PEDIATRICS Vol. 120 No. 3 September 2007, pp. 689-690 (doi:10.1542/peds.2007-1646)
LETTER TO THE EDITOR |
The DART Study of Low-Dose Dexamethasone Therapy
Karin J. Rademaker, MD, PhD1Floris Groenendaal, MD, PhD1
Frank van Bel, MD, PhD1
Linda S. de Vries, MD, PhD1
1 Department of Neonatology
Cuno S. P. M. Uiterwaal, MD, PhD2
2 Julius Center for Health Sciences and Primary Care
Wilhelmina Children's Hospital/University Medical Center Utrecht
3508 AB Utrecht, Netherlands
To the Editor.—
We read with interest the article "Outcome at 2 Years of Age of Infants From the DART Study: A Multicenter, International, Randomized, Controlled Trial of Low-Dose Dexamethasone" by Doyle et al.1 The authors reported on neurosensory outcome, growth, and health in 35 dexamethasone-treated children versus a control group of 35 children who were chronically ventilated. Of these children, 59 survived to 2 years of age and 58 were assessed at follow-up. One of the objectives of the DART (Dexamethasone: A Randomized Trial) study was to determine the long-term effects of low-dose dexamethasone. Unfortunately, the study had to be stopped at an inclusion rate of <10% because of difficulties in recruitment. The 70 included children were part of 192 eligible infants from 11 centers in 3 countries.
Careful study of the characteristics of the included patients reveals that within the 2 groups the administered amounts of dexamethasone varied widely, whereas the treatment differences between the 2 groups were possibly much smaller than intended, as argued below.1,2
Of the dexamethasone-treated children, 17.1% received prerandomization steroids. The study design allowed for a repeat course of the same blinded drug as a therapeutic option for the attending clinicians. Moreover, use of open-label corticosteroids after randomization was allowed to facilitate the study in the first place but resulted in 25.7% of them receiving late rescue systemic corticosteroid therapy. The dose of rescue corticosteroids (dexamethasone) exceeded the low-dose dexamethasone regimen to which children were randomly assigned. Thus, some children would have had 1 course of low-dose dexamethasone, some would have had 2 courses of low-dose dexamethasone, some would have had 1 short course of low-dose dexamethasone and a late rescue treatment of systemic corticosteroid therapy, and some would have had 2 short low-dose courses combined with the late rescue systemic treatment. Also, combinations of prerandomization steroids and different courses were possible. Therefore, within a group of 35 infants, 8 different dexamethasone regimens were possible, resulting in very different cumulative doses of dexamethasone.
At the same time, 14.3% of the patients in the control group already had received corticosteroids before randomization, and as many as 40% received late rescue treatment with systemic corticosteroids. Thus, these children also had very different corticosteroid treatments: some did not receive corticosteroids at any time, some received corticosteroids before randomization, some received late rescue systemic treatment with dexamethasone, and some could have had a combination of prerandomization and late rescue treatment. Thus, in the control group of 35 infants, 4 different corticosteroid regimens were possible, again yielding very different cumulative doses. It is important to note that a large proportion of children in the placebo group received a higher dose of dexamethasone than those in the dexamethasone group.
The outcome at 2 years of age was described for 56 survivors: 29 in the dexamethasone group and 27 in the placebo group. Nowhere in the follow-up article1 was it mentioned that a very high proportion of the control group received either steroids before randomization or late rescue systemic corticosteroid therapy, even at a higher dose than many children in the dexamethasone group.
The authors rightfully analyzed by the intention-to-treat principle for the intended effects and found beneficial intended effects despite much variation in treatment regimens. However, they did not seem to recognize that through their study design the dexamethasone treatment contrast between the groups may have become so small that a powerful randomized comparison of unintended effects, such as neurologic complications, may have become impossible. Moreover, open-label contamination would also increase the steroid dose received by infants in the index treatment arm, making interpretation even more difficult.
The authors found the very small patient groups to be the most important weakness of the study. However, we also find that unless the authors can show that there was a meaningful continued dexamethasone treatment contrast between the groups, the data do not support their conclusion on long-term outcome. For now, we feel that the only proper conclusion that should be drawn is that no conclusion can be drawn.
We do agree with the authors that it is very important to have a definite trial of corticosteroids to answer urgent questions about the safety of this therapy. Unfortunately, the DART study did not seem to answer these questions.
REFERENCES
- Doyle LW, Davis PG, Morley CJ, McPhee A, Carlin JB; DART Study Investigators. Outcome at 2 years of age of infants from the DART study: a multicenter, international, randomized, controlled trial of low-dose dexamethasone.
Pediatrics. 2007;119
:716
–721
[Abstract/Free Full Text] - Doyle LW, Davis PG, Morley CJ, McPhee A, Carlin JB; DART Study Investigators. Low-dose dexamethasone facilitates extubation among chronically ventilator-dependent infants: a multicenter, international, randomized, controlled trial.
Pediatrics. 2006;117
:75
–83
[Abstract/Free Full Text]
PEDIATRICS (ISSN 1098-4275). ©2007 by the American Academy of Pediatrics
CiteULike 
Connotea 
Del.icio.us 
Digg 
Facebook 
Reddit 
Technorati 
Twitter What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
