Cervical cancer screening tests: Techniques for cervical cytology and human papillomavirus testing

INTRODUCTION — Cervical cancer screening detects precancerous changes of the cervix (eg, cervical dysplasia), often making treatment possible before cervical cancer develops. Screening uses human papillomavirus (HPV) testing, cervical cytology (Pap test), or a combination of the two tests (eg, "co-testing").

Techniques for obtaining specimens for cervical cytology and HPV testing are reviewed here. Screening strategies and interpretation of the cervical cytology report are discussed separately. (See "Screening for cervical cancer in resource-rich settings" and "Cervical cancer screening: The cytology and human papillomavirus report".)

HOW TO OBTAIN A SAMPLE — Cell samples for cervical cytology and HPV testing are obtained during the speculum examination. Most often, the same specimen can be used for analysis of both cytology and HPV.

Specimens for cytology — There are two methods for preparing a specimen for cervical cytology (see 'Sample collection' below). For both methods, cells are obtained from the external surface of the cervix (ectocervix) and the cervical canal (endocervix) to evaluate the transformation zone (squamocolumnar junction), the area at greatest risk for neoplasia.

Collection device — Several collection devices are available for cervical cytology sampling. A spatula and a separate endocervical brush provide a specimen with more endocervical cells than when only a spatula is used [1]. It is also slightly better for detecting any grade of cervical intraepithelial neoplasia (CIN) than the single broom device [1]. Cotton tipped swabs should be avoided because they collect fewer endocervical cells and do not detect CIN as well as other devices [2-4]. A meta-analysis of 36 randomized trials and six observational studies in patients undergoing conventional Pap smears found that the most commonly used spatula (Ayre spatula) (figure 1) collected fewer endocervical cells than spatulas with extended tips (eg, Aylesbury), but both spatula types yielded similar diagnostic results [2].

Sample collection — To obtain cells from the cervix:

●Use the spatula to circumferentially scrape the ectocervix (for liquid-based samples, use a plastic rather than a wooden spatula; wood or plastic is fine for conventional smears) [5]. Sampling the ectocervix before the endocervix will minimize bleeding during sample collection. Obscuring blood in the sample interferes with interpretation of conventional Pap smears more than with liquid-based specimens.

●Insert the endocervical brush into the endocervix so that the bristles nearest the examiner are inserted to the level of the external cervical os. Rotate the brush 180 degrees to obtain a sample.

●Alternatively, if a broom is used, insert the central bristles into the endocervix with the outer bristles in contact with the ectocervix. Rotate the broom in the same direction for five turns [6]. Other devices on the market (eg, SpiraBrush, SoftBiopsy) have yet to be adequately studied with respect to safety and efficacy.

Preparation methods — There are two methods for preparing a specimen for cervical cytology: liquid-based thin layer preparation and conventional Pap smear .

●For liquid-based thin layer cytology, the collecting device is placed into a liquid fixative solution and vigorously swirled or rotated ten times in the solution [7] (figure 2). When the liquid is processed by the cytology laboratory, loose cells are trapped onto a filter and then plated in a monolayer onto a glass slide.

●For conventional Pap smears, the ectocervical spatula is smeared and the endocervical brush is rolled uniformly onto a single slide promptly after obtaining the specimens (figure 3). The slide is then rapidly fixed to avoid air-drying; the usual fixatives are either ethyl ether plus 95 percent ethyl alcohol or 95 percent ethyl alcohol alone. If spray fixatives are used, the spray should be held at least 10 inches away from the slide to prevent disruption of cells by the propellant. If co-testing is being performed, an HPV specimen is collected and sent separately.

For both methods, cells are obtained from the external surface of the cervix (ectocervix) and the cervical canal (endocervix) to evaluate the transformation zone (squamocolumnar junction), the area at greatest risk for neoplasia.

An advantage of some liquid-based systems is the ability to use a single specimen for cytology and testing for HPV. With conventional smears, a separate HPV test specimen has to be obtained if co-testing is performed. (See "Screening for cervical cancer in resource-rich settings", section on 'Types of screening and frequency'.)

Evidence regarding the screening efficacy with conventional and liquid-based Pap tests is discussed separately. (See "Screening for cervical cancer in resource-rich settings", section on 'Relative risks and benefits of each method'.)

Sample processing — Cytopathologists review cervical cytology slides. The interpretation of cytologic smears is subject to considerable interobserver variability, particularly in the case of nondiagnostic squamous and glandular atypias (atypical squamous cells of undetermined significance and atypical glandular cells of undetermined significance) [8].

In some settings, cytotechnologists and/or automated slide interpretation systems that use proprietary algorithms to perform initial review of the cytology slides to identify a subgroup of slides for subsequent evaluation by a cytopathologist. This subgroup consists of slides with high-powered fields that have specific abnormal cellular characteristics.

One example of an automated slide interpretation system is the ThinPrep Imaging System, which is approved by the US Food and Drug Administration (FDA) for primary screening of slides. This system uses programmed algorithms to review each slide for areas of most concern. If abnormalities are found, the whole slide is reviewed by a cytopathologist. In one study, use of this device increased detection of high-grade squamous intraepithelial lesions (HSIL) by 38 percent and low-grade squamous intraepithelial lesions by 46 percent compared with manual screening [9]. In another study, use of the imager resulted in fewer unsatisfactory slides than with conventional cytology (1.8 versus 3.1 percent) and better detection of HSIL [10]. The clinical effectiveness of automated systems and their role in cervical cancer screening have not been definitively established.

In the United States, quality assurance regulations require that laboratories rescreen 10 percent of randomly selected cervical cytology smears that were originally interpreted as negative [11]. Manual rescreening of all negative cytology smears is time consuming, although rapid manual rescreening (30 to 120 seconds per slide) is feasible and practiced in the United Kingdom and elsewhere.

Standardized terminology for reporting cervical cytology results was introduced with the Bethesda System in 1988, which was last revised in 2014 (table 1) [12-14]. (See "Cervical cancer screening: The cytology and human papillomavirus report".)

Most laboratories offer other tests to assist in reading abnormal cytology slides that have an increased detection of cervical abnormalities. These include commercial tests that detect molecular markers (eg, dual stain p16/Ki-67) that are highly associated with clinically relevant cervical neoplasia; testing negative for these markers is associated with almost no risk of disease [15-18]. Such testing is quickly becoming routine for cytopathologists to order as it aids in their diagnosis of equivocal cervical cytology (eg, atypical squamous cells of undetermined significance [ASCUS], low-grade squamous intraepithelial lesion [LSIL]).

HPV testing — HPV testing identifies oncogenic (ie, high-risk) HPV subtypes that are associated with cervical cancer (table 2). The subtypes that are tested have slight variation across the various testing systems, but all test for at least the 13 most common types. HPV genotyping refers to testing for individual HPV types, usually HPV 16 or 18, but some tests may also include HPV 45. (See "Virology of human papillomavirus infections and the link to cancer", section on 'HPV Genotypes and risk of cancer' and "Virology of human papillomavirus infections and the link to cancer", section on 'Molecular pathogenesis'.)

HPV testing systems are approved for either primary HPV testing (without cervical cytology) or co-testing (with cervical cytology) (table 3). In the United States, there are only two FDA-approved tests for primary HPV screening and five that are approved for co-testing. Tests that are FDA approved for co-testing are also suitable for reflex testing in response to a cervical cytology result of atypical squamous cells of undetermined significance (ASC-US). This is discussed in more detail elsewhere. (See "Screening for cervical cancer in resource-rich settings", section on 'Types of screening and frequency'.)

Clinical trial data vary for each HPV test. While there is also variability among laboratories regarding performance (ie, identifying cervical intraepithelial neoplasia [CIN] 2 or greater), all of the tests are highly sensitive in detecting HPV. As testing technology evolves, new tests are developed, or there are expanded approvals for previously approved tests, cervical screening testing management guidelines may take these subtle performance differences into account [19].

Cervical testing — Specimens for HPV testing can be collected from the endocervix using a cervical spatula or cervical brush, which is then placed in HPV test transport medium [20]. With some liquid-based cytology sampling systems, the same specimen can be used for HPV testing and cytology.

Self-collection of an HPV sample by the patient can also be performed; while this practice is well described in resource-limited settings, its use in resource-rich settings is more limited and self-sampling tests are not approved by the FDA. For self-collection, patients can collect samples from the vagina using a tampon, Dacron or cotton swab, cytobrush, or cervicovaginal lavage. Self-collection may increase screening uptake, especially for patients with barriers to health care. In randomized trials in the United States evaluating cervical cancer screening uptake, those with access to self-sampling HPV kits compared with usual care (ie, cervical cancer screening reminders) had higher rates of screening uptake [21-23]. A large, national, multicentric screening trial is underway [24]. Self-collection is discussed in more detail separately. (See "Screening for cervical cancer in resource-rich settings", section on 'Types of screening and frequency' and "Screening for cervical cancer in resource-limited settings", section on 'Self-collected samples'.)

Investigational strategies

●Urine testing – Urine testing for HPV has been proposed, but is not clinically available [25,26]. While urine testing does not appear to be as sensitive as cervical testing, it may have potential in large research studies or as an alternative test in settings where routine cervicovaginal examinations are not economically feasible or less likely to be performed due to cultural barriers.

●Menstrual blood sampling – Menstrual blood sampling for HPV testing has been described and may be a convenient approach for cervical cancer screening in premenopausal patients [27,28], but further studies are needed.

Additional tests — Additional testing that may be performed during examination of the cervix includes:

Gonorrhea, chlamydia, and trichomonas — It is common practice to collect the cytology sample before testing for cervical infection, if indicated. However, there is no evidence that the order in which the samples are obtained affects cytology results [29]. Liquid-based cytology systems allow testing for cytology, HPV, gonorrhea, chlamydia, and trichomonas from a single specimen. (See "Screening for sexually transmitted infections" and "Clinical manifestations and diagnosis of Neisseria gonorrhoeae infection in adults and adolescents", section on 'Performance on urogenital specimens' and "Clinical manifestations and diagnosis of Chlamydia trachomatis infections", section on 'Diagnosis of chlamydial infections' and "Trichomoniasis: Clinical manifestations and diagnosis", section on 'Less accurate'.)

Biopsy of visible lesions — During Pap testing, any lesion that is raised, friable, or has the appearance of condyloma should be biopsied, regardless of previous cytology results or other risk factors for cervical cancer [30]. The only visible lesions that do not require biopsy are Nabothian cysts and only when this diagnosis is confirmed by an experienced examiner. (See "Benign cervical lesions and congenital anomalies of the cervix", section on 'Nabothian cysts'.)

Anatomic barriers — In some patients, the cervix is difficult to visualize on pelvic examination. Factors that may make visualization difficult include:

●High body mass index [31].

●Prior cesarean section.

●A uterus that is sharply anteverted or retroverted.

●Obliteration of the vaginal fornices (from menopause-induced vaginal atrophy, prior pelvic radiation, or vaginal graft-versus-host disease).

If the clinician cannot see the cervix, options include the following:

●Use a longer Graves or Pederson speculum to reach the vaginal apex; press the speculum along the posterior vaginal wall until the apex is reached and then open the speculum slowly.

●Perform a bimanual examination to palpate the cervix and identify its location. In patients with obliteration of the vaginal fornices, palpation often allows the examiner to differentiate the firm cervical tissue from the surrounding vaginal walls. Lubricant is sometimes avoided as it can interfere with the ability to analyze the Pap specimen. (See 'Gel lubricants and other contaminants' below.)

●Improve visualization by optimizing the patient's position. In dorsal lithotomy, the following modifications can be used:

•Ensure that the patient's legs are sufficiently abducted. The patient may need to move toward the examiner. Care should be taken if the patient has knee or hip mobility issues.

•Elevate the sacrum by placing an object (bedpan, folded up sheet or towel) under the patient's hips.

●Confirm that the patient has a cervix (some patients who have undergone a total hysterectomy do not give an accurate surgical history).

In patients with cervical stenosis, it may be difficult to obtain an endocervical sample, thus resulting in an insufficient result. When it is difficult to insert the sampling device into the endocervix, one of the following techniques may facilitate collection of an endocervical sample:

●Perform Pap testing during menses. Menstrual blood often slightly dilates the cervix.

●Grasp the anterior or posterior lip of the cervix with a single-tooth tenaculum. Applying gentle traction will stabilize the cervix and may provide enough counter-tension to insert a sampling device into the external cervical os.

●Use a small mechanical dilator to dilate the cervix.

●Administer a para- or intracervical block to help with patient discomfort.

Follow-up in patients with a Pap test with absence of endocervical cells/transformation zone component is discussed separately [32-35]. (See "Cervical cancer screening: The cytology and human papillomavirus report", section on 'Absent EC/TZ'.)

FACTORS WITH LIMITED OR NO EFFECTS ON SAMPLING — There is perception that any action that may remove cells from the cervix (eg, prior Pap sampling, cervical cultures, swabbing) will impair Pap test cellularity, and thus compromise efficacy for cervical cancer screening. However, data do not support these concerns.

The factors discussed in this section relate to the effects on cytology or HPV testing or both.

Menses or other genital tract bleeding — Historically, patients planning to have screening cytology for cervical cancer have been advised to avoid testing during menses or other genital tract bleeding. We suggest performing rather than deferring the test, unless the blood cannot be cleaned from the cervix. Cleaning the cervix with a large cotton swab will remove obscuring blood and appears to have a minimal or no effect on sample cellularity and has no effect on detection of HPV [36,37].

If there is obscuring blood, conventional Pap smears are more likely to be unsatisfactory for interpretation than liquid-based methods because liquid-based techniques filter out red blood cells. This was demonstrated in a population-based retrospective study in the Netherlands in which over 100,000 patients who reported having regular menstrual cycles were screened for cervical cancer using the conventional Pap smear [5]. The rate of unsatisfactory smears was 23 percent during cycle days 0 to 3 versus 2 percent for the remainder of the cycle.

For liquid-based Pap tests, timing during the menstrual cycle does not appear to have a clinically significant effect on cytologic results [38,39]. This was illustrated by a large study in which 5060 patients with initial cytology showing atypical squamous cells of undetermined significance (ASC-US) or low-grade squamous intraepithelial lesions (LSIL) had over 20,000 liquid-based Pap tests [39]. The phase of the menstrual cycle did not have a significant effect on the rate of unsatisfactory specimens. Although the detection of LSIL or more severe abnormalities was slightly higher in the mid- versus early or late cycle (mid-cycle: 20 percent, early and late cycle: 18 percent), this difference is unlikely to be clinically significant.

Interval between Pap tests — A Pap test may need to be repeated after a brief interval if the sample is unsatisfactory or at the time of colposcopy. A short interval between Pap tests (15 to 30 days) does not appear to affect sensitivity for detection of cervical neoplasia [40,41].

The concern about a short interval between Pap tests is based on the hypothesis that previous scraping for cytology will remove the most superficial layer of cervical cells, where a potential abnormality is most likely to occur. It will then take a period of time (generally estimated as up to two months) for the superficial cells to regenerate [42]. Thus, if sampling is performed too soon, the underlying cells may be sampled and appear normal, yielding a false negative test.

The most informative study of the optimum interval between Pap smears evaluated liquid-based Pap tests and HPV testing results in 5055 patients with initial Pap tests that showed either ASC-US or LSIL [41]. Sensitivity of repeat specimens was assessed by comparing cytology with histology in patients with cervical intraepithelial neoplasia 3 or carcinoma. The interval from initial to repeat Pap test in these patients was 8 to 184 days. Adequacy of the sample did not appear to be affected, since cellularity of the sample and HPV viral load did not vary with Pap intervals. In addition, Pap test interval (15 to 120 days) did not significantly affect the finding of abnormal cytology. In fact, the likelihood of a Pap result of LSIL was higher after a shorter versus a longer interval (≤30 days [30 percent] versus >120 days [20 percent]); this likely represents regression of LSIL with time.

Gel lubricants and other contaminants — Contaminants, such as gel lubricant (eg, from an examiner's hand before a Pap test is performed), vaginal discharge, semen, spermicide, or intravaginal medications, have been thought to affect cervical sampling. On a conventional smear, the concern is that these may make the smear thick and difficult to read or, in the case of lubricants that include carbomers or carbopol polymers, may interfere with sample interpretation. In addition, in our experience, a sample is often returned by the laboratory with a note regarding difficulty in interpretation because of lubricant. Thus, if large amounts of vaginal contaminants are present, the discharge should be removed gently with a large cotton swab. Routine removal of a small amount of discharge or other contaminants is unnecessary. Clinicians should also check with their cytology laboratory to determine approved lubricants.

In general, studies have not shown an adverse impact of lubricants on cervical cytology interpretation [43-47]. Studies of use of nonoxynol-9 spermicide have had conflicting results regarding cervical cytology changes [48,49]. The spermicide nonoxynol-9 is not active against HPV, but detergents, such as sodium dodecyl sulfate (SDS), do inactivate HPV, and a spermicide/SDS combination could be useful in preventing HPV transmission [50,51]. Nonoxynol-9 spermicide is discussed in detail separately. (See "Pericoital (on demand) contraception: Diaphragm, cervical cap, spermicides, and sponge", section on 'Vaginal spermicide and pH regulator gel'.)

As discussed above, testing for Neisseria gonorrhea and Chlamydia trachomatis cervical infection is often performed concurrently with cervical cytology. Many clinicians avoid use of gel lubricants prior to testing for these bacteria, since some lubricants are bacteriostatic (eg, those containing chlorhexidine gluconate). However, one of the studies described above also evaluated DNA probe testing in 5535 patients in whom gel or water was used to lubricate the speculum prior to testing [43]. No significant differences were found between groups in the detection rate of chlamydia (1.5 percent in both groups); the gonorrhea infection rate was too low to analyze. (See "Screening for sexually transmitted infections".)

Vaginal intercourse, douching, and tampon use — While historically patients were advised to refrain from vaginal activities (eg, douching, tampon use, sexual intercourse) during the 48 hours prior to a Pap test, advising patients to avoid vaginal activities may be cumbersome to clinicians and make timely scheduling of Pap tests difficult for patients. There are few data that directly assess the effect of vaginal activities on cervical cytology or HPV testing, and we do not inquire about these activities prior to performing these tests.

In a study of HPV testing in which patients performed self-sampling with synthetic polyester swabs or a tampon, there was no significant effect on HPV detection when vaginal intercourse occurred within 48 hours of sampling [52]. Although it is possible that HPV detected in cytology samples following recent sexual activity could be derived from the male partner, this should not alter the standard management algorithm. In a subsequent study evaluating 132 patients with abnormal cervical cytology, HPV detection was similar before and after douching (sensitivity was 98 percent before douching; 96 percent after douching); results of the effect of douching on cytology results were not reported [53].

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.)

●Basics topics (see "Patient education: Cervical cancer screening tests (The Basics)")

●Beyond the Basics topics (see "Patient education: Cervical cancer screening (Beyond the Basics)")

SUMMARY AND RECOMMENDATIONS

●Clinical significance – Cervical cancer screening tests use human papillomavirus (HPV) testing, cervical cytology (Pap test), or a combination of the two tests (eg, "co-testing") to detect cellular changes of the cervix (eg, cervical dysplasia) or infection with types of HPV that may predispose patients to invasive cervical cancer. (See 'Introduction' above.)

●Specimens for cytology

•With liquid-based methods, the specimen is placed into a liquid fixative solution (figure 2). Conventional cervical smears are performed by smearing the specimen on a slide. Both methods are referred to as cervical cytology or a Pap test. (See 'Preparation methods' above.)

•Several types of collection devices can be used for cervical cytology sampling. (See 'Collection device' above.)

●HPV testing – HPV testing detects types of the virus that are associated with a high risk of cervical neoplasia. HPV testing systems approved for either primary HPV testing (without cervical cytology) or co-testing (with cervical cytology) are shown in the table (table 3). (See 'HPV testing' above.)

●Factors with limited or no effects on sampling

•Menses (or other vaginal bleeding), sexual intercourse, douching, tampon use, and a short interval between Pap testing do not appear to diminish the ability to diagnose cervical abnormalities or HPV infection. (See 'Factors with limited or no effects on sampling' above.)

•Use of some lubricants before performing a Pap test may interfere with results of cytology. Clinicians should check with their cytology laboratory for approved lubricants. (See 'Gel lubricants and other contaminants' above.)