Postnatal Corticosteroids for Preterm Infants — Do What We Say, Not What We Do
http://www.100md.com
《新英格兰医药杂志》
Physicians caring for pregnant women and their preterm infants have used corticosteroids since Liggins and Howie reported in 1972 that prenatal betamethasone decreased the incidence of the respiratory distress syndrome and increased the survival of preterm infants.1 This infatuation with corticosteroids is not unique to perinatal medicine; corticosteroids have been used for almost every human disease for which there is no effective therapy. Despite an astounding number of randomized, controlled trials of corticosteroids in perinatal medicine, we remain uncertain about their appropriate use. The Cochrane Review of 18 trials of prenatal corticosteroids demonstrated compelling benefits for infants delivered before 32 to 34 weeks of gestation.2 The main controversy over the use of prenatal corticosteroids is the risk–benefit ratio for various retreatment strategies if preterm delivery does not occur within one week after the initial corticosteroid treatment.
The three Cochrane meta-analyses of the postnatal use of corticosteroids to prevent or treat the syndrome of chronic lung injury that occurs in preterm infants (called bronchopulmonary dysplasia) include 39 trials.3,4,5 Nevertheless, the role of corticosteroids for the treatment of preterm infants remains poorly defined. The basic principles of drug evaluation are first to evaluate low doses used for short treatment intervals and then, if the drug is found to be safe, to evaluate increases in the dose or the duration of treatment. In contrast, the initial trials of corticosteroids for bronchopulmonary dysplasia used dexamethasone at a high dose for 42 days, and subsequent trials over the past 20 years have gradually evaluated lower doses and a shorter regimen.
The choice of dexamethasone also may have been unfortunate. Although dexamethasone and betamethasone are stereoisomers, they seem to have different effects when used before birth. Prenatal use of dexamethasone has been associated with an increase in periventricular leukomalacia, whereas betamethasone has not.6 Betamethasone seems to be more effective than dexamethasone for reducing the risk of complications of preterm birth. Although the two corticosteroids have a similar affinity for the nuclear glucocorticoid receptor, their nonnuclear effects differ strikingly. Watterberg has argued persuasively that the kinder and gentler choice of corticosteroid for the postnatal treatment of preterm infants is hydrocortisone.7
Given that the trials randomly assigned infants to what were probably the wrong doses of the wrong corticosteroid for the wrong duration of treatment, what do the meta-analyses tell us? Halliday et al. grouped the trials into three meta-analyses on the basis of the time at which corticosteroid therapy was started.3,4,5 Treatment within 96 hours after birth to prevent bronchopulmonary dysplasia was evaluated in 20 trials, in which a total of 3072 preterm infants were randomly assigned to treatment groups.5 Corticosteroid treatment decreased the incidence of bronchopulmonary dysplasia and facilitated extubation. However, short-term adverse effects (gastrointestinal bleeding and perforation, hyperglycemia, hypertension, hypertrophic cardiomyopathy, and growth failure), together with increased risks of abnormal findings on neurologic examinations and cerebral palsy at 18 months, suggest an adverse risk–benefit ratio.
Seven trials, involving a total of 669 infants, evaluated corticosteroid treatment begun at 7 to 14 days of age to prevent or treat bronchopulmonary dysplasia.3 This strategy decreased mortality and the incidence of bronchopulmonary dysplasia but with the same short-term adverse effects as those associated with early corticosteroid treatment. Very little information on the long-term outcome is available for this treatment strategy. Nine trials, involving a total of 562 infants, evaluated corticosteroid treatment of bronchopulmonary dysplasia beginning after three weeks of age.4 The corticosteroid-treated infants were extubated earlier and had less need for open-label corticosteroid treatment, with a lower rate of death or bronchopulmonary dysplasia at 36 weeks. A trend toward an increased risk of cerebral palsy was offset by a decreased risk of early death in the corticosteroid-treated infants.
A major conclusion of these analyses was that high-quality follow-up information was critically needed because of concern about neurodevelopmental outcomes. Follow-up data were not available for many of the infants and were often of low quality. Furthermore, many trials permitted the open-label use of corticosteroids, which means that many of the infants in the control group received corticosteroids.
Despite the decreased mortality with the treatments begun at seven days of age or later, the American Academy of Pediatrics and the Canadian Paediatric Society published guidelines in 2002 stating that "the routine use of systemic dexamethasone for the prevention or treatment of chronic lung disease in infants with very low birth weight is not recommended."8 This conclusion was based primarily on the concern about adverse neurodevelopmental effects. In this issue of the Journal, Yeh et al.9 report the outcomes at school age for infants who had been randomly assigned to receive dexamethasone (0.5 mg per kilogram of body weight per day for one week) or placebo, with the first dose given within 12 hours after birth and tapering of the dose over a period of three weeks. They report clear adverse effects on growth, IQ, and motor performance. The quality of the assessments is high, and the authors' conclusion — that a 28-day regimen of high-dose corticosteroid therapy begun within 12 hours after birth should not be recommended — cannot be refuted. These results strengthen the findings on meta-analysis of this early treatment strategy.5
Despite the recent statements by the American Academy of Pediatrics and the Canadian Paediatric Society and other recent pleas to stop using postnatal corticosteroids,8,10 clinicians continue to use them. The Vermont Oxford Network of 408 centers reported that 23 percent of 14,321 infants with a birth weight between 501 g and 1000 g received postnatal corticosteroids in 2002 (Soll R: personal communication). Similarly, the Neonatal Research Network data base of the National Institute of Child Health and Human Development indicates that in 2002, among infants in the same weight range, 19 percent of those who survived for more than 12 hours (range for all centers reporting information, 5 to 49 percent) received postnatal corticosteroid therapy.
Do what we say, not what we do. But what are clinicians actually doing? There is no solid information, but clinicians are probably not using corticosteroids primarily as early treatment. It is likely that corticosteroids are primarily used to treat infants in whom bronchopulmonary dysplasia is developing, and one hopes that a low-dose, short regimen is used. For better or worse, clinicians continue to use corticosteroids because they work, as demonstrated by the short-term pulmonary benefits identified in the meta-analyses.3,4 The report on outcomes at school age by Yeh et al. should end any remaining enthusiasm for early treatment with high-dose dexamethasone.9
Where do we go from here with postnatal corticosteroid therapy for bronchopulmonary dysplasia? The use of a lower dose of dexamethasone (0.15 mg per kilogram per day for three days, followed by tapering of the dose over seven days) with concurrent use of indomethacin resulted in an increased risk of gastrointestinal perforation.11 The 18-month follow-up of infants in this study showed no differences in neurodevelopmental outcomes between the dexamethasone group and the control group.12 However, this result is tempered by the observation that exposure to dexamethasone or open-label use of corticosteroids was associated with abnormal findings on neurologic examination. Although low-dose hydrocortisone seemed to decrease bronchopulmonary dysplasia in a pilot trial,13 a multicenter trial of hydrocortisone was recently stopped because of an increased risk of gastrointestinal perforation (Watterberg K: personal communication). However, in both these trials, corticosteroid therapy was started soon after birth.
If additional trials are conducted, they should target the infants who are currently receiving corticosteroids — those who are dependent on mechanical ventilation at two to three weeks of age and who have progressive lung injury. First, we must understand the pharmacokinetics of several corticosteroids in this age group, and then we should design trials that use the lowest possible effective doses for the shortest period. Neurodevelopmental complications on follow-up constitute the essential safety issue. Since clinicians will not stop using corticosteroids, we need to understand how to use them safely, if that is possible.
Source Information
From the Division of Pulmonary Biology and Neonatology, Cincinnati Children's Hospital Medical Center, Cincinnati.
References
Liggins GC, Howie RN. A controlled trial of antepartum glucocorticoid treatment for prevention of respiratory distress syndrome in premature infants. Pediatrics 1972;50:515-525.
Crowley P. Prophylactic corticosteroids for preterm birth. Cochrane Database Syst Rev 2000;2:CD000065-CD000065.
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-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-CD001145.
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-CD001146.
Baud O, Foix-L'Helias L, Kaminski M, et al. Antenatal glucocorticoid treatment and cystic periventricular leukomalacia in very premature infants. N Engl J Med 1999;341:1190-1196.
Watterberg KL. Adrenocortical function and dysfunction in the fetus and neonate. Semin Neonatol 2004;9:13-21.
Committee on Fetus and Newborn. Postnatal corticosteroids to treat or prevent chronic lung disease in preterm infants. Pediatrics 2002;109:330-338.
Yeh TF, Lin YJ, Lin HC, et al. Outcomes at school age after postnatal dexamethasone therapy for lung disease of prematurity. N Engl J Med 2004;350:1304-1313.
Barrington KJ. The adverse neuro-developmental effects of postnatal steroids in the preterm infant: a systematic review of RCTs. BMC Pediatr 2001;1:1-1.
Stark AR, Carlo WA, Tyson JE, et al. Adverse effects of early dexamethasone treatment in extremely-low-birth-weight infants. N Engl J Med 2001;344:95-101.
Stark AR, Carlo W, Vohr BR, et al. Neurodevelopmental outcome and growth at 18-22 months in infants treated with early dexamethasone. Pediatr Res 2001;49:Suppl:388A-388A. abstract.
Watterberg KL, Gerdes JS, Gifford KL, Lin HM. Prophylaxis against early adrenal insufficiency to prevent chronic lung disease in premature infants. Pediatrics 1999;104:1258-1263.
Related Letters:
Postnatal Dexamethasone for Lung Disease of Prematurity
Lommatzsch M., Klotz J., Virchow J. C. Jr., Watchko J. F., Brozanski B. S., Gordon P. V., Yeh T. F., Lin H. C., Huang C. C., Jobe A. H.(Alan H. Jobe, Ph.D.)
The three Cochrane meta-analyses of the postnatal use of corticosteroids to prevent or treat the syndrome of chronic lung injury that occurs in preterm infants (called bronchopulmonary dysplasia) include 39 trials.3,4,5 Nevertheless, the role of corticosteroids for the treatment of preterm infants remains poorly defined. The basic principles of drug evaluation are first to evaluate low doses used for short treatment intervals and then, if the drug is found to be safe, to evaluate increases in the dose or the duration of treatment. In contrast, the initial trials of corticosteroids for bronchopulmonary dysplasia used dexamethasone at a high dose for 42 days, and subsequent trials over the past 20 years have gradually evaluated lower doses and a shorter regimen.
The choice of dexamethasone also may have been unfortunate. Although dexamethasone and betamethasone are stereoisomers, they seem to have different effects when used before birth. Prenatal use of dexamethasone has been associated with an increase in periventricular leukomalacia, whereas betamethasone has not.6 Betamethasone seems to be more effective than dexamethasone for reducing the risk of complications of preterm birth. Although the two corticosteroids have a similar affinity for the nuclear glucocorticoid receptor, their nonnuclear effects differ strikingly. Watterberg has argued persuasively that the kinder and gentler choice of corticosteroid for the postnatal treatment of preterm infants is hydrocortisone.7
Given that the trials randomly assigned infants to what were probably the wrong doses of the wrong corticosteroid for the wrong duration of treatment, what do the meta-analyses tell us? Halliday et al. grouped the trials into three meta-analyses on the basis of the time at which corticosteroid therapy was started.3,4,5 Treatment within 96 hours after birth to prevent bronchopulmonary dysplasia was evaluated in 20 trials, in which a total of 3072 preterm infants were randomly assigned to treatment groups.5 Corticosteroid treatment decreased the incidence of bronchopulmonary dysplasia and facilitated extubation. However, short-term adverse effects (gastrointestinal bleeding and perforation, hyperglycemia, hypertension, hypertrophic cardiomyopathy, and growth failure), together with increased risks of abnormal findings on neurologic examinations and cerebral palsy at 18 months, suggest an adverse risk–benefit ratio.
Seven trials, involving a total of 669 infants, evaluated corticosteroid treatment begun at 7 to 14 days of age to prevent or treat bronchopulmonary dysplasia.3 This strategy decreased mortality and the incidence of bronchopulmonary dysplasia but with the same short-term adverse effects as those associated with early corticosteroid treatment. Very little information on the long-term outcome is available for this treatment strategy. Nine trials, involving a total of 562 infants, evaluated corticosteroid treatment of bronchopulmonary dysplasia beginning after three weeks of age.4 The corticosteroid-treated infants were extubated earlier and had less need for open-label corticosteroid treatment, with a lower rate of death or bronchopulmonary dysplasia at 36 weeks. A trend toward an increased risk of cerebral palsy was offset by a decreased risk of early death in the corticosteroid-treated infants.
A major conclusion of these analyses was that high-quality follow-up information was critically needed because of concern about neurodevelopmental outcomes. Follow-up data were not available for many of the infants and were often of low quality. Furthermore, many trials permitted the open-label use of corticosteroids, which means that many of the infants in the control group received corticosteroids.
Despite the decreased mortality with the treatments begun at seven days of age or later, the American Academy of Pediatrics and the Canadian Paediatric Society published guidelines in 2002 stating that "the routine use of systemic dexamethasone for the prevention or treatment of chronic lung disease in infants with very low birth weight is not recommended."8 This conclusion was based primarily on the concern about adverse neurodevelopmental effects. In this issue of the Journal, Yeh et al.9 report the outcomes at school age for infants who had been randomly assigned to receive dexamethasone (0.5 mg per kilogram of body weight per day for one week) or placebo, with the first dose given within 12 hours after birth and tapering of the dose over a period of three weeks. They report clear adverse effects on growth, IQ, and motor performance. The quality of the assessments is high, and the authors' conclusion — that a 28-day regimen of high-dose corticosteroid therapy begun within 12 hours after birth should not be recommended — cannot be refuted. These results strengthen the findings on meta-analysis of this early treatment strategy.5
Despite the recent statements by the American Academy of Pediatrics and the Canadian Paediatric Society and other recent pleas to stop using postnatal corticosteroids,8,10 clinicians continue to use them. The Vermont Oxford Network of 408 centers reported that 23 percent of 14,321 infants with a birth weight between 501 g and 1000 g received postnatal corticosteroids in 2002 (Soll R: personal communication). Similarly, the Neonatal Research Network data base of the National Institute of Child Health and Human Development indicates that in 2002, among infants in the same weight range, 19 percent of those who survived for more than 12 hours (range for all centers reporting information, 5 to 49 percent) received postnatal corticosteroid therapy.
Do what we say, not what we do. But what are clinicians actually doing? There is no solid information, but clinicians are probably not using corticosteroids primarily as early treatment. It is likely that corticosteroids are primarily used to treat infants in whom bronchopulmonary dysplasia is developing, and one hopes that a low-dose, short regimen is used. For better or worse, clinicians continue to use corticosteroids because they work, as demonstrated by the short-term pulmonary benefits identified in the meta-analyses.3,4 The report on outcomes at school age by Yeh et al. should end any remaining enthusiasm for early treatment with high-dose dexamethasone.9
Where do we go from here with postnatal corticosteroid therapy for bronchopulmonary dysplasia? The use of a lower dose of dexamethasone (0.15 mg per kilogram per day for three days, followed by tapering of the dose over seven days) with concurrent use of indomethacin resulted in an increased risk of gastrointestinal perforation.11 The 18-month follow-up of infants in this study showed no differences in neurodevelopmental outcomes between the dexamethasone group and the control group.12 However, this result is tempered by the observation that exposure to dexamethasone or open-label use of corticosteroids was associated with abnormal findings on neurologic examination. Although low-dose hydrocortisone seemed to decrease bronchopulmonary dysplasia in a pilot trial,13 a multicenter trial of hydrocortisone was recently stopped because of an increased risk of gastrointestinal perforation (Watterberg K: personal communication). However, in both these trials, corticosteroid therapy was started soon after birth.
If additional trials are conducted, they should target the infants who are currently receiving corticosteroids — those who are dependent on mechanical ventilation at two to three weeks of age and who have progressive lung injury. First, we must understand the pharmacokinetics of several corticosteroids in this age group, and then we should design trials that use the lowest possible effective doses for the shortest period. Neurodevelopmental complications on follow-up constitute the essential safety issue. Since clinicians will not stop using corticosteroids, we need to understand how to use them safely, if that is possible.
Source Information
From the Division of Pulmonary Biology and Neonatology, Cincinnati Children's Hospital Medical Center, Cincinnati.
References
Liggins GC, Howie RN. A controlled trial of antepartum glucocorticoid treatment for prevention of respiratory distress syndrome in premature infants. Pediatrics 1972;50:515-525.
Crowley P. Prophylactic corticosteroids for preterm birth. Cochrane Database Syst Rev 2000;2:CD000065-CD000065.
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-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-CD001145.
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-CD001146.
Baud O, Foix-L'Helias L, Kaminski M, et al. Antenatal glucocorticoid treatment and cystic periventricular leukomalacia in very premature infants. N Engl J Med 1999;341:1190-1196.
Watterberg KL. Adrenocortical function and dysfunction in the fetus and neonate. Semin Neonatol 2004;9:13-21.
Committee on Fetus and Newborn. Postnatal corticosteroids to treat or prevent chronic lung disease in preterm infants. Pediatrics 2002;109:330-338.
Yeh TF, Lin YJ, Lin HC, et al. Outcomes at school age after postnatal dexamethasone therapy for lung disease of prematurity. N Engl J Med 2004;350:1304-1313.
Barrington KJ. The adverse neuro-developmental effects of postnatal steroids in the preterm infant: a systematic review of RCTs. BMC Pediatr 2001;1:1-1.
Stark AR, Carlo WA, Tyson JE, et al. Adverse effects of early dexamethasone treatment in extremely-low-birth-weight infants. N Engl J Med 2001;344:95-101.
Stark AR, Carlo W, Vohr BR, et al. Neurodevelopmental outcome and growth at 18-22 months in infants treated with early dexamethasone. Pediatr Res 2001;49:Suppl:388A-388A. abstract.
Watterberg KL, Gerdes JS, Gifford KL, Lin HM. Prophylaxis against early adrenal insufficiency to prevent chronic lung disease in premature infants. Pediatrics 1999;104:1258-1263.
Related Letters:
Postnatal Dexamethasone for Lung Disease of Prematurity
Lommatzsch M., Klotz J., Virchow J. C. Jr., Watchko J. F., Brozanski B. S., Gordon P. V., Yeh T. F., Lin H. C., Huang C. C., Jobe A. H.(Alan H. Jobe, Ph.D.)