CHICAGO — A prospective study of 5,300 women has provided the first biological evidence of the mechanisms underlying the statistically established association between preterm delivery and the mother's future risk of heart disease and stroke.
Low HDL cholesterol and elevated homocysteine levels surfaced as key factors associated with spontaneous preterm birth, Dr. Michael S. Kramer of McGill University reported in a plenary session at the annual meeting of the Society for Pediatric and Perinatal Epidemiologic Research.
In addition, a significantly higher proportion of women with concentrations of homocysteine above the median showed signs of decidual vasculopathy (13.0% vs. 6.8%), Dr. Kramer said.
The study compared frozen plasma samples and fixed and stained placentas from 207 cases of spontaneous preterm birth with 444 term controls, approximately 2 per case.
Researchers analyzed homocysteine, folate, cholesterol (total, LDL, and HDL), and thrombin-antithrombin complexes and blindly assessed fixed and stained placentas for histologic evidence of infarction and decidual vasculopathy.
Both elevated homocysteine and low HDL cholesterol levels were significantly and independently associated with twice the risk of preterm birth, Dr. Kramer reported. “Similar vasculopathic risk factors may underlie preterm birth and adult coronary heart disease and stroke,” he said.
Women who delivered preterm had significantly higher plasma homocysteine (4.0 vs. 3.7 mmol/L; P = .001) and lower HDL cholesterol (1.6 vs. 1.8 mmol/L; P = .0001) levels, compared with women who delivered at term.
In addition, a higher proportion of women with high homocysteine concentrations (but not low HDL) had decidual vasculopathy (13.0% vs. 6.8%).
“The same factors that we know lead to stroke and heart disease were found to be elevated in the second trimester in mothers who subsequently gave birth preterm,” said Dr. Kramer in an interview.
The fact that their placentas showed evidence of vasculopathy on the mother's side was a major finding, because it provides a biological link with the vasculopathic plasma markers, he said.
However, “even if [these results] are robust, we still don't know whether homocysteine and HDL are pathologically involved in a biological sense with the preterm birth, or whether they're just markers of the mother's increased risk,” he said.
“In adults, when HDL and homocysteine damage blood vessels, they do it over decades,” he said. “With pregnancy, we're talking about months. How do [these factors] get translated biologically into an increased risk for preterm birth? It may be the homocysteine and HDL themselves that are acting on blood vessels in the placenta, or it may be something else that's causing the preterm birth.”
Dr. Kramer noted that the differences in HDL and homocysteine levels between the two groups were statistically significant but modest. For example, there was a less than 10% difference between the cases and controls in homocysteine (4.0 vs. 3.7 mmol/L). In addition, “the homocysteine concentrations were not high in terms of what is known or suspected to cause vascular damage, which is why we're underlining the fact that we don't know if it's the homocysteine,” he said. “These were not the sky-high levels associated with very high risks of coronary heart disease.”
The findings need to be replicated to determine whether they are robust, Dr. Kramer said. “However, I think it's unlikely that they were just a statistical fluke, because they were in the direction you'd expect,” he said.
Existing serum banks for large populations would offer a relatively easy and inexpensive method of linking pregnancy outcomes with HDL and homocysteine concentrations, he said.