 |
|
 |
|
 |
 |
|
 |
 |
 |
 |
 |
 |
 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Irene Roberts, MD
Department of Haematology,
Imperial College,
Hammersmith Hospital
London W12 ONN, United Kingdom
Simon Stanworth, MD
National Blood Service,
John Radcliffe Hospital,
Oxford OX3 9BQ, United Kingdom
To the Editor.—
Bell et al1 are to be congratulated for conducting and reporting their recent trial comparing the effects of restrictive and liberal thresholds for red blood cell transfusion in preterm neonates. Such patients receive large amounts of blood-component therapy, but despite considerable research in this patient group, virtually all blood-component therapy in preterm neonates remains opinion based rather than truly evidence based. Thus, any well-conducted research into the potential benefits and risks of blood components in preterm neonates must be welcomed with open arms.
The critical question for neonatologists is: What level of confidence can be attached to the findings in this article? The study confirms the known relationship between neonatal anemia and apnea but, far more importantly, shows an apparent relationship between restrictive red blood cell transfusion practice in preterm neonates and the development of the most severe grades of intraventricular hemorrhage and periventricular leukomalacia. Clearly, if confirmed, this would be of stellar importance to neonatal practice, because these are the cerebral lesions most commonly associated with future impaired neurodevelopment.
Fortunately (from a patient viewpoint), the number of neonates in the study with these complications is small, and as a result the study is clearly underpowered to demonstrate a separate significant correlation between restrictive red blood cell transfusion practice and either severe intraventricular hemorrhage or periventricular leukomalacia. Indeed, the primary end point for sample-size determination was a transfusion end point and not a clinical outcome end point. This difficulty is compounded further by the fact that only just more than 50% of the study population underwent late (day-42) cranial ultrasound examinations that potentially reduced the pick-up of severe cerebral lesions within the study. In addition, the chosen cranial ultrasound examination schedule for preterm neonates who were of <27 weeks' gestation (days 7, 14, and 42) may underestimate the prevalence of cystic periventricular leukomalacia, because the cysts of this lesion can develop and regress between days 7 and 42.
Despite these methodologic difficulties, the possible finding of major adverse cerebral outcomes associated with the increasingly common practice of restrictive red blood cell transfusion in preterm neonates should not simply be ignored. It therefore is reassuring that no such correlation was found in another recent large study of red blood cell transfusion practice in preterm neonates (although it has only been reported in abstract form).2
This suggests that perhaps the most important finding of this study is, yet again, how little we actually know about the risks and benefits of one of the most common therapies administered to preterm neonates and the pressing need for high-quality research trials using clinically relevant outcomes to evaluate blood-component therapy for these vulnerable, but potentially long-lived, patients.
REFERENCES
Related articles in Pediatrics:
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
||
|
||
Home | My Pediatrics | Journal Information | Current Issue | Past Issues | Subscriptions & Services | Contact Us Click here for faster international access © 2005 American Academy of Pediatrics. All rights reserved. |
||
|
||
