| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ORIGINAL RESEARCH |
From the Divisions of Gynecologic Oncology and Maternal-Fetal Medicine, Departments of Obstetrics and Gynecology, Radiation Oncology, and Pharmacology, University of Iowa Hospitals and Clinics, Iowa City, Iowa.
Address reprint requests to: Anil K. Sood, MD Division of Gynecologic Oncology Department of Obstetrics and Gynecology 200 Hawkins Drive, 4630 JCP Iowa City, IA 52242 E-mail: anil-sood{at}uiowa.edu
 |
Abstract |
|---|
Top
 Abstract Materials and MethodsResultsDiscussionReferences |
|---|
Methods: A matched case-control study of women with cervical cancer diagnosed during pregnancy or within 6 months of delivery was performed. Fifty-six women had cervical cancer diagnosed during pregnancy and 27 within 6 months after delivery. Controls (cervical cancer diagnosed at least 5 years since last delivery) were matched one-to-one with cases based on age, histology, stage, treatment, and time of treatment.
Results: Among postpartum women, four had stage IA disease, 15 had stage IB1 or IB2, and eight had stage IIA or higher disease. Eleven had radical hysterectomies and 14 had radiation therapy. Two with stage IA1 disease were treated with vaginal hysterectomies. One of seven patients who had cesareans developed a local and distant recurrence. In contrast, ten of 17 (59%) who delivered vaginally developed recurrences (P = .04). In multivariate analysis, vaginal delivery was the most significant predictor of recurrence (odds ratio [OR] 6.91; 95% confidence interval [CI] 1.45, 32.8), followed by high stage (OR 4.66; 95% CI 1.05, 20.8). The survival for patients diagnosed in the postpartum period was significantly worse than for controls.
Conclusion: Women diagnosed postpartum had worse survival than those diagnosed during pregnancy and were at significant risk of recurrent disease, particularly if they delivered vaginally. Therefore, pregnant women with cervical cancer should be delivered by cesarean.
Invasive cervical carcinoma during pregnancy is relatively uncommon but remains the most common malignancy associated with pregnancy. Its incidence is about 0.05% among all pregnant women.1 Thus, most institutions have limited experience treating these women.
Pregnancy is a good time for cervical screening because cytology routinely is done during prenatal care. However, 48–49% of cervical cancers associated with pregnancy are diagnosed within 6 months of delivery.2,3 Most authors have combined women diagnosed postpartum with those diagnosed during pregnancy. It is extremely unlikely that all these women developed new cancers subsequent to delivery. Approximately half of cervical cancers associated with pregnancy are incorrectly diagnosed during pregnancy, the reasons for which have not been explored. The effect of delivery mode on outcome and prognosis has been described only in small case reports.4
Most authors agree that pregnancy does not alter the outcomes of women with cervical cancer.2,3,5,6 Some suggested that prognoses might be worse for women diagnosed in later pregnancy.7 Concerns about vaginal delivery through a cervix with malignancy include hemorrhage and tumor dissemination at delivery. Some authors have recommended that cesarean be performed solely for obstetric indications. Most studies have not analyzed separately the women who were diagnosed postpartum. The objectives of our study were to evaluate the outcomes of women diagnosed with cervical cancer during pregnancy and within 6 months after delivery, to identify possible reasons for lack of diagnosis during pregnancy, and to assess the effect of vaginal delivery on recurrence risk and prognosis.
|
Materials and Methods |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
We sought to evaluate possible effects of pregnancy on the course of cervical cancer. To do so, we matched pregnant women, separately from those diagnosed within 6 months of delivery, to nonpregnant women with cervical cancer. Controls, who were at least 5 years since last delivery, were drawn from the University of Iowa tumor registry and matched by age (±5 years), histology, stage of disease, treatment, and year of treatment (±5 years). Controls were matched for aborted radical hysterectomy as well.
Diagnoses were made by biopsy and verified by pathology review at the institutional gynecologic oncology tumor board. Stage was determined after examination by a gynecologic oncologist and radiation oncologist. For consistency, the League of Nations staging of patients in the early years of the study was converted to the most recent International Federation of Gynecology and Obstetrics clinical staging system.8
Subjects were counseled on treatment options. The choice of treatment was based on stage, histology, extent of disease, and medical condition. For women with cervical cancer diagnosed during pregnancy, treatment decisions also were based on the desire for the pregnancy. If invasive cancer was diagnosed before 20 weeks’ gestation, immediate treatment usually was recommended. After 20 weeks’ gestation, timing of treatment depended on stage, histology, and lesion size. Women with stage IA or small IB squamous lesions were given the option to delay treatment to achieve fetal viability. They were examined carefully at 3–4-week intervals for evidence of progression. During the study, treatment of pregnant women was similar to that of nonpregnant women. Records were reviewed with particular attention to initial history and physical examination, histopathology, postoperative treatment, and follow-up. For pregnant women, prenatal course, delivery information, and fetal outcomes also were reviewed. For women diagnosed with cancer postpartum, prenatal and delivery records were reviewed.
Women were observed by a gynecologic oncologist every 3 months during the first year after treatment, every 4 months during the second year, every 6 months in the third through fifth years, and yearly thereafter. Women treated with radiation also were observed by a radiation oncologist. Papanicolaou smears were collected at each visit and chest radiographs were done yearly. Long-term follow-up was achievable for all women.
Chi-square, Fisher exact, McNemar, or paired t tests were used as appropriate to determine differences between variables with the BMDP statistical software program (BMDP Statistical Software, Los Angeles, CA). Kaplan-Meier survival plots were generated and comparisons between survival curves were made with the log-rank statistic. The Cox proportional hazards model was used for stepwise multiple linear regression to select significant prognostic factors for survival. Logistic regression was used to estimate multivariate odds ratios (ORs) with 95% confidence intervals (CIs) for risk of recurrence as a function of significant univariate variables. P < .05 was considered statistically significant.
|
Results |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
presents demographic data and clinical characteristics of the women diagnosed during pregnancy (n = 56) or postpartum (n = 27) and of the controls. There were no statistically significant differences between cases and controls in age, gravidity, parity, and smoking status (Table 1). Diagnosis was made by cytologic screening much more commonly in pregnant women (45%) than in controls (25%) (P = .03). Histologic diagnoses among pregnant women included squamous cell carcinoma in 49 (88%), adenocarcinoma in four (7%), and adenosquamous carcinoma in three (5%). Squamous tumors also were diagnosed in 24 (89%) of the postpartum women, and three (11%) were adenocarcinomas.
|
. Six women did not receive prenatal care, 11 had normal prenatal Papanicolaou smear results, and four did not have prenatal Papanicolaou smears. Six women had abnormal Papanicolaou smear results, and three of these were observed during pregnancy with colposcopy for severe dysplasia. These three were diagnosed with high-grade dysplasia (cervical intraepithelial neoplasia III) based on colposcopic biopsies during pregnancy, and they had cone biopsies 4–12 weeks after delivery. Two women were subsequently diagnosed with stage IB1 squamous cell carcinoma and one with stage IA1 disease. At delivery, cervical examinations were normal in 18 women, abnormal in seven, and unknown in two. The many factors that led to diagnoses of cancer are listed in Table 2. Of those women with normal prenatal Papanicolaou smear results, two had abnormal postpartum cytology results that led to cone biopsies, five had lesions at postpartum examinations that were biopsied, and four had persistent bleeding postpartum that led to biopsies and diagnoses of cancer.
|
Thirteen women had planned treatment delays ranging from 3 to 32 weeks. Details of the treatments and outcomes of these women have been reported.5,6 Among women diagnosed prenatally who delivered in the third trimester, only 3 of 32 (9%) had vaginal deliveries (Table 3). In contrast, 17 of 24 (71%) with advanced gestation, diagnosed postpartum, had vaginal deliveries, and only seven had abdominal deliveries for obstetric (not oncologic) indications (repeat cesarean in three and labor abnormalities in four). Estimated blood loss during vaginal deliveries ranged from 250 to 700 mL (mean 432 mL). No women required blood transfusions.
|
Among postpartum women, 11 had radical hysterectomies and 14 had radiation therapy. Two with stage IA1 disease were treated with vaginal hysterectomies. Among women treated with radical hysterectomies, two had microscopically positive lymph nodes. Both were treated postoperatively with radiation; one is alive 7 years after treatment with no evidence of recurrence and the other had recurrence in the pelvis at 5 months and died. Among women treated primarily with radiation, one with stage IB2 disease had exploration for a radical hysterectomy that was aborted because of grossly positive lymph nodes, which were resected. She developed pelvic and distant recurrences 23 months after completion of radiation and died of her disease. There were no long-term complications among women treated surgically. Among the 14 treated with radiation, two (14%) developed long-term complications and both developed bowel obstructions.
The patterns of recurrence for the postpartum patients based on stage are presented in Table 4. Only one woman with stage IB1 or less had recurrent disease. Sixty-two percent of stage IB2 patients had recurrent disease: one pelvic, two distant, and two pelvic and distant. Sixty-seven percent of women with stage IIB disease had recurrences, including one pelvic and three pelvic and distant. One with stage IIA cervical cancer was treated with radiation therapy, but 5 months after delivery she developed a recurrence at her episiotomy site and died of progressive disease despite further radiation. The woman with stage IIIB disease had a central pelvic recurrence at 11 months and underwent exenteration, but died of recurrent disease within 12 months. As in women diagnosed during pregnancy, the salvage rate for recurrent disease was poor, with no survivors.
|
. Vaginal delivery and high stage were significant risk factors. Diagnosis in the postpartum period approached significance as a risk factor for recurrence. There was no correlation between recurrence and smoking status or histology. These factors were considered possible independent predictors of recurrence using logistic regression. Vaginal delivery was the most significant predictor of recurrence, with an OR of 6.91 (95% CI 1.45, 32.8), followed by high stage (OR 4.66; 95% CI 1.05, 20.8).
|
) or for those diagnosed during pregnancy (P = .03). Survival based on mode of delivery is shown in Figure 2. Women who delivered vaginally had significantly worse survival than those delivered by other means (P = .001). The risk of distant recurrence was particularly high. Sixty percent of women who had treatment failure after delivering vaginally had distant recurrence alone or with pelvic recurrence. Significant variables for survival (stage, mode of delivery, and time of diagnosis) were evaluated using the Cox proportional hazards model. High stage (P < .001) was the most significant adverse prognostic factor, followed by vaginal delivery (P = .03).
|
|
|
Discussion |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Pregnancy appears to be an ideal setting for cervical cancer screening as part of prenatal care. This is reflected by the predominance of low-stage disease in women diagnosed during pregnancy; almost half (45%) (Table 1
) are diagnosed by cytologic screening. However, a substantial number of patients are not diagnosed with cancer during pregnancy, and their cancers are detected postpartum.2,3 The reasons for the delay of diagnosis have not been evaluated and should be assessed carefully to enhance prenatal screening programs for cervical cancer. Among the postpartum women in this study, six (22%) had no prenatal care. Thus, cervical cancer detection must be improved. Three women (11%) who had abnormal Papanicolaou smear results prenatally and were observed colposcopically during the pregnancy had early-stage cervical cancer and have been disease-free for more than 5 years after completion of treatment. We have shown previously that pregnant women with high-grade cervical dysplasia have a high likelihood of persistent disease postpartum, and up to 8% have invasive cancer.11 Given the high rate of underestimation of disease during pregnancy, even with colposcopy, cervical biopsy should be done at colposcopy. This group of women, with asymptomatic and microscopic disease diagnosed postpartum, appear to have an excellent prognosis with vaginal delivery.11 The remainder of cancers were missed during prenatal screening for many reasons, ranging from lack of cytologic screening to lack of investigation of persistent symptoms postpartum (Table 2). With effective cytologic screening or follow-up of symptoms, careful visualization and palpation of the cervix, and biopsies when indicated, most of the remainder of cases could have been detected prenatally. Physician error was most commonly due to inappropriate attribution of symptoms of cervical cancer to benign causes common in pregnancy, such as postcoital bleeding and discharge.
Modes of delivery should be selected carefully for pregnant women with cervical cancer. Concern has been expressed about possible infection, hemorrhage, obstructed labor, and dissemination of tumor cells caused by dilation of a cervix with cancer.1 Because of the relative rarity of cervical cancer in pregnancy, there are no randomized studies addressing the benefits of cesarean versus vaginal delivery. Previous studies conflicted regarding the optimal mode of delivery. Results have varied from no difference in survival2,3,12 to improved survival13 or worse survival14,15 when vaginal delivery was compared with cesarean. Jones et al7 recently reported poorer outcome with vaginal delivery than with cesarean (55% versus 75%, P not reported). Our study suggests that vaginal delivery is associated with poorer survival for both those diagnosed prenatally and postpartum. When only women diagnosed postpartum were considered, vaginal delivery remained a significant adverse prognostic factor for recurrent disease and poor survival.
Other potential concerns, such as recurrence at the episiotomy site after vaginal delivery, also should be considered. Although some authors reported successful treatment of episiotomy site recurrence,16,17 these recurrences can be associated with high morbidity and mortality.4 This adverse result is confirmed by the death of our subject who had an episiotomy site recurrence, the 11th such patient in the literature according to a MEDLINE search from 1960 to 1997 using the terms "cervix neoplasms" and "episiotomy." Recurrence is directly related to vaginal delivery, similar to skin metastasis after laparoscopy.18 Thus, all women diagnosed with cervical cancer postpartum who had vaginal deliveries should be examined carefully, and the episiotomy site should be palpated. Cesarean delivery might eliminate the risk of episiotomy site recurrence.
Previous studies have not evaluated recurrence patterns based on modes of delivery. In our postpartum subjects, recurrent cancer after cesarean was rare compared with after vaginal delivery. Most recurrences after vaginal delivery also involved distant sites. Thus, vaginal delivery might play a role in the dissemination of tumor cells. Salvage after recurrent disease was difficult, and all patients with recurrent disease died. Therefore, women with cervical cancer who have vaginal deliveries should be considered for chemotherapy because of the high rate of recurrence and poor salvage after distant metastasis.
The relation between outcome and time of diagnosis (prenatal or postpartum) is controversial. In many studies, women diagnosed postpartum had worse survival than those diagnosed during pregnancy.12,13,19 Jones et al7 found worse survival for women diagnosed in the third trimester than for those diagnosed in the first two trimesters. Sivanesaratnam et al20 reported that only one of 16 women diagnosed during pregnancy died, compared with three of four deaths (75%) among those diagnosed within 20 weeks after delivery. Nisker and Shubat21 found no effect of gestational age or prenatal or postpartum diagnosis on survival in 43 women with cervical cancer. Similarly, van der Vange et al2 found no difference in survival between women diagnosed during pregnancy and those diagnosed after delivery or abortion. In our study, univariate analysis showed that women diagnosed postpartum had worse survival than those diagnosed during pregnancy. However, this effect most likely was caused by more advanced stage and vaginal delivery because in multivariate analysis, only stage and vaginal delivery were significant variables.
In summary, cesarean should be the delivery mode of choice for pregnant women with cervical cancer. Cervical cytology, careful inspection and palpation of the cervix, careful follow-up, and a low threshold for cervical biopsy are recommended during prenatal care.
|
Footnotes |
|---|
Received August 11, 1999. Received in revised form December 7, 1999. Accepted December 23, 1999.
|
References |
|---|
|
Top
AbstractMaterials and MethodsResultsDiscussionReferences |
|---|
2. van der Vange N, Weverling GJ, Ketting BW, Ankum WM, Samlal R, Lammes FB. The prognosis of cervical cancer associated with pregnancy: A matched cohort study. Obstet Gynecol 1995;85: 1022–6.[Abstract]
3. Lee RB, Neglia W, Park RC. Cervical carcinoma in pregnancy. Obstet Gynecol 1981;58:584–9.
4. Cliby WA, Dodson MK, Podratz KC. Cervical cancer complicated by pregnancy: Episiotomy site recurrences following vaginal delivery. Obstet Gynecol 1994;84:179–82.
5. Sood AK, Sorosky JI, Krogman S, Anderson B, Benda J, Buller RE. Surgical management of cervical cancer complicating pregnancy: A case-control study. Gynecol Oncol 1996;64:294–8.
6. Sood AK, Sorosky JI, Mayr N, Krogman S, Anderson B, Buller RE, et al. Radiotherapeutic management of cervical carcinoma that complicates pregnancy. Cancer 1997;80:1073–8.[Medline]
7. Jones WB, Shingleton HM, Russell A, Fremgen AM, Clive RE, Winchester DP, et al. Cervical carcinoma and pregnancy. A national patterns of care study of the American College of Surgeons. Cancer 1996;77:1479–88.[Medline]
8. Creasman WT. New gynecologic cancer staging. Gynecol Oncol 1995;58:157–8.[Medline]
9. Sorosky JI, Squatrito R, Ndubisi BU, Anderson B, Podczaski ES, Mayr N, et al. Stage I squamous cell cervical carcinoma in pregnancy: Planned delay in therapy awaiting fetal maturity. Gynecol Oncol 1995;59:207–10.[Medline]
10. Sorosky JI, Cherouny PH, Podczaski ES, Hackett TE. Stage IB cervical carcinoma in pregnancy: Awaiting fetal maturity. J Gynecol Tech 1996;2:155–8.
11. Coppola A, Sorosky J, Casper R, Anderson B, Buller RE. The clinical course of cervical carcinoma in situ diagnosed during pregnancy. Gynecol Oncol 1997;67:162–5.[Medline]
12. van Praagh IGL, Harvey MH, Vernon CP. Carcinoma of the cervix associated with pregnancy. J Obstet Gynaecol Br Commonw 1965;72:75–80.[Medline]
13. Bosch A, Marcial VA. Carcinoma of the uterine cervix associated with pregnancy. Am J Roentgenol 1966;96:92–9.
14. Barber HRK, Brunschwig A. Gynecologic cancer complicating pregnancy. Am J Obstet Gynecol 1963;85:156–64.[Medline]
15. Mikuta JJ. Invasive carcinoma of the cervix in pregnancy. South Med J 1967;60:843–7.[Medline]
16. Gordon AN, Jensen R, Jones HW 3d. Squamous carcinoma of the cervix complicating pregnancy: Recurrence in episiotomy after vaginal delivery. Obstet Gynecol 1989;73:850–2.[Medline]
17. Van Dam PA, Irvine L, Lowe DG, Fisher C, Barton DPJ, Shepherd JH. Carcinoma in episiotomy scars. Gynecol Oncol 1992;44:96–100.[Medline]
18. Patsner B, Damien M. Umbilical metastases from a stage IB cervical cancer after laparoscopy: A case report. Fertil Steril 1992;58:1248–9.[Medline]
19. Gustafsson DC, Kottmeier HL. Carcinoma of the cervix associated with pregnancy. Acta Obstet Gynecol Scand 1962;41:1–21.
20. Sivanesaratnam V, Jayalakshmi P, Loo C. Surgical management of early invasive cancer of the cervix associated with pregnancy. Gynecol Oncol 1993;48:68–75.[Medline]
21. Nisker JA, Shubat M. Stage IB cervical carcinoma and pregnancy: Report of 49 cases. Am J Obstet Gynecol 1983;145:203–6.[Medline]
This article has been cited by other articles:
|
|
 |
|
  A. K. Sood Cancer in Surgical Scar: Insights and Clinical Challenges Obstet. Gynecol., August 1, 2007; 110(2): 453 - 454. [Full Text] [PDF]
|
|
|
|
|
|
G. Neumann, K. L. Rasmussen, and L. K. Petersen Cervical Adenosquamous Carcinoma: Tumor Implantation in an Episiotomy Scar Obstet. Gynecol., August 1, 2007; 110(2): 467 - 469. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Sadler and P. Sykes How little is known about cervical cancer in pregnancy? Ann. Onc., March 1, 2005; 16(3): 341 - 343. [Full Text] [PDF]
|
|
|
|
|
|
N. Germann, C. Haie-Meder, P. Morice, C. Lhomme, P. Duvillard, K. Hacene, and A. Gerbaulet Management and clinical outcomes of pregnant patients with invasive cervical cancer Ann. Onc., March 1, 2005; 16(3): 397 - 402. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. I. Sorosky Cervical Cancer Complicating Pregnancy: Two Patients With Three Clinical Objectives Obstet. Gynecol., November 1, 2004; 104(5): 1127 - 1128. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
