|
|
||||||||
Ann Thorac Surg 2003;75:250-256
© 2003 The Society of Thoracic Surgeons
a Department of Thoracic Surgery, University Hospital of Lille, Lille, France
b Department of Neonatology, University Hospital of Lille, Lille, France
Accepted for publication August 8, 2002.
* Address reprint requests to Dr Jaillard, Clinique Chirurgicale, Hôpital Calmette, Centre Hôpitalier Régional et Universitaire de Lille, 59037 Lille, France.
e-mail: sjaillard{at}chru-lille.fr
| Abstract |
|---|
|
|
|---|
METHODS: Follow-up was performed in 85 newborn infants with CDH admitted in our neonatal intensive care unit between January 1991 and December 1998. Early (< 2 months) and late mortality (
2 months), and respiratory, nutritional, musculoskeletal, and neurosensory outcome at 2 years were recorded.
RESULTS: Surgical repair was performed in 59 infants (69%) at a median postnatal age of 124 (range, 38 to 246) hours. Extracorporeal membrane oxygenation was used in 26 (30%) newborn infants. Survival at 2 years was 51 of 85 (60%) (early death, 28/85 [33%]; late death, 6/85 [7%]). Late deaths occurred because of persistent pulmonary hypertension or iatrogenic complications. Twelve of 51 (24%) newborn infants were oxygen dependant at the postnatal age of 28 days, and 1 of 51 (1.9%) was still oxygen dependant at 2 years. Growth failure was noted in 9 of 51 (18%), mainly related to severe gastro-esophageal reflux and oral aversion. Scoliosis was diagnosed in 2 infants. Neurologic examination at 2 years was normal in 45 of 51 (88%). Cerebral palsy and developmental delay were observed in 2 and 4 infants, respectively. Four infants (8%) experienced associated problems. Respiratory, nutritional, and musculoskeletal morbidity was higher in infants treated by extracorporeal membrane oxygenation (p < 0.05).
CONCLUSIONS: CDH infants are at risk for adverse nutritional and respiratory outcome. Despite severe respiratory failure at birth, prolonged oxygen therapy above 2 years of age is uncommon. Conversely, failure to thrive related at least in part to gastroesophageal reflux and oral dysfunction remains the major problem at 2 years of age. However, both nutritional and respiratory problems tend to improve with age.
| Introduction |
|---|
|
|
|---|
Since 1990, all the surviving newborn infants with CDH admitted at our institution are prospectively enrolled into a follow-up study. The aim of our study was to assess the outcome at 2 years of infants with CDH since the introduction of new management strategies, and to identify children with multidisciplinary health problems. Mortality and respiratory, nutritional, musculoskeletal abnormalities, and neurosensory outcome at 2 years were recorded.
| Patients and methods |
|---|
|
|
|---|
ECMO was used in case of severe hypoxemia (postductal PaO2 < 40 mm Hg or postductal SpO2 < 85% for more than 2 hours) despite optimal medical management, including the use of inotropic drugs, inhaled nitric oxide, high-frequency oscillatory ventilation, and exogenous surfactant. Both venovenous (before 1993) and venoarterial ECMO were used. Exclusion criteria for ECMO or surgery were birth weight below 1,800 g or gestational age less than 34 weeks, lack of "honeymoon" period defined by postductal PaO2 above 60 mm Hg and PaCO2 less than 50 mm Hg despite optimal medical management, and severe intra cranial complications. Right common carotid artery (RCCA) was reconstructed after venoarterial ECMO. Surgery was performed in the NICU and was always conducted under the control of the same surgeon (SJ). Repair was usually carried out within 24 hours of stability. A transabdominal approach was systematically used. A large "oversized" prosthetic patch (Gore-Tex; W.L. and Associates, Inc, Flagstaff, AZ) was used as soon as diaphragmatic repair would create a tension. Gentle suction was achieved through a chest tube according to mediastinal shift.
Early and late mortality was recorded. Early death was arbitrarily defined as a death within the first 2 months of life, and late death as a death after this period.
All the surviving infants were included into a follow-up study. Respiratory, nutritional, musculosqueletal, and neurosensory outcomes were assessed at 2 years. Chronic lung disease was defined as any requirement for supplemental oxygen 28 days after birth. Growth, including weight, length, and head circumference, was compared with the reference curves of Sempé [11]. Growth retardation was defined as weight less than fifth percentile. An esophageal pH monitoring was systematically performed to detect gastro-esophageal reflux (GER) before discharge, or controlled if clinically indicated. Oral dysfunction was defined as sucking-swallowing reflex anomalies with oral aversion. Assessment of neurologic outcome was made using items from Touwen [12] and Amiel-Tison and Grenier [13]. Neurologic outcome was defined as normal, developmental delay [14], or cerebral palsy according to Hagberg and associates [15]. All the infants treated by ECMO were evaluated by a psychologist in order to measure the developmental quotient (DQ) with the Brunet-Lezine scale [16]. Neurosensorial evaluation was systematically performed, including ocular and fundus oculi examination, hearing evaluation by Boel test, and brainstem auditory-evoked potential (BEAP).
Statistical analysis was performed using the SAS software (Statistical Software, Cary, NC) for comparison of means with the Students t test, and difference of proportions with
2 test or Fishers exact test. Mann-Whitney test was used for the comparison of groups of patients. A p < 0.05 was considered significant.
| Results |
|---|
|
|
|---|
Twenty-six among the 85 CDH infants (30%) failed to reach operation (Fig 1) for the following reasons. (1) Twenty infants had severe respiratory failure noted immediately after birth, in whom preductal PaO2 more than 60 mm Hg and PaCO2 less than 50 mm Hg were never reached despite optimal ventilatory and hemodynamic management using high-frequency oscillatory ventilation, inhaled NO, surfactant, et cetera. The use of ECMO or surgery was precluded because of lack of reasonable expectation of potential reversible respiratory failure. (2) Three infants had severe anoxo-ischemic brain injury precluding the use of ECMO or surgical repair. (3) Three other infants died without surgical repair while on ECMO support (n = 1) or immediately after ECMO was discontinued (n = 2) from extensive superior caval syndrome or recurrent refractory hypoxemia.
|
Fifty-seven infants (67%) were alive at 2 months. Mortality rates are summarized in Figure 1. Six late deaths were observed. Five of these 6 infants were treated by ECMO in the neonatal period. These late deaths were due to severe respiratory failure in 3 infants, with death between 5 and 18 months. Two infants died during late surgical intervention, 1 during general anesthesia for jejunostomy at 12 months, and 1 with mediastinal hemorrhage after insertion of a central venous line at 24 months. One infant, who did not required ECMO initially, died from sudden infant death at 3 months. Thus, 51 infants (60%) were alive at 2 years. Among the inborn population, survival rate was 73% (25/34) at 2 years. Among all the 2-year survivors, median age at discharge was 34 (range, 12 to 930) days, median time to extubation was 10 (range, 1 to 930) days, and median duration of oxygen therapy was 12 (range, 1 to 930) days. The use of ECMO was associated with an increase risk of death at 2 years (31% vs 8.3%, p = 0.04 by Fishers exact test). Perinatal risk factors for overall mortality are summarized in Table 1. A decreased survival was also observed with the use of a patch during initial surgery (Table 2).
|
|
|
|
| Comment |
|---|
|
|
|---|
Our follow-up study highlights that significant long-term morbidity exists in CDH infants. Half of the surviving infants presented adverse pulmonary, gastro-intestinal, nutritional, or musculoskeletal outcome at 2 years. Not surprisingly, the need for ECMO was predictive of more significant morbidity. These results are consistent with recent reported experiences from the multidisciplinary follow-up team of the Boston Childrens Hospital [4, 6]. The earlier studies reported by Van Meurs and associates [10], then later by dAgostino and associates [7], on the long-term respiratory outcome in neonates treated by ECMO showed, respectively, a 67% and 63% chronic lung disease rate. The authors found that diagnosis of CDH further increases the risk of chronic lung disease in infants treated by ECMO. More recently, Muratore and associates [4] have shown that the need for ECMO and the presence of a patch repair are both predictive of more significant pulmonary morbidity, but that non-ECMO CDH survivors also require frequent attention to pulmonary issues beyond the neonatal period. Furthermore, obstructive airway disease was found in about 25% of CDH infants at 5 years of age. Our data are in accordance with these results [4, 10] and support the findings that the rate of chronic lung disease was higher when patch repair was required. Chronic lung disease in CDH appears mainly related to pulmonary hypoplasia [23] and to lung damage due to mechanical ventilation [24]. Supporting this, our group of infants with chronic lung disease had higher oxygenation index during the first 2 days of life.
Nutritional morbidity has been more recently described in this population [6, 25]. It seems, however, to have been highly underestimated, and in our cohort, nearly 20% of our patients and 40% of the ECMO survivors experienced growth retardation with a weight less than the fifth percentile at 2 years of age. The pathophysiology of this growth retardation is complex, including catabolic stress in the neonatal period, oral aversion, GER, and persistant pulmonary morbidity [6]. The management of these patients is difficult, and there is no overall agreement in the literature. In his cohort, Muratore and associates [6] found that 56% of the CDH patients was below the 25th percentile for height and weight during the first year of life and one-third of the population required monthly visits to the multidisciplinary clinic for day-to-day management of growth and nutritional issues. Thirty-three percent of the patients required a gastrostomy tube, which is, much more than our 7% of patients. We were very restrictive to perform gastrostomy, as this generally increases oral aversion. However, this policy needs to be more carefully evaluated and counterbalanced with new hypotheses concerning maturation of the sucking-swallowing reflex [26]. Clinical evaluation by speech therapist revealed a hypersensibility of the oral zone and of the nauseous reflex, suggesting significant oral dysfunction. Whether early oral rehabilitation could prevent such oral problems remains speculative, but prevention with feeding protocols inspired by those recently described for preterm babies [27] should be discussed. GER plays a major role in the appearance of nutritional morbidity. In previous series, its incidence ranges from 12% to 69% in CDH infants [5, 28]. Obviously, the incidence of GER depends on the diagnostic methods. In some studies, systematic pH probe and acid clearance were performed [28]. In other studies, only symptomatic GER were recorded. Nagaya and associates [5] explained GER in CDH infants by slow pulmonary expansion of the affected side with a mediastinal shift, increase in intraabdominal pressure, and malposition of the stomach. An oversimplification of GER pathophysiology is based on the utero changes associated with the diaphragmatic defect. Mass effect created by the herniated viscera into the thorax is presumed to give an obstruction of the distal esophagus [29]. Stolar and associates [28] also reported serious anatomic and functional abnormalities of the esophagus observed on plain chest roentgenogram as a mediastinal mass. Contrast studies may identify the mediastinal anomalies as ectatic esophagus. Low gastric emptying and gut dysmotility increase the nutritional problem, and may require prolonged enteral tube feeding for the most unfavorable cases [3, 30]. As previously described [3, 7], we found that GER was more frequent in CDH infants who required ECMO. Moreover, 79% of the severe GER in CDH survivors were found in infants treated with a prosthetic patch. Most of them had either a large diaphragmatic defect or diaphragmatic agenesis. Accordingly, during recent surgical repair, anatomical findings urge us to propose antireflux surgery in these forms. The principle consists of esophagus fixation on the pillar of the healthy diaphragm (the other pillar is fragile or used for prosthetic patch fixing) to prevent the lower esophagus from sliding [5]. Then, we choose to perform a "hemi-Nissen" (Thal fundoplication), because the gastric fundus is abnormal and too small to be wrapped circumferentially around the esophagus. Although this added surgical procedure seems to reduce symptomatic GER during the postoperative course, long-term results are presently unknown.
Orthopedics anomalies such pectus deformities and scoliosis were also found in CDH infants [30]. Only two cases of scoliosis were observed in our population. These two infants had associated GER and growth failure, which increased the difficulties in management. It is likely that other orthopedics anomalies may occur later with development. As suggested by Lund and associates [30], scoliosis or pectus deformities may be related to diaphragmatic tension when closing hernia defect.
Most of our CDH infants (45/51) were found functionally normal on motor testing at 2 years of age, and ECMO did not alter the neurologic outcome. Our rate of cerebral palsy (CP) (4%) was in accordance with those described by Graziani and associates (9%) [31] or Glas and associates (5%) [32] in two cohorts of infants treated by ECMO. Some studies have suggested that infants with an initial diagnosis of CDH requiring ECMO presented poor neurodevelopmental outcome [3, 8], but follow-up was less than 2 years of age. Lund and associates [30] noted that motor delays improve with age, and for Graziani and associates [31], early delay in motor skills noted in infants with CDH was not predictive of cognitive development at school age. Follow-up is still in progress to assess school performances and quality of life at a later age in this cohort. We were concerned by the absence of hearing loss in our group of patients. A striking high rate of hearing dysfunction was reported in CDH infants, whether treated or not by ECMO [9, 34]. Prolonged use of pancuronium [35] as well as diuretics in infants with chronic lung disease [9] have been mentioned as possible risk factors. Both of these medications are rarely used in our unit, and this could explain at least in part the lack of hearing disorders in our cohort. However, the progressive nature of the loss in the ECMO survivors [36] requires a long-term follow-up to precisely appreciate the whole hearing ability.
The most striking feature of our results was the occurrence of associated problems in nearly 8% of our patients. In all of the cases, management was very complex, with intricacy between oral aversion, GER, growth failure, chronic lung disease, and developmental delay, associated with an increase parental stress. This has been underestimated because of the lack of coordinated follow-up programs in most of the cohorts studied. As already mentioned by Muratore and associates [6], "this leads to fragmented and inefficient care of complex issues by community pediatricians who are often unfamiliar with these problems." The key for optimal care and management of infants with CDH probably lies in the development of multidisciplinary follow-up programs, which could also be allowed to evaluate management protocols.
In conclusion, CDH infants are at risk of multiple medical problems whose management remains complex and unsolved. Oral aversion is probably one of the most difficult issues to prevent at present and requires specific evaluation. Late deaths can occur because of persistent pulmonary hypertension or iatrogenic complications. Although both nutritional and respiratory problems tend to improve with age, long-term follow-up by a multidisciplinary team is still required in CDH infants survivors to determine through adolescence and early adulthood the need for additional services and the health-related quality of life [33].
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
I M Balfour-Lynn, D J Field, P Gringras, B Hicks, E Jardine, R C Jones, A G Magee, R A Primhak, M P Samuels, N J Shaw, et al. BTS guidelines for home oxygen in children Thorax, August 1, 2009; 64(Suppl_2): ii1 - ii26. [Full Text] [PDF] |
||||
![]() |
Section on Surgery and the Committee on Fetus and Postdischarge Follow-up of Infants With Congenital Diaphragmatic Hernia Pediatrics, March 1, 2008; 121(3): 627 - 632. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Colvin, C. Bower, J. E. Dickinson, and J. Sokol Outcomes of Congenital Diaphragmatic Hernia: A Population-Based Study in Western Australia Pediatrics, September 1, 2005; 116(3): e356 - e363. [Abstract] [Full Text] [PDF] |
||||
![]() |
I M Balfour-Lynn, R A Primhak, and B N J Shaw Home oxygen for children: who, how and when? Thorax, January 1, 2005; 60(1): 76 - 81. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |