The true value of colposcopy in pregnancy is under debate; the examination may be more difficult depending on the gestation at which a woman presents. Cervical intraepithelial neoplasia does not have an accelerated progression during pregnancy, and treatment is usually deferred until postpartum. The prevalence of cervical intraepithelial neoplasia is greater in women with immune compromise. Those with human immunodeficiency have a higher prevalence, more persistence and less regression of human papillomavirus-related infections. Cervical cancer remains an AIDS-defining illness. Women who have had renal transplants also have a higher risk of developing cervical intraepithelial neoplasia. By contrast, other chronic illnesses that require immunosuppressant therapy do not seem to show this added risk. In young women, human papillomavirus infection is common and cervical intraepithelial neoplasia is also evident, but regression of these lesions is frequent and so conservative review may be appropriate. At the menopause, colposcopy is often unsatisfactory. The use of human papillomavirus testing for triage of low-grade cytological abnormalities may benefit this age group.
Colposcopy and pregnancy
The physiological changes of pregnancy (with the increased rate of squamous metaplasia, the increase in vascularity and the changes in the size and shape of the cervix) together conspire to make both cytological interpretation and colposcopic assessment a particular challenge. Benign lesions may appear to be suspicious of abnormality, but simply represent deciduosis, whereas active squamous metaplasia may be associated with a fine mosaic or punctuate surface pattern that may be indistinguishable from a low-grade cervical intraepithelial neoplasia (CIN) ( Fig. 1 ). An awareness of this is reflected in the National Health Service Cervical Screening Programme guidance for management: ( Table 1 ). In countries in which routine screening is not available, however, opportunistic screening, whether by cytology or visual inspection of the cervix, with acetic acid may be of value.
| If a woman has been called for routine screening and she is pregnant then the test should be deferred |
| The colposcopist may wish to perform only colposcopy at a follow-up appointment in pregnancy. |
| If repeat cytology is due and the woman has missed or defaulted her appointment prior to pregnancy consideration should be given to that woman having her cytology or a colposcopy during pregnancy. |
Although colposcopy is a safe and effective method of evaluating abnormal cytology, carrying it out in pregnancy may produce particular technical challenges. The vaginal walls are often lax, and there may be vulval and vaginal varicosities. On examination it is recommended that a large speculum be used. To keep the vaginal walls apart, a latex glove with the tip of the finger portion removed may be inserted in order to cover the speculum blades. It is then opened once inserted into the vagina. A condom instead of the finger of a glove may also be used.
The value of colposcopy in pregnancy
The aim of colposcopy during pregnancy is to exclude malignancy. Paraskevaidis et al. investigated the evolution of CIN and evaluated the safety of cytological and colposcopic surveillance of women with CIN during pregnancy. Ninety-eight women with antenatal cytological, colposcopic impression of CIN, or both, were followed up during pregnancy with cytology and colposcopy every 2 months. A cytological and colposcopic re-evaluation 2 months postpartum was carried out, and large loop excision of the transformation zone (LLETZ) (or loop electrosurgical excision procedure as it is known in North America), carried out if appropriate. Punch or loop biopsies were only taken in pregnancy if micro-invasion was suspected. In 14 out of 39 (35.9%) and in 25 out of 52 (48.1%) women with an antenatal impression of CIN I and CIN 2 and 3, respectively, a postnatal impression of regression was evident. Seven women with findings suspicious of micro-invasion underwent small loop biopsies during pregnancy, but early stromal invasion (< 1 mm) was seen in just one case. One more case of micro-invasion (1.5 mm of stromal invasion) was diagnosed postnatally in which the antenatal impression was of CIN 3. The investigators concluded that a considerable regression rate of CIN occurs after pregnancy, possibly attributable to the loss of the dysplastic cervical epithelium during cervical ‘ripening’ and vaginal delivery. However, without biopsy confirmation of the epithelial abnormality in pregnancy, and with the possibility of an over-call of cytological features in the pregnancy smear, the true regression rates may have been overestimated. In concordance with other earlier studies, the investigators concluded that frequent cytological and colposcopic evaluation seemed to be safe, and small loop biopsies are recommended for tissue diagnosis in suspected cases of possible micro-invasion.
Colposcopic biopsy in pregnancy
Several studies point to the relative safety of directed biopsies, although the value of such biopsies in pregnancy remains open to debate. Economos et al. looked at 612 pregnant women with abnormal cytology, retrospectively. Of their patients, 112 had surgical procedures that provided a surgical specimen to compare the cervical smear, colposcopic impression and directed biopsies. Cytology and colposcopic impression had a 91% concordance within one degree of severity. Colposcopic impression and directed biopsies had a 95% concordance within one degree of severity. No cases of invasive carcinoma were missed. Complications of biopsies were minimal. Three cases of delayed bleeding were resolved with the application of pressure. The investigators concluded that directed biopsies were safe in women referred with abnormal smears.
Other investigators have found similar results, but all stress the importance of the colposcopist being familiar with the physiological changes of pregnancy. A lack of experience can lead to over-estimation of the severity of the lesion and a mistaken diagnosis of invasive disease for pregnancy-associated changes.
In the UK, directed biopsies are not usually considered part of standard management. The rationale for UK policy is that, if there is serious concern of the presence of early or frankly invasive disease, a substantial biopsy will be required. Small directed punch biopsies may miss the invasive carcinoma or fail to give a representative sample of the most severe lesion. Multiple punch biopsies may also cause trauma and significant bleeding.
When malignancy is suspected, a possibly better option is to carry out a wedge biopsy using a diagnostic loop, needle diathermy or a knife, sampling the worst affected area of epithelium identified colposcopically. This is usually carried out in the operating theatre with ready access to haemostatic techniques and general anaesthesia if needed. The cervix is infiltrated with a solution of xylocaine and dilute adrenaline (1:80 000) to aid haemostasis. Even this technique, however, cannot completely exclude early invasive disease in the remaining cervical tissue. All this adds weight to the importance of colposcopy in pregnancy being carried out by expert colposcopists and consideration being given to some form of conisation.
Although one published study refers to the use of endocervical currettage in pregnancy with no adverse effect, the procedure is contraindicated in pregnancy because robust data on its safety are lacking. The concern is risk of premature rupture of membranes, preterm labour and haemorrhage.
The guidelines of the American Society for Colposcopy and Cervical Pathology (ASCCP) differ ( Table 2 ).
| Colposcopy is recommended for pregnant women with high-grade cervical squamous intraepithelial lesion; colposcopic evaluation preferably to be carried out by clinicians experienced in the evaluation of colposcpic changes induced by pregnancy. |
| Biopsy of lesions suspicious for CIN2, 3 or cancer is preferred; biopsy of other lesions is acceptable; |
| Endocervical curettage is unacceptable in pregnant women. |
| Diagnostic excision is unacceptable unless invasive cancer is suspected based on the referral cytology, colposcopic appearance, or cervical biopsy. Re-evaluation with cytology and colposcopy is recommended no sooner than 6 weeks postpartum for pregnant women with high-grade cervical squamous intraepithelial lesion in whom cervical intraepithelial neoplasia 2, 3 is not diagnosed. |
Wetta et al. recently questioned the value of carrying out colposcopy at all in pregnancy. In their study of 625 pregnant women, the most common referral cytology was low-grade squamous intraepithelial lesions (LSIL) (41%) followed by atypical squamous cells of undetermined significance (ASCUS) (34%). One hundred and thirty-eight women (22%) underwent a colposcopically directed biopsy at first visit. Of these, 43 (31.8%) had CIN 1, 28 (20.3%) had CIN 2 and 23 (16.7%) women had CIN 3. None of the women had invasive cancer. None of the women had progression of disease on repeat colposcopy. The evidence form several studies is that pregnant women with low-grade cytology rarely have biopsy-proven CIN 2 or 3, and are at low risk for the development of cervical cancer during pregnancy. The investigators point out that the ASCCP guidelines from 2006 state that colposcopic evaluation of ASCUS or LSIL cytology in pregnancy can be deferred to 6 weeks postpartum. Their study, however, has limitations, as the low incidence of cervical cancer in pregnancy makes it unlikely that a case would be found in a population of 625 women. Overall, only 78 women with ASCUS or LSIL underwent a biopsy and only 47 attended for repeat colposcopy. Their findings, however, coupled with the ASCCP guidance that diagnostic excision of high grade squamous intraepithelial lesions (HSIL) in pregnancy is unacceptable, add weight to the concept of potentially deferring follow-up colposcopy until the postpartum period.
Colposcopic biopsy in pregnancy
Several studies point to the relative safety of directed biopsies, although the value of such biopsies in pregnancy remains open to debate. Economos et al. looked at 612 pregnant women with abnormal cytology, retrospectively. Of their patients, 112 had surgical procedures that provided a surgical specimen to compare the cervical smear, colposcopic impression and directed biopsies. Cytology and colposcopic impression had a 91% concordance within one degree of severity. Colposcopic impression and directed biopsies had a 95% concordance within one degree of severity. No cases of invasive carcinoma were missed. Complications of biopsies were minimal. Three cases of delayed bleeding were resolved with the application of pressure. The investigators concluded that directed biopsies were safe in women referred with abnormal smears.
Other investigators have found similar results, but all stress the importance of the colposcopist being familiar with the physiological changes of pregnancy. A lack of experience can lead to over-estimation of the severity of the lesion and a mistaken diagnosis of invasive disease for pregnancy-associated changes.
In the UK, directed biopsies are not usually considered part of standard management. The rationale for UK policy is that, if there is serious concern of the presence of early or frankly invasive disease, a substantial biopsy will be required. Small directed punch biopsies may miss the invasive carcinoma or fail to give a representative sample of the most severe lesion. Multiple punch biopsies may also cause trauma and significant bleeding.
When malignancy is suspected, a possibly better option is to carry out a wedge biopsy using a diagnostic loop, needle diathermy or a knife, sampling the worst affected area of epithelium identified colposcopically. This is usually carried out in the operating theatre with ready access to haemostatic techniques and general anaesthesia if needed. The cervix is infiltrated with a solution of xylocaine and dilute adrenaline (1:80 000) to aid haemostasis. Even this technique, however, cannot completely exclude early invasive disease in the remaining cervical tissue. All this adds weight to the importance of colposcopy in pregnancy being carried out by expert colposcopists and consideration being given to some form of conisation.
Although one published study refers to the use of endocervical currettage in pregnancy with no adverse effect, the procedure is contraindicated in pregnancy because robust data on its safety are lacking. The concern is risk of premature rupture of membranes, preterm labour and haemorrhage.
The guidelines of the American Society for Colposcopy and Cervical Pathology (ASCCP) differ ( Table 2 ).
| Colposcopy is recommended for pregnant women with high-grade cervical squamous intraepithelial lesion; colposcopic evaluation preferably to be carried out by clinicians experienced in the evaluation of colposcpic changes induced by pregnancy. |
| Biopsy of lesions suspicious for CIN2, 3 or cancer is preferred; biopsy of other lesions is acceptable; |
| Endocervical curettage is unacceptable in pregnant women. |
| Diagnostic excision is unacceptable unless invasive cancer is suspected based on the referral cytology, colposcopic appearance, or cervical biopsy. Re-evaluation with cytology and colposcopy is recommended no sooner than 6 weeks postpartum for pregnant women with high-grade cervical squamous intraepithelial lesion in whom cervical intraepithelial neoplasia 2, 3 is not diagnosed. |
Wetta et al. recently questioned the value of carrying out colposcopy at all in pregnancy. In their study of 625 pregnant women, the most common referral cytology was low-grade squamous intraepithelial lesions (LSIL) (41%) followed by atypical squamous cells of undetermined significance (ASCUS) (34%). One hundred and thirty-eight women (22%) underwent a colposcopically directed biopsy at first visit. Of these, 43 (31.8%) had CIN 1, 28 (20.3%) had CIN 2 and 23 (16.7%) women had CIN 3. None of the women had invasive cancer. None of the women had progression of disease on repeat colposcopy. The evidence form several studies is that pregnant women with low-grade cytology rarely have biopsy-proven CIN 2 or 3, and are at low risk for the development of cervical cancer during pregnancy. The investigators point out that the ASCCP guidelines from 2006 state that colposcopic evaluation of ASCUS or LSIL cytology in pregnancy can be deferred to 6 weeks postpartum. Their study, however, has limitations, as the low incidence of cervical cancer in pregnancy makes it unlikely that a case would be found in a population of 625 women. Overall, only 78 women with ASCUS or LSIL underwent a biopsy and only 47 attended for repeat colposcopy. Their findings, however, coupled with the ASCCP guidance that diagnostic excision of high grade squamous intraepithelial lesions (HSIL) in pregnancy is unacceptable, add weight to the concept of potentially deferring follow-up colposcopy until the postpartum period.
Excisional diagnosis and treatment in pregnancy
Knife cone biopsy
If cervical conisation is considered during pregnancy, it is for the exclusion of invasive cancer. In the past, cold-knife cone biopsy was the only available excisional treatment modality. It was shown to be associated with high complication rates and high rates of residual disease, even when carried out in the first trimester. Hannigan et al. reported a high complication rate for cone biopsy undertaken during pregnancy in a large study of 82 pregnant women.
Excess haemorrhage was reported in 8.9% of cases (range 5.2–13.9%). The reported spontaneous miscarriage rate was up to 33%. Some investigators advocate deferring treatment to the second trimester to reduce this rate. To decrease the amount of bleeding, cervical cerclage at the time of conisation has also been suggested. Some practitioners continue to use a standard cold-knife cone biopsy technique, but without the traditional two laterally placed haemostatic sutures, instead using local intra-cervical infiltration with xylocaine and adrenaline before the knife cone and diathermy to the base afterwards and on occasion a vaginal pack. Cold-knife cone biopsy offers a high diagnostic accuracy but, as previously mentioned, the potential for complications for mother and fetus is significant.
Laser cone biopsy
Though less commonly available, carbon dioxide (CO 2 ) laser cone biopsy allows a more precise excision of the transformation zone, with less anatomical distortion to the cone base and the cervix itself. It is, therefore, less likely to be associated with medium- and long-term complications. In expert hands, it should be as effective as a cold knife cone biopsy for effecting simultaneous diagnosis and treatment of suspected micro-invasion. The technique, however, does seem to be associated with increased intra-operative blood loss, particularly with high-grade lesions, and this may be compounded by the increased vascularity of pregnancy. For this reason, some may prefer to use an electrical method in pregnancy.
The most recent publication on excision in pregnancy evaluated the safety and efficacy of CO 2 laser conisation in pregnant women with CIN 3 and carcinoma in situ (CIS) or micro-invasive carcinoma (MIC). A total of 49 pregnant women with biopsy-proven CIN3 and CIS (30 women) or MIC (19 women) were studied retrospectively. The median age and median week of gestation were 31 years (range: 22–39 years) and 17 weeks (range: 7–33 weeks), respectively. The median length of cervix resected by conisation, median duration of surgery and median blood loss were 14 mm (range: 5–23 mm ), 20 min (range: 7–35 min) and 78 ml (range: 10–797 ml), respectively. One biopsy-proven CIN3 and CIS patient was diagnosed with Federation of Gynecology and Obstetrics (FIGO) Ia2, and three biopsy-proven women with MIC were diagnosed with FIGO Ib1 based on conisation specimens. A total of 35 women could be followed until delivery, of which 27 (77.1%) women delivered transvaginally. Although eight women (22.9%) had a caesarean section and six women (17.1%) delivered preterm, no conisation-related obstetric complications were observed. The investigators felt that CO 2 laser conisation within 20 mm of length can be considered a safe procedure for pregnant women.
Large loop excision of the transformation zone and loop electrosurgical excision procedure
Only three published studies have evaluated LLETZ and LEEP during the index pregnancy. Robinson et al. reported on the outcome of 20 women who underwent LLETZ between 8 and 34 weeks gestation. All but two of their procedures were carried out under general or regional anaesthesia, and cervical sutures were placed in the cervix at 12, 3, 6 and 9 o’clock. The loop excisions were carried out excising the anterior lip as one specimen and the posterior lip as a second specimen. Ball electrodes were used for cauterisation to achieve haemostasis and a surgicel-type membrane left in the loop bed, following which the 3 and 9 o’clock sutures were tied together in the midline. They showed a significant complication rate of 25%. All five women with a significant morbidity had their procedure carried out between 27 and 34 weeks. Only 14 out of 20 (70%) of the specimens contained CIN and eight out of 15 (57%) had involved margins. Residual disease was seen in nine women, including three women whose initial loop specimen did not contain CIN.
In a small study of nine women who underwent a LLETZ procedure under local anaesthesia within the first 14 weeks of pregnancy, histological assessment showed CIN1 in two, CIN3 in five and MIC in two women. Margins were incompletely excised in two women who went on to have a repeat procedure postpartum. None of the women had spontaneous miscarriage, preterm labour or excessive haemorrhage. One woman developed cervical incompetence at 21 weeks gestation and required cervical cerclage. All pregnancies continued to term. Six women delivered vaginally. The reasons for three caesarean sections were previous section and breech presentation. In the retrospective study of Paraskevaidis et al., seven women with findings suspicious of micro-invasion underwent small loop biopsies in pregnancy. Histology showed decidual reaction in five cases, CIN3 in one and early stromal invasion (< 1 mm) in one more at 26 weeks gestation. This woman delivered at 37 weeks gestation after induction of labour. An LLETZ was carried out 6 weeks postpartum. Histology confirmed stage 1aI disease with clear margins.
The investigators did not advocate LLETZ in pregnancy because of the high risks of bleeding and possibility of incomplete excision. Instead, they recommended using the smallest loop wire to obtain a sufficient sample to examine for possible micro-invasion instead of a conventional punch biopsy.
The use of the diathermy loop in some of these studies has been described as a LLETZ. Strictly speaking, however, this is not the same procedure as a diathermy loop excision, which would be more equivalent to a cold-knife cone biopsy. In light of the recent new classification of transformation zones, further research that comments on the type of transformation zone treated and the depth of specimen may make more meaningful comparisons of techniques possible.
Glandular lesions in pregnancy
Few data are available on the significance of glandular atypia during pregnancy and in the post-partum period. Chhieng et al. reported on the outcome for 35 women diagnosed with an atypical glandular cells of uncertain origin smear in pregnancy between 1995 and 1997. This represented 0.26% of all smears carried out on pregnant women. The mean age was 29 years, with a range of 17–45 years. Eleven had a concomitant minor squamous lesion (one LSIL, and 10 ASCUS). Seventeen women underwent colposcopy and directed punch biopsy; repeat cytology was carried out in 10 women. Eight women were lost to follow up. Of those who had a biopsy, two had evidence of a high-grade squamous intra-epithelial lesion. In this small study of pregnant women referred with an atypical glandular cells of uncertain origin smear, none had evidence of a significant glandular lesion.
Adenocarcinoma-in-situ and primary adenocarcinoma of the cervix are uncommon, accounting for up to 15% of all cervical neoplasms. Few cases have been reported occurring in pregnancy. Even in the non-pregnant state, colposcopic evaluation of glandular abnormalities is challenging and often unreliable. Usually, there is a concomitant squamous lesion. In view of the high rate of pathology in women with glandular smear abnormalities, and the difficulty of colposcopic assessment, formal conisation might need to be considered for the management of these women in pregnancy, but each case should be discussed as an individual with multidisciplinary team input.
Until there is a better understanding of the natural history and implications of glandular atypia in pregnancy, close follow-up with colposcopy during pregnancy and post-partum with directed biopsy of suspicious areas remains important.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
