PEDIATRICS Vol. 115 No. 3 March 2005, pp. 830a-831 (doi:10.1542/10.1542/peds.2004-2524)
Physiologically Redefining Bronchopulmonary Dysplasia
Robert C. Beckerman, MDDepartment of Pediatrics,
Tulane University School of Medicine,
New Orleans, LA 70112
To the Editor.—
I was initially excited after reading the multicenter study published in the November 2004 issue of Pediatrics by Walsh et al, "Impact of a Physiologic Definition on Bronchopulmonary Dysplasia Rates."1 After more careful thought, I wished to express my concerns about the limitations of the authors' proposed "physiologic definition" of bronchopulmonary dysplasia (BPD) and the potential implications on infant morbidity and mortality. The general pediatrician and the pediatric pulmonologist are the physicians who are most frequently responsible for the follow-up and long-term care of high-risk postneonatal infants who have BPD and respiratory control disorders. After leaving the security of the NICU or "step-down" unit, in which they are closely monitored and observed, they are often discharged to high-risk home environments. Their morbidity and mortality rates, in at least the first year of life, remain high. Some succumb to episodes of acute cardiorespiratory failure, while others may die, suddenly and unexpectedly, in homes or in day care centers in which they may be poorly supervised and/or exposed to stresses such as viral infections or continued tobacco smoke or are left sleeping on unsafe surfaces and positions. I am afraid that limiting the definition of BPD to those infants who can maintain an oxygen saturation of >90% for 30 minutes in room air may discourage use of supplemental oxygen needed by many vulnerable preterm infants ready for discharge to the home.
The original article in the New England Journal of Medicine2 first described the clinical, radiographic, and pathologic entity of BPD. Before that, there was Wilson-Mikity syndrome and afterward, chronic pulmonary insufficiency of prematurity. The high-risk neonatal population of infants who have BPD/chronic lung disease have changed greatly over the past 35 years. The gestational ages and birth weights of surviving infants have become lower and lower. We now have some treatments that may alter the time of exposure to factors such as higher fraction of inspired oxygen or airway pressures, which predispose to the development of the more severe forms of BPD. The addition of a "physiologic" component may contribute something to the previous historical-only definition of BPD. That BPD was diagnosed when there was a continued supplemental oxygen requirement at 28 days of age or 36 weeks' postconception.
My major concern lies with the inadequacy of only a 30-minute exposure to room air after a weaning protocol to room air that may not have taken into account the effects of feeding, crying, or active sleep phases. An oxygen saturation in room air of only 90% over 30 minutes to exclude BPD seems too low a level and too short a time. Arterial oxygen saturation of 90% usually correlates with partial pressure of oxygen (arterial) (PaO2) of 60 mm Hg in a normal oxyhemoglobin saturation curve but is at the steep portion of the curve from which the patient can fall rapidly. That implies a normal partial pressure of carbon dioxide (PCO2), pH, and hemoglobin concentration. A dropping hemoglobin concentration, which is so common in the first 2 to 3 months postdischarge, can shift the curve so that PaO2 may actually be lower than expected. Yes, there may be increased release of oxygen to the tissues, but O2 content is lowered. What if the PCO2 rises, the Haldane effect? Then more O2 is released to the tissues. A lower-than-predicted PaO2 may cause pulmonary hypertension. The authors also suggested in their discussion that concurrent use of medications such as methylxanthines, diuretics, and bronchodilators can reduce the amount of supplemental oxygen needed. In contrast, when premature infants, especially those with BPD, go home, their activity and metabolic rates may increase. Methylxanthines and ß-agonists may actually increase metabolic rate, oxygen consumption, and even pulmonary blood flow, thereby increasing ventilation/perfusion ratios and hypoxemia. There is also the problem of abnormal ventilatory responses to hypercarbia and hypoxia. Diuretics can cause hypochloremic metabolic alkalosis and reduce ventilatory response to PCO2 in chronic hypercapnic lung diseases. If hypoxic response is immature or abnormal, then the response to sudden or additional lowering of oxygen saturation to <90%, especially during respiratory illnesses or sleep, especially prone, may make infants who have BPD more vulnerable by reducing arousals to central, obstructive, or mixed apnea and hypoxemia.
Simakajornboon et al,3 in Tulane's Sleep Center, studied convalescing and asymptomatic preterm infants. They were subjected to 6 to 8 hours of polysomnography before discharge. They demonstrated clinically adverse cardiorespiratory events during sleep in room air during the first half of the night, especially in active sleep. Those events and the percentage of quiet sleep were improved with use of supplemental oxygen during the second half of the night. Simakajornboon and coworkers showed in a another study4 that asymptomatic preterm infants who were exposed to tobacco smoke (nicotine) during gestation had a lesser chance of arousal after apnea and hypoxemia events. Again, those studied were not infants who had BPD but were "healthy" preterm infants. My point is that any "stress" test to determine need for supplemental oxygen or even to define BPD should be performed for at least 6 hours, during which there were 1 or 2 feedings and periods of both active and quiet sleep. After 25 years of seeing these apparently "healthy," oxygen-independent preterm infants being readmitted in respiratory distress, failure, or arrest days, weeks, or months after discharge from a 2- or 3-month NICU stay, I've come to think that lowering the threshold for diagnosis of BPD and use of supplemental oxygen may be safer. As the Louisiana Sudden Infant Death Syndrome Risk Reduction Medical Director since 1994, I have reviewed most of the death-scene investigations and autopsies of infants who died suddenly and unexpectedly. With a newly proposed classification system for Sudden Infant Death Syndrome (SIDS),5 the rate of sudden, unexpected infant death, not "classic" SIDS type Ia but rather type II SIDS, may not be declining at all but rather may be on the rise. Preterm infants who had BPD but did not use supplemental oxygen at all or properly seem to contribute to the sudden, unexpected infant death pool.
REFERENCES
- Walsh MC, Yao Q, Gettner P, et al. Impact of a physiologic definition on bronchopulmonary dysplasia rates.
Pediatrics. 2004;114
:1305
–1311
[Abstract/Free Full Text] - Northway WH Jr, Rosan RC. Treatment of hyaline-membrane disease. N Engl J Med. 1969;280 :331
- Simakajornboon N, Beckerman RC, Mack, C, Sharon S, Gozal D. Effect of supplemental oxygen on sleep architecture and cardiorespiratory events in preterm infants.
Pediatrics. 2002;110
:884
–888
[Abstract/Free Full Text] - Sawnani H, Jackson T, Murphy T, Beckerman R, Simakajornboon N. The effect of maternal smoking on respiratory and arousal patterns in preterm infants during sleep.
Am J Respir Crit Care Med. 2004;169
:733
–738
[Abstract/Free Full Text] - Krous HF, Beckwith B, Byard RW, et al. Sudden infant death syndrome and unclassified sudden infant deaths: a definitional and diagnostic approach.
Pediatrics. 2004;114
:234
–238
[Abstract/Free Full Text]
PEDIATRICS (ISSN 1098-4275). ©2005 by the American Academy of Pediatrics
CiteULike 
Connotea 
Del.icio.us 
Digg 
Facebook 
Reddit 
Technorati 
Twitter What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
