Preterm delivery is a major cause of neonatal morbidity and mortality1 that is increasing in incidence,1 and a major economic burden.2 The cost of preterm birth in England and Wales has recently been estimated at £2.9 billion a year.2 There is increasing evidence that excisional procedures for treatment of cervical intraepithelial neoplasia (CIN) are associated with preterm delivery3 and subsequent perinatal mortality.4 The two original articles in this issue add to the existing data and highlight the problem of obstetric management in pregnancies subsequent to excisional cervical surgery. The small but detailed study by Van de Vijver et al.,5 on women in Belgium found that conisation of the cervix was associated with a seven times higher relative risk of preterm birth before 37 weeks. Because of its small sample size, it did not have enough power to detect a statistically significant increased risk of more serious preterm delivery, before 34 weeks. However, although delivery between 34 and 37 weeks is rarely associated with significant sequelae,2 the study does add weight to the findings of a recent meta-analysis that found a significant association between excisional cervical surgery and preterm birth and consequent perinatal mortality.3,4 A larger study also reported in this issue by Ørtoft et al.,6 from Denmark also supports the meta-analysis by finding a three-fold increased risk of preterm delivery before 37 weeks after one cervical conisation and a ten-fold increased risk after two conisations. These Danish data add evidence to the concept that the more cervix that is removed the worse the risk of preterm labour.3 Importantly, they report similarly increased risks for severe preterm delivery (before 28 weeks of gestation) and perinatal mortality.
A potential problem with the reported association between cervical surgery for CIN and preterm birth is that of confounding factors as both preterm birth and CIN are known to be more prevalent in smokers and women of poor social economic class.1 However, Ørtoft et al.6, accounted for these confounding factors in their regression analysis and compared pregnancy outcomes in the same women before and after cervical surgery and still found an increased risk of preterm birth.6 Consequently, both studies in this issue support the concept that it is excisional cervical treatment (as opposed to ablation) that is associated with preterm labour and delivery. The mechanism by which this association occurs is not yet clarified. Ascending infection, from the vagina into the fetal–placental unit is commonly seen and presumed to be causative in preterm labour. The cervix is thought to be an important physical and immunological barrier to ascending infection,7 hence any weakening of the cervical barrier could increase the risk of infection and preterm labour and delivery.
The question that arises is how should obstetricians manage women who have had excisional cervical treatment for CIN in any subsequent pregnancy? The two- to ten-fold increased relative risk of preterm delivery in these women is very similar in magnitude to that of women with a past history of previous preterm birth.8 In current UK obstetric practice, women with a past history of preterm birth are treated as high-risk patients and given a package of care that can include warning of the risk of recurrence, referral for serial transvaginal cervical scanning and/or fetal fibronectin screening, and preventative treatments such as cervical cerclage, progesterone, bed rest and antenatal corticosteroid therapy. The problem with this screen-and-treat approach is that it has not been shown to improve pregnancy outcome, and is therefore probably not cost effective; further research in this area has been recommended.8 Warning women of an increased risk of preterm birth may increase their anxiety but also increase the likelihood that they seek medical help should symptoms of preterm labour occur. A short cervical length, as detected by transvaginal ultrasound scan in second-trimester, is known to be a good predictor of preterm birth,9,10 although treatment based on this finding has not been shown to be effective in improving perinatal outcomes. Although second--trimester cervical shortening was also found to predict preterm birth in women with previous excisional cervical surgery,11 the lack of a validated intervention to improve outcome makes screening unlikely to be helpful at present.12 Prediction of preterm birth could allow clinicians sufficient time to administer antenatal corticosteroids, which are effective at improving perinatal morbidity and mortality, but the high false-positive rate of present screening methods means that many mothers and babies would receive steroids, with all their potential short-term and long-term complications, without the possibility of benefit because they would deliver at term. The issue is particularly complicated by the lack of trials of preterm labour prevention that include long-term paediatric outcomes. There is an additional concern that prolonging pregnancy in women at risk of preterm deliver may simply keep the fetus in an unfavourable intrauterine environment and thereby increase paediatric morbidity. Hence, further research is required to determine the optimal management of women in pregnancy with a history of excisional cervical surgery for CIN.
The data reported in this issue are a timely reminder that treatment for CIN has obstetric consequences that should be considered in young women. Policies of ‘see and treat’ may have been attractive in organisational terms, but more conservative approaches, which include biopsy to confirm CIN before treatment, assessment of human papillomavirus (HPV) presence (absence of HPV means that recurrence is less likely and conservative treatment may be more appropriate, especially in older women), an awareness that false-positive rates are higher in very young women in whom CIN more often regresses, and the need to minimise the amount of cervical tissue that is removed, may be the best way forward if we are to minimise the damage to future pregnancy outcomes.