Docosahexaenoic acid (DHA) has been associated with downregulation of inflammatory responses.
To report the effect of DHA supplementation on long-term atopic and respiratory outcomes in preterm infants.
This study is a multicenter, randomized controlled trial comparing the outcomes for preterm infants <33 weeks' gestation who consumed expressed breast milk from mothers taking either tuna oil (high-DHA diet) or soy oil (standard-DHA) capsules. Data collected included incidence of bronchopulmonary dysplasia (BPD) and parental reporting of atopic conditions over the first 18 months of life.
Six hundred fifty-seven infants were enrolled (322 to high-DHA diet, 335 to standard), and 93.5% completed the 18-month follow-up. There was a reduction in BPD in boys (relative risk [RR]: 0.67 [95% confidence interval (CI): 0.47–0.96]; P = .03) and in all infants with a birth weight of <1250 g (RR: 0.75 [95% CI: 0.57–0.98]; P = .04). There was no effect on duration of respiratory support, admission length, or home oxygen requirement. There was a reduction in reported hay fever in all infants in the high-DHA group at either 12 or 18 months (RR: 0.41 [95% CI: 0.18–0.91]; P = .03) and at either 12 or 18 months in boys (RR: 0.15 [0.03–0.64]; P = .01). There was no effect on asthma, eczema, or food allergy.
DHA supplementation for infants of <33 weeks' gestation reduced the incidence of BPD in boys and in all infants with a birth weight of <1250 g and reduced the incidence of reported hay fever in boys at either 12 or 18 months.
Comments
High-dose DHA supplementation in preterm infants for respiratory and allergy outcomes
I read with interest the results of the respiratory and allergy outcomes of the trial of high-dose Docosahexaenoic Acid (DHA) supplementation in preterm infants presented by Davis, et al(1). I am concerned that the positive results presented in this publication could be misleading for reasons cited below.
For respiratory outcome, the authors conclude that the incidence of BPD was lower for boys and for all infants with birth weight <1250 grams. However, these results are based on post-hoc subgroup analysis of BPD data published earlier(2) rather than a pre-specified outcome as per trial's registration information(3). Therefore these results could at best be considered hypothesis generating which require validation in further studies. This becomes particularly important as the number of days on oxygen or the proportion of infants needing oxygen at discharge was not different between the two study groups.
For allergy outcomes, the authors test for multiple outcomes (i.e., hay fever, asthma, eczema and food allergies) at several time points (12 months, 18 months, and at either 12 or 18 months), at a great risk of ?- error (type 1 error). Although, authors acknowledge this concern in their text but don't provide the likely magnitude of this error with the number of analyses conducted, which approximates to about 50% (i.e., the probability that the reported results for hay-fever at "either 12 or 18 month" are false-positive). It is also difficult to fathom how the sample size (n) for reporting this outcome is lesser than the total number of infants included for reporting at 18 months alone.
My greatest concern is regarding the dissociation between the manner in which information is presented in the abstract and the main text of the study. It is well known that a large number of practitioners rely on the information contained in study abstracts as their main source of information due to lack of access to full subscription and various other reasons(4). Therefore, it is considered important that the conclusions stated in the abstract should accurately reflect the message of the study(5,6). Whilst, the authors state caution in believing study results by use of several phrases in the main manuscript such as, "Given the number of tests performed and the consequent inflation of type I error, we advise caution..." (page e73), "...cannot exclude the possibility, however, that these findings are due to chance..."(page e74), and "...these findings are potentially important and deserve further investigation..." (page e74); the abstract conveys a sense as if the observed treatment benefits are conclusive without any mention of need for further studies to validate those results. Furthermore, abstract hides pertinent information from readers that many of the analyses were post-hoc from an existing data set.
I will like to conclude by cautioning readers that any of the stated claims towards efficacy of high-dose DHA supplementation from this study could only be considered hypothesis generating and should not be used to change clinical practice.
References
1. Manley BJ, Makrides M, Collins CT, McPhee AJ, Gibson RA, Ryan P, Sullivan TR, Davis PG; for the DINO Steering Committee. High-Dose Docosahexaenoic Acid Supplementation of Preterm Infants: Respiratory and Allergy Outcomes. Pediatrics. 2011 Jul;128(1):e71-e77.
2. Makrides M, Gibson RA, McPhee AJ, Collins CT, Davis PG, Doyle LW, Simmer K, Colditz PB, Morris S, Smithers LG, Willson K, Ryan P. Neurodevelopmental outcomes of preterm infants fed high-dose docosahexaenoic acid: a randomized controlled trial. JAMA. 2009 Jan 14;301(2):175-82.
3. Australian New Zealand Clinical Trials Registry. ACTRN12606000327583. Available at http://www.anzctr.org.au/trial_view.aspx?ID=1427
4. Hopewell S, Clarke M, Moher D, Wager E, Middleton P, Altman DG, Schulz KF; CONSORT Group. CONSORT for reporting randomised trials in journal and conference abstracts. Lancet. 2008 Jan 26;371(9609):281-3.
5. Hopewell S, Clarke M, Moher D, Wager E, Middleton P, Altman DG, Schulz KF; CONSORT Group. CONSORT for reporting randomized controlled trials in journal and conference abstracts: explanation and elaboration. PLoS Med. 2008 Jan 22;5(1):e20.
6. Haynes RB, Mulrow CD, Huth EJ, Altman DG, Gardner MJ. More informative abstracts revisited. Ann Intern Med. 1990 Jul 1;113(1):69-76.
Conflict of Interest:
None declared