Mortality in infants with congenital diaphragmatic hernia: a study of the United States National Database



To examine the effect of regionalization of care on outcomes of neonates with congenital diaphragmatic hernia (CDH).

Study Design:

We analyzed the National Inpatient Sample and the ‘Kids’ database for the years 1997 to 2004. Infants with CDH were grouped based on whether they underwent surgical repair at the hospital of birth, or at another facility. Groups were compared using chi-square, t-test and logistic regression.


A total of 2140 infants were included: 41% were females, 42% were Caucasians, 48% were transported, 20% reported the use of extracorporeal membrane oxygenation (ECMO)and 33% died. Only 79% underwent operative repair, in which 85% survived after surgery. Survival among operated patients who used ECMO was 40%. Transported infants used more ECMO than non-transported ones (25 vs 15%; adjusted odds ratio (OR) 1.46; confidence interval 1.1 to 1.9, P=0.007), and had higher mortality after surgery (16 vs 13%; adjusted OR 1.46; confidence interval 1.1 to 2, P=0.02).


The utilization of neonatal transport of CDH patients is associated with increased mortality and increased need for ECMO. This study supports the need for regionalization of care, and favors maternal transport before delivery of CDH newborns.


Congenital diaphragmatic hernia (CDH) accounts for 8% of all major congenital anomalies and carries an incidence of 1 in 2000 to 4000 births.1 Despite remarkable improvement in the survival rate of CDH patients over the past 20 years,2, 3 the mortality of this complex malformation remains high yet underestimated because of a ‘hidden mortality’ related with those infants who die before being operated. Recognized factors contributing to the high mortality and morbidity of CDH patients include the severity of pulmonary hypoplasia, pulmonary artery hypertension and the presence of any associated fetal anomalies.4, 5 The management strategies for CDH patients have changed dramatically over the years. Some centers reported better outcomes in association with the implementation of gentle ventilation, permissive hypercapnea and delaying surgical repair after achieving physiologic stability.6, 7, 8 The effect of neonatal transport on outcomes of CDH infants has not been directly addressed, although multiple reports debated the difference in outcomes associated with transport, suggesting that in utero transport to tertiary care centers can be advantageous to both mothers and infants.4 The thermal and hemodynamic instabilities that routinely occur during transport can be compromising to neonates; especially in the setting of CDH patients who have hypoplastic lungs and pulmonary hypertension. Therefore, in this cohort study, we aimed to compare surgical outcomes of CDH infants born and managed at the same tertiary care center to those born at a primary care hospital who were then transported for surgical management. We also tested the effect of timing of surgical repair on outcomes. We used a national in-patient sample database in this study.


The data used in this study were obtained from the de-identified national in-patient sample and the ‘Kids’ database for the years 1997 to 2004. Data were produced as part of the Healthcare Cost and Utilization Project Inpatient Discharge Data of the Agency for Healthcare Research and Quality. These data were all-payer database that contained information from millions of in-patient discharges from approximately 1000 hospitals across the United States. Hospitals were sampled to represent a 20% stratified sample of all community hospitals. Data represented about 37 states across the nation.

In this study we extracted data on gender, race, geographic region, coexisting diagnoses, use of extracorporeal membrane oxygenation (ECMO), complications and length of stay, age at transport, age at surgery, gestational age and birth weight.

Patient selection

Inclusion criteria for the analysis were: (1) infants diagnosed with CDH according to the International Classification of Disease-Ninth Revision, (2) age at admission of <8 days and (3) patients operated for CDH repair within the first 6 weeks of life and/or died before surgery. From the data, we excluded infants with brain anomalies, abdominal wall defects, multiple congenital anomalies and chromosomal disorders. Included infants were classified into two groups based on whether they did not undergo transport (group 1), or were transported from the birth hospital to the surgical facility (group 2). They were further stratified into three groups according to their age at operation (0 to 2 days, 3 to 7 and >7 days). In addition, transferred infants were counted only once at the recipient hospital but not at the referring hospital.

Data management and analysis

Data were managed using SAS for Windows version 9 (SAS Institute, Cary, NC, USA), Comparisons between groups were performed for categorical variables using chi-square and for continuous variables using t-test. A logistic regression model was structured to calculate adjusted odds ratios (OR), controlling for confounders including gender, race, birth weight category, age at operation, spontaneous pneumothorax, persistent pulmonary hypertension of the newborn, sepsis, necrotizing enterocolitis (NEC) and birth asphyxia.


A total of 4741 infants had been diagnosed with CDH. Only 3249 infants had records of CDH repair or mortality, whereas 1492 infants did not have surgery and were discharged home without a confirmation of the CDH diagnosis. Of the 3249 CDH infants, we excluded 988 because of: being transferred out from the hospital of birth (n=539) to avoid being counted twice, admission beyond first week of life (n=144), surgery beyond 6 weeks of life (n=70), brain anomalies (n=86), abdominal wall defects (n=51), multiple congenital anomalies (n=18) and chromosomal disorders (n=90). There were 121 infants without data on whether they were transported or not. Therefore, only the remaining 2140 were included in the analysis. The dominant race of the cases was Caucasian (42%) and the dominant gender was males (59%). There were 52% of inborn cases (group 1), and 48% were transported (group 2). The overall mortality was 33%. The characteristics of the two groups are included in Table 1. When compared with inborn infants, the transported group encountered more cases of NEC (0.4 vs 1.3%, P=0.02) and more cases of persistent pulmonary hypertension of the newborn (8.9 vs 11.8%, P=0.03).

Table 1 Characteristics of the study population by groups (n=2140)

Only 1682 (79%) infants underwent surgical repair of which 85% survived. Compared with the transported group, inborn infants had lower mortality after surgery (12.8 vs 16.2%; adjusted OR 1.46; confidence interval 1.05 to 2.02, P=0.02), and less use of ECMO in operated infants (16.6 vs 23.3%; adjusted OR 1.46, confidence interval 1.1 to 1.9, P=0.007). The overall survival on ECMO was 40.3%. In those who underwent surgical repair, ECMO infants had a better survival rate in the inborn group when compared with the transported group (57.3 vs 46.5%, P=0.03; Table 2).

Table 2 CDH infants who were treated with ECMO (n=425)

Mortality in relation to the age at the time of operation did not differ between the first two groups: 21.1% for the 0 to 2 days group; and 30% for the 3 to 7 days group (P=0.58). The third group that was operated at >7 days encountered greater mortality (48.9%) when compared with the other two age groups (P<0.001) (Figure 1). In addition to the status of transport and day of operation, risks of mortality increased with African-American race, persistent pulmonary hypertension of the newborn, spontaneous pneumothorax, congenital heart diseases and NEC (Table 3).

Figure 1

Rates of mortality and use of ECMO in the study population stratified by date of surgery. The upper panel represents the percentage of infants who used ECMO at the three different categories for age of operation. The lower panel represents the percentage of infants who died at the three different categories for age of operation. *Risks for mortality and use of ECMO are increased when CDH repair was delayed by >7 days (P<0.001). Postnatal day of surgery was documented in only 1321 infants (1682 infants underwent surgical repair).

Table 3 Logistic regression analysis for factors associated with mortality in CDH infants


This study showed an increased mortality and use of ECMO in CDH infants who used transport for their management. Despite being an important subject, the effect of transport on these patients has not been previously addressed. We were able to examine such an effect using a relatively large CDH database that uniformly represents different categories of hospitals in 37 states across the country. We noticed a significant increase in mortality after surgery in transported infants (16.2 vs 12.8, adjusted OR 1.45). Such increase in mortality was noted after controlling for possible confounders that could have affected outcomes, including birth weight category, gender, race, asphyxia, respiratory distress syndrome, persistent pulmonary hypertension, sepsis, congenital heart diseases, spontaneous pneumothorax and postnatal age of surgery. Our data clearly showed a remarkable increase in the use of ECMO in patients who were transported for further care when compared with those who were born and managed at the same tertiary care center. CDH infants who have a better chance of survival are transferred to other institutions whereas non-viable ones remain at the birth facility. Considering that, it is ironic to observe that the use of ECMO was increased in transported infants. The rationale could be related to the increased risk for associated complications that arose in the group being transported; these include NEC and pulmonary hypertension (Table 1).

In this study, 51% of infants who had surgical repair and used ECMO survived. This study provides figures that might be beneficial to physicians and surgeons when discussing with families the possibilities and outcomes in CDH infants who used ECMO. Previous studies showed better figures for survival in ECMO-supported CDH infants.9, 10 However, our study represents the national figure. Infants who used ECMO have greater use of diaphragmatic and abdominal patches during surgical repair, and also frequently undergo other procedures such as fundoplications and gastrostomy tubes.11 Neurological outcomes of CDH infants who were managed with ECMO differed among studies.12 Hearing deficits were increased in CDH infants supported with ECMO.13 In our study, it is important to note that transported infants not only had more utilization of ECMO (P<0.0001), but also had higher mortality rates among those who used ECMO (P=0.03) (Tables 1 and 2).

In the past 20 years, despite regionalization of health care throughout the country, high-risk newborns are increasingly delivered in community hospitals that offer midlevel care.14 Paneth et al.15 compared early (0 to 4 h) and late (4 to 28 days) neonatal deaths among three different levels of care. Early neonatal death significantly increased in level 1 hospitals, implying that deaths in the first hours of life more closely reflect skills in intrapartum management and neonatal resuscitation. Therefore, even the most immediate transfers are too late and unlikely to have much effect on survival. In our study, transported infants used more ECMO. This can be explained by the inevitable stress that associates with transport, especially in the CDH population who are easily prone to develop pulmonary hypertension. Our data showed a significantly increased incidence of persistent pulmonary hypertension of the newborn in transported infants. In addition, it is likely that transported infants are exposed to periods of hypoxia and/or hemodynamic instability that occur during transport. These periods of hemodynamic instability can explain the increased incidence of NEC among transported infants. Resuscitation technique in the first hours of life is a crucial factor that determines neonatal outcomes. When resuscitation in the immediate postnatal period occurs at a primary care facility that is inexperienced with critical cases such as CDH, outcomes could change. Furthermore, en route during transport, infants are frequently exposed to manual positive pressure ventilation, often with high distending pressure, that can precipitate lung damage and respiratory failure. This may explain the increased incidence of spontaneous pneumothorax.

Management of CDH has changed in which emergent surgical correction is no longer necessary and waiting for the infant to achieve a physiologic respiratory stability is recommended.8, 16 Our data show no statistical difference in mortality, when comparing a group of babies operated before 3 days of life or between the ages of 3 to 7 days of life. However, when babies were operated on after 7 days of life, their mortality and use of ECMO increased significantly. It is possible that infants who stabilized after 7 days had a more significant form of the disease before surgery and consequently encountered death. Hence, this study was not structured to show that delayed surgery was an independent risk factor for mortality. However, it is important to note that all previous reports on improved outcomes with delayed surgery were retrospective in nature that used historical controls. Of course, care of CDH needs more individualized decisions, in which timing of surgery will depend on the degree of pulmonary hypoplasia, the severity of pulmonary hypertension and the overall general condition of the infant. Meanwhile, it is not clear whether the timing of surgery itself matter or whether the condition of the infant at the time of surgery is more important. Once the infant is stabilized, there is no clear benefit in waiting for an additional time period before surgical repair. This study prompts the need for a prospective study on the relationship between postnatal age and clinical conditions of infants at the time of surgery and their clinical outcomes.

Similar to any multicenter registry, the network data presented in this study have limitations. ‘Hidden mortality’ has been a problem to account for in different outcome studies for CDH population. Our study showed the effect of this hidden mortality, in which 21% of the population died before undergoing surgical repair. It is to be noted that the number of deaths in the transport group did not reflect the fetal deaths, stillbirths, infants who did not survive resuscitation or the ones who received comfort care.15

We conclude that regionalization of care is associated with improved outcomes of CDH infants. This study does not support a beneficiary effect for delaying surgical repair beyond 7 days.


  1. 1

    Doyle NM, Lally KP . The CDH Study Group and advances in the clinical care of the patient with congenital diaphragmatic hernia. Semin Perinatol 2004; 28: 174–184.

  2. 2

    Chen C, Jeruss S, Chapman JS, Terrin N, Tighiouart H, Glassman E et al. Long-term functional impact of congenital diaphragmatic hernia repair in children. Pediatr Surg 2007; 42: 657–665.

  3. 3

    Javid PJ, Jaksic T, Skasgard ED, Lee S, Canadian Neonatal Network. Survival rate in congenital diaphragmatic hernia: the experience in Canadian Neonatal Network. J Pediatr Surg 2004; 39: 657–660.

  4. 4

    Boloker J, Bateman DA, Wung JT, Stolar CJ . Congenital diaphragmatic hernia in 120 infants treated consecutively with permissive hypercapnea/ spontaneous respiration/elective repair. J Pediatr Surg 2002; 3: 357–366.

  5. 5

    Bagolan P, Casaccia G, Crescenzi F, Nahom A, Trucchi A, Giorlandino C . Impact of current treatment protocol on outcome of high-risk congenital diaphragmatic hernia. J Pediatr Surg 2004; 39: 313–318.

  6. 6

    Sakai H, Tamura M, Hosokowa Y, Bryan AC, Barker GA, Bohn DJ . Effect of surgical repair on respiratory mechanics in congenital diaphragmatic hernia. J Pediatr 1987; 111: 432–438.

  7. 7

    Nakayama DK, Motoyama EK, Tagge EM . Effect of preoperative stabilization on respiratory system compliance and outcome in newborn infants with congenital diaphragmatic hernia. J Pediatr 1991; 118: 793–799.

  8. 8

    Wung JT, Sahni R, Moffitt St, lipsitz E, Stolar CJ . Congenital diaphragmatic hernia: survival treated with very delayed surgery, spontaneous respiration, and no chest tube. J Pediatr Surg 1995; 30: 406–409.

  9. 9

    Heiss K, Manning P, Oldham KT, Coran Ag, Polley Jr TZ, Wesley JR et al. Reversal of mortality for congenital diaphragmatic hernia with ECMO. Ann Surg 1989; 209: 225–230.

  10. 10

    vd Staak FHJM, de Haan AFJ, Geven WB, Doesburg WH, Festen C . Improving survival for patients with high-risk congenital diaphragmatic hernia by using extracorporeal membrane oxygenation. J Pediatr Surg 1995; 30: 1463–1467.

  11. 11

    McGahren ED, Mallik K, Rodgers B . Neurological outcome is diminished in survivors of congenital diaphragmatic hernia requiring extracorporeal membrane oxygenation. J Pediatr Surg 1997; 32: 1216–1220.

  12. 12

    Bennett CC, Johnson A, Field DJ, Elbourne D, UK Collaborative ECMO Trial Group. UK collaborative randomized trial of neonatal extracorporeal membrane oxygenation: follow up to age 4 years. Lancet 2001; 357: 1094–1096.

  13. 13

    Rasheed A, Tindall S, Cueny DL, Klein MD, Delaney-Black V . Neurodevelopmental outcome after congenital diaphragmatic hernia: extracorporeal membrane oxygenation before and after surgery. J Pediatr Surg 2001; 36: 539–544.

  14. 14

    Phibbs CS, Baker LC, Caughey AB, Danielsen B, Schmitt SK, Phibbs RH . Level and volume of neonatal intensive care and mortality in very-low-birth-weight infants. N Engl J Med 2007; 356: 2165–2175.

  15. 15

    Paneth N, Kiely JL, Susser M . Age at death used to assess the effect of interhospital transfer of newborns. Pediatrics 1984; 73: 854–861.

  16. 16

    Logan JW, Rice HE, Goldberg RN, Cotton CM . Congenital diaphragmatic hernia: a systematic review and summary of best-evidence practice strategies. J Perinatol 2007; 27: 535–549.

Download references

Author information

Correspondence to H Aly.

Ethics declarations

Competing interests

The authors declare no conflict of interest.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Aly, H., Bianco-Batlles, D., Mohamed, M. et al. Mortality in infants with congenital diaphragmatic hernia: a study of the United States National Database. J Perinatol 30, 553–557 (2010) doi:10.1038/jp.2009.194

Download citation


  • CDH
  • transport
  • delayed surgery
  • ECMO
  • infants
  • regionalization

Further reading