| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
LETTER TO THE EDITOR |
South East Sydney Illawarra Area Health Service
PO Box 1958
Wollongong West, NSW 2500
Australia
E-mail: robert.moses{at}sesiahs.health.nsw.gov.au
Department of Statistics
Macquarie University
Sydney, NSW
Australia
Human Nutrition Unit
University of Sydney
Sydney, NSW
Australia
Dear Sir:
In 2006, we reported in the Journal (1) the results of a study examining the effects of a low-glycemic-index (low-GI) diet on pregnancy outcomes. Women who followed a low-GI diet from the beginning of the second trimester had a lower rate of large-for-gestational age (LGA) infants (P = 0.01), a lower birth centile (P = 0.005), and a lower ponderal index (P = 0.003). Women who completed the study were not asked to continue their respective diets or offered any specific dietary advice for the post-partum period. However, a copy of the published report with an explanation of the results was forwarded after publication.
The aims of this follow-up study were twofold. The first was to ascertain whether women who followed a low-GI diet during pregnancy had continued with this diet. The second was to ascertain whether there were any differences in infant size related to maternal diet during pregnancy. The women and infants were enrolled in the follow-up study 
2 y (mean: 22 mo; range: 16–29 mo) after the completion of the original study.
Analyses were carried out using SPSS software (version 14; SPSS Inc, Chicago, IL). Independent-sample t tests were used to compare the characteristics of mothers who were available for follow-up with the characteristics of those who were not available. Repeated-measures analyses of variance were used to investigate group differences between mothers at the end of the original study and in the follow-up study.
Analysis of variance models were used to investigate nutritional differences at the time of follow-up between mothers in the original low-GI and conventional GI (high fiber and low sugar) groups, with adjustment for possible differences due to current pregnancy. Analysis of variance models were also used to investigate differences in the characteristics of the infants (eg, height and weight) between mothers in the original low-GI and higher-GI groups: in this case, the current age of the infant and the months of breastfeeding were included as covariates. P < 0.05 was taken to indicate statistical significance. Results were expressed as means ± SEMs.
Of the 62 women who completed the original study, 19 (30%) did not wish to participate in the follow-up study. The women who did not wish to participate were of a similar age and body mass index (BMI; in kg/m2) but had a higher parity (1.16 ± 0.21 and 0.63 ± 0.11, respectively; P = 0.014) than did the 43 women who agreed to participate and complete a 3-d food record. Of those 43 women, 23 had been in the low-GI treatment group in the original study. Nine (39%) of these 23 women currently were pregnant, whereas only 3 (15%) of the 20 women in the previous higher-GI group were pregnant. The rates of pregnancy did not differ significantly (P = 0.099).
The 23 women in the previous low-GI group had a current GI of 55.30 ± 1.04, which did not differ significantly from the current GI of 56.43 ± 1.25 in the 20 women who had been in the previous higher-GI diet group. In addition, there were no significant differences between the groups with respect to glycemic load, kilojoules, or carbohydrate, fat, protein, or fiber content. Pregnancy also did not make any significant difference. The current GI of the diet for both groups did not differ significantly from the GI of the diet at the start of the original study.
For infant size (weight and height), the age is significant (P = 0.001), but neither the group (P = 0.09) nor the number of months of breastfeeding (P = 0.30) is significant. In the original study, there were 11 LGA infants (10 from women following the higher-GI diet). Nine of these infants were in the follow-up group. LGA was a significant (P = 0.037) predictor of current infant weight after adjustment for group and age (13.6 kg compared with 12.4 kg for the infants who were not LGA at birth). With the same adjustments, infant height was not significantly different (LGA: 87.5 and 84.2 cm, respectively; P = 0.125).
We conclude that women who were able to follow either diet during pregnancy did not wish to continue after the end of the study. All of the women reverted to a diet with a composition that was approximately the same as the diet they were following before enrollment in the original study. They made this change despite having been informed of the apparent health advantages of a low-GI diet and despite the fact that 9 of the 23 women had become pregnant again.
Women on the higher-GI diet had infants with a higher birth centile and a significantly higher rate of LGA. The concern is that unfavorable maternal nutrition during pregnancy may lead to unfavorable metabolic changes in the offspring. The numbers in this evaluation are small, women who were on a higher-GI diet were more likely to have an LGA infant, and these infants were heavier but not taller at a mean of 22 mo.
ACKNOWLEDGMENTS
None of the authors had a personal or financial conflict of interest.
REFERENCE
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
