Improving outcome in congenital diaphragmatic hernia – experience of a tertiary center without ECMO

Abstract

BACKGROUND:

Congenital diaphragmatic hernia (CDH) has a high mortality rate, representing a therapeutic challenge. Prenatal diagnosis (PND) is essential in defining optimal perinatal strategy, particularly delivery planning. Hospital Pediátrico de Coimbra is the referral centre for all neonatal surgery, particularly CDH, for the central region of Portugal. The aim was to evaluate clinical management and outcome of newborns with CDH.

METHODS:

An exploratory retrospective study made up of newborns admitted to PICU with CDH was undertaken between January 1995 and December 2014. Two groups were formed based on their year of admission: group A (1995– 2004) and group B (2005– 2014) and were compared.

RESULTS:

The mean birth weight of the 69 newborns admitted was 2.762 ± 696 g; the median of the gestational age was 38 weeks. Associated malformations were observed in 28 (40.5%) and 15 (21.7%) had a right-sided diaphragm defect. The global mortality was 13.0%; in group A was significantly higher than in group B (22.2 vs 3.0%; p = 0.029). A reduction in mortality throughout the years was confirmed after adjusting for POS score (OR = 0.77; 95% CI: 0.62– 0.96, p = 0.021). PND was made in 30.6% of cases in the group A and 66.7% in the group B (p = 0.03). Tertiary perinatal hospital birth was achieved in 60% of newborns in the group A versus 84.8% in group B (p = 0.022). Maximum fraction of inspired oxygen showed a statistically significant difference between the two study groups (60% vs 40%; p = 0.009).

CONCLUSIONS:

A significant decrease in mortality was observed throughout the study. The authors highlight the increase in prenatal diagnosis and an improvement in perinatal care with planning delivery as important contributors to these results.

Keywords

Congenital diaphragmatic hernia prenatal diagnosis therapeutic strategy mortality

1 Introduction

Congenital diaphragmatic hernia (CDH) is a rare and complex malformation with a high mortality rate, even in reference centers, representing a therapeutic challenge. According to EUROCAT (European Surveillance of Congenital Anomalies), the total prevalence of CDH was 2.3 (95% CI 2.2 to 2.4) per 10.000 births and 1.6 (95% CI 1.6 to 1.7) for isolated cases. Data from this registry also reveal that CDH was associated with a chromosomal anomaly, genetic syndrome or microdelection in 10.4% and with other major structural anomalies in 28.2% [1].

For decades, CDH was addressed only as a surgical problem and the investigation on this field was focused on the development of techniques for its repair. In the last 20 years, the recognition of bilateral pulmonary involvement (despite the most affected ipsilateral lung) and presence of pulmonary hypoplasia and subsequent pulmonary hypertension (PHT) as the basis of CDH physiopathology lead to improve the CDH medical therapeutic strategy [2 –5]. The progresses in the treatment of this pathology includes minimal handling, sedation, “gentle” ventilation, specific pulmonary hypertension measures, associated with elective surgery. Since the results of the VICI-trial, conventional mechanical ventilation is considered the optimal initial ventilation strategy and high-frequency oscillatory ventilation (HFOV) should be used as a resource approach when theconventional ventilation fails [6, 7].

Extracorporeal membrane oxygenation (ECMO) only available in reference centers has been used in the stabilization of the NB with CDH. Some authors reserve this technique for medical treatment failure, due to related high morbidity and absence of long term benefits [3 , 9]. We have no ECMO facility in our tertiary centre and this technique was only available for newborns in Portugal since 2010 [10].

During the last 30 years, all newborns (NB) with CDH of Portugal?s Central Region have been referred to the Pediatric Intensive Care Unit (PICU) of Hospital Pediátrico de Coimbra. Since 1995, “gentle” ventilation: synchronized pressure-controlled ventilation, peak inspiratory pressure below 28 cmH₂O, lower possible tidal volume (4–6 mL/kg) and higher respiratory rate without causing air trapping, aiming a pH >7.2 and arterial pCO₂ <70 mmHg and the minimum FiO₂ to achieve preductal saturation between 80 and 90%; elective surgery and rescue HFOV are a routine in our unit. The use of inhaled nitric oxide (iNO) was implemented after in 1998. Our hospital participates since 2008, in the “European CDH Consortium” whose objective is the standardization of the therapeutic approach of NB with CDH, including the indications for rescue HFVO (inability to achieve the goals mentioned above for conventional mechanical ventilation) and iNO use (pre and postductal saturation difference above 10%). The sildenafil was limited to the chronic phase of pulmonary hypertension.

A recognized perinatal strategy in CDH management consists on the development of a neonatal network regarding the delivery in a maternity hospital with differentiated perinatal support [4, 11].

This study was intended to identify the reality of this pathology in Portugal’s Central Region in the last twenty years and to analyze it at the light of nowadays knowledge. The aims include the characterization of the NB with CDH admitted to PICU, evaluation of therapeutic strategies used and determination of mortality, as well as its evolution throughout the study time.

2 Methodology

An exploratory analysis made up of NB admitted to PICU between January 1995 and December 2014 was undertaken. The method used in data collection was retrospective, through the consultation of clinical files and PICU database.

Variables related to the NB’s characteristics (gestational age, gender, birth weight, place of birth, need of neonatal support, prenatal diagnosis, liver position, local of the defect, associated malformations, severity of PHT and length of hospital stay, POS score – probability of survival by the Congenital Diaphragmatic Hernia Study Group), therapeutic strategies (ventilation – conventional mechanical ventilation / HFOV, minimal pH, maximum PaCO₂, maximum PIP, maximum FiO₂, maximum oxygenation index and maximum mean airway pressure; sedation and muscle paralysis; timing of surgery; iNO and sildenafil; inotropic or vasopressor support) and mortality at discharge were analyzed. Elective surgery was considered when the intervention was carried out after medical stabilization of PHT.

PHT was indirectly established by the echocardiographic determination of pulmonary systolic arterial pressure (PSAP – gradient right ventricle/right atrium ± 5 mmHg) in relation with systemic blood pressure (SBP). The worse measure was considered.

Based on the admission’s year, two groups were formed – group A (admissions between 1995 and 2004) and group B (admissions between 2005 and 2014). Demographic data, therapeutic strategies and mortality were compared. Statistical analysis was carried out with IBM SPSS^® statistics 21.0 and included absolute and relative frequencies, measures of central trend and dispersion and the application of statistical tests (Student? t or Mann-Whitney U for quantitative variables as appropriated after performing a normality test; Chi-square or Fisher exact test for qualitative variables according to Cochran?s rules). A logistic regression model was applied to assess the mortality throughout the study period after adjusting to POS score. A level of significance of 0.05 wasadopted.

3 Results

Of the 69 NB with CDH admitted to PICU, 38 (55.1%) were boys. Thirty three cases (47.8%) had prenatal diagnosis at a mean gestational age of 25.7 ± 6.8 weeks. In two of them fetal endoscopic tracheal occlusion was performed. The mean birth weight was 2.762 ± 696 g with 19 NB (27.9%) presenting low birth weight. The median gestational age was 38 weeks (Q₁–Q₃ 36.2–39) and 18 (25.8%) were less than 32 weeks. Forty nine children (70.1%) were born in a hospital with differential perinatal support.

In all children with PND the endotracheal intubation and nasogastric tube placement occurred in the first minutes of life. Bag-mask resuscitation wascarried out in 18.8% of the NB, all of them having no established PND.

Malformations were found in 28 NB (40.5%), the most common was congenital heart disease (26.6%). In four children (4.3%) genetic defects compatible with Fryns (3) and Cornelia de Lange (1) syndromes were identified. Fifty three NB (76.8%) had a left-sided defect, 15 (21.7%) a right-sided and one (1.4%) presented with a bilateral defect. The liver position was thoracic in 24 of the 59 (34.8%) with available information. The PSAP/SAP was evaluated in 43 cases (62.3%) with a mean of 0.85 ± 0.33. The median of POS score was 0.85 (Q₁–Q₃ 0.67–0.89).

Concerning therapeutic strategies, almost all NB (95.6%) were submitted to conventional ventilation with a median duration of 3.0 days (Q₁–Q₃ 2.0–11.0). Twenty six (39.4%), were moved to HFOV during a mean period of 5.9 ± 5.4 days. iNO was used in 14 NB (20.3%) and in eight (11.6%) was used sildenafil. Sedation and muscle paralysis were used in 65 (94.2%) and 28 (40.6%) respectively. Inotropic or vasopressor support was required in 40 (57.9%). Milrinone was used in 4 (5.8%).

Sixty one (88.4%) were submitted to surgical correction and in 55 (90.2%) surgery was performed electively at a median of 3.0 days of life (Q₁–Q₃ 2.0–4.0).

The median of PICU stay was ten days (Q₁–Q₃ 5.0–21.0). During the PICU stay, nine NB (13.0%) died and their annual distribution can be seen in Fig. 1. Seven NB died in the first hours of life.

Fig.1

Annual mortality at discharge of newborns with congenital diaphragmatic hernia admitted to PICU.

The mortality in group A was significantly higher than in group B (22.2% versus 3.0%; p = 0.029; Fisher’s test). According to logistic regression model applied, the mortality reduction over the study period was confirmed after adjusting for the POS (OR = 0.77; 95% CI: 0.62–0.96; mortality decrease each additional year; p = 0.021 – Table 1).

Table 1

Logistic regression model for mortality

Variables	β ± standard error	OR	95% CI	p-value
Year	–0.260 ± 0.113	0.771	0.618–0.962	0.021
POS score	–0.708 ± 1.811	0.493	0.014–17.134	0.696

OR – odds ratio; CI – confidence interval; POS – probability of survival.

The NB’s characteristics of the two groups were compared (Table 2). The PND was established in 30.6% of NB in group A and in 66.7% of NB in group B (p = 0.03; Chi-square). Birth in a tertiary perinatal hospital was significantly higher in NB in group B when compared with those in group A (84.8% versus 60%; p = 0.022; Qui-square).

Table 2

Characteristics of the newborns with CDH

	Group A (n = 36)	Group B (n = 33)	p-value
Gestational Age (weeks) – median	38.0	38.0	0.656^¥
IQR (Q₁–Q₃)	36.0–39.0	36.5–39.0
Birth Weight (gram) – median	2980	2720	0.615^¥
IQR (Q₁–Q₃)	2050–3275	2410–3115
PND (%)	30.6	66.7	0.03^§
Age at PND (weeks) – mean ± SD	26.9 ± 6.1	25.1 ± 7.3	0.467^ba
Tertiary Perinatal Hospital Birth (%)	60.0	84.8	0.022^§
ET 1st minutes (%)	56.5	87.5	0.018^§
Right-sided defect (%)	25.0	21.2	0.710^§
Associated malformations (%)	30.6	51.5	0.077^§
Genetic syndrome (%)	10.0	11.8	1.000^¤
Intrathoracic liver (%)	48.1	34.4	0.283^§
sPAP/SBP – median	0.83	0.705	0.881^¥
IQR (Q₁–Q₃)	0.56–1.21	0.59–1.10
POS score – median	0.79	0.86	0.163^¥
IQR (Q₁–Q₃)	0.49–0.90	0.73–0.89
Length of PICU stay (days) – median	11	10	0.559^¥
IQR (Q₁–Q₃)	2.5–22	6–20.8

^baT-test; ^¥Mann-Whitney U; ^§Qui-square; ^¤Fisher exact test; IQR – interquartile range; PND – prenatal diagnosis; ET – endotracheal intubation; sPAP- pulmonary artery systolic estimated pressure; SBP – systemic blood pressure; POS – probability of survival; PICU – pediatric intensive care unit.

In group A, endotracheal intubation was performed in 56.5% of the NB versus 87.5% in the group B (p = 0.018; Chi-square).

Regarding therapeutic strategies, it was observed that in group B the values of maximum FiO₂ were significantly lower (median 60.0% versus 40.0%; p = 0.009, Mann-Whitney). A similar tendency was observed in maximum PIP (Table 3).

Table 3

Therapeutic strategies implemented

	Group A (n = 36)	Group B (n = 33)	p-value
iNO (%)	19.4	21.2	0.855^c
HFOV (%)	42.9	33.3	0.419^c
Min pH – mean ± SD	7.26 ± 0.14	7.23 ± 0.14	0.072^a
Max PaCO₂ (mmHg) – median	60.3	58.5	0.783^b
IQR (Q₁– Q₃)	48.9– 84.2	51.2– 74.0
Max FiO₂ – median	60	40	0.009^b
IQR (Q₁– Q₃)	40.0– 100.0	40.0– 100.0
Max OI – median	10.20	11.70	0.831^b
IQR (Q₁– Q₃)	3.5– 26.8	4.3– 29.1
Max PIP (cmH₂O) – median	25.0	21.0	0.079^b
IQR (Q₁– Q₃)	19.0– 30.0	18.0– 24.0
MAP (cmH₂O) – mean ± SD	10.1 ± 3.3	10.8 ± 3.6	0.467^a
Days of ventilation – median	3.5	3.0	0.758^b
Q₁– Q₃	0.7– 12.0	0.8– 11.0
Sedation (%)	100	90	0.133^d
Muscle paralysis (%)	58.6	27.5	0.009^c
Inotropic/ vasopressor support (%)	82.7	40	<0.001^c
Elective surgery (%)	82.1	97.0	0.085^d

^aT-test; ^bMann-Whitney U; ^cQui-square; ^dFisher exact test; IQR – interquartile range; iNO – inhaled nitric oxide; HFOV – high-frequency oscillatory ventilation; PaCO2 – carbon dioxide partial pressure; FiO2 – inspired oxygen fraction; PIP – peak inspiratory pressure; MAP – mean airway pressure; OI – oxygenation index; Max – maximum; Min – minimum.

4 Discussion

The CDH diagnosis is often established by routine prenatal ultrasound that is part of the protocol used for pregnancy supervising [3 , 12].

Ultrasound evaluation in a tertiary center allows, not only the diagnosis, but also to establish the exact location and size of the defect, the herniated content, position of the liver, presence of associated malformations and observed-to-expected lung-to-head ratio. Comprehensive assessment will also include invasive sampling for high-resolution genetic testing. PND and survival predictors defined by a multidisciplinary team are important for planning the strategy to be adopted, including prenatal counselling and scheduling the labor in a tertiary perinatal hospital in order to benefit from a specialized care [2 , 13–15]. Although, the PND was performed on less than half of the NB admitted to our PICU (47.8%), a significant increase from period A (30.6%) to period B (66.7%) was noticed, achieving values very closed to the ones reported in the literature (66%) [16, 17]. Morphologic ultrasound is not always performed in a reference center and may contribute to this result. The percentage of right-sided defects in this study (21.7%) was higher than described in literature, what could partially justify the lower percentage of PDN [3 , 18]. The detection of these defects by ultrasound is disturbed by the presence of intrathoracic liver, whose density can easily be confused with the lung. Additionally, although the liver may cause a mediastinum shift to the left, the heart remains in its usual position, which makes diagnosis more difficult [12].

The overall survival rate of the NB admitted to our PICU was 87% , which is consistent with the one found in some referral centers (80–90%) [19 –21]. Some authors claim that it can be difficult to exclude the bias caused by prenatal “selection” or perinatal deaths that occurred before the transference and which could explain the higher rate of survival[19, 22]. In Europe, the most optimistic survival rates published by major reference centers with more than 25 cases per year and possibility of ECMO reaches up to 70–75% [23].

Over the years we assisted to a significant decline in mortality of CDH. It could be explained by increasing percentage of PND, tertiary center referral and adoption of well-established protocols [24]. For example, in a study conducted in France “in utero” referral policy to a tertiary center allowed a significant survival increase (from 41% to 66%) and a study by the Canadian Neonatal Network showed that centers with a high number of cases have a survival rate 19% higher than centers with low number of cases [18, 27]. Also, the consensus statement of CDH EURO consortium recommends the delivery and neonatal treatment in high case volume centers (≥6 CDH patients per year) [7]. Despite our low volume of cases per year, the improvement of the PND and perinatal network care associated to an increase of TPH births in recent years, probably lead to a better NB first stabilization by an experienced team previously to the PICU admission, which might contribute to better survival. In addition, the absence of differences into the two groups regarding general characteristics and survival predictors, reflecting a similar severity and the tendency to a lower maximum PIP and FiO₂, which indirectly reveals an increase in “gentle ventilation” strategy, may also have some influence in this result. Permissive hypercapnia and gentle ventilation have been reported to increase survival in neonates with CDH [28, 29]. Furthermore a decreased in ECMO utilization was also noticed [30].

As expected, no differences were observed between the two groups regarding iNO use. There are limitations to provide recommendations for iNO use in CDH newborns since it might transiently improve oxygenation, however does not reduce neither mortality nor ECMO need. Despite this evidence, iNO continues to be used regularly in CDH patients and the CDH EURO consortium consensus recommends its use to control pulmonary hypertension in cases with an oxygenation index of 20 or higher or a pre and postductal saturation difference of 10% or more as a trial of at least 1 hour. If there is no or insufficient response to iNO, intravenous prostacyclinor sildenafil should be considered [7, 30].

The limitations in our study include its retrospective nature and a large period of time considered. A bias of selection might be present since most severe cases may underwent pregnancy termination in later years, which could contribute to the high survival rate achieved.

5 Conclusion

A high survival rate was observed in NB with CDH throughout the study period, despite non available ECMO, which is consistent with data reported by the VICI-trial.

Our study highlights the importance of an organised perinatal network care with PDN and referral to tertiary centers with trained experts, in order to provide a delivery planning and an optimal management of NB with CDH through standard protocols, including “gentle ventilation” strategies.

Given the improved survival and associated co-morbidities (pulmonary, gastrointestinal and neurological) of children with CDH, future emphasis should be placed on finding a better quality of life for these children and in regular meetings with an experienced multidisciplinary team.

Disclosure statement

The authors declare that there are no proprietary, financial, professional or other personal interests of any nature or kind in any product, service and/or company that could be construed as influencing the content of the manuscript.

Footnotes

Acknowledgments

The authors thank Margarida Marques, from the Statistics Department of the Centro Hospitalar e Universitário de Coimbra, for her precious help in the logistic regression model.

References

McGivern

, Best

, Rankin

, et al. Epidemiology of congenital diaphragmatic hernia in Europe: A register-based study. Arch Dis Child Fetal Neonatal Ed.F. 2015;100:137–44.

Hedrick

Management of prenatally diagnosed congenital diaphragmatic hernia. Sem Fetal Neonatal Med. 2010;15:21–7.

Waag

, Loff

, Zahn

, et al. Congenital diaphragmatic hernia: A modern day approach. Semin Pediatr Surg. 2008;17:244–54.

Rottier

, Tibboel

. Fetal lung and diaphragm development in congenital diaphragmatic hernia. Semin Perinatol. 2005;29:86–93.

Rollins

Recent advances in the management of congenital diaphragmatic hernia. Curr Opin Pediatr. 2012;24:379–85.

Snoek

, Capolupo

, van Rosmalen

, et al. Conventional mechanical ventilation versus high-frequency oscillatory ventilation for congenital diaphragmatic hernia: A Randomized clinical trial (The VICI-trial). Ann Surg. 2016;263:867–74.

Snoek

, Reiss

, Greenough , et al. Standardized postnatal management of infants with congenital diaphragmatic hernia in Europe: The CDH EURO ConsortiumConsensus – Update. Neonatology. 2016;110:66–74.

Lally

Congenital diaphragmatic hernia. Curr Opin Pediatr. 2002;14:486–90.

Sluiter

, van de Ven

, Wijnen

RMH

, et al. Congenital diaphragmatic hernia: Still a moving target. Semin Fetal Neonatal Med. 2011;16:139–44.

10.

Rocha

, Soares

, Henriques-Coelho

, et al. Neonatal extracorporeal membrane oxygenation: Initial experience of Hospital de São João. Rev Port Pneumol. 2014;20(6: 336–40.

11.

Stevens

, Chess

, McConnochie

, et al. Survival in early- and late-term infants with congenital diaphragmatic hernia treated with extracorporeal membrane oxygenation. Pediatrics. 2002;110:590–6.

12.

Deprest

, Gratacos

, Nicolaides

, et al. Changing perspectives on the perinatal management of isolated congenital diaphragmatic hernia in Europe. Clin Perinatol. 2009;36:329–47.

13.

Doné

, Gucciardo

, Van

Mieghem

, et al. Prenatal diagnosis, prediction of outcome and in utero therapy of isolated congenital diaphragmatic hernia. Prenat Diagn. 2008;28:581–91.

14.

Skarsgard

, Harrison

. Congenital diaphragmatic hernia: The surgeon’s perspective. Pediatr Rev. 1999;20:71–8.

15.

Moya

, Lally

. Evidence-based management of infants with congenital diaphragmatic hernia. Semin Perinatol. 2005;29:112–7.

16.

Van den Hout

, Sluiter

, Gischler

, et al. Can we improve outcome of congenital diaphragmatic hernia? Pediatr Surg Int. 2009;25:733–43.

17.

Dillon

, Renwick

, Wright

. Congenital diaphragmatic herniation: Antenatal detection and outcome. Br J Radiol. 2000;73:360–5.

18.

Gallot

, Boda

, Ughetto

, et al. Prenatal detection and outcome of congenital diaphragmatic hernia: A French registry-based study. Ultrasound Obstet Gynecol. 2007;29(3:276–83.

19.

Van Meurs

, Short

. Congenital diaphragmatic hernia: The neonatologist"s perspective. Pediatr Rev. 1999;20:79–87.

20.

Bagolan

, Casaccia

, Crescenzi

, et al. Impact of a current treatment protocol on outcome of high-risk congenital diaphragmatic hernia. J Pediatr Surg. 2004;39:313–8.

21.

Boloker

, Bateman

, Wung

, et al. Congenital diaphragmatic hernia in 120 infants treated consecutively with permissive hipercapnea/spontaneous respiration/elective repair. J Pediatr Surg. 2002;37:357–66.

22.

Downard

, Jaksic

, Garza

, et al. Analysis of an improved survival rate for congenital diaphragmatic hernia. J Pediatr Surg. 2003;38:729–32.

23.

Chiu

, Sauer

, Mihailovic

, et al. The price of success in the management of congenital diaphragmatic hernia: Is improved survival accompanied by an increase in long-term morbidity? J Pediatr Surg 2006;41:888–92.

24.

Sartoris

, Varnholt

, Dahlheim

, et al. CDH in Mannheim – algorithm and results. Monatsschr Kinderheilkd. 2006;153:717.

25.

Lally

, Lasky

, Lally

, et al. Congenital Diaphragmatic Hernia Study Group: Standardized reporting for congenital diaphragmatic hernia – an international consensus. J Pediatr Surg. 2013;48:2408–15.

26.

Nasr

, Langer

. Canadian Pediatric Surgery Network: Influence of location of delivery on outcome in neonates with congenital diaphragmatic hernia. J Pediatr Surg. 2011;46:814–6.

27.

Grushka

, Laberge

, Puligandla

, et al. Canadian Pediatric Surgery Network: Effect of hospital case volume on outcome in congenital diaphragmatic hernia: The experience of the Canadian Pediatric Surgery Network. J Pediatr Surg. 2009;44:873–6.

28.

Guidry

, Hranjec

, Rodgers

, et al. Permissive hypercapnia in the management of congenital diaphragmatic hernia: Our institutional experience. J Am Coll Surg. 2012;214:640–5; discussion 646–7.

29.

Lupo

, Castoldi

, Maestri

, et al. Outcome of congenital diaphragmatic hernia: Analysis of implicated factors. Minerva Pediatr. 2013;65:279–85.

30.

Puligandla

, Grabowski

, Austin

, et al. Management of congenital diaphragmatic hernia: A systematic review from the APSA outcomes and evidence based practice committee. J Pediatr Surg. 2015;50(11), 1958–70.