Endothelial dysfunction is a form of subclinical cardiovascular disease that may be involved in preterm birth and small-for-gestational-age deliveries. However, concentrations of biomarkers of endothelial dysfunction before pregnancy have rarely been measured. We hypothesized that higher levels of biomarkers of endothelial dysfunction (cellular adhesion molecules and selectins) would be associated with odds of preterm birth and/or small-for-gestational-age deliveries. We included 235 women from the Coronary Artery Risk Development in Young Adults (CARDIA) study who were nulliparous at Y7, reported ≥1 live birth through Y25, and had ≥1 biomarker measured at Y7. We tested for associations between individual biomarkers and an averaged z-score representing total endothelial dysfunction with preterm birth and/or small-for-gestational-age deliveries using Poisson regression, adjusted for demographic and clinical characteristics at the exam immediately preceding index birth. At Y7, total evidence of endothelial dysfunction was similar in women who did (n = 59) and did not have (n = 176) preterm birth and/or small-for-gestational-age deliveries. There was no association between biomarkers of endothelial dysfunction (either individual biomarker or total score) with odds of preterm birth and/or small-for-gestational-age deliveries after adjustment: IRR = 1.01, 95% CI: 0.74, 1.39, p = 0.93 for total endothelial biomarker score. Associations were not modified by race. We conclude that biomarkers of endothelial dysfunction in nulliparous women, measured ~3 years before pregnancy, did not identify women at risk for preterm birth and/or small-for-gestational-age deliveries. This suggests that the maternal endothelial dysfunction that is believed to contribute to these birth outcomes may not be detectable before pregnancy.