Objectives: To find out whether supplementation of formula milk by long-chain polyunsaturated fatty acids (LCPUFA) affects neurodevelopment at 18 months of age in term or preterm infants by an individual patient data (IPD) meta-analysis.
Materials and Methods: Data of 870 children from 4 large randomised clinical trials for formula milk with and without LCPUFAs allowed for assessing the effect of LCPUFA with adjustment for potential confounders and extensive subgroup analysis on prematurity, LCPUFA source, and dosage. Any additional clinical trials examining the effect of LCPUFA supplementation on Bayley Scales of Infant Development at 18 months were regarded as relevant. Two relevant studies were identified by MEDLINE, but were not available to us. An IPD meta-analysis was performed with subgroup analyses by preterm delivery, very low birth weight (<1500 g), trials with higher amounts of docosahexaenoic acid (DHA) and arachidonic acid (AA), and specific sources of LCPUFA. The sample size of 870 children was sufficient to detect clinically relevant differences in Bayley Scales even in subgroups.
Results: There were no significant differences in mental or psychomotor developmental indexes between LCPUFA-supplemented and control groups for all children or in subgroups. This was confirmed with adjustment for the possible confounders: sex, gestational age, birth weight, maternal age, and maternal smoking. The adjusted mean differences in mental developmental index and psychomotor developmental index for all of the children were −0.8 (95% confidence interval −2.8 to 1.2) and −1.0 (−2.7 to 0.7), respectively.
Conclusions: These data based on considerable sample size provide substantial evidence that LCPUFA supplementation of infant formula does not have a clinically meaningful effect on the neurodevelopment as assessed by Bayley scores at 18 months. Inclusion of all relevant data should not have led to differing conclusions except, possibly, for very-low-birth-weight infants.
*Division of Epidemiology, Institute of Social Paediatrics and Adolescent Medicine, Ludwig-Maximilians University of Munich, Munich, Germany
†Department of Paediatrics, Division of Developmental Neurology, University Medical Center Groningen, Groningen, The Netherlands
‡Medical Research Council, Institute of Child Health, London, United Kingdom
§University of Munich, Dr von Hauner's Children's Hospital, Munich, Germany.
Received 15 May, 2008
Accepted 30 April, 2009
Address correspondence and reprint requests to Andreas Beyerlein, MSc, Division of Epidemiology, Institute for Social Paediatrics and Adolescent Medicine, Ludwig-Maximilians University of Munich, Heiglhofstr 63, 81377 Munich, Germany (e-mail: email@example.com).
The 4 trials are being followed up at school age in the Early Nutrition Programming Project. The long-term follow-up of the 4 intervention trials as well as the project on the IPD meta-analysis are funded under the Food Quality and Safety Priority of the Sixth Framework Programme for Research and Technical Development of the European Community (FOOD-CT-2005-007036, www.metabolic-programming.org).
This study was supported by the Early Nutrition Programming Project, in which all co-authors are involved.
A.B. wrote the first draft, did the analysis, and is the guarantor. M.H.A. and H.B. were responsible for the Groningen trial, and M.F., K.K., A.S., and A.L. for the British trials. E.R. organized cooperation and data collection. M.H.A., M.F., K.K., B.K., and R.v.K. contributed to subsequent drafts of the article. A.B. and R.v.K. wrote the final draft together.
The authors report no conflicts of interest.