Meta-analysis of dietary essential fatty acids and long-chain polyunsaturated fatty acids as they relate to visual resolution acuity in healthy preterm infants

Objective. To derive combined estimates of visual resolution acuity differences between healthy preterm infants consuming different compositions and ratios of essential fatty acids (EFAs) and docosahexaenoic acid (DHA), an omega-3 (n-3) long-chain polyunsaturated fatty acid (LCPUFA). Data Sources. Electronic biomedical reference database (Medline and Health Star from 1965 to July 1999) searches with index terms omega-3, n-3, infant, vision, acuity, and human. Current review article, monograph, and book chapter bibliography/reference section hand searches. Study Selection. A total of 5 original articles and 4 review chapters were reviewed for details on study design, conduct, and outcome. Four prospective trials of EFA/LCPUFA supplementation were included in these analyses. For behaviorally based outcomes, there were 2 randomized comparisons each at ≤1, 2, 6, 9, and 12 months of corrected age and 4 randomized comparisons at 4 months of corrected age. For electrophysiologically based outcomes (visual-evoked potential), there were 2 randomized comparisons each at ≤1 and ~4 months of corrected age. Data Extraction. Dietary composition and EFA/ LCPUFA balance, study design, and analytic characteristics (duration of feeding, source of EFAs/LCPUFAs, number of subjects in study population, number of subjects analyzed, and basis for estimating age), and experiment-based characteristics (location, number or sites, design, vision tests employed, testing protocol, and ophthalmic examination) were recorded independently by 2 researchers with a standardized protocol. Data Synthesis. The relative difference in visual resolution acuity between groups of infants who received a source of dietary EFAs/LCPUFAs and groups who did not was computed and then analyzed with the DerSimonian and Laird random-effects method. Results. Analysis of the randomized comparisons (DHA-supplemented formula vs DHA-free formula) showed significant differences in visual resolution acuity at 2 and 4 months of age. Combined estimates of behaviorally based visual resolution acuity differences at these ages were .47 ± .14 octaves and .28 ± .08 octaves, respectively. A 1-octave difference is a reduction in the width of the stimulus elements by 50%. Conclusion. These results support efficacy of n-3 LCPUFA intake in early visual system development, although supplementation safety issues still must be addressed through larger randomized trials. Whether n-3 intake confers lasting advantage in visually based process development across the life-span is still to be determined.