All excisional treatments for cervical intraepithelial neoplasia (CIN), including loop electrosurgical excision procedure (LEEP) conization, predispose to low birth weight and preterm birth.1–4 Whether this is truly causal, not just an association, remains unclear.5 Women treated for CIN have many risk factors for preterm birth, such as smoking and low socioeconomic position. Risk-taking behavior may further predispose to sexually transmitted and other infections. In our previous study, long-term mortality was increased among these women, reflecting risk-taking behavior.6
Using healthy women as controls may overestimate the risk for preterm birth after treatment for CIN. Women giving birth are generally healthy, because health problems predispose to subfertlity.7 Some studies have used patients from colposcopy clinics as controls, and in those studies, preterm birth rates have been high also among controls. In such studies, the risk for preterm birth8 and premature rupture of the membranes (PROM)9 was, however, increased significantly after treatment for CIN, but this risk seemed to be lower than in other studies. This suggests that these women may have high risk for preterm birth even without cervical surgical treatments. In some studies, the risk for preterm birth was related to the size of the cone.2,9–11 Retrospective evaluation of the cone size is, however, difficult.
In this study, we wanted to find out if LEEP conization predisposes to preterm birth. The ability to use internal controls allowed us to control for confounding factors. We further evaluated the effect of cone size and repeat treatment.
MATERIALS AND METHODS
Our material constituted of 672 women having an outpatient treatment for CIN during 1997–2003 and a subsequent delivery until 2006 at the Department of Obstetrics and Gynaecology, Helsinki University Hospital, and at the Maternity Hospital, Helsinki, Finland. We identified from the Hospital Discharge Register women who had undergone LEEP conization. We identified subsequent deliveries from the Medical Birth Register. We collected additional data from medical records and maternity cards, because the available number of variables in administrative registers is limited. We used these data sources to obtain the most accurate information on health behavior, pregnancy history, and CIN, including procedures, diagnoses, and excised cone sizes. We used preterm birth rates for singleton pregnancies from 1997 to 2006, as registered in the Medical Birth Register, to calculate expected rates for our study groups. The information on the Medical Birth Register and Hospital Discharge Register has good validity, making them suitable for research use.12,13 Details of these registers are described elsewhere.4
We excluded three women in whom treatment was performed during pregnancy, 30 women with a delivery during the study year but before the treatment, one woman because of earlier cold-knife conization, and 14 women owing to multiple pregnancy. Two women had delivered abroad, and data on previous deliveries was unavailable. We studied the following groups: all women with previous LEEP conization (n=624, of whom 75 had preterm birth), one subgroup consisting of primiparous women (n=364 women, of whom 44 had preterm birth), and another subgroup consisting of women with delivery before and after the treatment (n=258 women, of whom 17 had preterm birth before the procedure and 31 after the procedure). The last group was chosen to control for the confounding factors. Finally, we combined these two subgroups (including those two without information on the deliveries before treatment) and had a total of 624 women who had delivered after the treatment.
Data on smoking, alcohol consumption, and drug and substance abuse was obtained from the maternity cards, and was used as a proxy for the average use during pregnancy. The main outcome measure was preterm birth rate in the different subgroups. Deliveries before the 37th gestational week were defined as preterm birth. The duration of pregnancy was usually based on ultrasound determination, and the best clinical estimation was recorded both in medical records and in the Medical Birth Register. We further classified deliveries to moderately preterm (from 32 to 36 gestational weeks), very preterm (from 28 to 31 gestational weeks), and extremely preterm (less than 28 gestational weeks). We divided the cause for preterm birth into spontaneous preterm birth, preterm rupture of membranes, and induced preterm birth.
We further evaluated the effect of excised cone size on preterm birth risk. Our estimation on the cone size was based on the information of LEEP wire color (ie, size) available from the medical records. Other information of the cone size was used when reported by pathologists. These methods were combined for the best available estimation of the cone size. In general, these estimates were in agreement. Information on the cone size was available for 340 women (54.5% of all women). We considered cone size less than 10×10 mm as small, 15×12 mm as middle sized, and 20×12 mm or more as large. We also analyzed repeat LEEP conizations separately (n=59).
We used general preterm birth rates in Finland for singletons in 5-year age and parity groups from time period 1997–2006 to calculate the expected numbers of preterm births (4.6%, n=25,780 preterm births), and to derive a standardized incidence ratio. This was also calculated adjusting for previous preterm birth and parity. For women having a birth before and after the procedure, we used the delivery before LEEP conization as a reference. We calculated risk ratios (RRs) and 95% confidence intervals (CIs) for preterm birth; we further adjusted them for age and parity. The statistical comparisons were performed using the test for relative proportions (testing the differences between shares among case and control groups).14 The differences for continuous variables between groups were tested with two-tailed t test, and the differences in preterm birth rates were tested with the test for relative proportions. We also calculated number needed to treat for harm for the key variables.
The University of Helsinki Institutional Review Board approved the study. Furthermore, the register-keeping organization STAKES (National Research and Development Centre for Welfare and Health), currently THL (National Institute for Health and Welfare) gave permission to use health register data.
Women who had at least one delivery after the LEEP conization (n=624 women) were relatively young at the time of the treatment, 27.6±5.0 years and 30.1±5.0 years at the subsequent delivery. The age distribution is presented in Table 1. The majority of women had their first delivery after the treatment (n=364, 58.3%). Almost one half of the women (n=280, 44.9%) were lower-class white collar workers, but many were (n=150, 24.0%) upper-class white collar workers, as well. More than 6% were students and housewives; these rates are lower than the average rate in the Medical Birth Register. Smoking, alcohol consumption, and drug abuse during pregnancy among these women is shown in Table 2. More than one fourth smoked during pregnancy (n=152, 24.3%), and the majority (n=98, 64.5%) smoked more than five cigarettes per day. Less than 10% (n=45, 7.2%) reported alcohol consumption. Drug abuse was reported by 11 women (1.8%). Data on substance use were, however, missing in 20% of the cases.
We were able to find definite CIN diagnoses (excluding all diagnostic procedures) for almost all women (92.3%). Figure 1 presents the diagnoses in different time periods. The number of treated low-grade lesions showed decreasing tendency, as in accordance with the current guidelines.15,16 Most of the conizations were performed because of CIN 2 (n=199, 31.9%) or CIN 1 (n=191, 30.6%) (Table 3). Only 18 (2.9%) women had the treatment because of carcinoma in situ or invasive cervical cancer. A total of 59 (9.4%) women had repeat treatment. Short-term complications were rare; only 16 (2.6%) women had heavy bleeding requiring another appointment, and four (0.6%) developed a postoperative infection.
Altogether, 75 preterm births (12.0%) were recorded in the study group. The majority (n=58, 77.3%) of them were moderately preterm, 11 (14.7%) very preterm and only few (n=6, 8.0%) were extremely preterm. The mean±standard deviation (SD) gestational age for preterm birth was 33.3 weeks (±3.3). The main cause for preterm birth was preterm premature rupture of membranes (n=34, 45.3%). There were 19 (25.3%) spontaneous preterm deliveries and 22 (29.3%) induced preterm births. Compared with the Medical Birth Register data (of which 4.61% were preterm), the risk for preterm birth was 2.61-fold (CI 2.02–3.20; number needed to treat for harm, 14) after LEEP conization. Adjusting for age, parity, or both did not change the results (data not shown).
When considering the cone size, the risk for preterm birth was 2.45-fold (CI 1.38–3.53) when comparing large or repeat cones (n=103) with small or medium-sized ones (n=238) (Table 3).
In our study population, 59 repeat LEEP conizations were performed. These women had 14 (23.7%) preterm births. More than a fivefold increased risk for preterm birth was observed (RR 5.15, CI 2.45–7.84, number needed to treat for harm 5) when compared with the general preterm birth rates (4.61%).
In a subgroup analysis, we had 258 women who had a delivery both before and after the treatment. Table 1 presents the background characteristics of these women. The average±SD age at delivery before treatment was 25.7±5.1 years, and 30 of them (11.6%) were aged younger than 20 years. Most women had only one delivery before the treatment. Average age at treatment was 29.7±4.8 years and 31.7±5.0 years at the first delivery after the treatment. These women had a relatively low socioeconomic status; most of them (n=114, 44.2%) were lower-class white collar workers. Many were either students or housewives. These women smoked often (n=88, 34.1%) (Table 2), and almost one fourth smoked more than five cigarettes per day. A total of 23 women (8.9%) reported alcohol use during pregnancy. Drug or substance use was rare (1.6%).
These 258 women had 17 (6.5%) preterm births before and 31 (12.0%) after LEEP conization. The mean (±SD) gestational age was 33.0 (±4.1) before the treatment and 32.7 (±3.9) after the treatment. The risk for preterm birth was increased after LEEP conization when compared with deliveries before treatment (RR 1.94, CI 1.10–3.40, number needed to treat for harm 18) (Table 4). The risk was increased when compared with the background rates based on Medical Birth Register (RR 2.61, 1.69–3.52, number needed to treat for harm 13). Adjusting by age or parity or both did not change our results.
Thus, five of the 17 women (29.4%) having a preterm delivery before LEEP conization had recurrent preterm birth after the procedure. Of the 241 women who had term birth before the procedure, 26 had preterm birth after LEEP conization (10.8%).
We analyzed our data by parity and history of previous preterm birth and compared these rates to the general preterm rates in the Medical Birth Register. These data show that conization is a strong risk factor for preterm birth. The preterm birth rate among primiparous women was 6.0% in our study, which did not differ from that among primiparous women in the Medical Birth Register (5.6%, P=.82). After LEEP conization, the preterm birth rate was increased threefold (RR 3.38, CI 2.31–4.94) among all women with no previous preterm birth. In the second birth, the preterm birth rate was 10.7% among women with a previous term birth (RR=3.66, CI 2.27–5.91, P<.001) and 22.2% among women with a previous preterm birth (P=.63). These rates were higher than among the total population, 2.9% and 16.5%, respectively. Similar risk was observed for third and fourth births, even though the number of cases were small (data not shown).
Another subgroup consisted of women who had their first delivery after the treatment (ie, primiparous, n=364). As expected, they were older than the other subgroup, the average age being 26.1±4.5 years at the treatment and 28.8±4.5 years at the delivery (Table 1). These women were more educated; one third were upper-class white collar workers. Smoking was less common (n=64, 17.5%). Also, alcohol consumption was less common (n=22, 6.0%). Drug abuse was at approximately the same low level as in the other subgroup (n=7, 1.9%). There were 44 preterm births (12.0%). The risk for preterm birth was 2.63-fold (CI 1.85–3.41) when compared with the background rate (Medical Birth Register). Adjusting for age and parity did not change the risk.
We found that the risk for preterm birth was increased almost threefold after LEEP conization. When analyzing women with a delivery before and after LEEP conization and adjusting for age and parity, the risk for preterm birth was still almost twofold after LEEP conization. These women had a somewhat higher preterm birth rate (6.5%) before the treatment when compared with the background rate based on Medical Birth Register (4.6%, P=.075). Increased risk was also found among women without previous preterm birth. This suggests that the procedure as such is a stronger risk factor than maternal background characteristics. Furthermore, we found a tendency for increased preterm birth rates related to the cone size; repeat treatments were especially harmful.
Our data included deliveries from two different clinics. The Department of Obstetrics and Gynaecology of the Helsinki University Hospital is a tertiary referral university hospital, where preterm deliveries are referred; the Maternity Hospital is a local hospital with central hospital facilities. These two hospitals cover practically all deliveries in Helsinki. We assume that the inclusion of deliveries from both hospitals diminished selection bias, and our data can be considered to represent the population-based data of an urban population with an easy access to treatment. We had an opportunity to use Finnish high-quality register data. We collected this information from the medical records to find data on maternal background characteristics and on the procedure, including the cone size.
Women smoked and used alcohol during pregnancy more often than women in the general population. These self-reported rates, however, may be underestimates. Smoking was common among all women treated for CIN, which is in accordance with other studies.9,10 Primiparous women smoked and used less alcohol than the women having more than one pregnancy during the study period. The percentage of women reporting drug abuse was low. Drug abuse may have been underestimated in clinical practice, even though drug abuse is relatively rare in Finland.17
The weakness of our study was that the cone size was estimated and may have been inaccurate, and this information was not available in approximately 45% of all women. Many,2,9–11 but not all studies18 have reported that excised cone size is related to preterm birth. This is accordance with our study. Repeat treatments were especially harmful. Information on the diagnoses was available on virtually all (92.3%) women. Those women with CIN 3 or carcinoma in situ or carcinoma had clearly higher preterm birth rates (15.1% and 22.2%, respectively), which is clinically plausible. However, the differences remained statistically insignificant because of the small number of women. Sjoborg et al11 also reported a tendency for deeper cones with increasing CIN grade and shorter pregnancy duration. Samson et al18 did not find increased risk for preterm birth after repeat LEEP conization, but the risk was increased after LEEP and another procedure.
Some authors have suggested that the increased risk for preterm delivery is associated with background characteristics, such as low socioeconomic status of these women, rather than treatment for CIN.8,18,19 We determined the socioeconomic status of the women from two data sources and found that their position was higher than average. The primiparous women had highest socioeconomic position: 70% of them were either upper- or lower-class white collar workers. Our results suggest that surgical treatment is more important than any of the studied epidemiologic or demographic characteristics of the women. In addition, the increased risk for preterm birth after LEEP conization was highest among women without previous preterm birth. In many other studies previous preterm birth has been the strongest risk factor for preterm birth.8,10,20,21 Preterm birth rates after LEEP conization were the same among primiparous (12.1%) and other women (12.0%), which does not support the theory that primiparity is a strong risk factor for preterm birth.21
In the current study we found high proportion (45%) of preterm PROM after LEEP conization, which is in accordance with Sadler et al.9 It has been estimated that in the Nordic countries approximately 30% of all preterm births results from preterm PROM (Morken NH. Public health aspects of preterm birth. Studies using Scandinavian population-based data. Academic dissertation, University of Gothenburg 2008. p. 15). Why these treatments predispose to preterm PROM remains unknown. In some,22,23 but not in all24,25 studies, cervical shortening after LEEP conization has been reported. This may predispose to ascending infection. Decreased mucus production may lead to a defect in a local immune defense system, which may predispose to infection and finally lead to preterm PROM.
The cervix has been traditionally considered to regenerate rapidly. Treatments for CIN have been considered to be safe and easy, and not impairing pregnancy outcomes. However, LEEP conization clearly predisposes to preterm birth. We found almost a twofold risk for preterm birth after LEEP conization by using internal controls. Our results suggest that surgery itself and not the background characteristics explain the increased risk. Repeat treatments are extremely harmful. The risk increased especially among women without previous preterm birth. Human papillomavirus infections are increasing, and women deliver at older ages, which translates into increasing numbers of parturients with a history of LEEP conization. Therefore, unnecessary or “see and treat” procedures should be avoided.
1. Arbyn M, Kyrgiou M, Simoens C, Raifu AO, Koliopoulos G, Martin-Hirsch P, et al. Perinatal mortality and other severe adverse pregnancy outcomes associated with treatment of cervical intraepithelial neoplasia: meta-analysis. BMJ 2008;337:a1284.
2. 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.
3. Albrechtsen S, Rasmussen S, Thoresen S, Irgens LM, Iversen OE. Pregnancy outcome in women before and after cervical conisation: population based cohort study. BMJ 2008;337:a1343.
4. Jakobsson M, Gissler M, Sainio S, Paavonen J, Tapper AM. Preterm delivery after surgical treatment for cervical intraepithelial neoplasia [published erratum appears in Obstet Gynecol 2008;112:945]. Obstet Gynecol 2007;109:309–13.
5. Jakobsson M, Bruinsma F. Adverse pregnancy outcomes after treatment for cervical intraepithelial neoplasia. BMJ 2008;337:a1350.
6. Jakobsson M, Gissler M, Paavonen J, Tapper AM. Long-term mortality in women treated for cervical intraepithelial neoplasia. BJOG 2009;116:838–44.
7. Gissler M, Berg C, Bouvier-Colle MH, Buekens P. Pregnancy-associated mortality after birth, spontaneous abortion, or induced abortion in Finland, 1987–2000. Am J Obstet Gynecol 2004;190:422–7.
8. Bruinsma F, Lumley J, Tan J, Quinn M. Precancerous changes in the cervix and risk of subsequent preterm birth. BJOG 2007;114:70–80.
9. 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.
10. Nohr B, Tabor A, Frederiksen K, Kjaer SK. Loop electrosurgical excision of the cervix and the subsequent risk of preterm delivery. Acta Obstet Gynecol Scand 2007;86:596–603.
11. Sjoborg KD, Vistad I, Myhr SS, 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. Gissler M, Shelley J. Quality of data on subsequent events in a routine Medical Birth Register. Med Inform Internet Med 2002;27:33–8.
13. Keskimäki I, Aro S. Accuracy of data on diagnoses, procedures, and accidents in the Finnish Hospital Discharge Register. Int J Health Sci 1991;2:15–21.
14. Odeh RE, Owen DB, Birnbaum ZW, Fisher L. Pocket book of statistical tables. New York (NY): Marcel Dekker Inc.; 1977. p. 2–10.
15. Jordan J, Arbyn M, Martin-Hirsch P, Schenck U, Baldauf JJ, Da Silva D, et al. European guidelines for quality assurance in cervical cancer screening: recommendations for clinical management of abnormal cervical cytology, part 1. Cytopathology 2008;19:342–54.
16. Wright TC Jr, Massad LS, Dunton CJ, Spitzer M, Wilkinson EJ, Solomon D, et al. 2006 consensus guidelines for the management of women with cervical intraepithelial neoplasia or adenocarcinoma in situ. Am J Obstet Gynecol 2007;197:340–5.
17. Rönkä S, Virtanen A, Perälä R, Vihmo J. Finland—Drug Situation 2007. National Report to the EMCDDA. Available at: http://www.stakes.fi/tilastot/tilastotiedotteet/reitox/DRUG20SITUATION%202007%20FINLAND.pdf
. Retrieved June 5, 2009.
18. 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.
19. Cruickshank ME, Flannelly G, Campbell DM, Kitchener HC. Fertility and pregnancy outcome following large loop excision of the cervical transformation zone. Br J Obstet Gynaecol 1995;102:467–70.
20. Goldenberg RL, Culhane JF, Iams JD, Romero R. Epidemiology and causes of preterm birth. Lancet 2008;371:75–84.
21. Ananth CV, Peltier MR, Getahun D, Kirby RS, Vintzileos AM. Primiparity: an “intermediate” risk group for spontaneous and medically indicated preterm birth. J Matern Fetal Neonatal Med 2007;20:605–11.
22. Ricciotti HA, Burke L, Kobelin M, Slomovic B, Ludmir J. Ultrasound evaluation of cervical shortening after loop excision of the transformation zone (LETZ). Int J Gynaecol Obstet 1995;50:175–8.
23. Mazouni C, Bretelle F, Blanc K, Heckenroth H, Haddad O, Agostini A, et al. Transvaginal sonographic evaluation of cervix length after cervical conization. J Ultrasound Med 2005;24:1483–6.
24. 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.
25. Gentry DJ, Baggish MS, Brady K, Walsh PM, Hungler MS. The effects of loop excision of the transformation zone on cervical length: implications for pregnancy. Am J Obstet Gynecol 2000;182:516–20.