Precancerous Changes in the Cervix and Risk of Subsequent Preterm Birth

Abstract

Preterm birth is associated with increased perinatal morbidity and mortality and an elevated risk of disability. Women with precancerous changes of the cervix who have had treatment such as conization or a loop electrosurgical excision procedure (LEEP) are believed to be at increased risk of preterm birth. However, because many of the known risk factors for preterm birth, such as sexually transmitted disease, smoking, or prior induced abortion, also are associated with an increased risk of precancerous changes in the uterine cervix, it has been difficult to determine the degree risk due solely to cervical treatment from the degree of risk due to the other risk factors for preterm birth. This retrospective study was intended to compare rates of preterm birth in women having precancerous cervical changes to those of the general age-standardized population, and to detect any effect of cervical treatment on the time of delivery. Preterm deliveries, defined as those taking place before 37 weeks' gestation, were analyzed in 5548 women referred to a single institution in the years 1982–2000 who had either an abnormal Pap smear or a cervix that appeared abnormal. Compared with the general population of parturients, women referred to the dysplasia clinic were significantly more likely to have a preterm birth, whether or not they were treated. The standardized prevalence ratio was 2.0 (95% confidence interval [CI], 1.8–2.3) for treated women and 1.5 (95% CI, 1.4–1.7) for those not treated. This finding was verified when analysis was limited to women referred with positive cytology who had a confirmed lesion. Of the women seen in the dysplasia clinic, the adjusted odds ratio (OR) for preterm birth in treated versus untreated women was 1.23 (95% CI, 1.01–1.51), a significant association. Other factors increasing the risk of preterm birth included past abortion (induced or spontaneous), illicit drug use while pregnant, and major maternal disease. Cone biopsy, loop electrosurgical excision, and diathermy all were associated with an increased risk of preterm birth but, after adjusting for these possible confounding factors, the only significant correlation was with diathermy (OR, 1.72; 95% CI, 1.36–2.17). Laser ablation did not increase the risk of preterm birth. It appears that women who present with precancerous cervical changes are likelier to have a preterm birth, whether or not they have had treatment, but those who have had treatment that destroyed or removed a substantial amount of cervical tissue are at highest risk. Using precise methods of cervical ablation may help to reduce this risk.