 |
|
 |
|
 |
 |
|
 |
 |
 |
 |
 |
 |
 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
To the Editor.—
We congratulate Bell et al1 for undertaking this difficult but important clinical study. They conclude that their "finding of more frequent adverse neurologic events in the restrictive RBC [red blood cell]-transfusion group suggests that the practice of restrictive transfusions may be harmful to preterm infants." The real question is whether such a conclusion causally linking the transfusion practices to the adverse neurologic outcome can be reached on the basis of the data presented in the article.
The major difference in transfusion practices among neonatologists is not in the acutely sick, ventilator-dependent, very low birth weight infants in the first 2 weeks of life. The difficult question facing them is when to transfuse a growing low birth weight infant with chronic anemia who may or may not be symptomatic. In this regard, the data that are not presented may be more meaningful in helping the readers to decide whether the restrictive practices are harmful to preterm infants.
Almost all intracranial hemorrhage occurs in the first week of life. If one assumes that the grade 4 hemorrhage was detected on the first ultrasound at 7 days of age (exact timing not given), the authors have to at least present the data on transfusion practices in these 4 infants before the diagnosis of intraventricular hemorrhage (IVH) to support their conclusion. It is also important to know when these 4 patients were entered into the study (median age of entry was 3 days), when the first transfusion was performed, and what the hematocrit value was before the transfusion. Because hematocrit levels are typically maintained above 35% in these sick, ventilated infants during the first few days of life, early transfusion is likely to be dictated by the severity of their illness (with increased iatrogenic blood loss) or a drop in hematocrit caused by a large IVH. Neither IVH (grades 1–4) nor severe IVH (grades 3–4) was significantly different between the 2 groups. Similarly, there was no significant difference in the incidence of periventricular leukomalacia (PVL). However, the incidence of grade 4 IVH alone and combined grade 4 IVH and PVL (an unusual grouping) were statistically significant between the 2 groups. Was this an a priori hypothesis or merely a posthoc-analysis finding? Was appropriate correction done for such analysis? In summary, we are at a loss to understand the basis on which the increased incidence of grade 4 IVH was attributed to the restrictive-transfusion practices. To our knowledge, none of the previous studies on transfusion practices have observed this finding.2,3
The trend for increased PVL (not statistically significant) in the restrictive-transfusion group is intriguing. The important clinical question is whether anemia or its sequelae has anything to do with this finding. Anemia increases heart rate, cardiac output, stroke volume, and shortening fraction (no such trends are seen in the present study), attenuating the effects of packed cell volume on systemic oxygen-carrying capacity in preterm infants. The authors report a small but significant difference in the incidence of apnea and its sequelae. Providing the incidence of bradycardia and/or oxygen desaturation would have been more meaningful from a pathophysiological standpoint. One is always concerned whether the recurrent episodes of desaturation and bradycardia caused by apnea predispose very low birth weight infants to PVL. The elegant study by Poets et al,4 in which mildly anemic (8.2–12.0 g/dL) infants were monitored polygraphically before and after transfusion (not just by the nursing documentation), failed to show a difference in apnea (>4 seconds), bradycardia (<2/3 of baseline), or desaturation (<80%). Also, they did not find any significant change in the combined frequency of bradycardia and desaturation (the primary end points) after transfusion in a more severely anemic group (6.3–9.8 g/dL).5 Furthermore, Bell et al state that "although this difference [in apnea frequency] is statistically significant, the absolute increase in frequency of apnea, <1 event per day, was so small as to be of questionable clinical importance."1 No attempt has been made to show that these 4 infants had more severe anemia or more severe apnea from mild anemia. Yet, they attribute the adverse neurologic finding to the apnea.
Even more importantly, alternate hypotheses were not adequately explored. Could the increased incidence of grade 4 IVH be caused by other confounding variable(s) and unrelated to transfusion practices? Did it occur before transfusion practices could have an impact? Was the increased apnea frequency the result of IVH rather than anemia? Could the occurrence of PVL be related to clinically significant patent ductus arteriosus caused by ductal steal? How many of these 4 infants belong to the group of 15 infants treated for patent ductus arteriosus? Could PVL be the result of hypotension/ischemia/reperfusion injury? How many of these infants were given inotropic support in the first week of life? The fact that 2 infants had both IVH and PVL may suggest a common pathophysiologic mechanism. What role, if any, did autoregulation of cerebral blood flow (or the lack thereof) play in the genesis of these events? Did the ventilatory strategies used contribute to the IVH and PVL through their cardiovascular interaction? We are concerned about high-frequency ventilation with hypocarbia, impedance of venous return, and alteration in cardiac output.
We conclude by saying that the authors did a disservice to the readers by suggesting that transfusion practices may be causally linked to the grade 4 IVH and PVL for the reasons stated above.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
||
|
||
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. |
||
|
||
