BETHESDA, Md. -- Low maternal cholesterol early in pregnancy may increase the risk of preterm delivery among otherwise low-risk white women, researchers found.
BETHESDA, Md., Oct. 1 -- Low maternal cholesterol early in pregnancy may increase the risk of preterm delivery among otherwise low-risk white women, researchers found.
Total serum cholesterol below the 10th percentile in the second trimester was the strongest predictor of preterm delivery in low-risk mothers, reported Maximilian Muenke, M.D., of the National Human Genome Research Institute of the National Institutes of Health here, and colleagues in the October issue of the journal Pediatrics.
The association was almost entirely among white women (odds ratio 5.63, P<0.0001), with no significant effect among black women (OR 0.81, P=0.79).
Because risk was also elevated for high maternal serum cholesterol in this and previous studies, "the concept of an optimal range for maternal serum cholesterol during pregnancy may have merit," the researchers wrote.
"These results could support the implementation of nutritional interventions among pregnant women to reduce the risk of preterm birth," particularly in malnourished populations, commented Mario Merialdi, M.D., Ph.D., of the World Health Organization, and Jeffrey Murray, M.D., of the University of Iowa in Iowa City, Iowa, in an accompanying editorial.
However, "given the well-recognized risks between cardiovascular disease and elevated cholesterol, interventions to raise cholesterol might best be carried out only in high risk, low serum cholesterol populations of women early in pregnancy," they wrote.
Because no studies had looked at a lower boundary for cholesterol in pregnancy, the researchers conducted a pilot study among a cohort of women referred to South Carolina prenatal clinics for routine second-trimester serum screening.
The retrospective study included 118 women with total cholesterol below the 10th percentile (159 mg/dL at a mean gestational age 17.6 weeks) and 940 women with higher total cholesterol.
The researchers included only nonsmoking women ages 21 to 34 without diabetes who gave birth to a single, live infant. Total cholesterol was adjusted for gestational age at screening.
The 6.6% prevalence of preterm delivery (before 37 weeks gestation) was "considerably less" than the 12% rate seen in the South Carolina counties from which the cohort was drawn, testifying to the highly selected study population.
Preterm birth was 2.93-times more common with low total maternal cholesterol than with midrange cholesterol (159 to 261 mg/dl) after adjustment for potential confounders such as maternal age (12.7% versus 5.0%, P=0.001).
Preterm birth was even more common among women in the third percentile or lower for total cholesterol (16.2% preterm, OR 3.62 versus midrange cholesterol, P=0.008), particularly for white women (28.6% preterm, OR 7.69, P=0.0003).
As expected from previous studies, high maternal cholesterol (greater than 90th percentile, 261 mg/dL at 17.6 weeks gestation) was associated with elevated risk of preterm birth as well (OR 2.66, P=0.003).
Other differences correlated with cholesterol levels were also found. Term infants born to mothers with low total cholesterol had a 147-g lower average birth weight than those born to mothers with higher cholesterol after adjustment for specific gestational age, infant gender, and other factors (P=0.0006).
Microcephaly was also about twice as common among infants born to mothers with low cholesterol, but the trend was not significant. Congenital anomalies were not associated with total maternal serum cholesterol.
Although the pilot study included a highly selected population and requires validation, "the consistent finding that low maternal serum cholesterol was the strongest predictor of preterm delivery in every model tested" supports the validity of the association, Dr. Muenke and colleagues said.
However, "Whether this finding can be replicated or extended to other populations of pregnant women will be essential to examine," they added.
The mechanism remains "speculative, but biologically credible," Drs. Merialdi and Murray said, because cholesterol is a precursor of placental progesterone critical to maintaining pregnancy and is a major component of plasma membranes in the placenta.
The authors pointed out that "limitations are inherent in our methods and findings. The specific numeric thresholds for low and high total cholesterol are based on a mean sampling gestational age of 17.6 weeks and cannot be generalized to other populations with differing profiles of gestational risk. The ascertainment of potential study subjects was incomplete, and the low maternal serum cholesterol group differed in baseline characteristics from the control group: the specific effects of any resulting selection biases cannot be predicted."