Effective treatment of cervical intraepithelial neoplasia (CIN) underpins the success of population-based cervical screening in reducing the incidence of and mortality from invasive cervical cancer. All screening programs need to balance the potential increase in relative risk of preterm delivery in women after treatment for CIN compared with morbidity in untreated women. Both ablative and excisional modalities have been shown to be equally effective in eradicating CIN and reducing the risk of invasive cancer.1 There are concerns regarding the potential effects of loop electrosurgical excision procedure (LEEP) on reproductive function, and this may have a detrimental effect on current clinical practice. Other concerns regarding LEEP include risk of cesarean delivery (owing to failed cervical dilatation), low birth weight (LBW), and perinatal mortality and morbidity.2–12
Results of existing studies on reproductive outcomes after treatment for CIN2–18 are conflicting. All are retrospective observational studies, and most are underpowered to provide conclusive evidence on the effect of LEEP on spontaneous preterm delivery. Many cannot adjust for confounders such as maternal age, smoking, social deprivation, and previous preterm delivery. There are inconsistencies in the definition of outcomes (preterm delivery or spontaneous preterm delivery, gestational age at least 20 weeks or 24 weeks) and the selection of appropriate comparison groups.
The aim of this study was to estimate obstetric and perinatal outcomes in women with CIN3 compared with those in women with no history of CIN by linking two large databases—the National Health Service Scotland Information and Statistics Division and the Scottish Cancer Registry. Linkage of the two databases offers an ideal opportunity to explore obstetric and perinatal outcomes after a diagnosis of CIN3. In addition, we compared outcomes after treatment by an excisional method with an untreated cohort and with those treated by ablation.
MATERIALS AND METHODS
This retrospective cohort analysis was based on data linkage between the Scottish Cancer Registry and the Scottish Morbidity Record (SMR). The National Health Service Scotland Information and Statistics Division records data on all hospital patients receiving National Health Service treatment in Scotland. All obstetric data are recorded in the SMR02 database, and details of all women with CIN3 are recorded within the Scottish Cancer Registry. Data from the Scottish Cancer Registry have been judged to be of high accuracy, and SMR02 data are subject to regular quality assurance checks and have been more than 99% complete since the late 1970s.19,20 The exposed cohort comprised all women in Scotland with CIN3 diagnosed by histology followed by a first pregnancy in the period 1980 to 2005. The unexposed cohort included the remaining women, without a record of CIN, whose first pregnancies ended between 1980 and 2005. The data were generated as an anonymized dataset by the Information and Statistics Division. A diagnosis of CIN3 was determined as follows: ICD10 code for carcinoma in situ includes CIN, grade III, with or without mention of severe dysplasia. ICD10 codes used to identify cases were: 1) D06.7 930101, carcinoma in situ, other parts of cervix; 2) D06.9 930101, carcinoma in situ, cervix, unspecified; and 3) D06.0 930101, carcinoma in situ, endocervix. Permission to access the data was granted by the Information and Statistics Division Privacy Advisory Committee. As all data were anonymized, the Grampian Research Ethics Committee did not feel that the project required ethical review under the Governance Arrangements for Research Ethics Committees in the United Kingdom.
The primary outcomes were spontaneous preterm delivery (spontaneous delivery at a gestational age between 24 weeks and less than 37 weeks) and preterm premature rupture of membranes (PROM) (spontaneous rupture of membranes before labor at a gestational age between 24 weeks and less than 37 weeks). We also planned to look at secondary outcomes of rates of cesarean delivery, LBW, and perinatal mortality and morbidity.
Data analysis was performed using SPSS 13.0 (SPSS, Inc., Chicago, IL). Inclusion criteria were women who delivered between the ages of 20 and 45 years inclusive, at a gestational age of 24–43 weeks, neonatal birth weight more than 350 g, and, in case of CIN3, women who were diagnosed at the age of 20 or older.
Variables collected were 1) demographic details: age at diagnosis, maternal age at delivery, smoking, year of delivery, and socioeconomic status (the deprivation measure that has been used predominantly in the Information and Statistics Division is the Carstairs and Morris index, which uses a deprivation score that is derived at postcode sector level21); 2) antenatal details: malpresentation, preeclampsia and eclampsia, antepartum hemorrhage, preterm PROM, and preterm delivery (defined as delivery at a gestational age between at least 24 weeks and less than 37 weeks); and 3) labor details: type of onset of labor, mode of delivery, gestational age at delivery, neonatal birth weight, and perinatal mortality and morbidity. Descriptive statistics of all variables were computed. Missing data were excluded before analyses. It was anticipated that the ratio of exposed to unexposed cases would be at least 1:3. The statistical comparisons were done by using χ2 test and independent t-test. The difference of mean with 95% confidence interval (CI) was calculated for maternal age at delivery and neonatal birth weight. Odds ratios (ORs) with 95% CIs were calculated for all outcomes. These then were adjusted by logistic regression to adjust for confounders. For this study, a decision was made a priori to adjust for maternal age at delivery, smoking, socioeconomic status, year of delivery, birth weight, malpresentation, spontaneous preterm delivery, and preterm PROM where relevant. P≤.05 was considered statistically significant.
Assuming the overall preterm delivery rate to be 6%5 in the general population (unexposed control group), to detect a clinically meaningful difference in preterm delivery rates between this group and women with CIN3 (an increase in preterm delivery in women with CIN3 by at least 2.5 times over the general population or an OR of 2.55), we estimated that we would require 334 women in each group for a two-group comparison with 90% power and 5% level of significance.
The final cohort for analysis comprised 122,329 women (3,113 cases of CIN3 and 119,216 controls) (Fig. 1). The excisional-treatment group includes LEEP (n=1,097), laser excision (n=4), and cone biopsy (n=2); the ablative treatment group includes cold coagulation (n=181), laser ablation (n=84), and diathermy coagulation (n=20).
Women with CIN3 were older and had smaller neonates in their first pregnancies as compared with those in the unexposed cohort. These differences were not detected when women with different types of treatment were compared within the exposed cohort. There was no difference in the socioeconomic status of women with CIN3 when compared with the general population or in women who had excisional treatment when compared with those who had no treatment; however, those who had excisional treatment were more likely to be in a lower socioeconomic group as compared with those who had ablative treatment. There was no difference in the smoking habits in any of these comparison groups (Tables 1, 2, and 3). The mean (±standard deviation) time interval between treatment for CIN3 and delivery was 4.5 (±3.2) years.
Women with CIN3 were significantly more likely to have preterm delivery, spontaneous preterm delivery, and preterm PROM as compared with those in the general population (12% compared with 7%, 11% compared with 6%, and 8% compared with 6%, respectively). This difference remained significant even after adjusting for maternal age at delivery, birth weight, year of delivery, deprivation level, and smoking before pregnancy (Table 1). These differences were not seen within the different treatment groups in the same cohort of women with CIN3 (Tables 2, 3).
Women with CIN3 were less likely to undergo cesarean delivery when compared with those in the general population (19.1% compared with 20.8%; P=.04) (Table 1). There were no differences in cesarean delivery rates between the types of treatment or between excisional treatment compared with no treatment (Tables 2, 3). There were no differences in the rates of LBW neonates and perinatal mortality and morbidity in any of the comparison groups (Tables 2, 3), except in those women with excisional treatment, who had significantly lower rates of perinatal mortality and morbidity as compared with those who had no treatment (0.6% compared with 4.6%, P=.005) (Table 3).
Our study from a national cohort compares obstetric outcomes in women with CIN3 with those of women in the general population, all of whom are invited to participate in a public screening program for women aged 20–60 years. The comparison groups are unique, and risk is estimated for these women by comparing the risk of preterm delivery in women with CIN3 with that in women in the general population, addressing the effect of treatment per se. Women with CIN3 had increased obstetric risks that may relate to any of the risk factors, such as age, socioeconomic class, and smoking,22 shared with women without CIN3.23 Treatment for CIN3 was not associated with similarly increased risks.
Most studies compare obstetric outcomes of women treated for CIN with those of untreated women without CIN, which prevents distinguishing between the inherent risks of preterm delivery in women with CIN and those of treatment. Three systematic reviews report the effects of LEEP on preterm delivery.15,24,25 Crane in 200324 reports an increased rate of preterm delivery from 7% to 12.6% (OR 1.81, 95% CI 1.18–2.76). Kyrgiou et al in 200615 report an increase from 7% to 11.0% (OR 1.70, 95% CI 1.24–2.35) and an increased risk of preterm PROM from 2% to 5% (OR 2.69, 95% CI 1.62–4.46). The most recent meta-analysis25 found no increase in the risk of severe preterm delivery (less than 32/34 weeks) and could not pool data from the available heterogeneous results for extreme preterm delivery (less than 28/30 weeks). Of studies5,18,26 that compared treatment with no treatment in women with CIN, one18 shows no difference in rates of preterm delivery and spontaneous preterm delivery but higher rates of preterm PROM after LEEP. An increased risk of preterm delivery after treatment5 may relate to greater tissue destruction with radical diathermy and laser ablation, a finding supported by the observed fall in the excess risk of preterm delivery over the study period26 as the use of cold knife conization waned.
Our study is susceptible to inherent errors associated with all observational studies based on large national registries—unknown confounders, lack of data on behavioral and social factors, and information bias resulting from inaccuracies in reporting, entry, and retrieval of data.27 Our study has provided us with the baseline spontaneous preterm delivery rate for women with CIN3, allowing us to perform a retrospective power calculation. Therefore, it has sufficient power to show a difference in preterm delivery between women having excisional treatment and those having ablative treatment (Table 2), with an overall spontaneous preterm delivery rate of 11% (Table 1) in women with CIN3. Assuming that ablative treatment does not affect the spontaneous preterm delivery rate, to detect a doubling of the preterm delivery rate in the excision group (an OR of 2.0), we estimated that we required 194 women in the ablative group and 388 in the excision group for a two-group comparison with 90% power and 5% level of significance. Data on the risk of preterm delivery among treated women compared with the general population often are used in counseling patients and are clinically useful. In agreement with previous reports,15,24 we have established that women with CIN3 have poorer obstetric outcomes. However, contrary to the recent message on LEEP,5,18,26 there was no difference in the same outcomes between the excision and ablation treatment groups. If the treatment modality has an additional effect, this is yet to be established. The small number of women who had no treatment is an important comparator with no difference in the spontaneous preterm delivery and preterm PROM rates.
We do not have data regarding the depth of excision or the total number of treatment procedures each woman underwent before the first pregnancy. Excision should be to a depth of 7–8 mm, but this information is not recorded. Although all colposcopists in the United Kingdom must be certificated by the national society,28 the same level of training is not followed in all countries. The depth of a LEEP specimen is deeper in the center than at the periphery and could remove more of the endocervical canal than with ablation, which may have led to the hypothesis that LEEP increases the risk of preterm delivery. Our results in the two treatment groups (Table 2) should be interpreted with caution because the CIs are wide and the odds of spontaneous preterm delivery in the excisional group may be higher or lower by 50% when compared with the ablative-treatment group. Prerequisites for ablation mean that women with type 3 transformation zones, glandular abnormalities, or previous cervical treatments must undergo excision. However, these account for a minority of treatments. Other concerns associated with treatment for CIN include an increased risk of cesarean delivery, LBW, and perinatal mortality and morbidity. We confirmed that treatment does not increase the risk of cesarean delivery.2,3,6,7,9,10,12 However, there is conflicting evidence about the risk of LBW and perinatal mortality and morbidity.2–4,6–10,12,13 Although the risk of these does not differ in women with or without CIN3 or in those treated by excision or ablation, we found an increased risk of perinatal mortality and morbidity in those women who had no treatment, indicating common risk factors between women who default to treatment and those who are at increased risk of poor obstetric outcomes.
Prospective studies do not exist and are difficult to run owing to the number of women required for recruitment and retention over the period of follow-up necessary to identify relatively uncommon obstetric outcomes. Current concerns regarding the effect of treatment of CIN need to be balanced against the success of cervical screening programs, and data from large cohorts are required urgently to offer clear guidance in counseling women when they undergo colposcopy. We are concerned that a change in colposcopy practice is being promoted without adequate supporting data. This study provides information allowing it to be put into perspective by comparing effects of excisional treatment with CIN3 overall. Women with a history of CIN3 should be counseled for poorer pregnancy outcomes and treated as having high-risk pregnancies for purposes of antenatal follow-up. They also should be counseled adequately before treatment but can be reassured regarding the treatment of CIN on the risk of preterm delivery.
1. Kyrgiou M, Tsoumpou I, Vrekoussis T, Martin-Hirsch P, Arbyn M, Prendiville W, et al. The up-to-date evidence on colposcopy practice and treatment of cervical intraepithelial neoplasia: The Cochrane colposcopy & cervical cytopathology collaborative group (C5 group) approach. Cancer Treat Rev 2006;32:516–23.
2. Acharya G, Kjeldberg I, Hansen S, Sørheim N, Jacobsen B, Maltau J. Pregnancy outcome after loop electrosurgical excision procedure for the management of cervical intraepithelial neoplasia. Arch Gynecol Obstet 2005;272:109–12.
3. Blomfield P, Buxton J, Dunn J, Luesley D. Pregnancy outcome after large loop excision of the cervical transformation zone. Am J Obstet Gynecol 1993;169:620–5.
4. Braet P, Peel J, Fenton D. A case controlled study of the outcome of pregnancy following loop diathermy excision of the transformation zone. BJOG 1994;14:79–82.
5. Bruinsma F, Lumley J, Tan J, Quinn M. Precancerous changes in the cervix and risk of subsequent preterm birth. BJOG 2007;114:70–80.
6. Cruickshank M, Flannelly G, Campbell D, Kitchener H. Fertility and pregnancy outcome following large loop excision of the cervical transformation zone. BJOG 1995;102:467–70.
7. Haffenden D, Bigrigg A, Codling B, Read M. Pregnancy following large loop excision of the transformation zone. Br J Obstet Gynaecol 1993;100:1059–60.
8. Jakobsson M, Gissler M, Sainio S, Paavonen J, Tapper A. Preterm delivery after surgical treatment for cervical intraepithelial neoplasia. Obstet Gynecol 2007;109:309–13.
9. Paraskevaidis E, Koliopoulos G, Lolis E, Papanikou E, Malamou-Mitsi V, Agnantis N. Delivery outcomes following loop electrosurgical excision procedure for microinvasive (FIGO stage IA1) cervical cancer. Gynecol Oncol 2002;86:10–3.
10. Samson S, Bentley J, Fahey T, McKay D, Gill G. The effect of loop electrosurgical excision procedure on future pregnancy outcome. Obstet Gynecol 2005;105: 325–32.
11. Sjøborg K, Vistad I, Myhr S, Svenningsen R, Herzog C, Kloster-Jensen A, et al. Pregnancy outcome after cervical cone excision: a case-control study. Acta Obstet Gynecol Scand 2007;86:423–8.
12. Tan L, Pepra E, Haloob R. The outcome of pregnancy after large loop excision of the transformation zone of the cervix. J Obstet Gynaecol 2004;24:25–7.
13. Balasubramani L, Orbell S, Hagger M, Brown V, Tidy J. Do women with high-grade cervical intraepithelial neoplasia prefer a see and treat option in colposcopy? Br J Obstet Gynaecol 2007;114:39–45.
14. Gentry D, Baggish M, Brady K, Walsh P, Hungler M. The effects of loop excision of the transformation zone on cervical length: implications for pregnancy. Am J Obstet Gynecol 2000;182:516–20.
15. Kyrgiou M, Koliopoulos G, Martin-Hirsch P, Arbyn M, Prendiville W, Paraskevaidis E. Obstetric outcomes after conservative treatment for intraepithelial or early invasive cervical lesions: systematic review and meta-analysis. Lancet 2006;367:489–98.
16. Nøhr B, Tabor A, Frederiksen K, Kjaer S. Loop electrosurgical excision of the cervix and the subsequent risk of preterm delivery. Acta Obstet Gynecol Scand 2007;86:596–603.
17. Paraskevaidis E, Bilirakis E, Koliopoulos G, Lolis ED, Kalantaridou S, Paschopoulos M, et al. Cervical regeneration after diathermy excision of cervical intraepithelial neoplasia as assessed by transvaginal sonography. Eur J Obstet Gynecol Reprod Biol 2002;102:88–91.
18. Sadler L, Saftlas A, Wang W, Exeter M, Whittaker J, McCowan L. Treatment for cervical intraepithelial neoplasia and risk of preterm delivery. JAMA 2004;291:2100–6.
19. Brewster D, Stockton D, Harvey J, Mackay M. Reliability of cancer registration data in Scotland, 1997. Eur J Cancer 2002;38:414–7.
20. Cole S.. Scottish maternity and neonatal records. In: Chalmers I, McIlwaine GM, editors. Perinatal audit and surveillance. London (UK): Royal College of Obstetricians and Gynaecologists; 1980; p.39–51.
21. Carstairs V, Morris R. Deprivation and health. BMJ 1989;289:1462.
22. Steer P, Flint C. ABC of labour care: Preterm labour and premature rupture of membranes. BMJ 1999;318:1059–62.
23. Arends M, Buckley C, Wells M. Aetiology, pathogenesis and pathology of cervical neoplasia. J Clin Pathol 1998;51:96–103.
24. Crane J. Pregnancy outcome after loop electrosurgical excision procedure: a systematic review. Obstet Gynecol 2003;102:1058–62.
25. Arbyn M, Kyrgiou M, Simoens C, Raifu A, Koliopoulos G, Martin-Hirsch P, et al. Perinatal mortality and other adverse pregnancy outcomes associated with treatment of cervical intraepithelial neoplasia: a meta-analysis. BMJ 2008;337:a1284.
26. Albrechtsen S, Rasmussen S, Thoresen S, Irgens L, Iversen O. Pregnancy outcome in women before and after cervical conisation: population based cohort study. Br Med J 2008;337:a1343.
27. Mamdani M, Sykora K, Li P, Streiner D, Austin P, Rochon P et al. Reader's guide to critical appraisal of cohort studies: 2. Assessing potential for confounding. BMJ 2005;330:960–2.
28. British Society of Colposcopy and Cervical Pathology. Available at: http://www.bsccp.org.uk
. Retrieved May 1, 2009.