 |
|
 |
|
 |
 |
|
 |
 |
 |
 |
 |
 |
 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
From the Divisions of Pediatric Pulmonology and Neonatal Medicine, Childrens Hospital Los Angeles, University of Southern California Keck School of Medicine, Los Angeles, California
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Study Design. We studied 50 children at 11.1 ± 1.1 years (mean ± SD) who had been treated with neonatal ECMO for meconium aspiration syndrome (38%), sepsis (18%), sepsis with pneumonia (12%), congenital diaphragmatic hernia (12%), congenital heart disease (8%), persistent pulmonary hypertension of the newborn (6%), and respiratory distress syndrome (4%) and 27 healthy controls (10.8 ± 1.6 years). All subjects completed a respiratory questionnaire and performed pulmonary function and graded cardiopulmonary exercise testing.
Results. Neonatal ECMO survivors had hyperinflation (median residual volume: 131%), airway obstruction (median forced expired volume in 1 second: 79%), lower oxygen saturation with exercise, and lower peak oxygen consumption than controls. The ECMO group achieved similar exercise minute ventilation to controls, with more rapid and shallow breathing. ECMO survivors had an increased frequency of exercise-induced bronchospasm. Those who required higher inspired oxygen tension and ventilator pressures after weaning from ECMO had lower forced expired volume in 1 second and oxygen saturation values.
Conclusion. Neonatal ECMO survivors experience lung injury lasting into later childhood. Lung dysfunction correlates with the extent and duration of barotrauma and oxygen exposure as neonates.
Key Words: ECMO • cardiopulmonary exercise test • pulmonary sequela
Abbreviations: ECMO, extracorporeal membrane oxygenation • MAS, meconium aspiration syndrome • CDH, congenital diaphragmatic hernia • CHLA, Childrens Hospital Los Angeles • 
E, minute ventilation • SPO2, oxygen saturation • CO2, carbon dioxide production • O2, oxygen consumption • RQ, respiratory exchange ratio • OI, oxygen index • FIO2, fraction of inspired oxygen • RV, residual volume • FEV1, forced expired volume in 1 second • FEF25%–75%, forced expiratory flow between 25% and 75% of vital capacity • TLC, total lung capacity • VT, tidal volume
Early-infancy lung injury is associated with long-term pulmonary sequelae.1–6 Infants surviving neonatal respiratory failure often experience significant immediate and long-term respiratory morbidity characterized as chronic lung disease in infancy and reactive airways disease subsequently. Although the use of neonatal extracorporeal membrane oxygenation (ECMO) has decreased with the development of new treatment modalities, it is still used as life-saving therapy for term or near-term neonates with reversible respiratory failure and a high mortality risk when treated only with conventional management.7, 8 ECMO is used as rescue therapy for persistent pulmonary hypertension of the newborn, meconium aspiration syndrome (MAS), respiratory distress syndrome, congenital pneumonia, and congenital diaphragmatic hernia (CDH). It has been speculated that reducing barotrauma and hyperoxia prevents additional injury to the lungs and promotes lung healing.9, 10 Although ECMO and lung rest may modify the severity of chronic lung disease, Garg et al11 reported that many infants treated with neonatal ECMO still have abnormal pulmonary mechanics at 6 months of age. Beardsmore et al12 supported these findings and documented a broad spectrum of abnormalities of respiratory function at 1 year of age in neonatal ECMO survivors.
Currently, there is little published data regarding long-term pulmonary follow-up of patients who were treated with ECMO for newborn respiratory failure. Also, to our knowledge, there are no published studies of exercise performance of their lung function in later childhood. We hypothesize that children treated with neonatal ECMO will have airway obstruction and lung hyperinflation, with decreased exercise tolerance, limited by pulmonary mechanisms. Finally, the severity of pulmonary function and exercise-test abnormalities will correlate with the severity of neonatal respiratory failure.
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
Pulmonary Function Testing
Pulmonary function testing was performed in the Pulmonary Physiology Laboratory at CHLA, located at sea level. Children were tested if they had not had a respiratory tract infection for at least 3 weeks before testing. None of the subjects had used bronchodilators within 8 hours of pulmonary function testing. Vital capacity and maximal expiratory flow rates were measured with a mass-flow sensor (Vmax 229, Sensor Medics, Yorba Linda, CA). Functional residual capacity was measured by pressure-volume body plethysmography (2800 Autobox, Sensor Medics).15 We measured nitrogen wash-out to assess the uniformity of ventilation and single-breath diffusing capacity of lung for carbon monoxide to assess pulmonary vascular disease.
Graded Cardiopulmonary Exercise Testing
After pulmonary function testing, children performed a graded exercise test on a treadmill, during which speed and slope were increased in stepwise progression. Children inhaled and exhaled through a mouthpiece from which inspired and expired gas concentrations were analyzed continuously with a rapid-response Zirconium O2 analyzer and infrared CO2 analyzer by a computerized breath-by-breath system (Vmax, Sensor Medics). Inhaled and exhaled tidal volumes were measured with a turbine digital volume transducer (Sensor Medics). From these, the following gas-exchange parameters were measured on a breath-by-breath basis: minute ventilation (E), oxygen consumption (O2), carbon dioxide production (CO2), and respiratory exchange ratio (RQ = CO2/O2) and ventilatory equivalents for oxygen (E/O2) and carbon dioxide (E/CO2). Electrocardiogram, transcutaneous oxygen, and carbon dioxide tension (Transend cutaneous gas system, Sensor Medics) and oxygen saturation (SPO2) by pulse oximeter (BCI International, Waukesha, WI) were monitored continuously during the exercise stress test.16, 17 Subjects were studied for 3 minutes at rest and then during exercise. Treadmill speed and slope were increased every minute until the child could exercise no longer. Lung volumes and expiratory flow rates were measured 5, 10, and 15 minutes after exercise. Changes in maximal expiratory flow rates after exercise were considered abnormal if they exceeded the range established for normal children.18 When maximal expiratory flow rates decreased after exercise, a bronchodilator was administered and maximal expiratory flow rates were measured again.
Statistical Methods
All results are expressed as medians (quartiles 25th and 75th). Individual tests were analyzed and considered abnormal if they were less or more than 2 ± SD from available reference values appropriate for height, gender, and age. The difference between groups was tested by means of the Mann-Whitney U test. Correlations between clinical factors such as duration of ECMO and pulmonary function values were analyzed by using linear regression analysis. P values of <.05 were considered significant.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Primary neonatal diagnoses of the children who were treated with ECMO were MAS (38%), sepsis (18%), sepsis with pneumonia (12%), CDH (12%), congenital heart disease (8%), persistent pulmonary hypertension of the newborn (6%), and respiratory distress syndrome (4%). Specific diagnoses of the patients with congenital heart disease were transposition of great arteries, total anomalous pulmonary venous return, tetralogy of Fallot, and left coronary sinus atrioventricular malformation. The patient with the left coronary sinus atrioventricular malformation had had coil embolization before ECMO treatment. The other patients with cardiac malformations received treatment with ECMO before their corrective surgeries. All congenital heart disease patients had had complete surgical correction of their cardiac defect. The mean gestational age at birth for the ECMO group was 39.4 ± 2.2 weeks (range: 35-42 weeks), and the mean birth weight was 3203 ± 509 g (range: 2240-4300 g). ECMO was indicated on the basis of an oxygen index (OI = mean airway pressure x fraction of inspired oxygen [FIO2]/partial pressure of oxygen, arterial) >40 for 4 hours in 70% of the cases. In the remainder of the patients, barotrauma, acute deterioration, and failure to respond to conventional ventilation precipitated the initiation of ECMO before 4 hours of OI > 40.
Two of the children, who had had MAS, were unable to perform pulmonary function and cardiopulmonary exercise tests because they did not tolerate the mouthpiece. Of 6 CDH patients, 5 had had a left-sided diaphragmatic hernia.
Respiratory Questionnaire
All 50 children in the ECMO group completed a detailed respiratory questionnaire. Seven children (14%) in the ECMO group had hay fever, and there was a history of atopy in the parents or siblings (asthma, eczema, or hay fever) of 8 children (16%). Eight of the children (16%) in the ECMO group had prenatal or postnatal exposure to cigarette smoke. Fifty percent of children complained of wheezing at some point in their life, and 32% had wheeze during the previous 12 months. Fifty-eight percent of children complained of cough or wheeze during exercise, and 24% had limited exercise activity because of their respiratory symptoms. Thirty-two percent of children who were treated with ECMO had physician-diagnosed asthma and used bronchodilators and inhaled steroids. Eight (16%) children were hospitalized beyond the neonatal period for treatment of pneumonia or asthma.
Pulmonary Function Testing
Forty-eight children in the ECMO group and 27 children in control group performed pulmonary function testing. The ECMO group had higher residual volume (RV), lower forced expired volume in 1 second (FEV1), forced expiratory flow between 25% and 75% of vital capacity (FEF25%–75%), and lower SPO2 during rest than controls (Table 2). When children with congenital heart disease and CDH were excluded from analysis, there were no significant changes in the pulmonary function test results. Although there were no significant differences in mean total lung capacity (TLC) or vital capacity between the ECMO group and controls, 3 children in the ECMO group had restrictive disease (defined as TLC < 77% predicted). Two of these children had sepsis as their neonatal diagnoses, and 1 had left-sided CDH. Forty-six percent of children in the ECMO group had hyperinflation (RV > 123% predicted), and 52% had airway obstruction (FEV1 < 82% predicted and/or FEF25%–75% < 68% predicted). The single-breath diffusing capacity of the lung for carbon monoxide was within the normal range in both groups. The slope of phase III of the single-breath nitrogen wash-out curve was increased in 41% of the patients in the ECMO group, indicating relatively nonuniform distribution of ventilation. None of the healthy controls had an increased slope of phase III.
|
O2 was significantly lower in the ECMO group than controls (Table 2). Although both groups had similar maximal E, the ECMO group had lower tidal volumes and higher respiratory rates during exercise (Table 2). Eleven children (22.9%) in the ECMO group but none of the control subjects had oxygen desaturation (SPO2 < 90%) during exercise. As a group, children in the ECMO group had lower SPO2 during exercise than controls (Table 2). Children who were treated with ECMO were more likely to have exercise-induced bronchospasm than controls. Twenty-three children (47.9%) in the ECMO group and 6 children (22.2%) in the control group had exercise-induced bronchospasm (P = .004). There was no correlation between FEV1 at rest and maximal O2. Children in the ECMO group who had lower FEV1 values also had lower SPO2 during exercise (r = 0.294; P = .04).
Correlation of Neonatal Data With Pulmonary Function Testing and Graded Cardiopulmonary Exercise Testing
All neonatal data (gestational age, birth weight, pre-ECMO blood gases, pre-ECMO ventilator settings, OI, age when ECMO was started, time on ECMO, ventilator settings 6 hours after ECMO, time to extubation after ECMO, age on extubation, and total hospital days) of children who were treated with ECMO were analyzed. Children who required higher peak ventilator pressures and higher-than-ambient inspired oxygen concentrations 6 hours after being weaned from ECMO had lower FEV1 in this study. Baseline FEV1 correlated inversely with inspiratory pressure (r = –0.391; P = .009) and FIO2 (r = –0.356; P = .018). In addition, children who required higher inspired oxygen concentrations 6 hours after weaning from ECMO had lower SPO2 during exercise (r = –0.378; P = .01). Finally, children who had longer neonatal hospitalizations had lower FEV1 (r = –0.327; P = .03). There was no correlation between the other neonatal data and pulmonary function and cardiopulmonary exercise data.
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Pulmonary Function and Graded Cardiopulmonary Exercise Testing
In this study, children who were treated with ECMO had airway obstruction, hyperinflation, and hypoxia at rest when compared with healthy controls. Almost half of the patients in the ECMO group had hyperinflation and airway obstruction. Garg et al11 reported the presence of abnormal pulmonary mechanics in children who were treated with ECMO when they were 6 months old. In a report of respiratory follow-up of survivors of a randomized, controlled trial of ECMO, Beardsmore et al12 found that although children treated with ECMO had abnormalities of lung function at 1 year of age, their lung function was slightly better than that of infants treated with other methods of ventilation. We chose to study these children when they were 
9 years old, because alveolar development is complete by 8 years of age, and any pulmonary abnormalities documented after this age imply long-standing impairment of lung function that will persist into adulthood.21, 22 Long-term pulmonary function abnormalities such as hyperinflation and airway obstruction are well recognized after neonatal respiratory failure secondary to lung injury from prematurity, neonatal pneumonia, and MAS.2–4, 23, 24 Treatment modalities such as supplemental oxygen and mechanical ventilation play a major role in the pathogenesis of the chronic lung disease in these children.25, 26 It is commonly believed that the avoidance of continued exposure to high inspired oxygen concentration and barotrauma during the course of ECMO has both reduced mortality and encouraged lung healing and recovery and potentially recovery of normal lung function. However, our study reveals that despite the use of ECMO, the abnormalities of lung function persist beyond infancy into late childhood and possibly into adulthood.
To our knowledge, this is the first study of graded cardiopulmonary exercise testing in children who were treated with ECMO. In this study, children in the ECMO group had lower aerobic capacity values compared with controls. Although only 1 child with CDH had low peak O2 values in the Marven et al study,6 Zaccara et al27 found significantly lower peak O2 values in patients who had CDH. Children with mild MAS achieved normal aerobic capacity in the Swaminathan et al study.4 The greater abnormalities seen in our present study may be due to more severe neonatal respiratory illness and longer duration and greater hyperoxic exposure and barotraumas, because the patients in the Swaminathan et al study had only an average exposure to assisted ventilation for 36 hours after birth.
There were qualitative and quantitative differences in the ventilatory response to exercise in children treated with ECMO compared with healthy controls. The ECMO group had higher E achieved through a lower tidal volume (VT) and higher respiratory rates. In CDH patients during exercise, Marven et al6 found reduced VT and increased frequency of breathing, which they thought was compensatory. Pianosi and Fisk28 also reported a higher respiratory rate with normal tidal volume during exercise in prematurely born children. They attributed the higher respiratory rate to small airway obstruction with both increased resistive and elastic loads to overcome. Rather than raising tidal volume, they maintained ventilation by higher breathing frequencies.
Another striking finding in our study was the presence of oxygen desaturation during exercise in the ECMO group. One quarter of the children in the ECMO group had a decline in SPO2 during exercise. Children with lower FEV1 values at rest had lower SPO2 levels during exercise. Oxygen desaturation during exercise was present in follow-up studies of children who were born prematurely and who had bronchopulmonary dysplasia.3, 28 Last, in this study, children in the ECMO group were more likely to have exercise-induced bronchospasm than controls. Other studies have documented the presence of airway hyperreactivity as a sequela to neonatal respiratory failure.3–5, 19
In our study, we were not able to study a control group of children who had received conventional treatment but also had a similar degree of neonatal respiratory failure. However, the Collaborative Extracorporeal Membrane Oxygenation Trial confirmed the survival advantage of ECMO over conventional management, with ECMO-treated children having slightly better lung function at 1 year of age than that of infants treated conventionally.12 Therefore, we can speculate that these abnormalities might have been more pronounced in the rare surviving children with the same severity of neonatal respiratory failure treated with conventional methods.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
We thank Ha Pham, RPFT, and Gladys Ng, RPFT, for performance of the pulmonary function studies and exercise stress tests.
|
FOOTNOTES |
|---|
Address correspondence to Arnold C. G. Platzker, MD, Division of Pediatric Pulmonology, Childrens Hospital Los Angeles, 4650 Sunset Blvd, MS #83, Los Angeles, CA 90027. E-mail: aplatzker{at}chla.usc.edu
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  C. S. Beardsmore, J. Westaway, H. Killer, R. K. Firmin, and H. Pandya How Does the Changing Profile of Infants Who Are Referred for Extracorporeal Membrane Oxygenation Affect Their Overall Respiratory Outcome? Pediatrics, October 1, 2007; 120(4): e762 - e768. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
||
|
||
Home | My Pediatrics | Journal Information | Current Issue | Past Issues | Subscriptions & Services | Contact Us Click here for faster international access © 2004 American Academy of Pediatrics. All rights reserved. |
||
|
||
