Article Text

This article has a correction. Please see:

Which course of dexamethasone?
  2. D B KNIGHT,
  1. Department of Paediatrics, National Women’s Hospital
  2. Claude Road, Auckland, Private Bag 92019
  3. New Zealand

    Statistics from

    Editor—Skelton et al 1 state that “most units start [dexamethasone] by the second week of life, weaning over 2–3 weeks rather than 6” and that such a regimen “has been shown to be comparable in effectiveness with a six week course.” The reference cited is a review paper that provides no evidence for the first statement, and discusses studies that showed a benefit in time to extubation but not in the duration of oxygen treatment.2Time of extubation is irrelevant if the incidence of chronic lung disease has not been reduced, and, to our knowledge, a three week course has not been shown to have this beneficial effect. Indeed, Cummings et al, in the study in which he reported beneficial effects of the 6 week course, found that an 18 day course was not effective at reducing time in oxygen.3 There are now at least four dexamethasone regimens described which do reduce the incidence of chronic lung disease. In these days of evidence based medicine, should we not be using one of these?3-6

    The cardiac effects of a short course of dexamethasone have been reported before, by Brozanski et al, in their trial of a three day repeatable pulsed course.4 They also found significant hypertrophy which regressed by discharge. Definition of myocardial hypertrophy by statistical comparison with the “control” group in the study of Skelton et al is of doubtful value. Controls were very different babies, being a median of 3 weeks older and 400 g heavier at birth. Their echocardiograms were performed at different gestational and postnatal ages. Furthermore, the increases in interventricular septal and left ventricular posterior wall thicknesses are described for a 21 day period in the control group and for a variable period (to maximal hypertrophy) in the treatment group. Inspection of the data provided in the figures suggests that comparison between groups after 21 days may not have shown a significant difference.


    Drs Skelton, Parsons, and Gill reply:

    Editor—We feel that Bloomfield et al may have missed the main purpose of our study. Evans1-7 recommended screening for left ventricular haemorrhage for infants receiving dexamethasone. We examined the safety of current prevailing practice. In the UK dexamethasone courses have been reduced in length due to concerns about side effects. Pulsed dexamethasone1-8 is not often used. Whatever its merits and despite many studies of different regimens, ours reflects current use. Although we hope to prevent chronic lung disease, in practice, early extubation and the stabilisation short courses produce are not “irrelevant.”

    Due to the nature of infants receiving dexamethasone, well matched controls were not possible. However, the changes on real time echocardiography were very clear cut. Our controls, as well matched as possible, were included to provide comparison and emphasis of the severity of the left ventricular haemorrhage, rather than prove that it had occurred. In trying to compare groups at 21 days, Bloomfieldet al seem to have misunderstood. Twenty eight day old controls was chosen to try to coincide with maximum hypertrophy, at around 10 days after starting dexamethasone. Two time points were felt sufficient, as we found no left ventricular haemorrhage in a normal preterm population of a similar age.1-9

    Brozanski’s1-8 study is not comparable. The study population and dexamethasone use differed: repeated doses 3 days every 10 days up to 36 weeks is hardly “short course.” The ethics of placebos is of concern and a low incidence of left ventricular haemorrhage (24%) suggests different natural history or inappropriate reference ranges.1-9 Our study is the first to examine in depth left ventricular haemorrhage using current dexamethasone practice.

    Finally, these studies emphasise our poor knowledge of dexamethasone and left ventricular haemorrhage. Greater emphasis on the mechanism of action of dexamethasone rather than differing regimens may improve use in small susceptible infants.


    1. 1-7.
    2. 1-8.
    3. 1-9.

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Linked Articles

    • Correction
      BMJ Publishing Group Ltd and Royal College of Paediatrics and Child Health