Colposcopy

INTRODUCTION — Colposcopy is a diagnostic procedure in which a colposcope (a dissecting microscope with various magnification lenses) is used to provide an illuminated, magnified view of the cervix, vagina, vulva, or anus (picture 1) [1]. The primary goal of colposcopy is to identify precancerous and cancerous lesions so that they may be treated early.

Colposcopy of the cervix is the main focus of this topic and is used as further evaluation of abnormal cervical screening tests (cytology and/or human papillomavirus testing). Colposcopy of the vagina and vulva is also reviewed here.

Colposcopic evaluation is based on the finding that malignant and premalignant epithelium have specific visual characteristics in terms of contour, color, and vascular pattern that are recognizable using colposcopy. The improved visualization of epithelial surfaces with colposcopy compared with gross visual examination enhances the colposcopist's ability to distinguish normal from abnormal areas and to obtain directed biopsies.

Related topics are discussed in detail separately, including:

●Cervical cancer screening (see "Screening for cervical cancer in resource-rich settings")

●Follow-up of abnormal cervical screening tests (see "Cervical cancer screening: Risk assessment, evaluation, and management after screening" and "Cervical cytology: Evaluation of atypical and malignant glandular cells")

●Management of cervical intraepithelial neoplasia (CIN) (see "Cervical intraepithelial neoplasia: Management")

●Anal neoplasia (see "Anal squamous intraepithelial lesions: Epidemiology, clinical presentation, diagnosis, screening, prevention, and treatment")

INDICATIONS — Colposcopy is used as a follow-up test to evaluate abnormal cervical cancer screening tests (cytology and/or human papillomavirus testing [HPV]) or abnormal findings on gross examination of the cervix, vagina, or vulva. It has not been found to be an effective screening tool for cervical cancer when used alone [2].

The indications for cervical colposcopy based on abnormal results of cervical cytology or HPV testing are discussed in detail separately. (See "Cervical cancer screening: The cytology and human papillomavirus report" and "Cervical cancer screening: Risk assessment, evaluation, and management after screening" and "Cervical cytology: Evaluation of atypical and malignant glandular cells".)

Common indications for colposcopy, other than abnormal cervical cytology or HPV-positive testing, include [3]:

●Evaluation of a palpably or visually abnormal cervix, vagina, or vulva.

●Evaluation after an abnormal test for cervical neoplasia other than cervical cytology or HPV testing. These methods are most commonly used in low-resource settings and include visual inspection with acetic acid or Lugol iodine, cervicography, or speculoscopy. (See "Cervical cancer screening tests: Visual inspection methods".)

●In conjunction with treatment of cervical neoplasia to ensure that known lesions are completely removed or treated, to detect any other lesions in surrounding areas, and for posttreatment surveillance. (See "Cervical intraepithelial neoplasia: Management".)

CONTRAINDICATIONS — There are few contraindications to colposcopy. (See 'Complications' below.)

●Cervicitis – It is advisable to treat active cervicitis (eg, Trichomonas vaginalis), if present, before colposcopy as inflammation and infection can impede accurate assessment of epithelial abnormalities. While other vaginal infections may not affect visualization with colposcopy, treatment may allow the patient to be more comfortable during the examination. (See "Acute cervicitis" and "Vaginitis in adults: Initial evaluation".)

●Anticoagulation or bleeding diatheses – Anticoagulation or bleeding diatheses are not absolute contraindications to colposcopy. Bleeding is usually minimal, even in patients with these issues. However, a full history should be obtained in patients that are at high risk for bleeding. Medications can sometimes be held prior to colposcopy; for example, in our practice, we tell some patients to skip their morning dose of apixaban on the day of their colposcopy/biopsy. This is done only after discussion with their prescribing provider.

●Pregnancy – The physiologic changes to the cervix during pregnancy (eg, hyperemia) make colposcopy and identifying cancerous lesions more difficult. In addition, the morbidity associated with cervical conization or biopsy during pregnancy is substantial. Biopsy is only performed if invasive disease is suspected. This is discussed in more detail elsewhere. (See "Cervical intraepithelial neoplasia: Management", section on 'Pregnant patients'.)

●Immunosuppression – Patients who are severely immunosuppressed (absolute neutrophil count <500 cells/microL) are at risk for bacterial translocation and potential bacteremia from a speculum examination. Colposcopy should be avoided in these patients in most circumstances.

RELEVANT ANATOMY AND HISTOLOGY — The uterine cervix is a tubular fibromuscular structure that serves as the conduit between the endometrial cavity and the vagina. The superior portion is continuous with the uterus. The cervical canal opens into the endometrial cavity at the internal os and into the vagina at the external os.

The inferior portion of the cervix protrudes into the vagina. In some patients (eg, postmenopausal, following pelvic radiation), the cervix may appear flush with the vagina on examination rather than protruding.

The ectocervix is the surface of the cervix that protrudes into the vagina (figure 1). It is covered with squamous epithelium. The endocervix is the cervical canal, which is lined with columnar (glandular) epithelium.

The squamocolumnar junction (SCJ) of the cervix (junction of squamous and glandular cells, generally at the external cervical os) and the transformation zone (the transformation zone is the area between the original SCJ and the current one) are the areas at greatest risk for neoplasia (figure 2 and figure 3 and picture 2) [4]. The transformation zone is an area of squamous metaplasia that lies between the SCJ and the remainder of the squamous epithelium [5]. The transformation zone contains embryonic cells that may be especially vulnerable to infection with human papillomavirus and to oncogenic transformation [4].

A comparison of normal, low-grade, and high-grade squamous lesions by cervical cytology and histology is shown in the figure (figure 4).

The vagina is tubular and the proximal portion is the widest. The areas of the vagina that surround the protruding cervix are called the fornices. The vagina is covered with squamous epithelium.

The vulva comprises the external female genitalia, including the labia majora, labia minora, clitoris, vulvar vestibule, external urethral meatus, and vaginal orifice. These structures are covered with squamous epithelium.

Female pelvic anatomy is discussed in detail separately. (See "Surgical female pelvic anatomy: Uterus and related structures" and "Surgical female urogenital anatomy".)

INSTRUMENTATION AND EQUIPMENT

Colposcope — The colposcope is a lighted binocular microscope that magnifies the tissue of interest (eg, cervical, vaginal, or vulvar epithelium), thereby helping to identify features suggestive of abnormal tissue (picture 1). Specifications and set-up of the colposcope include:

●At the start of the procedure, the clinician should adjust the intraocular distance between the two eyepieces on the colposcope to ensure binocular vision.

●The colposcope should be placed approximately 30 cm from the tissue that is being visualized. Most colposcopes have a focal length (ie, working distance between the lens and target tissue) of 30 cm. This focal length allows the colposcopist the appropriate amount of room to comfortably reach the cervix with instruments.

●Colposcopes differ by manufacturer, and magnification can range from 3.5x to 30x. On conventional colposcopes, the head of the colposcope is moved closer and farther from the patient, and this changes the focus, and sometimes the magnification. One type of video colposcope automatically adjusts the magnification, depending on how close the scope is to the patient, and this is based on the focus. Fine focusing can be done by turning a knob; coarse focusing is performed by moving the instrument toward or away from the patient.

●Low power (2x to 10x) is often used to obtain an overall impression of surface architecture and for examination of the vulva. Medium (10x to 20x) and high (20x to 25x) powers are utilized to evaluate the vagina and cervix, with high power being particularly useful for close inspection of vascular patterns, which can signify high-grade or invasive disease. A consistent plan for magnification (usually starting at 15x) is important so that examinations are comparable and reproducible. In general we use a magnification of 15x for the cervix and vagina and 3.5 to 7.5x for evaluation of the vulva, as the organ is closer and thus requires less magnification.

●The examination should begin with white light at a low-power setting for a global view of the cervix. A green filter switch is present to toggle between the two light settings. Green filter light may help identify vascular changes.

"Smart" colposcopes are now available and allow the clinician to capture and annotate images and videos and incorporate these into the patient's electronic medical record; one such device is the Enhanced Visual Assessment System (EVA) colposcope, which is cleared by the US Food and Drug Administration [6].

Other supplies and instruments — Other supplies and instruments include biopsy instruments and solutions needed for the examination (eg, acetic acid) or to control bleeding (eg, Monsel solution) (table 1). Prior to starting the examination, all necessary equipment and supplies should be readily available to the colposcopist, and an assistant should be present throughout the procedure.

PREPROCEDURE PREPARATION

Review of history and records — A focused medical history and relevant medical records should be obtained. The biggest risk factor for cervical neoplasia is a history of cervical neoplasia, especially recurrent or persistent abnormal results over a period of years.

The lower genital tract history and review of records should include:

●Prior cervical cytology (Pap test) and human papillomavirus (HPV) tests, normal and abnormal

●Prior cervical, vulvar, and vaginal biopsy results

●History of lower genital tract cancers or precancers

●History of condyloma (genital warts)

●Treatments to the cervix, vagina, and vulva

If possible, documentation of prior abnormal results should be obtained to confirm the history. For most patients, the written reports of previous cytology and histology are sufficient. In some cases, in our practice, we request the slides and have them reviewed by our pathology laboratory. In general, we do this when the patient is seeing us for a second opinion or when there is a discrepancy between the cytology and biopsy result (especially when the cytology results is a high-grade squamous epithelial lesion or adenocarcinoma in situ and we are unable to find a lesion on colposcopic examination that explains this abnormality).

In addition, a relevant medical history should be obtained, including:

●Obstetric and gynecologic history – Last menstrual period, history of sexually transmitted infections, birth control method (if appropriate), history of cervical or pelvic surgery, and HPV vaccination status, with dates of injections if available. Vaccination status does not appear to affect colposcopic evaluation. In one retrospective study including 160 patients in Sweden presenting for colposcopy, colposcopic impression was similar for vaccinated (56 percent) compared with unvaccinated patients; the prevalence of HPV 16/18 infection was lower in the vaccinated subgroup (8 versus 33 percent) [7].

●Immunosuppression – Immunosuppression increases the risk of progression of cervical neoplasia (eg, human immunodeficiency virus infection, autoimmune disease, history of transplant or cancer, or immunosuppressant medications).

●Smoking history – The risk of cervical neoplasia is increased in cigarette smokers.

●Factors that impact the safety of the procedure – Anticoagulation, bleeding disorders, allergy to iodine.

Informed consent and counseling — The procedure should be explained to the patient. Questions should be answered about the procedure, possible results and prognosis, and follow-up. The discussion is documented in the medical record and a consent form is signed.

Procedural anxiety — Anxiety, pain, and compliance with follow-up after colposcopy appear to be improved by several measures, including showing an informational video before colposcopy, playing music during colposcopy, viewing video colposcopy during the procedure, or providing a visual distraction such as a picture on the ceiling [8-11]. Although printed information leaflets did not reduce anxiety levels, they did increase knowledge levels.

CERVICAL COLPOSCOPY — The colposcope is used to examine the entire surface for the cervix, but the most emphasis is put on examining the squamocolumnar junction (SCJ) and transformation zone. (See 'Relevant anatomy and histology' above.)

Recommendations have been made by the American Society for Colposcopy and Cervical Pathology (ASCCP) Standards Committee to address shortcomings in colposcopic practice. These include [12]:

●Terminology in colposcopic practice – To standardize terminology for colposcopic practice and reporting of findings.

●Risk-based colposcopy practice approach – To adapt colposcopic practice based on previous level of risk or probability of finding precancer/cancer.

●Colposcopy procedures and adjuncts approach – Drafted components for comprehensive and minimum practice for colposcopy examination based on expert opinion.

The cervical colposcopy procedure includes:

●Repeat cervical cytology (with human papillomavirus [HPV] testing), if indicated. (See 'Repeat cytology and HPV testing' below.)

●Examination of vulva, vagina, and cervix under gross visualization.

●Colposcopic examination of cervix and upper one-third of the vagina.

●Biopsy and/or endocervical sampling, as indicated.

●Documentation of findings.

Abnormal colposcopic findings are used to choose the sites to biopsy. However, colposcopic findings themselves are not diagnostic of cervical neoplasia; this can only be established definitively by pathologic examination of the biopsied tissue.

Colposcopy by an experienced colposcopist will usually detect sites of significant disease (cervical intraepithelial neoplasia [CIN] 2,3; cancer), but some patients with a normal colposcopic examination can still have a significant histologic abnormality. For this reason, it is important to reconcile the cytologic, colposcopic, and histologic findings in formulating a diagnosis and management plan for each patient.

To begin the procedure, the patient is placed in the dorsal lithotomy position. The vulva is examined under gross visualization for any suspicious lesions or findings that require biopsy or colposcopic examination. A speculum is placed in the vagina; it is best to use the largest speculum that the patient can easily tolerate so that the entire cervix and vaginal fornices may be visualized.

Colposcopy is typically well-tolerated, and anesthesia or analgesia are not usually required. A review of 19 randomized trials found no difference in pain scores in patients given an oral analgesic compared with placebo [13].

Procedure

Repeat cytology and HPV testing — Cervical cytology is repeated (with HPV testing), if indicated (table 2). In our practice, we repeat cytology if the interval between colposcopy and the original Pap smear is greater than six weeks. In general, we do not repeat HPV more frequently than annually. The Pap test is repeated because abnormal cervical cytology findings (low-grade squamous intraepithelial lesion [LSIL] and high-grade squamous intraepithelial lesion [HSIL]) may resolve or progress over time. A current cytologic result enables the colposcopist to make the most accurate correlation of cytologic, colposcopic, and histologic findings.

Collection of a cervical specimen is presented separately. The authors do not use devices marketed for single use removal of the squamous epithelium (commercial names SpiraBrush and SoftBiopsy) as they have yet to be adequately studied with respect to safety and efficacy. (See "Cervical cancer screening tests: Techniques for cervical cytology and human papillomavirus testing", section on 'How to obtain a sample'.)

Traditionally, it was advised that the interval between cervical cytology tests be at least six weeks, but it appears that shorter intervals do not compromise the adequacy of the specimen. (See "Cervical cancer screening tests: Techniques for cervical cytology and human papillomavirus testing", section on 'Interval between Pap tests'.)

Gross visualization — The cervix and vagina are first examined under gross visualization with a bright light, without application of solutions. If the view of the cervix is occluded (mucus, blood, discharge, or debris), cotton soaked in saline may be used to cleanse it.

The clinician should look for and define the following features/findings:

●Erosion

●Ulceration

●Areas in which the ectocervix has an irregular surface (the surface of the ectocervix is normally smooth) – This may represent mitotically active tissue, such as a neoplasm or condylomata

●Leukoplakia

●Pigmented lesions

●Exophytic growth

Ulcerations or erosions may be due to cervical neoplasia, but may also be due to trauma or infection. Gross inspection can often identify abnormalities. Any abnormality that is detected should be biopsied.

Colposcopic examination

Technique — The colposcope is used to examine the entire surface of the cervix, but the most emphasis is put on examining the SCJ and transformation zone. The SCJ (junction of squamous and glandular cells, generally at the external cervical os) and the transformation zone (the transformation zone is the area between the original SCJ and the current one) are the areas at greatest risk for neoplasia (figure 2 and figure 3 and picture 2) [4]. (See 'Relevant anatomy and histology' above.)

Visualization is aided by:

●Application of acetic acid – The cervix is examined first without, and then with, a solution of 3 to 5 percent acetic acid. This allows improved colposcopic visualization of abnormal areas. The acetic acid is applied generously to the cervix using a cotton swab [14]. After 30 to 60 seconds, the acidic solution dehydrates cells so that squamous cells with relatively large or dense nuclei (eg, metaplastic cells, dysplastic cells, cells infected with HPV) reflect light and thus appear white [15]. This is referred to as "acetowhite change." Blood vessels and columnar cells are not affected but become easier to visualize against the white background. Fading of acetowhite changes usually occurs after three minutes [16], and therefore, acetic acid should be reapplied, as needed, after this time. If excessive acetic acid pools in the vagina, it should be removed with dry swabs as it can cause irritation.

●Differential light absorption by squamous versus glandular cells – A tissue characteristic that helps to differentiate between cell types is that squamous and glandular cells have a differential absorption of light. The squamous cells of the ectocervix have a smooth grey-pink appearance, and the glandular cells of the endocervix have a pink-red cobblestone appearance (figure 1).

●Green or blue filter – Use of the green or blue filter on the colposcope can accentuate abnormal vasculature. Visualized through the green or blue filter, the blood vessels appear darker, and this sharpens the contrast between vessels and the surrounding epithelium.

●If no lesions are seen after acetic acid is applied, a dilute Lugol or Schiller solution may be applied to the cervix and vagina to aid in detection. Lugol iodine consists of 5 g of iodine and 10 g of potassium iodide in 100 mL distilled water [17]. Uniform uptake of stain would confirm the colposcopist's impression that no lesion is present. Glycogen-containing cells will take up iodine and become dark brown. Nonglycogenated cells, such as normal columnar or glandular cells, high-grade lesions, and many low-grade lesions, will not take up iodine and remain light yellow. Thus, they can be easily differentiated from normal tissue for sampling or treatment purposes. Iodine staining should not reveal any lesions the examiner has not previously identified with saline or acetic acid.

After the cervix is examined, the upper one-third of the vagina, in particular the lateral fornices, is also examined. The transformation zone and SCJ may extend into the upper vagina, particularly in younger patients. Abnormal vaginal findings include adenosis, polyps, cysts, diethylstilbestrol morphology, condyloma, and changes suggestive of preinvasive or invasive disease. (See 'Vaginal colposcopy' below.)

Approach to a difficult colposcopy — In some patients, it may be difficult to visualize the entire cervix or the SCJ. Guidance to address these challenges includes:

●The SCJ is a circumferential region around the external cervical os. However, the distance between the os and the edges of the junction varies. In some patients, the distance between the os and the edges of the junction is narrow, or the SCJ is not visible because it recedes within the endocervical canal. The location and size of the junction are altered by several factors:

•Hormonal factors – Conditions or medications that result in high serum estrogen levels (eg, pregnancy, oral contraceptives) cause the SCJ to be more everted. If a great degree of eversion is present, the columnar epithelium is exposed to the vaginal milieu; this is referred to as ectropion. Ectropion is common during pregnancy or in adolescents. In contrast, menopause may cause the squamocolumnar to recede into the cervical canal. (See "Benign cervical lesions and congenital anomalies of the cervix", section on 'Ectropion'.)

•Prior cervical treatments – Cervical conization with removal of transformation zone may make it difficult to visualize the current SCJ.

If it is difficult to visualize the SCJ or the upper margin of a lesion, an endocervical speculum may be used or a wet cotton swab can be placed on the ectocervix to pull the tissue aside and open the os (picture 3). The cotton swab is often less traumatic than the endocervical speculum.

●If the patient has a hypertrophied cervix or markedly redundant vagina, it is sometimes necessary to use a large cotton swab to move the cervix or attach a single tooth tenaculum to the cervix to allow manipulation and visualization of the entire cervix and vagina.

Findings — The locations of abnormalities are noted as on the face of the clock (eg, at 2:00 or 10:00).

The type and characteristics of an abnormality may correspond with the grade of the abnormality. Thus, the colposcopist forms an impression of the severity of disease based on visualization. However, a colposcopic impression is not diagnostic, and biopsies must be performed to obtain histologic results. No colposcopic findings are pathognomonic for cervical cancer. ASCCP Colposcopy Standards include both comprehensive and minimum recommendations for documenting the colposcopic evaluation (table 3) [12].

Abnormalities that should be noted during cervical colposcopy include:

●Acetowhite epithelium – Acetowhite lesions with sharp borders are more likely to be high-grade lesions, while diffuse borders suggest low-grade disease.

●Abnormal vascular patterns; these include:

•Mosaicism – Punctate (figure 5) and mosaic (figure 6) vessels (picture 3 and picture 4) in a field of acetowhite epithelium in the transformation zone are suggestive of CIN, low-grade (picture 5A-B) or high-grade (picture 6A-C).

•Punctation – Findings consistent with punctation should be differentiated with a stippled appearance beyond the transformation zone that may be found in patients with vaginitis (picture 7).

•Atypical vessels – Atypical vessels that have increased caliber (coarseness); display irregular and abrupt changes in direction; are widely spaced; suddenly terminate; or have corkscrew, comma, or hairpin patterns are suggestive of microinvasive or invasive disease (picture 8).

It should be noted that inflammatory changes on the cervix can appear similar to high-grade lesions, with diffuse mosaicism and cervical friability. If the cervix appears inflamed or if there are prior biopsies suggestive of acute inflammation, reactive changes, or follicular cervicitis, in our practice, we perform testing for gonorrhea and chlamydia and treat the patient empirically for cervicitis with a single dose of azithromycin (oral, 1 g) prior to performing the colposcopy. If testing for gonorrhea is positive, further treatment is required. (See "Acute cervicitis" and "Treatment of Chlamydia trachomatis infection" and "Treatment of uncomplicated gonorrhea (Neisseria gonorrhoeae infection) in adults and adolescents".)

The 2011 International Federation for Cervical Pathology and Colposcopy (IFCPC) colposcopic terminology of the cervix includes the components of the examination and the terminology to describe abnormal findings (table 4). This classification includes the following findings that are likely to correlate with abnormal histology [18]:

●Features suggestive of low-grade disease – Thin acetowhite epithelium; irregular, geographic border; fine mosaic; fine punctation (picture 5A-B).

●Features suggestive of high-grade disease – Dense acetowhite epithelium, rapid appearance of acetowhitening, cuffed crypt (gland openings), coarse mosaic, coarse punctation, sharp border, inner border sign, ridge sign (picture 6A-C).

●Features suggestive of invasive cancer – typical vessels, fragile vessels, irregular surface, exophytic lesion, necrosis, ulceration (necrotic), tumor/gross neoplasm (picture 8).

●Nonspecific – Leukoplakia (keratosis, hyperkeratosis), erosion.

Biopsies — Cervical biopsies are obtained using long biopsy instruments that are able to reach the cervix. The Kevorkian (picture 9) or Tischler (picture 10) biopsy instruments are preferred by most providers. The amount of tissue removed is approximately 1 to 2 mm. These biopsies are sometimes referred to as punch biopsies.

Local anesthesia is not routinely used for biopsies of the cervix and upper vagina, since injection of the anesthetic is probably as painful as the biopsy. Injection of an anesthetic may also disrupt the epithelium, making visualization of the lesion more difficult. Distraction techniques (eg, verbal distraction, visual distraction [11], music, asking the patient to cough during the biopsy) are as effective as local anesthetic injection, require less time, and have no risk of intravascular injection of anesthetic agents [19]. Topical and oral analgesics are ineffective [20].

Each specimen is individually labeled according to its location on the cervix as if taken from the face of a clock (eg, biopsy at 2:00 or 10:00) and each biopsy specimen is placed in a separate, labeled container containing a permanent fixative.

The ASCCP recommends targeted biopsies of all acetowhite areas and obtaining at least two and up to four biopsies [21]. In patients who are low risk for cervical cancer (LSIL or less on cervical cytology and HPV 16/18 negative) and have no visible lesion on colposcopy, untargeted (randomized) biopsies are not recommended.

In observational data from a randomized trial of the evaluation of abnormal cervical cytology, the sensitivity of two (82 percent) or three or more (83 percent) biopsies was superior to one biopsy (68 percent) [22]. In addition, a prospective study demonstrated that biopsy of multiple versus single sites was not associated with an increased likelihood of acquiring new HPV infections [23].

In our practice, we generally biopsy two lesions unless there is a significant suspicion of cancer or many areas of abnormality that appear distinct from each other. Biopsies may be taken from any part of the most abnormal appearing area or from more than one abnormal appearing area. If there is no visible lesion, we do not generally obtain random biopsies, but we do obtain an endocervical sample, even if the transformation zone is well visualized. A thorough vaginal and vulvar colposcopy is needed, especially in patients with a high-grade lesion on cervical cytology or HPV test demonstrating HPV 16 or 18. (See 'Effectiveness' below and 'Endocervical sampling' below.)

The order of taking the biopsies should be from posterior to anterior; this is the best order to avoid bleeding that blocks visualization of the other biopsy sites. Ferric subsulfate (Monsel solution) is applied with either small or large cotton swabs to the biopsied areas to control bleeding. In case of significant bleeding, having surgical and/or vaginal packs available is advisable.

For patients at an increased risk of bleeding, care should be taken to obtain very small biopsies and limit the number of biopsies obtained. Patients with risk factors for complications should be referred to an experienced clinician for colposcopy, if available. Factors that increase the risk of bleeding from cervical biopsies include:

●Anticoagulants or bleeding disorders. Consider having Monsel solution close by with larger Scopettes (Q-tips).

●Pregnant patients.

●Biopsy of an invasive lesion can lead to severe bleeding. If a lesion that is suspicious for cancer is found, the patient should be referred to an experienced clinician and the biopsy performed in a setting that is well equipped to deal with bleeding if it occurs, including access to vaginal packing, transfusion, an operating room, and radiation therapy.

Endocervical sampling

●Patient selection – Endocervical sampling is performed according to the 2019 ASCCP management guidelines [21]. In our practice, we perform endocervical sampling at colposcopy in all nonpregnant patients at the first visit. Approximately 5 to 15 percent of patients with high-grade CIN are diagnosed solely based on the endocervical specimen [22,24-27]. Some data suggest that endocervical sampling increases the sensitivity of the examination, particularly in older patients [24,25]. In addition, endocervical sampling may sample cells from "skip lesions" (noncontiguous lesions) that are typical of glandular neoplasia [28].

Some clinicians do not perform endocervical sampling in all patients. If selective use of endocervical sampling is performed, the indications are:

•Atypical squamous cells cannot exclude HSIL (ASC-H).

•HSIL.

•Atypical glandular cells; adenocarcinoma in situ.

•Atypical squamous cells of undetermined significance (ASC-US) or LSIL plus any of the following:

-No lesion is visualized during the colposcopic examination.

-Colposcopy is inadequate.

-Ablative treatment is contemplated.

•Transformation zone is not clearly visible.

•Ablative therapy is being considered, and when a lesion has not been found (regardless of preceding Pap).

Endocervical sampling is not performed in pregnant patients. (See 'Pregnant patients' below.)

●Procedure – Either a curette or endocervical brush may be used to perform sampling. In a meta-analysis including randomized trials and observational studies, use of a curette or brush had similar diagnostic accuracy, inadequate sampling rate, and patient discomfort, although the quality of this evidence was low [29].

The instrument is inserted into the endocervical canal and the four quadrants of the canal are sampled. If a curette is used, an endocervical brush is then inserted and rotated to remove any exfoliated tissue. These specimens should be collected and labeled as a specimen separate from other specimens obtained (ie, biopsies). A drop of mucus and blood is often seen at the os, and this should be included in the specimen.

Endocervical samples are intended to be from the endocervical canal but may be contaminated by cells from the ectocervix (figure 1). This is particularly likely in patients in whom the transformation zone has receded into the cervical canal (eg, postmenopausal patients, patients with a prior cone biopsy or ablative therapy). The specimen should be sent even if contamination is suspected. However, if neoplasia is found, it may be unclear whether the neoplasia is present on the ectocervix or endocervix. Thus, the clinician must use their judgment and their level of suspicion that the sample was contaminated. In such cases, positive results may not have the same implications; for example, the clinician might be more willing to follow rather than treat a patient.

An endocervical sample that does not show endocervical tissue is an inadequate result and cannot be construed as a negative result. In such cases, if the result may impact management, repeat sampling should be performed.

Documentation — Colposcopic examinations should be documented in a consistent and reproducible way [30]. Standardized ASCCP colposcopic practice includes a general assessment, evaluation for acetowhite changes, and description of colposcopic findings (table 5) [30].

Documentation should include:

●Visualization of the SCJ (figure 3 and picture 2) (see 'Relevant anatomy and histology' above). Whether the SCJ was fully or not fully visualized has been referred to previously as "adequate" or "inadequate" examination. The ASCCP no longer uses the term "adequate" and rather requires documentation that the SCJ was or was not fully visualized.

●Size and location of abnormalities. The margins of any visible lesions must be fully identified and the histologic results from biopsies of the lesions should explain the abnormal cytology (ie, if the cytology was high grade and all biopsies are normal) (table 5).

●Whether the vagina or vulva was evaluated. This should be clearly noted. For all patients undergoing cervical colposcopy, at minimum, evaluation of the upper fornices should be performed.

The locations of abnormalities are noted as on the face of the clock (eg, at 2:00 or 10:00). The approximate size is documented in millimeters.

The format of documentation will vary by practice of institution, depending on whether a paper chart or electronic health record (EHR) is used. Some EHRs have diagrams that can be drawn on. Some allow photographs or videos. This technology is particularly useful in educational settings. The clinician should be aware of potential medicolegal issues as future retrospective review of images may identify areas of abnormality that were initially missed. An advantage, however, is that the images may be stored as part of an EHR so that changes may be followed over time. Alternatively, the examination findings can be described in a note. An example of a comprehensive paper form is shown in the figure (form 1).

We also document the type of speculum used so that the appropriate speculum can be selected at subsequent visits.

Follow-up — We ask patients to avoid vaginal intercourse for 48 hours after cervical biopsies are taken to minimize trauma to the cervix, which may result in bleeding. Patients should be given both verbal and written instructions about how and when they will receive their results.

The provider should review the cytology and histology results and colposcopic findings with consideration of the patient's age; medical conditions; and prior cervical cytology, HPV test results, and histology history before establishing a diagnosis and management plan.

For patients with cytology that is either suggestive of possible malignancy or shows frankly malignant cells, the colposcopic examination will not be considered complete unless a biopsy is done that confirms the presence of cancer. In this situation, if invasive disease is not found and a lesser lesion such as CIN 3 is detected, repeat colposcopy by a more experienced colposcopist and review of the referring cytology slide are appropriate before considering more invasive intervention. Alternatively, if cytologic review is not possible, a cold knife cone or loop electrosurgical excision procedure can be performed (with one intact specimen to allow evaluation of margins). (See "Cervical intraepithelial neoplasia: Diagnostic excisional procedures".)

Management recommendations are discussed separately [31]. (See "Cervical cancer screening: The cytology and human papillomavirus report" and "Cervical intraepithelial neoplasia: Management".)

Complications — Potential complications include bleeding or infection at the biopsy site.

Significant bleeding and infection are rare. For bleeding, the first measure is typically to apply ferric subsulfate (Monsel solution) with either small or large cotton swabs to the biopsied areas. If necessary, sustained pressure with a cotton swab, silver nitrate sticks, or oxidized regenerated cellulose (eg, Surgicel) may also be used. In rare cases of significant bleeding, the vagina can be packed to apply pressure to the cervix. Ferric subsulfate solution and silver nitrate interfere with interpretation of biopsy specimen, so these substances should not be applied until after all biopsies have been taken.

Infection may present as postprocedure pelvic pain, purulent discharge, or abnormal bleeding or spotting. The patient should be evaluated for cervicitis with a pelvic examination and cervical testing for infection. In our practice, we treat these patients in the same manner as nonspecific cervicitis (eg, azithromycin monotherapy, doxycycline monotherapy, or a combined regimen of levofloxacin and metronidazole). Coverage for possible gonorrhea infection is included in empiric therapy if the patient is at increased risk for infection. (See "Acute cervicitis", section on 'Physical examination' and "Acute cervicitis", section on 'Treatment'.)

Effectiveness — Colposcopy is a diagnostic test used as a follow-up for patients with abnormal cervical cytologic results. It has not been found to be an effective screening tool for cervical cancer when used alone [2].

The efficacy of colposcopy to detect CIN and cervical cancer depends upon the experience and training of the colposcopist. An important factor is the ability of the colposcopist to interpret the colposcopic findings and obtain properly directed biopsies. Among experienced colposcopists, there is good interobserver agreement for normal epithelium; CIN 2,3; and invasive cancer [32]. There is more interobserver variation in diagnosis of CIN 1 [33]. However, this lack of agreement is also true for histopathologic diagnosis of these entities.

Colposcopy using the 2011 IFCPC criteria was able to distinguish normal cervix from CIN/carcinoma with a sensitivity of 86 percent and specificity of 30 percent, and to distinguish normal cervix/low-grade lesions (CIN 1) from high-grade lesions (CIN 2,3 or carcinoma) with sensitivity of 61 percent and specificity of 94 percent [18].

One explanation for limitations of the specificity of colposcopy is that acetowhite epithelium is often observed on the anterior and posterior ectocervix, even in the absence of CIN; this may lead to unnecessary biopsies, particular in these areas [34]. However, CIN appears to be similarly distributed in all four quadrants.

The diagnostic performance of colposcopy for detection of cervical neoplasia was evaluated in a meta-analysis of 32 studies including almost 8000 colposcopic punch biopsies; all patients subsequently underwent excisional biopsy (cone biopsy or loop electrosurgical excision procedure) and the results of excisional biopsy were used as the reference standard. To detect CIN 2 or higher in the excisional biopsy, a colposcopic punch biopsy of CIN 1 or higher had a sensitivity of 91 percent and a specificity of 25 percent and a punch biopsy result of CIN 2 or higher had a sensitivity of 80 percent and a specificity of 63 percent [35].

Performing two or more biopsies appears to increase sensitivity [22,24]. Options include performing additional biopsies from another part of the most abnormal-appearing lesion, from more than one abnormal-appearing area, or randomly from normal-appearing quadrants [22]. The best approach has not been studied, but at this time random biopsies from normal-appearing cervix do not appear warranted, since this would lead to biopsy of very large numbers of patients without disease and in the absence of information about the additional cost and risks of this approach [36].

VAGINAL COLPOSCOPY — Colposcopy of the vagina is very similar to that of the cervix, but the natural rugae and folds can make the procedure more time-consuming and difficult in premenopausal patients. Postmenopausal patients can have more atrophy or mucosal thinning, resulting in abrasions and tears from the speculum. In such patients, a limited course (eg, three months) of vaginal estrogen therapy may be used to improve colposcopic evaluation.

A systematic approach, beginning in one quadrant and gradually working to cover all four quadrants, involves applying acetic acid repeatedly, and noting areas of abnormal staining. Adenosis, pigmented lesions, and cysts are usually seen without the need of special stains. It is necessary to rotate the speculum to allow the anterior and posterior vagina to be visualized. Lugol staining is helpful in delineating lesions, and in ensuring that all abnormal areas have been identified.

Opening the speculum as much as is feasible pulls the vaginal tissue taut so the rugated surface appears nearly flat and facilitates colposcopy; however, when biopsy is attempted, it is helpful to partially close the speculum to allow the vagina to fold in on itself. Keeping the vagina taut makes biopsy difficult. Cysts and adenosis are usually easy to sample. The site of the lesion is documented using the vaginal apex, urethra, and fourchette as landmarks. Appropriate terminology should be used similar to cervical colposcopy (table 4 and table 5) [37].

Practice varies regarding whether anesthesia is used for vaginal biopsies. Some clinicians use local anesthesia for all vaginal biopsies, while others use it for only distal vaginal biopsies (the distal vagina has somatic innervation, so biopsies may be more painful; the proximal vagina has autonomic innervation).

Ferric subsulfate solution and pressure are usually all that are required for hemostasis.

VULVAR COLPOSCOPY — Vulvar colposcopy is indicated in patients with:

●Visible abnormalities of the vulva

●No abnormalities of the cervix or vagina that can account for the abnormal cervical cytology

●Focal vulvar itch, pain, or burning, without a clear etiology

Vulvar colposcopy is performed in a similar manner to cervical and vaginal colposcopy. Three to 5 percent acetic acid is applied to the external genitalia. It may take from three to five minutes for the solution to permeate the cells and for the lesion to appear white. Particular attention should be paid to periurethral and perianal areas.

If clearly delineated lesions are noted, a vulvar biopsy is performed under local anesthesia using a 3 to 5 mm punch biopsy and silver nitrate sticks for hemostasis. Alternately, a lesion may be excised with a scalpel and sutures placed. This may allow a lesion to be removed in its entirety. We typically apply a eutectic mixture of local anesthetics (EMLA) to the vulva approximately 20 to 30 minutes prior to the anesthetic skin injection and find that patients are comfortable during the biopsy/excision.

Specimens are individually labeled according to their location on the vulva (eg, numbered and diagrammed in the patient's record) and placed in separate specimen containers. With the shift to electronic medical records, we recommend that a picture be taken and placed in the patient's medical record if possible.

PREGNANT PATIENTS — If there is any possibility of pregnancy, a pregnancy test should be obtained. If pregnancy is possible and pregnancy status is uncertain, patients should be treated as if they are pregnant.

If colposcopy is indicated during pregnancy, in terms of timing, we prefer to see these patients early in the second trimester (12 to 15 weeks of gestation).

In general, biopsies of the ectocervix may bleed more during pregnancy. In our practice, we perform a biopsy in pregnant patients only if there are colposcopic features of high-grade or invasive disease. Endocervical curettage is not performed in pregnant patients because of potential trauma to the gestational sac and heavy bleeding.

The indications for colposcopy during pregnancy are discussed in detail separately. (See "Cervical cancer screening: Risk assessment, evaluation, and management after screening", section on 'Pregnant patients' and "Cervical cytology: Evaluation of atypical and malignant glandular cells".)

ALTERNATIVES — Colposcopy is the "gold standard" diagnostic tool in the United States for diagnosing cervical intraepithelial neoplasia following abnormal cervical cytology [38]; however, it is resource-intensive. The cervix can also be evaluated by other modalities. These may be divided into visualization techniques utilizing broad-band light (eg, direct visualization, speculoscopy, cervicography, and colposcopy) and technologies utilizing electronic detection methods (eg, Polarprobe and in-vivo spectroscopy). Four-quadrant biopsy of the squamocolumnar junction is also an acceptable alternative where equipment and experienced personnel to perform colposcopy are not available, but pathology evaluation of the specimens is available [24,34].

●Visualization techniques – Visual inspection techniques have lower specificity than cytology and are generally used in low-resource settings where cervical cytology and colposcopy are not feasible [39]. Visual inspection of the cervix is discussed in detail separately. (See "Cervical cancer screening tests: Visual inspection methods".)

●Electronic screening techniques

•Polarprobe and TruScan – These devices are not widely used. They utilize the principle that both normal and abnormal cervical tissue have characteristic electrical and optical properties which can be measured using a combination of various light and electronic frequencies. The probe is placed onto the surface of the cervix and signals are transmitted between the cervix and the portable console. The emitted tissue "signals" are then compared algorithmically with known signals stored in a databank of cervical tissue types.

•Spectroscopy – Spectroscopy relies on the principle of differential light emission by various tissue types. The targeted tissues have an "optical signature" determined by the amount of light they absorb and emit. Biochemical and structural changes that underlie properties of various tissue types permit detection of the differences between normal and abnormal cervical tissues using a specialized fluorescent probe. In comparative trials of fluorescence spectroscopy versus colposcopy [40] and other diagnostic techniques, spectroscopy performed better. However, because these approaches add cost but do not provide a pathologic diagnosis, their use in clinical practice is limited.

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Cervical cancer screening, prevention, and management".)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Beyond the Basics topics (see "Patient education: Management of a cervical biopsy with precancerous cells (Beyond the Basics)" and "Patient education: Follow-up of low-grade abnormal Pap tests (Beyond the Basics)" and "Patient education: Follow-up of high-grade or glandular cell abnormal Pap tests (Beyond the Basics)")

SUMMARY AND RECOMMENDATIONS

●Overview – Colposcopy is a diagnostic procedure in which a colposcope is used to provide an illuminated, magnified view of the cervix, vagina, and vulva to look for malignant and premalignant epithelium. Malignant and premalignant epithelium have specific macroscopic characteristics relating to contour, color, and vascular pattern that can be identified by the colposcopist for directed biopsy. (See 'Introduction' above.)

●Indications – Indications for colposcopy include further assessment of cervical cytologic abnormalities or abnormal human papillomavirus (HPV) tests, patients exposed to diethylstilbestrol in utero, patients with gross abnormalities of the lower genital tract, and as an adjunct to surgery of the cervix, vagina, and vulva. (See 'Indications' above.)

●Procedure

•The cervix and vagina are examined with a bright light and then with the colposcope, using saline if necessary. Pigmented areas and obvious lesions are noted. Three to 5 percent acetic acid is applied to the cervix and the cervix is reexamined. A green-filter examination is subsequently performed to accentuate abnormal vasculature. (See 'Gross visualization' above and 'Colposcopic examination' above.)

Metaplastic cells, dysplastic cells, and cells infected with HPV reflect light and thus appear white (picture 5A and picture 6A). Areas of white epithelium are further evaluated for abnormal vascular patterns, such as punctation, mosaicism, or abnormally appearing vessels (picture 3 and picture 4). (See 'Colposcopic examination' above.)

•The most abnormally appearing areas are biopsied. Biopsies are relatively contraindicated in patients on anticoagulation medication, who have a known bleeding disorder, or who are pregnant. (See 'Biopsies' above and 'Contraindications' above.)

•Endocervical sampling is performed according to the 2019 American Society for Colposcopy and Cervical Pathology (ASCCP) management guidelines, if ablative treatment is contemplated, and in those with a colposcopic examination in which the entire squamocolumnar junction (SCJ) could not be visualized or a lesion enters the endocervical canal. (See 'Endocervical sampling' above.)

•Documentation of the colposcopic examination needs to include a description of the SCJ (fully visualized or not (figure 3)), the size and specific qualities of any abnormalities seen, and whether the vagina or vulva was evaluated. For all patients undergoing cervical colposcopy, at minimum, evaluation of the upper fornices should be performed. (See 'Documentation' above.)

●Effectiveness – Colposcopy by an experienced colposcopist will usually detect patients with significant disease (ie, high-grade squamous intraepithelial lesion [HSIL], cancer), but some patients with a normal colposcopic examination can still have a significant histologic abnormality. For this reason, it is important to reconcile the cytologic, colposcopic, and histologic findings in formulating a diagnosis and management plan for each patient and following these patients closely. (See 'Effectiveness' above.)

●Follow-up – Patients with persistent abnormalities of cervical cytology over time or who repeatedly test positive for high-risk HPV, and whose initial colposcopy fails to reveal a significant lesion, should have multiple colposcopies and biopsies over time, and may even require a diagnostic excisional procedure, to rule out a high-grade lesion. (See 'Effectiveness' above and 'Follow-up' above.)