The widespread use of cervical cancer screening and subsequent treatment of cervical intraepithelial neoplasia (CIN) have lowered the incidence of cervical cancer significantly in developed countries over the past 50 years. However, treatment methods that remove or destroy cervical tissue may modify the structural and functional integrity of the cervix, with the potential for increased risk of preterm delivery. Most recently, the American Society for Colposcopy and Cervical Pathology 2006 consensus guidelines for the management of women with CIN takes the position that women who have undergone a loop electrosurgical excision procedure (LEEP) are at increased risk for future adverse pregnancy outcomes, including preterm delivery, low birth weight, and premature rupture of membranes.1
One of the difficulties in assessing the possible adverse reproductive effects of LEEP is determining the effects of potentially confounding factors. For example, is it possible that characteristics intrinsic to the woman undergoing LEEP might be associated with the adverse pregnancy effects rather than the procedure itself? Our purpose was to design a retrospective study that permitted us to identify women with LEEP either before or after a pregnancy and compare their pregnancy outcomes to those of our general obstetric population. This design would permit us to assess whether or not the risk of preterm birth associated with LEEP was intrinsic to the procedure or to the woman undergoing the procedure.
MATERIALS AND METHODS
Records for all women undergoing LEEP in the Parkland Health & Hospital System between January 1992 and May 2008 were identified and entered into an electronic file. The first date corresponds to the introduction of LEEP at Parkland Hospital. The records in this file then were checked against a computerized database for an obstetric delivery at Parkland. The records of women with LEEP and a delivery at Parkland were subfiled as to whether the LEEP preceded the identified pregnancy resulting in delivery at Parkland or followed the identified delivery. All women delivered of singletons during the same time period without LEEP constituted the general obstetric population. Obstetric outcomes for deliveries before or after LEEP were ascertained by using this same computerized database. At Parkland, selected obstetric and neonatal outcomes for all deliveries are entered into this database. Nurses attending each delivery complete an obstetric data sheet. Database nurse specialists subsequently assess the accuracy and completeness of these data by chart review before entering the information into the database.
All LEEPs were performed within the Parkland Health & Hospital System. The great majority of these were performed in outpatient clinic settings under local anesthesia by faculty physicians or by residents under direct faculty supervision. A few were performed in an operating room owing to individual patient risk factors or anatomic considerations that contraindicated an outpatient procedure. There were no LEEPs performed during pregnancy.
Diathermy loop electrodes were chosen according to the size and location of the cervical transformation zone. The majority of procedures were performed using a 15-mm×12-mm rounded loop. A 10-mm×10-mm square or rounded loop was used alone for small transformation zones or in addition to a larger loop for excision of additional ectocervical tissue or the distal endocervical canal. Most excisions were performed under colposcopic guidance using a blended current. Multiple excisional passes were used to preserve cervical volume if excision of a large transformation zone in a single pass would have removed an unnecessary amount of cervical stroma.
There was no modification of prenatal care or surveillance for patients with a history of LEEP compared with that received by the general obstetric population; ultrasonographic cervical-length measurements were not obtained, and LEEP patients were not targeted for prophylactic cervical cerclage.
Statistical analysis using SAS 9.1 (SAS Institute, Cary, NC) included χ2, Wilcoxon rank sum, and multiple logistic regression. Multiple comparisons were adjusted using the method of Bonferroni. P<.05 was considered significant.
Parkland Hospital is a tax-supported institution providing health care to the citizens of Dallas County, Texas. The obstetrics and gynecology service is staffed by resident physicians and faculty members of the Department of Obstetrics and Gynecology at the University of Texas Southwestern Medical School. This analysis was approved by the institutional review board of Southwestern Medical Center.
Gestational age was defined as the clinical estimate of fetal age used for management during labor and delivery. This clinical estimate is calculated based on the last menses when corroborated by clinical landmarks or ultrasonography. Ultrasound estimates were used when gestational age based on last menstrual period was not consistent with the ultrasound estimate. Preterm births were defined as those before 37 weeks and were subdivided into those 33 weeks or less and those late-preterm, ie, 34 to 36 weeks. Reasons for preterm birth included spontaneous labor with intact membranes on presentation, ruptured membranes on presentation, and indicated preterm birth when complications such as hypertensive disorders prompted preterm delivery.
Neonatal death rates were calculated based on deaths of liveborn neonates up to 28 days of life. Stillbirth rates were calculated based on total births, as was perinatal mortality.
A total of 241,701 women were delivered of singletons at Parkland Hospital during the study period. As shown in Figure 1, 3,080 women underwent LEEP; 1,353 (44%) of those also had a delivery at Parkland Hospital and constitute the study population. The remaining 1,727 women with a history of LEEP had no record of a delivery at Parkland and were not included. Among the study patients, 511 underwent LEEP before the index pregnancy and 842 had LEEP after the index pregnancy. Because there were no twin pregnancies in women with LEEP, these women were compared with 240,348 women with singleton pregnancies but without LEEP.
As shown in Table 1, women with LEEP preceding pregnancy were significantly different compared with the general obstetric population. They were older, more often African-American, and less often nulliparous. Women with LEEP after a pregnancy had age profiles similar to the those in the general obstetric population but were more often African-American and nulliparous. These demographic differences were corrected for by using logistic regression in subsequent comparisons. The mean interval from LEEP to subsequent pregnancy was 113 weeks; 11% had an interval of 90 days or less.
The rate of preterm birth in women undergoing LEEP are compared with that in the general obstetric population in Table 2. There was no evidence of an increased risk of preterm birth in women undergoing LEEP either before or after pregnancy when compared with the general obstetric population. Indeed, the only significant differences were in the opposite direction and suggested that LEEP before pregnancy was associated with a decreased rate of spontaneous preterm birth between 34 and 36 weeks of gestation. After Cox regression analysis and adjustment for age, race, and nulliparity, Kaplan-Meier survival curve analysis showed no significant difference among the three study groups with regard to preterm birth. There were no significant associations of LEEP with perinatal mortality or any of its components (Table 3).
This analysis of 1,353 women who underwent LEEP either before or after pregnancy does not support an increase in preterm births related to the LEEP itself or related to factors intrinsic to women undergoing the procedure. Specifically, there was no increased risk of preterm births in any subgroup when based on gestational age or the reason for preterm birth. This result is not at odds with other reports on the effects of LEEP on preterm births.
The four largest and most recent studies of pregnancy after LEEP used differing methodologies and varied in their conclusions. Samson et al2 demonstrate an increased risk of spontaneous preterm birth, preterm delivery after rupture of membranes, and low birth weight neonates. Sadler et al3 found no increase in preterm birth. Bruinsma et al4 performed a population-based study and show an increased risk of preterm birth in women referred for evaluation of CIN regardless of whether or not they underwent a previous LEEP. This result suggests that the risk of preterm birth was intrinsic to the women who develop CIN. Most recently, Jakobsson and colleagues5 report an increased risk for preterm birth among women who had undergone prior LEEP. Risk was increased further by both a relatively large excision and by repeat procedures. This study compared delivery outcomes of a local index population with a calculated preterm birthrate for the general population in Finland. Such conflicting results between studies have prompted three meta-analyses that consider pregnancy outcomes after LEEP.6–8 All three combine a small number of studies that differ widely in methodologies, participant numbers, control groups, and primary outcomes. Crane and colleagues6 report an increased risk of preterm birth after LEEP based on combined data from five reports. Kyrgiou and colleagues7 analyzed these same five reports plus three more. The combined meta-analytic relative risk of preterm birth based on these eight reports and related to LEEP was 1.70 (95% confidence interval 1.24–2.35). However, the relative risk crossed 1.0 in seven of the eight analyzed studies. Thus, the preponderance of individual studies in this meta-analysis suggested that LEEP was not associated with preterm birth; it is only when the studies were aggregated that preterm birth was associated with LEEP. In contrast, Arbyn and colleagues8 found no increased risk of preterm birth before 34 weeks in their meta-analysis.
Our study of the effects of LEEP on preterm births is by no means ideal and indeed suffers all the difficulties associated with retrospective analyses. Specifically, we do not have information on some potential confounders such as history of tobacco, alcohol, or drug use; socioeconomic status; or prior preterm births. Additionally, we do not have data on LEEP depth, volume of tissue excised, or the degree of neoplasia prompting the LEEP. Given that the ultimate study of LEEP effects, ie, a randomized trial, cannot be done because women cannot be randomly assigned to a sham procedure, retrospective analyses such as ours are the only alternative. We conclude from our results as well as other reports in the literature that LEEP cannot definitively be implicated as a cause of preterm birth.
1. 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.
2. 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.
3. 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.
4. Bruinsma F, Lumley J, Tan J, Quinn M. Precancerous changes in the cervix and risk of subsequent preterm birth. BJOG 2007;114:70–80.
5. Jakobsson M, Gissler M, Paavonen J, Tapper AM. Loop electrosurgical excision procedure and the risk for preterm birth. Obstet Gynecol 2009;114:504–10.
6. Crane JM. Pregnancy outcome after loop electrosurgical excision procedure: a systematic review. Obstet Gynecol 2003;102:1058–62.
7. 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.
© 2010 The American College of Obstetricians and Gynecologists
8. 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.