery, sex, gestation, Apgar score, presence of associated anomalies, site of defect, contents of defect, and age at surgery were recorded. All relevant clinical parameters such as the presence of persistent pulmonary hypertension of the newborn (PPHN), shock, pneumothorax, alveolar–arterial oxygen gradient (AaDO2), OI, and ventilatory settings were recorded on a predesigned pro forma.

Definitions used in the study

PPHN: Based on tricuspid regurgitation jet and shunt across patent ductus arteriosus in 2D ECHO, it can be classified into four categories:

Mild: Estimated right ventricular (RV) pressure <2/3rd of systolic BP

Moderate: Estimated RV pressure >2/3rd of the systolic BP

Severe: Estimated RV pressure is less than systolic BP[6]

Shock: Mean BP <50th centile for that gestation or evidence of tissue hypoperfusion as lactate >5 mmol/L, pH <7.25, and urine output <1 ml/kg/h.

Maximum OI presurgery, highest AaDO2presurgery, and best partial pressure of oxygen (PaO2) presurgery were taken for the analysis. Highest ventilatory setting before surgery was assessed in the analysis.

Statistical analysis was done using SPSS for Windows version 17.0 (SPSS Inc., Chicago, USA). For comparison of categorical variables, Chi-square test and Fisher's exact test were used. For continuous variables, Mann–Whitney test and t-test were used. P < 0.05 was considered as statistically significant.

RESULTS

Baseline details

During the study period, 30 neonates with CDH fulfilled the criteria for inclusion, of which 16 (53.3%) were inborn. The mean gestation age at birth and birth weight was 37.12 ± 1.76 weeks and 2726.22 ± 579.67 g, respectively.

Seventeen (56.66%) neonates were diagnosed antenatally with mean gestation at detection slightly later in survivor group (31.1 ± 4.88 weeks vs. 27.2 ± 4.54 weeks) in nonsurvivors group. Twenty-seven (90%) neonates had left-sided CDH, 2 (6.66%) had right-sided hernia, and 1 (3.33%) had bilateral diaphragmatic hernia. The survival rate was 60%. Eight (26.66%) patients expired before surgery and 4 (13.33%) patients expired postsurgery. Baseline characteristics are shown in Table 1.

Table 1

Factors associated with mortality

Parameters	Survivors (n=18), n (%)	Nonsurvivors (n=12), n (%)	P
Male	12 (66.66)	8 (66.66)	1.00
>37 weeks	7 (38.88)	4 (33.33)	0.76
>2.5 kg	14 (77.77)	8 (66.66)	0.50
LSCS	16 (88.88)	11 (91.66)	0.577
Antenatal detection	8 (44.44)	9 (75)	0.098
Content of hernia (stomach/liver)	2 (11.11)	9 (75)	<0.001
Apgar <5 at 5 min	1 (5.55)	7 (58.33)	0.003
Moderate-to-severe PPHN	3 (16.66)	12 (100)	<0.001
Shock	9 (50)	12 (100)	0.003
Pneumothorax	2 (11.11)	5 (41.66)	0.05

PPHN: Persistent pulmonary hypertension of the newborn, LSCS: Lower segment caesarean section

Associated congenital anomalies noted were tetralogy of Fallot (1), situs inversus (1), and single umbilical artery (1). One patient had a history of CDH in elder sibling too. However, genetic workup could not be done for that patient.

Ventilation details

Mean airway pressure (MAP) and FiO2requirement were higher in nonsurvivor group prior to surgery [Table 2]. High-frequency oscillatory ventilation (HFOV) was used in four patients. One patient required HFOV presurgery. iNO was used in 2 patients; in 1 neonate presurgery, and postsurgery in the other.

Table 2

Ventilatory parameters

Parameter	Outcome		P
	Survived	Nonsurvivors
MAP, mean±SD	9.79±2.78	12.52±1.99	<0.0001
FiO2, mean±SD	60.69±22.2	90.00±19.49	<0.0001
pH, mean±SD	7.302±0.08	7.215±0.16	0.221
pCO2, mean±SD	44.68±14.6	56.75±25.3	0.458
PaO2, mean±SD/median (IQR)	134.5±59.99	50.1 (24-62.5)	<0.0001

MAP: Mean airway pressure, SD: Standard deviation, IQR: Interquartile range

Ventilatory indices associated with poor prognosis were low PaO2presurgery, high AaDO2 [Figure 1], and high OI [Figure 2].

Figure 1

The mean AaDO2 in first 24 hours of admission was significantly higher in non survivor group

Figure 2

The mean Oxygenation index within first 24 hours was significantly higher in non survivors

Surgical and postoperative details

CDH surgery was done electively in all patients. Twenty-two patients underwent open surgery and three of them had undergone minimally invasive surgery (thoracoscopic). Median age at surgery was 51 h with interquartile range of 48–78 h in survivors and mean age of 54 ± 5.88 h in nonsurvivors group. Two patients developed postoperative chylothorax. One patient developed chronic lung disease with oxygen requirement for 40 days. However, no patient required oxygen postdischarge.

Three patients developed culture-positive sepsis during the hospital stay with methicillin-resistant Staphylococcus, Escherichia coli, and Pseudomonas species. One patient developed preoperative pneumothorax, whereas six patients developed postoperative pneumothorax.

The predictors of adverse outcome were low Apgar score at 5 min, presence of moderate-to-severe PPHN, need for higher ventilatory settings, and shock. Antenatal detection of diaphragmatic hernia did not impact survival rates. As noted from Figure 3, majority of the nonsurvirors died within 10 days of admission.

Figure 3

Survival analysis showing most of the babies expired within 10 days of admission

DISCUSSION

The present study highlights a single-center experience from Chennai. Survival rate of 60% was noted in our cohort. Studies from other Indian centers have reported survival rate ranging from 58% to 78%.[234] On the other hand, studies from developed countries have reported a survival rate of 85%–90% with protocolized care.[67]

Babies with antenatal diagnosis have better prognosis, especially if the occurrence of diaphragmatic hernia is later in gestation.[8] In this study, we did not find significant difference in survival rate in babies with or without antenatal detection. This may be explained by the reason that majority of them were diagnosed early in gestation with higher rates of pulmonary hypoplasia and pulmonary hypertension. Similar results were observed by Chandrasekaran et al.[2] Contents of hernia including liver and stomach have been reported to have a poor prognosis.[8910] In the present study also, babies with liver or stomach as the content of hernia were at significantly higher risk of mortality.

Clinical parameters associated with poor prognosis in various studies are low Apgar score at birth, presence of severe PPHN, shock, and need for extracorporeal membrane oxygenation (ECMO).[257] In our study, low Apgar score (<5 at 5 min), presence of moderate-to-severe PPHN, and shock were associated with poor prognosis.

Need for high ventilatory setting before surgery, low PaO2, high AaDO2(>500), and high OI >20 within first 24 h were found to be significantly associated with mortality in this study. Incidence of pneumothorax was 23.33% in this study, which was comparable to 22% incidence observed in a study by Chandrasekaran et al.[2]

CONCLUSIONS

The present study highlights that the predictors of adverse outcome in CDH include low Apgar score (<5 at 5 min), presence of moderate-to-severe PPHN, need for higher ventilatory settings, and shock. Antenatal detection of diaphragmatic hernia did not impact survival rates.

Limitations

The study has not addressed the long-term morbidity in CDH neonates and has a small sample size compared to other published Indian studies from public sector hospitals.

Despite these limitations, the study does show that well-defined and protocol-based management of CDH in a corporate hospital setup is feasible with good results. The study underscores the importance of previously described predictive factors of survival in CDH.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.