Cervical dysplasia in pregnancy: a multi-institutional evaluation




Objective


This study was undertaken to identify the prognostic indicators associated with postpartum regression of cervical dysplasia diagnosed in pregnancy.


Study Design


A retrospective cohort study of pregnant women referred for colposcopy from 2004-2007 at four academic centers.


Results


One thousand seventy-nine patients were identified. Colposcopic impression by cervical cytology is detailed later in the text. Of patients who underwent biopsy, results correlated with or were less severe than colposcopic impression in 83% with CIN 1 and 56% with CIN 2/3. Fifty-seven percent had follow-up postpartum, with 61% reverting to normal. Resolution of cervical dysplasia was inversely associated with smoking ( P = .002). No progression to cancer occurred during pregnancy.


Conclusion


Colposcopic impression in pregnancy correlated with cervical biopsy results and postpartum colposcopic findings when performed by expert colposcopists. A high proportion of cervical dysplasia regressed postpartum. Cervical biopsies in pregnancy may not be necessary unless invasive cancer is suspected.


Approximately 2-3 million abnormal Papanicolaou (Pap) smears are diagnosed each year in the United States and 5-13% of these abnormal tests are detected in pregnant women. Of the approximately 4 million women who become pregnant in the United States each year, between 2-20% will have abnormal cervical cytology noted on a Pap test. Cervical cancer is the most common malignancy diagnosed during pregnancy, and cervical neoplasia, which includes carcinoma-in-situ and invasive carcinoma, is estimated to occur in 1.5-12 of every 100,000 pregnancies.


Most cervical abnormalities in pregnancy are discovered as a result of routine screening at the initiation of prenatal care. Management guidelines for cervical dysplasia in pregnancy are not well defined and are derived from data in nonpregnant women, expert opinion, anecdotal experience or retrospective series of pregnant women. The latter studies, the majority of which are small, retrospective reports, demonstrate varied postpartum regression rates of 12-97% and progression rates of 2-60%. Overall, the prevalence of abnormal cervical cytology in pregnancy is similar to that of age-matched, nonpregnant women. However, adequate data does not exist to develop definitive, evidence-based guidelines for cervical dysplasia diagnosed in pregnancy. Therefore, the purpose of this study was to generate the largest report in the literature of pregnant women who had cervical dysplasia diagnosed and to identify the prognostic indicators associated with postpartum regression, persistence or progression of various cytologic abnormalities.


Materials and Methods


A retrospective cohort study of consecutively treated pregnant woman who underwent colposcopy from 2004-2007 at 4 US academic centers was performed. Institutional review board approval was obtained at each participating site. All sites served as referral centers for outlying communities and contained high-volume colposcopy specialty clinics run by expert colposcopists.


Patients were identified via colposcopy databases. Data abstracted from medical records included patient demographics, tobacco and sexual history, gestational age (EGA) at time of colposcopy, referral cytology, colposcopic impression, biopsy results (if performed), recommendations, and follow-up cytology or follow-up colposcopic impression. All study subjects had at least 1 antepartum cervical cytology and colposcopy performed. Select patients underwent colposcopic biopsy at the discretion of the treating physician and most patients underwent a postpartum colposcopic evaluation. Directed cervical biopsy was performed most commonly on the following patients: (1) those with colposcopic evidence of invasion or cervical intraepithelial neoplasia (CIN) 3; (2) those with significant discordance between colposcopy and cytology; and (3) those in whom noncompliance postpartum was a concern.


Statistics were primarily descriptive. Patient demographics, cervical cytology and colposcopy impressions, and regression/progression rates were compared between pregnant smokers and nonsmokers by Student t test, χ 2 test, or Fisher’s exact test. The χ 2 or Fisher’s exact test were used to test independent variables in univariate analysis to identify prognostic factors. Cox proportional hazards regression model was used for multivariate analysis. All statistical tests were performed using SAS software (SAS Institute, Inc., Cary, NC), and 2-tailed P values < .05 was considered statistically significant.




Results


One thousand seventy-nine patients were identified. Patient demographic data are presented in Table 1 . The median patient age was 22.5 years, and the median EGA at the time of first colposcopy was 21 weeks with trimester breakdown as follows: 13% in first, 67% in second, and 20% of patients evaluated in the third trimester. Fifty-one percent of patients were in their first pregnancy. The majority of patients were white, and they had a median age of first intercourse at age 16 with a median number of 7 sexual partners. Thirty-two percent of patients had a sexually transmitted infection history, and almost 5% had undergone a prior ablative or excisional cervical procedure. Thirty-four percent of women were current or past tobacco users with a mean pack per day use of 0.5 (standard deviation [SD], 0.33).



TABLE 1

Patient demographics (n = 1079)

































































































Variables No. (Range, SD, or %)
Mean patient age, y 22.6 (4.46)
Mean estimated gestational age (wk) at first colposcopy 21 (7.1)
Race
White 351 (41.7%)
African American 273 (31.5%)
Hispanic 155 (18.4%)
Asian 18 (2.1%)
Native American 5 (0.6%)
Other 39 (4.6%)
Unknown 243 (22.5%)
No. of pregnancies
1 555 (51.5%)
2 280 (26%)
3 148 (13.7%)
≥4 94 (8.7%)
Mean age at first sexual intercourse, y 16.3 (3.11)
Mean no. of lifetime partners 7 (12.4)
STI history
Yes 334 (31.6%)
No 724 (68.4%)
Previous cervical procedure
Cryotherapy 21 (2%)
Excisional procedure 27 (2.6%)
No prior treatment 999 (95.4%)
Tobacco use
None 704 (66.2%)
Current 294 (27.6%)
Past 66 (6.2%)
Unknown 15 (1.4%)
Median follow-up time, mo 15 (1-46)

STI , sexually transmitted infection.

Fader. Cervical dysplasia in pregnancy. Am J Obstet Gynecol 2010.


All patients were referred for antepartum colposcopy because of an abnormal cervical cytology. Referral cytology distribution is shown in Table 2 . Thirty percent of patients (n = 325) were diagnosed with an atypical squamous cells (ASC variant). Specifically, 6.8% had 2 consecutive atypical squamous cells of undetermined significance (ASCUS) cytology, 80.6% had an ASCUS cytology with positive high-risk human papilloma virus (HPV) test and 4.6% patients had an atypical squamous cells-Cannot Exclude High Grade (ASC-H) cytology result. Fifty-five percent of overall patients (n = 589) were diagnosed with low-grade squamous intraepithelial lesion (LSIL) and 15% (n = 164) were found to have high-grade squamous intraepithelial lesion (HSIL) on their respective cervical cytology tests. The distribution of colposcopic impression by referral cytology is detailed in Table 2 . The vast majority of patients with a diagnosis of either an ASC variant or LSIL (88%) had a colposcopic impression of a normal cervix or low-grade dysplasia (CIN 1). Only 1% had an impression of CIN 3. Of patients with a HSIL cytology, most (44%) had a colposcopic impression of CIN 2 and 31% were believed to have a normal cervix or CIN 1.


Tags:
Jul 6, 2017 | Posted by in GYNECOLOGY | Comments Off on Cervical dysplasia in pregnancy: a multi-institutional evaluation

Full access? Get Clinical Tree
Get Clinical Tree app for offline access