Associations between cervical intraepithelial neoplasia during pregnancy, previous excisional treatment, cone-length and preterm delivery: a register-based study from western Sweden

摘要

Background Excisional treatment of cervical intraepithelial neoplasia (CIN) has been associated with increased risk of preterm delivery (PTD), although the underlying mechanism is as yet unclear. Studies on formalin-fixed excised tissue indicate that the risk increases with cone-length, but the magnitude of increase is uncertain, especially in case of minor excisions (<= 10 mm), as well compared to women with untreated CIN during pregnancy. This study assesses the impact of cone-length at previous treatment for CIN as well as diagnosis of CIN during pregnancy on the risk of PTD. Methods A register-based cohort study in western Sweden linking cervical cytology, histology, and treatment data from the Swedish National Cervical Screening Registry to data on obstetric outcomes in singleton pregnancies 2008-2016 from the Swedish Medical Birth Registry. These groups were compared for PTD and other obstetric outcomes: (1) women with one excisional treatment (n=3250, including a subgroup (n=2408) with cone-length measured before fixation; (2) women with untreated CIN diagnosed during pregnancy (n=1380); and (3) women with normal cytology (n=42,398). Logistic regression analyses were adjusted for socioeconomic and health-related confounders. Results Treated women had increased risk of PTD (adjusted odds ratio (aOR) 1.60, 95 confidence interval (CI) 1.21-2.12), spontaneous PTD (aOR 1.95, 95 CI 1.40-2.72) and preterm prelabor rupture of membranes (pPROM) (aOR 2.74, 95 CI 1.66-4.51) compared to the CIN during pregnancy group. ORs were similar when compared to the normal cytology group. Risks of these outcomes increased with cone-length. Mean cone-length was 9.1 mm. Cone-length <= 10 mm was associated with increased risk of PTD (aOR 1.41, 95 CI 1.02-1.94), spontaneous PTD (aOR 1.73, 95 CI 1.18-2.54), and pPROM (aOR 2.44, 95 CI 1.40-4.28), compared to the CIN during pregnancy group. The PTD risk was similar for cone-lengths 3-10 mm, thereafter increasing by 15 with each additional millimeter. Conclusions This study suggests that all excisional treatment, including small cones, are associated with increased risk of PTD and pPROM. Risks increase further with cone-length. In women of reproductive age, clinicians should aim to remove all CIN but minimal healthy cervical tissue. Cone-length should be recorded at treatment, for future prenatal risk estimation.