Cervical intraepithelial neoplasia (CIN) is a common complication in women of reproductive age.1 Treatment options have evolved and include cold knife conization, laser ablation, cryotherapy, and loop electrosurgical excision procedure (LEEP).1 Cold knife conization of the cervix for treatment of CIN has been associated with adverse outcomes in subsequent pregnancy, including preterm delivery, low birth weight infants, incompetent cervix, and cervical stenosis.2–8 Several small studies that have evaluated LEEP and subsequent pregnancy outcomes have had mixed results.9–13 A systematic review found LEEP may be associated with subsequent preterm birth; however, there were many shortcomings of the individual studies, including lack of control for potential confounders.14 A larger study found LEEP and laser cone treatments associated with significantly increased rates of premature rupture of membranes before 37 weeks of gestation leading to preterm birth, but no increase in spontaneous preterm labor with intact membranes leading to preterm delivery.15 A recent study found LEEP to be associated with preterm birth and low birth weight.16 There is limited information on the effect of cryotherapy on subsequent pregnancy, suggesting no increased risk of preterm birth.17,18
Preterm birth is a major cause of perinatal morbidity and mortality, and despite improvements in perinatal management in the last two decades, the rate of preterm birth has not declined.19 Research has found that transvaginal ultrasound measurement of cervical length predicts preterm birth.20 There is limited information on the effect of LEEP on cervical length as measured by transvaginal ultrasonography, with conflicting results.21,22 One recent study has suggested that transvaginal ultrasonography of the cervix is predictive of preterm birth in women with prior cone biopsy (various methods combined).23
The objective of this study was to estimate whether cervical length measured by transvaginal ultrasonography in women having had LEEP, cold knife conization, or cryotherapy predicts spontaneous preterm birth.
MATERIALS AND METHODS
This prospective cohort study recruited pregnant women with singleton gestations from June 2001 to June 2004 at the Women’s Health Centre of the Health Care Corporation of St. John’s. Inclusion criteria were women having previously had LEEP, cryotherapy, or cold knife conization for cervical dysplasia and now pregnant with singleton gestations. Two comparison groups were also included: 1) women with a history of spontaneous preterm birth not having had treatment for cervical dysplasia and 2) a low-risk control group without history of spontaneous preterm birth or treatment for cervical dysplasia. The study was approved by the Human Investigation Committee of Memorial University of Newfoundland and the ethics committee of the hospital. Written informed consent was obtained before enrollment in the study. Eligible women were approached by their attending physicians and informed of the study, and consenting women were referred for transvaginal ultrasonography. Women in the low-risk group were chosen from attending physicians’ offices, randomly selected, and consenting to transvaginal ultrasonography.
Transvaginal ultrasonography was performed by 1 of 2 of the coauthors (J.M.G.C. or T.D.), both maternal–fetal medicine specialists, between 24 and 30 weeks of gestation with the ATL HDI 5000 Ultrasound System (Phillips Medical Systems, Markham, Ontario, Canada) or Voluson 730 Ultrasound System (GE Medical Systems, Milwaukee, WI) using a 5–9 MHz transvaginal probe. With the maternal bladder empty, the cervical length was measured in the sagittal plane after visualizing simultaneously the internal and external cervical os. Three measurements were obtained. Next, suprapubic pressure was applied, displacing the presenting fetal part to determine whether funneling occurred, defined as a “V-” or “U-”shaped indentation of the internal os by the amniotic membranes. The shortest of these measurements was considered the cervical length. Cervical length measurements were not blinded to enrolled women or attending physicians. Gestational age was determined by known last menstrual period or dating ultrasonography at less than 20 weeks of gestation. Complications of each pregnancy were recorded, including antepartum bleeding after 20 weeks of gestation, the use of corticosteroids for fetal lung maturation, the use of tocolytics, hydramnios, diabetes (both gestational and pre-gestational), and maternal smoking. Demographics including gravidity, parity, maternal age, and ethnicity were recorded. The primary outcome of the study was spontaneous preterm birth (less than 37 weeks of gestation), with transvaginal ultrasound cervical length being the primary exposure variable of interest. Secondary outcomes included the occurrence of spontaneous preterm birth less than 34 weeks of gestation, with the presence of funneling on transvaginal ultrasonography being a secondary exposure variable of interest. Other neonatal outcomes included gestational age at delivery, birth weight, Apgar score, neonatal intensive care unit (NICU) admission, cord arterial pH, and perinatal morbidity and mortality. Perinatal morbidity was defined as at least one of the following: 5-minute Apgar score less than 7, cord arterial pH less than 7.10, bacterial infection within 72 hours of delivery, NICU admission more than 24 hours, seizure, ventilation after initial resuscitation, or evidence of end organ dysfunction less than 72 hours after delivery (eg, hepatic, cardiac, renal, coagulation, or hypotension). Other maternal outcomes included the route of delivery and need for induction.
Sample size was based on detecting a 5-mm difference in cervical length with a standard deviation of 8 mm (based on a previous study24), requiring 42 women per group, and on detecting a difference of spontaneous preterm birth from 7 % (based on the control group in a previous systematic review14) to 25 % (based on estimated risk of spontaneous preterm birth if there is a history of spontaneous preterm birth in a previous pregnancy,25,26), requiring 63 women per group. Sample size was based on adequate numbers of women in the LEEP group, previous spontaneous preterm birth group, and low-risk control group.
Statistical analysis was performed with Statistix 7.1, 2002 (Analytic Software, Tallahassee, Fl) and PEPI 3.01, 2000 (Computer Programs for Epidemiologists, Stone Mountain, GA). Continuous variables that were normally distributed were described and compared with Student t tests and analysis of variance (ANOVA). If the null hypothesis was rejected in ANOVA, pair-wise comparisons were performed using Tukey’s HSD test. Categorical variables were compared with χ2 or Fisher exact test, where appropriate. Ordinal variables and continuous variables that were not normally distributed were described as medians and compared with Kruskal-Wallis test. A P value less than .05 was considered statistically significant. Multiple logistic regression was used to control for potential confounders and determine which variables significantly predicted the primary outcome of spontaneous preterm birth less than 37 weeks of gestation. Variables included in the initial models were maternal age, gestational age at the time of transvaginal ultrasonography, parity, smoking, antepartum bleeding after 20 weeks of gestation, and previous spontaneous preterm birth. Variables were retained in the final models if they had a P value less than .10. Receiver operator characteristic curves were used to determine the best cutoff point for cervical length for prediction of spontaneous preterm birth less than 37 weeks. The number of women “needed to treat” for 1 additional preterm birth to occur was calculated.27
Two hundred seventy-six women were enrolled in the study—75 in the LEEP group, 21 in the cold knife conization group, 36 in the cryotherapy group, 63 in the previous spontaneous preterm birth group, and 81 low-risk controls. Maternal characteristics of the treatment and comparison groups are shown in Table 1. All woman were white except 1. The cold knife conization and cryotherapy groups had greater mean maternal ages, the previous spontaneous preterm birth group had greater median gravidity and parity, the cold knife conization and control groups had higher rates of antepartum bleeding after 20 weeks of gestation, and the cryotherapy group had a higher rate of hydramnios. With regard to the diagnosis of cervical pathology in the treatment groups, women who had LEEP tended to have more advanced CIN. Cervical pathology was not known in 5 women in the LEEP group, 6 in the cold knife conization group, and 20 women in the cryotherapy group.
Table 2 reveals the transvaginal ultrasound findings of the treatment and comparison groups. Women in the LEEP, cold knife conization, cryotherapy, and previous spontaneous preterm birth groups had significantly shorter mean cervical lengths compared with the low-risk control group (ANOVA P = .001, Tukey’s HSD P < .001, P = .03, P = .02 and P = .01, respectively). Funneling was not found to be significantly associated with any of the treatments. The presence of a cervical length less than 3.0 cm was found more frequently in the LEEP, cold knife conization, cryotherapy, and previous spontaneous preterm birth groups compared with the control group (P = .002, .03, .002, and .001, respectively).
Maternal outcomes for the groups are shown in Table 3. Women having had LEEP, cold knife conization, or a previous spontaneous preterm birth were significantly more likely to have spontaneous preterm birth less than 37 weeks of gestation compared with the low-risk control group. For every 9 LEEPs performed and subsequent pregnancy occurring, there would be 1 additional preterm birth (95% confidence interval [CI] 6–35). For every 6 cold knife conizations performed and subsequent pregnancy occurring, there would be 1 additional preterm birth (95 % CI 3–112). Women having had cryotherapy were not more likely to have spontaneous preterm birth less than 37 weeks of gestation. There were no significant differences in other maternal outcomes, including spontaneous preterm birth less than 34 weeks of gestation, gestational age at delivery, rate of cesarean delivery, or induction of labor. A total of 5 women received tocolytic therapy—3 in the LEEP group, 1 in the cryotherapy group, and 1 in the control group (P = .53). No woman in the study received progesterone for preterm birth prevention. A subgroup analysis excluding women with a history of previous spontaneous preterm birth in the cervical treatment groups found that LEEP and cold knife conization were still significantly associated with spontaneous preterm birth less than 37 weeks compared with the control group (P = .007 and .003, respectively).
Table 4 shows neonatal outcomes of the treatment and comparison groups, revealing no significant differences among the groups, including birth weight, low birth weight, Apgar score, NICU admission, cord arterial pH, or significant perinatal morbidity or mortality. Neonatal data (other than gestational age at delivery) was not available for 4 women who delivered at other centers. There were three perinatal deaths in the study—1 each in the LEEP group, the previous spontaneous preterm birth group, and the control group. One patient had spontaneous rupture of membranes at 27 weeks and 3 days gestation, resulting in a cord prolapse and stillbirth. A second patient had a stillbirth diagnosed at 32 weeks of gestation with a true knot noted in the cord at delivery. A third woman had a cesarean delivery for transverse lie at term, and a major congenital heart defect was diagnosed in the baby in the neonatal period. This child had palliative care and died at 32 days of age.
Multiple logistic regression was performed to control for potential confounders to identify predictors of spontaneous preterm birth less than 37 weeks of gestation. When the LEEP and low-risk control groups were included in the multiple logistic regression model, only the group (ie, LEEP or low-risk control) was found to be statistically significant with an odds ratio of 3.45 (95 % CI 1.28–10.00, P = .02). When evaluating women having undergone cone knife conization compared with the low-risk control group, multiple logistic regression controlling for potential confounders found the only significant variable was the group (ie, cold knife conization or low-risk control group), with an odds ratio of 2.63 (95 % CI 1.28–5.56, P = .009).
Receiver operator characteristic curves were developed to predict spontaneous preterm birth (less than 37 weeks of gestation) in women having had LEEP and in women having had cold knife conization, finding the best cutoff was a cervical length less than 3.0 cm for LEEP (Fig. 1). Sensitivity, specificity, positive and negative predictive values, and likelihood ratios for this cutoff for LEEP are summarized in Table 5. A cutoff for cold knife conization could not be easily identified due to the small sample size.
Several studies of the association of LEEP and subsequent pregnancy outcomes have had shortcomings, including small sample sizes and lack of control for potential confounders (such as smoking).9–14 In a larger study Sadler et al15 found LEEP and laser conization were associated with preterm premature rupture of membranes but not preterm labor with intact membranes leading to preterm birth. Samson et al16 found women with LEEP more likely to deliver less than 37 weeks of gestation and have infants weighing less than 2,500 g. Neither of these studies evaluated the use of transvaginal ultrasonography in these women. Berghella et al23 evaluated the effect of prior cone biopsy on the prediction of preterm birth with transvaginal ultrasonography of the cervix. They found that in women with cone biopsy by various methods combined (LEEP, cold knife conization, and laser conization), a cervical length of less than 25 mm was predictive of spontaneous preterm birth at less than 35 weeks of gestation. They did not have adequate power to evaluate specific types of cone biopsy and so did not see a significant correlation in these subgroups. They also did not examine women having undergone cryotherapy, a low-risk control group, or a subgroup of women who had had spontaneous preterm birth previously without having had cervical treatment. They did not control for smoking or other antenatal complications. They did find that funneling was significantly associated with preterm birth in their group overall.
Our current study addresses some of the shortcomings of earlier studies. We calculated a sample size a priori and had adequate numbers of women in the LEEP group, low-risk control group, and previous spontaneous preterm birth group to evaluate the primary outcomes of cervical length and preterm birth less than 37 weeks of gestation. As well, we chose to include a comparison group of women having had spontaneous preterm birth previously but not having had treatment to their cervices. Studies have shown that a history of spontaneous preterm birth is a predictor of preterm birth in subsequent pregnancy25,26 and that transvaginal ultrasound assessment of cervical length in this group of women is predictive of this outcome.20 Using multiple logistic regression, we controlled for several potential confounders. This is particularly important because the demographic characteristics of the various groups (as shown in Table 1) differed significantly on a number of characteristics. We also performed a subgroup analysis including women who had cervical treatment but had not had previous spontaneous preterm birth.
This study provides important information regarding expected cervical lengths after various cervical procedures. We found that women who had had LEEP, cold knife conization, or cryotherapy all had shorter cervical lengths than low-risk control women and similar cervical lengths to women who had a history of previous spontaneous preterm birth. Loop electrosurgical excision procedure and cold knife conization, but not cryotherapy, were associated with spontaneous preterm birth less than 37 weeks of gestation. We had an 83.7% power to detect a 20% rate of spontaneous birth in the cryotherapy group, based on the rate of spontaneous preterm birth in our control group. We found a cutoff of less than 3.0 cm to be the best cutoff to predict spontaneous preterm birth less than 37 weeks in the LEEP group.
It is important that the shortcomings of this study be addressed. We recognize that we did not have adequate numbers of women in the cold knife conization or cryotherapy groups, yet despite this we saw that both had significantly shorter cervical lengths than a low-risk control group and that cold knife conization was associated with spontaneous preterm birth less than 37 weeks of gestation. Second, we were able to control for a number of potential confounders, but we did not have information on socioeconomic status, which may be a potential confounder. In the current study, we were not able to rule out the possibility that cervical dysplasia itself may increase the risk of preterm delivery. To assess this, a control group with untreated dysplasia would be needed. It is also possible that we were not able to control for other possible confounding factors that may be associated with cervical dysplasia and preterm birth. We recognize that we did not have adequate power to see a significant difference in some of the secondary outcomes, such as significant neonatal morbidity and mortality and spontaneous preterm birth less than 34 weeks of gestation. To see an increase in perinatal morbidity from 5 % to 10 %, one would need 474 women per group, and to see a doubling in spontaneous preterm birth less than 34 weeks of gestation from 2.5 % to 5.0 %, one would need 984 women per group. An increased risk of preterm birth before 34 weeks of gestation would be more clinically important than preterm birth before 37 weeks. Also, we did not have adequate power to differentiate the different subclasses of spontaneous preterm birth, specifically, premature rupture of the membranes before the onset of labor compared with preterm labor with intact membranes. Because women were enrolled between 24 and 30 weeks of gestation we were not able to assess the risk of extreme premature delivery or cervical incompetence. Thus, this study might underestimate the influence of prior cervical surgery. Finally, we recognize that we were not able to comment on the specific characteristics of the excisional procedure (LEEP or cold knife conization), such as date of the procedure and the depth and width of the tissue sample removed, and any history of pregnancy before or between the cervical procedures and the current pregnancy, because this information was not available on all women.
The pathophysiology of preterm birth is multifactorial. Loop electrosurgical excision procedure and cold knife conization may lead to spontaneous preterm birth by reducing mechanical support through shortening of the cervix. Also, the vaginal bacterial flora may be altered and local immunologic defense mechanisms may be impaired by destruction of glandular epithelia.15,28 Perhaps cryotherapy causes less damage to the underlying cervical connective tissue, or the immunoglobulin secretory function is less damaged by cryotherapy.
This study has implications for women with abnormal Pap tests who may be considering pregnancy in the future and also for women who have already had excisional treatment to their cervices for abnormal Pap test and who are pregnant or who are considering pregnancy. It is important that guidelines on the diagnosis and treatment of cervical intraepithelial neoplasia be followed appropriately.29,30 Excisional procedures should not be routinely used to treat women with atypical squamous cells of uncertain significance in the absence of biopsy-confirmed CIN. For women with CIN, the method of treatment of cervical dysplasia should be carefully considered. For women who have undergone excisional procedures, it is important that they be counseled about the potential effect on future pregnancy. These data may be used to select high-risk women for future trials of interventions to reduce preterm birth. If future studies confirm our findings, health care providers may wish to consider closer surveillance of pregnant women who have had LEEP or cold knife conization.
1. Bristow RE, Karlan BY. Disorders of the uterine cervix. In: Scott JR, Dicsaia PJ, Hammond CB, Spellacy WN, editors. Danforth’s obstetrics and gynecology. 8th ed. Philadelphia (PA): Lippincott, Williams & Wilkins; 1999. p. 805–35.
2. Jones JM, Sweetnam P, Hibbard BM. The outcome of pregnancy after cone biopsy of the cervix: a case-control study. Br J Obstet Gynaecol 1979;86:913–6.
3. Ludviksson K, Sandstrom B. Outcome of pregnancy after cone biopsy—a case-control study. Eur J Obstet Gynecol Reprod Biol 1982;14:135–42.
4. Kristensen J, Langhoff-Roos J, Kristensen FB. Increased risk of preterm birth in women with cervical conization. Obstet Gynecol 1993;81:1005–8.
5. Lee NH. The effect of cone biopsy on subsequent pregnancy outcome. Gynecol Oncol 1978;6:1–6.
6. 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.
7. 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.
8. Luesley DM, McCrum A, Terry PB, Wade-Evans T, Nicholson HO, Mylotte MJ, et al. Complications of cone biopsy related to the dimensions of the cone and the influence of prior colposcopic assessment. Br J Obstet Gynaecol 1985;92:158–64.
9. 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.
10. Haffenden DK, Bigrigg A, Codling BW, Read MD. Pregnancy following large loop excision of the transformation zone. Br J Obstet Gynaecol 1993;100:1059–60.
11. Braet PG, Peel JM, Fenton DW. A case controlled study of the outcome of pregnancy following loop diathermy excision of the transformation zone. J Obstet Gynaecol 1994;14:79–82.
12. 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.
13. Paraskevaidis E, Koliopoulos G, Lolis E, Papanikou E, Malamou-Mitsi V, Agnantis NJ. Delivery outcomes following loop electrosurgical excision procedure for microinvasive (FIGO stage IA1) cervical cancer. Gynecol Oncol 2002;86:10–3.
14. Crane JM. Pregnancy outcome after loop electrosurgical excision procedure: a systematic review. Obstet Gynecol 2003;102:1058–62.
15. 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.
16. Samson SLA, 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.
17. Cox JT. Management of cervical intraepithelial neoplasia. Lancet 1999;353:857–9.
18. Montz FJ. Impact of therapy for cervical intraepithelial neoplasia on fertility. Am J Obstet Gynecol 1996;175:1129–36.
19. Joseph KS, Kramer MS, Marcoux S, Ohlsson A, Wen SW, Allen A, et al. Determinants of preterm birth rates in Canada from 1981 through 1983 and from 1992 through 1994. N Engl J Med 1998;339:1434–9.
20. Owen J, Iams JD. National Institute of Child Health and Human Development Maternal-Fetal Medicine Units Network. What we have learned about cervical ultrasound. Semin Perinatol 2003;27:194–203.
21. 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.
22. 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.
23. Berghella V, Pereira L, Gariepy A, Simonazzi G. Prior cone biopsy: prediction of preterm birth by cervical ultrasound. Am J Obstet Gynecol 2004;191:1393–7.
24. Iams JD, Goldenberg RL, Meis PJ, Mercer BM, Moawad A, Das A, et al. The length of the cervix and the risk of spontaneous premature delivery. National Institute of Child Health and Human Development Maternal Fetal Medicine Unit Network. N Engl J Med 1996;334:567–72.
25. Mercer BM, Goldenberg RL, Moawad AH, Meis PJ, Iams JD, Das AF, et al. The preterm prediction study: effect of gestational age and cause of preterm birth on subsequent obstetric outcome. National Institute of Child Health and Human Development Maternal-Fetal Medicine Units Network. Am J Obstet Gynecol 1999;181:1216–21.
26. Adams MM, Elam-Evans LD, Wilson HG, Gilbertz DA. Rates of and factors associated with recurrence of preterm delivery. JAMA 2000;283:1591–6.
27. Sackett DL, Richardson WS, Rosenberg W, Haynes RB. Evidence-based medicine: how to practice and teach EBM. New York (NY): Churchill Livingstone; 1997. p. 135,179–81.
28. Svare JA, Andersen LF, Langhoff-Roos J, Jensen ET, Bruun B, Lind I, et al. The relationship between prior cervical conization, cervical microbial colonization and preterm premature rupture of the membranes. Eur J Obstet Gynecol Reprod Biol 1992;47:41–5.
29. Wright TC, Cox JT, Massad LS, Twiggs LB, Wilkinson EJ. ASCCP-Sponsored Consensus Conference. 2001 consensus guidelines for the management of women with cervical cytological abnormalities. JAMA 2002;287:2120–9.
© 2006 The American College of Obstetricians and Gynecologists
30. Wright TC Jr, Cox JT, Massad LS, Carlson J, Twiggs LB, Wilkinson EJ, et al. 2001 consensus guidelines for the management of women with cervical intraepithelial neoplasia. Am J Obstet Gynecol 2003;189:295–304.