Phototherapy and Neurodevelopmental Outcomes in Extremely LBW Infants
Phototherapy and Neurodevelopmental Outcomes in Extremely LBW Infants
Morris BH, Oh W, Tyson JE, et al; NICHD Neonatal Research Network
N Engl J Med. 2008;359:1885-1896
Some studies have indicated that a serum bilirubin level as low as 5 mg/dL may lead to central nervous system injury, but conclusions from these previous studies were not clear. Data from the 1970s suggested that aggressive treatment of hyperbilirubinemia by phototherapy was associated with increased risk for death in infants weighing < 2500 g, but that phototherapy did not lead to improved neurodevelopmental outcomes.
The current study was a multicenter, randomized trial that looked at a composite outcome of death or neurodevelopmental impairment (assessed at 18-22 months, corrected age). The combined outcome was chosen because infants who died would not complete the neurodevelopmental assessment.
Sixteen centers that were members of the Neonatal Research Network enrolled newborns who weighed 501-1000 g. Infants with severe perinatal complications or congenital anomalies were not enrolled. The randomization was stratified by birthweight (501-750 g and 751-1000 g) and by center. All enrolled infants began phototherapy within the first 14 days of life, and phototherapy was continued for at least 24 hours once begun. The bilirubin level that would trigger phototherapy use varied by weight stratum, age of the patient, and then treatment group. For example, for infants 501-750 g, infants in the aggressive therapy group began phototherapy whenever their level was ≥ 5 mg/dL compared with 8 mg/dL as the cutoff for beginning phototherapy in the conservative infants weighing 501-750 g.
Bilirubin levels that would trigger phototherapy were slightly higher for infants > 7 days old and for infants > 751 g, but the highest level that would trigger phototherapy (infants 751 g, > 7 days, in the conservative therapy group) was 10 mg/dL. At day 5 of treatment, a serum sample was obtained to test for unbound bilirubin. Infants underwent exchange transfusions for bilirubin levels of 13 mg/dL (501-750 g) and 15 mg/dL (751-1000 g) if they did not respond to increasing phototherapy irradiance levels. Trained individuals, masked to treatment assignment, completed neurodevelopmental assessment at 18-22 months (corrected) age.
The following conditions constrituted "neurodevelopmental impairment": blindness (uni- or bilateral); severe hearing loss (uni- or bilateral); moderate or severe cerebral palsy, eg, unable to walk without devices or unable to walk, respectively; or a predetermined "low score" on the neurodevelopmental assessment.
Over a roughly 2.5-year period, the study enrolled 1974 infants (69% of eligible). Of interest, 22% of the "conservative" therapy group never received phototherapy compared with < 1% of the aggressive therapy group. There was no difference in the primary outcome between the 2 treatment groups, with 52% of the infants in the aggressive therapy group and 55% of the infants in the conservative therapy group either dying or experiencing neurodevelopmental impairment (P = .15). This difference remained nonsignificant even after adjusting for confounding variables, such as race/ethnicity, sex, and whether the infants were "inborn" or "outborn" relative to the study center. Rates of death were 24% in the aggressive therapy group and 23% in the conservative therapy group. However, the rate of neurodevelopmental impairment alone was lower in the aggressive therapy group at 26% compared with 30% (relative risk 0.86, 95% confidence interval 0.74-0.99).
When looking at prespecified secondary outcomes, the aggressive therapy group had lower rates of hearing loss, athetosis, and were less likely to have neurodevelopmental scores below a desired cutoff. When looking at secondary outcomes by weight group, there was no benefit among the infants who weighed 751-1000 g nor in the subgroup that weighed 501-750 g; both confidence intervals included 1.0. In additional post hoc analyses, the authors raised the concern that the lower-weight group actually appeared to be at risk for higher rates of death.
The authors concluded that there was no significant difference in the rate of the combined outcome (death or neurodevelopmental impairment) between the infants treated with aggressive phototherapy and those treated with conservative therapy. The absolute death rate in the smaller infants was higher (difference NS) among the aggressive phototherapy group, prompting the authors to raise concern that the benefit (again, difference not sigificant) with regard to neurodevelopmental impairment in the infants weighing 501-750 g may be offset by a higher death rate in that group.
This was a large, complicated trial that was not powered for some of the major subgroup analyses, such as the subgroup analyses by weight. The study was powered relative to the primary outcome. Therefore, whereas the trends in the subgroups are hypothesis-forming for future studies, they cannot be relied on as firm conclusions relative to this study. In summary, these data neither support nor refute the value of aggressive phototherapy among very-low-birthweight infants, and the trial demonstrates once again how difficult it can be to obtain firm conclusions even from a large, difficult-to-complete trial.
Abstract
Aggressive vs. Conservative Phototherapy for Infants With Extremely Low Birth Weight
Morris BH, Oh W, Tyson JE, et al; NICHD Neonatal Research Network
N Engl J Med. 2008;359:1885-1896
Summary
Some studies have indicated that a serum bilirubin level as low as 5 mg/dL may lead to central nervous system injury, but conclusions from these previous studies were not clear. Data from the 1970s suggested that aggressive treatment of hyperbilirubinemia by phototherapy was associated with increased risk for death in infants weighing < 2500 g, but that phototherapy did not lead to improved neurodevelopmental outcomes.
The current study was a multicenter, randomized trial that looked at a composite outcome of death or neurodevelopmental impairment (assessed at 18-22 months, corrected age). The combined outcome was chosen because infants who died would not complete the neurodevelopmental assessment.
Sixteen centers that were members of the Neonatal Research Network enrolled newborns who weighed 501-1000 g. Infants with severe perinatal complications or congenital anomalies were not enrolled. The randomization was stratified by birthweight (501-750 g and 751-1000 g) and by center. All enrolled infants began phototherapy within the first 14 days of life, and phototherapy was continued for at least 24 hours once begun. The bilirubin level that would trigger phototherapy use varied by weight stratum, age of the patient, and then treatment group. For example, for infants 501-750 g, infants in the aggressive therapy group began phototherapy whenever their level was ≥ 5 mg/dL compared with 8 mg/dL as the cutoff for beginning phototherapy in the conservative infants weighing 501-750 g.
Bilirubin levels that would trigger phototherapy were slightly higher for infants > 7 days old and for infants > 751 g, but the highest level that would trigger phototherapy (infants 751 g, > 7 days, in the conservative therapy group) was 10 mg/dL. At day 5 of treatment, a serum sample was obtained to test for unbound bilirubin. Infants underwent exchange transfusions for bilirubin levels of 13 mg/dL (501-750 g) and 15 mg/dL (751-1000 g) if they did not respond to increasing phototherapy irradiance levels. Trained individuals, masked to treatment assignment, completed neurodevelopmental assessment at 18-22 months (corrected) age.
The following conditions constrituted "neurodevelopmental impairment": blindness (uni- or bilateral); severe hearing loss (uni- or bilateral); moderate or severe cerebral palsy, eg, unable to walk without devices or unable to walk, respectively; or a predetermined "low score" on the neurodevelopmental assessment.
Over a roughly 2.5-year period, the study enrolled 1974 infants (69% of eligible). Of interest, 22% of the "conservative" therapy group never received phototherapy compared with < 1% of the aggressive therapy group. There was no difference in the primary outcome between the 2 treatment groups, with 52% of the infants in the aggressive therapy group and 55% of the infants in the conservative therapy group either dying or experiencing neurodevelopmental impairment (P = .15). This difference remained nonsignificant even after adjusting for confounding variables, such as race/ethnicity, sex, and whether the infants were "inborn" or "outborn" relative to the study center. Rates of death were 24% in the aggressive therapy group and 23% in the conservative therapy group. However, the rate of neurodevelopmental impairment alone was lower in the aggressive therapy group at 26% compared with 30% (relative risk 0.86, 95% confidence interval 0.74-0.99).
When looking at prespecified secondary outcomes, the aggressive therapy group had lower rates of hearing loss, athetosis, and were less likely to have neurodevelopmental scores below a desired cutoff. When looking at secondary outcomes by weight group, there was no benefit among the infants who weighed 751-1000 g nor in the subgroup that weighed 501-750 g; both confidence intervals included 1.0. In additional post hoc analyses, the authors raised the concern that the lower-weight group actually appeared to be at risk for higher rates of death.
The authors concluded that there was no significant difference in the rate of the combined outcome (death or neurodevelopmental impairment) between the infants treated with aggressive phototherapy and those treated with conservative therapy. The absolute death rate in the smaller infants was higher (difference NS) among the aggressive phototherapy group, prompting the authors to raise concern that the benefit (again, difference not sigificant) with regard to neurodevelopmental impairment in the infants weighing 501-750 g may be offset by a higher death rate in that group.
Viewpoint
This was a large, complicated trial that was not powered for some of the major subgroup analyses, such as the subgroup analyses by weight. The study was powered relative to the primary outcome. Therefore, whereas the trends in the subgroups are hypothesis-forming for future studies, they cannot be relied on as firm conclusions relative to this study. In summary, these data neither support nor refute the value of aggressive phototherapy among very-low-birthweight infants, and the trial demonstrates once again how difficult it can be to obtain firm conclusions even from a large, difficult-to-complete trial.
Abstract
Source...