Abstract
OBJECTIVE: To investigate the risk for preterm birth associated with vaginal infections in pregnancies after a loop electrosurgical excision procedure (LEEP), compared with women with no prior LEEP. DESIGN: Multicentre retrospective cohort study. SETTING: USA. POPULATION: Women with LEEP between 1996 and 2006 were compared with two unexposed groups who had cervical biopsy or Pap test, without any other cervical procedure, in the same calendar year. METHODS: The first pregnancy progressing beyond 20 weeks of gestation in women with prior LEEP was compared with pregnancy in women without LEEP. Stratified analysis according to the presence or the absence of vaginal infection during pregnancy was used to investigate whether the risk for preterm birth differed according to the presence or the absence of infection. The interaction between LEEP and vaginal infection was investigated using multivariable logistic regression with interaction terms, as well as the Mantel-Haenszel test for homogeneity. MAIN OUTCOME MEASURES: Spontaneous preterm birth (<37 and <34 weeks of gestation). RESULTS: Of 1727 patients who met the inclusion criteria, 34.4% (n = 598) underwent LEEP prior to an index pregnancy. There was no increased risk for vaginal infections among women with LEEP compared with women without LEEP. Chlamydia infection and LEEP demonstrated significant interaction, suggesting that the presence of chlamydia infection in women with a history of LEEP augments the risk for preterm birth, compared with women with no history of LEEP. CONCLUSIONS: Vaginal infections during pregnancy in women with a history of LEEP may be associated with an increased risk for preterm birth, compared with women with no history of LEEP.