Jakobsson, Maija MD; Gissler, Mika MSocSci, PhD; Sainio, Susanna MD, PhD; Paavonen, Jorma MD, PhD; Tapper, Anna-Maija MD, PhD
Preterm deliveries have been rising in Nordic countries in recent years, particularly in Norway and Denmark.1 Preterm delivery has a major impact on future health of the newborn.2 The potential causative factors of preterm delivery are multiple and difficult to elucidate. A recent meta-analysis shows that all excisional procedures to treat cervical intraepithelial neoplasia (CIN) increase the preterm delivery rate.3 As most such women are young, it is important to investigate the possible adverse effects of these procedures on subsequent pregnancy outcome.
Randomized trials are difficult to organize, which makes large register-based studies necessary. We had an opportunity to utilize high-quality, population-based register data from Finland. The aim of our study was to assess the effect of different surgical treatment modalities used for treatment of CIN on the outcome of subsequent pregnancies, especially preterm deliveries.
MATERIALS AND METHODS
We used Hospital Discharge Register for 1986–2003 to identify reproductive-aged (15–49 years) women who underwent treatment for CIN. These cases were linked to the Finnish Medical Birth Register for 1987–2004 to identify all subsequent births. Both registers are run by the National Research and Development Centre for Welfare and Health (STAKES), and they cover the total population, including hospital care and deliveries in the private sector. The data linkage was performed by using women's unique identification number existing in both registers.
The Hospital Discharge Register collects information on all inpatient episodes in health care facilities (since 1967). According to the latest data quality study from the late 1980s, in total 95% hospitalizations were registered, and 97% of main diagnoses concerning pregnancy, birth, and puerperium were correctly reported at the three-digit International Classification of Diseases code level.4 Increased use of electronic patient journals in hospitals has further improved its completeness and validity.4 Since 1994, the Hospital Discharge Register includes all day surgical procedures, and since 1998, all hospital outpatient visits as well. The medical procedures performed in the hospitals were identified by using surgical procedure codes (ie, in 1986–1996 based on the national classification of Finnish Hospital League and since 1996 on the Finnish version of Nordic Classification on Surgical Procedures).
The Medical Birth Register collects baseline data information on mothers under a doctor's or midwife's supervision and interventions during pregnancy and delivery and on the newborn's outcome during the first 7 days. The Medical Birth Register data are complied at the time of birth, using the mother's prenatal charts as one of the information sources. Data are collected from all delivery units in Finland. Less than 0.1% of all newborns are missing from the Medical Birth Register. Information on such cases is routinely obtained from the Central Population Register and the Cause-of-Death Register kept by Statistics Finland. The register is complete after these data linkages. Data are checked at the National Research and Development Centre for Welfare and Health, and seemingly incorrect information is sent back for revision. For most variables, the data corresponds well with information in hospital records.5
The aim of this study was to examine the outcome of subsequent birth, especially preterm deliveries, after cervical ablative or excisional treatments for precancerous lesions. The main outcomes of the study were preterm delivery (duration of pregnancy less than 37 weeks), subclassified further into moderately preterm delivery (32–37 weeks of gestation), very preterm delivery (28–31 weeks of gestation), and extremely preterm delivery (less than 28 weeks of gestation). Low birth weight rate and perinatal mortality rate were also analyzed. Newborns weighing less than 2,500 g were defined as low birth weight babies. The perinatal death rate included stillbirths from 22 weeks of gestation and early neonatal deaths (during the first 7 days after delivery).
The study material included 25,827 reproductive-aged women on whom cervical treatment was performed during 1986–2003. A total of 5,835 women had subsequent deliveries recorded in the Medical Birth Register. The number of newborns was 8,405, of which 8,210 were singletons. Only the results for the singleton deliveries are reported here. Our control group consisted of all other women without history of treatment for CIN who delivered during 1987–2004 (n=1,056,855 singletons).
Women with history of treatment for CIN (n=25,827) were divided into three categories. Our first category (n=16,145 women) included excisional therapies such as conization of cervix using knife (cold-knife conization), large loop excision of the transformation zone (LEEP), and laser conization. In these treatments, tissue is excised from the cervix. According to the Finnish classification (1983–1996) these procedures had a common treatment number (8326). The current classification based on the Nordic Classification of Surgical Procedures, has different codes for cold-knife conization (LDC00) and LEEP (LDC03). Since the introduction of the LEEP procedure, cold-knife conizations and laser conizations have become extremely rare in Finland. In the second category (n=9,028 women), we included ablative treatments, such as cryotherapy, electrocoagulation, and laser vaporization (8325, LDB10, LDB20). In these procedures, the surface of the lesion is destroyed without excision. The third category included other excisional treatments of the cervix, such as cervical amputation (8327/LDC10) and other cervical excisions (8339/LDC96). Only one code was problematic (ie, excision of lesion of cervix uteri [8339/LDB00] which included women with different kinds of cervical problems). In this category, the main diagnoses were reviewed carefully, and cases with irrelevant cervical treatments (such as abortion, excisions of polyps) were excluded. In total, 654 women were finally included in the third group. Thus, the final study groups were as follows:
* Conization group (16,145 women and 4,545 newborns),
* Ablation group (9,028 women and 3,425 newborns),
* Other group (654 woman and 240 newborns).
The statistical comparisons were done by using the χ2 test, the test for relative proportions (testing the differences between proportions among case and control groups6) and the Fisher exact test. We calculated risk ratios (RRs) with 95% confidence intervals (CIs) for all outcomes. We further calculated odds ratios (ORs) with 95% CIs by logistic regression to adjust for confounders. We used maternal age at birth, previous deliveries, and maternal smoking during pregnancy (as a proxy of socioeconomic status) as confounding variables in the logistic regression analyses. We also studied if a short time interval between the procedure and the delivery (during the same year or the year before) affects the perinatal outcomes. All the analyses were repeated by including only the first deliveries of cases after treatment for CIN and the first deliveries of the control women.
This study was approved by the University of Helsinki Institutional Review Board. The data linkage between Medical Birth Register and Hospital Discharge Register was performed after the register keeping organization the National Research and Development Centre for Welfare and Health had given the authorization required by the legislation.
The characteristics of women with cervical treatment (cases) or without cervical treatment (controls) are presented in Table 1. The case women with CIN treatment were slightly older. The cases had fewer previous deliveries than the controls and were more often nulliparous than the controls (42.8% versus 30.9%). Case women with cervical treatment were almost twice as often smokers (26.6%) than control women (15.2%). Information on the socioeconomic status was available from October 1990 onward. Case women had lower socioeconomic status than controls (Table 1).
Case women had 724 (8.8%) preterm deliveries and 472 (5.7%) low birth weight babies (Table 2). The corresponding rates among controls were 49,257 (4.6%) and 32,976 (3.1%). Perinatal mortality rate was 7.7 among cases and 6.7 among controls.
The risk of preterm delivery was increased after any cervical treatment (ie, after conization [RR 1.99, 95% CI 1.81–2.20], ablation [RR 1.60, 95% CI 1.41–1.82], and other treatments [RR 1.97, 95% CI 1.29–3.02]).When adjusted for age, parity, and smoking, the corresponding ORs were 1.78 (95% CI 1.60–1.97), 1.47 (95% CI 1.29–1.67), and 1.82 (95% CI 1.17–2.84). No differences between different treatment modalities were observed.
The risk increase after conization was highest for extremely preterm delivery and very preterm delivery (RR 2.10, 95% CI 1.47–2.99 and RR 2.86, 95% CI 2.22–3.70, respectively). For moderately preterm deliveries, the risk was also increased, but less. In the ablative group, the risk of preterm delivery increased progressively (Fig. 1). Adjusting for maternal age, parity, and maternal smoking did not affect these results.
In the subgroup analysis from 1997 onward, we could discriminate different treatment modalities within treatment categories. The preterm delivery rate was increased after all treatment modalities; after loop conization 7.8% (2,690 deliveries), after cold-knife conization 8.6% (93 deliveries), and after amputation of the cervix 30.8% (39 deliveries).
Mainly due to increased risk of prematurity, the risk of low birth weight was also increased after conization (RR 2.06, 95% CI 1.83 –2.31), ablation (RR 1.39, 95% CI 1.18–1.63), and other treatments (RR 2.16, 95% CI 1.32–3.54). Perinatal mortality rate was increased only in the conization group (RR 1.74, 95% CI 1.30–2.32). Adjusting for the confounding factors did not change these results.
The analyses done by including only the first deliveries of women after CIN treatment did not change our results. Time interval between treatment and delivery had no effect on the preterm delivery rate (data not shown).
We found a significantly increased risk of preterm delivery after all surgical treatments for CIN. After conization, the risk was highest for very preterm delivery and extremely preterm delivery.
Finland has an excellent health information system mainly based on register data. In this study, we used national, population-based register data with excellent data quality and coverage. The absence of reporting bias, recall bias, or participation bias further increases the credibility of our results.
Our material included procedures that were performed in hospitals and outpatient clinics during an 18-year period. This database is substantially larger compared with previously published studies. We had information on all day surgical patients from 1994 onward and on all outpatient patients since 1998. Previously, most treatments were performed as inpatient, but more recently most treatments are performed in outpatient settings. We admit, that some procedures might have been performed in settings not captured by the database. However, in Finland only few procedures are performed in private clinics outside hospital settings which would not have been captured by the database. It is unlikely that the missing cases had affected our results.
We were able to adjust for maternal age, parity, and smoking as the most important confounding factors. Unfortunately, we were unable to define the socioeconomic status for all women. The group “other” is heterogeneous including students and housewives, and socioeconomic status was not included in the Medical Birth Register before 1990. Therefore, we used maternal smoking as a surrogate variable. In Finland, socioeconomic status and smoking are strongly correlated.7 Our database did not include other risk factors that may be common to CIN and preterm delivery.
Preterm delivery is a serious obstetric complication, not only because it causes major neonatal mortality but also because it has an impact on future health in general. Recent data show that the rates of preterm delivery have been rising in Nordic countries.1 If this trend is real and also observed in other countries, societal and health impacts will be considerable.
Preterm delivery is associated with many causative factors which are poorly understood.2 Some investigators have found a significant correlation between cervical conization and preterm delivery,8–11 whereas others have not been able to confirm this.12–18 These results are based on small retrospective studies. In early studies, the cold-knife conization technique seemed to increase the risk of preterm delivery.8–10 In one study from Norway, the risk of preterm delivery was increased up to 6-fold after laser conization.11 A larger retrospective cohort study from New Zealand showed no significant increase in risk of total preterm delivery for any cervical treatment. The risk of premature rupture of membranes, however, was significantly increased after laser conization and LEEP, but not after laser ablation.19 In a retrospective cohort study from Canada, LEEP was associated with a 3.5-fold increase in the risk of overall preterm delivery.20
Recent systematic review and meta-analysis showed that all excisional procedures, but not ablative procedures, were associated with pregnancy-related morbidity.3 The former finding was confirmed also in our study. However, we found an increased risk of preterm delivery after all cervical treatments, including ablation. Strikingly, perinatal mortality was also increased after conization.
The risk of early preterm delivery was highest after excisional treatments, which may indicate that cone size is of importance. In fact, some studies,3,10,19,21–22 but not all,11 have reported an increased risk of preterm delivery by cone size. However, the cone size is often difficult to measure, and it is almost impossible to evaluate retrospectively. The present study does not allow further analysis of the size of tissue removed, as this information is not collected in the registers.
Human papillomavirus (HPV) infections causing CIN have become highly prevalent in young women.23 Cervical intraepithelial neoplasia is a relatively early manifestation of HPV infection.23 The loop electrosurgical excision procedure is generally considered an easy and safe method for treating these lesions, and even so called “see and treat” management has become increasingly popular.12 Thus, an increasing number of women of reproductive age face such procedures, altering cervical anatomy and subsequently affecting pregnancy outcomes. Increasing evidence suggests that HPV-associated cervical lesions have a high rate of spontaneous regression, especially among young women.23–24 Preterm delivery has serious consequences on the future health of women and their newborns. It is important to avoid unnecessary cervical trauma when treating young women for CIN. To do no harm should be kept in mind particularly in the management with cervical low-grade lesions. In conclusion, we found that any treatment for CIN increases the risk of adverse pregnancy outcome.
1. Langhoff-Roos J, Kesmodel U, Jacobsson B, Rasmussen S, Vogel I. Spontaneous preterm delivery in primiparous women at low risk in Denmark: population based study. BMJ 2006;22:937–9.
2. Shennan AH, Bewley S. Why should preterm deliveries be rising? BMJ 2006;332:924–5.
3. 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.
4. Gissler M, Haukka J. Finnish health and social welfare registers in epidemiological research. Norsk Epidemiologi 2004;14:1–8.
5. Gissler M, Shelley J. Quality of data on subsequent events in a routine Medical Birth Register. Med Inform Internet Med 2002;27:33–8.
6. Odeh RE, Owen DB, Birnbaum ZW, Fisher L. Pocket book of statistical tables. New York: Marcel Dekker Inc; 1977. p. 2–10.
7. Jaakkola N, Jaakkola MS, Gissler M, Jaakkola JJ. Smoking during pregnancy in Finland: determinants and trends, 1987–1997. Am J Public Health 2001;91:284–6.
8. Lee NH. The effect of cone biopsy on subsequent pregnancy outcome. Gynecol Oncol 1978;6:1–6.
9. Jones CJ, Brinton LA, Hamman RF, Stolley PD, Lehman HF, Levine RS, et al. Risk factors for in situ cervical cancer: results from a case-control study. Cancer Res 1990;50:3657–62.
10. Leiman G, Harrison NA, Rubin A. Pregnancy following conization of the cervix: complications related to cone size. Am J Obstet Gynecol 1980;136:14–8.
11. Hagen B, Skjeldestad FE. The outcome of pregnancy after CO2 laser conisation of the cervix. Br J Obstet Gynaecol 1993;100:717–20.
12. Gunasekera PC, Phipps JH, Lewis BV. Large loop excision of the transformation zone (LEEP) compared to carbon dioxide laser in the treatment of CIN: a superior mode of treatment. Br J Obstet Gynaecol 1990;97:995–8.
13. Sagot P, Caroit Y, Winer N, Lopes P, Boog G. Obstetrical prognosis for carbon dioxide laser conization of the uterine cervix. Eur J Obstet Gynecol Reprod Biol 1995;58:53–8.
14. Althuisius SM, Schornagel IJ, Dekker GA van Geijn HP, Hummel P. Loop electrosurgical excision procedure of the cervix and time of delivery in subsequent pregnancy. Int J Gynaecol Obstet 2001;72:31–4.
15. Blomfield PI, Buxton J, Dunn J, Luesley DM. Pregnancy outcome after large loop excision of the cervical transformation zone. Am J Obstet Gynecol 1993;169:620–5.
16. Ferenczy A, Choukroun D, Falcone T, Franco E. The effect of cervical loop electrosurgical excision on subsequent pregnancy outcome: North American experience. Am J Obstet Gynecol 1995;172:1246–50.
17. Tan L, Pepra E, Haloob RK. The outcome of pregnancy after large loop excision of the transformation zone of the cervix. J Obstet Gynaecol 2004;24:25–7.
18. Mathevet P, Chemali E, Roy M, Dargent D. Long-term outcome of a randomized study comparing three techniques of conization: cold knife, laser, and LEEP. Eur J Obstet Gynecol Reprod Biol 2003;106:214–8.
19. 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.
20. Samson SL, Bentley JR, Fahey TJ, McKay DJ, Gill GH. The effect of loop electrosurgical excision procedure on future pregnancy outcome. Obstet Gynecol 2005;105:325–32.
21. Raio L, Ghezzi F, Di Naro E, Gomez R, Luscher KP. Duration of pregnancy after carbon dioxide laser conization of the cervix: influence of cone height. Obstet Gynecol 1997;90:978–82.
22. Myllynen L, Karjalainen O. Pregnancy outcome after combined amputation and conization of the uterine cervix. Ann Chir Gynaecol 1984;73:345–9.
23. Auvinen E, Niemi M, Malm C, Zilliacus R, Trontti A, Fingerroos R, et al. High prevalence of HPV among female students in Finland. Scand J Infect Dis 2005;37:873–6.
24. Moscicki AB, Shiboski S, Hills NK, Powell KJ, Jay N, Hanson EN, et al. Regression of low-grade squamous intra-epithelial lesions in young women. Lancet 2004;364:1678–83.
© 2007 by The American College of Obstetricians and Gynecologists. Published by Wolters Kluwer Health, Inc. All rights reserved.