Drs Murray, Roberts, and Stanworth underscore the importance of our finding of increased risk of intracerebral hemorrhage and periventricular leukomalacia among small preterm infants transfused with red blood cells using more restrictive criteria, allowing their hematocrit levels to fall to lower threshold values before transfusing.1 They also ask about the degree of confidence that we should have in our result. The answer to their question is that the clustering of major adverse neurologic events in the restrictive-transfusion group may have occurred by chance, but only a 1.2% chance (P = .012); our finding would be expected to occur by chance only 1 in 83 times.
Murray et al point out that the association we found might have been weaker had every subject received a late cranial ultrasound examination and if all subjects had received more frequent ultrasound examinations between 7 and 42 days of age. Although this is true, it is also possible that more complete data would have shown the association to be even stronger.
We agree absolutely that this association of restrictive transfusion practice with adverse short-term neurologic events must be confirmed by additional study but that also it cannot be brushed aside because of design or power issues. We, too, are pleased that another, larger study of the impact of transfusion practice on outcome in preterm infants was completed recently.2 We hope that the Premature Infants in Need of Transfusion (PINT) study will include careful analysis of this important association of restrictive transfusion practice with serious adverse neurologic events. If the findings of the PINT study differ from ours, it will be important to examine whether any such difference might be explained by differences in the design of the PINT study compared with ours. All interested practitioners anxiously await publication of the details of the PINT study.
Both we and the PINT investigators have the responsibility of tracking our subjects to look for any longer-term impact of transfusion practice on neurologic and developmental outcome. We plan to recall our subjects for careful evaluation when they reach school age.
REFERENCES
Related articles in Pediatrics:
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||