CHICAGO – A diet high in fiber and low in fat reduces the risk of gestational diabetes appreciably, as does exercise, Michelle A. Williams, Sc.D., said at the annual scientific sessions of the American Diabetes Association.
The number of observational studies that have shown that diet and exercise can affect the risk of gestational diabetes has reached a critical mass that should be taken seriously, said Dr. Williams, professor of epidemiology at the University of Washington, Seattle.
There is an “impressive consistency” in the data, she said.
“This is a time for action, and we should be thinking about moving toward translational approaches from this research,” she said. “Pregnancy really is an ideal time for promoting healthful activities.”
A connection between a high-fat diet and gestational diabetes was first documented in an observational study in 1997 (Diabetes Care 1997;20:1647-50). Since then, two subsequent studies have confirmed that connection. One study did not, but it involved women in China, where high-fat diets include polyunsaturated fat, she said.
In a United States prospective cohort study by researchers at the University of North Carolina at Chapel Hill, 1,698 pregnant women were asked about their diet. The investigators found that women who had gestational diabetes had a higher-fat diet than those who did not, with a mean percentage of 35% of their total calories from fat, compared with 33% for pregnant women without gestational diabetes (Am. J. Clin. Nutr. 2004:79:479-86). The researchers estimated that a woman with a fat intake of 40% of total calories had more than twice the risk of gestational diabetes than one whose fat intake was 20% of total calories. This was true even when the number of calories remained exactly the same. At the 40% fat level, more than 10% of women could be expected to have gestational diabetes.
In 2004, Dr. Williams' group reported results of a survey suggesting that low vitamin C intake also resulted in increased risk of gestational diabetes. After interviewing 67 patients with gestational diabetes and 260 controls, they reported that women who took fewer than 70 mg of vitamin C a day–an inadequate intake–had a 3.7 times higher risk of gestational diabetes than those who took more (J. Reprod. Med. 2004;49:257-66).
The group followed up that study by actually measuring plasma ascorbic acid in 755 women at an average 13 weeks' gestation. That study showed that the women in the lowest quartile of plasma ascorbic acid had a rate of gestational diabetes of 9%, compared with a rate of 3% in women in the highest quartile (Epidemiology 2004;15:597-604).
Regarding fiber in the diet, a group that looked at Nurses' Health Study data found that when the women had an intake of 22 gm or more of fiber per day prior to their pregnancy, their risk of gestational diabetes was reduced by 33% (Diabetes Care 2006;29:2223-30).
Dr. Williams said she found much the same reduction in risk in her own cohort of 288 cases of gestational diabetes and 444 controls. The data, which have not yet been published, shows that women in the highest quintile for fiber intake had a 40% reduction in risk of gestational diabetes, relative to women in the lowest quintile, and that the risk reduction followed a very linear trend down from the highest quintile. Less than half of the women, cases and controls alike, met the recommendation that the daily diet should include five fruits and vegetables.
“Suffice it to say, if they were consuming five a day, their risk for gestational diabetes would [have been] reduced quite substantially,” she said.
Six observational studies have shown a benefit from exercise before and during pregnancy, Dr. Williams said. One found a risk reduction of 47% when women who were obese exercised during pregnancy (Am. J. Epidemiol. 1997;146:961-65).
Analysis of her own data on 155 cases and 386 controls showed a 55% reduction in risk of gestational diabetes in those women who reported that they engaged in regular leisure time physical activity prior to their pregnancy. Leisure time physical activity during pregnancy resulted in a 48% reduction in risk, she added.
Dr. Williams' data also address the question of whether activity must be vigorous to produce a benefit, or whether it can be moderate, she said. Her analysis found that risk was reduced by 71% when that activity was vigorous, but risk was reduced by only 35% with moderate activity, such as walking.
In observational studies, it is probably impossible to control the analysis completely for variables that may confound the investigators' conclusions, hard as the researchers may try, Dr. Williams noted.