 |
|
 |
|
 |
 |
|
 |
 |
 |
 |
 |
 |
 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
COMMENTARY |
Department of Pediatrics, McGill University, Royal Victoria Hospital, Montreal Quebec, Canada
Therapies for hypotension in very preterm infants are among the most common interventions given to these fragile patients. The proportion of very low birth weight infants who receive potent and potentially toxic interventions varies dramatically from one NICU to another.1
A prospective cohort study of extremely low gestational age newborns (the ELGAN study), results of which are published in this issue of Pediatrics,2 shows even greater variation in interventions for hypotension than previously reported. The proportion of infants under 28 weeks' gestation who were treated ranged from 32% to 98%, and the proportion that received either hydrocortisone or vasopressor agents ranged from 8% to 86%. The investigators examined the reasons for these variations and were unable to show that hypotension was more frequent or severe in the high-intervention centers. Indeed, the only consistent factor associated with intervention was the NICU in which the infant was. It is remarkable that in 7 of the participating NICUs, <10% of the infants under 28 weeks' gestation were considered to have a good enough cardiovascular status to avoid treatment.
These enormous variations in practice reflect the differences between the criteria for treatment and the favorite interventions reported in a survey of neonatologists.3 The most common criterion was a mean arterial pressure in mmHg less than the gestational age in weeks (either with or without clinical signs), which is a simplistic threshold that ignores the well-described spontaneous increase in normal blood pressure that occurs in the first few days of life and has no empirical support in the literature. More than 50% of very immature infants will have a blood pressure below this threshold on the first day of life.4 Using such a criterion is also inconsistent with the well-described lack of association of blood pressure with systemic blood flow.5 Therefore, it is very evident, unfortunately, that we truly have no idea which preterm newborns might benefit from cardiovascular support6 or what interventions are effective in improving clinical outcomes.
Most neonatologists start treatment for hypotension with at least 1 fluid bolus and then commence dopamine.3 Additional evidence of the uncertainty surrounding hypotension treatment is shown by the fact that the starting dose of dopamine given by some individuals is the same as the maximum dose used by others; the same pattern is seen for dobutamine, epinephrine, volume of saline given as a fluid bolus, and (dramatically so) steroids (hydrocortisone starting dose: 0.1–5 mg/kg; maximum: 0.5–15 mg/kg).
Unfortunately, the authors of this study have not provided data to indicate if the described variations in practice are associated with variations in clinically important outcomes, but other data suggest that such associations do exist. The Canadian neonatal network, for example, has shown an odds of developing severe intraventricular hemorrhage that varied >10-fold between units even after adjusting for severity of illness and demographic factors known to affect risk.7 That study suggested that pressor treatment of hypotension was associated with a worse outcome in preterm infants; this remains true even when the severity of hypotension is accounted for.4 Previous observational data have also shown an increase in intraventricular hemorrhage8 and mortality9 in preterm infants that is associated with fluid-bolus administration.
Eighty-two percent of the infants in the ELGAN study received treatment for hypotension; if this is generally true in the United States (where 
0.4% of the 4.1 million births [ie, 16000 infants] are born at extremely low gestational age annually), then >13000 very high-risk infants are receiving treatment of unknown efficacy and toxicity for uncertain indications each year. Similar dramatic figures could be calculated for other countries.
A critical review of the literature reveals that there have never been randomized trials:
A recent National Institute of Child Health and Human Development/Food and Drug Administration–sponsored conference6 suggested models for prospective trials to address these areas of ignorance. The details of the trials suggested may be debatable, but the importance of actually performing them surely is not.
We cannot let this ignorance continue.
 |
FOOTNOTES |
|---|
Address correspondence to Keith Barrington, MBChB, McGill University, Department of Pediatrics, Room C7.68, Royal Victoria Hospital, 687 Pine Ave W, Montreal, Quebec, Canada H3A 1A1. E-mail: keith.barrington{at}mcgill.ca
The author has indicated he has no financial relationships relevant to this article to disclose.
Opinions expressed in these commentaries are those of the authors and not necessarily those of the American Academy of Pediatrics or its Committees.
|
REFERENCES |
|---|
 TOP REFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
||
|
||
Home | My Pediatrics | Journal Information | Current Issue | Past Issues | Subscriptions & Services | Contact Us Click here for faster international access © 2007 American Academy of Pediatrics. All rights reserved. |
||
|
||
