Colposcopy 

Updated: Jan 31, 2018
Author: Stephen A Metz, MD, PhD; Chief Editor: Michel E Rivlin, MD 

Overview

History of the Procedure

Literally translated, colposcopy (colpo: vagina; scope: to look) means to look into the vagina. Colposcopy was first described by Hans Hinselman of Germany in 1925 as a screening tool for cervical cancer.[1] Hinselman suspected that endophytic or exophytic lesions of the cervix were likely precursors of cervical carcinoma, and he hoped that by magnifying these tissues, precursor lesions might be identified early enough to allow effective treatment before invasive disease developed or spread. His theories on the genesis of cervical cancer were incorrect, and his protocol for colposcopic evaluation was clinically impractical, so the search for alternative methods for cervical cancer screening continued.

In 1920, Dr. George Papanicolaou began his investigation of vaginal and cervical cytology. In collaboration with Herbert Traut, he published his seminal monograph Diagnosis of Uterine Cancer by the Vaginal Smear in 1943.[2] In the United States, the Papanicolaou test became the primary screening test for detection of cervical cancer and its precursors. Colposcopy was essentially unknown in this country until the 1960s, when it was introduced in its current role as a confirmatory test for evaluation of women with abnormal cervical cytologic findings. Currently, it has near-universal acceptance as the most effective follow-up test for women suspected of having premalignant or malignant cervical lesions.

Problem

Until the early 1970s, the usual management of women found to have an abnormal Papanicolaou test result was cone biopsy of the cervix. In most cases, examination of the biopsy specimen revealed only minor epithelial changes. Therefore, most women were being overtreated at significant human, as well as financial, cost. The acceptance of colposcopic evaluation as an intermediate step to identify patients who truly require surgical therapy has resulted in a dramatic decrease in such unnecessary surgery.

Epidemiology

Frequency

The importance of accurate screening and evaluation for cervical cancer cannot be overstated. At the beginning of the 20th century, cervical cancer was the most common cancer to affect women worldwide, as well as the most common malignancy among females in the United States. In large part due to the widespread implementation of effective cervical cytologic screening programs (using the Papanicolaou test), the frequency of cervical cancer has decreased dramatically in the United States. Cervical cancer remains a problem in unscreened populations.

The National Cancer Institute estimates that in 2009 there were 11,270 new cases of cervical cancer diagnosed in the United States, and that approximately 4070 women died of this disease. Additionally, 45,000 women are found to have high-grade premalignant cervical epithelial lesions. Outside the United States, especially in developing countries, the prevalence of cervical cancer remains high. At the end of the 20th century, it remained the secondmost common cause of cancer death in women worldwide.

Etiology

The initiating event for cervical cancer in almost all cases is infection of the cervical epithelium by one of several serotypes of human papillomavirus (HPV). HPV DNA is found in more than 99.7% of all cervical cancers. More than 100 serotypes exist in this family of DNA viruses, approximately 30 of which have a predilection for the female genital tract. Approximately 15 of these serotypes have been found in cervical cancer. Serotypes 16 and 18 account for almost 70% of cervical cancer.

Immunosuppressed women, such as patients with AIDS, and women exposed to certain environmental factors (notably cigarette smoke), are at increased risk.

Pathophysiology

Squamous cell carcinoma accounts for approximately 80% of cervical cancer cases. Significant advances have occurred in the understanding of the pathophysiology and molecular basis of this disease. These advances have resulted in modification of management recommendations. The previously held concept that cervical cancer was the end-stage of an inexorable progression of cervical morphology from atypia through progressively more severe intraepithelial neoplasia to eventually become invasive squamous carcinoma, is overly simplistic.

As noted, the critical step in the initiation of the malignant process is epithelial infection with an oncogenic virus. When a cervical epithelial cell of a susceptible host is infected by one of these high-risk viruses, the viral genome is incorporated into the host cell genome. Inactivation of the p53 tumor suppressor gene and retinoblastoma gene by the viral genome results in immortalization of the infected cells. Absent effective immune surveillance by the host, the affected tissue may eventually undergo malignant transformation. These patients typically present with higher-grade lesions (CIN II, III, or frank cancer).

Conversely, low-grade cervical intraepithelial neoplasia (CIN I) lesions are almost always the result of infection with a nononcogenic virus, typically serotypes 6 and 11. Infection of cervical epithelial cells by one of these serotypes may result in exuberant cell proliferation during the course of viral replication (essentially resulting in formation of a wart), and these patients will present with CIN I lesions. However, the normal process of cell maturation and apoptosis is preserved, and malignant transformation is very rare. Spontaneous resolution of CIN I lesions is common. Conversely, spontaneous resolution of higher-grade lesions becomes progressively less certain, and active therapy is usually indicated.

Indications

Colposcopic evaluation is indicated when the presence of a malignant or premalignant lesion in the cervix, vagina, or vulva is suspected. Uncommonly, this situation may arise when an unusual cervical lesion is detected as the result of inspection of the vagina and cervix for an unrelated reason.[3]

Another specific indication is evaluation of a patient who presents with postcoital vaginal bleeding. Although this symptom is most commonly the result of vaginal or cervical trauma, postcoital bleeding is a classic symptom of cervical cancer.

The most common situation arises as the result of cervical cytological screening as part of a routine health maintenance evaluation.

Almost all United States cytopathology laboratories use the Bethesda System for reporting findings of cervical cytology. The version in current use is the 2001 revision.[4, 5]

Following are pertinent components of reports issued under this system:

  • Specimen adequacy

    • Satisfactory/unsatisfactory for evaluation

    • Presence or absence of endocervical or transformation

      • Zone components

  • General categorization

    • Negative for intraepithelial lesion or malignancy

    • Epithelial cell abnormality

  • Interpretation/result

    • Epithelial cell abnormality

      • Atypical squamous cells, undetermined

      • significance (ASC-US)

      • Atypical squamous cells, cannot exclude high-grade squamous intraepithelial lesion (ASC-H)

      • Low-grade squamous intraepithelial lesion (LSIL)

      • High-grade squamous intraepithelial lesion (HSIL)

      • Squamous cell carcinoma

    • Glandular cell abnormality

      • Atypical glandular cells (AGC)

      • Endocervical adenocarcinoma in situ (AIS)

      • Adenocarcinoma

Evaluation of the patient with an abnormal pap smear has been evolving. In September 2006, a panel of experts representing 29 organizations and professional societies issued a set of consensus guidelines reflecting current understanding of the pathophysiology and natural history of cervical dysplasia.[6]

Importantly, the cytologic classifications do not correspond to histologic evaluation of tissue samples, and patient management ultimately must be based on histologic diagnosis.[7] In the special case of adolescent women, who are at very low risk for epithelial cervical cancer, and in whom low-grade changes usually resolve spontaneously, colposcopy is recommended to be reserved for high-grade SIL. In the special case of pregnant women, deferring colposcopy until after delivery is reasonable unless cytologic evaluation suggests significant risk of invasive disease.[8]

The detailed guidelines and rationale are available through the American Society for Colposcopy and Cervical Pathology (ASCCP) or from the following reference:

Massad LS, Einstein MH, Huh WK, et al. 2012 updated consensus guidelines for the management of abnormal cervical cancer screening tests and cancer precursors. J Low Genit Tract Dis. Apr 2013;17(5 Suppl 1):S1-S27.[9]

They may be summarized as follows:

  • Negative for intraepithelial abnormality: These patients should continue to receive routine cytologic screening.

  • ASC-H, LGSIL, HGSIL, squamous cell cancer: Because of the increased incidence of significant histologic abnormality in patients who have these diagnoses, colposcopic evaluation and biopsy of abnormal sites is recommended. In the special case of postmenopausal women, atrophic changes may result in the cytologic diagnosis of LGSIL. Managing these women similar to those with a diagnosis of ASC-US is appropriate.

  • ASC-US: As with the other abnormal Papanicolaou test classification, colposcopic evaluation is an option. However, because this is the least reproducible of the cytologic categories, and because of the very low risk of invasive cancer in patients who have this diagnosis, less aggressive management strategies are also options. These include the following:

    • Repeat cytologic evaluation at 6 and 12 months

    • “Reflex” testing for the presence of high-risk HPV serotypes.

    • If either of these strategies results in abnormal findings, colposcopy is indicated. Should no abnormality be discerned, the patient may resume routine surveillance.

  • AGC: This finding is not uncommonly associated with significant glandular or squamous cell abnormality somewhere in the lower genital tract. As neither repeat cytologic testing nor HPV testing has been found to have sufficient sensitivity to detect endocervical or endometrial abnormalities, these patients are recommended to undergo colposcopic, endocervical and endometrial evaluation, and sampling in addition to HPV testing.

  • AIS, adenocarcinoma: These patients may require excisional procedures to assure adequate sampling should the above measures not be diagnostic.

Relevant Anatomy

The ring of cervical epithelium between the native squamous epithelium of the outer cervix (or cervical portio) and the columnar epithelium of the endocervical canal is a region of significant morphologic fluctuation.

The columnar epithelium of the endocervical canal often extends out onto the cervical portio prior to menarche. Under the influence of environmental factors in the vagina, especially increased acidity associated with menarche, the exposed columnar epithelium undergoes morphologic change referred to as squamous metaplasia, a thin layer of stratified squamous epithelium overlying the columnar epithelium. See the image below

Early metaplastic change of columnar epithelium. Early metaplastic change of columnar epithelium.

The ring of squamous metaplasia is referred to as the transformation zone. The border between the metaplastic epithelium and the endocervical columnar epithelium is referred to as the squamocolumnar junction. Because viral infections most commonly affect cells in rapidly dividing epithelium, the transformation zone is typically the locus of dysplastic change. Therefore, complete visualization of the entire transformation zone, most specifically the squamocolumnar junction is required for adequate assessment.

Colposcopic evaluation is aided by the differential reaction of the cells of mature squamous tissue and actively dividing squamous tissue to dilute acetic acid and iodine. Actively dividing cells are characterized by an increased nuclear–cytoplasm ratio. This is especially true of HPV–infected cells (see image below).

Human papilloma virus (HPV)–infected epithelium wi Human papilloma virus (HPV)–infected epithelium with increased light reflectance.

Exposure of such tissue to dilute acetic acid results in increased light reflectivity and, therefore, visual contrast with normal tissue (see image below). Mature squamous epithelial cells contain a large amount of glycogen, which stains a characteristic mahogany color when exposed to iodine (the Schiller reaction), whereas rapidly dividing cells, which contain relatively little glycogen, remain unstained.

Normal squamocolumnar junction. Tissue response to Normal squamocolumnar junction. Tissue response to applied light.

Another hallmark of intraepithelial neoplasia and invasive disease is alteration of the microvasculature of the transformation zone. Columnar epithelium is characterized by a single layer of columnar cells overlying stromal fronds. Each of these villous fronds protruding into the columnar-lined endocervix has its own capillary supply.

During the process of squamous metaplasia, these vessels become covered over and therefore less visible. Cellular proliferation associated with viral infection or dysplastic transformation (shown in the image below) is associated with vascular changes.

Dysplastic epithelium with increased reflectance o Dysplastic epithelium with increased reflectance of applied light.

There is in-growth of the perpendicular vessels into the epithelium. Viewed end-on under magnification, these vessels appear as small red dots (referred to as “punctation,” depicted in the image below).

Punctation in cervical lesions. Left is fine, righ Punctation in cervical lesions. Left is fine, right is coarse (likely high grade or invasive).

The vessels can become interconnected in intricate patterns that resemble a tiled floor (referred to as “mosaicism”, depicted in the image below).

Mosaicism in cervical lesions; left is fine, right Mosaicism in cervical lesions; left is fine, right is coarse (likely high grade or invasive).

Some relationship appears to exist between increased vascular spacing and degree of dysplasia, but this has proven to be an insensitive marker. On the other hand, the presence of a vessel course irregularity (eg, sudden vessel termination, “hairpin” or “comma” type vessels, abnormal branching, or increasing diameter of a vessel previously diminishing in caliber) is suggestive of malignant transformation.

Contraindications

No absolute contraindications to the performance of a colposcopic examination exist, although special precautions may be required in special circumstances, such as a pregnant patient who harbors a placenta previa. The patient's ability to tolerate a standard speculum examination is the only true limiting factor. Pregnancy by itself is not a contraindication. Active cervicitis and vulvovaginitis should be treated before undertaking the examination, because inflamed tissues can alter the ability to obtain an accurate assessment, and it can also make the discomfort of the examination markedly worse.

 

Workup

Laboratory Studies

A report by Castle et al indicates that the Xpert human papillomavirus (HPV) assay is a sensitive and reliable diagnostic tool for detecting high-risk HPV (hrHPV) DNA as well as grade 2 or greater cervical intraepithelial neoplasia (CIN2+) in a colposcopy referral population.[10] The investigators collected 2 cervical specimens from 708 US women and tested the samples with either the Xpert hrHPV DNA assay or the assay in conjunction with the Hybrid Capture 2 and the cobas HPV tests.

They found no statistical difference in positive results between the two specimens for the hrHPV DNA assay.[10] The sensitivity for identifying CIN2+ was 89.0% in the specimens tested only with the Xpert hrHPV DNA assay; in the specimens evaluated with all three tests, the sensitivity was 90.4% each for the Xpert hrHPV assay and the cobas HPV test, and it was 81.6% for Hybrib Capture 2 test.

Diagnostic Procedures

Colposcopic evaluation of the cervix should be a simple procedure. However, it is associated with high levels of patient anxiety which can have consequences such as pain and discomfort with the procedure and failure to return for followup. Anxiety may be reduced by a variety of interventions including playing music and viewing the procedure on a TV monitor. Providing patients with information leaflets increases knowledge levels and reduces psychosexual dysfunction.[11]

The colposcopic evaluation must be performed thoroughly and accurately. Accomplishing this for every patient is most reliably accomplished if a systematic repeatable routine is developed and followed for each patient. Below is one such protocol.

  • Explain to the patient the indication for and nature of the procedure.

  • Position the patient as comfortably as possible in the lithotomy position.

  • Carefully insert a speculum of the appropriate size. For patient comfort, a water-based lubricant thinly applied to the speculum blades can be of benefit. This substance should not distort the subsequent evaluation in any way. Care should be taken to avoid any trauma to the cervix on insertion or opening of the speculum. Normal columnar epithelium and dysplastic epithelium can be very fragile, and even minor trauma from speculum placement can cause enough oozing of blood to obscure findings.

  • Inspect the vagina and cervix visually with the naked eye. Gently remove any excess mucus or discharge with a large cotton-tipped applicator moistened with saline. Document any clinical findings on this gross inspection.

  • Liberally apply 3-5% acetic acid with a large cotton swab saturated with the solution. This must be in place for at least 60 seconds prior to inspecting for changes. If the evaluation takes more than 3-5 minutes, acetic acid should be reapplied because the cellular effects it creates are transient in nature.

  • Position the colposcope and focus on the cervix with the desired magnification (7X-15X).

  • Inspect carefully to ensure that the entire transformation zone (TZ) can be observed (ie, that the squamocolumnar junction is visible for its entire circumference). If the TZ extends into the cervical canal, the use of an endocervical speculum can aid in visualization. If the entire TZ cannot be observed, or if the full extent of any lesion cannot be visualized, the evaluation must be considered unsatisfactory.

  • Identify and document with drawings and description the presence of any acetowhite lesions and their internal vascular patterns. Use of the green filter at this point can improve the ability to identify lesion margins and vascular patterns. Many expert colposcopists also place Lugol solution (dilute iodine) on the cervix after initial examination and before any biopsies are obtained. This is a rather nonspecific staining process. It can be useful to provide additional information regarding the extent of abnormal epithelial changes.

  • Biopsy samples should be obtained from all abnormal lesions. As noted, the visual appearance of lesions is a poor predictor of degree of dysplastic change. Histologic analysis of all lesions is necessary to optimize sensitivity of the examination and minimize the risk of missing a significant abnormality. With experience, the examiner may become comfortable in determining if 2 locations are identical, and therefore not have to biopsy each location. Biopsy samples can typically be obtained without the need for anesthetic, but its use is not precluded. Biopsy instruments should be of a 2-bladed type (eg, Tischler, Burke, Kevorkian) and kept sharp and in good working condition. Specimens should be removed from the instrument and placed in an appropriate fixative in a labeled container and submitted for histologic evaluation.

  • Cytologic sampling of the endocervix with a cytobrush, or, alternatively, endocervical curettage may be used to evaluate the patient for endocervical pathology. Studies indicate that cytologic endocervical sampling with a cytobrush has increased sensitivity but decreased specificity compared with endocervical curettage. With either technique, the sample is submitted in fixative. Pregnancy is a relative contraindication for such endocervical sampling.

  • A hemostatic agent can be applied to each biopsy site immediately after sample collection. Monsel paste (dehydrated ferric subsulfate) and silver nitrate are effective measures. If multiple biopsies are indicated, initial samples should be taken from the inferior aspect of the cervix to prevent bleeding or chemical application from running down and obscuring adequate visualization.

    • Special precautions should be taken to optimize hemostasis when colposcopy is being performed on a pregnant woman. As is true for nongravid women, biopsy of the cervix should be performed for any lesion suspected of being invasive. Bleeding from even small biopsy sites can be brisk and persistent, so special preparation for this likelihood should be undertaken. While positioning the biopsy forceps to obtain a sample, a cotton swab saturated with Monsel paste should be readied immediately adjacent to the instrument. As soon as the biopsy is taken and before removal of the specimen, the swab should be firmly applied to the wound bed. A large second swab should be ready to put in place after removal of the first. Should this not control bleeding, equipment for placement of a small suture should be immediately available. If proper precautions are taken, bleeding is seldom a significant problem, even in the case of pregnant patients.

  • The speculum is removed, and patient instructions are provided. Spotting and a light discharge can be anticipated. Coitus should be avoided for 7-10 days, and a follow-up examination and discussion of pathologic findings should take place in 1-2 weeks.

  • Results should be reviewed to confirm that they correspond to the cytologic reports. The colposcopic examination should be considered inadequate for any of the following reasons:

    • The biopsy histology is less severe (typically by ≥2 grades) than that predicted from the cytologic sample.

    • The endocervical margin of any abnormal area cannot be visualized completely.

    • The squamocolumnar junction cannot be visualized completely.

    • There is evidence of endocervical disease not visualized at the time of colposcopy.

    • Histologic or cytologic suggestion of invasive disease not confirmed by biopsy.

    • Histologic or cytologic suggestion of adenocarcinoma or adenocarcinoma in situ.

 

Treatment

Surgical Therapy

Following complete colposcopic evaluation of the cervical lesion and histologic confirmation of the diagnosis, an appropriate treatment regimen can be developed. This should be based on the extent and degree of the pathologic findings, taking into account, insomuch as possible, the individual patient’s desire for future childbearing.

Follow-up

Regardless of the treatment modality chosen, patients undergoing therapy for preinvasive cervical lesions are at risk for recurrence. Follow-up algorithms are available from various sources, such as the American Society for Colposcopy and Cervical Pathology.

For patient education resources, see Cancer Center and Women's Health Center, as well as Colposcopy, Cervical Cancer, and Pap Smear.

Complications

Complications from colposcopic procedures are exceedingly rare. Occasionally, bothersome bleeding can occur following biopsy. This tends to be problematic only with procedures performed during pregnancy or with large excisional procedures. Infection of biopsy sites is also exceedingly rare, although it can occur following laser ablation or LEEP procedures. The most worrisome complication is inadequate or inaccurate evaluation leading to the missed diagnosis of invasive cancer. This obviously can lead to treatment delays and poorer outcomes. Another complication is the overestimation of lesion severity by inexperienced practitioners. This can put the patient on a treatment course that may not be necessary and has the potential for adverse sequelae. Many of these sequelae center around future fertility limitations such as cervical stenosis or incompetence.

The infrequent but preventable lack of adequate evaluation is the only real controversy surrounding the procedure today. Questions concern who should perform the examination and what training requirements must be met before instituting the procedure on patients. Because of the prevalence of HPV disease and the frequency of abnormal findings on Papanicolaou tests, this becomes both an economic issue and a quality issue. Some have recommended as many as 200 supervised procedures to gain competence followed by regular performance of at least 25 procedures a year to maintain competence. The learning curve undoubtedly is practitioner-dependent, and, currently, no adequate studies have identified minimum criteria for certification. All practitioners performing this procedure should put mechanisms in place to ensure their own competence and safety.

Future and Controversies

The colposcope can also be helpful in evaluating lesions of the vagina or vulva. The vaginal epithelium is a nonkeratinizing squamous epithelium similar to that of the exocervix. Acetowhite changes and vascular patterns can be observed that are similar to those found on the cervical portio. Because vaginal lesions do not originate in metaplastic tissue, vascular patterns previously described are not diagnostically reliable. When premalignant changes are suspected, all acetowhite lesions should be biopsied.

The vagina is more sensitive to pain than the cervix, so prebiopsy injection of local anesthesia should be considered. However, in a 2014 study involving 214 women with abnormal cervical cytologic findings that required colposcopy and directed cervical punch biopsy with/without endocervical curettage, Oz et al found no significant differences in pain scores after the colposcopic cervical biopsies with endocervical curettage between women who received 10% topical lidocaine spray and those who received placebo.[12]

The vulva is also a potential site for development of preinvasive disease. These tissues also can show acetowhite changes, but, because of the thickness of the epithelium and its keratin surface, acetic acid should be applied for a greater length of time and in a higher concentration (eg, 5%) to be effective in bringing about this change. Altered vascular patterns are uncommon on the vulva; but, when they are observed, biopsies should be performed liberally. Again, because of the sensitivity of these tissues, all biopsies should be obtained under local anesthetic.

Another use of the colposcope is in the evaluation of a victim of sexual assault. This has gained popularity, especially in the evaluation of children suspected of being assaulted. At low magnification, the colposcope can assist in identifying tissue trauma that might be too subtle to be detected by the naked eye. Careful, thorough, and gentle examination, especially of hymenal tissues, can usually be accomplished with minimal discomfort. Attached cameras for recording findings can be helpful from an evidentiary perspective.

An exploratory study by Mueller et al reported that cervix images from the low-cost point-of-care tampon colposcope (POCkeT) were comparable to a standard-of-care colposcope.[13]