Outcomes of Extremely Low Birth Weight (<1 kg) and Extremely Low Gestational Age (<28 Weeks) Infants With Bronchopulmonary Dysplasia: Effects of Practice Changes in 2000 to 2003
OBJECTIVE. The goal was to evaluate whether changes in neonatal intensive care have improved outcomes for children with bronchopulmonary dysplasia (oxygen dependence at corrected age of 36 weeks).
METHODS. We compared outcomes of extremely low birth weight (<1 kg) and extremely low gestational age (<28 weeks) infants with bronchopulmonary dysplasia between 2 periods (period I, 1996–1999: extremely low birth weight, n = 122; extremely low gestational age, n = 118; period II, 2000–2003: extremely low birth weight, n = 109; extremely low gestational age, n = 107).
RESULTS. For both groups, significant practice changes between period I and period II included increased prenatal and decreased postnatal steroid therapy and increased surfactant therapy, indomethacin therapy, and patent ductus arteriosus ligation. Significant morbidity changes included decreased rates of severe cranial ultrasound abnormalities and increased rates of ventilator dependence. Rates of bronchopulmonary dysplasia did not change (52% vs 53%). Follow-up evaluation revealed significantly lower rates of neurosensory abnormalities during period II (extremely low birth weight: 29% vs 16%; extremely low gestational age: 31% vs 16%). There were no changes in rates of Mental Developmental Index scores of <70 (extremely low birth weight: 42% vs 42%; extremely low gestational age: 37% vs 45%) or overall developmental impairment (extremely low birth weight: 51% vs 49%; extremely low gestational age: 50% vs 51%). For the extremely low gestational age group, predictors of neurosensory abnormalities were severe cranial ultrasound abnormality and postnatal steroid therapy. Predictors of overall impairment included severe cranial ultrasound abnormalities, ventilator dependence, postnatal steroid therapy, and patent ductus arteriosus ligation. For the extremely low birth weight group, the only predictor of neurosensory abnormalities was severe cranial ultrasound abnormality. Predictors of overall impairment included multiple birth, ventilator dependence, and severe cranial ultrasound abnormalities.
CONCLUSIONS. Neurosensory outcomes of infants with bronchopulmonary dysplasia improved during 2000 to 2003 but overall neurodevelopmental outcomes did not change.
The rates of bronchopulmonary dysplasia (BPD), the most common chronic condition affecting preterm infants, increased concomitantly with the improved survival of smaller, less mature infants in the 1990s.1 BPD is a major predictor of poor neurodevelopmental outcomes, including increased rates of cerebral palsy and other neurosensory and motor abnormalities,2–8 poor cognitive outcomes during early childhood and school age,5,9,10 and impaired long-term respiratory function.11,12 In addition, chronic health problems associated with BPD have a major impact on the daily life of families that persists beyond the neonatal period.13
Although BPD clearly is associated with poorer long-term outcomes, relative to unaffected preterm and term control subjects, recent changes in NICU practices, including increased prenatal steroid use14 and practices to decrease the rates of nosocomial infections,15 might have modified the outcomes for children with BPD. Furthermore, although the use of postnatal corticosteroid therapy to prevent or to treat BPD was high in the 1990s, this has since declined dramatically, after reports of associated long-term sequelae, including cerebral palsy, neuromotor dysfunction, and developmental delay,16–20 and the joint statement of the American Academy of Pediatrics and the Canadian Paediatric Society that postnatal corticosteroid therapy is not recommended without additional randomized trials.21 Reduction in postnatal steroid use thus might have improved long-term outcomes for children with BPD.
We reported recently on improved outcomes for the total cohort of extremely low birth weight (ELBW) infants admitted to our NICU since 2000.22 In the present study we sought to determine specifically whether the early childhood outcomes of children with BPD since the year 2000 have improved, compared with those of children with BPD born before that time. We hypothesized that the early childhood outcomes of children with BPD would have improved as a result of the new neonatal treatment protocols. We also sought to examine whether outcomes differed when assessed according to gestational age or birth weight subgroups.
The population included infants admitted to the NICU at Rainbow Babies and Children's Hospital during the 4 year periods of 1996 to 1999 (period I) and 2000 to 2003 (period II). We examined the outcomes of ELBW infants selected according to the traditional birth weight-based cutoff point of <1 kg (ELBW group) and according to the gestational age-based criterion of <28 weeks of gestation (extremely low gestational age [ELGA] group). Selection of a gestational age-based group allowed us to exclude infants with birth weights of <1 kg who had a gestational age of ≥28 weeks because of intrauterine growth failure. Potential bias associated with considering only a <1-kg birth weight cutoff group could thus be examined. Infants with birth weights of <500 g were excluded.
ELBW Group (Birth Weight of 500–999 g)
During periods I and II, 302 and 299 ELBW infants, respectively, were admitted to the NICU, of whom 15 (5%) and 12 (4%) were excluded because of major congenital malformations. Of the remaining infants, 230 (80%) in period I and 228 (79%) in period II survived to corrected age (CA) (postmenstrual age plus postnatal age) of 20 months, of whom 126 (55%) and 117 (52%), respectively, had BPD, defined as oxygen dependence at CA of 36 weeks.23 This definition includes the moderate and severe disease categories of BPD, according to the National Institutes of Health consensus definition.24 Of the infants with BPD, 122 (97%) in period I and 109 (93%) in period II were monitored to CA of 20 months and underwent a complete neurodevelopmental assessment, including evaluation with the Bayley Scales of Infant Development II.25
ELGA Group (Gestational Age of <28 Weeks)
During periods I and II, 315 and 298 ELGA infants, respectively, were admitted to the NICU, of whom 14 (4%) and 11 (4%) were excluded because of major congenital malformations. Of the remaining infants, 242 (80%) in period I and 228 (79%) in period II survived to CA of 20 months, of whom 126 (52%) and 120 (53%), respectively, had BPD. Of the infants with BPD, 118 (94%) in period I and 107 (89%) in period II were monitored to CA of 20 months and underwent a complete neurodevelopmental assessment.
One infant in period II, who was in both the ELBW and ELGA groups, was excluded because the infant had been transferred to another hospital and the total duration of oxygen dependence was not known. Four infants in period I and 8 in period II died at CA of >36 weeks but <20 months. The cause of death for the 4 infants who died in period I, all of whom were in both the ELBW and ELGA groups, was related to their severe BPD. One infant also had superimposed sepsis. Seven of the 8 infants who died in period II were in both the ELBW and ELGA groups. Four died as a result of severe BPD, 1 as a result of late-onset necrotizing enterocolitis, and 1 as a result of disseminated intravascular coagulation. The cause of death of 1 infant who died after neonatal discharge was not known. One additional ELGA infant died as a result of severe BPD.
Infants were monitored until CA of 18 to 20 months, according to the protocol of the ongoing follow-up program at Rainbow Babies and Children's Hospital. They received a complete physical and neurologic examination. Major neurosensory abnormalities were defined as cerebral palsy (spastic diplegia, hemiplegia, triplegia, or quadriplegia),26 persistent hypotonia or hypertonia, shunt-dependent hydrocephalus without neurologic abnormalities, and unilateral or bilateral blindness or deafness requiring a hearing aid. Cerebral palsy was defined as a disorder of movement and posture attributable to a defect or lesion of the immature brain.26 Hypotonia and hypertonia were included as major neurologic abnormalities because these conditions are often considered to represent variants of cerebral palsy. Shunt-dependent hydrocephalus without neurologic abnormalities was also considered a neurosensory abnormality.27 Cognitive and motor development was assessed at CA of 18 to 20 months by using the Bayley Scales of Infant Development Mental Developmental Index (MDI) and Psychomotor Developmental Index (PDI).25 The developmental specialist was blinded with respect to perinatal complications and neonatal history. Children with neurosensory abnormalities and/or a MDI score of <70 were considered to have neurodevelopmental impairment.
Perinatal data were extracted from the hospital charts at the time of discharge from the neonatal nursery. Demographic data included maternal age, marital status, education level, and race. Birth data included birth weight and gestational age, gender of the infant, delivery method, and multiple birth status. Gestational age was determined from the date of the mother's last menstrual period and was confirmed with obstetric measures and abdominal ultrasonography in most cases. Neonatal morbidities included patent ductus arteriosus (PDA), confirmed with echocardiography; sepsis, defined as a positive blood culture in the presence of clinical signs of sepsis; and necrotizing enterocolitis, according to the definition described by Bell et al.28 Periventricular hemorrhage was categorized as grades I through IV.29 A severe cranial ultrasound abnormality was defined as the presence of grade III or IV periventricular hemorrhage, periventricular leukomalacia, or persistent ventriculomegaly at the time of discharge from the hospital. Perinatal therapies examined included prenatal and postnatal steroid therapy, surfactant therapy, indomethacin therapy, and PDA ligation. The study was approved by the institutional review board of the University Hospitals of Cleveland, and signed parental consent was obtained for all study participants.
Comparisons of the outcomes of children with BPD between periods I and II were undertaken separately for the ELBW and ELGA infants. The unpaired t test or Wilcoxon rank sum test was used to compare continuous variables, the χ2 test or Fischer's exact test was used to compare categorical variables, and the Cochran-Armitage trend test was used to compare distributions of ordinal categorical variables.30 Pearson or bipoint biserial correlations were performed between potential neonatal risk factors and neurodevelopmental outcomes, including cerebral palsy, MDI of <70, mean MDI score, neurosensory abnormalities, and neurodevelopmental impairment. Postnatal steroid therapy was evaluated as whether the infant received steroid therapy (yes or no) and according to the number of days received, as a surrogate for dose effect. Multivariate analyses were conducted to examine the effects of significant (P < .05) changes in therapies between the 2 periods, sociodemographic factors (maternal education less than high school and race), gender of the child, and time period (period I versus period II).
Sociodemographic, Obstetric, and Infant Birth Data
Sociodemographic measures, including maternal age, race, education, and marital status, did not differ between periods for either the ELBW or ELGA group (Table 1). There were no significant differences between periods in either group with respect to maternal obstetric factors or infant birth data other than an increase in prenatal steroid therapy and the rates of multiple births during period II. Birth weight, gestational age, gender, and mean weight z score (a measure of intrauterine growth)31 did not change.
Neonatal Therapies and Morbidities
Rates of periventricular hemorrhage decreased significantly from period I to period II for both the ELBW and ELGA groups, resulting in a decrease in the overall rates of severe cranial ultrasound abnormalities. The rates of periventricular leukomalacia and ventricular dilation did not change, and neither did the rates of other morbidities, including sepsis and necrotizing enterocolitis (Table 2). Surfactant therapy use increased significantly in period II for both groups of infants, and the rate of postnatal steroid therapy and the median days of treatment for infants who received postnatal steroid therapy showed significant decreases. Total days of ventilator dependence (an indicator of BPD severity), the use of continuous positive airway pressure therapy, and total days of oxygen dependence increased for both groups in period II. Indomethacin therapy and PDA ligation also increased across the 2 periods for both groups, although the rates of PDA did not change.
Neurosensory and Developmental Outcomes at CA of 20 Months
In both the ELBW and ELGA groups, children lost to follow-up monitoring did not differ from those monitored in terms of maternal education, race, gender of the infant, and gestational age. Children who did not have a MDI score were considered lost to follow-up monitoring. The birth weights of infants monitored versus those not monitored in the ELGA group did not differ during period I; in period II, however, the 107 infants monitored had a lower birth weight than did the 13 infants not monitored (748.3 ± 143.4 g vs 912.7 ± 195.3 g; P < .05). Birth weights in the ELBW group did not differ in either period.
The rates of cerebral palsy at CA of 20 months, shunt-dependent hydrocephalus, and blindness did not differ significantly between the 2 periods for either group, although there was a nonsignificant decline in the rates of cerebral palsy from period I to period II. Both the ELBW and ELGA groups showed significant decreases in the rates of deafness and overall neurosensory abnormalities in period II (Table 3). There were no differences between periods in the mean Bayley MDI scores or in the rates of subnormal (<70) MDI scores (Table 4). This finding persisted even when children with neurosensory abnormalities were excluded from the analyses. The mean PDI score was significantly higher in period II for both groups, although there were no differences in the rates of PDI scores of <70. The PDI scores did not include the entire population, because some children with cerebral palsy could not be tested on the Bayley motor scale. In the ELBW group, the PDI was not scored for 6 children (5%) in period I and 5 (5%) in period II; in the ELGA group, the PDI was not scored for 7 children (6%) in period I and 8 (7%) in period II. When the children with neurosensory abnormalities were excluded from the analyses, the PDI scores did not differ between periods in either group. Similarly, the rates of rehospitalization before CA of 20 months did not differ in either group. There were no significant changes in overall neurodevelopmental impairment (MDI of <70 and/or overall neurosensory abnormalities) in either group.
Of the 9 children who did not have a MDI score during period I and therefore were excluded from the analyses, 4 were not testable because of severe behavior problems and/or mental retardation and 5 were lost to follow-up monitoring. Of the 13 children who did not have a MDI score during period II, 4 were not testable, 7 were lost to follow-up monitoring, and 2 did not participate because of parental refusal. Therefore, the numbers of children who had severe developmental delays or behavior problems and therefore lacked full assessments did not differ between the 2 periods.
Effects of Changes in Neonatal Risk Factors and Therapies on Neonatal Outcomes
Multivariate analyses were performed to examine the effects of significant changes in neonatal risk factors and therapies, including multiple birth status, duration of ventilator dependence, severe cranial ultrasound abnormalities, indomethacin therapy, PDA ligation, and prenatal and postnatal steroid therapy, on outcomes. These factors were entered into the multivariate analysis together with the time period (period I versus period II) and maternal education, race, and gender. Because of limitations in the number of variables allowed for multivariate logistic regression analyses, based on the number of affected children, surfactant therapy (which did not predict neurodevelopment in the univariate analyses) was excluded. For similar reasons, race and gender were excluded in the analysis of neurosensory outcomes.
Results for the ELBW subgroups revealed the significant predictors of subnormal MDI (<70) to be duration of ventilator dependence and severe cranial ultrasound abnormalities (Table 5). The only predictor of neurosensory abnormalities was a severe cerebral ultrasound abnormality. Predictors of neurodevelopmental impairment included multiple birth, duration of ventilator dependence, and a severe cerebral ultrasound abnormality (Table 5). Results of the gestational age (ELGA) subgroup analyses were similar, with the exception that multiple births did not predict overall impairment, PDA ligation predicted neurodevelopmental impairment, and exposure to postnatal steroid therapy predicted both neurosensory impairment and neurodevelopmental impairment (Table 6). The duration of postnatal steroid therapy for children who received steroids was not predictive (data not shown).
We sought to examine the early childhood outcomes of ELBW and ELGA children with moderate to severe BPD after the therapeutic changes in neonatal intensive care that have occurred since the year 2000. These changes include increases in the use of surfactant therapy, prenatal steroid therapy, and indomethacin therapy and decreases in the use and duration of postnatal steroid therapy. These changes have been associated with decreases in the rates of periventricular hemorrhage but not the rates of BPD. Similar experiences have been reported by others.32,33 No other major changes in care were identified during the periods of study. Our results indicated that rates of neurosensory abnormalities among children with BPD decreased in period II. This was attributable to significant decreases in the rates of deafness and nonsignificant decreases in the rates of cerebral palsy; because the MDI scores did not change, however, neurodevelopmental outcomes at CA of 20 months remained fairly constant.
We recently reported on the changes in outcomes for the total cohort of ELBW infants born at our center during the years 2000 to 2002, compared with 1990 to 1999, regardless of BPD status.22 In this population, there were significant decreases in the rates of periventricular hemorrhage and severe cranial ultrasound abnormalities and also significant decreases in the rates of other neonatal morbidities, including sepsis, meningitis, and ventricular dilation. These were associated with a significant decrease in neurosensory abnormalities at CA of 20 months. Similar to our present results, there was no improvement in cognitive function.
Our findings of only a nonsignificant decrease in the rates of cerebral palsy in the subgroup of children with BPD are surprising, particularly considering the considerable decrease in postnatal steroid use. A meta-analysis of randomized, controlled trials of postnatal steroid therapy revealed that postnatal steroid therapy increased the risk of death or cerebral palsy significantly.34–36 However, Doyle et al37 reported that this occurred only for infants who had <35% risk of developing BPD; among infants for whom the risk exceeded 65%, steroid therapy reduced the rate of death or cerebral palsy. Because, by definition, our cohort all had moderate to severe BPD, this might explain our lack of a significant decrease in the rate of cerebral palsy, although there was a significant decrease in the overall rate of neurosensory impairment, which includes deafness. We are unable to explain the decrease in the rate of deafness during period II, because the treatment changes we identified are not known to be associated with hearing loss.7 It is possible that practices that we did not measure (such as changes in antibiotic or diuretic use) might have contributed to the decrease in the rate of hearing loss. Furthermore, we identified only severe hearing loss requiring a hearing aid.
Postnatal corticosteroid therapy has been shown to expedite weaning from mechanical ventilation and thus reduce the rates of BPD.34–36 The significant increase in the duration of ventilator dependence in period II in our populations might be attributable to the decreased rate and duration of postnatal steroid therapy since 2000, although our rates of BPD did not change. Because the duration of ventilator dependence may be considered an indicator of disease severity for infants with BPD, this increase in severity may explain why the outcomes of the children with BPD did not improve during period II. However, we lack other measures of the severity of BPD, such as pulmonary function test or lung biopsy results. Because postnatal steroid therapy may, in certain subgroups of infants, decrease the rates and severity of BPD,37 hydrocortisone therapy, which does not seem to have a deleterious effect on neurodevelopmental outcomes38,39 or on MRI findings,39 has been suggested to prevent or to treat BPD.40
Both the ELBW and ELGA groups showed significant decreases in rates of periventricular hemorrhage and overall severe cranial ultrasound abnormalities in period II. This is most likely attributable to the significant increase in both prenatal steroid and indomethacin therapy in period II. Prophylactic indomethacin therapy has been shown to reduce the incidence of periventricular hemorrhage41; however, it has not been associated with changes in neurodevelopmental outcomes, such as cognitive delay, cerebral palsy, blindness, or deafness.41,42
A possible explanation for the lack of improvement in cognitive or overall neurodevelopment outcomes in our population with BPD may be that poor outcomes for infants with BPD persist, regardless of treatment changes, because of the nature of the disease itself. O'Shea et al9 reported outcomes for children with BPD at 4 to 5 years of age, after exclusion of those with abnormal cranial ultrasound findings or neurologic abnormalities. Even with these potentially confounding factors removed, children with BPD continued to have poorer cognitive performance, compared with children without BPD. Because poor outcomes persist even in the absence of identified neurologic abnormalities, other disease factors may play a role. Infants receiving mechanical ventilation experience frequent transient episodes of oxygen desaturation and hypoxemia, even after the acute phase of respiratory failure has resolved.43 These frequent oxygen desaturations may contribute to poor neurodevelopmental outcomes, which persist despite the recently improved NICU practices. To our knowledge, there are no available data on the effects of repeated episodes of desaturation and hypoxemia on the outcomes of preterm infants. However with the more-recently developed pulse oximeters, quantifiable data should be available in the future. Additional factors that may contribute to poorer outcomes for children with BPD include the high rates of nosocomial sepsis, poor growth, and maternal separation associated with prolonged neonatal hospitalization and repeated later rehospitalizations.
Although research on the outcomes of preterm infants typically includes birth weight-defined populations, we chose also to consider an ELGA population, thus eliminating the bias associated with inclusion of ELBW infants with severe intrauterine growth restriction and exclusion of those born at <28 weeks of gestation with birth weights of >1 kg. Throughout our analyses, findings for the ELBW and ELGA groups were similar. This similarity in outcomes likely reflects the substantial overlap in the compositions of the 2 groups, with 213 (88%) of the total sample being included in both groups. In the ELBW group, there were 12 infants (10%) in period I and 6 infants (6%) in period II who had a gestational age of >28 weeks and thus were not included in the ELGA group. Similarly, the ELGA group had 8 infants (7%) in period I and 4 infants (4%) in period II with birth weights of >1000 g, who were not included in the ELBW group. Despite the overlap between groups, however, the gestational age subgroup, which was not biased by the inclusion of infants with growth failure born at ≥28 weeks of gestation, revealed a deleterious effect of postnatal steroid use on neurosensory and overall impairment, which was not evident in the birth weight-based (ELBW) group.
The strengths of our study include the high follow-up rates and detailed collection of perinatal risk factors for various outcomes. Weaknesses include the lack of a measure of neonatal pulmonary function and detailed assessment of hearing loss. We did not find a decrease in the rates of sepsis, but we do not have information on whether the numbers of episodes of infection decreased in period II. The follow-up period was relatively short. School age testing of cognitive function may reveal improved cognitive function, although rates of cerebral palsy remain fairly stable.44 Furthermore, our definition of BPD does not align with the recent National Institutes of Health consensus conference definition, which divides BPD into mild, moderate, and severe categories.24 These categories of BPD require data on the fraction of oxygen delivered to the infant, which was not available in our database. However, our definition of BPD as oxygen dependence at CA of 36 weeks overlaps with the moderate to severe categories of BPD in the consensus definition.45,46
Because the outcomes of children with BPD are poor regardless of the recent therapeutic changes, it will be important to focus attention on strategies that specifically prevent the development of this condition. Respirator use may induce barotrauma and the risk of BPD. Recent practices thus include more-frequent use of nasal continuous positive airway pressure, rather than endotracheal assisted ventilation. In fact, in 2004 to 2005, the rates of BPD for ELBW and ELGA infants in our unit decreased to 27% and 30%, respectively. The therapeutic changes since 2000 resulted in a decrease in the rate of periventricular hemorrhage; however, the rates of periventricular leukomalacia, a major determinant of severe ultrasound abnormalities and poor neurodevelopmental outcomes, have not changed. Therefore, research also needs to continue to minimize this type of brain injury.
This work was supported by grants from the National Institutes of Health (grants T35 HL082544, M01 RR00080, and HD21364).
Special thanks go to Angelia Williams, Alpher Torres, Bonnie Siner, RN, and Bonnie Tarantino for their assistance and to Dr Richard Martin for review of the manuscript.
- Accepted June 22, 2007.
- Address correspondence to Maureen Hack, MBChB, Division of Neonatology, Rainbow Babies and Children's Hospital, University Hospitals of Cleveland Case Medical Center, 11100 Euclid Ave, Cleveland, OH 44106-6010. E-mail:
The authors have indicated they have no financial relationships relevant to this article to disclose.
- ↵Schmidt B, Asztalos EV, Roberts RS, et al. Impact of bronchopulmonary dysplasia, brain injury, and severe retinopathy on the outcome of extremely low-birth-weight infants at 18 months: results from the trial of indomethacin prophylaxis in preterms. JAMA.2003;289 :1124– 1129
- ↵Singer L, Yamashita T, Lilien L, Collin M, Baley J. A longitudinal study of developmental outcome of infants with bronchopulmonary dysplasia and very low birth weight. Pediatrics.1997;100 :987– 993
- ↵Kilbride HW, Powers R, Wirtschafter DD, et al. Evaluation and development of potentially better practices to prevent neonatal nosocomial bacteremia. Pediatrics.2003;111(4) . Available at: www.pediatrics.org/cgi/content/full/111/4/e504
- Shinwell ES, Karplus M, Reich D, et al. Early postnatal dexamethasone treatment and increased incidence of cerebral palsy. Arch Dis Child Fetal Neonatal Ed.2000;83 :F177– F181
- O'Shea TM, Kothadia JM, Klinepeter KL, et al. Randomized placebo-controlled trial of a 42-day tapering course of dexamethasone to reduce the duration of ventilator dependency in very low birth weight infants: outcome of study participants at 1-year adjusted age. Pediatrics.1999;104 :15– 21
- Yeh TF, Lin YJ, Huang CC, et al. Early dexamethasone therapy in preterm infants: a follow-up study. Pediatrics.1998;101(5) . Available at: www.pediatrics.org/cgi/content/full/101/5/e7
- ↵American Academy of Pediatrics, Committee on Fetus and Newborn. Postnatal corticosteroids to treat or prevent chronic lung disease in preterm infants. Pediatrics.2002;109 :330– 338
- ↵Wilson-Costello D, Friedman H, Minich N, et al. Improved neurodevelopmental outcomes for extremely low birth weight infants in 2000–2002. Pediatrics.2007;119 :37– 45
- ↵Shennan AT, Dunn MS, Ohlsson A, Lennox K, Hoskins EM. Abnormal pulmonary outcomes in premature infants: prediction from oxygen requirement in the neonatal period. Pediatrics.1988;82 :527– 532
- ↵Bayley N. Bayley Scales of Infant Development. 2nd ed. San Antonio, TX: Psychological Corp; 1993
- ↵World Health Organization. International Classification of Impairments, Disabilties and Handicaps. Geneva, Switzerland: World Heath Organization; 1980
- ↵Agresti A. Categorical Data Analysis. New York, NY: John Wiley and Sons; 1990:100– 102
- ↵Kramer MS, Platt RW, Wen SW, et al. A new and improved population-based Canadian reference for birth weight for gestational age. Pediatrics.2001;108(2) . Available at: www.pediatrics.org/cgi/content/full/108/2/e35
- ↵Kaempf JW, Campbell B, Sklar RS, et al. Implementing potentially better practices to improve neonatal outcomes after reducing postnatal dexamethasone use in infants born between 501 and 1250 grams. Pediatrics.2003;111(4) . Available at: www.pediatrics.org/cgi/content/full/111/4/e534
- ↵Halliday HL, Ehrenkranz RA, Doyle LW. Early postnatal (<96 hours) corticosteroids for preventing chronic lung disease in preterm infants. Cochrane Database Syst Rev.2003;(1):CD001146
- Halliday HL, Ehrenkranz RA, Doyle LW. Moderately early (7–14 days) postnatal corticosteroids for preventing chronic lung disease in preterm infants. Cochrane Database Syst Rev.2003;(1):CD001144
- ↵Halliday HL, Ehrenkranz RA, Doyle LW. Delayed (>3 weeks) postnatal corticosteroids for chronic lung disease in preterm infants. Cochrane Database Syst Rev.2003;(1):CD001145
- ↵Doyle LW, Halliday HL, Ehrenkranz RA, Davis PG, Sinclair JC. Impact of postnatal systemic corticosteroids on mortality and cerebral palsy in preterm infants: effect modification by risk for chronic lung disease. Pediatrics.2005;115 :655– 661
- ↵Watterberg KL, Shaffer ML, Mishefske MJ, et al. Growth and neurodevelopmental outcomes after early low-dose hydrocortisone treatment in extremely low birth weight infants. Pediatrics.2007;120 :40– 48
- ↵Ment LR, Oh W, Ehrenkranz RA, et al. Low-dose indomethacin and prevention of intraventricular hemorrhage: a multicenter randomized trial. Pediatrics.1994;93 :543– 550
- ↵Fowlie PW, Davis PG. Prophylactic intravenous indomethacin for preventing mortality and morbidity in preterm infants. Cochrane Database Syst Rev.2002;(3):CD000174
- ↵Hack M, Taylor HG, Drotar D, et al. Poor predictive validity of the Bayley Scales of Infant Development for cognitive function of extremely low birth weight children at school age. Pediatrics.2005;116 :333– 341
- ↵Ehrenkranz RA, Walsh MC, Vohr BR, et al. Validation of the National Institutes of Health consensus definition of bronchopulmonary dysplasia. Pediatrics.2005;116 :1353– 1360
- Copyright © 2008 by the American Academy of Pediatrics