Impact of a Physiologic Definition on Bronchopulmonary Dysplasia Rates
Objective. Bronchopulmonary dysplasia (BPD) is the endpoint of many intervention trials in neonatology, yet the outcome measure when based solely on oxygen administration may be confounded by differing criteria for oxygen administration between physicians. We previously reported a technique to standardize the definition of BPD between sites by using a timed room-air challenge in selected infants. We hypothesized that a physiologic definition of BPD would reduce the variation in observed rates of BPD among different neonatal centers.
Methodology. A total of 1598 consecutive inborn premature infants (501–1249 g birth weight) who remained hospitalized at 36 weeks' postmenstrual age were prospectively assessed and assigned an outcome with both a clinical definition and physiologic definition of BPD. The clinical definition of BPD was oxygen supplementation at exactly 36 weeks' postmenstrual age. The physiologic definition of BPD was assigned at 36 ± 1 weeks' postmenstrual age and included 2 distinct subpopulations. First, neonates on positive pressure support or receiving >30% supplemental oxygen with saturations between 90% and 96% were assigned the outcome BPD and not tested further. Second, those receiving ≤30% oxygen or effective oxygen >30% with saturations >96% underwent a room-air challenge with continuous observation and oxygen-saturation monitoring. Outcomes of the room-air challenge were “no BPD” (saturations ≥90% during weaning and in room air for 30 minutes) or “BPD” (saturation <90%). At the conclusion of the room-air challenge, all infants were returned to their baseline oxygen levels. Safety (apnea, bradycardia, increased oxygen use) and outcomes of the physiologic definition versus the clinical definition were assessed.
Results. A total of 560 (35.0%) neonates were diagnosed with BPD by the clinical definition of oxygen use at 36 weeks' postmenstrual age. The physiologic definition diagnosed BPD in 398 (25.0%) neonates in the cohort. All infants were safely studied. There were marked differences in the impact of the definition on BPD rates between centers (mean reduction: 10%; range: 0–44%). Sixteen centers had a decrease in their BPD rate, and 1 center had no change in their rate.
Conclusions. The physiologic definition of BPD reduced the overall rate of BPD and reduced the variation among centers. Significant center differences in the impact of the physiologic definition were seen, and differences remained even with the use of this standardized definition. The magnitude of the change in BPD rate is comparable to the magnitude of treatment effects seen in some clinical trials in BPD. The physiologic definition of BPD facilitates the measurement of BPD as an outcome in clinical trials and the comparison between and within centers over time.
Survival in very low birth weight (VLBW) neonates has improved steadily, with 84% of all VLBW neonates surviving.1 Although most of these survivors are healthy, chronic lung injury remains an important health burden, with ∼25% developing bronchopulmonary dysplasia (BPD).1,2
The lack of a precise definition of BPD clouds any study in which BPD is an outcome variable. Most neonates with BPD do not undergo lung biopsy or any physiologic test, and thus their pulmonary disease is defined clinically on the basis of the sustained need for supplemental oxygen at 36 weeks' postmenstrual age. The validity of this definition is supported by evidence that oxygen dependence at 36 weeks' postmenstrual age is predictive of long-term impairment in pulmonary function.2 An inherent limitation of this definitional approach to diagnosis is that the need for oxygen is determined by individual physicians rather than on the basis of a physiologic assessment. The assumption that the criteria on which the decision to administer oxygen is uniform and applied similarly across institutions is erroneous.3 Because there is no consensus in the literature, neonatologists have widely divergent practices regarding oxygen-saturation targets. Indeed, published literature cites acceptable saturation ranges from 84% to 98%.3–11 At least 1 study suggests that VLBW neonates do just as well, and may actually have improved outcomes, when managed with lower oxygenation saturation (70–90%).12
We previously reported the development of a physiologic definition of BPD based on oxygen-saturation monitoring in selected infants in low oxygen.13 Therefore, we hypothesized that applying a physiologic definition of BPD would reduce the variation in observed rates of BPD among different neonatal centers.
From October 2000 to June 2002, 2582 consecutive inborn neonates (501–1249 g birth weight) were followed prospectively at the 17 centers of the National Institute of Child Health and Human Development Neonatal Research Network. The mean birth weight was 899 ± 212 g (mean ± SD) and postmenstrual age was 27 ± 2.4 weeks. At 36 ± 1 week's postmenstrual age, respiratory outcomes were defined in survivors as “BPD” or “no BPD” using 2 different definitions. The first definition of BPD used was a traditional clinical definition based on oxygen administration at 36 weeks' postmenstrual age. The second definition was a new physiologic definition based on combined measurement of oxygen saturation and oxygen administration. One hundred twenty-five infants with incomplete data were excluded from the analyses; 354 infants had died before 36 weeks' postmenstrual age, 151 infants were transferred (53 in oxygen), and 354 infants were discharged (12 in oxygen) before 36 weeks' postmenstrual age. This yielded a total cohort of 1598 in whom the physiologic definition was compared with the clinical definition.
Procedure for the Physiologic Definition of BPD
For this study, BPD was defined on a physiologic basis that combined oxygen and ventilation support with an assessment of saturation in selected infants that we previously reported.13 Briefly, at 35 to 37 weeks' postmenstrual age, infants treated with mechanical ventilation, continuous positive airway pressure, or supplemental oxygen concentration ≥30% and oxygen saturations between 90% and 96% were diagnosed with BPD without additional testing. Infants in supplemental oxygen <30% at rest with oxygen saturations between 90% and 96% or ≥30% with saturations >96% underwent a timed stepwise reduction to room air. For infants receiving oxygen by hood, oxygen was weaned in 2% increments. For infants receiving oxygen by nasal cannulae, flow was weaned initially in increments (for flow of 1.0–2.0: step down 0.5 liters per minute [lpm]; for flow 0.1-0.99 lpm: step down 0.1 lpm), and then oxygen concentration was reduced in increments of 20% to room air. Cannulae were removed from the nares for the remainder of the challenge. Oxygen that was given only during feedings was not included for the purposes of eligibility. Those who failed the reduction were diagnosed with BPD. No BPD was defined as treatment with room air with oxygen saturation ≥90% or passing a timed, continuously monitored oxygen-reduction test. No formal assessment of sleep state was made. Neonates were monitored continuously with a cardiorespiratory monitor and pulse oximeter (Nellcor, Mallinckrodt Inc, Hazelwood, MO) and observed directly by a trained neonatal research nurse throughout the reduction test. Values for heart rate, respiratory rate, oxygen saturation, frequency of apnea (cessation of breathing for >20 seconds), and bradycardia (heart rate <80 beats per minute for >10 seconds) were recorded every 60 seconds for a 15-minute baseline period. All occurrences of movement artifact (defined as visible motion of the infant and loss of the monitor's plethysmograph signal) were recorded, and corresponding saturation values were deleted. The test was performed in 4 parts: baseline, reduction phase, room-air observation phase, and return to usual oxygen. When piloting the physiologic definition, we arbitrarily chose the period of each oxygen-weaning step as 10 minutes and the period in room air as 60 minutes, allowing sufficient time at each step for the saturation to equilibrate. In the pilot data, saturation equilibrated by 5 minutes in the weaning step, and all infants who failed in the room-air phase did so before 30 minutes.13 Therefore, in the current study, each weaning step was shortened to 5 minutes and the period of observation in room air was decreased to 30 minutes.
From October 2000 to May 2001, failure was defined as saturation <88% for 5 consecutive minutes. Twenty-eight infants were studied using this definition. After this initial pilot phase, failure criteria were changed to improve the safety of infants studied by limiting exposure to extreme desaturation events. Beginning in June 2001, failure was defined as oxygen saturation 80% to 89% for 5 consecutive minutes or <80% for 15 seconds. There were 2 methods for passing during the room-air observation phase: a rapid pass and a pass after full monitoring. Rapid-pass criteria were met by successfully weaning to room air with all saturation ≥96% for 15 minutes. If the saturations were 90% to 95%, the infant was monitored for 30 minutes in room air and defined a pass when all saturations exceeded 90% in that 30-minute period.
Delivered oxygen concentration was measured directly in infants receiving oxygen by hood using an oxygen analyzer placed directly above the infant's head inside the hood. Among infants receiving oxygen by nasal cannula, the delivered oxygen concentration (termed the effective oxygen concentration) was calculated by using the technique described by Benaron and Benitz as modified for the Supplemental Therapeutic Oxygen for Prethreshold Retinopathy of Prematurity (STOP-ROP) trial.10,14 This technique uses weight, oxygen liter flow, and oxygen concentration to calculate an effective concentration delivered by nasal cannula.
The attending physician individualized the neonate's management. Some infants received oxygen by hood and some by nasal cannula at the discretion of the clinical care team. Infants cared for in nasal cannula received oxygen delivered by either blended oxygen (15 centers) or 100% oxygen (2 centers) at variable flow rates (range: 0.01-2.00 lpm). In 6 centers, ≥1 infants were treated with nasal cannula with room air and variable flow (n = 16; range of flow: 0.03-2.00); all these infants were tested with a reduction test.
The clinical definition of BPD was assessed at 36 weeks' and 0 days' postmenstrual age and assigned in any neonate receiving oxygen supplementation for any portion of the day.
Adverse events, defined before the study began, were collected in every infant and consisted of apnea (cessation of breathing >20 seconds), bradycardia (heart rate <80 beat per minute for >10 seconds), and increase in baseline fraction of inspired oxygen (Fio2) of 5% persisting for >1 hour after the reduction test. Each infant was evaluated for adverse events in the 1 hour before the oxygen-reduction challenge and the 1 hour after the reduction. In addition, if any event occurred together with a saturation 80% to 89% for 5 consecutive minutes or <80% for 15 seconds during the challenge, the test was declared a failure and the infant was returned to the prior oxygen level immediately.
Human Subject Protections
The institutional review boards (IRBs) of all participating hospitals approved the study. Infants were studied after informed permission (written or verbal as required by the local IRB) was obtained from the parent or guardian. The risk of study participation was minimized by the continuous presence of an experienced neonatal research nurse. All infants were returned to their baseline supplemental oxygen at the end of the oxygen-reduction challenge. Results of the room-air challenge were handled differently at different centers in conjunction with local IRB requirements. In 12 centers, the results of the test were given to the primary physician caring for the infant; at the remaining 5 centers, results were concealed. The approach at an individual center was consistent throughout the study period. Additional modifications in treatment after the oxygen-reduction phase, if any, were made at the primary physician's discretion.
Continuous variables were described by mean and SD. Group differences on continuous measures were compared with Student's t tests/analysis of variance. Categorical variables were assessed with frequency distribution, and group differences were compared with the χ2 test. The McNemar test was used for comparing BPD rates by clinical definition and physiologic definition and for differences in the rates of medication use. The impact of baseline effective Fio2 and saturation on oxygen-reduction test results was assessed with regression models to adjust for potential confounding variables. A posthoc power analysis was done based on the McNemar statistic.15 The sample size was sufficient to detect an absolute rate difference of 6% at 80% power and type I error of 0.05.
The outcomes of the neonates at 36 weeks' postmenstrual age are shown in Fig 1. The physiologic definition diagnosed BPD in 398 (25.0%) of 1598 neonates. Of these 398 patients, 272 were diagnosed with BPD because of their continued use of ventilator, continuous positive airway pressure, or oxygen support ≥30% and 126 because of a failed oxygen-reduction test. In contrast, the clinical definition diagnosed BPD in 560 (35.0%) of these 1598 infants. Overall, using the physiologic definition reduced the number of infants diagnosed with BPD from 35% to 25% (P < .0001). The physiologic definition also modestly reduced the variation among centers (variation: 15–66% with the clinical definition and 9–57% with the physiologic definition; P < .0001), although substantial variation remained.
During the physiologic-definition stage, 227 infants were studied with an oxygen-reduction test at a mean of 36.4 ± 3.4 weeks' postmenstrual age. At the time of the oxygen-reduction test, 11 of the infants were receiving oxygen by hood, and 216 (92%) were receiving oxygen by nasal cannula with flow rates at a mean of 0.49 lpm (range: 0.02–2.00 lpm). Of 227 infants tested, 101 (44%) passed the oxygen-reduction phase and were successfully weaned to room air. Of the 126 infants who failed, 83 (66%) failed during the reduction phase, 39 failed during the room-air phase, and 4 had incomplete information on the time of failure. The mean duration of the oxygen-reduction test was 34 ± 24 minutes. As expected, the duration of the study was significantly shorter in those who failed compared with those who passed (23 ± 16 vs 49 ± 24 minutes; P < .001). Infants who failed during the room-air phase did so at a mean of 20 ± 11 minutes.
Both the incidence of BPD and the magnitude of the impact of the physiologic definition on BPD rates varied by center (P < .0001; Fig 2). The overall reduction was 10%, with a range of 0% to 44%. Of 17 centers, 16 had a decrease in their BPD rate and 1 remained the same. The center with the largest impact had a reduction in the BPD rate from 60% with the clinical definition to 16% with the physiologic definition. This center routinely supplements oxygen to achieve saturations >98% in patients with stage 2 or greater retinopathy of prematurity.
For the infants who had the oxygen-reduction test performed, we compared those who passed and those who failed to determine if test failure could be predicted by a preexisting demographic characteristic (Table 1). Infants who failed were of similar birth weight and gestation to those who passed. Baseline heart rate and respiratory rate were also similar among those who failed and those who passed. Saturation before the reduction was significantly higher in those who passed compared with those who failed (97.0 ± 2.3 vs 95.8 ± 2.8; P < .001), and the amount of oxygen supplementation was significantly less (23% ± 3% vs 26 ± 5%; P < .001).
Safety of the Oxygen-Reduction Test
To assess the safety of the oxygen-reduction challenge, we assessed the frequency of predefined potential adverse events in every infant in the 1 hour before and 1 hour after the challenge. Results of the safety monitoring are shown in Table 2. The number of infants with apnea or bradycardia was not different in the periods before and after the challenge. Desaturation events, defined as saturation <88% for >5 consecutive minutes, did occur in more infants after the challenge (3 vs 15 infants; P < .0012). However, the frequency of increased oxygen support in the period before and after the challenge was not different, suggesting that the events were not viewed as significant by the clinical team caring for the infant. No infant had ventilation instituted after the challenge.
Validity of the Physiologic Definition as a Measure of Pulmonary Morbidity
To further assess the validity of the physiologic definition, we correlated the definition with other markers of ongoing pulmonary disease and health-resource use. Infants diagnosed with BPD by the physiologic definition compared with those with no BPD were more likely to remain in hospital at 120 days of age (44.7% vs 6.6%; P < .0001), and more likely to be discharged from the hospital in oxygen (53.3% vs 3.7%; P < .0001). Medication use at 36 weeks' postmenstrual age was also higher in those diagnosed with BPD (Table 3). The frequency of medication use did vary by center (P < .001).
We further analyzed the subset of 227 who received the oxygen-reduction challenge. As previously shown, infants who passed and infants who failed the oxygen-reduction challenge were of similar birth weight, postmenstrual age, and age at the time of study (Table 1). Diuretic use at 36 weeks (39% vs 49%; P = .11) and steroid use at 36 weeks (2% vs 3%; P = .58) were not different between those who passed and those who failed. However, significantly more of those who failed the room-air challenge were discharged from the hospital in oxygen than in those who passed (58% vs 26%; P < .001) The physiologic definition correlates with ongoing pulmonary morbidity and health-resource utilization.
Potential Impact of Varying Saturation Cutoff for Failure
Because the optimal oxygen-saturation value for convalescent preterm infants is not known, we assessed the impact of varying oxygen-saturation cutoff values between 88% and 92% on the test results. The failure criteria of oxygen saturation <90% for 5 consecutive minutes or saturation <80% for 15 seconds proved to be a robust cutoff point. When we assessed the potential impact of lowering the failure criteria to <89% for 5 consecutive minutes (while still failing any <80% for 15 seconds), only 1 additional infant passed the challenge. No additional infants passed at a failure cutoff of <88% for 5 consecutive minutes. Similarly, when we assessed higher cutoff values of 91% and 92% for a failure cutoff, no additional infants failed the test.
The physiologic definition of BPD standardized the definition of BPD among centers and led to a reduction in the overall rate of BPD from 35% to 25%. Variation in rates of BPD among centers was reduced modestly, but significant variation among centers remained even when using the physiologic definition.
The clinical definition was assigned at 36 weeks' and 0 days' postmenstrual age, whereas the physiologic definition was assigned between 35 and 37 weeks' postmenstrual age. The wider window of time allowed infants who were steadily weaning from oxygen to reach room air in the window and also ensured that infants were not studied during an acute deterioration. This wider window resulted in more infants reaching room air and thus being assigned the outcome “no BPD.” Thus, the number assigned the clinical definition of BPD is greater than the sum of those on positive pressure support, oxygen, and failed physiologic challenge, as shown in Fig 1. The difference is clinically relevant, because the overall goal of the definition is to identify infants with ongoing pulmonary morbidity rather than those with transient deterioration.
The difference in BPD rates determined by the clinical definition and those identified by the physiologic definition, which considers both oxygen administration and oxygen saturation, suggests that differences in oxygen-delivery practices and oxygen-saturation goals among centers contribute to differences in BPD rates. The physiologic definition minimizes the impact of the variation of such practices, yet even with this improved definition, significant differences in BPD rates between centers remained. This suggests that other treatment variables may also contribute to the variation seen. One variable that clearly impacted the rates of BPD seen in this study was the use of oxygen for extrapulmonary indications such as the treatment of apnea and bradycardia and retinopathy of prematurity. An additional variable is the widespread use of nasal cannula for oxygen delivery. Little is known about optimal nasal cannula oxygen delivery, and this gap in knowledge produces marked variations among centers in prescriptions for flow and concentration that may directly influence the duration of oxygen therapy.
The magnitude of the change in the definition of BPD is clinically important. Indeed, it is similar to the effect size seen in clinical trials of therapeutic modalities such as vitamin A that demonstrated a 7% reduction in BPD.16 The use of the physiologic definition minimizes the variation in the diagnosis of BPD among centers. Such variation may have masked or minimized efficacious effects of therapies tested in clinical trials in the past.
In this evaluation, the test proved to be safe with only transient desaturation events detected. In fact, one could argue that neonates monitored under this protocol with continuous direct nursing observation were safer than infants who are weaned in routine clinical practice when direct observation is performed intermittently. The test was required in only 14% of this cohort of 1598 infants and thus is feasible to perform within the constructs of a clinical trial.
A limitation of the current study may be the selection of the saturation cutoff point of <90% as the failure definition. We choose this cutoff value after surveying the principal investigators in the National Institute of Child Health and Human Development Neonatal Research Network to ascertain a safe and acceptable lowest saturation value used at their center. We believe that the cutoff point has merit on several levels. First, there is no consensus on an acceptable saturation for infants with BPD. Second, the saturation values used in this trial are similar to those used in 2 other large trials assessing the impact of saturation. The STOP-ROP trial randomized patients with prethreshold retinopathy of prematurity to be studied at saturation levels of 89% to 94% or 96% to 99%. The study demonstrated no meaningful differences in outcome between the groups managed with saturations in these ranges except for a higher rate of adverse pulmonary events in the group managed with higher saturations.10 In the BOOST trial, infants born <30 weeks of age who were oxygen-dependent at 32 weeks were randomized to ranges of 91% to 94% or 95% to 98% and demonstrated no difference in growth but did show a higher rate of oxygen dependence at 36 weeks' postmenstrual age (64% vs 46%).11 Third, at least 1 study has suggested that infants managed with lower saturations in the range of 70% to 90% had improved outcomes.12 The analysis of the impact of the saturation cutoff point indicates that there is very little impact on the passing rate with a cutoff varying between 88% and 92% saturation.
A second potential limitation is that the observation period in room air is relatively brief. For the purposes of this study, in which all infants were returned to their prior oxygen treatment, the period was sufficient to identify infants who desaturate in room air. The test correlated with more meaningful measures of later pulmonary morbidity such as length of oxygen use, hospital stay, and discharge home in oxygen. An additional possible limitation is that a large number of neonates were receiving treatment with medications including methylxanthines, diuretics, and bronchodilators at the time of the physiologic definition, and the rates of medication use varied between centers. Although there are no existing data that such medication use can reduce the amount of supplemental oxygen needed, it is possible that this might occur and influence the test results. It is reassuring that the frequency of medication use was similar between those who passed and those who failed the oxygen-reduction test.
The physiologic definition creates a window of evaluation in which all infants are assessed similarly. This definition will minimize practice variation between centers for the short period of the study. Minimizing variation may allow trials to more accurately detect real treatment differences that were masked previously by such variation. The physiologic definition is not intended to be used as a guide for clinical management. The infants studied, including those who passed the challenge, were all returned to their baseline oxygen and managed at the discretion of their attending physician. It was not the intent of this study to evaluate the safety of allowing infants who passed to remain in room air, and thus we can not comment on the safety of that practice nor would we recommend it based on these data. Additional work is needed to determine if this timed oxygen reduction would be of benefit in weaning oxygen in clinical practice. We would strongly encourage clinicians to await additional data before instituting this weaning protocol in clinical practice. The relationship of the BPD outcome based on a physiologic diagnosis to long-term pulmonary function or neurodevelopmental outcome is also unknown. Evaluation of these infants at 18 months of age is in progress.
The physiologic definition proposed here is an improvement on the current definition of oxygen use at 36 weeks' postmenstrual age. However, ultimately better tools are needed to identify these infants. Such tools may include biomarkers of lung injury or simplified tests of infant pulmonary function that may ultimately replace this simple clinical standard. This study confirms that the physiologic definition is safe, results in an overall reduction in the rate of BPD, and modestly decreases variation in the diagnosis of BPD among centers. Improving the precision of the diagnosis of BPD will facilitate the measurement of BPD outcome in clinical trials and the comparison of outcomes between and within centers over time.
M. C. Walsh, MD, MS, and A. A. Fanaroff, MD (Case Western Reserve University, Cleveland, OH); A. H. Jobe, MD, PhD (University of Cincinnati, Cincinnati, OH); R. Higgins, MD (National Institute of Child Health and Human Development, Bethesda, MD); N. Finer, MD (University of California, San Diego, CA); and K. Poole, PhD (Research Triangle Institute, Research Triangle Park, NC).
N. Kazzi, MD, K. Hayes-Hart, RN, M. Betts, RRT, S. Shankaran, MD, principal investigator, and G. Muran, RN (Wayne State University, Detroit, MI); A. Laptook, MD, M. Martin, RN, and J. Allen, RRT (University of Texas Southwestern, Dallas, TX); W. A. Engle, MD, L. Miller, RN, R. Hooper, RRT, and J. Lemons, MD, principal investigator (Indiana University, Indianapolis, IN); W. Rhine, MD, C. Kibler, RN, J. Parker, RRT, D. Stevenson, MD, principal investigator, and B. Ball, BS (Stanford University, Palo Alto, CA); M. Rasmussen, MD, M. Grabarczyk, BSN, C. Joseph, RRT, and K. Arnell, BSN (Sharp Mary Birch Hospital for Women, San Diego, CA); G. Heldt, MD, R. Bridge, RN, J. Goodmar, RRT, N. Finer, MD, and C. Henderson, RCP (University of California, San Diego, CA); and S. Buchter, MD, M. Berry, RN, I. Seabrook, RRT, B. Stoll, MD, principal investigator, and E. Hale, RN (Emory University, Atlanta, GA).
S. Duara, MD, and R. Everette, RN (University of Miami, Miami, FL); W. Carlo, MD, and M. Collins, RN (University of Alabama, Birmingham, AL); and W. Oh, MD, and A. Hensman, RN (Brown University, Providence, RI).
M. T. O'Shea, MD, MPH, and N. Peters, RN (Wake Forest University, Winston-Salem, NC); J. Tyson, MD, MPH, and G. McDavid, RN (University of Texas, Houston, TX); A. A. Fanaroff, MD, M. C. Walsh, MD, MS, and N. Newman, RN (Case Western Reserve University, Cleveland, OH); D. Phelps, MD, and L. Reubens, RN (University of Rochester, Rochester, NY); R. A. Ehrenkranz, MD, and P. Gettner, RN (Yale University, New Haven, CT); C. Michael Cotten, MD, and K. Auten, RN (Duke University, Durham, NC); and E. Donovan, MD, and C. Grisby, RN (University of Cincinnati, Cincinnati, OH).
Qing Yao, PhD, and Ken Poole, PhD (Research Triangle Institute, Research Triangle Park, NC).
This work was supported in part by Specialized Clinical Investigator Award HD21364-15SI from the National Institutes of Child Health and Development. Neonatal Network centers were funded under awards from the National Institute of Child Health and Development (U10 HD21397, U10 HD34216, U10 HD27853, M01 RR 08084, U10 HD27871, M01 RR 06022, U10 HD21364, U10 HD40461, U10 HD40689, U10 HD27856, M01 RR 00750, U10 HD27904, U10 HD40498, U10 HD40521, U01 HD36790, U10 HD21385, and M01 RR 00070).
We thank the infants and their families who participated in the study. We also acknowledge the invaluable assistance of the research nurses of the National Institutes of Child Health and Development Neonatal Research Network and the nursing and medical staff in the participating centers, without whom the study could not have been performed. We also thank Wally Carlo, MD, Krisa VanMeurs, MD, and David Stevenson, MD, for critical review and comments.
- ↵Lemons JA, Bauer CR, Oh W, et al. Very low birth weight outcomes of the National Institute of Child Health and Human Development Neonatal Research Network, January 1995 through December 1996. Pediatrics.2001;107(1) . Available at: www.pediatrics.org/cgi/content/full/107/1/e1
- ↵Shennan AT, Dunn MS, Ohlsson A, Lennox K, Hoskins EM. Abnormal pulmonary outcomes in premature infants: prediction from oxygen requirements in the neonatal period. Pediatrics.1988;82 :527– 532
- Garg M, Kurzner SI, Bautista DB, Keens TG. Clinically unsuspected hypoxia during sleep and feeding in infants with bronchopulmonary dysplasia. Pediatrics.1988;81 :635– 642
- Moyer-Mileur LJ, Nielson DW, Pfeffer KD, Witte MK, Chapman DL. Eliminating sleep-associated hypoxemia improves growth in infants with bronchopulmonary dysplasia. Pediatrics.1996;98 :779– 783
- Singer L, Martin RJ, Hawkins SW, et al. Oxygen desaturation complicates feeding in infants with bronchopulmonary dysplasia after discharge. Pediatrics.1992;90 :380– 384
- ↵STOP-ROP Multicenter Study Group. Supplemental Therapeutic Oxygen for Prethreshold Retinopathy of Prematurity (STOP-ROP), a randomized, controlled trial. I: primary outcomes. Pediatrics.2000;105 :295– 310
- ↵Askie LM, Henderson-Smart DJ, Irwig L, Simpson JM. The effect of differing oxygen saturation targeting ranges on long term growth and development of extremely preterm, oxygen dependent infants: the BOOST Trial [abstract]. Pediatr Res.2003;51 :2203A
- ↵Tin W, Milligan DW, Pennefather P, Hey E. Pulse oximetry, severe retinopathy, and outcome at one year in babies of less than 28 weeks gestation. Arch Dis Child Fetal Neonatal Ed.2001;84 :F106– F110
- ↵Woolson R. Statistical Methods for the Analysis of Biomedical Data. New York, NY: John Wiley & Sons; 1987
- Copyright © 2004 by the American Academy of Pediatrics