HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by EL-BASTAWISSI, A. Y. Articles by DALING, J. R. Search for Related Content PubMed PubMed Citation Articles by EL-BASTAWISSI, A. Y. Articles by DALING, J. R.

Effect of Cervical Carcinoma In Situ and Its Management on Pregnancy Outcome

From the Fred Hutchinson Cancer Research Center, Swedish Medical Center, Seattle, Washington, and Oregon Health Sciences University, Portland, Oregon.

Address reprint requests to: Janet Daling, PhD Fred Hutchinson Cancer Research Center 1100 Fairview Avenue North MP381, PO Box 19024 Seattle, WA 98109-1024 E-mail: jdaling{at}fhcrc.org

	Abstract

Top Abstract Materials and Methods Results Discussion References

 
Objective: To assess the effect of cervical carcinoma in situ (CIS) and its management on subsequent pregnancy outcome.

Methods: We used a population-based retrospective cohort design that included record linkage between cancer data and birth records. The Cancer Surveillance System records of women with CIS (n = 1851, 312 diagnosed during pregnancy) diagnosed between 1984 and 1992, were linked to birth certificates of their first subsequent deliveries after CIS diagnosis. The comparison group (n = 9201) was a random sample of women without CIS who gave birth during the same years. The outcome measures were preterm and low birth weight infants subsequent to CIS diagnosis and treatment. Treatments included no therapy, dilation and curettage or endocervical curettage, cryosurgery or laser vaporization, and conization.

Results: Women with CIS who were not treated with conization had a small increased risk of preterm delivery (odds ratio [OR] 1.4, 95% confidence interval [CI] 1.0, 2.0) and no increased risk of low birth weight infant (OR 1.0, 95% CI 0.7, 1.6), compared with women without CIS, after adjusting for maternal smoking, race, parity, marital status, and history of induced pregnancy termination. Women with CIS who had conization were more likely to deliver premature infants (OR 1.6, 95% CI 1.2, 2.0) than women without CIS, after adjusting for the same confounding factors. The apparently increased risk of low birth weight (OR 1.8, 95% CI 1.4, 2.4) seemed to be a reflection of premature delivery.

Conclusion: The risk of prematurity increased after conization for CIS and did not increase when women with CIS had other procedures.

Most cervical cancer cases are diagnosed at the in situ (87.8%) and local (6.7%) stages because of widespread cytologic screening.¹ Age-specific incidence rates for carcinoma in situ (CIS) are highest for women aged 15–44 years, ie, during their reproductive years, and peak at 200 cases per 100,000 women 25 years old.¹ Many women with CIS have not completed their desired family size. These women want to complete a current pregnancy or delay definitive treatment until the births of one or more children. Despite the increase in CIS detection, few studies have assessed the effect of CIS and the impact of its treatment on a current or future pregnancy, labor and delivery, and birth outcome.² Consequently, clinicians have few data on birth outcomes subsequent to CIS diagnosis and treatment with which they can counsel patients with CIS.

Current treatments for CIS include locally destructive methods, such as cryosurgery or laser ablation, or excisional methods, such as cold-knife or laser conization. Since the early 1990s, the loop electrosurgical excision procedure has become the procedure of choice. The extent of destruction or excision greatly affects structure and function of the cervix and might affect pregnancy outcome.

The purpose of our study was to assess pregnancy outcomes (frequency of preterm birth and low birth weight) and methods of delivery after CIS diagnosis and management.

	Materials and Methods

Top Abstract Materials and Methods Results Discussion References

 
We used a population-based, record-linkage, retrospective cohort design for this analysis.³ Records of women less than 50 years old with CIS (n = 9957) diagnosed between 1984 and 1992 were identified by the Cancer Surveillance System at the Fred Hutchinson Cancer Research Center in Seattle, Washington. These records were linked to birth certificates from the Department of Health in Washington state to determine which women delivered live singletons between 1984 and 1995 after CIS diagnosis. A random sample of women (randomized using ranuni function [SAS software, SAS Institute, Cary, NC] with an arbitrary primary number as a seed) without cervical cancer (neither CIS nor invasive cervical cancer were documented in CSS from 1974 to the time of our analysis) who gave birth during the same years was selected from birth certificates and served as a comparison group. The CIS cohort and comparison group were frequency-matched on their country of birth (foreign versus United States) and on age at delivery (mean for women with CIS 28.6 years versus 26.8 for the comparison group). To determine whether women with CIS were more likely to have poor pregnancy outcomes independent of their disease, or the factors we could adjust for, we compared the risks of poor outcomes in the pregnancies immediately preceding CIS diagnoses to pregnancies immediately preceding the index pregnancies in the comparison group, among women who had previous deliveries. Records of CIS cases were linked to birth certificates of previous deliveries closest to diagnosis from 1980 forward, whereas records of the comparison group were linked to birth certificates of immediately previous deliveries. To account for paternal genetic factors that contribute to infant birth weight, we only linked records of women who indicated the same father of the index infant and previous children. Multiple births were excluded. CIS cases and comparison women were eligible only if they resided in the 13 counties of Washington state covered by the Cancer Surveillance System. The comparison group to CIS case ratio was 5:1. If women with CIS could not be linked to their subsequent deliveries, it was due to failure to conceive, pregnancy termination, fetal death, multiple deliveries, delivery outside Washington state, or failure of our protocol to track name change.

The Cancer Surveillance System is a population-based cancer registry established in 1973 that collects information on all newly diagnosed cancer cases in 13 counties in western Washington state and has a population base of more than 3.3 million. The registry is a participant in the National Cancer Institute’s Surveillance, Epidemiology, and End Results program, which routinely monitors the quality of data for accuracy and completeness of coverage. Approximately 98% of all incident cancer cases in the 13-county area are identified by the Cancer Surveillance System. It records data through medical record abstraction and automatic reporting systems from pathology laboratories within the catchment area. The protocol by the Surveillance, Epidemiology, and End Results program for recording first courses of therapy is that if medical records indicate a treatment plan, the most extensive treatment is recorded.

The Washington state birth certificate has been a check-box format since 1980, which has provided more complete reporting of birth complications.⁴ It contains information on maternal characteristics, obstetric history (including information on spontaneous and induced pregnancy terminations and cervical incompetence), behavioral risks, labor and delivery, the newborn, and pregnancy outcome. It does not include information on cerclage. Birth certificates have been revised since 1984; therefore, we presented the results to show the changing denominators. Information on birth weight, gestation length, and delivery method was complete for 98.8%, 83.2%, and 93.8% of women with CIS versus 99.7%, 86.7%, and 94.7% for comparison women, respectively. Data on cervical incompetence were missing for 9.6% of women with CIS versus 7.8% of comparison women (P = .03), from 1989–1995.

Preterm birth was defined as birth before 37 weeks’ gestation, whereas moderate and extreme preterm births were defined as 34–36 6/7 and less than 34 weeks’ gestation, respectively. Low birth weight was defined as less than 2500 g. We categorized available diagnostic and therapeutic procedures for CIS into four groups. This categorization was based on the degree of potential destruction to the cervix caused by the procedure and on the distribution of data for each procedure. The four groups were 1) no surgical procedure (including incisional, needle, or aspiration biopsy); 2) dilation and curettage (D&C) or endocervical curettage (ECC); 3) cryosurgery or laser vaporization (with or without D&C); and 4) conization (cold knife, laser, or loop electrosurgical excision procedure, referred to here as loop excision). Our system pooled the different techniques of conization in one code, so that we could not differentiate between the primary means of conization. Information on D&C and ECC was combined. For women with CIS diagnosed during pregnancy, we assumed that they had ECC, because it seemed highly unlikely they could have D&C for pregnancy termination and immediately conceive the child we linked to. We calculated the interval in months between date of CIS diagnosis and delivery date (an interval of 9 months or less indicated diagnosis of CIS during pregnancy). CIS diagnosis was confirmed histologically for 99.5% of the cases. The remaining cases were confirmed by positive exfoliative cytologic findings. None of the women with CIS had cervical intraepithelial neoplasia grade I or II, and none had lesions less than severe dysplasia.

We used conditional logistic regression for analysis. When the analysis involved a nonbinary pregnancy outcome (such as the variable combining birth weight and gestational age in Table 3), we used polytomous logistic regression. Because the results of both models were similar, we presented the results of the conditional logistic regression.

	Results

Top Abstract Materials and Methods Results Discussion References

The most common procedure in women with CIS before the index birth was conization (62.9%). Of the 1525 CIS cases with known procedures and gestational ages of neonates at birth, 185 (12.1%) had preterm births. Women who had conization had a slightly higher proportion of premature births (13.2%) compared with women with CIS who had no surgical procedures (11.0%) (P = .333). Of the 1832 women with CIS with known procedures and birth weight of the neonate, 137 (7.5%) had low birth weight infants. Women who had conization had more low birth weight infants (8.5%) compared with women with CIS who had no surgical procedures (5.5%) (P =.044). Of the 1723 women with CIS with known procedures and delivery methods, 379 (22.0%) delivered by cesarean. Women who underwent conization had more cesarean deliveries (23.6%) compared with women with CIS who had no surgical procedures (19.9%) (P = .168). Women in whom CIS was diagnosed during pregnancy were more likely to have no surgical procedures and less likely to have cryosurgery or laser vaporization or conization, compared with women who became pregnant after CIS diagnosis (P = .000, .01, and .000, respectively). There were no significant differences in birth outcome or delivery method among treatment groups, when women with CIS during pregnancy were compared with women who became pregnant after CIS diagnosis, except that women who were pregnant at diagnosis and had cone biopsies were less likely to deliver by cesarean, when compared with women who became pregnant after diagnosis and had cone biopsies (P = .033). Table 2 illustrates the frequencies of various procedures by birth outcome, delivery method, and pregnancy status at CIS diagnosis.

We analyzed a subset of CIS and comparison groups, from 1992 to 1995, to evaluate the effect of further adjusting for maternal education (12 or fewer versus more than 12 years of education) and socioeconomic status (Medicaid-charity versus not Medicaid-charity) on pregnancy outcome. These variables have been collected on birth certificates only since 1992. Further adjustment for maternal education only increased the risk of low birth weight from 2.1 (95% CI 1.4, 3.3) to 2.4 (95% CI 1.5, 3.8), among women with CIS who had cone biopsies. There was no substantial change in risk of adverse pregnancy outcome among women with CIS who did not have cone biopsies.

Among the 1825 women with CIS and known parity, 955 (52.3%) had one or more previous deliveries, of which 607 (63.6%) linked to previous birth certificates, and 387 (63.8%) of those had the same fathers. Of the 9055 comparison women with known parity, 5605 (61.9%) had one or more previous deliveries of which 3698 (66.0%) linked to previous birth certificates, and 2851 (77.1%) of those had the same fathers. Women with CIS were more likely to have different or unknown fathers of their previous deliveries than women without CIS (different fathers: OR 1.2, 95% CI 0.9, 1.7; unknown fathers: OR 2.1, 95% CI 1.7, 2.5). Women with CIS did not have more preterm births (OR 0.8, 95% CI 0.5, 1.2) or low birth weight infants (OR 0.7, 95% CI 0.4, 1.2) before CIS diagnosis compared with control women’s deliveries immediately before the index delivery.

	Discussion

Top Abstract Materials and Methods Results Discussion References

 
The most important findings of this study are that women with CIS who did not have cone biopsies were neither at increased risk of adverse pregnancy outcome, nor were they at increased risk of cesarean delivery compared with women without CIS. Women with CIS who had cone biopsies were more likely to have preterm infants by cesarean delivery. Women with CIS had increased unadjusted risk of cervical incompetence compared with women without CIS.

Comparable to our findings, several studies found no association between non-cone procedures and adverse pregnancy outcomes.^5,6 These studies were hospital based, were limited by small sample size and lack of control for confounding factors, and investigated cervical intraepithelial neoplasia rather than CIS. Our findings on the association of CIS with adverse pregnancy outcomes when cone biopsies were performed are consistent with those of other investigators.^7–9 One study found no adverse pregnancy outcomes associated with loop excision for CIS.¹⁰ Those studies had similar limitations.

Because women with CIS were not at increased risk for adverse pregnancy outcome before CIS diagnosis, the observed increased risk subsequent to diagnosis is most likely related to CIS and its treatment. The lack of considerable damage to the structure or function of the cervix among women with CIS who did not have cone biopsies might explain their favorable pregnancy outcomes. The association of cone biopsy with extreme prematurity suggests that some degree of cervical incompetence might have had a precipitating effect on those preterm deliveries; however, no specific test or criterion for accurate diagnosis of cervical incompetence exists.¹¹ Practitioners might have selectively misclassified women with CIS as having incompetent cervices on the basis of the knowledge that these women had previous CIS, leading to overestimating the risk of cervical incompetence. Conversely, women with CIS who had severe cervical incompetence might have had fetal losses we did not determine; hence, the risk of cervical incompetence among those women might have been underestimated. The number of women with cervical incompetence was also too small for us to draw meaningful conclusions. We are limited by not having information on cerclage. About 80% (547 of 681) of women with CIS who did not have cone biopsies have had some kind of procedure that inflicted cervical trauma of varying degrees. This might explain the increased risk of cervical incompetence among these women.¹²

Comparable to other studies,⁷ our data showed a significantly increased risk of cesarean delivery among women with CIS who had cone biopsies. However, our data did not allow us to determine whether the increased risk of cesarean delivery was caused by obstetric complications or by physicians’ or women’s choice. It seems reasonable to assume that increased cesarean delivery among women with CIS who had cone biopsies is more likely caused by obstetric complications, perhaps resulting from cervical dystocia from scarring.⁷ Women with CIS who had a cone biopsies were more likely to have prolonged labor compared with those who did not have a cone biopsies and controls. The significant difference in cesarean deliveries between women who had cone biopsies during pregnancy versus before pregnancy suggests that the latter women had enough time to develop scar tissue at the site of conization, which the former women lacked. The increased risk of cesarean could also have been from complicated deliveries associated with prematurity, such as breech presentations, placental abruptions, and preeclampsia, which occurred more frequently among women with CIS who had a cone biopsies compared with both women with CIS who did not have cone biopsies and controls.

Population-based data are less subject to potential selection bias than hospital-based studies of patients from referral centers.¹³ Recall bias or nonparticipation was minimized using the record-linkage approach, which was a powerful tool to identify large numbers of women with CIS who delivered infants after diagnosis of CIS. Valuable data were obtained inexpensively and quickly, without jeopardizing quality. This large study allowed for adequate statistical power to examine the effect of CIS on birth outcome.

The most important limitation to our study was the lack of information to adjust for age at first intercourse, number of sexual partners, and history of sexually transmitted diseases.¹⁴ Our data showed no difference between women with CIS and the comparison group for genital herpes. We further controlled for this limitation by using the available information that might be considered markers for those exposures, such as marital status and smoking. Information on attempts to conceive after CIS diagnosis was also not available; therefore, we could not assess the impact of CIS on women’s fertility. The lower parity of women with CIS and the fewer living children those women had at index delivery suggest that they have not completed their families and are likely to attempt conception.

Future research should focus on differentiating between the effects of various means of conization on pregnancy outcomes subsequent to CIS diagnosis. It would be also important to determine the frequency of fetal losses for those women.

	Footnotes

 
Funding was provided by The Surveillance, Epidemiology, and End Results (SEER) and the National Institutes of Health/National Cancer Institute, contract number NO1 CN 67009.

PII S0029-7844(98)00386-X

Received May 1, 1998. Received in revised form July 23, 1998. Accepted August 6, 1998.

	References

Top Abstract Materials and Methods Results Discussion References

 
1. Wiggins CL, Thomas DB, Potts MS. Cancer in western Washington state 1974–1991. Seattle, Washington: Cancer Surveillance System, Fred Hutchinson Cancer Research Center 1992:36–37.

2. Ueki M, Ueda M, Kumagai K, Okamoto Y, Noda S, Matsuoka M. Cervical cytology and conservative management of cervical neoplasias during pregnancy. Int J Gynecol Pathol 1995;14:63–9.[Medline]

3. Mueller BA, Schwartz SM. Risk of recurrence of birth defects in Washington state. Paediatr Perinat Epidemiol 1997;11:107–18.

4. Frost F, Starzyk P, George S, McLaughlin JF. Birth complication reporting: The effect of birth certificate design. Am J Public Health 1984;74:505–6.[Abstract/Free Full Text]

5. Hemmingsson E. Outcome of third trimester pregnancies after cryotherapy of the uterine cervix. Br J Obstet Gynaecol 1982;89: 675–7.[Medline]

6. Anderson MC, Horwell DH, Broby Z. Outcome of pregnancy after laser vaporization conization. Colposcopy Gynecol Laser Surg 1984;1:35–39.

7. Myllynen L, Karjalainen O. Pregnancy outcome after combined amputation and conization of the uterine cervix. Ann Chir Gynaecol 1984;73:345–9.[Medline]

8. Hagen B, Skjeldestad FE. The outcome of pregnancy after CO2 laser conisation of the cervix. Br J Obstet Gynaecol 1993;100:717–20.[Medline]

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.[Medline]

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.[Medline]

11. Creasy RK, Resnik R. Maternal-fetal medicine; principles and practice. 3rd ed. Philadelphia, Pennsylvania: WB Saunders, 1994.

12. Caspi E, Schneider D, Sadovsky G, Weinraub Z, Bukovsky I. Diameter of cervical internal os after induction of early abortion by laminaria or rigid dilatation. Am J Obstet Gynecol 1983;146:106–8.[Medline]

13. Hickok DE, Gordon DC, Milberg JA, Williams MA, Daling JR. The frequency of breech presentation by gestational age at birth: A large population-based study. Am J Obstet Gynecol 1992;166: 851–2.[Medline]

14. Eschenbach DA, Gravett MG, Chen KCS, Hoyme UB, Holmes KK. Bacterial vaginosis during pregnancy: An association with prematurity and postpartum complications. Scand J Urol Nephrol 1984; 86:213–22.

This article has been cited by other articles:

				S.-L. A. Samson, J. R. Bentley, T. J. Fahey, D. J. McKay, and G. H. Gill The Effect of Loop Electrosurgical Excision Procedure on Future Pregnancy Outcome Obstet. Gynecol., February 1, 2005; 105(2): 325 - 332. [Abstract] [Full Text] [PDF]

				L. Sadler, A. Saftlas, W. Wang, M. Exeter, J. Whittaker, and L. McCowan Treatment for Cervical Intraepithelial Neoplasia and Risk of Preterm Delivery JAMA, May 5, 2004; 291(17): 2100 - 2106. [Abstract] [Full Text] [PDF]

				Cervical Cone Biopsy and Subsequent Preterm Birth Journal Watch Women's Health, March 1, 1999; 1999(301): 6 - 6. [Full Text]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by EL-BASTAWISSI, A. Y. Articles by DALING, J. R. Search for Related Content PubMed PubMed Citation Articles by EL-BASTAWISSI, A. Y. Articles by DALING, J. R.