The purpose of this study was to determine the relationship between maternal inflammation and first or second trimester pregnancy loss. A nested case-control analysis was performed among the cohort enrolled in the Oral Conditions and Pregnancy study. We compared maternal serum C-reactive protein concentration between women with a pregnancy loss at < 21 weeks gestation to control women without gestational diabetes or preeclampsia who delivered at term. Participants were 2:1 frequency matched to cases by maternal age, race, and prior preterm birth. Median concentration of serum C-reactive protein between cases and controls was compared using Wilcoxon rank sum test. The potential effects of maternal smoking, gestational age at blood draw, and insurance status were evaluated and an adjusted odds ratio (95% confidence interval) for pregnancy loss was calculated using multivariable logistic regression. Among 1224 participants, 102 (9.8%) experienced pregnancy loss and 44 had complete information available. Median serum C-reactive protein concentration was significantly higher in controls compared with all cases (3.2 versus 0.5 microg/mL; p < 0.001). However, when stratified by gestational age at the time of the loss, median serum C-reactive protein level among controls was similar to those with a loss at less than 12 weeks (3.2 versus 2.0 microg/mL) but significantly higher compared with those whose loss occurred at greater than 12 weeks gestation (3.2 versus 0.5 microg/mL; p < 0.05). After adjusting for maternal smoking, gestational age at blood draw, and insurance status, women whose serum C-reactive protein level was greater than the 75% percentile had a decreased odds ratio for pregnancy loss (0.20; 95% confidence interval, 0.06 to 0.65). Pregnancy loss is not associated with increased systemic inflammation as measured by maternal serum C-reactive protein. Future study should be directed at determining the role of maternal inflammation during early pregnancy development and placentation.