The two most striking conclusions that the authors make in this review1 are that prophylactic phototherapy reduces the need for exchange transfusion, and reduces long-term neurodevelopmental impairment.
With regard to the effect on exchange transfusions this result is not considered clinically relevant. The finding is largely due to one study2, started in 1974, in which the threshold for doing an exchange transfusion was very low, and the phototherapy used would have been far less effective than that used now. Nowadays exchange transfusions are done at much higher bilirubin levels and are a rare event. In our hospital during the period 2000 to 2011 inclusive only 2.5 per 1000 preterm infants had an exchange transfusion, and only 1.7 per 1000 low birth weight infants had an exchange transfusion. Studies done before the late 1990s would almost certainly have used a far less effective form of phototherapy. The effectiveness of phototherapy has continued to improve over the years showing dramatic improvement in the 1990s with the change from fluorescent phototherapy lights to halogen lights. The more relevant results are found in the subgroup of two studies (both reported in this century) that had a high threshold for exchange transfusion where there was no difference in the proportion of infants that had an exchange transfusion.
The effect of prophylactic therapy on neurodevelopmental impairment is more interesting. Only one study, in extremely low birth weight infants, reported this outcome3: the study was large (total N = 1974) and of good quality. The relative risk of neurodevelopmental impairment was 0.86 (95% CI 0.74–0.99). The number-needed-to-treat to prevent one case of impairment at 18–22 months of age was 25. This is a small effect but an important one. The report of the original study3 is worth reading and serious consideration should be given to the use of prophylactic and aggressive phototherapy in extremely low birth weight infants. There is good evidence that the benefit outweighs any harm in infants with a birth weight of 751 to 1000 g. In infants with a birth weight of 501 to 750 g the evidence is less clear and the benefit of prophylactic and aggressive phototherapy may be offset by a slight increase in mortality in this group.
The other important findings of this review are that the overall duration of phototherapy use and the duration of hospital stay were not increased in infants receiving prophylactic phototherapy.
Two further subgroup analyses based on the following would have been more useful:
- 1.The quality of studies, especially the subgroup of studies where group allocation was actually random and well concealed. The authors do state that they ‘performed a sensitivity analysis to explore the effect of trial quality on the results’, but the results of this sensitivity analysis are not reported;
- 2.Recent studies, especially from 2000 onwards, compared with the rest. This would be valuable given the significant changes in the effectiveness of phototherapy at the end of the 20th century.