Study design

We conducted a retrospective case-control study. All women who had previously given blood for routine genetic multiple marker screening and subsequently delivered at the University of North Carolina–Chapel Hill from January 2004 to November 2008 were eligible. This study was approved by the institutional review board of the University of North Carolina–Chapel Hill prior to data collection, and permission was obtained to use banked serum from these women for research purposes.

Healthy women with term deliveries (37 weeks or longer) were used as controls and matched 3:1 by age, race, parity, and tobacco use to women with spontaneous preterm birth. Cases were women with spontaneous preterm birth defined as noniatrogenic delivery 23 0/7 or longer and gestation of 34 0/7 weeks or less.

Exclusion criteria included medically indicated preterm delivery, multiple gestation, major congenital fetal anomalies, placenta previa, preeclampsia, pregestational hypertension, kidney disease, diabetes mellitus, known thrombophilias, or any other significant preexisting chronic medical disease.

Nonfasting blood samples were collected for routine genetic multiple marker screening between 15 and 20 weeks’ gestation, and serum aliquots were barcoded and frozen at –70°C until assayed. Maternal demographic and medical data were chart abstracted.

Laboratory analyses

PON 1 activity in maternal serum was determined by the rate of hydrolysis of 2 substrates: paraoxon (1 mM; PON 1 activity) and phenylacetate (5 mM; arylesterase activity), respectively.

PON 1 activity in serum was determined as described previously, using paraoxon as a substrate and by measuring the increase in absorbance at 412 nm because of the formation of 4-nitrophenol. Briefly, a concentrated stock solution of paraoxon (120 mM) was prepared by dissolving paraoxon in acetone. A working solution of paraoxon (2 mM) was then prepared by diluting the concentrated paraoxon stock solution with distilled water. A 100 μL aliquot of paraoxon assay buffer (100 mM Tris-HCl, 2 mM CaCl ₂ , pH 8) was added to each well of a 96-well plate. A 6 μL aliquot of each serum sample was then added to the wells of the microtiter plate in quadruplicate.

To initiate the reaction, a 100 μL volume of the working solution of paraoxon was added to each well using a manifold pipette. The solution was mixed and incubated at 25°C for 1 minute. The absorbance at 412 nm was then monitored at 11 second intervals for 5 additional minutes using a VERSA _Max tunable microtiter plate reader (Molecular Devices, Sunnyvale, CA). The rate of increase in absorbance was linear throughout this period. PON 1 activity catalyzed by the PON 1 enzyme was reported as the change in absorbance at 412 nm over time (dA per minute). Inter- and intraassay coefficients of variation averaged 4.5% and 3.5%, respectively, for this assay.

The arylesterase activity of PON 1 is used as a proxy for enzyme concentration and was determined using phenylacetate as a substrate as described previously. Briefly, a solution of phenylacetate (15 mM) was prepared daily and maintained at 25°C. To conduct the assay, a 2 mL aliquot of arylesterase assay buffer (30 mM Tris-HCl, 1.5 mM CaCl ₂ , pH 8) was combined with 5 μL of serum in a cuvette and incubated in triplicate at 25°C for 5 minutes in the Peltier temperature-controlled compartment of a spectrophotometer (DU640; Beckman Instruments, Inc, Palo Alto, CA).

To initiate the reaction, 1 mL of the phenylacetate solution was forcibly added to the sample, and the solutions were mixed by rapidly aspirating the solution 3 times into a disposable pipette. The rate of increase in absorbance at 270 nm was then monitored at 15 second intervals for 3 minutes and was linear throughout this period. The arylesterase activity of the PON 1 enzyme was reported as the change in absorbance at 270 nm over time (dA per minute). Inter- and intraassay coefficients of variation averaged 1.5% and 2.8%, respectively, for this assay.

Lipid profiles were analyzed using the Cholestech LDX System (Lipid Profile Cassette, Hayward, CA). Specific measures included total cholesterol (TC), triglycerides, very low-density lipoprotein (VLDL), LDL, and HDL. The coefficients of variation for the different lipid parameters range from 2% to 4%.

Statistical analysis

Data were analyzed using SAS software (version 9.1.3; SAS Institute, Inc, Cary, NC). χ ² and Fisher’s exact test were used for analysis of categorical variables; the Student t test was used to compare group means. P < .05 was considered statistically significant.