HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Lammi-Keefe, C. J Articles by Thoman, E. B Search for Related Content PubMed Articles by Lammi-Keefe, C. J Articles by Thoman, E. B Agricola Articles by Lammi-Keefe, C. J Articles by Thoman, E. B

Reply to HS Hansen and SF Olsen

Department of Nutritional Sciences
University of Connecticut
Storrs, CT 06269
E-mail: clammi{at}canr.uconn.edu

Evelyn B Thoman

Department of Biobehavioral Sciences
University of Connecticut
Storrs, CT 06269

Dear Sir:

We thank Hansen and Olsen for bringing to our attention their article on the effects of fish-oil supplements on the prolongation of gestation (1). In that study, Danish women in their 30th wk of pregnancy were randomly assigned to 1 of 3 groups to receive fish-oil capsules, olive oil capsules, or nothing. Fish-oil supplementation prolonged gestation by 4.0 d, on average, relative to gestation in the olive oil control group. Thus, fish-oil supplementation can prolong gestation.

In our study we reported significant associations between plasma phospholipid docosahexaenoic acid (DHA) concentrations in pregnant women and several sleep measures of their infants on days 1 and 2 of life (2). Because women with higher DHA concentrations are expected to have longer gestations, Hansen and Olsen wondered whether the associations with the sleep measures could be due, in part or entirely, to the prolongation of gestation rather than to DHA concentrations per se. They suggest determining 1) whether sleep measures correlate with gestational age (GA) and 2) whether the correlations disappear when adjusted for GA. These suggestions are necessary but not sufficient to yield meaningful interpretations of the data, because a reduction in a correlation between DHA and a sleep measure via an adjustment for GA may remove part of the effects of DHA because GA is dependent on DHA concentrations. To clarify this matter, a third statistical evaluation is needed: it needs to be determined whether the correlations between sleep measures and GA disappear after adjustment for DHA concentrations.

Correlations between DHA values (with the use of the ratio of the long-chain polyunsaturated fatty acids n-6 and n-3) in pregnancy and sleep measures on days 1 and 2 are shown in Table 1. In recalculating these correlations, we found, to our embarrassment, that we had failed to delete the data from 3 infants on day 2 because they lacked sufficient time in their cribs to give us reliable sleep measures. Therefore, the correlations in Table 1 are based on data from 17 infants on day 1 and from 14 infants on day 2. (Importantly, in our reanalysis without the data from the 3 infants who had insufficient crib time, we found no significant differences in the sleep measures or in the correlations. In fact, some of the correlations were even greater. Thus, our interpretation of the original data remains unaffected by our reanalysis.) We then obtained correlations between GA (determined by ultrasound in all but 2 cases) and the sleep measures. The first 2 rows of Table 1 list all of the significant correlations (P < 0.05) obtained with either DHA or GA. The third row lists the partial correlations of DHA and sleep measures with the linear effects of GA removed, and the fourth row lists the partial correlations of GA and sleep measures with the effects of DHA removed.

The partial correlations fell into 2 patterns. Two sleep states—active sleep and wake—remained correlated with DHA concentrations after the effects of GA were removed, whereas 2 measures of arousal—arousal in active sleep and in quiet sleep—and the ratio of active sleep to quiet sleep remained associated with GA after the association with DHA concentrations was removed. On day 1, the partial correlations of 2 sleep measures with both of the predictor variables (DHA and GA) were not significant. This finding indicates that DHA indirectly affects these sleep measures via an increase in GA.

These analyses suggest the existence of 2 independent pathways and 1 interdependent pathway that relate DHA and GA to sleep measures in infants. Further research is necessary to determine the validity and generality of these conclusions, particularly in light of the lack of effect of DHA supplementation on GA in the study of Olsen et al, in which the comparison of fish-oil supplementation was made with women who received no oil (1), and in other studies (3, 4).

REFERENCES

1. Olsen SF, Sorensen JD, Secher NJ, et al. Randomised controlled trial of effect of fish-oil supplementation on pregnancy duration. Lancet 1992;339:1003-7.[Medline]
2. Cheruku SR, Montgomery-Downs HE, Farkas SL, Thoman EB, Lammi-Keefe CJ. Higher maternal plasma docosahexaenoic acid during pregnancy is associated with more mature neonatal sleep-state patterning. Am J Clin Nutr 2002;76:608-13.[Abstract/Free Full Text]
3. Onwude JL, Lilford RJ, Hjartardottir H, Staines A, Tuffnell D. A randomized double blind placebo controlled trial of fish oil in high risk pregnancy. Br J Obstet Gynaecol 1995;102:95-100.[Medline]
4. Helland IB, Saugstad OD, Smith L, et al. Similar effects on infants of n-3 and n-6 fatty acids supplementation to pregnant and lactating women. Pediatrics 2001;108:82-91.

This Article Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Lammi-Keefe, C. J Articles by Thoman, E. B Search for Related Content PubMed Articles by Lammi-Keefe, C. J Articles by Thoman, E. B Agricola Articles by Lammi-Keefe, C. J Articles by Thoman, E. B