Preterm birth is defined as delivery prior to 37 weeks’ completed gestation. It affects an average of 11% of pregnancies worldwide and is the leading cause of death in children under age 5. Many studies have identified associations between pregnancy lipid levels and increased risk for preterm birth. This thesis investigates the role of genetic variability associated with lipids and its relationship with preterm birth, and the relationship between pre-pregnancy dyslipidemia and risk for preterm birth.
Genetic variability in the form of single-nucleotide polymorphisms, previously identified by genome-wide association studies for association with lipid levels, was analyzed for association with risk for preterm birth. The study population included 992 women in California with banked 2nd trimester serum samples. Serum lipid levels and DNA were used. Genetic risk scores were constructed for each subject using published SNPs associated with lipid levels as an indicator of genetic burden. These genetic risk scores were then analyzed for association with risk for preterm birth. The GRS were not associated with the overall risk for preterm birth. However, a higher HDL-C GRS was associated with increased risk for spontaneous preterm birth. Higher triglyceride and total cholesterol GRS were associated with decreased risk for spontaneous preterm birth.
The relationship between pre-pregnancy dyslipidemia and risk for preterm birth was assessed in a cohort of 2,962,434 women giving birth in the state of California from 2007-2012. Dyslipidemia, as defined by medical diagnostic codes, was associated with a 1.5-fold increase in risk for preterm birth. This association was consistent across race/ethnicity, body mass index, type of dyslipidemia, and type of preterm birth.
This thesis identified counter-intuitive associations between lipid GRS and spontaneous preterm birth, while also identifying a strong relationship between pre-pregnancy dyslipidemia and all types of preterm birth including spontaneous. Together, these findings suggest that the previously reported associations between lipids and preterm birth may be reflecting unidentified dyslipidemias. One possible interpretation of the counter-intuitive genetic findings is that while extreme dyslipidemia predisposes to preterm birth a genetic predisposition to low total cholesterol also confers increased risk for spontaneous preterm birth. An alternative explanation is that these results are simply an artefact of the data and additional genetic loci and lifestyle factors confer stronger effects on risk for spontaneous PTB than the effects of the genetic loci included in this thesis.