Source: Shiell AW, et al. High-meat, low-carbohydrate diet in pregnancy. Relation to adult blood pressure in the offspring. Hypertension 2001;38:1282-1288.
A study of 626 men and women in Motherwell, Scotland, evaluated maternal dietary characteristics during pregnancy on blood pressure of progeny at age 27-30 years. Mothers had participated in a dietary intervention program in which they had been advised to eat 1 lb of red meat per day and to avoid carbohydrate-rich foods.
The offspring of women reporting greater consumption of meat and fish during the second half of pregnancy had higher systolic blood pressures as adults. The regression coefficient was 0.19 mm Hg per portion per week. Diastolic blood pressure in adults was associated with higher maternal pregnancy consumption of fish but not meat; the regression coefficient was 1 mm Hg per serving per week.
The associations were independent of maternal blood pressure, body size, and tobacco use. The associations may reflect metabolic stress associated with high essential amino acid intake without adequate nutrients to utilize those amino acids.
Comments by Anthony R. Scialli, MD
It would not be surprising to learn that maternal nutrition is important in pregnancy outcome; however, the proposal that meat intake during pregnancy plays an important role in the health of adult offspring is not intuitive. Certainly, low birth weight has been shown to be associated with hypertension in later life, but low birth weight might be due to severe maternal undernutrition or to a number of other pregnancy disorders. Fetal starvation might be imagined to produce compensatory changes that could result in long-lasting alterations in an individual’s cardiovascular system. But the proposition that normally grown babies might be affected three decades later by imbalances in maternal diet appears far-fetched given the important postnatal factors that influence the health of adults, such as diet, exercise, and the use of tobacco and ethanol.
The test of the influence of a high meat diet during pregnancy was made possible by advice given to women in parts of Scotland that preeclampsia could be prevented by a high intake of animal protein, specifically red meat. The idea appears foolish in the 21st century, particularly in view of the current notion that animal products are highly toxic to all human life forms, but between 1952 and 1976, it was one of the many ridiculous things that obstetricians told pregnant women.
Compliance with the dietary advice was assessed during prenatal visits using specially designed forms for recording diet. Numbers of servings per week were recorded for 10 food groups: meat, fish, eggs, cheese, green vegetables, potatoes, bread, cakes (including scones and biscuits), sweets, and milk. Women were discouraged from eating potatoes, rolls, scones, cakes, or biscuits, under the assumption that carbohydrate-rich foods were bad. Fish, eggs, and cheese were encouraged and 1 lb of red meat per day was prescribed. The evaluation of offspring ages 27-30 years included information on ethanol intake. Height, weight, and blood pressure were measured by field workers. Three blood pressure readings were taken using an automated cuff, and the average of the three was used. Cuff size, sex, alcohol intake, and body mass index were adjusted with linear regression analysis.
The relationship between maternal intake of meat and fish and blood pressure was evaluated using multiple regression techniques. Thus, there was no control group in a traditional sense; rather, the relationship between dietary intake and blood pressure was evaluated as a continuous correlation. Other correlations also were evaluated. For example, number of servings of carbohydrate was inversely proportional to pregnancy weight gain.
The increase in systolic blood pressure with meat and fish, and the increase in diastolic blood pressure with fish were modest but statistically significant. The effect was particularly noteworthy if the mother had a low intake of green vegetables. Mean systolic blood pressure in offspring of women who consumed 11 or fewer portions of meat and fish per week and at least seven servings of green vegetables per week was 118, compared to a mean blood pressure of 124 in women consuming more than 21 servings of meat and fish per week and fewer than seven servings of green vegetables.
Is a 6 mm Hg increase in systolic blood pressure important? It could be, if it represented a shift in blood pressure distribution upwards. Even though 124 is a normal blood pressure, an equivalent increase across the blood pressure distribution would result in an increase in the number of people diagnosed as hypertensive, with consequent morbidity either from the effects of the high blood pressure or from the effects of the medications used to treat high blood pressure.
How reliable are the results of this study? Could women have made up their dietary intakes at the time of their prenatal visits, perhaps to please the staff or to avoid being scolded?
This possibility is unlikely to explain the results, because such an explanation would require a made-up dietary intake to correlate with offspring blood pressure 30 years later. Also, the correlation between meat intake and blood pressure held only for the second half of pregnancy, and not for the first half, suggesting a physiologic event unrelated to maternal lying. In addition, urinary protein excretion was measured in 10 women and showed the expected correlation with reported meat intake. Finally, this study was undertaken after publication of a study from Aberdeen, in which reported maternal meat consumption showed a similar relation to offspring blood pressure. In the Aberdeen study, dietary intakes were recorded without the women having received any particular dietary instruction, and represented freedom of food choice.
The authors’ proposal that high animal protein intake during pregnancy can result in long-lasting cardiovascular changes in the fetus is based on the premise that excess essential amino acids require oxidation, a folate-dependent process. In women with low intake of potatoes and green vegetables, inadequate folate is postulated to have resulted in nutritional "stress" (their term) and consequent maternal cortisol secretion. How exactly maternal cortisol would act on the fetal cardiovascular system in a long-lasting manner is not stated; the image of a fetus on steroids is as far as the authors take us mechanistically.
There are other explanations that have not been adequately considered. The first is the salt content of the maternal diets. The specific instruction to the pregnant women in this clinic was to use canned corned beef in place of fruit or biscuits for between meal snacks. The amount of corned beef or other canned meat that was used by these women was not evaluated, and salt seems as attractive a villain as essential amino acids. To be fair, the authors considered this possibility, but could go nowhere with it given the diet records available to them.
A more compelling concern is that of maternal wealth. Eating 1 lb/d of red meat costs more than eating green vegetables. It is likely that women in this clinic were equally brow-beaten about their diets, and perhaps only the wealthier of clients could afford to comply. If so, the study might be more a correlation of family wealth with blood pressure. Family wealth could certainly have effects on health that extended decades after birth.
Finally, there is a possibility that maternal dietary habits during pregnancy were taught to offspring. Kids born to women who used canned corned beef instead of fruit may be more likely themselves to adopt such dietary preferences. Although offspring body mass index was included in the regression, dietary fat and salt intake in the offspring may still have played a role.
In spite of these alternative explanations, this paper represents a particularly competent study in which statistical analyses were used as tools and aids rather than as poorly understood window-dressing. The most important negative comment goes to forcing this sentence on an otherwise flawless presentation: "An expanded Methods section can be found in an online data supplement available at http://www.hypertensionaha.org." Call me old-fashioned, but I still read paper. Given the importance of the methods section in the evaluation of a scientific paper, I expect editors to give me everything.