To estimate the relationship between maternal periodontal disease and both early spontaneous preterm birth and selected markers of upper genital tract inflammation. In this case-control study, periodontal assessment was performed in 59 women who experienced an early spontaneous preterm birth at less than 32 weeks of gestation, in a control population of 36 women who experienced an early indicated preterm birth at less than 32 weeks of gestation, and in 44 women with an uncomplicated birth at term (>or = 37 weeks). Periodontal disease was defined by the degree of attachment loss. Cultures of the placenta and umbilical cord blood, cord interleukin-6 levels, and histopathologic examination of the placenta were performed for all women. Severe periodontal disease was more common in the spontaneous preterm birth group (49%) than in the indicated preterm (25%, P =.02) and term control groups (30%, P =.045). Multivariable analyses, controlling for possible confounders, supported the association between severe periodontal disease and spontaneous preterm birth (odds ratio 3.4, 95% confidence interval 1.5-7.7). Neither histologic chorioamnionitis, a positive placental culture, nor an elevated cord plasma interleukin-6 level was significantly associated with periodontal disease (80% power to detect a 50% difference in rate of histological chorioamnionitis, alpha = 0.05). Women with early spontaneous preterm birth were more likely to have severe periodontal disease than women with indicated preterm birth or term birth. Periodontal disease was not associated with selected markers of upper genital tract inflammation. II-2